Learning

Computing
With
Robots

Edited by
Deepak Kumar

Institute for Personal Robots in Education

www.roboteducation.org
This text is provided as is with no guarantees of any kind.

Permission is hereby granted to reproduce facsimiles of this text in any form for distribution
provided an e-mail note is sent to the editor at [email protected] describing the
manner and intended purpose of distribution.

Permission is also granted to modify the content of this text in any manner seemed desirable
by instructors. We recommend that all derivations from this text be distributed through IPRE
in the same manner as this text. To obtain the permission we require the same process as laid
out for distribution above.

We strongly recommend distributing the text and its modifications with a not for profit
motive.

We are making this text available in the manner described above so as to enable the widest
possible accessibility of the content as well as the ideas illustrated in this text. We hope that
users will respect and honor and promote such practices with their own works.

We have taken great care to attribute all images used in this text to their original sources. This
text is intended for educational purposes and such use constitutes fair use. Where feasible, we
have obtained permissions to use copyrighted images. Please contact the editor in case there
are any pending issues in this regard. All images not explicitly attributed in the text are
created by members of the IPRE team.

This version was produced in August, 2008. Revised in June 2009.

ISBN-10: 1607028832
ISBN-13: 978-1607028833
Institute for Personal Robots in Education
Fall 2008, Revised June 2009
Contributors
IPRE was created to explore the use of personal robots in education
with primary funding from Microsoft Research, Georgia Institute of
Technology, and Bryn Mawr College. This text would not have been
possible without the hardware, software, and course materials
developed by members of the IPRE team.

Ben Axelrod, Georgia Institute of Technology

Tucker Balch, Georgia Institute of Technology
Douglas Blank, Bryn Mawr College
Natasha Eilbert, Bryn Mawr College
Ashley Gavin, Bryn Mawr College
Gaurav Gupta, Georgia Institute of Technology
Mansi Gupta, Bryn Mawr College
Mark Guzdial, Georgia Institute of Technology
Jared Jackson, Microsoft Research
Ben Johnson, Georgia Institute of Technology
Deepak Kumar, Bryn Mawr College
Marwa Nur Muhammad, Bryn Mawr College
Keith O’Hara, Georgia Institute of Technology
Shikha Prashad, Bryn Mawr College
Richard Roberts, Georgia Institute of Technology
Jay Summet, Georgia Institute of Technology
Monica Sweat, Georgia Institute of Technology
Stewart Tansley, Microsoft Research
Daniel Walker, Georgia Institute of Technology

i
Learning Computing With Robots

ii
 Learning Computing With Robots

Contents

Preface v

Chapter 1
The World of Robots 1

Chapter 2
Personal Robots 21

Chapter 3
Building Robot Brains 43

Chapter 4
Sensing From Within 71

Chapter 5
Sensing The World 97

Chapter 6
Insect‐Like Behaviors 129

Chapter 7
Behavior Control 157

iii
Learning Computing With Robots

Chapter 8
Sights & Sounds 187

Chapter 9
Image Processing & Perception 215

Chapter 10
Artificial Intelligence 245

Chapter 11
Computers & Computation 273

Chapter 12
Fast, Cheap & Out of Control 309

Python Overview 321

Myro Overview 329

Index 340

iv
 Preface

What is a computer? A personal computer?

What is a robot? A personal robot?
What is computing? What is personal computing?

This book will introduce you to the world of computers, robots, and
computing. Computers, as we know them today have been around less than 75
years. To our young audience for whom this book is designed, that may seem
like an eternity. While computers are all around us, the world of computing is
just in its infancy. It is a baby!

In this book, you will learn that computing is no more about computers than
astronomy is about telescopes. In that respect, the roots of computing could be
traced back more than 2000 years to Euclid in 300 BC. Could it be that
computing is actually as old as or even older than some of the other
disciplines? Prior to the development of modern computers, the word
computing mostly implied numerical calculation. Today, computing
encompasses all forms of electronic manipulation of objects: numbers, text,
sound, music, photographs, movies, genes, motors, etc., and even ideas! Every
day we are discovering new things that can be computed.

Robots are not that much older than computers. We will learn that, contrary to
popular belief, a robot is any mechanism that is guided by automated control.

v
Preface

A robot can be controlled as a result of computation. In fact, in today’s robots

computation is a key component.

Advances in the past few decades have brought about many changes in the
way we use and live with technology. Computers have become smaller,
cheaper, faster, and more personal. The same is also true for robots. In this
book we have created a confluence of these two technologies to use as a basis
for introducing the world of computing to students.

The word personal is key to this new confluence. Computers and robots are
no longer the realm of large corporate offices and industrial manufacturing
scenarios. They have become personal in many ways: they help you write
your term papers, store and organize your photographs, your music collection,
your recipes, and keep you in touch with your friends and family. Many
people might even dismiss the computer as an appliance: like a toaster oven,
or a car. However, as you will learn in this book, a computer is far more
versatile than most appliances. It can be used as an extension of your mind.
We were not kidding above when we mentioned that computers can be used
to process ideas. It is in this form, that a computer becomes an enormously
powerful device regardless of your field of interest. How personal you can
make a computer or a robot is entirely up to you and your abilities to control
these devices. This is where this book comes in. You will learn the basics of
how a computer and a robot is controlled and how you can use these ideas to
personalize your computers even further, limited only by your imagination.

How to use this book

We have strived in creating this book to make the learning experience

engaging, entertaining, and enlightening. This book comes with a robot kit
and you will make extensive use of your personal robot as you go through the
book. As a part of your learning process, you will be bringing all of the
material in this book to life in the form of computer programs and robot
behaviors. You will create new and interesting behaviors for your robots. In
order to get the most out of this book, you will need the following:

vi
 • This book.
• A robot. This book uses the Scribbler robot.
• A Fluke Lancet dongle.
• A Bluetooth enabled computer. If your computer does not have built-
in Bluetooth, you can purchase a USB Bluetooth dongle to add that
functionality. You can use any of the following operating systems:
MACOS (Version 10.4 or later), Windows (XP or Vista), or Linux.
• Software: Python+Myro. This book is based on version 2.4 of Python.
• Some accessories: A basic gamepad controller, Sharpie permanent
color markers, a flashlight, spare AA batteries, poster or any other
large drawing paper, a carrying case, some craft materials for
decorating the robot, etc.

You can order the robot, the Fluke dongle, the Bluetooth dongle, as well as a
software CD, and the book from Georgia Robotics Inc. (georgiarobotics.com).
For more information on additional sources, documentation (including a copy
of this book), and latest updates visit the IPRE web site at roboteducation.org.

While our approach is tightly integrated with the use of a personal robot we
will also take frequent excursions into other forms of computing: numbers,
text, sounds, music, images, etc. The thing to remember is that all the
concepts of computing introduced in this book can be used to do computing
on all kinds of objects.

To get the most out of this book we would recommend reading each chapter
first and then bringing all the ideas discussed in the chapter to life by actually
creating computer programs and robot behaviors. Before proceeding to the
next chapter, try and do most of the exercises at the end of each chapter and
think about further applications that come to mind. Try out a few of those and
then proceed to the next chapter. We would also like to encourage you to
work in an environment where you can share your work and ideas with others
around you. After all, you are going to be creating several robot behaviors.
Find the opportunity to demonstrate your robot’s newly programmed
capabilities to your friends, classmates, and even the wider world (through
web sites like youtube.com). Make learning from this book a social activity
Preface

and you will find yourself learning more. The bottom-line keyword here is
doing. Computing is an abstract concept, but the only way to grasp it is by
doing it. Like that famous Nike commercial, Just do it!

What you will learn

By using this book as described above you will learn a number of

fundamental concepts in computing and robotics. We have written the text for
entry-level students into computing. Other than being able to use a computer
to do something, we do not expect our readers to have any additional
background in computing. Some of the material in the book does rely on basic
high school level mathematics. It is ideally suited for a college-level
introductory computing course which is widely recognized as CS1 in the
computing education community. By the time a reader works through the
entire book they will have covered the following topics:

• Fundamentals of computing: problem solving, program design, basic

syntax and semantics of the programming language Python; the role of
names; data values: numbers, strings, and boolean; data structures:
sequences, lists, tuples, dictionaries, objects; functions, parameters,
basic scoping rules, assignment, expressions, control (sequential,
conditional and iteration constructs), modules and libraries (APIs);
Simple input-output, file I/O, and other forms of I/O devices; testing
and error checking; interactive design.
• Program development process: using an IDE, interactive style
computing, debugging, and documentation.
• Basic robotics: robot sensing and control; sensor types, synchronous
and asynchronous operations; control paradigms: direct or reactive,
Braitenberg style, subsumption style, neural network; behavior design;
real-time/embedded control.
• Media computation: numbers, sound and music, image: generation,
processing, and understanding; gamepad controllers and game playing.
• Social implications of computing and robotics.
• Computation: history and essentials of algorithms; efficiency; limits of
computing; other issues and open problems in computing.

viii
For instructors, it should be clear that the above set of concepts provide
sufficient coverage of the CS1 topics recommended by the latest ACM/IEEE
Curricular guidelines1 as well as the guidelines developed for liberal arts
degrees in computer science2. In many respects the content of the book goes
well beyond the recommended CS1 guidelines. It does not explicitly mention
recursion or sorting algorithms. We have deliberately chosen to omit these
topics. Sorting of lists and other sequential structures is built-in in Python and
students will use these to do sorting. We often find that students discover
recursion in the course of doing exercises from the text. Rather than provide a
separate chapter or a section on recursion, we have left it as a topic to be
introduced by the instructor if she/he so chooses. We do use recursion is
example programs in Chapter 10. It would also be fairly trivial for instructors
to include class discussions of sorting algorithms if they feel so inclined.

Further, for instructors, the coverage and sequence of topics in this book will
vastly differ from that found in most traditional CS1 texts. We have
deliberately taken a contextual approach to introduce ideas. The nature of the
context, robots and computers, has driven the nature, ordering, and
sequencing of topics. We have found that this is a more engaging, exciting,
and enlightening framework. We hope that you will be able to share this
excitement while you are teaching a course based on this material. Further
details on how to use this book and additional materials is provided in the
Instructor’s manual accompanying this text. Much of this work is an outcome
of research conducted under the banner of IPRE specifically for developing a
fresh new approach to CS1. We invite you to refer to our project web site
(roboteducation.org) and read some of the papers and articles that elaborate
much more on the underlying research and its goals.

To students, we have taken efforts to make the material engaging and exciting
for you. We hope that we have succeeded somewhat in this. This is an open
source development project, so we hope that you will take the time to provide

1
Computing Curricula 2001, ACM Journal on Educational Resources in Computing (JERIC)
Vol. 1 , No. 3, Fall 2001.
2
A Model Curriculum for a Liberal Arts Degree in Computer Science, LACS, ACM JERIC, Vol.
7, No. 2, June 2007.
Preface

valuable feedback, as well as ideas and content for further improving this
book. We do hope that you will exploit all the fun learning opportunities
provided in this book.

x
 The World of Robots

I wouldn't ever want them to be brought back to Earth. We built them for
Mars, and Mars is where they should stay. But Spirit and Opportunity have
become more than just machines to me. The rovers are our surrogates, our
robotic precursors to a world that, as humans, we're still not quite ready to
visit.

-: Steve Squyres in Roving Mars, Hyperion, 2005.

Opposite page: Mars Rover.

Photo courtesy of NASA/JPL‐Caltech

1
Chapter 1

The picture on the opposite page is among one of thousands sent back by
Spirit and Opportunity from the surface of Mars. It goes without saying that it
will probably be several years, decades or even more, before a human ever
sets foot on Mars. The rovers Spirit and Opportunity landed on Mars in
January 2004 as robot geologists whose mission was to analyze rocks and
soils of the red planet in search for clues to past presence of water on the
planet. The two robots were expected to last about 90 days. Years later, they
were still exploring the planet's surface and sending invaluable geological and
pictorial data from the planet.

Back on Earth, the same month as the rovers landing on Mars, the
Tumbleweed robot rover traveled 40 miles across Antarctica’s polar plateau
transmitting local meteorological data back to its base station via satellite.
Besides surviving adverse conditions on mars and Antarctica, robots are
slowly becoming household consumer items. Take for instance, the Roomba
from iRobot Corporation. Introduced in 2002, several million Roombas have
been sold to vacuum and clean floors.

One commonality among the robots mentioned above is that they are all
designed for very specific tasks: analyze rocks and soils on the surface of
Mars, meteorology on the polar cap, or vacuuming a room. Yet, the core of
robot technology is almost as easy to use as computers. In this course you will
be given a personal robot of your own. Through this personal robot, you will
learn to give it instructions to do a variety of tasks. Like the robots mentioned
above your robot is also a rover. However, unlike the robots above, your
personal robot does not come pre-programmed to do any specific task. It has
certain basic capabilities (that you will learn about) and it can be programmed
to make use of its capabilities to do various tasks. We hope that the process of
learning about the capabilities of your robot and making it do different things
will be exciting and fun for you. In this chapter, we introduce you to the world
of robots and then introduce you to your own personal robot and some of its
capabilities.

2
 The World of Robots

The rim of Victoria Crater on Mars.

The rover Opportunity has been superimposed on the crater rim to show scale.
Photo courtesy of JPL/NASA/Cornell University, October, 2006.

What is a robot?

The Merriam-Webster Online Dictionary gives the following definitions of

the word robot:

1. a machine that looks like a human being and performs various complex acts
(as walking or talking) of a human being; also a similar but fictional machine
whose lack of capacity for human emotions is often emphasized; and also
an efficient insensitive person who functions automatically
2. a device that automatically performs complicated often repetitive tasks
3. a mechanism guided by automatic controls

In today's world, the first two definitions will probably be considered archaic
(the third interpretation in the first definition notwithstanding). Robots were
originally conceived as human-like entities, real or fictional, devoid of
emotions, that performed tasks that were repetitive or full or drudgery.
Today's robots come in all kinds of shapes and sizes and take on all kinds of
tasks (see below for some examples). While many robots are put to use for
repetitive or dull tasks (including the Roomba; unless you enjoy the
therapeutic side effects of vacuuming :-), robots today are capable of doing a
lot more than implied by the first two definitions above. Even in fictional
robots the lack of emotional capacity seems to have been overcome (see for
instance Steven Spielberg’s movie, Artificial Intelligence).
3
Chapter 1

For our purposes, the third definition

is more abstract and perhaps more
appropriate. A robot is a mechanism
or an artificial entity that can be
guided by automatic controls. The
last part of the definition, guided by Today, it is hard to imagine life
automatic controls, is what we will without a web search engine. While
focus on in this course. That is, given there are several search engines
a mechanism capable of such available, the one provided by Google
guidance, what is involved in
Inc. has become synonymous with
creating its controls?
web searching. So much so that
A Short History of Robots people use it as a common phrase:
“Google it!”
Modern robots were initially
You may have your own personal
conceived as industrial robots
preference for a search engine. Go
designed to assist in automated
manufacturing tasks. The first ahead use it and search for the items
commercial robot company, suggested here.
Unimation, was created nearly 50
years ago. As the use of robots in
industrial manufacturing grew, people also started experimenting with other
uses of robots. Earlier industrial robots were mainly large arms that were
attached to a fixed base. However, with the development of mobile robots
people started to find uses for them in other domains. For instance, in
exploring hazardous environments ranging from radioactive sites, volcanoes,
finding and destroying mines, military surveillance, etc. We started this
chapter by introducing you to two Mars rover robots. The first ever planetary
rover landed on Mars in 1997. Increasingly in the last decade or so robots
have ventured into newer and more exciting areas like medicine (Google:
robotic surgery, robot wheelchair, etc.), toys and entertainment (Google:
Pleo, SONY Aibo, LEGO Mindstorms, etc.), and even education (Google:
IPRE). Some of the most exciting developments in robotics are still in
research stages where, for example, in Artificial Intelligence research people
are trying to develop intelligent robots and also using robots to understand and

4
 The World of Robots

explore models of human intelligence. Here we have provided some pointers

(do the searches mentioned above) for examples of various robots and their
uses. There are numerous web sites where you can look up more about the
history of robots. We will leave that as an exercise.

Robots and Computers

In the last few decades computers

have become increasingly
ubiquitous. Most likely you are
reading this sentence on a
computer. If you’re reading this text
online, the text itself is coming to
you from another computer (located
somewhere on the western banks of
the Delaware River in south-eastern
parts of the state of Pennsylvania in
the United Stated of America). On
its journey from the computer in A Postage stamp titled World of
Pennsylvania to your computer, the Invention (The Internet) was issued by
text has probably traveled through UK’s Royal Mail on March 1, 2007
several computers (several dozen if honoring the development of the
you are outside the state of World Wide Web.
Pennsylvania!). What makes this
journey of this text almost
instantaneous is the presence of communication networks over which the
internet and the World Wide Web operate. Advances in the technologies of
wireless communication networks make it possible to access the internet from
nearly any place on the planet. The reason that you are sitting in front of a
computer and learning about robots is primarily because of the advent of these
technologies. While robots are not quite as ubiquitous as computers, they are
not too far behind. In fact, it is precisely the advances in computers and
communications technologies that have made it feasible for you to become
more familiar with the world of robots.

5
Chapter 1

The relationship between robots and computers is the basis for the use of the
phrase automatic controls in describing a robot. Automatically controlling a
robot almost always implies that there is a computer involved. So, in the
process of learning about and playing with robots you will also uncover the
world of computers. Your robot has a computer embedded in it. You will be
controlling the robot through your computer. Moreover, you will do this over
a wireless communication technology called bluetooth. Initially, for our
purposes, learning to automatically control a robot is going to be synonymous
with learning to control a computer. This will become more obvious as we
proceed through these lessons.

Automating control involves specifying, in advance, the set of tasks the robot
or the computer is to perform. This is called programming. Programming
involves the use of a programming language. Today, there are more
programming languages than human languages! Perhaps you have heard of
some of them: Java, C, Python, etc. In this course, we will do all our robot
programming in the programming language Python. Python, named after the
popular Monty Python TV shows, is a modern language that is very easy to
learn and use.

While we are talking about computers and languages, we should also mention
the Myro (for My robot) software system. Myro was developed by us to
simplify the programming of robots. Myro provides a small set of robot
commands that extend the Python language. This makes it easy, as you will
see, to specify automatic controls for robots.

A Robot of Your Own: The Scribbler

The scribbler robot, shown here is also a rover.

It can move about in its environment. The
wheels, and its other functions, can be
controlled through a computer via a wireless
interface. Your laboratory assistants will
provide you with a Scribbler and the required
The Scribbler Robot
components to enable wireless

6
 The World of Robots

communication. Once configured, you will be able to control the robot's

movements (and all other features) through the computer. Besides moving,
your robot can also play sounds (beeps) and, with the help of a pen inserted in
its pen port, it can draw a line wherever it goes (hence its name, Scribbler).
The robot can move forward, backward, turn, spin, or perform any
combination of these movements
giving it adequate functionality to
travel anywhere on the surface of an Dear Student:
environment. Besides roving, the
Every chapter in this book will include
Scribbler robot can also sense certain
features of its environment. For several hands‐on activities. These are
example, it is capable of sensing a short learning exercises designed
wall or an obstacle, or a line on the carefully to explore some of the
floor. We will discuss the Scribblers concepts presented in the chapter. It
sensing capabilities later. is important that you do all of the
activities in the chapter before moving
Do This on to the next chapter.

The first few activities show you We would also recommend trying out
how you to set up the computer and some (or all) of the exercises
the robot and will help you get suggested at the end to gain a better
familiarized with your Scribbler. understanding.
This will involve the following four
activities:

1. First things first: Setup Myro

2. Name your robot
3. Drive your robot around
4. Explore a little further

You may need the assistance of your instructor for the first activity to ensure
that you know how to set up and use your robot for the remainder of the text.

7
Chapter 1

1. First things first: Setup Myro

At the time you received your robot, its

software and hardware was configured for
use. The software we will be using for
controlling the robot is called, Myro (for
My Robot) which works in conjunction
with the Python language. In this, the first
exercise, we will start the robot and the
software and ensure that the software is
able to successfully communicate with the
robot through your computer. If Myro has The Fluke Dongle adds
not been installed on your computer, you
Bluetooth and other
should go ahead and obtain a copy of it (by
inserting the Myro CD into your computer capabilities to the Scribbler.
or following directions from the Myro
Installation Manual.

In a typical session, you will start

the Python software, connect to
the robot through the Myro
library, and then control the
robot through it. We have set up
the system so that all
communication between the
computer and the robot occurs
wirelessly over a Bluetooth
connection. Bluetooth
technology is a common wireless
communication technology that The Scribbler robot with the Fluke Dongle.
enables electronic devices to talk
to each other over short
distances. For example, Bluetooth is most commonly used in cell phones to
enable wireless communication between a cell phone (which may be in your
pocket) and your wireless headset. This kind of communication requires two

8
 The World of Robots

physical devices that serve as receivers and transmitters. In the Scribbler kit
you received, there is a pair of these Bluetooth devices: one connects into the
scribbler (Fluke Dongle) and the other into the USB port of your computer. If
your computer has a built-in Bluetooth capability, you may not need the one
that goes into your computer. Go ahead and make sure that these devices are
plugged in, the robot is turned on, and so is your computer. Open up the
Bluetooth configuration facility on your computer to get the port number
being used for communication. On Windows this will be something like
COM12 (the actual number will vary). Detailed instructions for setting up the
Bluetooth connection came with your robot. Make sure you have established
the connection and obtained the port number before proceeding.

2. Name your robot

In this exercise, we will connect to the robot and make it do something

simple, like make it beep. Then, we will give the robot a name to personalize
it. These tasks can be performed by using the following steps:

1. Start Python
2. Connect to the robot
3. Make the robot beep
4. Give the robot a name

Since this is your very first experience with using robots, we will provide
detailed instructions to accomplish the task outlined above.

1. Start Python: When you installed the software, a file called Start
Python.pyw was created. You should copy this file into a folder where you
plan to store all your robot programs. Once done, navigate to that folder and
open it. In it you will find the Start Python icon. Go ahead and double-click
on it. The following window should appear on your computer screen:

9
Chapter 1

What you see above is the Python interaction window or the Python Shell.
This particular shell is called IDLE (notice that it reports above that you are
using IDLE Version 1.1.4.). You will be entering all Python commands in
this IDLE window. The next step is to use Myro to coonnect to the robot.

2. Connect to the robot: Make sure your robot and the computer have their
Bluetooth dongles inserted and that your robot is turned on. To connect to the
robot enter the following command into the Python shell:

>>> from myro import *

This interaction is shown below (the Myro version number will be different):

10
 The World of Robots

That is, you have now informed the Python Shell that you will be using the
Myro library. The import statement/command is something you will use each
time you want to control the robot. After issuing the import, some useful
information is printed about Myro and then the Shell is ready for the next
Python command. Now it is time to connect to the robot by issuing the
following command:

>>> initialize("comX")

where X is the port number using which your computer is using to

communicate with the robot. If you need help figuring out the port number,
consult with your instructor. The example below shows how to issue the
command when the port com5 is being used:

11
Chapter 1

When you issue the initialize command, the computer attempts to

communicate with the robot. If this is successful, the robot responds with the
Hello... line shown above. As you can see, the robot's name is BluePig.
You can give your robot whatever name you like. We will do that later. First,
let us give it a command to make a beep so that we know that we are in
control of the robot:

3. Make the robot beep: In the Python Shell, enter the command:

>>> beep(1, 880)

The command above directs the robot to make a beeping sound at 880 Hertz
for 1 second. Go ahead and try it. Your robot will beep for 1 second at 880
Hz. Go ahead and try the following variations to hear different beeps:

beep(0.5, 880)
beep(0.5, 500)
etc.

12
 The World of Robots

So now, you should realize that you are in control of the robot. By issuing
simple commands like the ones above, you can make the robot perform
different behaviors. Now, we can learn the command to give the robot a new
name.

4. Give the robot a name: Suppose we wanted to name the robot Shrek. To
do this, all you have to do it give it the following command:

>>> setName("Shrek")

Whatever name you decide to give your robot, you can specify it in the
command above replacing the words Shrek. From now on, that will be the
name of the robot. How do we know this for a fact? Go ahead and try asking it
its name:

>>> getName()

It will also report than name each time you connect to it using the initialize
command:

>>> initialize("com5")
Waking robot from sleep...
Hello, I'm Shrek!
>>>

Congratulations! You have now completed the first Exercise and you are well
on your way to more fun and exciting things with your robot. Before we move
on, it would be a good idea to review what you just did. Each session with a
robot begins by starting the Python software (Step 1 above), followed by
importing the Myro library and initializing the robot. From then on, you can
issue any command to the robot.

The Myro library contains dozens of commands that enable various kinds of
robot behaviors. In the next few weeks we will be learning several robot
commands and learning how to use them to program complex robot
behaviors. One thing to remember at this juncture is that all commands are

13
Chapter 1

being issued in the Python language. Thus, as you learn more about your
robot and its behaviors, you will also be learning the Python language.

One characteristic of programming languages (like Python) is that they have a

very strict way of typing commands. That is, and you may already have
experienced this above, the language is very precise about what you type and
how you type it. Every parenthesis, quotation mark, and upper and lower case
letter that makes up a command has to be typed exactly as described. While
the rules are strict luckily there aren't too many of them. Soon you will get
comfortable with this syntax and it will become second nature. The precision
in syntax is required so that the computer can determine exactly one
interpretation for the command resulting in desired action. For this reason,
computer languages are often distinguished from human languages by
describing them as formal languages (as opposed to natural languages that
are used by humans).

3. Drive the robot around

In this exercise, we will introduce you to

a way of making the robot move about in
its environment manually controlled by a
game pad device (see picture on right).
As above, place the robot on an open
floor, turn the robot on, start Python as
above and connect to the robot. You may
already have this from Exercise 2 above.
Also, plug the game pad controller into
an available USB port of your computer.
At the prompt, enter the following
command: The game pad controller.

>>> gamepad()

In response to this command, you will get some help text printed in the IDLE
window showing what would happen if you pressed various game pad
buttons. If you look in the picture of the game pad controller above, you will
14
 The World of Robots

notice that it has eight (8) blue buttons (numbered 1 through 8 in the picture),
and an axis controller (the big blue swivel button on the left). The axis
controller can be used to move the robot around. Go ahead and try it. Pressing
each of the numbered buttons will result in different behaviors, some will
make the robot beep, some will make the computer speak or say things.
Button#1 will result in the robot taking a picture of whatever it is currently
seeing through its camera and display it on your computer screen. Button#8
will quit from the game pad control mode.

Spend some time experimenting with the various game pad control features.
See how well you can navigate the robot to go to various places, or follow a
wall, or go round and round something (like yourself!). You may also place
the robot on a big piece of paper, insert a Sharpie pen in its pen port and then
move it around to observe its scribbling. Can you scribble your name (or
initials)? Try a pattern or other shapes.

Without creating a program, this is an effective remote way of controlling the

movements of your robot. The next exercise asks you to try and issue
commands to the robot to move.

4. Explore a little further

OK, now you are on your own. Start Python, import Myro, connect to the
robot, and give commands to move forward, backward, turn left and right, and
spin. Use the commands: forward(SPEED), backward(SPEED),
turnLeft(SPEED), turnRight(SPEED), and rotate(SPEED). SPEED can be
any number between -1.0...1.0. These and all other robot commands are
detailed in the Myro Reference Manual. This would be a good time to review
the descriptions of all the commands introduced in this section.

Myro Review

from myro import *

This command imports all the robot commands available in the Myro library.
We will use this whenever we intend to write programs that use the robot.

15
Chapter 1

initialize(<PORT NAME>)
init(<PORT NAME>)
This command establishes a wireless communication connection with the
robot. <PORT NAME> is determined at the time you configured your software
during installation. It is typically the word com followed by a number. For
example, "com5". The double quotes (") are essential and required.

beep(<TIME>, <FREQUENCY>)
Makes the robot beep for <TIME> seconds at frequency specified by
<FREQUENCY>.

getName()
Returns the name of the robot.

setName(<NEW NAME>)
Sets the name of the robot to <NEW NAME>. The new name should be enclosed
in double quotes, no spaces, and not more than 16 characters long. For
example: setName("Bender").

gamepad()
Enables manual control of several robot functions and can be used to move
the robot around.

Python Review

Start Python.pyw
This is the icon you double-click on to start a Python Shell (IDLE).

>>>
The Python prompt. This is where you type in a Python command.

Note: All commands you type (including the Myro commands listed above)
are essentially Python commands. Later, in this section we will list those
commands that are a part of the Python language.

16
 The World of Robots

Exercises

1. Where does the word robot come from? Explore the etymology of the
words robot and robotics and write a short paper about it.
2. What are Asimov's Laws of robotics? Write a viewpoint essay on
them.
3. Look up the Wikipedia entry on robots, as well as the section in AI
Topics (see links above). Write a short essay on the current state of
robots.
4. Write a short essay on a robot (real or fictional) of your choice. Based
on what you have learned from your readings, evaluate its capabilities.
5. Spirit and Opportunity were not the
first rovers to land on Mars. On July 4,
1997, the Mars Pathfinder landed on
Mars with a payload that included the
Sojourner rover. The United States
Postal Service issued the stamp shown
here to commemorate the landing.
This is perhaps the first ever real robot Facsimile of the Mars
to appear on a postage stamp! Find out Pathfinder Postage Stamp
what you can about the Mars
Pathfinder mission and compare the Sojourner rover with Spirit and
Opportunity.
6. Through the exercises, you have experienced a subset of capabilities
of the Scribbler robot. Reflect/write about the physical capabilities of
the Scribbler and the kinds of tasks you could make it perform.
7. Insert a pen (provided in your kit) in the robot's pen port. Place the
robot on a surface where it is OK to write/draw. Drive the robot
around with the game pad controller. It will scribble on the paper as it
moves. Observe its scribbles by moving it forward and then
backwards. Does it trace its path exactly? Why or why not?
8. Using the game pad operation, make your robot Scribble your name on
the floor. You may find this difficult for several reasons. Try to make
the robot write your initials instead. Also, see if you can guide the
robot to draw a five point star. This task is in some sense not too

17
Chapter 1

different from controlling a robot to perform surgery. Research the

capabilities of today's surgical robots and write a short paper about it.
9. Using the game pad controller draw the Bluetooth logo (see picture)
using a pen inserted in the Scribbler robot. Do a web search for Harald
Blåtand and read more about the runic alphabets.

The Bluetooth logo is derived from runic alphabet

letters H and B juxtaposed together. HB for Harald
Harald Blåtand Gormson Blåtand a Scandinavian King (from the 10th century
What’s in a name? AD) who was legendary in uniting Denmark and
Norway. The wireless technology we use today is
named in his honor (Blåtand means “Bluetooth”)
because the technology itself was developed by
Ericsson, a Scandinavian company. The technology is
designed to unite computers and telecomm devices.
Bluetooth devices are most commonly found in cell
phones. We’re using it here to communicate
between your robot and the computer.

Further Reading

1. Wikipedia entry on Robots (http://en.wikipedia.org/wiki/Robot)

2. AI Topics: Robots from the American Association for Artificial
Intelligence (AAAI) (http://www.aaai.org/AITopics/html/robots.html)
3. Social Robots are robots that interact with and learn from people
around them. Here is an interview with Cynthia Breazeal who heads
the Robotic Life Group at MIT's Media Lab.
(http://www.pbs.org/saf/1510/features/breazeal.htm)
4. Visit the online Robot Hall of Fame and find out more about the real
and fictional robots that have been inducted into it.
(http://www.robothalloffame.org/)

18
The World of Robots

19
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Personal Robots

Every Pleo is autonomous. Yes, each one begins life as a newly-hatched baby
Camarasaurus, but that's where predictability ends and individuality begins.
Like any creature, Pleo feels hunger and fatigue - offset by powerful urges to
explore and be nurtured. He'll graze, nap and toddle about on his own -when
he feels like it! Pleo dinosaur can change his mind and his mood, just as you
do.

From: www.pleoworld.com

Opposite page: Pleo robots

Photo courtesy of UGOBE Inc.

21
Chapter 2

Most people associate the personal computer (aka the PC) revolution with the
1980’s but the idea of a personal computer has been around almost as long as
computers themselves. Today, on most college campuses, there are more
personal computers than people. The goal of One Laptop Per Child (OLPC)
Project is to “provide children around the world with new opportunities to
explore, experiment, and express themselves” (see www.laptop.org). Personal
robots, similarly, were conceived several decades ago. However, the personal
robot ‘revolution’ is still in its infancy. The picture on the previous page
shows the Pleo robots that are designed to emulate behaviors of an infant
Camarasaurus. The Pleos are marketed mainly as toys or as mechatronic
“pets”. Robots these days are being used in a variety of situations to perform a
diverse range of tasks: like mowing a lawn; vacuuming or scrubbing a floor;
entertainment; as companions for elders; etc. The range of applications for
robots today is limited only by our imagination! As an example, scientists in
Japan have developed a baby seal robot (shown on the opposite page) that is
being used for therapeutic purposes for nursing home patients.

Your Scribbler robot is your personal robot. In this case it is being used as an
educational robot to learn about robots and computing. As you have already
seen, your Scribbler is a rover, a robot that moves around. Such robots have
become more prevalent in the last few years and represent a new dimension of
robot applications. Roaming robots have been used for mail delivery in large
offices and as vacuum cleaners in homes. Robots vary in the ways in which
they move about: they can roll about like small vehicles (like the lawn mower,
Roomba, Scribbler, etc.), or even ambulate on two, three, or more legs (e.g.
Pleo). The Scribbler robot moves on three wheels, two of which are powered.
In this chapter, we will get to know the Scribbler in some more detail and also
learn about how to use its commands to control its behavior.

The Scribbler Robot: Movements

In the last chapter you were able to use the Scribbler robot through Myro to
carry out simple movements. You were able to start the Myro software,
connect to the robot, and then were able to make it beep, give it a name, and
move it around using a joystick. By inserting a pen in the pen port, the

22
 Personal Robots

scribbler is able to trace its path of movements

on a piece of paper placed on the ground. It
would be a good idea to review all of these
tasks to refresh your memory before
proceeding to look at some more details about
controlling the Scribbler.

If you hold the Scribbler in your hand and

take a look at it, you will notice that it has The Paro Baby Seal Robot.
three wheels. Two of its wheels (the big ones Photo courtesy of National
on either side) are powered by motors. Go Institute of Advanced
ahead turn the wheels and you will feel the Industrial Science and
resistance of the motors. The third wheel (in Technology, Japan (paro.jp).
the back) is a free wheel that is there for
support only. All the movements the Scribbler
performs are controlled through the two motor-driven wheels. In Myro, there
are several commands to control the movements of the robot. The command
that directly controls the two motors is the motors command:

motors(LEFT, RIGHT)

LEFT and RIGHT can be any value in the range [-1.0...1.0] and these values
control the left and right motors, respectively. Specifying a negative value
moves the motors/wheels backwards and positive values move it forward.
Thus, the command:

motors(1.0, 1.0)

will cause the robot to move forward at full speed, and the command:

motors(0.0, 1.0)

will cause the left motor to stop and the right motor to move forward at full
speed resulting in the robot turning left. Thus by giving a combination of left
and right motor values, you can control the robot's movements. Myro has also

23
Chapter 2

provided a set of often used movement commands that are easier to remember
and use. Some of them are listed below:

forward(SPEED)
backward(SPEED)
turnLeft(SPEED)
turnRight(SPEED)
stop()

Another version of these commands takes a second argument, an amount of

time in seconds:

forward(SPEED, SECONDS)
backward(SPEED, SECONDS)
turnLeft(SPEED, SECONDS)
turnRight(SPEED, SECONDS)

Providing a number for SECONDS in the commands above specifies how long
that command will be carried out. For example, if you wanted to make your
robot traverse a square path, you could issue the following sequence of
commands:

forward(1, 1)
turnLeft(1, .3)
forward(1, 1)
turnLeft(1, .3)
forward(1, 1)
turnLeft(1, .3)
forward(1, 1)
turnLeft(1, .3)

of course, whether you get a square or not will depend on how much the robot
turns in 0.3 seconds. There is no direct way to ask the robot to turn exactly 90
degrees, or to move a certain specified distance (say, 2 ½ feet). We will return
to this later.

You can also use the following movement commands to translate (i.e. move
forward or backward), or rotate (turn right or left):

24
 Personal Robots

translate(SPEED)
rotate(SPEED)

Additionally, you can specify, in a single command, the amount of translation

and rotation you wish use:

move(TRANSLATE_SPEED, ROTATE_SPEED)

In all of these commands, SPEED can be a value between [-1.0...1.0].

You can probably tell from the above list that there are a number of redundant
commands (i.e. several commands can be specified to result in the same
movement). This is by design. You can pick and choose the set of movement
commands that appear most convenient to you. It would be a good idea at this
point to try out these commands on your robot.

Do This: Start Myro, connect to the robot, and try out the following
movement commands on your Scribbler:

First make sure you have sufficient room in front of the robot (place it on the
floor with a few feet of open space in front of it).

>>> motors(1, 1)
>>> motors(0, 0)

Observe the behavior of robot. Specifically, notice if it does (or doesn't) move
in a straight line after issuing the first command. You can make the robot
carry out the same behavior by issuing the following commands:

>>> move(1.0, 0.0)

>>> stop()

Go ahead and try these. The behavior should be exactly the same. Next, try
making the robot go backwards using any of the following commands:

25
Chapter 2

motors(-1, -1)
move(-1, 0)
backward(1)

Again, notice the behavior closely. In rovers precise movement, like moving
in a straight line, is difficult to achieve. This is because two independent
motors control the robot's movements. In order to move the robot forward or
backward in a straight line, the two motors would have to issue the exact same
amount of power to both wheels. While this technically feasible, there are
several other factors than can contribute to a mismatch of wheel rotation. For
example, slight differences in the mounting of the wheels, different resistance
from the floor on either side, etc. This is not necessarily a bad or undesirable
thing in these kinds of robots.
Under similar circumstances even Do humans walk straight?
people are unable to move in a
precise straight line. To illustrate Find a long empty hallway and make
this point, you can try the sure you have a friend with you to help
experiment shown on right. with this. Stand in the center of the
hallway and mark your spot. Looking
For most people, the above straight ahead, walk about 10‐15 paces
experiment will result in a variable without looking at the floor. Stop,
movement. Unless you really
mark your spot and see if you walked
concentrate hard on walking in a
in a straight line.
straight line, you are most likely to
display similar variability as your Next, go back to the original starting
Scribbler. Walking in a straight
spot and do the same exercise with
line requires constant feedback
and adjustment, something your eyes closed. Make sure your
humans are quite adept at doing. friend is there to warn you in case you
This is hard for robots to do. are about to run into an object or a
Luckily, roving does not require wall. Again, note your spot and see if
such precise moments anyway. you walked in a straight line.

Do This: Review all of the other movement commands listed above and try
them out on your Scribbler. Again, note the behavior of the robot from each of
these commands. In doing this activity, you may find yourself repeatedly

26
 Personal Robots

entering the same commands (or IDLE Tip

simple variations). IDLE provides a
convenient way to repeat previous
You can repeat a previous command
commands (see the Tip in the box
by using IDLE's command history
on the right).
feature:
Defining New Commands ALT‐p retrieves previous command
ALT‐n retrieves next
Trying out simple commands
(Use CTRL‐p and CTRL‐n on MACs)
interactively in IDLE is a nice way
to get to know your robot's basic Pressing ALT‐p again will give the
features. We will continue to use
previous command from that one and
this each time we want to try out
so on. You can also move forward in
something new. However, making a
robot carry out more complex the command history by pressing ALT‐
behaviors requires several series of n repeatedly. You can also click your
commands. Having to type these cursor on any previous command and
over and over interactively while press ALT‐ENTER to repeat that
the robot is operating can get command.
tedious. Python provides a
convenient way to package a series
of commands into a brand new command called a function. For example, if
we wanted the Scribbler to move forward and then move backward (like a
yoyo), we can define a new command (function) called yoyo as follows:

>>> def yoyo():

forward(1)
backward(1)
stop()

The first line defines the name of the new command/function to be yoyo. The
lines that follow are slightly indented and contain the commands that make up
the yoyo behavior. That is, to act like a yoyo, move forward and then
backward and then stop. The indentation is important and is part of the Python

27
Chapter 2

syntax. It ensures that all indented commands are part of the definition of the
new command. We will have more to say about this later.

Once the new command has been defined, you can try it by entering the
command into IDLE as shown below:

>>> yoyo()

Do This: If you have your Scribbler ready, go ahead and try out the new
definition above by first connecting to the robot, and then entering the
definition above. You will notice that as soon as you type the first line, IDLE
automatically indents the next line(s). After entering the last line hit an extra
RETURN to end the definition. This defines the new command in Python.

Observe the robot's behavior when you give it the yoyo() command. You
may need to repeat the command several times. The robot momentarily moves
and then stops. If you look closely, you will notice that it does move forward
and backwards.

In Python, you can define new functions by using the def syntax as shown
above. Note also that defining a new function doesn't mean that the
commands that make up the function get carried out. You have to explicitly
issue the command to do this. This is useful because it gives you the ability to
use the function over and over again (as you did above). Issuing the new
function like this in Python is called, invocation. Upon invocation, all the
commands that make up the function's definition are executed in the sequence
in which they are listed in the definition.

28
 Personal Robots

How can we make the robot's yoyo And now for something
behavior more pronounced? That is, completely different
make it move forward for, say 1 second,
and then backwards for 1 second, and
then stop? You can use the SECONDS
option in forward and backward
movement commands as shown below:

>>> def yoyo():

forward(1, 1)
backward(1, 1)
stop()

The same behavior can also be

accomplished by using the command,
wait which is used as shown below:

wait(SECONDS)

where SECONDS specifies the amount of

DVD Cover, from
time the robot waits before moving on
http://Wikipedia.com
to the next command. In effect, the
robot continues to do whatever it had IDLE is the name of the editing and
been asked to do just prior to the wait
Python shell program. When you
command for the amount of time
specified in the wait command. That is, double‐click Start Python you are
if the robot was asked to move forward really starting up IDLE. Python is
and then asked to wait for 1 second, it the name of the language that we
will move forward for 1 second before will be using, and gets its name
applying the command that follows the from Monty Python's Flying Circus.
wait. Here is the complete definition of IDLE supposedly stands for
yoyo that uses the wait command: Interactive DeveLopment
Environment, but do you know to
whom else it might be homage?

29
Chapter 2

>>> def yoyo():

forward(1) Scribbler Tip:
wait(1)
backward(1)
Remember that your Scribbler runs on
wait(1)
stop() batteries and with time they will get
drained. When the batteries start to
Do This: Go ahead and try out the run low, the Scribbler may exhibit
new definitions exactly as above erratic movements. Eventually it stops
and issue the command to the responding. When the batteries start
scribbler. What do you observe? In
to run low, the Scribbler's red LED light
both cases you should see the robot
starts to blink. This is your signal to
move forward for 1 second
followed by a backward movement replace the batteries.
for 1 second and then stop.

Adding Parameters to Commands

Take a look at the definition of the yoyo function above and you will notice
the use of parentheses, (), both when defining the function as well as when
using it. You have also used other functions earlier with parentheses in them
and probably can guess their purpose. Commands or functions can specify
certain parameters (or values) by placing them within parentheses. For
example, all of the movement commands, with the exception of stop have
one or more numbers that you specify to indicate the speed of the movement.
The number of seconds you want the robot to wait can be specified as a
parameter in the invocation of the wait command. Similarly, you could have
chosen to specify the speed of the forward and backward movement in the
yoyo command, or the amount of time to wait. Below, we show three
definitions of the yoyo command that make use of parameters:

>>> def yoyo1(speed):

forward(speed, 1)
backward(speed, 1)

30
 Personal Robots

>>> def yoyo2(waitTime):

forward(1, waitTime)
backward(1, waitTime)

>>> def yoyo3(speed, waitTime):

forward(speed, waitTime)
backward, waitTime)

In the first definition, yoyo1, we specify the speed of the forward or backward
movement as a parameter. Using this definition, you can control the speed of
movement with each invocation. For example, if you wanted to move at half
speed, you can issue the command:

>>> yoyo1(0.5)

Similarly, in the definition of yoyo2 we have parameterized the wait time. In

the last case, we have parameterized both speed and wait time. For example, if
we wanted the robot to move at half speed and for 1 ½ seconds each time, we
would use the command:

>>> yoyo3(0.5, 1.5)

This way, we can customize individual commands with different values

resulting in different variations on the yoyo behavior. Notice in all o fthe
definitions above that we did not have to use the stop() command at all.
Why?

Saving New Commands in Modules

As you can imagine, while working with different behaviors for the robot, you
are likely to end up with a large collection of new functions. It would make
sense then that you do not have to type in the definitions over and over again.
Python enables you to define new functions and store them in files in a folder
on your computer. Each such file is called a module and can then be easily
used over and over again. Let us illustrate this by defining two behaviors: a
parameterized yoyo behavior and a wiggle behavior that makes the robot
wiggle left and right. The two definitions are given below:

31
Chapter 2

# File: moves.py
# Purpose: Two useful robot commands to try out as a module.

# First import myro and connect to the robot

from myro import *

init()

# Define the new functions...

def yoyo(speed, waitTime):

forward(speed)
wait(waitTime)
backward(speed)
wait(waitTime)
stop()

def wiggle(speed, waitTime):

rotate(-speed)
wait(waitTime)
rotate(speed)
wait(waitTime)
stop()

All lines beginning with a '#' sign are called comments. These are simply
annotations that help us understand and document the programs in Python.
You can place these comments anywhere, including right after a command.
The # sign clearly marks the beginning of the comment and anything
following it on that line is not interpreted as a command by the computer.
This is quite useful and we will make liberal use of comments in all our
programs.

Notice that we have added the import and the init commands at the top. The
init command will always prompt you to enter the com-port number.

Do This: To store the yoyo and wiggle behaviors as a module in a file, you
can ask IDLE for a New Window from the File menu. Next enter the text
containing the two definitions and then save them in a file (let’s call it
moves.py) in your Myro folder (same place you have the Start Python

32
 Personal Robots

icon). All Python modules end with the filename extension .py and you
should make sure they are always saved in the same folder as the Start
Python.pyw file. This will make it easy for you as well as IDLE to locate
your modules when you use them.

Once you have created the file, there are two ways you can use it. In IDLE,
just enter the command:

>>> from moves import *

and then try out any of the two commands. For example, the following shows
how to use the yoyo function after importing the moves module:

As you can see from above, accessing the commands defined in a module is
similar to accessing the capabilities of the myro module. This is a nice feature
of Python. In Python, you are encouraged to extend the capabilities of any
system by defining your own functions, storing them in modules and then
using them by importing them. Thus importing from the moves module is no
different that importing from the myro module. In general, the Python import

33
Chapter 2

command has two features that it specifies: the module name; and what is
being imported from it. The precise syntax is described below:

from <MODULE NAME> import <SOMETHING>

where <MODULE NAME> is the name of the module you are importing from, and
<SOMETHING> specifies the commands/capabilities you are importing. By
specifying a * for <SOMETHING> you are importing everything defined in the
module. We will return to this a little later in the course. But at the moment,
realize that by saying:

from myro import *

you are importing everything defined in the myro module. Everything defined
in this module is listed and documented in the Myro Reference Manual. The
nice thing that this facility provides is that you can now define your own set
of commands that extend the basic commands available in Myro to customize
the behavior of your robot. We will be making use of this over and over again
in this course.

Functions as Building Blocks

Now that you have learned how to define new commands using existing ones,
it is time to discuss a little more Python. The basic syntax for defining a
Python function takes the form:

def <FUNCTION NAME>(<PARAMETERS>):

<SOMETHING>
...
<SOMETHING>

That is, to define a new function, start by using the word def followed by the
name of the function (<FUCTION NAME>) followed by <PARAMETERS> enclosed
in parenthesis followed by a colon (:). This line is followed by the commands
that make up the function definition (<SOMETHING>...<SOMETHING>). Each
command is to be placed on a separate line, and all lines that make up the

34
 Personal Robots

definition should be indented (aligned) the same amount. The number of

spaces that make up the indentation is not that important as long as they are
all the same. This may seem a bit awkward and too restricting at first, but you
will soon see the value of it. First, it makes the definition(s) more readable.
For example, look at the following definitions for the yoyo function:

def yoyo(speed, waitTime):

forward(speed)
wait(waitTime)
backward(speed)
wait(waitTime)
stop()

def yoyo(speed, waitTime):

forward(speed); wait(waitTime)
backward(speed); wait(waitTime)
stop()

The first definition will not be accepted by Python, as shown below:

It reports that there is a syntax error and it highlights the error location by
placing the thick red cursor (see the third line of the definition). This is
because Python strictly enforces the indentation rule described above. The
second definition, however, is acceptable. For two reasons: indentation is
consistent; and commands on the same line can be entered separated by a
semi-colon (;). We would recommend that you continue to enter each
command on a separate line and defer from using the semi-colon as a
separator until you are more comfortable with Python. More importantly, you

35
Chapter 2

will notice that IDLE helps you in making your indentations consistent by
automatically indenting the next line, if needed.

Another feature built into IDLE that enables readability of Python programs is
the use of color highlighting. Notice in the above examples (where we use
screen shots from IDLE) that pieces of your program appear in different
colors. For example, the word def in a function definition appears in red, the
name of your function, yoyo appears in blue. Other colors are also used in
different situations, look out for them. IDLE displays all Python words (like
def) in red and all names defined by you (like yoyo) in blue.

The idea of defining new functions by using existing functions is very

powerful and central to computing. By defining the function yoyo as a new
function using the existing functions (forward, backward, wait, stop))
you have abstracted a new behavior for your robot. You can define further
higher-level functions that use yoyo if you want. Thus, functions serve as
basic building blocks in defining various robot behaviors, much like the idea
of using building blocks to build bigger structures. As an example, consider
defining a new behavior for your robot: one that makes it behave like a yoyo
twice, followed by wiggling twice. You can do this by defining a new
function as follows:

>>> def dance():

yoyo(0.5, 0.5)
yoyo(0.5, 0.5)
wiggle(0.5, 1)
wiggle(0.5, 1)

>>> dance()

Do This: Go ahead and add the dance function to your moves.py module.
Try the dance command on the robot. Now you have a very simple behavior
that makes the robot do a little shuffle dance.

36
 Personal Robots

Guided by Automated Controls

Earlier we agreed that a robot is a “mechanism guided by automated

controls”. You can see that by defining functions that carry out more complex
movements, you can create modules for many different kinds of behaviors.
The modules make up the programs you write, and when they are invoked on
the robot, the robot carries out the specified behavior. This is the beginning of
being able to define automated controls for a robot. As you learn more about
the robot’s capabilities and how to access them via functions, you can design
and define many kinds of automated behaviors.

Summary

In this chapter, you have learned several commands that make a robot move in
different ways. You also learned how to define new commands by defining
new Python functions. Functions serve as basic building blocks in computing
and defining new and more complex robot behaviors. Python has specific
syntax rules for writing definitions. You also learned how to save all your
function definitions in a file and then using them as a module by importing
from it. While you have learned some very simple robot commands, you have
also learned some important concepts in computing that enable the building of
more complex behaviors. While the concepts themselves are simple enough,
they represent a very powerful and fundamental mechanism employed in
almost all software development. In later chapters, we will provide more
details about writing functions and also how to structure parameters that
customize individual function invocations. Make sure you do some or all of
the exercises in this chapter to review these concepts.

37
Chapter 2

Myro Review

backward(SPEED)
Move backwards at SPEED (value in the range -1.0…1.0).

backward(SPEED,SECONDS)
Move backwards at SPEED (value in the range -1.0…1.0) for a time given in
SECONDS, then stop.

forward(SPEED)
Move forward at SPEED (value in the range -1.0..1.0).

forward(SPEED,TIME)
Move forward at SPEED (value in the range -1.0…1.0) for a time given in
seconds, then stop.

motors(LEFT,RIGHT)
Turn the left motor at LEFT speed and right motor at RIGHT speed (value in the
range -1.0…1.0).

move(TRANSLATE, ROTATE)
Move at the TRANSLATE and ROTATE speeds (value in the range -1.0…1.0).

rotate(SPEED)
Rotates at SPEED (value in the range -1.0…1.0). Negative values rotate right
(clockwise) and positive values rotate left (counter-clockwise).

stop()
Stops the robot.

translate(SPEED)
Move in a straight line at SPEED (value in the range -1.0…1.0). Negative
values specify backward movement and positive values specify forward
movement.

turnLeft(SPEED)
Turn left at SPEED (value in the range -1.0…1.0)
38
 Personal Robots

turnLeft(SPEED,SECONDS)
Turn left at SPEED (value in the range -1.0..1.0) for a time given in seconds,
then stops.

turnRight(SPEED)
Turn right at SPEED (value in the range -1.0..1.0)

turnRight(SPEED,SECONDS)
Turn right at SPEED (value in the range -1.0..1.0) for a time given in seconds,
then stops.

wait(TIME)
Pause for the given amount of TIME seconds. TIME can be a decimal number.

Python Review
def <FUNCTION NAME>(<PARAMETERS>):
<SOMETHING>
...
<SOMETHING>
Defines a new function named <FUNCTION NAME>. A function name should
always begin with a letter and can be followed by any sequence of letters,
numbers, or underscores (_), and not contain any spaces. Try to choose names
that appropriately describe the function being defined.

Exercises

1. Compare the robot's movements in the commands turnLeft(1),

turnRight(1) and rotate(1) and rotate(-1). Closely observe the robot's
behavior and then also try the motor commands:

>>> motors(-0.5, 0.5)

>>> motors(0.5, -0.5)
>>> motors(0, 0.5)
>>> motors(0.5, 0)

39
Chapter 2

Do you notice any difference in the turning behaviors? The rotate

commands make the robot turn with a radius equivalent to the width of the
robot (distance between the two left and right wheels). The turn command
causes the robot to spin in the same place.

2. Insert a pen in the scribbler's pen port and then issue it command to go
forward for 1 or more seconds and then backwards for the same amount. Does
the robot travel the same distance? Does it traverse the same trajectory?
Record your observations.

3. Measure the length of the line drawn by the robot in Exercise 2. Write a
function travel(DISTANCE) to make the robot travel the given DISTANCE.
You may use inches or centimeters as your units. Test the function on the
robot a few times to see how accurate the line is.

4. Suppose you wanted to turn/spin your robot a given amount, say 90

degrees. Before you try this on your robot, do it yourself. That is, stand in one
spot, draw a line dividing your two feet, and then turn 90 degrees. If you have
no way of measuring, your turns will only be approximate. You can study the
behavior of your robot similarly by issuing it turn/spin commands and making
them wait a certain amount. Try and estimate the wait time required to turn 90
degrees (you will have to fix the speed) and write a function to turn that
amount. Using this function, write a behavior for your robot to transcribe a
square on the floor (you can insert a pen to see how the square turns out).

5. Generalize the wait time obtained in Exercise 3 and write a function called
degreeTurn(DEGREES). Each time it is called, it will make the robot turn the
specified degrees. Use this function to write a set of instructions to draw a
square.

6. Using the functions travel and degreeTurn, write a function to draw the
Bluetooth logo (See Chapter 1, Exercise 9).

7. Choreograph a simple dance routine for your robot and define functions to
carry it out. Make sure you divide the tasks into re-usable moves and as much

40
 Personal Robots

as possible parameterize the moves so they can be used in customized ways in

different steps. Use the building block idea to build more and more complex
series of dance moves. Make sure the routine lasts for at least several seconds
and it includes at least two repetitions of the entire sequence. You may also
make use of the beep command you learned from the last section to
incorporate some sounds in your choreography.

8. Record a video of your robot dance and then dub it with a soundtrack of
your choosing. Use whatever video editing software accessible to you. Post
the video online on sites like YouTube to share with friends.

9. Lawn mower robots and even vacuuming robots can use specific
choreographed movements to ensure that they provide full coverage of the
area to be serviced. Assuming that the area to be mowed or cleaned is
rectangular and without any obstructions, can you design a behavior for your
Scribbler to provide full coverage of the area? Describe it in writing. [Hint:
Think about how you would mow/vacuum yourself.]

41
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Building Robot Brains

What a splendid head, yet no brain.

Aesop (620 BC-560 BC)

Opposite page: Home Simpson’s Brain

Photo courtesy of The Simpson’s Trivia (www.simpsonstrivia.com)

43
Chapter 3

If you think of your robot as a creature that acts in the world, then by
programming it, you are essentially building the creature's brain. The power
of computers lies in the fact that the same computer or the robot can be
supplied a different program or brain to make it behave like a different
creature. For example, a program like Firefox or Explorer makes your
computer behave like a web browser. But switching to your Media Player, the
computer behaves as a DVD or a CD player. Similarly, your robot will behave
differently depending upon the instructions in the program that you have
requested to run on it. In this chapter we will learn about the structure of
Python programs and how you can organize different robot behaviors as
programs.

The world of robots and computers, as you have seen so far is intricately
connected. You have been using a computer to connect to your robot and then
controlling it by giving it commands. Most of the commands you have used so
far come from the Myro library which is specially written for easily
controlling robots. The programming language we are using to do the robot
programming is Python. Python is a general purpose programming language.
By that we mean that one can use Python to write software to control the
computer or another device like a robot through that computer. Thus, by
learning to write robot programs you are also learning how to program
computers. Our journey into the world of robots is therefore intricately tied up
with the world of computers and computing. We will continue to interweave
concepts related to robots and computers throughout this journey. In this
chapter, we will learn more about robot and computer programs and their
structure.

Basic Structure of a Robot Brain

The basic structure of a Python program (or a robot brain) is shown below:

def main():
<do something>
<do something>
...

44
 Building Robot Brains

This is essentially the same as defining a new function. In fact, here, we are
adopting a convention that all our programs that represent robot brains will be
called main. In general, the structure of your robot programs will be as shown
below (we have provided line numbers so we can refer to them):

Line 1: from myro import *

Line 2: init()

Line 3: <any other imports>

Line 4: <function definitions>
Line 5: def main():
Line 6: <do something>
Line 7: <do something>
Line 8: ...

Line 9: main()

Every robot brain program will begin with the first two lines (Line 1 and
Line 2). These, as you have already seen, import the Myro library and
establish a connection with the robot. In case you are using any other libraries,
you will then import them (this is shown in Line 3). This is followed by the
definitions of functions (Line 4), and then the definition of the function,
main. Finally, the last line (Line 9) is an invocation of the function main.
This is placed so that when you load this program into the Python Shell the
program will start executing. In order to illustrate this, let us write a robot
program that makes it do a short dance using the yoyo and wiggle
movements defined in the last chapter.

# File: dance.py
# Purpose: A simple dance routine

# First import myro and connect to the robot

from myro import *

initialize("com5")

# Define the new functions...

def yoyo(speed, waitTime):

45
Chapter 3

forward(speed, waitTime)
backward(speed, waitTime)

def wiggle(speed, waitTime):

motors(-speed, speed)
wait(waitTime)
motors(speed, -speed)
wait(waitTime)
stop()

# The main dance program

def main():
print "Running the dance routine..."
yoyo(0.5, 0.5)
wiggle(0.5, 0.5)
yoyo(1, 1)
wiggle(1, 1)
print "...Done"

main()

We have used a new Python command in the definition of the main function:
the print command. This command will print out the text enclosed in double
quotes (") when you run the program. This program is not much different
from the dance function defined in the previous chapter except we are using a
spin motion to wiggle. However, instead of naming the function dance we are
calling it main. As we mentioned earlier, this is just a naming convention that
we are adopting that makes it easy to identify the main program in a program
file.

Do This: In order to run this program on the robot, you can start IDLE, create
a new window, enter the program in it, save it as a file (dance.py) and then
select the Run Module feature in the window's Run menu. Alternately, to run
this program, you can enter the following command in the Python Shell:

>>> from dance import *

This is essentially equivalent to the Run Module option described above.

When you run the program you will notice that the robot carries out the dance

46
 Building Robot Brains

routine specified in the main program. Also notice the two messages printed
in the IDLE window. These are the results of the print command. print is a
very useful command in Python and can be used to output essentially anything
you ask it to. While you are in this session, go ahead and change the print
command to the following:

speak("Running the dance routine")

speak is a Myro command that enables speech output from your computer.
Go ahead and change the other print command also to the speak command
and try your program. Once done, enter some other speak commands on the
IDLE prompt. For example:

speak("Dude! Pardon me, would you have any Grey Poupon?")

The speech facility is built into most computers these days. Later we will see
how you can find out what other voices are available and also how to change
to them.

Speaking Pythonese

We have launched you into the world of computers and robots without really
giving you a formal introduction to the Python language. In this section, we
provide more details about the language. What you know about Python so far
is that it is needed to control the robot. The robot commands you type are
integrated into Python by way of the Myro library. Python comes with several
other useful libraries or modules that we will try and learn in this course. If
you need to access the commands provided by a library, all you have to do is
import them.

The libraries themselves are largely made up of sets of functions (they can
contain other entities but more on that later). Functions provide the basic
building blocks for any program. Typically, a programming language (and
Python is no exception) includes a set of pre-defined functions and a
mechanism for defining additional functions. In the case of Python, it is the

47
Chapter 3

def construct. You have already seen several examples of function definitions
and indeed have written some of your own by now. In the def construct, when
defining a new function, you have to give the new function a name. Names
are a critical component of programming and Python has rules about what
forms a name.

What’s in a name?

A name in Python must begin with either an alphabetic letter (a-z or A-Z) or
the underscore (i.e. _) and can be followed by any sequence of letters, digits,
or underscore letters. For example,

iRobot
myRobot
jitterBug
jitterBug2
my2cents
my_2_cents

are all examples of valid Python names. Additionally, another important part
of the syntax of names is that Python is case sensitive. That is the names
myRobot and MyRobot and myrobot are distinct names as far as Python is
concerned. Once you name something a particular way, you have to
consistently use that exact case and spelling from then on. Well, so much
about the syntax of names, the bigger question you may be asking is what
kinds of things can (or should) be named?'

So far, you have seen that names can be used to represent functions. That is,
what a robot does each time you use a function name (like yoyo) is specified
in the definition of that function. Thus, by giving functions a name you have a
way of defining new functions. Names can also be used to represent other
things in a program. For instance, you may want to represent a quantity, like
speed or time by a name. In fact, you did so in defining the function yoyo
which is also shown below:

48
 Building Robot Brains

def yoyo(speed, waitTime):

forward(speed, waitTime)
backward(speed, waitTime)

Functions can take parameters that help customize what they do. In the above
example, you can issue the following two commands:

>>> yoyo(0.8, 2.5)

>>> yoyo(0.3, 1.5)

The first command is asking to perform the yoyo behavior at speed 0.8 for 2.5
seconds where as the second one is specifying 0.3 and 1.5 for speed and time,
respectively. Thus, by parameterizing the function with those two values, you
are able to produce similar but varying outcomes. This idea is similar to the
idea of mathematical functions: sine(x) for example, computes the sine of
whatever value you supply for x. However, there has to be a way of defining
the function in the first place that makes it independent of specific parameter
values. That is where names come in. In the definition of the function yoyo
you have named two parameters (the order you list them is important): speed
and waitTime. Then you have used those names to specify the behavior that
makes up that function. That is the commands forward, and backward use the
names speed and waitTime to specify whatever the speed and wait times are
included in the function invocation. Thus, the names speed and waitTime
represent or designate specific values in this Python program.

Names in Python can represent functions as well as values. What names you
use is entirely up to you. It is a good idea to pick names that are easy to read,
type, and also appropriately designate the entity they represent. What name
you pick to designate a function or value in your program is very important,
for you. For example, it would make sense if you named a function
turnRight so that when invoked, the robot turned right. It would not make
any sense if the robot actually turned left instead, or worse yet, did the
equivalent of the yoyo dance. But maintaining this kind of semantic
consistency is entirely up to you.

49
Chapter 3

Values

In the last section we saw that names can designate functions as well as
values. While the importance of naming functions may be obvious to you by
now, designating values by names is an even more important feature of
programming. By naming values, we can create names that represent specific
values, like the speed of a robot, or the average high temperature in the month
of December on top of the Materhorn in Switzerland, or the current value of
the Dow Jones Stock Index, or the name of your robot, etc. Names that
designate values are also called variables. Python provides a simple
mechanism for designating values with names:

speed = 0.75
aveHighTemp = 37
DowIndex = 12548.30
myFavoriteRobot = "C3PO"

Values can be numbers or strings (anything enclosed in double-quotes, ").

The above are examples of assignment statements in Python. The exact syntax
of an assignment statement is given below:

<variable name> = <expression>

You should read the above statement as: Let the variable named by
<variable name> be assigned the value that is the result of calculating the
expression <expression>. So what is an <expression>? Here are some
examples:

>>> 5
5
>>> 5 + 3
8
>>> 3 * 4
12
>>> 3.2 + 4.7
7.9
>>> 10 / 2
5

50
 Building Robot Brains

What you type at the Python prompt (>>>) is actually called an expression.
The simplest expression you can type is a number (as shown above). A
number evaluates to itself. That is, a 5 is a 5, as it should be! And 5 + 3 is 8.
As you can see when you enter an expression, Python evaluates it and then
outputs the result. Also, addition (+), subtraction (-), multiplication (*), and
division (/) can be used on numbers to form expressions that involve
numbers.

You may have also noticed that numbers can be written as whole numbers (3,
5, 10, 1655673, etc) or with decimal points (3.2, 0.5, etc) in them. Python (and
most computer languages) distinguishes between them. Whole numbers are
called integers and those with decimal points in them are called floating point
numbers. While the arithmetic operations are defined on both kinds of
numbers, there are some differences you should be aware of. Look at the
examples below:

>>> 10.0/3.0
3.3333333333333335
>>> 10/3
3
>>> 1/2
0
>>> 1.0/2
0.5

When you divide a floating point number by another floating point number,
you get a floating point result. However, when you divide an integer by
another integer, you get an integer result. Thus, in the examples above, you
get the result 3.3333333333333335 when you divide 10.0 by 3.0, but you get
3 when you divide 10 by 3. Knowing this, the result of dividing 1 by 2 (see
above) is zero (0) should not be surprising. That is, while the division
operation looks the same (/), it treats integers differently than floating point
values. However, if at least one of the numbers in an arithmetic operation is a
floating point number, Python will give you a floating point result (see last
example above). You should keep this in mind. More on numbers later, before
we get back to robots, let us quickly introduce you to strings.

51
Chapter 3

Computers came to be called so because they excelled in doing calculations.

However, these days, computers are capable of manipulating any kind of
entity: text, images, sounds, etc. Text is made of letters or characters and
strings are simply sequences of characters. Python requires that strings be
written enclosed in quotes: which could be single ('I am a string'), double
("Me too!"), or even triple quotes ('''I'm a string as well!''').
Treating a string as a value is a powerful feature of Python. Python also
provides some operations on strings using which you can write some useful
string expressions. Here are some examples:

>>> mySchool = "Bryn Mawr College"

>>> yourSchool = "Georgia Institute of Technology"
>>> print mySchool
Bryn Mawr College

>>> print yourSchool

Georgia Institute of Technology

>>> print mySchool, yourSchool

Bryn Mawr College Georgia Institute of Technology

>>> yourSchool+mySchool
'Georgia Institute of TechnologyBryn Mawr College'

>>> print yourSchool+mySchool

Georgia Institute of TechnologyBryn Mawr College

Pay special attention to the last two examples. The operation + is defined on
strings and it results in concatenating the two strings. The print command is
followed by zero or more Python expressions, separated by commas. print
evaluates all the expressions and prints out the results on the screen. As you
have also seen before, this is a convenient way to print out results or messages
from your program.

52
 Building Robot Brains

A Calculating Program

Ok, set your robot aside for just a few more minutes. You have now also
learned enough Python to write programs that perform simple, yet interesting,
calculations. Here is a simple problem:

On January 1, 2008 the population of the world was estimated at

approximately 6.650 billion people. It is predicted that at current rates of
population growth, we will have over 9 billion people by the year 2050. A
gross estimate of population growth puts the annual increase at +1.14% (it
has been as high as +2.2% in the past). Given this data, can you estimate by
how much the world’s population will increase in this year (2008)? Also, by
how much will it increase each day?

In order to answer the questions, all you have to do is compute 1.14% of

6.650 billion to get the increase in population this year. If you divide that
number by 366 (the number of days in 2008) you will get average daily
increase. You can just use a calculator to do these simple calculations. You
can also use Python to do this in two ways. You can use it as a calculator as
shown below:

>>> 6650000000*1.14/100.0
75810000.0

>>> 75810000.0/365.0
207131.1475409836

That is, in this year there will be an increase of 75.81 million in the world’s
population which implies an average daily increase of over 207 thousand
people). So now you know the answer!

Also, let us try and write a program to do the above calculations. A program
to do the calculation is obviously going to be a bit of overkill. Why do all the
extra work when we already know the answer? Small steps are needed to get
to higher places. So let’s indulge and see how you would write a Python
program to do this. Below, we give you one version:

53
Chapter 3

#File: worldPop.py
# Purpose:
# Estimate the world population growth in a year and
# also per day.
# Given that on January 1, 2008 the world's population was
# estimated at 6,650,000,000 and the estimated growth is
# at the rate of +1.14%

def main():
population = 6650000000
growthRate = 1.14/100.0

growthInOneYear = population * growthRate

growthInADay = growthInOneYear / 365

print "World population on January 1, 2008 is", population

print "By Jan. 1, 2009, it will grow by", growthInOneYear
print "An average daily increase of", growthInADay

main()

The program follows the same structure and conventions we discussed above.
In this program, we are not using any libraries (we do not need any). We have
defined variables with names population, and growthRate to designate the
values given in the problem description. We also defined the variables
grothInOneYear and growthInADay and use them to designate the results
of the calculations. First, in the main program we assign the values given,
followed by performing the calculation. Finally, we use the print commands
to print out the result of the computations.

Do This: Start Python, enter the program, and run it (just as you would run
your robot programs) and observe the results. Voila! You are now well on
your way to also learning the basic techniques in computing! In this simple
program, we did not import anything, nor did we feel the need to define any
functions. But this was a trivial program. However, it should serve to
convince you that writing programs to do computation is essentially the same
as controlling a robot.

54
 Building Robot Brains

Using Input The Energy Problem

The program we wrote above uses specific

values of the world’s population and the rate
of growth. Thus, this program solves only
one specific problem for the given values.
What if we wanted to calculate the results for
a different growth rate? Or even a different
estimate of the population? What if we
wanted to try out the program for varying
quantities of both? Such a program would be
much more useful and could be used over
and over again. Notice that the program
begins by assigning specific values to the The root cause of world
two variables: energy problems is growing
world population and energy
population = 6650000000 consumption per capita.
growthRate = 1.14/100.0
How many people can the
One thing you could do is simply modify earth support? Most experts
those two lines to reflect the different values. estimate the limit for long‐
However, typical programs are much more term sustainability to be
complicated than this one and it may require between 4 and 16 billion.
a number of different values for solving a
From: Science, Policy & The Pursuit
problem. When programs get larger, it is not
of Sustainability, Edited by Bent,
a good idea to modify them for every specific
Orr,and Baker. Illus. by Shetter.
problem instance but it is desirable to make Island Press, 2002.
them more useful for all problem instances.
One way you can achieve this is by using the
input facilities of Python. All computer programs typically take some input,
do some computation (or something), and then produce some output. Python
has a simple input command that can be used to rewrite the program above as
follows:

55
Chapter 3

#File: worldPop.py
# Purpose:
# Estimate the world population growth in a year and
# also per day.
# Given that on Jnauray 1, 2008 the world's population was
# estimated at 6,650,000,000 and the estimated growth is
# at the rate of +1.14%

def main():
# print out the preamble

print "This program computes population growth figures."

# Input the values

population = input("Enter current world population: ")
growthRate = input("Enter the growth rate: ")/100.0

# Compute the results

growthInOneYear = population * growthRate
growthInADay = growthInOneYear / 365

# output results
print "World population today is", population
print "In one year, it will grow by", growthInOneYear
print "An average daily increase of", growthInADay

main()

Read the program above carefully. Notice that we have added additional
comments as well as print statements. This improves the overall readability as
well as the interaction of this program. Notice the use of the input statements
above. The basic syntax of input is shown below:

<variable name> = input(<some prompt string>)

That is, the input is a function whose parameter is a string and the value it
returns is the value of the expression that will be entered by the user. When
executed, the computer will print out the prompt and wait for the user to enter
a Python expression. The user can enter whatever expression in response to

56
 Building Robot Brains

the prompt and then hit the RETURN or ENTER key. The expression is then
evaluated by Python and the resulting value is returned by the input function.
That value is then assigned to the variable <variable name>. The statement
above uses the same syntax as the assignment statement described above.
Python has made obtaining input from a user easy by defining input as a
function. Now, look at the use of the input function in the program above.
With this modification, we now have a more general program which can be
run again and again. Below, we show two sample runs:

Notice how you can re-run the program by just typing the name of the
main()function. There are other ways of obtaining input in Python. We will
see those a little later.

Robot Brains

Writing programs to control your robot is therefore no different from writing a

program to perform a computation. They both follow the same basic structure.
The only difference is that all robot programs you will write will make use of
the Myro library. There will be several robot programs that will require you to
obtain input from the user (see exercises below). You can then make use of
the input function as described above.

One characteristic that will distinguish robot programs from those that just do
computations is in the amount of time it will take to run a program. Typically,
a program that only performs some computation will terminate as soon as the

57
Chapter 3

computation is completed. However, it will be the case that most of the time
your robot program will require the robot to perform some task over and over
again. Here then, is an interesting question to ask:

Question How much time would it take for a vacuuming robot to vacuum a
16ft X 12ft room?

Seemingly trivial question but if you think about it a little more, you may
reveal some deeper issues. If the room does not have any obstacles in it (i.e.
an empty room), the robot may plan to vacuum the room by starting from one
corner and then going along the entire length of the long wall, then turning
around slightly away from the wall, and traveling to the other end. In this
manner, it will ultimately reach the other side of the room in a systematic way
and then it could stop when it reaches the last corner. This is similar to the
way one would mow a flat oblong lawn, or even harvest a field of crop, or re-
ice an ice hockey rink using a Zamboni machine. To answer the question
posed above all you have to do is calculate the total distance travelled and the
average speed of the vacuum robot and use the two to compute the estimated
time it would take. However, what if the room has furniture and other objects
in it?

You might try and modify the approach for vacuuming outlined above but
then there would be no guarantee that the floor would be completely
vacuumed. You might be tempted to redesign the vacuuming strategy to allow
for random movements and then estimate (based on average speed of the
robot) that after some generous amount of time, you can be assured that the
room would be completely cleaned. It is well known (and we will see this
more formally in a later chapter) that random movements over a long period
of time do end up providing uniform and almost complete coverage.
Inherently this also implies that the same spot may very well end up being
vacuumed several times (which is not necessarily a bad thing!). This is similar
to the thinking that a herd of sheep, if left grazing on a hill, will result, after a
period of time, in a nearly uniform grass height (think of the beautiful hills in
Wales).

58
 Building Robot Brains

On the more practical side, iRobot's Roomba robot uses a more advanced
strategy (though it is time based) to ensure that it provides complete coverage.
A more interesting (and important) question one could ask would be:

Question: How does a vacuuming robot know that it is done cleaning the
room?

Most robots are programmed to either detect certain terminating situations or

are run based on time. For example, run around for 60 minutes and then stop.
Detecting situations is a little difficult and we will return to that in the next
chapter.

So far, you have programmed very simple robot behaviors. Each behavior
which is defined by a function, when invoked, makes the robot do something
for a fixed amount of time. For example, the yoyo behavior from the last
chapter when invoked as:

>>> yoyo(0.5, 1)

would cause the robot to do something for about 2 seconds (1 second to go

forward and then 1 second to move backward). In general, the time spent
carrying out the yoyo behavior will depend upon the value of the second
parameter supplied to the function. Thus if the invocation was:

>>> yoyo(0.5, 5.5)

the robot would move for a total of 11 seconds. Similarly, the dance behavior
defined in the previous chapter will last a total of six seconds. Thus, the total
behavior of a robot is directly dependent upon the time it would take to
execute all the commands that make up the behavior. Knowing how long a
behavior will take can help in pre-programming the total amount of time the
overall behavior could last. For example, if you wanted the robot to perform
the dance moves for 60 seconds, you can repeat the dance behavior ten times.
You can do this by simply issuing the dance command 10 times. But that gets
tedious for us to have to repeat the same commands so many times.
Computers are designed to do repetitious tasks. In fact, repetition is one of the
59
Chapter 3

key concepts in computing and all programming languages, including Python,

provide simple ways to specify repetitions of all kinds.

Doing Repetition in Python

If you wanted to repeat the dance behavior 10 times, all you have to do is:

for i in range(10):
dance()

This is a new statement in Python: the for-statement. It is also called a loop

statement or simply a loop. It is a way of repeating something a fixed number
of times. The basic syntax of a for-loop in Python is:

for <variable> in <sequence>:

<do something>
<do something>
...

The loop specification begins with the keyword for which is followed by a
<variable> and then the keyword in and a <sequence> followed by a colon
(:). This line sets up the number of times the repetition will be repeated. What
follows is a set of statements, indented (again, indentation is important), that
are called a block that forms the body of the loop (stuff that is repeated).

When executed, the <variable> (which is called a loop index variable) is

assigned successive values in the <sequence> and for each of those values, the
statements in the body of the loop are executed. A <sequence> in Python is a
list of values. Lists are central to Python and we will see several examples of
lists later. For now, look at the dance example above and notice that we have
used the function range(10) to specify the sequence. To see what this
function does you can start IDLE and enter it as an expression:

>>> range(10)
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

60
 Building Robot Brains

The result of entering the range(10) is a sequence (a list) of ten numbers

0..9. Notice how range returns a sequence of values starting from 0 all the
way up to, but including, 10. Thus, the variable i in the loop:

for i in range(10):
dance()

will take on the values 0 through 9 and for each of those values it will execute
the dance() command.

Do This: Let us try this out on the robot. Modify the robot program from the
start of this chapter to include the dance function and then write a main
program to use the loop above.

# File: dance.py
# Purpose: A simple dance routine

# First import myro and connect to the robot

from myro import *

init()

# Define the new functions...

def yoyo(speed, waitTime):

forward(speed, waitTime)
backward(speed, waitTime)
stop()

def wiggle(speed, waitTime):

motors(-speed, speed)
wait(waitTime)
motors(speed, -speed)
wait(waitTime)
stop()

def dance():
yoyo(0.5, 0.5)
yoyo(0.5, 0.5)
wiggle(0.5, 1)
wiggle(0.5, 1)

61
Chapter 3

# The main dance program

def main():
print "Running the dance routine..."

for danceStep in range(10):

dance()

print "...Done"
IDLE Tip
main()
You can stop a program at any time by
Notice that we have used hitting the CTRL‐C keys (pronounced as
danceStep (a more meaningful Control‐see). That is, pressing the
name than i) to represent the CTRL‐key and then at the same time
loop index variable. When you pressing the c‐key.
run this program, the robot
should perform the dance routine In the case of a robot program this will
ten times. Modify the value also stop the robot.
specified in the range command
to try out some more steps. If you end up specifying a really large value,
remember that for each value of danceStep the robot will do something for 6
seconds. Thus, if you specified 100 repetitions, the robot will run for 10
minutes.

In addition to repeating by counting, you can also specify repetition using

time. For example, if you wanted the robot (or the computer) to do something
for 30 seconds. You can write the following command to specify a repetition
based on time:

while timeRemaining(10):
<do something>
<do something>
...

The above commands will be repeated for 10 seconds. Thus, if you wanted
the computer to say “Doh!” for 5 seconds, you can write:

62
 Building Robot Brains

while timeRemaining(5):
speak("Doh!", 0)

In writing robot programs there will also be times when you just want the
robot to keep doing its behaviors forever! While technically by forever we do
mean eternity in reality what is likely to happen is either it runs out of
batteries, or you decide to stop it (by hitting CTRL-C). The Python command
to specify this uses a different loop statement, called a while-loop that can be
written as:

while True:
<do something>
<do something>
...

True is also a value in Python (along with False) about which we will learn
more a little later. For now, it would suffice for us to say that the above loop
is specifying that the body of the loop be executed forever!

Do This: Modify the dance.py program to use each of the while-loops

instead of the for-loop. In the last case (while TRUE: ) remember to use
CTRL-C to stop the repetitions (and the robot).

As we mentioned above, repetition is one of the key concepts in computing.

For example, we can use repetition to predict the world population in ten
years by repeatedly computing the values for each year:

for year in range(10):

population = population * (1 + growthRate)

That is, repeatedly add the increase in population, based on growth rate, ten
times.

Do This: Modify the worldPop.py program to input the current population,

growth rate, and the number of years to project ahead and compute the
resulting total population. Run your program on several different values

63
Chapter 3

(Google: “world population growth” to get latest numbers). Can you estimate
when the world population will become 9 billion?

Summary

This chapter introduced the basic structure of Python (and robot) programs.
We also learned about names and values in Python. Names can be used to
designate functions and values. The latter are also called variables. Python
provides several different types of values: integers, floating point numbers,
strings, and also boolean values (True and False). Most values have built-in
operations (like addition, subtration, etc.) that perform calculations on them.
Also, one can form sequences of values using lists. Python provides simple
built-in facilities for obtaining input from the user. All of these enable us to
write not only robot programs but also programs that perform any kind of
computation. Repetition is a central and perhaps the most useful concept in
computing. In Python you can specify repetition using either a for-loop or a
while-loop. The latter are useful in writing general robot brain programs. In
later chapters, we will learn how to write more sophisticated robot behaviors.

Myro Review

speak(<something>)
The computer converts the text in <something> to speech and speaks it out.
<something> is also simultaneously printed on the screen. Speech generation
is done synchronously. That is, anything following the speak command is
done only after the entire thing is spoken.

speak(<something>, 0)
The computer converts the text in <something> to speech and speaks it out.
<something> is also simultaneously printed on the screen. Speech generation
is done asynchronously. That is, execution of subsequent commands can be
done prior to the text being spoken.

timeRemaining(<seconds>)
This is used to specify timed repetitions in a while-loop (see below).

64
 Building Robot Brains

Python Review

Values
Values in Python can be numbers (integers or floating point numbers) or
strings. Each type of value can be used in an expression by itself or using a
combination of operations defined for that type (for example, +, -, *, /, % for
numbers). Strings are considered sequences of characters (or letters).

Names
A name in Python must begin with either an alphabetic letter (a-z or A-Z) or
the underscore (i.e. _) and can be followed by any sequence of letters, digits,
or underscore letters.

input(<prompt string>)

This function prints out <prompt string> in the IDLE window and waits for
the user to enter a Python expression. The expression is evaluated and its
result is returned as a value of the input function.

from myro import *

initialize("comX")

<any other imports>

<function definitions>
def main():
<do something>
<do something>
...

main()
This is the basic structure of a robot control program in Python. Without the
first two lines, it is the basic structure of all Python programs.

print <expression1>, <expression2>, ...

Prints out the result of all the expressions on the screen (in the IDLE
window). Zero or more expressions can be specified. When no expression is
specified, it prints out an empty line.

65
Chapter 3

<variable name> = <expression>

This is how Python assigns values to variables. The value generated by
<expression> will become the new value of <variable name>.

range(10)
Generates a sequence, a list, of numbers from 0..9. There are other, more
general, versions of this function. These are shown below.

range(n1, n2)
Generates a list of numbers starting from n1…(n2-1). For example,
range(5, 10) will generate the list of numbers [5, 6, 7, 8, 9].

range(n1, n2, step)

Generates a list of numbers starting from n1…(n2-1) in steps of step. For
example, range(5, 10, 2) will generate the list of numbers [5, 7, 9].

Repetition

for <variable> in <sequence>:

<do something>
<do something>
...

while timeRemaining(<seconds>):
<do something>
<do something>
...

while True:
<do something>
<do something>
...

These are different ways of doing repetition in Python. The first version will
assign <variable> successive values in <sequence> and carry out the body
once for each such value. The second version will carry out the body for
<seconds> amount of time. timeRemaining is a Myro function (see above).
The last version specifies an un-ending repetition.

66
 Building Robot Brains

Exercises

1. Write a Python program to convert a temperature from degrees Celsius to

degrees Fahrenheit. Here is a sample interaction with such a program:

Enter a temperature in degrees Celsius: 5.0

That is equivalent to 41.0 degrees Fahrenheit.

The formula to convert a temperature from Celsius to Fahrenheit is: C/5=(F-

32)/9, where C is the temperature in degrees Celsius and F is the temperature
in degrees Fahrenheit.

2. Write a Python program to convert a temperature from degrees Fahrenheit

to degrees Celsius.

3. Write a program to convert a given amount of money in US dollars to an

equivalent amount in Euros. Look up the current exchange rate on the web
(see xe.com, for example).

4. Modify the version of the dance program above that uses a for-loop to use
the following loop:

for danceStep in [1,2,3]:

dance()

That is, you can actually use a list itself to specify the repetition (and
successive values the loop variable will take). Try it again with the lists [3,
2, 6], or [3.5, 7.2, 1.45], or [“United”, “States”, “of”,
“America”]. Also try replacing the list above with the string “ABC”.
Remember, strings are also sequences in Python. We will learn more about
lists later.

5. Run the world population program (any version from the chapter) and when
it prompts for input, try entering the following and observe the behavior of the
program. Also, given what you have learned in this chapter, try and explain
the resulting behavior.

67
Chapter 3

6. Use the values 9000000000, and 1.42 as input values as above. Except,
when it asks for various values, enter them in any order. What happens?

7. Using the same values as above, instead of entering the value, say
9000000000, enter 6000000000+3000000000, or 450000000*2, etc. Do you
notice any differences in output?

8. For any of the values to be input, replace them with a string. For instance
enter "Al Gore" when it prompts you for a number. What happens?

9. Rewrite your solution to Exercise 4 from the previous chapter to use the
program structure described above.

10. You were introduced to the rules of naming in Python. You may have
noticed that we have made extensive use of mixed case in naming some
entities. For example, waitTime. There are several naming conventions used
by programmers and that has led to an interesting culture in of itself. Look up
the phrase CamelCase controversy in your favorite search engine to learn
about naming conventions. For an interesting article on this, see The
Semicolon Wars (www.americanscientist.org/issues/pub/the-semicolon-wars).

11. Experiment with the speak function introduced in this chapter. Try giving
it a number to speak (try both integers and floating point numbers). What is
the largest integer value that it can speak? What happens when this limit is
exceeded? Try to give the speak function a list of numbers, or strings, or
both.

12. Write a Python program that sings the ABC song: ABCD…XYZ. Now I
know my ABC’s. Next time won’t you sing with me?

68
Building Robot Brains

69
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Sensing From Within

I see dead people.

Cole Sear (played by Haley Joel Osment) in Sixth Sense,
M. Night Shyamalan, 1999.

Opposite page: Candle Flame

Photo courtesy of Jon Sullivan (www.pdphoto.org)

71
Chapter 4

Cole Sear in Shyamalan's Sixth Sense is not referring to dead bodies lying in
front of him (for those who have not seen the movie). The five senses that
most humans relate to are: touch, vision, balance, hearing, and taste or smell.
In all cases our bodies have special sensory receptors that are placed on
various parts of the body to enable sensing. For example the taste receptors
are concentrated mostly on the tongue; the touch receptors are most sensitive
on hands and the face and least on the back and on limbs although they are
present all over the body, etc. Besides the difference in the physiology of each
kind of receptors there are also different neuronal pathways and thereby
sensing mechanisms built into our bodies. Functionally, we can say that each
type of sensory system starts with the receptors which convert the thing they
sense into electrochemical signals that are transmitted over neurons. Many of
these pathways lead to the cerebral cortex in the brain where they are further
processed (like, "Whoa, that jalapeno is hot!!"). The perceptual system of an
organism refers to the set of sensory receptors, the neuronal pathways, and the
processing of perceptual information
in the brain. The brain is capable of
combining sensory information from Proprioception
different receptors to create richer Sensing from within
experiences than those facilitated by
the individual receptors. Get something really delicious to eat,
like a cookie, or a piece of chocolate,
The perceptual system of any or candy (whatever you fancy!). Hold
organism includes a set of external it in your right hand, and let your
sensors (also called exteroceptors) right arm hang naturally on your
and some internal sensing side. Now close your eyes, real tight,
mechanisms (interoceptors or and try to eat the thing you are
proprioception). Can you touch your holding. Piece of cake! (well,
belly button in the dark? This is whatever you picked to eat :‐)
because of proprioception. Your
body's sensory system also keeps Without fail, you were able to pick it
up and bring it to your mouth, right?
track of the internal state of your
body parts, how they are oriented,
Give yourself a Snickers moment and
etc.
enjoy the treat.

72
 Sensing From Within

Sensing is an essential component of being a robot and every robot comes

built with internal as well as external sensors. It is not uncommon, for
example, to find sensors that are capable of sensing light, temperature, touch,
distance to another object, etc. An example of internal sensing in robots is the
measurement of movement relative to the robot's internal frame of reference.
Sometimes also called dead reckoning, it can be a useful sensing mechanism
that you can use to design robot behaviors.

Robots employ electromechanical sensors and there are different types of

devices available for sensing the same physical quantity. For example, one
common sensor found on many robots is a proximity sensor. It detects the
distance to an object or an obstacle. Proximity sensors can be made using
different technologies: infrared light, sonar, or even laser. Depending upon the
type of technology used, their accuracy, performance, as well as cost vary:
infrared (IR) is the cheapest, and laser is on the expensive side. Lets us take a
look at the perceptual system of your Scribbler robot starting with internal
sensors.

Proprioception in the Scribbler

The Scribbler has three useful internal sensory mechanisms: stall, time, and
battery level. When your program asks the robot to move it doesn’t always
imply that the robot is actually physically moving. It could be stuck against a
wall, for example. The stall sensor in the Scribbler enables you to detect this.
You have already seen how you can use time to control behaviors using the
timeRemaining and wait functions. Also, for most movement commands,
you can specify how long you want that movement to take place (for example
forward(1, 2.5) means full-speed forward for 2.5 seconds). Finally, it is
also possible to detect battery power level so that you can detect when it is
time to change batteries in the robot.

Time

All computers come built-in with an internal clock. In fact, clocks are so
essential to the computers we use today that without them we would not have

73
Chapter 4

computers at all! Your Scribbler robot can use the computer’s clock to sense
time. It is with the help of this clock that we are able to use time in functions
like timeRemaining, wait, and other movement commands. Just with these
facilities it is possible to define interesting automated behaviors.

Do This: Design a robot program for the Scribbler to draw a square (say with
sides of 6 inches). To accomplish this, you will have to experiment with the
movements of the robot and correlate them with time. The two movements
you have to pay attention to are the rate at which the robot moves, when it
moves in a straight line; and the degree of turn with respect to time. You can
write a function for each of these:

def travelStraight(distance):
# Travel in a straight line for distance inches
...

def degreeTurn(angle):
# Spin a total of angle degrees

That is, figure out by experimentation on your own robot (the results will vary
from robot to robot) as to what the correlation is between the distance and the
time for a given type of movement above and then use that to define the two
functions above. For example, if a robot (hypothetical case) seems to travel at
the rate of 25 inches/minute when you issue the command translate(1.0),
then to travel 6 inches you will have to translate for a total of (6*60)/25
seconds. Try moving your robot forward for varying amounts for time at the
same fixed speed. For example try moving the robot forward at speed 0.5 for
3, 4, 5, 6 seconds. Record the distance travelled by the robot for each of those
times. You will notice a lot of variation in the distance even for the same set
of commands. You may want to average those. Given this data, you can
estimate the average amount of time it takes to travel an inch. You can then
define travelStraight as follows:

74
 Sensing From Within

def travelStraight(distance):
# set up your robot’s speed
inchesPerSec = <Insert your robot’s value here>

# Travel in a straight line for distance inches

forward(1, distance/inchesPerSec)

Similarly you can also determine the time required for turning a given number
of degrees. Try turning the robot at the same speed for varying amounts of
time. Experiment how long it takes the robot to turn 360 degrees, 720 degrees,
etc. Again, average the data you collect to get the number of degrees per
second. Once you have figured out the details use them to write the
degreeTurn function. Then use the following main program:

def main():
# Transcribe a square of sides = 6 inches

for side in range(4):

travelStraight(6.0)
degreeTurn(90.0)

speak("Look at the beautiful square I made.")

main()

Run this program several times. It is unlikely that you will get a perfect square
each time. This has to do with the calculations you performed as well with the
variation in the robot's motors. They are not precise. Also, it generally takes
more power to move from a still stop than to keep moving. Since you have no
way of controlling this, at best you can only approximate this type of
behavior. Over time, you will also notice that the error will aggregate. This
will become evident in doing the exercise below.

Do This: Building on the ideas from the previous exercise, we could further
abstract the robot's drawing behavior so that we can ask it to draw any regular
polygon (given the number of sides and length of each side). Write the
function:

75
Chapter 4

def drawPolygon(SIDES, LENGTH):

# Draw a regular polygon with SIDES number of sides
# and each side of length LENGTH.

Then, we can write a regular polygon drawing robot program as follows:

def main():
# Given the number of sides and the length of each side,
# draw a regular polygon

# First, ask the user for the number of sides and

# side length
print “Given # of sides and side length I will draw”
print “a polygon for you. Specify side length in inches.”

nSides = input("Enter # of sides in the polygon: ")

sideLength = input("Enter the length of each side: ")

# Draw the polygon

drawPolygon(nSides, sidelength)

speak("Look! I can draw.")

main()

To test the program, first try drawing a square of sides 6 inches as in the
previous exercise. Then try a triangle, a pentagon, hexagon, etc. Try a
polygon with 30 sides of length 0.5 inches. What happens when you give 1 as
the number of sides? What happens when you give zero (0) as the number of
sides?

A Slight Detour: Random Walks

One way you can do interesting things with robot drawings is to inject some
randomness in how long the robot does something. Python, and most
programming languages, typically provide a library for generating random
numbers. Generating random numbers is an interesting process in itself but we
will save that discussion for a later time. Random numbers are very useful in
all kinds of computer applications, especially games and in simulating real life

76
 Sensing From Within

phenomena. For example, in estimating how many cars might be entering an

already crowded highway in the peak of rush hour? Etc. In order to access the
random number generating functions in Python you have to import the
random library:

from random import *

There are lots of features available in this library but we will restrict ourselves
with just two functions for now: random and randint. These are described
below:

random() Returns a random number between 0.0 and 1.0.

randint(A, B) Returns a random number in the range [A…B].

Here is a sample interaction with these two functions:

77
Chapter 4

As you can see, using the random number library is easy enough, and similar
to using the Myro library for robot commands. Given the two functions, it is
entirely up to you how you use them. Look at the program below:

def main():
# generate 10 random polygons
for poly in range(10):
# generate a random polygon and draw it
Print “Place a new color in the pen port and then…”
userInput = input("Enter any number: ")
sides = randint(3, 8)
size = randint(2, 6)
drawPolygon(sides, size)

# generate a random walk of 20 steps

for step in range(20):
travelStraight(random())
degreeTurn(randrange(0, 360))

The first loop in the program draws 10 random polygons of sizes ranging
from 3 to 8 sides and each side in the range 2 to 6 inches. The second loop
carries out a random walk of 20 steps.

Asking Questions?

As you can see from above, it is easy to program various kinds of movements
into the Scribbler. If there is a pen in the pen port, the Scribbler draws a path.
Also in the example above, you can see that we can stop the program
temporarily, pretend that we are taking some input and use that as an
opportunity to change the pen and then go on. Above, we used the Python
input command to accomplish this. There is a better way to do this and it
uses a function provided in the Myro library:

>>> askQuestion("Are you ready?")

When this function is executed, a dialog window pops up as shown below:

78
 Sensing From Within

When you press your mouse on any of the choices (Yes/No), the window
disappears and the function returns the name of the key selected by the user as
a string. That is, if in the above window you pressed the Yes key, the function
will return the value:

>>> askQuestion("Are you ready?")

'Yes'

The askQuestion command can be used in the program above as follows:

askQuestion("Change my pen to a different color and press

'Yes' when ready.")

While this is definitely more functional than our previous solution, we can
actually do better. For example, what happens when the user presses the No
button in the above interaction? One thing you know for sure is that the
function will return the string 'No'. However, the way we are using this
function, it really does not matter which key the user presses. askQuestion is
designed so it can be customized by you so that you can specify how many
button choices you want to have in the dialog window as well as what the
names of those buttons would be. Here is an illustration of how you would
write a better version of the above command:

askQuestion("Change my pen to a different color and press 'OK'

when ready", ["OK"])

79
Chapter 4

Now this is certainly better. Notice that the function askQuestion can be
used with either one parameter or two. If only one parameter is specified, then
the default behavior of the function is to offer two button choices: 'Yes' and
'No'. However, using the second parameter you can specify, in a list, any
number of strings that will become the choice buttons. For example,

askQuestion("What is your favorite ice cream flavor?",

["Vanilla", "Chocolate", "Mango", "Hazelnut", "Other"])

This will be a very handy function to use in many different situations. In the
next exercise, try and use this function to become familiar with it.

Do This: Write a Scribbler program of your own that exploits the Scribbler's
movements to make random drawings. Make sure you generate drawings with
at least three or more colors. Because of random movements, your robot is
likely to run into things and get stuck. Help your robot out by picking it up
and placing it elsewhere when this happens.

80
 Sensing From Within

Back to time…

Most programming languages also allow you to access the internal clock to
keep track of time, or time elapsed (as in a stop watch), or in any other way
you may want to make use of time (as in the case of the wait) function. The
Myro library provides a simple function that can be used to retrieve the
current time:

>>> currentTime()
1169237231.836

The value returned by currentTime is a number that represents the seconds

elapsed since some earlier time, whatever that is. Try issuing the command
several times and you will notice that the difference in the values returned by
the function represents the real time in seconds. For example:

>>> currentTime()
1169237351.5580001
>>> 1169237351.5580001 - 1169237231.836
119.72200012207031

That is, 119.722 seconds had elapsed between the two commands above. This
provides another way for us to write robot behaviors. So far, we have learned
that if you wanted your robot to go forward for 3 seconds, you could either
do:

forward(1.0, 3.0)

or
forward(1.0)
wait(3.0)
stop()

or

81
Chapter 4

while timeRemaining(3.0):
forward(1.0)
stop()

Using the currentTime function, there is yet another way to do the same
thing:

startTime = currentTime() # record start time

while (currentTime() - startTime) < 3.0:
forward(1.0)
stop()

The above solution uses the internal clock. First, it records the start time. Next
it enters the loop which first gets the current time and then checks to see if the
difference between the current time and start time is less than 3.0 seconds. If
so, the forward command is repeated. As soon as the elapsed time gets over
3.0 seconds, the loop terminates. This is another way of using the while-loop
that you learned in the previous chapter. In the last chapter, you learned that
you could write a loop that executed forever as shown below:

while True:
<do something>

The more general form of the while-loop is:

while <some condition is true>:

<do something>

That is, you can specify any condition in <some condition is true>. The
condition is tested and if it results in a True value, the step(s) specified in <do
something> is/are performed. The condition is tested again, and so on. In the
example above, we use the expression:

(currentTime() - startTime) < 3.0

If this condition is true, it implies that the elapsed time since the start is less
than 3.0 seconds. If it is false, it implies that more than 3.0 seconds have
82
 Sensing From Within

elapsed and it results in a False value, and the loop stops. Learning about
writing such conditions is essential to writing smarter robot programs.

While it may appear that the solution that specified time in the forward
command itself seemed simple enough (and it is!), you will soon discover that
being able to use the internal clock as shown above provides more versatility
and functionality in designing robot behaviors. This, for example is how one
could program a vacuum cleaning robot to clean a room for 60 minutes:

startTime = currentTime()
while (currentTime() - startTime)/60.0 < 60.0:
cleanRoom()

You have now seen how to write robot programs that have behaviors or
commands that can be repeated a fixed number of times, or forever, or for a
certain duration:

# do something N times
for step in range(N):
do something...

# do something forever
while True:
do something...

# do something for some duration

while timeRemaining(duration):
do something...

# do something for some duration

duration = <some time in seconds>
startTime = currentTime()
while (currentTime() - startTime) < duration:
do something...

All of the above are useful in different situations. Sometimes it becomes a

matter of personal preference.

83
Chapter 4

Writing Conditions

Let us spend some time here to learn about conditions you can write in while-
loops. The first thing to realize is that all conditions result in either of two
values: True or False (or, alternately a 1 or a 0). These are Python values,
just like numbers. You can use them in many ways. Simple conditions can be
written using comparison (or relational) operations: < (less than), <= (less than
or equal to), > (greater than), >= (greater than or equal to), == (equal to), and
!= (not equal to). These operations can be used to compare all kinds of values.
Here are some examples:

>>> 42 > 23
True
>>> 42 < 23
False
>>> 42 == 23
False
>>> 42 != 23
True
>>> (42 + 23) < 100
True
>>> a, b, c = 10, 20, 10
>>> a == b
False
>>> a == c
True
>>> a == a
True
>>> True == 1
True
>>> False == 1
False

The last two examples above also show how the values True and False are
related to 1 and 0. True is the same as 1 and 0 is the same as False. You can
form many useful conditions using the comparison operations and all
conditions result in either True (or 1) or False (or 0). You can also compare
other values, like strings, using these operations:

84
 Sensing From Within

>>> "Hello" == "Good Bye"

False
>>> "Hello" != "Good Bye"
True
>>> "Elmore" < "Elvis"
True
>>> "New York" < "Paris"
True
>>> "A" < "B"
True Unicode
>>> "a" < "A"
False Text characters have an
internal computer coding or
Study the above examples carefully. Two
representation that enforces
important things to notice are: strings are
compared using alphabetical ordering (i.e. lexicographic ordering. This
lexicographically). Thus "Elmore" is less internal encoding is very
than "Elvis" since "m" is less than "v" in important in the design of
those strings ("El" being equal in both). That computers and this is what
is also why "New York" is less than "Paris" enables all computers and
(since "N" is less than "P"). The second devices like iPhones etc. to
important thing to note is that uppercase exchange information
letters are less than their equivalent lowercase consistently. All language
counterparts ("A" is less than "a"). This is by characters in the world have
design (see box on right).
been assigned a standard
Besides relational operations you can build computer encoding. This is
more complex conditional expressions using called Unicode.
the logical operations (also called Boolean
operations): and, or, and not. Here are some examples:

>>> (5 < 7) and (8 > 3)

True
>>> not ( (5 < 7) and (8 > 3))
False
>>> (6 > 7) or (3 > 4)
False

85
Chapter 4

>>> (6 > 7) or (3 > 2)

True

We can define the meaning of logical operators as follows:

• <expression-1> and <expression-2>: Such an expression will

result in a value True only if both <expression-1> and
<expression-2> are True. In all other cases (i.e. if either one or both
of <expression-1> and <expression-2> are False) it results in a
False.

• <expression-1> or <expression-2>: Such an expression will

result in a value True if either <expression-1> or <expression-2>
are True or if both are True. In all other cases (i.e. if both of
<expression-1> and <expression-2> are False) it results in a
False.

• not <expression>: Such an expression will result in a value True if

<expression> is False or False if <expression> is True). I.e., it
flips or complements the value of expression.

These operators can be combined with relational expressions to form

arbitrarily complex conditional expressions. In fact, any decision making in
your programs boils down to forming the appropriate conditional expressions.
The logical operators were invented by the logician George Boole in the mid
19th century. Boolean algebra, named after Boole, defines some simple, yet
important laws that govern the behavior of logical operators. Here are some
useful ones:

• (A or True) is always True.

• (not (not A)) is just A
• (A or (B and C)) is the same as ((A or B) and (A or C))
• (A and (B or C)) is the same as ((A and B) or (A and C))
• (not (A or B)) is the same as ((not A) and (not B))
• (not (A and B)) is the same as ((not A) or (not B))

86
 Sensing From Within

These identities or properties can help you in simplifying conditional

expressions. The conditional expressions can be used to write several useful
conditions to control the execution of some program statements. These allow
you to write conditional repetitions as:

while <some condition is true>:

<do something>

Now you can see why the following is a way of saying, “do something
forever”:

while True:
<do something>

Since the condition is always True the statements will be repeated forever.
Similarly, in the loop below:

while timeRemaining(duration):
<do something>

As soon as the duration is up, the value of the timeRemaining(duration)

expression will become False and the repetition will stop. Controlling the
repetitions based on conditions is a powerful idea in computing. We will be
using these extensively to control the behaviors of robots.

Sensing Stall

We mentioned in the beginning of this chapter that the Scribbler also has a
way of sensing that it is stalled when trying to move. This is done by using the
Myro function getStall:

getStall() Returns True if the robot has stalled, False otherwise.

You can use this to detect that the robot has stalled and even use it as a
condition in a loop to control behavior. For example:

87
Chapter 4

while not getStall():

<do something>

That is, keep doing <do something> until the robot has stalled. Thus, you
could write a robot behavior that goes forward until it bumps into something,
say a wall, and then stops.

while not getStall():

forward(1.0)
stop()
speak("Ouch! I think I bumped into something! ")

In the above example, as long as the robot is not stalled, getStall() will
return False and hence the robot will keep going forward (since not False is
True). Once it does bump into something, getStall() will return True and
then the robot will stop and speak.

Do This: Write a complete program for the Scribbler to implement the above
behavior and then observe the behavior. Show this to some friends who are
not in your course. Ask them for their reactions. You will notice that people
will tend to ascribe some form of intelligence to the robot. That is, your robot
is sensing that it is stuck, and when it is, it stops trying to move and even
announces that it is stuck by speaking. We will return to this idea of artificial
intelligence in a later chapter.

Sensing Battery Power Levels

Your Scribbler robot runs on 6 AA batteries. As with any other electronic

device, with use, the batteries will ultimately drain and you will need to
replace with fresh ones. Myro provides an internal battery-level sensing
function, called getBattery that returns the current voltage being supplied by
the battery. When the battery levels go down, you will get lower and lower
voltages causing erratic behavior. The battery voltage levels of your Scribbler
will vary between 0 and 9 volts (0 being totally drained). What low means is
something you will have to experiment and find out. The best way to do this is
to record the battery level when you insert a fresh set of batteries. Then, over

88
 Sensing From Within

time, keep recording the battery levels as Disposing Batteries

you go.
Make sure that you dispose
The Scribbler also has some built-in battery- used batteries properly and
level indicator lights. The red LED on the responsibly. Batteries may
robot remains lit when the power levels are contain hazardous materials
high (or in the good range). It starts to flash
like cadmium, mercury, lead,
when the battery level runs low. There is
also an similar LED on the Fluke dongle. lithium, etc. which can be
Can you find it? Just wait until the battery deadly pollutants if disposed
levels run low and you will see it flashing. in landfills. Find out your
nearest battery recycling or
You can use battery-level sensing to define disposal option to ensure
behaviors for robots so that they are carried proper disposal.
out only when there is sufficient power
available. For example:
while (getBattery() >= 5.0) and timeRemaining(duration):
<do something>

That is, as long as battery power is above 5.0 and the time limit has not
exceeded duration, <do something>.

World Population, revisited

The ability to write conditional expressions also enables us to define more

sophisticated computations. Recall the world population projection example
from previous chapter. Given the population growth rate and the current
population, you can now write a program to predict the year when the world’s
population will increase to beyond a given number, say 9 billion. All you have
to do is write a condition-driven repetition that has the following structure:

89
Chapter 4

year = <current year>

population = <current population>
growthRate = <rate of growth>

# repeat as long as the population stays below 9000000000

while population < 9000000000:
# compute the population for the next year
year = year + 1
population = population * (1+growthRate)

print "By the year", year, "the world’s population"

print "will have exceeded 9 billion."

That is, add population growth in the next year if the population is below 9
billion. Keep repeating this until it exceeds 9 billion.

Do This: Complete the program above and compute the year when the
world’s population will exceed 9 billion. To make your program more useful
make sure you ask the user to input the values of the year, population, growth
rate, etc. In fact, you can even ask the user to enter the population limit so you
make use the program for any kinds of predictions (8 billion? 10 billion?).
How would you change the program so it prints the population projection for
a given year, say 2100?

Summary

In this chapter you have learned about proprioception or internal sensory

mechanisms. The Scribbler robot has three internal sensory mechanisms:
time, stall, and battery-level. You have learned how to sense these quantities
and also how to use them in defining automated robot behaviors. You also
learned about random number generation and used it to define unpredictable
robot behaviors. Later, we will also learn how to use random numbers to write
games and to simulate natural phenomena. Sensing can also be used to define
conditional repetitive behaviors using conditional expressions in while-loops.
You learned how to construct and write different kinds of conditions using
relational and logical operations. These will also become valuable in defining

90
 Sensing From Within

behaviors that use external sensory mechanisms and also enable us to write
more explicit decision-making behaviors. We will learn about these in the
next chapter.

Myro Review
randomNumber()
Returns a random number in the range 0.0 and 1.0. This is an alternative Myro
function that works just like the random function from the Python random
library (see below).

askQuestion(MESSAGE-STRING)
A dialog window with MESSAGE-STRING is displayed with choices: 'Yes' and
'No'. Returns 'Yes' or 'No' depending on what the user selects.

askQuestion(MESSAGE-STRING, LIST-OF-OPTIONS)
A dialog window with MESSAGE-STRING is displayed with choices indicated in
LIST-OF-OPTIONS. Returns option string depending on what the user selects.

currentTime()
The current time, in seconds from an arbitrary starting point in time, many
years ago.

getStall()
Returns True if the robot is stalled when trying to move, False otherwise.

getBattery()
Returns the current battery power level (in volts). It can be a number between
0 and 9 with 0 indication no power and 9 being the highest. There are also
LED power indicators present on the robot. The robot behavior becomes
erratic when batteries run low. It is then time to replace all batteries.

Python Review

True, False
These are Boolean or logical values in Python. Python also defines True as 1
and False as 0 and they can be used interchangeably.

91
Chapter 4

<, <=, >, >=, ==, !=

These are relational operations in Python. They can be used to compare
values. See text for details on these operations.

and, or not
These are logical operations. They can be used to combine any expression that
yields Boolean values.
random()
Returns a random number between 0.0 and 1.0. This function is a part of the
random library in Python.

randint(A, B)
Returns a random number in the range A (inclusive) and B (exclusive). This
function is a part of the random library in Python.

Exercises

1. Write a robot program to make your Scribbler draw a five point star. [Hint:
Each vertex in the star has an interior angle of 36 degrees.]

2. Experiment with Scribbler movement commands and learn how to make it

transcribe a path of any given radius. Write a program to draw a circle of any
input diameter.

3. Write a program to draw other shapes: the outline of a house, a stadium, or

create art by inserting pens of different colors. Write the program so that the
robot stops and asks you for a new color.

4. If you had an open rectangular lawn (with no trees or obstructions in it) you
could use a Zanboni like strategy to mow the lawn. Start at one end of the
lawn, mow the entire length of it along the longest side, turn around and mow
the entire length again, next to the previously mowed area, etc. until you are
done. Write a program for your Scribbler to implement this strategy (make the
Scribbler draw its path as it goes).

92
 Sensing From Within

5. Enhance the random drawing program from this chapter to make use of
speech. Make the robot, as it is carrying out random movements, to speak out
what it is doing. As a result you will have a robot artist that you have created!

6. Rewrite your program from the previous exercise so that the random
behavior using each different pen is carried out for 30 seconds.

7. The Myro library also provides a function called, randomNumber() that

returns a random number in the range 0.0 and 1.0. This is similar to the
function random() from the Python library random that was introduced in this
chapter. You can use either based on your own preference. You will have to
import the appropriate library depending on the function you choose to use.
Experiment with both to convince yourself that these two are equivalent.

8. In reality, you only need the function random() to generate random

numbers in any range. For example, you can get a random number between 1
and 6 with randRange(1,6) or as shown below:

randomValue = 1 + int(random()*6)

The function int() takes any number as its parameter, truncates it to a whole
number and returns an integer. Given that random() returns values between
0.0 (inclusive) and 1.0 (exclusive), the above expression will assign a random
value between 1..5 (inclusive) to randomValue. Given this example, write a
new function called myRandRange() that works just like randrange():
def myRandRange(A, B):

# generate a random number between A..B

# (just like as defined for randrange)

9. What kinds of things can your robot talk about? You have already seen how
to make the robot/computer speak a given sentence or phrase. But the robot
can also "talk" about other things, like the time or the weather.

One way to get the current time and date is to import another Python library
called time:

93
Chapter 4

>>> from time import *

The time module provides a function called localtime that works as follows:

>>> localtime()
(2007, 5, 29, 12, 15, 49, 1, 149, 1)

localtime returns all of the following in order:

1. year
2. month
3. day
4. hour
5. minute
6. seconds
7. weekday
8. day of the year
9. whether it is using daylight savings time, or not

In the example above, it is May 29, 2007 at 12:15pm and 49 seconds. It is also
the 1st day of the week, 149 day of the year, and we are using daylight
savings time. You can assign each of the values to named variables as shown
below:

year, month, day,…, dayOfWeek = localtime()

Then, for the example above, the variable year will have the value 2007;
month will have the value 5, etc. Write a Python program that speaks out the
current date and time.

94
Sensing From Within

95
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Sensing The World

I see all obstacles in my way.

From the song I can see clearly now,
Johnny Nash, 1972.

Opposite page: The Senses

Photo courtesy of Blogosphere (cultura.blogosfere.it)

97
Chapter 5

In the previous chapter you learned how proprioception: sensing time, stall,
and battery-level can be used in writing simple yet interesting robot behaviors.
All robots also come equipped with a suite of external sensors (or
exteroceptors) that can sense various things in the environment. Sensing
makes the robot aware of its environment and can be used to define more
intelligent behaviors. Sensing is also related to another important concept in
computing: input. Computers act on different kinds of information: numbers,
text, sounds, images, etc. to produce useful applications. Acquiring
information to be processed is generally referred to as input. In this chapter
we will also see how other forms of input can be acquired for a program to
process. First, let us focus on Scribbler’s sensors.

Scribbler Sensors

The Scribbler robot can sense the amount of ambient light, the presence (or
absence) of obstacles around it, and also take pictures from its camera.
Several devices (or sensors) are located on the Scribbler (see picture on the
previous page). Here is a short description of these:

Camera: The camera can take a still picture of whatever the robot is currently
“seeing”.

Light: There are three light sensors present on the robot. These are located in
the three holes present on the front of the robot. These sensors can detect the
levels of brightness (or darkness). These can be used to detect variations in
ambience light in a room. Also, using the information acquired from the
camera, the Scribbler makes available an alternative set of brightness sensors
(left, center, and right).

Proximity: There are two sets of these on the Scribbler: IR Sensors (left and
right) on the front; and Obstacle Sensors (left, center, and right) on the Fluke
dongle. They can be used to detect objects on the front and on its sides.

98
 Sensing the World

Getting to know the sensors

Sensing using the sensors provided in the Scribbler is easy. Once you are
familiar with the details of sensor behaviors you will be able to use them in
your programs to design interesting creature-like behaviors for your Scribbler.
But first, we must spend some time getting to know these sensors; how to
access the information reported by them; and what this information looks like.
As for the internal sensors, Myro provides several functions that can be used
to acquire data from each sensor device. Where multiple sensors are available,
you also have the option of obtaining data from all the sensors or selectively
from an individual sensor.

Do This: perhaps the best way to get a quick look at the overall behavior of
all the sensors is to use the Myro function
senses:
Scribbler Sensors
>>> senses()

This results in a window (see picture on

right) showing all of the sensor values
(except the camera) in real time. They are
updated every second. You should move the
robot around and see how the sensor values
change. The window also displays the values
of the stall sensor as well as the battery level.
The leftmost value in each of the sensor sets (light, IR, obstacle, and bright) is
the value of the left sensor, followed by center (if present), and then the right.

The Camera

The camera can be used to take pictures of

the robot’s current view. As you can see,
the camera is located on the Fluke dongle.
The view of the image taken by the camera

99
Chapter 5

will depend on the orientation of the robot (and the dongle). To take pictures
from the camera you can use the takePicture command:

takePicture()
takePicture("color")
takePicture("gray")

Takes a picture and returns a picture

object. By default, when no parameters
are specified, the picture is in color.
Using the "gray" option, you can get a
grayscale picture. Example:

>>> p = takePicture()
>>> show(p)

Alternately, you can also do:

>>> show(takePicture())

Once you take a picture from the camera, you can do many things with it. For
example, you may want to see if there is a laptop computer present in the
picture. Image processing is a vast subfield of computer science and has
applications in many areas. Understanding an image is quite complex but
something we do quite naturally. For example, in the picture above, we have
no problem locating the laptop, the bookcase in the background, and even a
case for a badminton racket (or, if you prefer racquet). The camera on the
Scribbler is its most complex sensory device that will require a lot of
computational effort and energy to use it in designing behaviors. In the
simplest case, the camera can serve as your remote “eyes” on the robot. We
may not have mentioned this earlier, but the range of the Bluetooth wireless
on the robot is 100 meters. In Chapter 9 we will learn several ways of using
the pictures. For now, if you take a picture from the camera and would like to
save it for later use, you use the Myro command, savePicture, as in:

>>> savePicture(p, “office-scene.jpg”)

100
 Sensing the World

The file office-scene.jpg will be saved in the same folder as your Start
Python folder. You can also use savePicture to save a series of pictures
from the camera and turn it into an animated “movie” (an animated gif
image). This is illustrated in the example below.

Do This: First try out all the commands for taking and saving pictures. Make
sure that you are comfortable using them. Try taking some grayscale pictures
as well. Suppose your robot has ventured into a place where you cannot see it
but it is still in communication range with your computer. You would like to
be able to look around to see where it is. Perhaps it is in a new room. You can
ask the robot to turn around and take several pictures and show you around.
You can do this using a combination of rotate and takePicture commands
as shown below:

while timeRemaining(30):
show(takePicture())
turnLeft(0.5, 0.2)

That is, take a picture and then turn for 0.2 seconds, repeating the two steps
for 30 seconds. If you watch the picture window that pops up, you will see
successive pictures of the robot’s views. Try this a few times and see if you
can count how many different images you are able to see. Next, change the
takePicture command to take grayscale images. Can you count how many
images it took this time? There is of course an easier way to do this:

N = 0
while timeRemaining(30):
show(takePicture())
turnLeft(0.5, 0.2)
N = N + 1
print N

101
Chapter 5

Now it will output the number of images it Pixels

takes. You will notice that it is able to take
many more grayscale images than color ones. Each image is made up of
This is because color images have a lot more several tiny picture elements
information in them than grayscale images or pixels. In a color image,
(see text on right). A 256x192 color image each pixel contains color
requires 256x192x3 (= 147, 456) bytes of data information which is made up
where as a grayscale image requires only of the amount of red, green,
256x192 (= 49,152) bytes. The more data you and blue (RGB). Each of these
have to transfer from the robot to the values is in the range 0..255
computer, the longer it takes. and hence it takes 3 bytes or
24‐bits to store the
You can also save an animated GIF of the information contained in a
images generated by the robot by using the single pixel. A pixel that is
colored pure red will have the
savePicture command by accumulating a
RGB values (255, 0, 0).
series of images in a list. This is shown
A grayscale image, on the
below:
other hand only contains the
Pics = [] level of gray in a pixel which
while timeRemaining(30): can be represented in a single
pic = takePicture() byte (or 8‐bits) as a number
show(pic)
ranging from 0..255 (where 0
Pics.append(pic)
turnLeft(0.5, 0.2) is black and 255 is white).
savePicture(Pics, “office-movie.gif”)

First we create an empty list called, Pics. Then we append each successive
picture taken by the camera into the list. Once all the images are accumulated,
we use savePicture to store the entire set as an animated GIF. You will be
able to view the animated GIF inside any web browser. Just load the file into a
web browser and it will play all the images as a movie.

There are many more interesting ways that one can use images from the
camera. In Chapter 9 we will explore images in more detail. For now, let us
take a look at Scribbler’s other sensors.

102
 Sensing the World

Light Sensing

The following functions are available to obtain values of light sensors:

getLight()Returns a list containing the three values of all light sensors.

getLight(<POSITION>) Returns the current value in the <POSITION> light

sensor. <POSITION> can either be one of 'left', 'center', 'right' or one
of the numbers 0, 1, 2. The positions 0, 1, and 2 correspond to the left, center,
and right sensors. Examples:

>>> getLight()
[135, 3716, 75]
>>> getLight('left')
135
>>> getLight(0)
135
>>> getLight('center')
3716
>>> getLight(1)
3716
>>> getLight('right')
75
>>> getLight(2)
75

The values being reported by these sensors can be in the range [0..5000]
where low values imply bright light and high values imply darkness. The
above values were taken in ambient light with one finger completely covering
the center sensor. Thus, the darker it is, the higher the value reported. In a
way, you could even call it a darkness sensor. Later, we will see how we can
easily transform these values in many different ways to affect robot behaviors.

It would be a good idea to use the senses function to play around with the
light sensors and observe their values. Try to move the robot around to see
how the values change. Turn off the lights in the room, or cover the sensors
with your fingers, etc.

103
Chapter 5

When you use the getLight function without any parameters, you get a list of
three sensor values (left, center, and right). You can use assign these to
individual variables in many ways:

>>> L, C, R = getLight()
>>> print L
135
>>> Center = getLight(“center”)
>>> print center
3716

The variables can then be used in many ways to define robot behaviors. We
will see several examples of these in the next chapter.

The camera present on the Fluke dongle can also be used as a kind of
brightness sensor. This is done by averaging the brightness values in different
zones of the camera image. In a way, you can think of it as a virtual sensor.
That is, it doesn’t physically exist but is embedded in the functionality of the
camera. The function getBright is similar to getLight in how it can be used
to obtain brightness values:

getbright()Returns a list containing the three values of all light sesnors.

getBright(<POSITION>) Returns the current value in the <POSITION> light

sensor. <POSITION> can either be one of 'left', 'center', 'right' or one of the
numbers 0, 1, 2. The positions 0, 1, and 2 correspond to the left, center, and
right sensors. Examples:

>>> getBright()
[2124047, 1819625, 1471890]
>>> getBright('left')
2124047
>>> getBright(0)
2124047
>>> getBright('center')
1819625
>>> getBright(1)
1819625
>>> getBright('right')

104
 Sensing the World

1471890
>>> getBright(2)
1471890

The above values are from the camera image of the Firefox poster (see picture
above). The values being reported by these sensors can vary depending on the
view of the camera and the resulting brightness levels of the image. But you
will notice that higher values imply bright segments and lower values imply
darkness. For example, here is another set
of values based on the image shown here
on the right.
>>> getBright()
[1590288, 1736767, 1491282]

As we can see, a darker image is likely to

produce lower brightness values. In the
image, the center of the image is brighter
than its left or right sections.

It is also important to note the differences in the nature of information being

reported by the getLight and getBright sensors. The first one reports the
amount of ambient light being sensed by the robot (including the light above
the robot). The second one is an average of the brightness obtained from the
image seen from the camera. These can be used in many different ways as we
will see later.

Do This: The program shown below uses a normalization function to

normalize light sensor values in the range [0.0..1.0] relative to the values of
ambient light. Then, the normalized left and right light sensor values are used
to drive the left and right motors of the robot.

# record average ambient light values

Ambient = sum(getLight())/3.0

# This function normalizes light sensor values to 0.0..1.0

def normalize(v):
if v > Ambient:

105
Chapter 5

v = Ambient

return 1.0 - v/Ambient

def main():
# Run the robot for 60 seconds
while timeRemaining(60):
L, C, R = getLight()
# motors run proportional to light
motors(normalize(L), normalize(R))
stop()

Run the above program on your Scribbler robot and observe its behavior. You
will need a flashlight to affect better reactions. When the program is running,
try to shine the flashlight on one of the light sensors (left or right). Observe
the behavior. Do you think the robot is behaving like an insect? Which one?
Study the program above carefully. There are some new Python features used
that we will discuss shortly. We will also return to the idea of making robots
behave like insects in the next chapter.

Proximity Sensing

The Scribbler has two sets of proximity detectors. There are two infrared (IR)
sensors on the front of the robot and there are three additional IR obstacle
sensors on the Fluke dongle. The following functions are available to obtain
values of the front IR sensors:

getIR()Returns a list containing

the two values of all IR sesnors.

getIR(<POSITION>) Returns the

current value in the <POSITION>
IR sensor. <POSITION> can either
be one of 'left' or 'right' or one of
the numbers 0, 1. The positions 0
and 1 correspond to the left,
center, and right sensors.

106
 Sensing the World

Examples:
>>> getIR()
[1, 0]
>>> getIR('left')
1
>>> getIR(0)
1
>>> getIR('right')
0
>>> getIR(1)
0

IR sensors return either a 1 or a 0. A value of 1 implies that there is nothing in

close proximity of the front of that sensor and a 0 implies that there is
something right in front of it. These sensors can be used to detect the presence
or absence of obstacles in front of the robot. The left and right IR sensors are
places far enough apart that they can be used to detect individual obstacles on
either side.

Do This: Run the senses function and observe the values of the IR sensors.
Place various objects in front of the robot and look at the values of the IR
proximity sensors. Take your notebook and place it in front of the robot about
two feet away. Slowly move the notebook closer to the robot. Notice how the
value of the IR sensor changes from a 1 to a 0 and then move the notebook
away again. Can you figure out how far (near) the obstacle should be before it
is detected (or cleared)? Try moving the notebook from side to side. Again
notice the values of the IR sensors.

The Fluke dongle has an additional set of obstacle sensors on it. These are
also IR sensors but behave very differently in terms of the kinds of values
they report. The following functions are available to obtain values of the
obstacle IR sensors:

getObstacle()Returns a list containing the two values of all IR sensors.

getObstacle(<POSITION>) Returns the current value in the <POSITION> IR

107
Chapter 5

sensor. <POSITION> can either be one of 'left', ‘center’, or 'right' or one of the
numbers 0, 1, or 2. The positions 0, 1, and 2 correspond to the left, center, and
right sensors.

Examples:
>>> getObstacle()
[1703, 1128, 142]
>>> getObstacle('left')
1703
>>> getObstacle(0)
1703
>>> getObstacle('center')
1128
>>> getObstacle(1)
1128
>>> getObstacle('right')
142
>>> getObstacle(2)
142

The values reported by these sensors range from 0 to 7000. A 0 implies there
is nothing in front of the sensor where as a high number implies the presence
of an object. The sensors on the sides can be used to detect the presence (or
absence of walls on the sides).

Do This: Place your Scribbler on the floor, turn it on, start Python, and
connect to it. Also connect the game pad controller and start the manual drive
operation (gamepad()). Next, issue the senses command to get the real time
sensor display. Our objective here is to really "get into the robot's mind" and
drive it around without ever looking at the robot. Also resist the temptation to
take a picture. You can use the information displayed by the sensors to
navigate the robot. Try driving it to a dark spot, or the brightest spot in the
room. Try driving it so it never hits any objects. Can you detect when it hits
something? If it does get stuck, try to maneuver it out of the jam! This
exercise will give you a pretty good idea of what the robot senses, how it can
use its sensors, and to the range of behaviors it may be capable of. You will
find this exercise a little hard to carry out, but it will give you a good idea as

108
 Sensing the World

to what should go into the brains of such robots when you actually try to
design them. We will try and revisit this scenario as we build various robot
programs.

Also do this: Try out the program below. It is very similar to the program
above that used the normalized light sensors.

def main():
# Run the robot for 60 seconds
while timeRemaining(60):
L, R = getIR()
# motors run proportional to IR values
motors(R, L)
main()

Since the IR sensors report 0 or 1 values, you do not need to normalize them.
Also notice that we are putting the left sensor value (L) into the right motor
and the right sensor value (R) into the left motor. Run the program and
observe the robot’s behavior. Keep a notebook handy and try to place it in
front of the robot. Also place it slightly on the left or on the right. What
happens? Can you summarize what the robot is doing? What happens when
you switch the R and L values to the motors?

You can see how simple programs like the ones we have seen above can result
in interesting automated control strategies for robots. You can also define
completely automated behaviors or even a combination of manual and
automated behaviors for robots. In the next chapter we will explore several
robot behaviors. First, it is time to learn about lists in Python.

Lists in Python

You have seen above that several sensor functions return lists of values. We
also used lists to accumulate a series of pictures from the camera to generate
an animated GIF. Lists are a very useful way of collecting a bunch of
information and Python provides a whole host of useful operations and
functions that enable manipulation of lists. In Python, a list is a sequence of

109
Chapter 5

objects. The objects could be anything: numbers, letters, strings, images, etc.
The simplest list you can have is an empty list:

>>> []
[]

or

>>> L = []
>>> print L
[]

An empty list does not contain anything. Here are some lists that contain
objects:

>>> N = [7, 14, 17, 20, 27]

>>> Cities = [“New York”, “Dar es Salaam”, “Moscow”]
>>> FamousNumbers = [3.1415, 2.718, 42]
>>> SwankyZips = [90210, 33139, 60611, 10036]
>>> MyCar = [“Toyota Prius”, 2006, “Purple”]

As you can see from above, a list could be a collection of any objects. Python
provides several useful functions that enable manipulation of lists. Below, we
will show some examples using the variables defined above:

>>> len(N)
5
>>>len(L)
0
>>> N + FamousNumbers
[7, 14, 17, 20, 27, 3.1415, 2.718, 42]
>>> SwankyZips[0]
90210
>>> SwankyZips[1:3]
[33139, 60611]
>>> 33139 in SwankyZips
True
>>> 19010 in SwankyZips
False

110
 Sensing the World

From the above, you can see that the function len takes a list and returns the
length or the number of objects in the list. An empty list has zero objects in it.
You can also access individual elements in a list using the indexing operation
(as in SwankyZips[0]). The first element in a list has index 0 and the last
element in a list of n elements will have an index n-1. You can concatenate
two lists using the ‘+’ operator to produce a new list. You can also specify a
slice in the index operation (as in SwankyZips[1:3] to refer to the sublist
containing elements from index 1 through 2 (one less than 3). You can also
form True/False conditions to check if an object is in a list or not using the
in operator. These operations are summarized in more detail at the end of the
chapter.

Besides the operations above, Python also provides several other useful list
operations. Here are examples of two useful list operations sort and
reverse:

>>> SwankyZips
[90210, 33139, 60611, 10036]
>>> SwankyZips.sort()
>>> SwankyZips
[10036, 33139, 60611, 90210]
>>> SwankyZips.reverse()
>>> SwankyZips
[90210, 60611, 33139, 10036]
>>> SwankyZips.append(19010)
>>> SwankyZips
[90210, 60611, 33139, 10036, 19010]

sort rearranges elements in the list in ascending order. reverse reverses the
order of elements in the list, and append appends an element to the end of the
list. Some other useful list operations are listed at the end of the chapter.
Remember that lists are also sequences and hence they can be used to perform
repetitions. For example:

>>> Cities = ["New York", "Dar es Salaam", "Moscow"]

>>> for city in Cities:
print city

111
Chapter 5

New York
Dar es Salaam
Moscow

The variable city takes on subsequent values in the list Cities and the
statements inside the loop are executed once for each value of city. Recall,
that we wrote counting loops as follows:

for I in range(5):
<do something>

The function range returns a sequence of numbers:

>>> range(5)
[0, 1, 2, 3, 4]

Thus the variable I takes on values in the list [0, 1, 2, 3, 4] and, as in the
example below, the loop is executed 5 times:

>>> for I in range(5):

print I

0
1
2
3
4

Also recall that strings are sequences. That is, the string:
ABC = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"

is a sequence of 26 letters. You can write a loop that runs through each
individual letter in the string and speaks it out as follows:
>>> for letter in ABC:
speak(letter)

112
 Sensing the World

There are also some useful functions that convert strings into lists. Say we
have a string containing a sentence:

>>> sentence = "Would you have any Grey Poupon"

You can convert the string above into individual words using the split
operation:

>>> sentence.split()
['Would', 'you', 'have', 'any', 'Grey', 'Poupon']

In light of the list operations presented above review some of the sensing
examples from earlier in the chapter. We will be using lists in many examples
in the remained of the text. For now, let us return to the topic of sensing.

Extrasensory Perception?

You have seen many ways of acquiring sensory information using the robot’s
sensors. In addition to the robot itself, you should be aware that your
computer also has several “sensors” or devices to acquire all kinds of data.
For example, you have already seen how, using the input function, you can
input some values into your Python programs:

>>> N = input("Enter a number: ")

Enter a number: 42
>>> print N
42

Indeed, there are other ways you can acquire information into your Python
programs. For example, you can input some data from a file in your folder. In
Chapter 1 you also saw how you were able to control your robot using the
game pad controller. The game pad was actually plugged into your computer
and was acting as an input device. Additionally, your computer is most likely
connected to the internet using which you can access many web pages. It is
also possible to acquire the content of any web page using the internet.
Traditionally, in computer science people refer to this is a process of input.
Using this view, getting sensory information from the robot is just a form of

113
Chapter 5

input. Given that we have at our disposal all of the input facilities provided by
the computer, we can just as easily acquire input from any of the modalities
and combine them with robot behaviors if we wish. Whether you consider this
as extra sensory perception or not is a matter of opinion. Regardless, being
able to get input from a diverse set of sources can make for some very
interesting and useful computer and robot applications.

Game Pad Controllers

The game pad controller you used in

Chapter 1 is a typical device that
provides interaction facilities when
playing computer games. These
devices have been standardized
enough that, just like a computer
mouse or a keyboard, you can
purchase one from a store and plug it
into a USB port of your computer.
Myro provides some very useful input
functions that can be used to get input
from the game pad controller. Game pads come is all kinds of flavors and
configurations with varying numbers of buttons, axes, and other devices on
them. In the examples below, we will restrict ourselves to a basic game pad
shown in the picture on previous page.

The basic game pad has eight buttons (numbered 1 through 8 in the picture)
and an axis controller (see picture on right). The buttons can be pressed or
released (on/off) which are represented by 1 (for on) and 0 (for off). The axis
can be pressed in many different orientations represented by a pair of values
(for the x-axis and y-axis) that range from -1.0 to 1.0 with [0.0, 0.0]
representing no activity on the axis. Two Myro functions are provided to
access the values of the buttons and the axis:

114
 Sensing the World

getGamepad(<device>)
getGamepadNow(<device>) Game Pad’s Axis Control
returns the values indicating the status of
the specified <device>. <device> can be
"axis" or "button".

The getGamepad function returns only after

<device> has been used by the user. That
is, it waits for the user to press or use that
device and then returns the values
associated with the device at that instant.
getGamepadNow does not wait and simply
returns the device status right away. Here are
some examples: Game Pad’s Axes

>>> getGamepadNow("axis")
[0.0, 0.0]
>>> getGamepad("axis")
[0.0, -1.0]
>>> getGamepadNow("button")
[0, 0, 0, 0, 1, 1, 0, 0]

Both getGamepad and getGamepadNow return the same set of values: axis
values are returned as a list [x-axis, y-axis] (see picture on right for
orientation) and the button values are returned as a list of 0, and 1’s. The first
value in the list is the status of button#1, followed by 2, 3, and so on. See
picture above for button numbering.

Do This: Connect the game pad controller to your computer, start Python, and
import the Myro module. Try out the game pad commands above and observe
the values. Here is another way to better understand the operation of the game
pad and the game pad functions:

while timeRemaining(30):
print getGamepad("button")

115
Chapter 5

Try out different button combinations. What happens when you press more
than one button? Repeat the above for axis control and observe the values
returned (keep the axes diagram handy for orientation purposes).

The game pad controller can be used for all kinds of interactive purposes,
especially for robot control as well as in writing computer games (see Chapter
X). Let us write a simple game pad based robot controller. Enter the program
below and run it.

def main():
# A simple game pad based robot controller
while timeRemaining(30):
X, Y = getGamePadNow("axis")
motors(X, Y)

The program above will run for 30 seconds. In that time it will repeatedly
sample the values of the axis controller and since those values are in the range
-1.0..1.0, it uses them to drive the motors. When you run the above program
observe how the robot moves in response to pressing various parts of the axis.
Do the motions of the robot correspond to the directions shown in the game
pad picture on previous page? Try changing the command from motors to
move (recall that move takes two values: translate and rotate). How does it
behave with respect to the axes? Try changing the command to move(-X, -
Y). Observe the behavior.

As you can see from the simple example above, it is easy to combine input
from a game pad to control your robot. Can you expand the program above to
behave exactly like the gamepad controller function you used in Chapter 1?
(See Exercise 6).

The World Wide Web

If your computer is connected to the internet, you can also use Python
facilities to access the content of any web page and use it as input to your
program. Web pages are written using markup languages like HTML and so
when you access the content of a web page you will get the content with the

116
 Sensing the World

markups included. In this section we will show you how to access the content
of a simple web page and print it out. Later we will see how you could use the
information contained in it to do further processing.

Go to a web browser and take a look at the web page:

http://www.fi.edu/weather/data/jan07.txt

This web page is hosted by the Franklin Institute of Philadelphia and contains
recorded daily weather data for Philadelphia for January 2007. You can
navigate from the above address to other pages on the site to look at daily
weather data for other dates (the data goes back to 1872!). Below, we will
show you how, using a Python library called urllib, you can easily access
the content of any web page. The urllib library provides a useful function
called urlopen using which you can access any web page on the internet as
follows:

>>> from urllib import *

>>> Data = urlopen("http://www.fi.edu/weather/data/jan07.txt")
>>> print Data.read()

January 2007
Day Max Min Liquid Snow Depth
1 57 44 1.8 0 0
2 49 40 0 0 0
3 52 35 0 0 0
… … … … … …
31 31 22 0 0 0
#days 31
Sum 1414 1005 4.18 1.80 1.10

The following two commands are important:

>>> Data = urlopen("http://www.fi.edu/weather/data/jan07.txt")

>>> print Data.read()

117
Chapter 5

The first command uses the function urlopen (which is imported from the
urllib) to establish a connection between your program and the web page.
The second command issues a read to read from that connection. Whatever is
read from that web page is printed out as a result of the print command.

A little more about Python functions

Before we move on, it would be good to take a little refresher on writing

Python commands/functions. In Chapter 2 we learned that the basic syntax for
defining new commands/functions is:

def <FUNCTION NAME>(<PARAMETERS>):

<SOMETHING>
...
<SOMETHING>

The Myro module provides you with several useful functions (forward,
turnRight, etc.) that enable easy control of your robot's basic behaviors.
Additionally, using the syntax above, you learned to combine these basic
behaviors into more complex behaviors (like wiggle, yoyo, etc.). By using
parameters you can further customize the behavior of functions by providing
different values for the parameters (for example, forward(1.0) will move the
robot faster than forward(0.5)). You should also note a crucial difference
between the movement commands like forward, turnLeft, and commands
that provide sensory data like getLight or getStall, etc. The sensory
commands always return a value whenever they are issued. That is:

>>> getLight('left')
221
>>> getStall()
0

Commands that return a value when they are invoked are called functions
since they actually behave much like mathematical functions. None of the
movement commands return any value, but they are useful in other ways. For
instance, they make the robot do something. In any program you typically

118
 Sensing the World

need both kinds of functions: those that do something but do not return
anything as a result; and those that do something and return a value. In Python
all functions actually return a value. You can already see the utility of having
these two kinds of functions from the examples you have seen so far.
Functions are an integral and critical part of any program and part of learning
to be a good programmer is to learn to recognize abstractions that can then be
packaged into individual functions (like drawPolygon, or degreeTurn) which
can be used over and over again.

Writing functions that return values

Python provides a return-statement that you can use inside a function to

return the results of a function. For example:

def triple(x):
# Returns x*3
return x * 3

The function above can be used just like the ones you have been using:

>>> triple(3)
9
>>> triple(5000)
15000

The general form of a return-statement is:

return <expression>

That is, the function in which this statement is encountered will return the
value of the <expression>. Thus, in the example above, the return-
statement returns the value of the expression 3*x, as shown in the example
invocations. By giving different values for the parameter x, the function
simply triples it. This is the idea we used in normalizing light sensor values in
the example earlier where we defined the function normalize to take in light
sensor values and normalize them to the range 0.0..1.0 relative to the observed
ambient light values:

119
Chapter 5

# This function normalizes light sensor values to 0.0..1.0

def normalize(v):
if v > Ambient:
v = Ambient

return 1.0 - v/Ambient

In defining the function above, we are also using a new Python statement: the
if-statement. This statement enables simple decision making inside computer
programs. The simplest form of the if-statement has the following structure:

if <CONDITION>:
<do something>
<do something>
...

That is, if the condition specified by <CONDITION> is True then whatever is

specified in the body of the if-statement is carried out. In case the
<condition> is False, all the statements under the if command are skipped
over.

Functions can have zero or more return-statements. Some of the functions

you have written, like wiggle do not have any. Technically, when a function
does not have any return statement that returns a value, the function returns a
special value called None. This is already defined in Python.

Functions, as you have seen, can be used to package useful computations and
can be used over and over again in many situations. Before we conclude this
section, let us give you another example of a function. Recall from Chapter 4
the robot behavior that enables the robot to go forward until it hits a wall. One
of the program fragments we used to specify this behavior is shown below:

while not getStall():

forward(1)
stop()

120
 Sensing the World

In the above example, we are using the value returned by getStall to help us
make the decision to continue going forward or stopping. We were fortunate
here that the value returned is directly usable in our decision making.
Sometimes, you have to do little interpretation of sensor values to figure out
what exactly the robot is sensing. You will see that in the case of light
sensors. Even though the above statements are easy to read, we can make
them even better, by writing a function called stuck() as follows:

def stuck():
# Is the robot stalled?
# Returns True if it is and False otherwise.

return getStall() == 1

The function above is simple enough, since getStall already gives us a

usable value (0/False or 1/True). But now if we were to use stuck to write
the robot behavior, it would read:

while not stuck():

forward(1)
stop()

As you can see, it reads much better. Programming is a lot like writing in this
sense. As in writing, there are several ways of expressing an idea in words.
Some are better and more readable than others. Some are downright poetic.
Similarly, in programming, expressing something in a program can be done in
many ways, some better and more readable than others. Programming is not
all about functionality, there can be poetry in the way you write a program.

Summary

In this chapter you have learned all about obtaining sensory data from the
robot’s perceptual system to do visual sensing (pictures), light sensing, and
proximity sensing. The Scribbler provides a rich set of sensors that can be
used to design interesting robot behaviors. You also learned that sensing is
equivalent to the basic input operation in a computer. You also learned how to

121
Chapter 5

get input from a game pad, the World Wide Web, and from data files.
Programs can be written to make creative use of the input modalities available
to define robot behaviors, computer games, and even processing data. In the
rest of the book you will learn how to write programs that make use of these
input modalities in many different ways.

Myro Review
getBright()
Returns a list containing the three values of all light sensors.
getBright(<POSITION>)
Returns the current value in the <POSITION> light sensor. <POSITION> can
either be one of 'left', 'center', 'right' or one of the numbers 0, 1, 2.

getGamepad(<device>)
getGamepadNow(<device>)
Returns the values indicating the status of the specified <device>. <device>
can be "axis" or "button". The getGamepad function waits for an event
before returning values. getGamepadNow immediately returns the current
status of the device.
getIR()
Returns a list containing the two values of all IR sesnors.

getIR(<POSITION>)
Returns the current value in the <POSITION> IR sensor. <POSITION> can
either be one of 'left' or 'right' or one of the numbers 0, 1.
getLight()
Returns a list containing the three values of all light sensors.
getLight(<POSITION>)
Returns the current value in the <POSITION> light sensor. <POSITION> can
either be one of 'left', 'center', 'right' or one of the numbers 0, 1, 2. The
positions 0, 1, and 2 correspond to the left, center, and right sensors.

122
 Sensing the World

getObstacle()
Returns a list containing the two values of all IR sesnors.
getObstacle(<POSITION>)
Returns the current value in the <POSITION> IR sensor. <POSITION> can
either be one of 'left', ‘center’, or 'right' or one of the numbers 0, 1, or 2.
savePicture(<picture>, <file>)
savePicture([<picture1>, <picture2>, …], <file>)
Saves the picture in the file specified. The extension of the file should be
“.gif” or “.jpg”. If the first parameter is a list of pictures, the file name
should have an extension “.gif” and an animated GIF file is created using
the pictures provided.

senses()
Displays Scribbler’s sensor values in a window. The display is updated every
second.

show(<picture>)
Displays the picture in a window. You can click the left mouse anywhere in
the window to display the (x, y) and (r, g, b) values of the point in the
window’s status bar.

takePicture()
takePicture(“color”)
takePicture(“gray”)
Takes a picture and returns a picture object. When no parameters are
specified, the picture is in color.

Python review

if <CONDITION>:
<statement-1>
...
<statement-N>
If the condition evaluates to True, all the statements are are performed.
Otherwise, all the statements are skipped.

123
Chapter 5

return <expression>
Can be used inside any function to return the result of the function.

<string>.split()
Splits <string> into a list.

urlopen(<URL>)
Establishes a stream connection with the <URL>. This function is to be
imported from the Python module urlopen.
<stream>.read()
Reads the entire contents of the <stream> as a string.

Lists:
[] is an empty list.

<list>[i]
Returns the ith element in the <list>. Indexing starts from 0.
<value> in <list>
Returns True if <value> is in the <list>, False otherwise.

<list1> + <list2>
Concatenates <list1> and <list2>.

len(<list>)
Returns the number of elements in a list.
range(N)
Returns a list of numbers from 0..N
range(N1, N2)
Returns a list of numbers starting from N1..(N2-1)

124
 Sensing the World

range(N1, N2, N3)

Returns a list of numbers starting from N1 and less than N3 incrementing by
N3.
<list>.sort()
Sorts the <list> in ascending order.
<list>.append(<value>)
Appends the <value> at the end of <list>.
<list>.reverse()
Reverses the elements in the list.

Exercises

1. The numbers assigned to the variable FamousNumbers in this chapter all

have names. Can you find them? What other famous numbers do you know?

2. Besides text, the speak command can also vocalize numbers. Try
speak(42) and also speak(3.1419). Try some really large whole numbers,
like speak(4537130980). How is it vocalized? Can you find out the limits of
numerical vocalization?

3. The Fluke camera returns pictures that are 256x192 (= 49,152) pixels.
Typical digital cameras are often characterized by the number of pixels they
use for images. For example, 8 megapixels. What is a megapixel? Do some
web search to find out the answer.

4. All images taken and saved using the Fluke camera can also be displayed in
a web page. In your favorite browser, use the Open File option (under the File
menu) and navigate to a saved picture from the Scribbler. Select it and view it
in the browser window. Try it with an animated GIF image file.

5. What does the following expression produce?

L = [5]*10

125
Chapter 5

Enter it as a Python command and then observe the value of L.

6. Modify the game pad input program from this chapter to make the axis
controller behave so that when the top of the axis is pressed, the robot goes
forward and when the bottom of the axis is pressed, it goes backward.

126
Sensing the World

127
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Insect-Like Behaviors

So you gotta let me know

Should I stay or should I go?
From the song, Should I stay or should I go, Mick Jones (The Clash), 1982.

Opposite page: Ladybug

Photo courtesy of Jon Sullivan (www.pdphoto.org)

129
Chapter 6

Designing robot behaviors is a challenging, yet fun process. There isn't a

formal methodology or a technique that one can follow. It involves creativity,
the ability to recognize the strengths and limitations of the physical robot, the
kind of environment the robot will be carrying out the behavior, and of course
the knowledge of available paradigms for programming robot behaviors.
Creativity is essential to the design of robot behaviors. You have already seen
how even a simple robot like the Scribbler can be programmed to carry out a
diverse range of behaviors. We have also spent a considerable effort so far in
exploring the range of possible functions a robot can perform. Where a robot
is placed when it is running can play an important role in exhibiting a
programmed behavior successfully. In this chapter, we take a different look at
robot behaviors.

Braitenberg Vehicles

In 1984, Valentino Braitenberg wrote a book titled: Vehicles: Experiments in

Synthetic Psychology (MIT Press). In it, he describes several thought
experiments that are centered around the creation of simple vehicles with very
simple control mechanisms that exhibit seemingly complex behavior. The
purpose of the thought experiments was to illustrate some fundamental
insights into the internal structure of animal (and human) brains. Each
experiment involves a descripion of a simple vehicle that is endowed with a
small suite of sensors (much like our Scribbler robot) and how the sensors can
be connected to the motors of these imaginary vehicles in ways that parallel
neurological connections in animals. He shows how the resulting vehicles are
capable of complex behaviors which can be described as: fear, aggression,
love, logic, free will, etc.

One central theme underlying Braitenberg's experiments is the demonstration

of what he calls the Law of uphill analysis and downhill invention: It is much
more difficult to guess the internal structure of an entity just by observing its
behavior than it is to actually create the structure that leads to the behavior.
That is, trying to postulate the internal structure purely by observing certain
behavior is an uphill (harder) task whereas trying to create an entity that
exhibits a certain behavior is a downhill (easy) task. While all of Braitenberg's

130
 Insect‐Like Behaviors

vehicles were imaginary and not really designed to be actually fabricated

people have found it a fun and intellectually interesting exercise to create
them. Personal robots like Scribblers make perfect platforms to do this and in
what follows we will describe some of Braitenberg's (and Braitenberg-type)
vehicles and design robot behaviors based on them.

Vehicle 1: Alive
Vehicle#1: Alive
The first vehicle Braitenberg
describes has one sensor and one
motor. The value transmitted by the
sensor directly feeds into the motor.
If the value being reported by the
sensor is a varying quantity (say
light), the vehicle will move at a
speed proportional to the amount of
quantity being detected by the sensor.

A schematic of the vehicle is shown above on the left. In order to design this
vehicle using the Scribbler, you can use the center light sensor and connect
what it reports directly to both motors of the robot. This is shown on the right.
That is, the same light reading is directly controlling both the motors by the
same amount. As you have already seen, there are a many different ways to
specify motor movement commands to the Scribbler. Suppose the value
obtained from the center light sensor is C, you can control both motors using
this value by using the command:

motors(C, C)

Alternately, you can also use the forward command:

forward(C)

Now that we know how the internal structure of this vehicle looks, we can
start to write a program that will implement it. But, before we get there, we
need to sort out a small issue of compatibility: light sensors report values in
131
Chapter 6

the range 0..5000 whereas motors and movement commands take values in
the range -1.0 to 1.0. In this example, we are only concerned with movements
that range from a complete stop to full speed forward, so the values range
from 0.0 to 1.0. We have to computationally normalize, or map the light
sensor values in this range. A first attempt at this would be to write a function
called normalize that operates as follows:

def normalize(v):
# normalizes v to range 0.0 to 1.0

Once we have this function, we can write the behavior for the vehicle as
follows:

def main():
# Braitenberg vehicle#1: Alive

while True:
L = getLight("center")
forward(normalize(L))

main()

Normalizing Sensor Values

It is time now to think about the task of the normalize function. Given a
value received from a light sensor, it has to transform it to a proportional
value between 0.0 and 1.0 so that the brighter the light, the higher the value
(i.e. closer to 1.0). Vehicle#1 moves in proportion to the amount of light it
receives. This is a good time to revisit the senses function of Myro to look at
the values reported by the light sensors. Go ahead and do this.

After examining the values returned by the light sensors you may notice that
they report small values (less than 50) for bright light and larger values (as
large as 3000) for darkness. In a way, you can say that the light sensor is
really a darkness sensor; the darker it is the higher the values reported by it.
The light sensors are capable of reporting values between 0 and 5000. Now,

132
 Insect‐Like Behaviors

we can certainly calibrate or normalize using these values using the following
definition of normalize:

def normalize(v):
# Normalize v (in the range 0..5000) to 0..1.0, inversely

return 1.0 - v/5000.0

That is, we divide the value of the light sensor by its maximum value and then
subtract that from 1.0 (for inverse proportionality). Thus a brighter light value,
say a value of 35, will get normalized as:

1.0 - 35.0/5000.0 = 0.9929

If 0.9929 is sent to the motors (as in the above program), the robot would
move full speed forward. Let us also compute the speed of the robot when it is
in total darkness. When you place a finger on the center sensor, you will get
values in the 2000-3000 range. For 3000, the normalization will be:

1.0 - 3000.0/5000.0 = 0.40

The robot will still be moving, although at nearly half the speed. Most likely,
you will be operating the robot in a room where there is sufficient ambient
light. You will notice that under ambient daylight conditions, the values
reported by the light sensors are in the 150-250 range. Using the above
normalization you will get:

1.0 - 200.0/5000.0 = 0.9599

That is almost full speed ahead. In order to experience the true behavior of the
above vehicle, we have to use a normalization scheme that takes into account
the ambient light conditions (they will vary from room to room). Further, let
us assume that in ambient light conditions, we will watch the robot respond to
a light source that we will control. A flashlight will work nicely. So, to make
the robot appropriately sensitive to the flashlight under ambient light
conditions you can write a better version of normalize as follows:

133
Chapter 6

def normalize(v):
if v > Ambient:
v = Ambient

return 1.0 - v/Ambient

That is, the darkest condition is represented by the ambient light value
(Ambient) and then normalization is done with respect to that value. You can
either set the ambient value by hand, or, a better way is to have the robot
sense its ambient light at the time the program is initiated. This is the same
version of normalize that you saw in the previous chapter. Now you know
how we arrived at it. The complete program for Vehicle#1 is shown below:

# Braitenberg Vehicle#1: Alive

from myro import *
initialize("com"+ask("What port?"))

Ambient = getLight("center")

def normalize(v):
if v > Ambient:
v = Ambient

return 1.0 - v/Ambient

def main():
# Braitenberg vehicle#1: Alive

while True:
L = getLight("center")
forward(normalize(L))

Do This: Implement the program above and observe the robot's behavior.
Does it respond as described above?

You may have also noticed by now that the three light sensors are not
necessarily closely matched. That is, they do not report exactly the same
values under the same conditions. When writing robot programs that use
multiple light sensors, it is a good idea to average the values returned by all
the light sensors to represent the ambient value. Modify the program above to
134
 Insect‐Like Behaviors

use the average of all three values as the ambient value. There shouldn't be a
noticeable difference in the robot's behavior. However, this is something you
may want to use in later programs.

Vehicle 2: Coward and Aggressive

Vehicle 2a: Coward

The next set of vehicles use two

sensors. Each sensor directly drives
one motor. Thus the speed of the
individual motor is directly
proportional to the quantity being
sensed by it sensor. There are two
possible ways to connect the sensors.
In the first case, Vehicle2a, the sensor
on each side connects to the motor on the same side. In the other case,
Vehicle2b, the connections are interchanged. That is, the left sensor connects
to the right motor and the right sensor connects to the left motor. Let us design
the control program for Vehicle 2a first:

# Vraitenberg Vehicle#2a
from myro import *
initialize("com"+ask("What port?"))

Ambient = sum(getLight())/3.0

def normalize(v):
if v > Ambient:
v = Ambient

return 1.0 - v/Ambient

def main():
# Braitenberg vehicle#2a: Coward

while True:
L = getLight("left")
R = getLight("right")
motors(normalize(L), normalize(R))

135
Chapter 6

The structure of the above program is very similar to that of Vehicle1. We

have modified the setting of the ambient light value to that of an average of
the three light values. Also, we use the motors command, to drive the left and
right motors proportional to the left and right light sensor values (after
normalizing).

Do This: Implement the control program for Vehicle 2a as shown above.

Observe the robot’s behaviors by shining the flashlight directly in front of the
robot and in each of the left and right sensors.

Next, write the control program for Vehicle 2b: Aggressive

Vehicle2b as shown here. This
requires one simple change from
the program of Vehicle2a: switch
the parameters of the motors
command to reflect the
interchanged connections. Again
observe the behaviors by shining
the flashlight directly ahead of the
robot and also a little to each side.

You will notice that the robots behave the same way when the light is placed
directly ahead of them: they are both attracted to light and hence move
towards the light source. However, Vehicle 2a will move away from the light
if the light source is on a side. Since the nearer sensor will get excited more,
moving the corresponding motor faster, and thereby turning the robot away.
In the case of Vehicle 2b, however, it will always turn towards the light
source and move towards it. Braitenberg calls these behaviors coward (2a)
and aggressive (2b).

Controlling Robot Responses

It is often necessary, when designing and testing robot behaviors, to properly

set up the robot's environment and the orientation of the robot in it. In simple
cases this is easily achieved by first placing the robot in the desired

136
 Insect‐Like Behaviors

orientation and then loading and executing the program. However prior to the
robot's actual behavior, the robot may need to perform some preliminary
observations (for example, sensing ambient light), it becomes necessary to re-
orient the robot properly before starting the execution of the actual behavior.
This can be easily accomplished by including some simple interactive
commands in the robot's program. The resulting program structure is shown
below:

# import myro library and establish connection with the robot

# define all functions here (like, normalize, etc.)
# set values of ambient conditions

def main():
# Description of the behavior...

# Give user the opportunity to set up the robot

askQuestion("Press OK to begin...", ["OK"])

# Write your robot's behavior commands here

Do This: Modify the programs for Vehicles 1, 2a, and 2b to include the
askquestion command above.

We have introduced a few basic programming patterns above that can be used
in many robot programming situations. The thing to remember is that, at any
point in the execution of a robot's program, you can also program appropriate
interjections to perform various experimental or control functions. We will
see several other examples later on.

Other Normalizations

All the normalizations of light sensor values shown above were used to
normalize the values in the range 0.0..1.0 in direct proportion to the amount of
light being sensed. That is, the darker it is, the closer the normalized values
are to 0.0 and the brighter it gets, the closer the normalized values get to 1.0.
This is just one way that one can relate the quantity being sensed to the
amount of speed applied to the robot's motors. You can imagine other

137
Chapter 6

relationships. The most obvious of course is an inverse relationship: the

darker it is the closer to 1.0 and vice versa. Braitenberg calls this inhibitory
(as opposed to excitatory) relationship: the more of a quantity being sensed,
the slower the robot's motors turn. As in Vehicles 2a and 2b above, there is
choice of two kinds of connections: straight and crossed. These are shown
below (a plus (+) sign next to a connector indicates an excitatory connection
and a minus sign (-) represents an inhibitory connection):

Writing the normalize function for an inhibitory connection is quite

straightforward:

def normalize(v): Vehicles 3a (Love) and 3b (Explorer)

if v > Ambient:
v = Ambient

return v/Ambient

Braitenberg describes the behavior

of the resulting vehicles as love
(Vehicle 3a) and explorer (Vehicle
3b). That is, if you were to observe
the behavior of the two vehicles,
you are likely to notice that Vehicle 3a will come to rest facing the light
source (in its vicinity) whereas vehicle 3b will come to rest turned away from
the source and may wander away depending on the presence of other light
sources.

In other variations on sensor value normalizations, Braitenberg suggests using

non-monotonic mathematical functions. That is, if you look at the excitatory
and inhibitory normalizations, they can be described as monotonic: more
light, faster motor speed; or more light, slower motor speed. But consider
other kinds of relationships for normalizations. Observe the function shown
on the next page. That is, the relationship is increasing in proportion to
sensory input but only up to a certain point and after that it decreases.
Incorporating such relationships in vehicles will lead to more complex

138
 Insect‐Like Behaviors

behavior (Braitenberg describes them as vehicles having instincts). The

following defines a normalization function, based on the curve shown:

A Non‐Monotonic Function

The above function is based on the

simpler function:

which in the first definition is stretched

to span the range 0..200 for values of x
with 100 being the point where it reports
the maximum value (i.e. 1.0). Mathematically this function is also known as
the bell curve or a Gaussian Curve in general. A bell curve is defined in terms
of a mean (π) and standard deviation (σ) as shown below:

Thus, in the normalization function we are using 100 as mean and 30 as

standard deviation. You can easily scale the curve for the range of sensor
values you desire using the following normalize function.

def normalize(v):
mean = Ambient/2.0
stddev = Ambient/6.0
if v >= Ambient:
v = Ambient
return exp(-(v - mean)**2 / 2*(stddev**2))

exp(x) is a Python function that computes the value of . It is available in

the Python math library. We will delve into the math library in more detail in
the next chapter. In order to use the exp function as shown above you have to
import the math library:

139
Chapter 6

from math import *

There are of course several other possibilities that one could try: a step
function; or a threshold; and any other mathematical combinations. The key
idea is that there is a clear mapping of the range of sensor values to motor
values in the range 0.0..1.0.

Robots using these normalizations and other variations are likely to exhibit
very interesting and sometimes unpredictable behaviors. Observers unaware
of the internal mapping mechanisms will have a hard time describing
precisely the robot's behavior and will tend to use anthropomorphic terms
(like, love, hate, instincts, etc.) to describe the behavior of robots. This is what
an uphill analysis means.

Multiple Sensors

Adding several sensors enriches the design space for robot behaviors. As a
designer, you now have a choice of different types of mappings: excitatory,
inhibitory, or more complex; and connections: straight or crossed. Suddenly
the resulting robot behavior will seem complex. On the Scribbler, for instance,
in addition to light sensors, you also have the stall sensor, and the IR sensors.
With the exception of light sensors, all of these other sensors are digital or
threshold sensors that are either ON or OFF (i.e. they report values that are
either 0 or 1 indicating the presence or absence or the thing they are sensing).
In a way you can think that the digital sensors are already normalized, but it is
still possible to invert the relationship if need be. You can design several
interesting behaviors by combining two or more sensors and deciding whether
to connect them straight or crossed.

Do This: In your design of vehicles 2a and 2b substitute the obstacle sensor in

place of the light sensors. Describe the behavior of the resulting vehicles. Try
the same for Vehicles 3a and 3b. Next, combine the behavior of the resulting
vehicles with the light sensors. Try out all combinations of connections, as
well as inhibitory and excitatory mappings. Which vehicles exhibit the most
interesting behaviors?

140
 Insect‐Like Behaviors

More Vehicles

Here are descriptions of several vehicles that are in the spirit of Braitenberg's
designs and also exhibit interesting behaviors. Using the concepts and
programming techniques from above, try to implement these on the Scribbler
robot. Once completed, you should invite some friends to observe the
behaviors of these creatures and record their reactions.

Timid

Timid is capable of moving forward in a straight line. It has one threshold

light sensor, pointing up. When the light sensor detects light, the creature
moves forward, otherwise, it stays still. The threshold of the light sensor
should be set to ambient light. That way, when the creature can "see" the
light, it will move. When it enters a shadow (which can be cast by a hand or
another object) it stops. If whatever is casting the shadow is moved, the
creature will move again. Therefore, timid is a shadow seeker.

Indecisive

Indecisive is similar to Timid, except, it never stops: its motors are always
running, either in forward direction, or in reverse direction, controlled by the
threshold light sensor. When the light sensor detects light, it moves forward,
otherwise, it moves backwards. When you run this creature, you will notice
that it tends to oscillate back and forth at shadow edges. Thus, Indecisive is a
shadow edge seeker.

Paranoid

Paranoid is capable of turning. This is accomplished by moving the right

motor forward and moving the left motor in reverse direction at the same
time. It has a single threshold light sensor. When the sensor detects light, it
moves forward. When the sensor enters a shadow, it reverses the direction of
its left motor, thus turning right. Soon the sensor will swing around, out of the

141
Chapter 6

shadow. When that happens, it resumes its forward motion. Paranoid, is a

shadow fearing creature.

This, That, or the Other

The if-statement introduced earlier in Chapter 5 is a way of making simple

decisions (also called one-way decisions). That is, you can conditionally
control the execution of a set of commands based on a single condition. The
if-statement in Python is quite versatile and can be used to make two-way or
even multi-way decisions. Here is how you would use it to choose among two
sets of commands:

if <condition>:
<this>
else:
<that>

That is, if the <condition> is true it will do the commands specified in

<this>. If, however, the <condition> is false, it will do <that>. Similarly,
you can extend the if-statement to help specify multiple options:

if <condition-1>:
<this>
elif <condition-2>:
<that>
elif <condition-3>:
<something else>
...
else:
<other>

Notice the use of the word elif (yes, it is spelled that way!) to designate "else
if". Thus, depending upon whichever condition is true, the corresponding
<this>, <that>, or <something else> will be carried out. If all else fails,
the <other> will be carried out.

142
 Insect‐Like Behaviors

Simple Reactive Behaviors

Using the three light sensors the robot can detect varying light conditions in
its environment. Let us write a robot program that makes it detect and orient
towards bright light. Recall from Chapter 5 that light sensors report low
values in bright light conditions and high values in low light. To accomplish
this task, we only need to look at the values reported by left and right light
sensors. The following describes the robot's behavior:

do for a given amount of time

if left light is brighter than right light
turn left
else
turn right

Thus, by making use of the if-else statement, we can refine the above into the
following:

while timeRemaining(30):
if left light is brighter that right light:
turnLeft(1.0)
else:
turnRight(1.0)

The only thing remaining in the commands above is to write the condition to
detect the difference between the two light sensors. This can be done using the
expression:

getLight('left') < getLight('right')

Do This: Write a complete program that implements the above behavior and
test it on your robot.

You may have noticed that even in uniform lighting conditions sensors tend to
report different values. It is generally a good idea to threshold the difference
when making the decision above. Say we set the threshold to a difference of at
least 50. That is, if the left and right sensors differ by at least 50 then turn

143
Chapter 6

towards the brighter sensor. What happens if the difference is less than the
threshold? Let us decide that in that case the robot will stay still. This
behavior can be captured by the following:

thresh = 50

while timeRemaining(30):
# Get sensor values for left and right light sensors
L = getLight('left')
R = getLight('right')

# decide how to act based on sensors values

if (L - R) > thresh:
# left is seeing less light than right so turn right
turnRight(1.0)
elif (R - L) > thresh:
# right is seeing less light than left, so turn left
turnLeft(1.0)
else:
# the difference is less than the threshold, stay put
stop()

Notice how we have used the variable thresh to represent the threshold
value. This is good programming practice. Since the performance of sensors
varies under different light conditions, this allows you to adjust the threshold
by simply changing that one value. By using the name thresh instead of a
fixed value, say 50, you only have to make such changes in one place of your
program.

In the statements above, there is a pattern that you will find recurring in many
programs that define robot behaviors using simple decisions:

while timeRemaing(<seconds>):
<sense>
<decide and then act>

Such behaviors are called reactive behaviors. That is, a robot is reacting to the
change in its environment by deciding how to act based on its sensor values.
A wide range of robot behaviors can be written using this program structure.

144
 Insect‐Like Behaviors

Below, we present descriptions of several interesting, yet simple automated

robot behaviors. Feel free to implement some of them on your robot.

Simple Reactive Behaviors

Most of the behaviors described below require selection among alternatives

using conditional expressions and if-statements.

Refrigerator Detective: As a child did you ever wonder if that refrigerator

light was always on? Or did it shut off when you closed the door? Well, here
is a way to find out. Build a refrigerator detective robot that sits inside the
fridge and tells you if the light is on or off!

Burglar Alarm Robot: Design a robot that watches your dorm door. As soon
as the door opens, it sounds an alarm (beeps).

Wall Detector: Write a robot program that goes straight and then stops when
it detects a wall in front. You will be using the IR sensors for this task.

Hallway Cruiser: Imagine your robot in

an environment that has a walled corridor
going around a 2 ft by 2 ft square box (see
picture on right). Write a program that will
enable the robot to go around this box. One
strategy you can use is to have the robot go
forward in a straight line until it bumps into
a wall. After a bump it will proceed to
make a 90 degree turn (you may need to
have it go backwards a little to enable
turning room) and then continue again in a
straight line.

Measuring Device: You have calibrated your robot with regards to how far it
travels in a given amount of time. You can use that to design a robot that

145
Chapter 6

measures space. Write a program that enables a robot to measure the width of
a hallway.

Follower: Write a robot program to exhibit the following behavior: The robot
prefers to stay near a wall (in front). If it does not have a wall in front of it, it
moves forward until it finds it. Test your program first by placing the robot in
a play pen. Ensure that your program behaves as described. Next, place it on a
floor and hold a blank piece of paper in front of it (close enough so the robot
can detect it). Now, slowly move the paper away from the robot. What
happens?

Designing Reactive Behaviors

Most of the robot behaviors that are implemented using the Braitenberg style
rely on a few simple things: selecting one or more sensors; choosing the kind
of wiring (straight or crossed); and selecting normalization functions for each
sensor. While you can guess the behavior that may result from these designs
the only way to confirm this is by actually watching the robot carry out the
behavior. You also saw how, using if-statements you can design simple, yet
interesting robot. In this section we will design additional reactive behaviors.

Light Following

To begin, it will be fairly straightforward to extend the behavior of the light

orienting behavior from above into one that results in a light follower robot.
That is, with a flashlight you will be able to guide the robot to follow you
around. Again, you have to start by observing the range of values reported by
the robot under various lighting conditions. If a flashlight is going to be the
bright light source, you will observe that the light sensors report very low
values when a light is shining directly on them (typically in the 0..50 range).
Thus, deciding which way to go (forward, turn left, or turn right) can be
decided based on the sensor readings from the three light sensors. The
structure of the program appears as follows:

146
 Insect‐Like Behaviors

# Light follower

from myro import *

initialize(ask("What port?"))

# program settings...

thresh = 50
fwdSpeed = 0.8
cruiseSpeed = 0.5
turnSpeed = 0.7 # left turn, -0.7 will be right turn

def main():
while True:
# get light sensor values for left, center, and right
L, C, R = getLight()

# decide how to act based on sensor values

if C < thresh:
# bright light from straight ahead, go forward
move(fwdSpeed, 0)
elif L < thresh:
# bright light at left, turn left
move(cruiseSpeed, turnSpeed)
elif R < thresh:
# bright light on right side, turn right
move(cruiseSpeed, -turnSpeed)
else:
# no bright light, move forward slowly (or stop?)
move(cruiseSpeed/2, 0)
main()

Notice that, in the program above, we have decided to set values for light
threshold (thresh) as well as movements to specific values. Also, in all cases,
we are using the move command to specify robot movement. This is because
the move command allows us to blend translation and rotation movement.
Additionally, notice that regardless of the sensor values, the robot is always
moving forward some amount even while turning. This is essential since the
robot has to follow the light and not just orient towards it. In the case where
there is no bright light present, the robot is still moving forward (at half the
cruise speed).

147
Chapter 6

Do This: Implement the light following program as described above and

observe the robot's behavior. Try adjusting the value settings (for threshold as
well as motor speeds) and note the changes in the robot's behaviors. Also, do
you observe that this behavior is similar to any of the Braitenberg vehicles
described above? Which one?

In the design of the light following robot above, we used a threshold value for
detecting the presence of bright light. Sometimes it is more interesting to use
differential thresholds for sensor values. That is, is the light sensor's value
different from the ambient light by a certain threshold amount? You can use
the senses function again observe the differences from ambient light and
modify the program above to use the differential instead of the fixed
threshold.

Here is another idea. Get several of your classmates together in a room with
their robots, all running the same program. Make sure the room has plenty of
floor space and a large window with a curtain. Draw close the curtains so the
outside light is temporarily blocked. Place the robots all over the room and
start the program. The robots will scurry around, cruising in the direction of
their initial orientation. Now, slowly draw the curtains open to let in more
light. What happens?

Avoiding Obstacles

Obstacles in the path of a robot can be detected using the IR sensors in front
of the robot. Then, based on the values obtained, the robot can decide to turn
away from an approaching obstacle using the following algorithm:

if obstacle straight ahead, turn (left or right?)

if obstacle on left, turn right
if obstacle on right, turn left
otherwise cruise

148
 Insect‐Like Behaviors

This can be implemented using the program below:

# Avoiding Obstacles

from myro import *

initialize(ask("What port?"))

# program settings...

cruiseSpeed = 0.6
turnSpeed = 0.5 # this is a left turn, -0.5 will be right
turn

def main():
while True:
# get sensor values for left and right IR sensors
L, R = getIR()
L = 1 - L
R = 1 - R

# decide how to act based on sensors values

if L and R:
# obstacle straight ahead, turn (randomly)
move(0, turnSpeed)
elif L:
# obstacle on left, turn right
move(cruiseSpeed, -turnSpeed)
elif R:
# obstacle on right, turn left
move(cruiseSpeed, turnSpeed)
else:
# no obstacles
move(cruiseSpeed, 0)
main()

As in the case of the light follower, observe that we begin by setting values
for movements. Additionally, we have flipped the values of the IR sensors so
that the conditions in the if-statements look more natural. Recall that the IR
sensors report a 1 value in the absence of any obstacle and a 0 in the presence
of one. By flipping them (using 1 - value) the value is 1 for an obstacle

149
Chapter 6

present and 0 otherwise. These values make it more natural to write the
conditions in the program above. Remember in Python, a 0 is equivalent to
False and a 1 is equivalent to True. Read the program above carefully and
make sure you understand these subtleties. Other than that, the program
structure is very similar to the light follower program.

Another way to write a similar robot behavior is to use the value of the stall
sensor. Recall that the stall sensor detects if the robots has bumped against
something. Thus, you can write a behavior that doesn't necessarily avoid
obstacles, but navigates itself around by bumping into things. This is very
similar to a person entering a dark room and then trying to feel their way by
touching or bumping slowly into things. In the case of the robot, there is no
way to tell if the bump was on its left or right. Nevertheless, if you use the
program (shown below) you will observe fairly robust behavior from the
robot.

# Avoiding Obstacles by bumping

from myro import *

initialize(ask("What port?"))

# program settings...

cruiseSpeed = 1.0
turnSpeed = 0.5 # this is a left turn, -0.5 will be right
turn

def main():
while True:

if getStall():
# I am stalled, turn (randomly?)
move(0, turnSpeed)
else:
# I am not stalled, cruise on
move(cruiseSpeed, 0)

main()

150
 Insect‐Like Behaviors

At times, you may notice that the robot gets stuck even when trying to turn.
One remedy for this is to stop the robot, back up a little, and then turn.

Do This: Implement the program above, observe the robot behavior. Next,
modify the program as suggested above (when stalled stop, backup, then
turn).

Maze Solver: Create a simple maze for your robot. Place the robot at one end
of the maze and use the obstacle avoidance programs from above (both
versions). Does you robot solve the maze? If not, note if your maze is right
handed or left handed (i.e. every turn is a right turn or left turn in the maze),
or both. Modify the obstacle avoidance programs to solve the right-handed,
left-handed mazes. How would you enable the robot to solve a maze that has
both right and left turns?

Corral Exiting
OK Corral?
Given that a simple obstacle
avoidance program can enable a
robot to solve simple mazes, we
can also design more interesting
behaviors on top of that. Imagine a
corral: an enclosed area with maze
like partitions and an entrance,
with a light source at the entrance
(see picture on right). Given the
robot's position, can we design a
behavior that will enable the robot
to exit the corral?

One can design a solution for the specific corral shown here: follow a wall
(any wall) until it sees bright light then switch to light seeking. Can the
Scribbler be designed to follow a wall? Remember the Fluke dongle has left
and right obstacle sensors that are pointing to its sides. Another approach will
be to combine the obstacle avoidance behavior from above with the light

151
Chapter 6

seeker behavior. That is, in the absence of any bright light, the robot moves
around the corral avoiding obstacles and when it sees a bright light, it heads
towards it. The hard part here will be to detect that it has exited the corral and
needs to stop.

Summary

Braitenberg uses very simple ideas to enable people to think about the way
animal and human brains and bodies are wired. For example, in humans, the
optic nerves (as do some others) have crossed connections inside the brain.
That is, the nerves from the left eye are connected to the right side of the brain
and vice versa. Actually they cross over and some information from either
side is also represented on the same side (that is there are straight as well as
crossed connections). However, it is still a puzzle among scientists as to why
this is the case and what, if any, are the advantages or disadvantages of this
scheme. Similarly, observing the behaviors of Vehicles 2a and 2b one can
easily see in them parallels in the behavior of several animals, like flies
orienting towards light/heat sources. Simple robot behaviors can provide deep
insights into complex behavior: that the observation and analysis of something
is an uphill task if one doesn't know the internal structure. And, by
constructing simple internal structures one can arrive at seemingly complex
behaviors. These seemingly complex behaviors have also been shown to
influence group behavior in insects (see the picture of article on next page).
That is, robots that do not look anything like insects, and not too different in
size than the Scribbler, can be used to influence insect behavior in many
situations.

In this chapter, we have attempted to give you a flavor for the idea of
synthetic psychology. At the same time you have also learned how to program
internal structures in a robot brain and learned several techniques for robot
control.

152
 Insect‐Like Behaviors

Story from Science Magazine, January 10, 2008

Background

All the numbered vehicles described here were developed in a set of thought
experiments designed by Valentino Braitenberg in his book, Vehicles:
Experiments in Synthetic Psychology, MIT Press, 1984.

Some of the other vehicles described here were designed by David Hogg,
Fred Martin, and Mitchel Resnick of the MIT Media Laboratory. Hogg et al

153
Chapter 6

used specialized electronic LEGO bricks to build these vehicles. For more
details, see their paper titled, Braitenberg Creatures.

To read more about robots influencing insect behavior see the November 16,
2007 issue of Science magazine. The primary article that is discussed in the
picture above is by Halloy et al, Social Integration of Robots into Groups of
Cockroaches to Control Self-Organized Choices, Science, November 16,
2007. Volume 318, pp 1155-1158.

Myro Review

There were no new Myro features introduced in this chapter.

Python Review

The if-statement in Python has the following forms:

if <condition>:
<this>

if <condition>:
<this>
else:
<that>

if <condition-1>:
<this>
elif <condition-2>:
<that>
elif <condition-3>:
<something else>
...
...
else:
<other>

The conditions can be any expression that results in a True, False, 1, or 0

value. Review Chapter 4 for details on writing conditional expressions.

154
 Insect‐Like Behaviors

Exercises

1. An even better way of averaging the ambient light conditions for purposes
of normalization is to have the robot sample ambient light all around it. That
is, turn around a full circle and sample the different light sensor values. The
ambient value can then be set to the average of all the light values. Write a
function called, setAmbient that rotates the robot for a full circle (or you
could use time), samples light sensor values as it rotates, and then returns the
average of all light values. Change the line:

Ambient = sum(getLight())/3.0

to the line:

Ambient = setAmbient()

Try out all of the earlier behaviors described in this chapter to see how this
new mechanism affects the robot's behavior.

2. Design and implement a program that exhibits the corral exiting behavior
described in this chapter.

3. Implement the refrigerator detective behavior described in this chapter.

4. Implement the Burglar alarm robot described in this chapter.

5. Implement the hallway cruiser behavior described in this chapter.

6. In addition to movements try to integrate music/sound output in your robot

behaviors and observe how the addition of sounds amplifies the perception of
the robot’s personality.

155
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Behavior Control

Oh, Behave!
Austin Powers (played by Mike Myers) in the movie
Austin Powers: International Man of Mystery,
New Line Cinema, 1997.

Opposite page: My Traffic Lights

Illustration by Yingshun Wong (yingshun.co.uk)

157
Chapter 7

Writing programs is all about exercising control. In the case of a robot’s brain
your program directs the operations of a robot. However, it is also important
to realize that the program itself is really controlling the computer. That is,
when you write Python programs you are controlling the computer that is then
communicating with the robot. Your program is directing the computer to
control the robot. If you take Myro out of the picture you are writing
programs to control the computer. This is the sense in which learning with
robots also leads to learning computing. Every program you write is doing
computation. Contrary to popular misconceptions computing is not just about
doing calculations with numbers. Controlling your robot is also computing, as
is predicting the world’s population, or composing an image, etc. This is one
aspect of control.

When writing robot control programs, the structure you use to organize the
program itself is a control strategy. Programming a robot is specifying
automated control. As a programmer or behavior designer you structure your
program to accomplish the goals of the behavior: how the sensors are used to
decide what to do next. This is another aspect of control. So far, you have
seen how to write control programs using Braitenberg style sensor-motor
wiring. You have also seen how to specify reactive control. These are
examples of two robot control paradigms.

In this chapter we delve further into the world of computation and robot
control paradigms. We will learn how to write robot control programs for
more complex and more robust robot tasks. We will also see how, using the
concepts learned so far, we can write useful and interesting computer
applications.

Behavior­based Control

When writing robot control programs, so far, you have used a very basic
technique in designing control programs:

158
 Behavior Control

def main():
# do forever or for some time
# or until a certain condition is satisfied

# sense and transform sensor values

# reason or decide what to do next
# do it

As you have seen, such a control program works well for many simple tasks.
However, you may have already run into many situations where, once the task
gets a little complex, it becomes difficult to structure a program in terms of a
single stream of control as shown above. For example, the corral exiting
behavior from the last chapter requires you to combine two simple behaviors:
solve a maze (avoid obstacles) and seek light to get out of the corral. As you
have seen before, it is fairly easy to program each of the individual behaviors:
obstacle avoidance; light following. But, when you combine these behaviors
to accomplish the corral exiting behavior two things happen: you are forced to
amalgamate the two control strategies into a single one and it may become
difficult to decide which way to combine them; additionally, the resulting
program is not very pretty and hard to read. In reality, hardly any robot
programs are written that way. In this section, we will look at a different way
of structuring robot programs that makes designing behaviors easy, and yet,
the resulting structure of the overall program is also clean and straightforward.
You can design some very sophisticated behaviors using these ideas.

People in the robotics community call the style of programming shown above
as reactive control or direct control. Also referred to as sensor fusion, the
resulting programs are purely sensor driven and hence appear to be too
bottom-up. That is, the values of the sensors drive the logic of control as
opposed to the goals of the robot tasks themselves. In behavior-based control
you get away from sensors and focus the design of your robot programs based
on the number and kinds of behaviors your robot has to carry out.

Let us look at how behavior-based control enables us to design the corral

exiting behavior. The robot essentially has to carry out three kinds of
behaviors: cruise (in the absence of any obstacles and/or light), avoid
obstacles (if present), and seek light (if present). In a behavior-based style of

159
Chapter 7

writing programs, you will define each of these behaviors as an individual

decision unit. Thus, each is quite simple and straightforward to write. Next,
the control program has to fuse the behaviors recommended by each
individual behavior unit. Look at the picture shown below:

In the diagram above, we have shown the three basic behaviors that we are
trying to combine: Cruise, Avoid, SeekLight. Each of these behaviors
outputs a triple: Yes/No, Translate Speed, Rotate Speed. A Yes implies that the
behavior module has a recommendation. No implies that it doesn't. That is, it
allows the possibility of a behavior having no recommendation. For example,
in the corral exiting situation, in the absence of a light source being sensed by
the robot, the SeekLight module will not have any recommendation. It then
becomes the task of the arbitrator (or the decision module) to decide which of
the available recommendations to use to drive the robot. Notice that in the end
to control the robot, all one has to do is decide how much to translate and
rotate. Many different arbitration schemes can be incorporated. We will use a
simple but effective one: Assign a priority to each behavior module. Then the
arbitrator always chooses the highest priority recommendation. This style of
control architecture is also called subsumption architecture. In the figure
above, we have actually drawn the modules in the order of their priority: the
higher the module is in the figure, the higher its priority. The lowest behavior,
cruise does not require any sensors and is always present: it wants the robot to
always go forward.

160
 Behavior Control

Arranging a control based on combining simple behaviors has several

advantages: you can design each individual behavior very easily; you can also
test each individual behavior by only adding that behavior and seeing how
well it performs; you can incrementally add any number of behaviors on top
of each other. In the scheme above, the control regime implies that the robot
will always cruise forward. But, if there is an obstacle present, it will override
the cruise behavior and hence try to avoid the obstacle. However, if there is a
light source detected, it will supersede all behaviors and engage in light
seeking behavior. Ideally, you can imagine that all the behaviors will be
running simultaneously (asynchronously). In that situation, the arbitrator will
always have one or more recommendations to adopt based on priority.

Let us develop the program that implements behavior-based control. First, we

define each behavior:

cruiseSpeed = 0.8
turnSpeed = 0.8
lightThresh = 80

def cruise():
# is always ON, just move forward
return [True, cruiseSpeed, 0]

def avoid():
# see if there are any obstacles
L, R = getIR()
L = 1 - L
R = 1 - R

if L:
return [True, 0, -turnSpeed]
elif R:
return [True, 0, turnSpeed]
else:
return [False, 0, 0]

161
Chapter 7

def seekLight():
L, C, R = getLight()

if L < lightThresh:
return [True, cruiseSpeed/2.0, turnSpeed]
elif R < lightThresh:
return [True, cruiseSpeed/2.0, -turnSpeed]
else:
return [False, 0, 0]

In the above, you can see that each individual behavior is simple and is easy
to read (and write). There are several ways to incorporate these into a
behavior-based program. Here is one:

# list of behaviors, ordered by priority (left is highest)

behaviors = [seekLight, avoid, cruise]

def main():

while True:
T, R = arbitrate()
move(T, R)

main()

The main program calls the arbitrate function that returns the chosen translate
and rotate commands which are then applied to the robot. The function
arbitrate is simple enough:

# Decide which behavior, in order of priority

# has a recommendation for the robot
def arbitrate():

for behavior in behaviors:

output, T, R = behavior()
if output:
return [T, R]

162
 Behavior Control

That is, it queries each behavior in order of priority to see if it has a

recommendation. If it does, that is the set of motor commands returned.

Do This: Implement the program above and see how well the robot behaves
in navigating around and exiting the corral. What happens if you change the
priority (ordering in the list) of behaviors? In writing the program above, we
have used two new Python features. We will review these next.

Names and Return values

In Chapter 3 you learned that in Python names can be used to represent

functions as well as numbers and strings. You also saw in Chapter 5 that lists,
and even pictures or images could be represented using names. A name is an
important programming concept. In Python a name can represent anything as
its value: a number, a picture, a function, etc. In the program above, when we
defined the variable behaviors as:

behaviors = [seekLight, avoid, cruise]

we used the names seekLight, avoid, and cruise to denote the functions
that they represented. We named these functions earlier in the program (using
def). Thus, the list named behaviors is a list of function names each of which
denote the actual function as its value. Next, look at the way we used the
variable behaviors in the function arbitrate:

for behavior in behaviors:

output, T, R = behavior()
…

Since behaviors is a list it can be used in the loop to represent the sequence
of objects (in this case functions). Thus in each iteration of the loop, the
variable behavior takes on successive values from this list: seekLight,
avoid, and cruise. When the value of the variable is seekLight, the function
seekLight is called in the statement:

output, T, R = behavior()

163
Chapter 7

There is no function named behavior() defined anywhere in the program.

However, since the value of the variable name behavior is set to one of the
functions in the list behaviors, the corresponding function is invoked.

The other new aspect of Python that we have used in the program above is the
fact that a function can return any object as its value. Thus the functions
cruise, avoid, seekLight, and arbitrate all return lists as their values.

Both of these are subtle features of Python. Many other programming

languages have restrictions on the kinds of things one can name as variables
and also on the kinds of values a function can return. Python gives uniform
treatment to all objects.

The Python Math Library

Most programming languages, Python included, provide a healthy selection of

libraries of useful functions so you do not have to write them. Such libraries
are often termed as an application programming interface or an API. Earlier
you were introduced to the random library provided by Python. It contains
several useful functions that provide facilities for generating random numbers.
Similarly, Python also provides a math library that provides often used
mathematical functions. In Chapter 6, we used the function exp from the math
library to normalize sensor values. Some of the commonly used functions in
the math library are listed below:

Commonly used functions in the math module

ceil(x) Returns the ceiling of x as a float, the smallest integer value greater
than or equal to x.

floor(x) Returns the floor of x as a float, the largest integer value less than
or equal to x.

exp(x) Returns .

164
 Behavior Control

log(x[, base]) Returns the logarithm of x to the given base. If the base is
not specified, return the natural logarithm of x (i.e., log ).

log10(x) Returns the base-10 logarithm of x (i.e. log .

pow(x, y) Returns .

sqrt(x) Returns the square root of x (√ ).

Some of the other functions available in the math module are listed at the end
of the chapter. In order to use any of these all you have to do is import the
math module:

import math

>>> math.ceil(5.34)
6.0
>>> math.floor(5.34)
5.0

>>> math.exp(3)
20.085536923187668

>>> math.log10(1000)
3.0
>>> math.log(1024, 2)
10.0
>>> math.pow(2, 10)
1024.0

>>> math.sqrt(7.0)
2.6457513110645907

Alternately, you can import the math module as shown below:

>>> from math import *

>>> ceil(5.34)
6.0
>>> floor(5.34)

165
Chapter 7

5.0
>>> exp(3)
20.085536923187668
>>> log10(1000)
3.0
>>> log(1024,2)
10.0
>>> sqrt(7.0)
2.6457513110645907

In Python, when you import a module using the command:

import <module>

You have to prefix all its commands with the module name (as the case in the
first set of examples above). We have also been using the form

from <module> import *

This imports all functions and other objects provided in the module which can
then be used without the prefix. You can also individually import specific
functions from a module using:

from <module> import <this>, <that>, …

Which version you use may depend on the context in which you use the
imported functions. You may run into a situation where two different modules
define functions with the same name but to do different things (or to do things
in a different way). In order to use both functions in the same module you will
have to use the module prefix to make clear which version you are using.

Doing Computations

Lets us weave our way back to traditional style computing for now. You will
see that the concepts you have learned so far will enable you to write lots of
different and more interesting computer applications. It will also give you a
clear sense of the structure of typical computer programs. Later, in Chapter

166
 Behavior Control

11, we will also return to the larger issue of the design of general computer
programs.

A Loan Calculator

Your current car, an adorable 1992 SAAB 93 was bought used and, for past
several months, you have had nothing but trouble keeping the car on the road.
Last night the ignition key broke off in the key slot when you were trying to
start it and now the broken piece would not come out (this used to happen a
lot with older SAAB's). The mechanic has to dismantle the entire ignition
assembly to get the broken key out and it could cost you upwards of $500.
The car's engine, which has done over 185,000 miles, has left you stranded on
the road many times. You have decided that this is it; you are going to go out
and get yourself a brand new reliable car. You have been moonlighting at a
restaurant to make extra money and have managed to save $5500.00 for
exactly this situation. You are now wondering what kind of new car you can
buy. Obviously, you will have to take a loan from a bank to finance the rest of
the cost but you are not sure how big a loan, and therefore what kind of car,
you can afford. You can write a small Python program to help you try out
various scenarios.

You can either dream (realistically) of the car you would like to buy, or you
can go to any of the helpful car buying sites on the web (www.edmunds.com
is a good place to start). Let us say that you have spent hours looking at
features and options and have finally narrowed your desires to a couple of
choices. Your first choice is going to cost you $22,000.00 and your second
choice is priced at $18,995.00. Now you have to decide which of these you
can actually afford to purchase.

First, you go talk to a couple of banks and also look at some loan offers on the
web. For example, go to bankrate.com and look for current rates for new car
loans.

Suppose the loan rates quoted to you are: 6.9% for 36 months, 7.25% for 48
months, and 7.15 for 60 months.

167
Chapter 7

You can see that there is a fair bit of variation in the rates. Given all this
information, you are now ready to write a program that can help you figure
out which of the two choices you may be able to make. In order to secure the
loan, you have to ensure that you have enough money to pay the local sales
tax (a 6% sales tax on a $20,000 car will add up to a hefty $1200!). After
paying the sales tax you can use the remainder of the money you have saved
up towards the down payment. The remainder of the money is the amount that
you would borrow. Depending on the type of loan you choose, your monthly
payments and how long you will make those payments will vary. There is a
simple formula that you can use to estimate your monthly payment:

Whoa! That seems complicated. However, given the formula, you can see that
it really requires two mathematical functions: log and , both of which
are available in the Python math module. Suddenly, the problem seems not
that hard.

Let us try and outline the steps needed to write the program: First, note the
cost of the car, the amount of money you have saved, and the sales tax rate
Also, note the financials: the interest rate, and the term of the loan. The
interest rate quoted is generally the annual percentage rate (APR) convert it to
monthly rate (by dividing it by 12). Next, compute the sales tax you will pay.
Use the money left to make a down payment. Then determine the amount you
will borrow. Plug in all of the values in the formula and compute the monthly
payment. Also, compute the total cost of the car. Output all the results. Next,
we can take each of the above steps and start to encode them into a program.
Here is a first order refinement:

def main():
# First, note the cost of the car (Cost),
# the amount of money you have saved (Cash),
# and the sales tax rate (TaxRate)

# Also, note the financials: the interest rate (APR),

# and the term of the loan (Term)

168
 Behavior Control

# The interest rate quoted is generally the annual

# percentage rate (APR)
# Convert it to monthly rate (by dividing it by 12) (MR)

# Next, compute the sales tax you will pay (SalesTax)

# Use the money left to make a down payment (DownPayment)
# Then determine the amount you will borrow (LoanAmount)

# Plug in all of the values in the formula and compute

# the monthly payment (MP)

# Also, compute the total cost of the car. (TotalCost)

# Output all the results

main()

Above, we have taken the steps and converted them into a skeletal Python
program. All the steps are converted to Python comments and where needed,
we have decided the names of variables that will hold the values that will be
needed for the calculations. This is useful because this also helps determine
how the formula will be encoded and also helps determine what values can be
programmed in and which ones you will have to supply as input. Making the
program require inputs will easily enable you to enter the different parameters
and then based on the outputs you get, you can decide which car to buy. Let
us encode all the inputs first:

def main():
# First, note the cost of the car (Cost),
Cost = input("Enter the cost of the car: $")

# the amount of money you have saved (Cash),

Cash = input("Enter the amount of money you saved: $")

# and the sales tax rate (TaxRate) (6% e.g.)

SalesTaxRate = 6.0

# Also, note the financials: the interest rate (APR),

# and the term of the loan (Term)
# The interest rate quoted is generally the annual
# percentage rate (APR)

169
Chapter 7

APR = input("Enter the APR for the loan (in %): ")

# Convert it to monthly rate (by dividing it by 12) (MR)

# Next, compute the sales tax you will pay (SalesTax)

# Use the money left to make a down payment (DownPayment)
# Then determine the amount you will borrow (LoanAmount)

# Plug in all of the values in the formula and compute

# the monthly payment (MP)

# Also, compute the total cost of the car. (TotalCost)

# Output all the results

main()

We have refined the program to include the inputs that will be needed for each
run of the program. Notice that we chose not to input the sales tax rate and
instead just assigned it to the variable SalesTaxRate. If you wanted, you
could also have that be entered as input. What you choose to have as input to
your program is your design decision. Sometimes the problem may be framed
so it explicitly specifies the inputs; sometimes you have to figure that out. In
general, whatever you need to make your program more versatile is what you
have to base your decisions on. For instance, fixing the sales tax rate to 6.0
will make the program usable only in places where that rate applies. If, for
example, you wanted your friend in another part of the country to use the
program, you should choose to make that also an input value. Let us go on to
the next steps in the program and encode them in Python. These are mainly
computations. The first few are simple. Perhaps the most complicated
computation to encode is the formula for computing the monthly payment. All
of these are shown in the version below.

From math import *

def main():
# First, note the cost of the car (Cost),
Cost = input("Enter the cost of the car: $")

# the amount of money you have saved (Cash),

170
 Behavior Control

Cash = input("Enter the amount of money you saved: $")

# and the sales tax rate (TaxRate) (6% e.g.)

SalesTaxRate = 6.0

# Also, note the financials: the interest rate (APR),

# and the term of the loan (Term)
# The interest rate quoted is generally the annual
# percentage rate (APR)
APR = input("Enter the APR for the loan (in %): ")

# Input the term of the loan (Term)

term = input("Enter length of loan term (in months): ")

# Convert it (APR) to monthly rate (divide it by 12) (MR)

# also divide it by 100 since the value input is in %
MR = APR/12.0/100.0

# Next, compute the sales tax you will pay (SalesTax)

SalesTax = Cost * SalesTaxRate / 100.0

# Use the money left to make a down payment (DownPayment)

DownPayment = Cash - SalesTax

# Then determine the amount you will borrow (LoanAmount)

LoanAmount = Cost - DownPayment

# Plug in all of the values in the formula and compute

# the monthly payment (MP)
MR = (LoanAmount * MR) / (1.0 - exp(-term * log(1.0+MR)))

# Also, compute the total cost of the car. (TotalCost)

TotalCost = SalesTax + DownPayment + MR * term

# Output all the results

print "Here are the details about your new car..."
print "------------------------------------------"
print
print "Money you have saved $", Cash
print "Cost of the car $", Cost
print "Sales Tax rate is", SalesTaxRate, "%"
print "Sales Tax on the car $", SalesTax
print "Your down payment will be $", DownPayment

171
Chapter 7

print "You will be borrowing $", LoanAmount

print "A", term, "month loan at", APR, "% APR"
print "Your monthly payment will be $", MP
print "Total cost will be $", TotalCost
print

main()

Do This: When you enter the above program and run it in Python, you can
enter the data about your car. Here is a sample run:

Enter the cost of the car: $20000.00

Enter the amount of money you saved: $5500.00
Enter the APR for the loan (in %): 6.9
Enter the length of loan term (in months): 36
Here are the details about your new car...
------------------------------------------

Money you have saved $ 5500.0

Cost of the car $ 20000.0
Sales Tax rate is 6.0 %
Sales Tax on the car $ 1200.0
Your down payment will be $ 4300.0
You will be borrowing $ 15700.0
A 36 month loan at 6.9 % APR
Your monthly payment will be $ 484.052914723
Total cost will be $ 22925.90493

It appears that at for the $20000.00 car, for a 36 month 6.9% loan you will
end up paying $484.05 (or $484.06 depending upon how your loan company
round pennies!).

When you need to restrict your output values to specific decimal places (two
in the case of dollars and cents, for example), you can use the string
formatting features built into Python. For example, in a string, you can specify
how to include a floating point value as follows:

"Value of PI %5.3f in 3 places after the decimal" % (math.pi)

The above is a Python expression that has the syntax:

172
 Behavior Control

<string> % <expression>

Inside the <string> there is a format specification beginning with a %-sign

and ending in an f. What follows the %-sign is a numerical specification of the
value to be inserted at that point in the string. The f in the specification refers
to the fact it is for a floating point value. Between the %-sign and the f is a
number, 5.3. This specifies that the floating point number to be inserted will
take up at least 5 spaces, 3 of which will be after the decimal. One of the
spaces is always occupied by the decimal. Thus, the specification %5.3f
specifies a value to be inserted in the following manner:

"This is the value of PI -.--- expressed in three places after

the decimal"

You can see the result of executing this in Python below:

>>> "Value of PI %5.3f in 3 places after the decimal" %

(math.pi)
'Value of PI 3.142 in 3 places after the decimal'

>>> "Value of PI %7.4f in 4 places after the decimal" %

(math.pi)
'Value of PI 3.1416 in 4 places after the decimal'

In the second example above, we replaced the specification with %7.4f.

Notice that the resulting string allocates seven spaces to print that value. If
there are more spaces than needed they get padded by blanks on the leading
edge (notice the extra space before 3 in the second example above). If the
space specified is less, it will always be expanded to accommodate the
number. For example:

>>> "Value of PI %1.4f in 4 places after the decimal" %

(math.pi)
'Value of PI 3.1416 in 4 places after the decimal'

We deliberately specified that the value be 1 space wide with 4 spaces after
the decimal (i.e. %1.4f). As you can see, the space was expanded to

173
Chapter 7

accommodate the value. What is assured is that the value is always printed
using the exact number of spaces after the decimal. Here is another example:

>>> "5 is also %1.3f with 3 places after the decimal." % 5

'5 is also 5.000 with 3 places after the decimal.'

Thus, the value is printed as 5.000 (i.e. the three places after the decimal are
always considered relevant in a specification like %1.3f). Similarly, for
specifying whole number or integer values you can use the letter-d, and for
strings you can use the letter-s:

>>> "Hello %10s, how are you?" % "Arnold"

'Hello Arnold, how are you?'

By default, longer specifications are right-justified. You can use a %-

specification to left-justify. For example:

>>> "Hello %-10s, how are you?" % "Arnold"

'Hello Arnold , how are you?'

Having such control over printed values is important when you are trying to
output tables of aligned values. Let us modify our program from above to use
these formatting features:

from math import *

def main():
# First, note the cost of the car (Cost),
Cost = input("Enter the cost of the car: $")

# the amount of money you have saved (Cash),

Cash = input("Enter the amount of money you saved: $")

# and the sales tax rate (TaxRate) (6% e.g.)

SalesTaxRate = 6.0

# Also, note the financials: the interest rate (APR),

# and the term of the loan (Term)

174
 Behavior Control

# The interest rate quoted is generally the annual

# percentage rate (APR)
APR = input("Enter the APR for the loan (in %): ")

# and the term of the loan (Term)

term = input("Enter length of loan term (in months): ")

# Convert it (APR) to monthly rate (divide it by 12) (MR)

# also divide it by 100 since the value input is in %
MR = APR/12.0/100.0

# Next, compute the sales tax you will pay (SalesTax)

SalesTax = Cost * SalesTaxRate / 100.0

# Use the money left to make a down payment (DownPayment)

DownPayment = Cash - SalesTax

# Then determine the amount you will borrow (LoanAmount)

LoanAmount = Cost - DownPayment

# Plug in all of the values in the formula and compute

# the monthly payment (MP)
MP = (LoanAmount * MR) / (1.0 - exp(-term * log(1.0+MR)))

# Also, compute the total cost of the car. (TotalCost)

TotalCost = SalesTax + DownPayment + MP * term

# Output all the results

print "Here are the details about your new car..."
print "------------------------------------------"
print
print "Money you have saved $%1.2f" % Cash
print "Cost of the car $%1.2f" % Cost
print "Sales Tax rate is %1.2f" % SalesTaxRate
print "Sales Tax on the car $", SalesTax, "%"
print "Your down payment will be $%1.2f" % DownPayment
print "You will be borrowing $%1.2f" % LoanAmount
print "A %2d month loan at %1.2f APR" % (term, APR)
print "Your monthly payment will be $%1.2f" % MP
print "Total cost will be $%1.2f" % TotalCost
print

main()

175
Chapter 7

When you run it again (say for a slightly different loan term), you get:

Enter the cost of the car: $20000.00

Enter the amount of money you saved: $5500.00
Enter the APR for the loan (in %): 7.25
Enter the length of loan term (in months): 48
Here are the details about your new car...
------------------------------------------

Money you have saved $5500.00

Cost of the car $20000.00
Sales Tax rate is 6.00
Sales Tax on the car $ 1200.0 %
Your down payment will be $4300.00
You will be borrowing $15700.00
A 48 month loan at 7.25 APR
Your monthly payment will be $377.78
Total cost will be $23633.43

You can see that for the same amount if you borrow it for a longer period you
can reduce your monthly payments by over $100 (but you pay about $700
more in the end).

Decision Making in Computer Programs

Decision making is central to all computer programs. In fact there is a famous

theorem in computer science that says that if any programmable device is
capable of being programmed to do sequential execution, decision making,
and repetition, it is capable of expressing any computable algorithm. This is a
very powerful idea that is couched in terms of very simple ideas. Given a
problem, if you can describe its solution in terms of a combination of the three
execution models then you can get any computer to solve that problem. In this
section, we will look at decision making in a classic game playing situation.

Computers have been extensively used in game playing: for playing games
against people and other computers. The impact of computers on game
playing is so tremendous that these days several manufacturers make game

176
 Behavior Control

playing consoles with computers in

them that are completely dedicated to
game playing.

Let us design a simple game that you

have most likely played while
growing up: Paper, Scissors, Rock!
The exact name of the game can
In this game, two players play against vary, with the three components
each other. At the count of three each appearing in a different order, or
player makes a hand gesture with “stone” in place of “rock”.
indicating that they have selected one Non‐English speakers may know
of the three items: paper (the hand is the game by their local words for
held out flat), scissors (two fingers are “rock, paper, scissors”, although
extended to designate scissors, or rock it is also known as Jankenpon in
(indicated by making a fist). The rules Japan, Rochambeau in France,
of deciding the winner are simple: if and in South Africa as Ching‐
both players pick the same object it is Chong‐Cha.. .
a draw; otherwise, paper beats rock, From:talkingtotan.wordpress.com/2007/08/
scissors beat paper, and rock beats
scissors. Let us write a computer program to pay this game against the
computer. Here is an outline for playing this game once:

# Computer and player make a selection

# Decide outcome of selections (draw or who won)
# Inform player of decision

If we represent the three items in a list, we can have the computer pick one of
them at random by using the random number generation facilities provided in
Python. If the items are represented as:

items = ["Paper", "Scissors", "Rock"]

Then we can select any of the items above as the computer's choice using the
expression:

177
Chapter 7

# Computer makes a selection

myChoice = items[<0 or 1 or 2 selected randomly>]

That is items[0] represents the choice "Paper", items[1] represents

"Scissors", and items[2] is "Rock". We saw, in Chapter 4, that we can
generate a random number in any range using the randint function from the
random library module in Python. Thus we can model the computer making a
random selection using the following statement:

from random import *

# Computer makes a selection

myChoice = items[randint(0,2)]

Recall that randint(n, m) returns a random numbers in the range [n..m].

Thus, randint(0, 2) will return either 0, 1, or 2. We can use the
askQuestion command in Pyro to ask the player to indicate their selection
(see Chapter 3).

# Player makes a selection

yourChoice = askQuestion("Pick an item by clicking on it.",
items)

Now that we know how to the computer and player make their selection, we
need to think about deciding the outcome. Here is an outline:

if both picked the same item then it is a draw

if computer wins then inform the player
if player wins then inform the player

Rewriting the above using if-statements we get the following first draft:

if myChoice == yourChoice:
print "It is a draw."
if <myChoice beats yourChoice>:
print "I win."
else:
print "You win."

178
 Behavior Control

All we need to do is figure out how to write the condition <myChoice beats
yourChoice>. The condition has to capture the rules of the game mentioned
above. We can encode all the rules in a conditional expression as follows:

if (myChoice == "Paper" and yourChoice == "Rock")

or (myChoice == "Scissors" and yourChoice == "Paper")
or (myChoice == "Rock" and yourChoice == "Scissors"):
print "I win."
else:
print "You win."

The conditional expression above captures all of the possibilities that should
be examined in order to make the decision. Another way of writing the above
decision would be to use the following:

if myChoice == "Paper" and yourChoice == "Rock":

print "I win."
elif myChoice == "Scissors" and yourChoice == "Paper":
print "I win."
elif myChoice == "Rock" and yourChoice == "Scissors":
print "I win."
else:
print "You win."

That is each condition is examined in turn until one is found that confirms that
the computer wins. If none such condition is true, the else-part of the if-
statement will be reached to indicate that the player won.

Another alternative to writing the decision above is to encode the decision in a

function. Let us assume that there is a function beats that returns True or
False depending on the choices. We could then rewrite the above as follows:

if myChoice == yourChoice:
print "It is a draw."
if beats(myChoice, yourChoice):
print "I win."
else:
print "You win."

179
Chapter 7

Let us take a closer look at how we could define the beats function. It needs
to return True if myChoice beats yourChoice. So all we need to do is encode
the rules of the game described above. Here is a draft of the function:

def beats(me, you):

# Does me beat you? If so, return True, False otherwise.

if me == "Paper" and you == "Rock":

# Paper beats rock
return True
elif me == "Scissors" and you == "Paper":
# Scissors beat paper
return True
elif me == "Rock" and you == "Scissors":
# Rock beats scissors
return True
else:
return False

Once again, we have used the if-statements in Python to encode the rules of
the game. Now that we have completely fleshed out all the critical parts of the
program, we can put them all together as shown below:

# A program that plays the game of Paper, Scissors, Rock!

from myro import *
from random import randrange

def beats(me, you):

# Does me beat you? If so, return True, False otherwise.

if me == "Paper" and you == "Rock":

# Paper beats rock
return True
elif me == "Scissors" and you == "Paper":
# Scissors beat paper
return True
elif me == "Rock" and you == "Scissors":
# Rock beats scissors
return True
else:
return False

180
 Behavior Control

def main():
# Play a round of Paper, Scissors, Rock!
print "Lets play Paper, Scissors, Rock!"
print "In the window that pops up, make your selection>"

items = ["Paper", "Scissors", "Rock"]

# Computer and Player make their selection...

# Computer makes a selection
myChoice = items[randrange(0, 3)]

# Player makes a selection

yourChoice = askQuestion("Pick an item.", items)

# inform Player of choices

print
print "I picked", myChoice
print "You picked", yourChoice

# Decide if it is a draw or a win for someone

if myChoice == yourChoice:
print "We both picked the same thing."
print "It is a draw."
elif beats(myChoice, yourChoice):
print "Since", myChoice, "beats", yourChoice, "..."
print "I win."
else:
print "Since", yourChoice, "beats", myChoice, "..."
print "You win."

print "Thank you for playing. Bye!"

main()

A few more print commands were added to make the interaction more natural.

Do This: Implement the Paper, Scissors, Rock program from above and play
it several times to make sure you understand it completely. Modify the
program above to play several rounds. Also, incorporate a scoring system that

181
Chapter 7

keeps track of the number of times each player won and also the number of
draws.

Summary

In this chapter you have seen a variety of control paradigms: behavior-based

control for robots, writing a simple, yet useful computational program, and
writing a simple computer game. Behavior-based control is a powerful and
effective way of describing sophisticated robot control programs. This is the
paradigm used in many commercial robot applications. You should try out
some of the exercises suggested below to get comfortable with this control
technique. The other two programs illustrate how, using the concepts you
have learned, you can design other useful computer applications. In the next
several chapters, we will explore several other dimensions of computing:
media computation, writing games, etc.

Myro Review

There was nothing new from Myro in this chapter.

Python Review

The math library module provides several useful mathematics functions.

Some of the commonly used functions are listed below:

ceil(x) Returns the ceiling of x as a float, the smallest integer value greater
than or equal to x.

floor(x) Returns the floor of x as a float, the largest integer value less than
or equal to x.

exp(x) Returns .

182
 Behavior Control

log(x[, base]) Returns the logarithm of x to the given base. If the base is
not specified, return the natural logarithm of x (i.e., log ).

log10(x) Returns the base-10 logarithm of x (i.e. log .

pow(x, y) Returns .

sqrt(x) Returns the square root of x (√ ).

Trigonometric functions

acos(x) Returns the arc cosine of x, in radians.

asin(x) Returns the arc sine of x, in radians.

atan(x) Returns the arc tangent of x, in radians.

cos(x) Returns the cosine of x radians.

sin(x) Returns the sine of x radians.

tan(x) Returns the tangent of x radians.

degrees(x) Converts angle x from radians to degrees.

radians(x) Converts angle x from degrees to radians.

The module also defines two mathematical constants:

pi The mathematical constant π.

e The mathematical constant e.

183
Chapter 7

Exercises

1. Replace the Avoid module in the behavior-based control program with a

module called: follow. The follow module tries to detect a wall to the robot’s
right (or left) and then tries to follow it at all times. Observe the resulting
behavior.

2. Do Population growth by modeling it as a difference equation.

3. Do population growth by modeling it as a continuous growth function.

4. Instead of using the statement:

yourChoice = items[randint(0,2)]

use the command:

yourChoice = choice(items)

choice(<list>) is another function defined in the random module. It

randomly selects an element from the supplied list.

5. Can you make the program better?

6. Implement the HiLo game: the computer picks a random number between 0
and 1000 and the user has to try and guess it. With every guess the computer
informs the user f their guess was high (Hi), low (Lo), or they got it.

7. Reverse the roles in the HiLo game from above. This time, the user guesses
a number between 0 and 1000 and computer program has to guess it. Think
about a good strategy for guessing the number based on high and low clues,
and then implement that strategy.

184
Behavior Control

185
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Sights & Sounds

Don't make music for some vast, unseen audience or market or ratings share
or even for something as tangible as money. Though it's crucial to make a
living, that shouldn't be your inspiration. Do it for yourself.
-Billy Joel

Opposite page: Be My Valentine

Fractal Art by Sven Geier (www.sgeier.net)

187
Chapter 8

We mentioned earlier that the notion of computation these days extends far
beyond simple numerical calculations. Writing robot control programs is
computation, as is making world population projections. Using devices like
iPods you are able to enjoy music, videos, and radio and television shows.
Manipulating sounds and images is also the realm of computation and in this
chapter we will introduce you to these. You have already seen how, using
your robot, you can take pictures of various scenes. You can also take similar
images from your digital camera. Using basic computational techniques you
have learned so far you will see, in this chapter, how you can do computation
on shapes and sound. You will learn to create images using computation. The
images you create can be used for visualizing data or even for aesthetic
purposes to explore your creative side. We will also present some
fundamentals of sound and music and show you how you can also do similar
forms of computation using music.

Sights: Drawing

If you have used a computer for any amount of time you must have used some
kind of a drawing application. Using a typical drawing application you can
draw various shapes, color them etc. You can also generate drawings using
drawing commands provided in the Myro library module. In order to draw
anything you first need a place to draw it: a canvas or a window. You can
create such a window using the command:

myCanvas = GraphWin()

If you entered the command above in IDLE, you will immediately see a small
gray window pop up (see picture on right). This window will be serving as
our canvas for creating drawings. By default, the window created by the
GraphWin command is 200 pixels high and 200 pixels wide and its name is
“Graphics Window”. Not a very inspiring way to start, but the GraphWin
command has some other variations as well. First, in order to make the
window go away, you can use the command:

188
 Sights & Sounds

myCanvas.close() A Graphics window

To create a graphics window of any size and

a name that you specify, you can use the
command below:

myCanvas = GraphWin(“My
Masterpiece”, 200, 300)

The command above creates a window

named “My Masterpiece” that will be 200
pixels wide and 300 pixels tall (see picture on
right). You can change the background color
of a graphics window as shown below:

myCanvas.setBackground(“white”) My Masterpiece 200x300

You can name any of a number of colors in the

command above, ranging from mundane ones
like “red”, “blue”, “gray”, “yellow”, to
more exotic colors ranging from
“AntiqueWhite” to “LavenderBlush” to
“WhiteSmoke”. Colors can be created in many
ways as we will see below. Several thousand
color names have been pre-assigned (Google:
color names list) that can be used in the
command above.

Now that you know how to create a canvas

(and make it go away) and even set a
background color, it is time to look at what we can draw in it. You can create
and draw all kinds of geometrical objects: points, lines, circles, rectangle, and
even text and images. Depending on the type of object you wish to draw, you
have to first create it and then draw it. Before you can do this though, you
should also know the coordinate system of the graphics window.

189
Chapter 8

In a graphics window with width, W and height H (i.e WxH pixels) the pixel
(0, 0) is at the top left corner and the pixel (199, 299) will be at the bottom
right corner. That is, x-coordinates increase as you go right and y-coordinates
increase as you go down.

The simplest object that you can create is a point. This is done as follows:

p = Point(100, 50)

That is, p is an object that is a Point whose x-coordinate is at 100 and y-

coordinate is at 50. This only creates a Point object. In order to draw it, you
have to issue the command:

p.draw(myCanvas)

The syntax of the above command may seem a little strange at first. You saw
it briefly when we presented lists in Chapter 5 but we didn’t dwell on it.
Certainly it is different from what you have seen so far. But if you think about
the objects you have seen so far: numbers, strings, etc. Most of them have
standard operations defined on them (like +, *, /, etc.). But when you think
about geometrical objects, there is no standard notation. Programming
languages like Python provide facilities for modeling any kind of object and
the syntax we are using here is standard syntax that can be applied to all kinds
of objects. The general form of commands issued on objects is:

<object>.<function>(<parameters>)

Thus, in the example above, <object> is the name p which was earlier
defined to be a Point object, <function> is draw, and <parameters> is
myCanvas. The draw function requires the graphics window as the parameter.
That is, you are asking the point represented by p to be drawn in the window
specified as its parameter. The Point objects have other functions available:

>>> p.getX()
100

190
 Sights & Sounds

>>> p.getY()
50

That is, given a Point object, you can get its x- and y-coordinates. Objects are
created using their constructors like the Point(x, y) constructor above. We
will use lots of constructors in this section for creating the graphics objects. A
line object can be created similar to point objects. A line requires the two end
points to be specified. Thus a line from (0, 0)
to (100, 200) can be created as:

L = Line(Point(0,0), Point(100,200))

And you can draw the line using the same

draw command as above:

L.draw(myCanvas)

The picture on the right shows the two objects

we have created and drawn so far. As for
Point, you can get the values of a line’s end
points:

>>> start = L.getP1()

>>> start.getX
0
>>> end = L.getP2()
>>> end.getY()
200

Here is a small Python loop that can be used to create and draw several lines:

for n in range(0, 200, 5):

L=Line(Point(n,25),Point(100,100))
L.draw(myCanvas)

In the loop above (the results are shown on the

right), the value of n starts at 0 and increases by
5 after each iteration all the way up to but not including 200 (i.e. 195). For
191
Chapter 8

each value of n a new Line object is created with starting co-ordinates (n,
25) and end point at (100, 100).

Do This: Try out all the commands introduced so far. Then observe the
effects produced by the loop above. Change the increment 5 in the loop above
to different values (1, 3, etc.) and observe the effect. Next, try out the
following loop:

for n in range(0, 200, 5):

L = Line(Point(n, 25), Point(100, 100))
L.draw(myCanvas)
wait(0.3)
L.undraw()

The undraw function does exactly as the name implies. In the loop above, for
each value that n takes, a line is created (as above), drawn, and then, after a
wait of 0.3 seconds, it is erased. Again, modify the value of the increment and
observe the effect. Try also changing the amount of time in the wait
command.

You can also draw several geometrical shapes: circles, rectangles, ovals, and
polygons. To draw a circle, (or any geometrical shape), you first create it and
then draw it:

C = Circle(centerPoint, radius)
c.draw(myCanvas)

centerPoint is a Point object and radius is specified in pixels. Thus, to draw

a circle centered at (100, 150) with a radius of 30, you would do the following
commands:

C = Circle(Point(100, 150), 30)

c.draw(myCanvas)

Rectangles and ovals are drawn similarly (see details at the end of the
chapter). All geometrical objects have many functions in common. For

192
 Sights & Sounds

example, you can get the center point of a circle, a rectangle, or an oval by
using the command:

centerPoint = C.getCenter()

By default, all objects are drawn in black. There are several ways to modify or
specify colors for objects. For each object you can specify a color for its
outline as well as a color to fill it with. For example, to draw a circle centered
at (100, 150), radius 30, and outline color red, and fill color yellow:

C = Circle(Point(100, 150), 30)

C.draw(myCanvas)
C.setOutline(“red”)
C.setFill(“yellow”)

By the way, setFill and setOutline have the same effect on Point and
Line objects (since there is no place to fill any color). Also, the line or the
outline drawn is always 1 pixel thick. You can change the thickness by using
the command setWidth(<pixels>):

C.setWidth(5)

The command above changes the width of the circle’s outline to 5 pixels.

Do This: Try out all the commands introduced here. Also, look at the end of
the chapter for details on drawing other shapes.

Earlier, we mentioned that several colors have been assigned names that can
be used to select colors. You can also create colors of your own choosing by
specifying their red, green, and blue values. In Chapter 5 we mentioned that
each color is made up of three values: RGB or red, green and blue color
values. Each of the these values can be in the range 0..255 and is called a 24-
bit color value. In this way of specifying colors, the color with values (255,
255, 255) (that is red = 255, green = 255, and blue = 255) is white; (255, 0, 0)
is pure red, (0, 255, 0), is pure blue, (0, 0, 0) is black, (255, 175, 175) is pink,
etc. You can have as many as 256x256x256 colors (i.e. over 16 million

193
Chapter 8

colors!). Given specific RGB values, you can create a new color by using the
command, color_rgb:

myColor = color_rgb(255, 175, 175)

Do This: The program below draws several circles of random sizes with
random colors. Try it out and see its outcome. A sample output screen is show
on the right. Modify the program to input a number for the number of circles
to be drawn. randrange(m,n) returns a random number in range [m..n-1].

# Program to draw a bunch of # random colored circles

from myro import *
from random import *

def makeCircle(x, y, r):

# creates a Circle centered at point (x, y) of radius r
return Circle(Point(x, y), r)

def makeColor():
# creates a new color using random RGB values
red = randrange(0, 256)
green = randrange(0, 256)
blue = randrange(0, 256)
return color_rgb(red, green,blue)

def main():
# Create and display a
# graphics window
width = 500
height = 500
myCanvas =
GraphWin(‘Circles’,width,height)

myCanvas.setBackground("white")

# draw a bunch of random

# circles with random
# colors.
N = 500
for i in range(N):
# pick random center

194
 Sights & Sounds

# point and radius

# in the window
x = randrange(0,width)
y = randrange(0,height)
r = randrange(5, 25)
c = makeCircle(x, y, r)
# select a random color
c.setFill(makeColor())

c.draw(myCanvas)

main()

Notice our use of functions to organize the program. From a design

perspective, the two functions makeCircle and makeColor are written
differently. This is just for illustration purposes. You could, for instance,
define makeCircle just like makeColor so it doesn’t take any parameters and
generates the values of x, y, and radius as follows:

def makeCircle():
# creates a Circle centered at point (x, y) of radius r
x = randrange(0,width)
y = randrange(0,height)
r = randrange(5, 25)

return Circle(Point(x, y), r)

Unfortunately, as simple as this change seems, the function is not going to

work. In order to generate the values of x, and y it needs to know the width
and height of the graphics window. But width and height are defined in the
function main and are not available or accessible in the function above. This is
an issue of scope of names in a Python program: what is the scope of
accessibility of a name in a program?

Python defines the scope of a name in a program textually or lexically. That is,
any name is visible in the text of the program/function after it has been
defined. Note that the notion of after is a textual notion. Moreover, Python
restricts the accessibility of a name to the text of the function in which it is

195
Chapter 8

defined. That is, the names width and height are defined inside the function
main and hence they are not visible anywhere outside of main. Similarly, the
variables red, green, and blue are considered local to the definition of
makeColor and are not accessible outside of the function, makeColor.

So how can makeCircle, if you decided it would generate the x and y values
relative to the window size, get access to the width and height of the window?
There are two solutions to this. First, you can pass them as parameters. In that
case, the definition of makeCircle will be:

def makeCircle(w, h):

# creates a Circle centered at point (x, y) of radius r
# such that (x, y) lies within width, w and height, h
x = randrange(0,w)
y = randrange(0,h)
r = randrange(5, 25)

return Circle(Point(x, y), r)

Then the way you would use the above function in the main program would
be using the command:

C = makeCircle(width, height)

That is, you pass the values of width and height to makeCircle as
parameters. The other way to define makeCircle would be exactly as shown
in the first instance:

def makeCircle():
# creates a Circle centered at point (x, y) of radius r
x = randrange(0,width)
y = randrange(0,height)
r = randrange(5, 25)

return Circle(Point(x, y), r)

However, you would move the definitions of width and height outside and
before the definitions of all the functions:

196
 Sights & Sounds

from myro import *

from random import *

width = 500
height = 500

def makeCircle():
…

def makeColor():
…

def main():
…

Since the variables are defined outside of any function and before the
definitions of the functions that use them, you can access their values. You
may be wondering at this point, which version is better? Or even, why bother?
The first version was just as good. The answer to these questions is similar in
a way to writing a paragraph in an essay. You can write a paragraph in many
ways. Some versions will be more preferable than others. In programming, the
rule of thumb one uses when it comes to the scope of names is: ensure that
only the parts of the program that are supposed to have access to a name are
allowed access. This is similar to the reason you would not share your
password with anyone, or your bank card code, etc. In our second solution, we
made the names width and height globally visible to the entire program that
follows. This implies that even makeColor can have access to them whether it
makes it needs it or not.

You may want to argue at this point: what difference does it make if you
make those variables visible to makeColor as long as you take care not to use
them in that function? You are correct, it doesn’t. But it puts an extra
responsibility on your part to ensure that you will not do so. But what is the
guarantee that someone who is modifying your program chooses to?

We used the simple program here to illustrate simple yet potentially

hazardous decisions that dot the landscape of programming like land mines.

197
Chapter 8

Programs can crash if some of these names are mishandled in a program.

Worse still, programs do not crash but lead to incorrect results. However, at
the level of deciding which variables to make local and which ones to make
global, the decisions are very simple and one just needs to exercise safe
programming practices. We will conclude this section of graphics with some
examples of creating animations.

Any object drawn in the graphics window can be moved using the command
move(dx, dy). For example, you move the circle 10 pixels to the right and 5
pixels down you can use the command:

C.move(10, 5)

Do This: Let us write a program that moves a circle about (randomly) in the
graphics window. First, enter this following program and try it out.

# Moving circle; Animate a circle...

from myro import *
from random import *

def main():
# create and draw the graphics window
w = GraphWin("Moving Circle", 500, 500)
w.setBackground("white")

# Create a red circle

c = Circle(Point(250, 250), 50)
c.setFill("red")
c.draw(w)

# Do a simple animation for 200 steps

for i in range(200):
c.move(randrange(-4, 5), randrange(-4, 5))
wait(0.2)

main()

198
 Sights & Sounds

Notice, in the above program, that we are moving the circle around randomly
in the x- and y directions. Try changing the range of movements and observe
the behavior. Try changing the values so that the circle moves only in the
horizontal direction or only in the vertical direction. Also notice that we had
to slow down the animation by inserting the wait command after every move.
Comment the wait command and see what happens. It may appear that
nothing did happen but in fact the 200 moves went so quickly that your eyes
couldn’t even register a single move! Using this as a basis, we can now write
a more interesting program. Look at the program below:

# Moving circle; Animate a circle...

from myro import *
from random import *

def main():
# create and draw the graphics window
winWidth = winHeight = 500
w = GraphWin("Bouncing Circle", winWidth, winHeight)
w.setBackground("white")

# Create a red circle

radius = 25
c = Circle(Point(53, 250), radius)
c.setFill("red")
c.draw(w)

# Animate it
dx = dy = 3
while timeRemaining(15):
# move the circle
c.move(dx, dy)

# make sure it is within bounds

center = c.getCenter()
cx, cy = center.getX(), center.getY()

if (cx+radius >= winWidth) or (cx-radius <= 0):

dx = -dx

199
Chapter 8

if (cy+radius >= winHeight) or (cy-radius <= 0):

dy = -dy

wait(0.01)

main()

For 15 seconds, you will see a red circle bouncing around the window. Study
the program above to see how we keep the circle inside the window at all
times and how the direction of the ball bounce is being changed. Each time
you change the direction, make the computer beep:

computer.beep(0.005, 440)

If you are excited about the possibility of animating a circle, imagine what
you can do if you have many circles and other shapes animated. Also, plug in
the game pad controller and see if you can control the circle (or any other
object) using the game pad controls. This is very similar to controlling your
robot. Design an interactive computer game that takes advantage of this new
input modality. You can also design multi-user games since you can connect
multiple game pad controllers to your computer. See the Reference
documentation for details on how to get input from several game pad
controllers.

Drawing Text & Images

Like shapes, you can also place text in a graphics window. The idea is the
same. You first create the text using the command:

myText = Text(<anchor point>, <string>)

and then draw it. You can specify the type face, size, and style of the text. We
will not detail these here. They are summarized in the reference at the end of
the text. When we get an opportunity, we will use these features below in
other examples.

200
 Sights & Sounds

Images in this system are treated just like any other objects. You can create an
image using the Image command:

myPhoto = Image(<centerPoint>, <file name>)

You have to have an already prepared image in one of the common image
formats (like JPEG, GIF, etc.) and stored in a file (<file name>). Once the
image object is created, it can be drawn, undrawn, or moved just like other
shapes.

What you do with all this new found functionality depends on your creativity.
You can use graphics to visualize some data: plotting the growth of world
population, for example; or create some art, an interactive game, or even an
animated story. You can combine your robot, the game pad controller, or even
sound to enrich the multi-media experience. The exercises at the end of the
chapter present some ideas to explore further. Below, we delve into sounds.

Sound

As we have seen, you can have your robot make beeps by calling the beep()
function, like so:

beep(1, 440)

This command instructs the robot to play a tone at 440 Hz for 1 second. Let us
first try and analyze what is in the 440 Hz tone. First, the letters Hz are an
abbreviation for Hertz. The name itself comes from a German physicist,
Heinrich Rudolph Hertz who did pioneering work in the production of
electromagnetic waves in the late 19th century. Today, we use Hertz (or Hz)
as a unit for specifying frequencies.

1Hertz = 1cycle / second

201
Chapter 8

The most common use of frequencies these days is in specifying the clock
speeds of computer CPU's. For example, a typical PC today runs at clock
speeds of a few GigaHertz (or GHz).

1GigaHertz = 109cycles / second

Frequencies are related to periodic (or repetitive) motions or vibrations. The

time it takes for a motion or vibration to repeat is called its time period.
Frequency and time period are inversely related. That is the number of cycles
or repetitions in a second is called the frequency. Thus 1 Hertz refers to any
motion or vibration that repeats every 1 second. In the case of computer clock
frequencies then, a computer running at 4 Gigahertz is repeating 4 billion
times a second! Other examples of periodic motions include: the earth's
rotation on its axis (1 cycle every 24 * 60 * 60 = 86400 seconds or at a
frequency of 0.00001157 cycles/second), a typical audio CD spins 400 times a
second, a CD drive on your computer rated at 52x spins the CD at 52 * 400 =
20800 times per second, hummingbirds can flap their wings at frequencies
ranging from 20-78 times/second (some can go even as high as 200!).

Sound is a periodic compression and refraction (or return to its original state)
of air (for simplicity, let us assume that the medium is air). One Cycle of a
sound comprises one compression and one refraction. Thus, producing a beep
at 440 Hz represents 440 complete cycles of compression and refraction.
Generally, a human ear is capable for hearing frequencies in the 20 Hz to
20000 Hz (or 20 Kilo Hertz) range. However the capability varies from
person to person. Also, many electronic devices are not capable for producing
frequencies in that entire range. 20-20KHz is considered hi-fidelity for stereo
or home theater audio components. Let us first examine the range of audible
sounds the Scribbler can produce. To make a sound out of the Scribbler, you
have to give a frequency and the duration (in seconds) that the sound should
be played. For example, to play a 440 Hz sound for 0.75 seconds:

beep(0.75, 440)

202
 Sights & Sounds

The human ear is capable of distinguishing sounds that differ only by a few
Hertz (as little as 1 Hz) however this ability varies from person to person. Try
the commands:

beep(1, 440)
beep(1, 450)

Can you distinguish between the two tones? Sometimes it is helpful to place
these in a loop so you can repeatedly hear the alternating tones to be able to
distinguish between them. The next exercise can help you determine what
frequencies you are able to distinguish.

Exercise: Using the example above, try to see how close you can get in
distinguishing close frequencies. As suggested, you may want to play the
tones alternating for about 5-10 seconds. Start with 440 Hz. Can you hear the
difference between 440 and 441? 442? Etc. Once you have established your
range, try another frequency, say 800. Is the distance that you can distinguish
the same?

Do This: You can program the Scribbler to create a siren by repeating two
different tones (much like in the example above). You will have to experiment
with different pairs of frequencies (they may be close together or far apart) to
produce a realistic sounding siren. Write your program to play the siren for 15
seconds. The louder the better!

You can also have Myro make a beep directly out of your computer, rather
than the robot, with the command:

computer.beep(1, 440)

Unfortunately, you can't really have the robot and computer play a duet. Why
not? Try these commands:

beep(1, 440)
computer.beep(1, 440)
beep(1, 880)
computer.beep(1, 880)

203
Chapter 8

beep(1, 440)
computer.beep(1, 440)

What happens? Try your solutions to the above exercises by making the
sounds on the computer instead of the Scribbler.

Musical Scales

In western music, a scale is divided into 12 notes (from 7 major notes:

ABCDEFG). Further there are octaves. An octave in C comprises of the 12
notes shown below:

C C#/Db D D#/Eb E F F#/Gb G G#/Ab A A#/Bb B

C# (pronounced "C sharp") is the same tone as Db (pronounced "D flat").

Frequencies corresponding to a specific note, say C, are multiplied (or

divided) by 2 to achieve the same note in a higher (or lower) octave. On a
piano there are several octaves available on a spread of keys. What is the
relationship between these two tones?

beep(1, 440)
beep(1, 880)

The second tone is exactly one octave the first. To raise a tone by an octave,
you simply multiply the frequency by 2. Likewise, to make a tone an octave
lower, you divide by 2. Notes indicating an octave can be denoted as follows:

C0 C1 C2 C3 C4 C5 C6 C7 C8

That is, C0 is the note for C in the lowest (or 0) octave. The fifth octave
(numbered 4) is commonly referred to as a middle octave. Thus C4 is the C
note in the middle octave. The frequency corresponding to C4 is 261.63 Hz.
Try playing it on the Scribbler. Also try C5 (523.25) which is twice the
frequency of C4 and C3 (130.815). In common tuning (equal temperament)

204
 Sights & Sounds

the 12 notes are equidistant. Thus, if the frequency doubles every octave, each
successive note is 21 / 12 apart. That is, if C4 is 261.63 Hz, C# (or Db) will be:

/
C#4/Db4 = 261.63*2 = 277.18

Thus, we can compute all successive note frequencies:

D4 = 277.18 * 21 / 12 = 293.66

/
D#4/Eb = 293.66*2 = 311.13

etc.

The lowest tone that the Scribbler can play is A0 and the highest tone is C8.
A0 has a frequency of 27.5 Hz, and C8 has a frequency of 4186 Hz. That's
quite a range! Can you hear the entire range?

beep(1, 27.5)
beep(1, 4186)

Exercise: Write a Scribbler program to play all the 12 notes in an octave

using the above computation. You may assume in your program that C0 is
16.35 and then use that to compute all frequencies in a given octave (C4 is
16.35 * 24). Your program should input an octave (a number from 0 through
8), produce all the notes in that octave and also printout a frequency chart for
each note in that octave.

Making Music

Playing songs by frequency is a bit of a pain. Myro contains a set of functions

to make this task a bit more abstract. A Myro song is a string of characters
composed like so:

205
Chapter 8

NOTE1 [NOTE2] WHOLEPART

where [] means optional. Each of these notes/chords is composed on its own

line, or separated by semicolons where:

NOTE1 is either a frequency or a NOTENAME

NOTE2 is the same, and optional. Use for Chords.
WHOLEPART is a number representing how much of
a whole note to play.

NOTENAMES are case-insensitive strings. Here is an entire scale of NOTENAMES:

C C#/Db D D#/Eb E F F#/Gb G G#/Ab A A#/Bb B C

This is the default octave. It is also the 5th octave, which can also be written
as:

C5 C#5/Db5 D5 D#5/Eb5 E5 F5 F#5/Gb5 G5 G#5/Ab5 A5 A#5/Bb5 B5

C6

The Myro Song Format replicates the keys on the piano, and so goes from A0
to C8. The middle octave on a keyboard is number 4, but we use 5 as the
default octave. See http://en.wikipedia.org/wiki/Piano_key_frequencies for
additional details. Here is a scale:

"C 1; C# 1; D 1; D# 1; E 1; F 1; F# 1; G 1; G# 1; A 1; A# 1; B
1; C 1;"

The scale, one octave lower, and played as a polka:

"C4 1; C#4 1/2; D4 1/2; D#4 1; E4 1/2; F4 1/2; F#4 1; G4 1/2;

G#4 1/2; A4 1; A#4 1/2; B4 1/2; C4 1;"

There are also a few other special note names, including PAUSE, REST, you can
leave the octave number off of the default octave notes if you wish. Use "#"
for sharp, and "b" for flat.

206
 Sights & Sounds

WHOLEPART can either be a decimal notation, or division. For example:

Ab2 .125

or

Ab2 1/8

represents the A flat in the second octave (two below middle).

As an example, try playing the following:

c 1
c .5
c .5
c 1
c .5
c .5
e 1
c .5
c .5
c 2
e 1
e .5
e .5
e 1
e .5
e .5
g 1
e .5
e .5
e 2

Do you recognize it??

You may leave blank lines, and comments should begin with a # sign. Lines
can also be separated with a semicolon.

207
Chapter 8

Using a song

For the following exercises, you will need to have an object to play the song.
You will need to initialize the robot in a slightly different way. Rather than:

initialize()

do:

robot = Scribbler()

Now that you have a song, you probably will want to play it. If your song is in
a file, you can read it:

s = readSong(filename)

and play it on the robot:

robot.playSong(s)

or on the computer:

computer.playSong(s)

You can also use makeSong(text) to make a song. For example:

s = makeSong("c 1; d 1; e 1; f 1; g 1; a 1; b 1; c7 1;")

and then play it as above.

If you want to make it play faster or slower, you could change all of the
WHOLENOTE numbers. But, if we just want to change the tempo, there is an
easier way:

robot.playSong(s, .75)

208
 Sights & Sounds

The second argument to playSong is the duration of a whole note in seconds.

Standard tempo plays a whole note in about .5 seconds. Larger numbers will
play slower, and smaller numbers play faster.

Summary

You can use the graphics window as a way of visualizing anything. In the
graphics window you can draw all kinds of shapes: points, line, circles,
rectangles, ovals, polygons, text, and even images. You can also animate these
shapes as you please. What you can do with these basic drawing capabilities is
limited only by your creativity and your programming ability. You can
combine the sights you create with sounds and other interfaces, like the game
pad controller, or even your robot. The multimedia functionalities introduced
in this chapter can be used in all kinds of situations for creating interesting
programs.

Myro Reference

Below, we summarize all of the graphics commands mentioned in this

chapter. You are urged to review the reference manual for more graphics
functionality.

GraphWin()
GraphWin(<title>, <width>, <height>)
Returns a graphics window object. It creates a graphics window with title,
<title> and dimensions <width> x <height>. If no parameters are specified,
the window created is 200x200 pixels.

<window>.close()
Closes the displayed graphics window <window>.

<window>.setBackground(<color>)
Sets the background color of the window to be the specified color. <color>
can be a named color (Google: color names list), or a new color created using
the color_rgb command (see below)

209
Chapter 8

color_rgb(<red>, <green>, <blue>)

Creates a new color using the specified <red>, <green>, and <blue> values.
The values can be in the range 0..255.

Point(<x>, <y>)
Creates a point object at (<x>, <y>) location in the window.

<point>.getX()
<point>.getY()
Returns the x and y coordinates of the point object <point>.

Line(<start point>, <end point>)

Creates a line object starting at <start point> and ending at <end point>.

Circle(<center point>, <radius>)

Creates a circle object centered at <center point> with radius <radius>
pixels.

Rectangle(<point1>, <point2>)
Creates a rectangle object with opposite corners located at <point1> and
<point2>.

Oval(<point1>, <point2>)
Creates an oval object in the bounding box defined by the corner points
<point1> and <point2>.

Polygon(<point1>, <point2>, <point3>,…)

Polygon([<point1>, <point2>, …])
Creates a polygon with the given points as vertices.

Text(<anchor point>, <string>)

Creates a text anchored (bottom-left corner of text) at <anchor point>. The
text itself is defined by <string>.

Image(<centerPoint>, <file name>)

Creates an image centered at <center point> from the image file <file
name>. The image can be in GIF, JPEG, or PNG format.

210
 Sights & Sounds

All of the graphics objects respond to the following commands:

<object>.draw(<window>)
Draws the <object> in the specified graphics window <window>.

<object>.undraw()
Undraws <object>.

<line-object>.getP1()
<line-object>.getP2()
Returns the end points of the <line-object>.

<object>.getCenter()
Returns the center point of the <object>.

<object>.setOutline(<color>)
<object>.setFill(<color>)
Sets the outline and the fill color of the <object> to the specified <color>.

<object>.setWidth(<pixels>)
Sets the thickness of the outline of the <object> to <pixels>.

<object>.move(<dx>, <dy>)
Moves the object <dx>, <dy> from its current position.

The following sound-related functions were presented in this chapter.

beep(<seconds>, <frequency>)
beep(<seconds>, <f1>, <f2>)
Makes robot beep for <seconds> time at frequency specified. You can specify
a single frequency <frequency> or a mix of two: <f1> and <f2>.

<robot/computer object>.beep(<seconds>, <frequency>)

<robot/computer object>.beep(<seconds>, <f1>, <f2>)
Makes robot or computer beep for <seconds> at frequency specified. You can
specify a single frequency <frequency> or a mix of two: <f1> and <f2>.

211
Chapter 8

robot.playSong(<song>)
Plays the <song> on the robot.

readSong(<filename>)
Reads a song file from <filename>.

song2text(song)
Converts a <song> to text format.

makeSong(<text>)
text2song(<text>)
Converts <text> to a song format.

Python reference

In this chapter we presented informal scope rules for names in Python

programs. While these can get fairly complicated, for our purposes you need
to know the distinction between a local name that is local within the scope of
a function versus a global name defined outside of the function. The text
ordering defines what is accessible.

Exercises

1. Using the graphics commands learned in this chapter, write a

program to generate a seasonal drawing: winter,
summer, beach, or a festival scene.

2. Write a program that has a function

drawRobot(x, y, w) such that it draws the
figure of a robot shown on the right. The robot is
anchored at (x, y) and is drawn in the window, w.
Also notice that the robot is entirely made up of
rectangles (8 of them).

3. Using the drawRobot function from the

previous exercise, draw a pyramid of robot as shown on the right.

212
 Sights & Sounds

4. Suppose that each robot from the previous exercise is colored using one of
the colors: red, blue, green, and yellow. Each time a robot is drawn, it uses the
next color in the sequence. Once the sequence is exhausted, it recycles. Write
a program to draw the above pyramid so robots appear in these colors as they
are drawn. You may decide to modify drawRobot to have an additional
parameter: drawRobot(x, y, c, w) or you can write it so that drawRobot
decides which color it chooses. Complete both versions and compare the
resulting programs. Discuss the merits and/or pitfalls of these versions with
friends. [Hint: Use a list of color names.]

5. The beep command can play two tones simultaneously. In this exercise we
will demonstrate an interesting phenomena. First, Try the following
commands:

beep(2, 660)
beep(2, 665)

You may or may not be able to tell the two tones apart. Next, try playing the
two tones together using the command:

beep(2, 660, 665)

You will hear pulsating beats. This phenomenon, called beating, is very
common when you play pure tone together. Explain why this happens.

6. Do a web search on fractals and write programs to draw some simple

fractals. For example, Koch Snowflakes, Sierpinski Triangles, Mandelbrot
Sets, Julia Sets, etc.

213
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Image Processing &
Perception

Seeing is believing.
Proverb

Opposite page: Rotating Snakes (A Visual Illusion)

Created by Akiyoshi Kitaoka (www.ritsumei.ac.jp/~akitaoka/index‐e.html)

215
Chapter 9

Your robot has a small digital camera that can be used to take pictures. A
picture taken by a digital camera is represented as an image. As you have seen
in the previous chapter images can be drawn and moved about in a graphics
window just as if it were any other graphics object (like a line, circle, etc.).
We also saw in Chapter 5 how an image taken from the Scribbler’s camera
can be viewed on your monitor in a separate window. In this chapter we will
learn how to do computing on images. We will learn how images are
represented and how we can create them via computation and also process
them in many different ways. The representation of images we will use is
same as those used by most digital cameras and cell phones and the same
representation can be used to display them in a web page. We will also learn
how an image taken by the robot’s camera can be used to serve as the
camera’s eye into its world using some image understanding techniques.
Image understanding is the first step in visual perception. A robot equipped
with even the most rudimentary visual perception capabilities can be designed
to carry out more interesting behaviors.

What is an Image?

In Myro you can issue a command for the robot to take a picture and display it
on the computer’s monitor using the commands:

pic = takePicture()
show(pic)

The picture on the next page shows an example image taken from the
Scribbler’s camera. An image is made up of several tiny picture elements or
pixels. In a color image, each pixel contains color information which is made
up of the amount of red, green, and blue (also called, RGB) values. Each of
these values can be in the range [0..255] and hence it takes 3 bytes or 24 bits
to store the information contained in a single pixel. A pixel that is colored
pure red will have the RGB values (255, 0, 0). A grayscale image, on the
other hand, only contains the level of gray in a pixel which can be represented
in a single byte (8 bits) as a number ranging from 0..255 where 0 is black and
255 is white. The entire image is just a 2-dimensional array of pixels. For
example, the images obtained from the Scribbler have 256x192 (WxH) or a
216
 Image Processing & Perception

total of 49,152 pixels. Since each pixel

requires 3 bytes of data, this image has a
size of 147,456 bytes.

All digital cameras are sold by specifying

the number of megapixels. For example,
the camera shown below is rated at 6.3
megapixels. This refers to the size of the
largest image that it can take. The more
pixels in an image the better the image
resolution can be when it is printed. With a
6.3 megapixel image you will be able to create good quality prints as large as
13x12 inches (or even larger). By comparison, a conventional 35mm
photographic film has roughly 4000x3000 or 12 million pixels. Professional
digital cameras easily surpass this resolution and that is why, in the last
decade or so, we have seen a rapid decline in film-based photography. For
displaying sharp images on a computer you need less than half the resolution
offered by the camera shown here.
The Scribbler camera, with an
image size of 147,456 bytes is only
about 0.14 megapixels. While this is
low resolution it is sufficient for
doing visual perception for the
robot.

To make electronic storage and

transfer of images (web, e-mail,
etc.) fast and convenient the data in an image can be compressed. Several
formats are available to electronically store images: JPEG, GIF, PNG, etc.
JPEG is the most common format used by digital cameras, including the
Scribbler’s. JPEG enables excellent compression coupled with a wider range
of colors compared with the GIF format making it useful for most image
applications. Myro supports both JPEG and GIF image formats. When we
intend to process an image, we will always use the JPEG format. We will
primarily use the GIF format for creating animated images.

217
Chapter 9

Myro Image Basics

After you take a picture with the Scribbler as above you can get some
information about the size of the picture with the getWidth() and
getHeight() functions:

picWidth = getWidth(pic)
picHeight = getHeight(pic)
print "Image WxH is", picWidth, "x", picHeight, “pixels.”

If you wish to save the image for later use, you can use the Myro command:

savePicture(pic, "OfficeScene.jpg")

The file OfficeScene.jpg will be saved in the current folder. The .jpg
extension signals the command to save the image as a JPEG image. If you
wanted to save it as a GIF image, you can use the .gif extension as shown
below:

savePicture(pic, "OfficeScene.gif")

Later, you can load the picture from disk with the makePicture() function:

mySavedPicture = makePicture("OfficeScene.jpg")
show(mySavedPicture)

A nice command combination that allows you to navigate and then select the
image to load is:

mySavedPicture = makePicture(pickAFile())
show(mySavedPicture)

The pickAFile command gives you a navigational dialog box, just like the
one you see when you open and select files in many other computer
applications. You can navigate to any folder and select a file to open as an
image. In fact, you can use the makePicture command to load any JPEG
picture file regardless of whether it was created from your own digital camera
218
 Image Processing & Perception

or one you downloaded from the web. Below, we show you how to load a
picture and display it:

lotusTemple = makePicture(pickAFile())
show(lotusTemple, “Lotus Temple”)

If you move your mouse and

click anywhere in the picture,
you can also get the x- and y-
coordinates of the pixel you
selected along with its RGB
values. This is shown in the
picture on the right. The
mouse was clicked on the
pixel at location (65, 196) and
its RGB values were
(168,174,104). Can you guess
where that is in the picture?
By the way, the show
command takes an optional
second parameter which is a
string that becomes the title of the image window. One advantage of being
able to load and view any image is that we can also learn to process or
manipulate such images computationally. We will return to image processing
later in this chapter.

A Robot Explorer

If you do not need a full color picture, you can tell Myro to capture a gray-
scale image by giving the takePicture() function the "gray" parameter.

grayPic = takePicture("gray")
show(grayPic)

You will notice that taking the gray-scale picture took less time than taking
the color picture. As we explained earlier, a gray scale picture only uses one

219
Chapter 9

byte per pixel, instead of three. Because gray-scale images can be transferred
from the robot to your computer faster than full color images, they can be
useful if you want the images to update quickly. For example, you can use the
joyStick() function combined with a loop that takes and displays pictures to
turn your robot into a remotely piloted explorer, similar to the mars rovers.

joyStick()
for i in range(25):
pic = takePicture("gray")
show(pic)

The above code will open a joy stick window so that you can control your
robot and then capture and show 25 pictures, one after the other. While the
pictures are being captured and displayed like a movie, you can use the
joystick to drive your robot around, using the pictures to guide it. Of course, if
you removed the "gray" parameter from the takePicture() function call,
you would get color pictures instead of grayscale pictures, but they would
take much longer to transfer from the robot to your computer, and make it
more difficult to control the robot.

Animated GIF movies

The savePicture() function also allows you to make an animated GIF,

which is a special type of picture that in a web browser will show several
pictures one after another in an animation. To save an animated GIF, you
must give the savePicture() function a list of pictures (instead of a single
picture) and a filename that ends in ".gif". Here is an example:

pic1 = takePicture()
turnLeft(0.5,0.25)
pic2 = takePicture()
turnLeft(0.5,0.25)
pic3 = takePicture()
turnLeft(0.5,0.25)
pic4 = takePicture()

listOfPictures = [pic1, pic2, pic3, pic4]

220
 Image Processing & Perception

savePicture(listOfPictures, "turningMovie.gif")

The best way to view an animated GIF file is to use a web browser. In your
favorite browser use the FILE->Open File menu, and then pick the
turningMovie.gif file. The web browser will show all frames in the movie,
but then stop on the last frame. To see the movie again, press the "Reload"
button. You can also use a loop to make a longer movie with more images:

pictureList = [] #Start with an empty list.

for i in range(15):
pic = takePicture()
pictureList = pictureList + [pic] #Append the new picture
turnLeft(0.5,0.1)

savePicture(pictureList,"rotatingMovie.gif")

The above commands create an animated GIF movie from 15 pictures taken
while the robot was rotating in place. This is a nice way to capture a complete
scene around the robot

Making Pictures

Since an image is just an array of pixels it is also possible to create or

compose your own images by coloring in each individual pixel. Myro
provides simple commands to fill in or examine a color or grayscale value in
individual pixels. You can use computations to determine what to fill in each
pixel. To start with, you can create a blank image as follows:

W = H = 100
newPic = makePicture(W, H, black)
show(newPic)

The 100x100 pixel image starts out with all its pixels
colored pure black (i.e. RGB = (0,0,0)). If you’d
rather like all pixels to be a different color, you can
specify its RGB values:

221
Chapter 9

newPic = makePicture(W, H, makeColor(R,G,B))

Alternately, if you ever need to, you can also set all the pixels to any color,
say white, using the loop:

for x in range(W)
for y in range(H):
pixel = getPixel(newPic, x, y)
setColor(pixel, white)
repaint(newPic)

The getPixel command returns the pixel at specified x- and y- locations in

the picture. setColor sets the given pixel’s color to any specified color.
Above we’re using the predefined color white. You can create a new color by
specifying its RGB values in the command:

myRed = makeColor(255, 0, 0)

To visually select a color and its corresponding RGB values, you can use the
command:

myColor = pickAColor()

A color palette will be displayed from which you can select a color of your
choosing. The palette also shows you the chosen color’s RGB values. After
you select a color and press OK, the value of myColor will be the selected
color.

The repaint command refreshes the displayed image so you can view the
changes you made. In the example loop above, we are using it after all the
pixels are modified. If you want to see the changes as they are being made,
you can include the repaint command inside the loop:

for x in range(W)
for y in range(H):
pixel = getPixel(newPic, x, y)
setColor(pixel, white)
repaint(newPic)

222
 Image Processing & Perception

You will be able to view each pixel being changed. However, you will also
notice that repainting this way takes a considerable amount of time even on
the small image we are creating. Thus, it is a good idea to refresh once all the
pixels are modified.

In addition to the setColor command, Myro also has the setRGB command
that can be used to set a pixel’s color. This command uses RGB values
themselves, instead of a color.

setRGB(pixel, (255, 0, 0))

There is also a corresponding command to get the RGB values of a given

pixel:

r, g, b = getRGB(pixel)

The getRGB command returns the triple (R,G,B) that can be assigned to
individual variables as shown above. Additionally, given a pixel, you get the
individual RGB values using the commands getRed, getGreen, and getBlue.
These are described in more detail at the end of this chapter and are illustrated
in examples below.

Many image creation and processing situations require the processing of every
pixel in the image. In the loops above, we are using the x- and y- variables to
reach every pixel in the image. An alternate way of accomplishing the same
result is to use the following form of the loop:

for pixel in getPixels(newPic):

setColor(pixel, gray)
repaint(newPic)

Like the timeRemaining function used in earlier chapters, getPixels returns

the next pixel to be processed each time around the loop thus guaranteeing
that all pixels of the image are processed. Just to watch how the loop above
works you may want to try and put repaint inside the loop. The difference
between the two methods is that the first loop gets a pixel at the given x- and

223
Chapter 9

y- coordinates, while the latter gives you a pixel at a time without worrying
about its x- and y- values.

Shades of Gray

Using the basic image pixel accessing and modifying commands one can
write computations to create interesting and creative images. To introduce you
to basic image creation and processing techniques, we will create some
images entirely in the grayscale spectrum. In a JPEG image, all shades of gray
have equal RGB values. The darkest shade of gray is (0,0,0) or black and the
brightest is (255,255,255) or white. Rather than worrying about the triple of
RGB values, we can think of just a single value in the range 0..255 to
represent the grayscale spectrum. We can write a simple function that will
return the RGB shade of gray as follows:

def gray(v):
# returns an rgb gray color for v
return makeColor(v, v, v)

Let’s create an image that displays the entire range of shades of gray using the
following rule:

pixel at x, y is colored using the grayscale x+y

That is, a pixel in the image at 0, 0 will get the shade 0+0=0 or black, which
the pixel at 50,50 will get the shade 50+50=100 which will be a midrange
gray. We can accomplish this using the following loop:

def main():
MAX = 255
W = 100
H = 100

# Create the blank image

myPic = makePicture(W, H)
show(myPic, “Shades of Gray”)

224
 Image Processing & Perception

# Fill in each pixel with x+y shade of gray

for x in range(W):
for y in range(H):
setPixel(myPic, x, y, gray((x+y)%(MAX+1))

We have used the variable MAX above to represent the maximum value of the
grayscale range. In the last line we take the remainder (x+y)%(MAX+1) to
ensure that the gray value will be in the range 0..255. The image generated by
the above commands is shown on the right. It may seem strange at first that
we have used the x- and y- locations of a pixel to compute its grayscale or
brightness value. Since x- and y- are numbers and since the grayscale value
itself is a number it is ok to do this. Recognizing that x- and y- values (and
their sum) might be larger than the range of the grayscale spectrum, we take
the remainder to wrap the values around.

Do This: Write a complete program and try this out. Once you have the
program, try it out on an image of size 123x123 (why?). Then again, try it out
on a much larger image, say 500x500.

The grayscale range basically represents levels of brightness. We can have

some more fun by changing the range, say to 0..50. Then
we can modify the gray function as shown below to
proportionately map any value in the range 0..50 to the
range 0..255:

MAX = 50
def gray(v):
# returns an rgb gray color for v
brightness = v*255/MAX
return makeColor(brightness, brightness, brightness)

Do This: Rewrite the program to use the above definition of gray and create a
500x500 image. Modify the value of MAX to 25 and create a 100x100 image
(shown above). Note that if you set MAX to 255, it will revert to the earlier
version of the function.

225
Chapter 9

Do This: Next use the rule:

pixel at x, y is colored using the

grayscale x*y

Remember to take remainder as above.

Change MAX to 250 and create a
500x500 image. Try other sizes and
values of MAX.

You can experiment with other rules,

geometric transformations, and
trigonometric functions. See exercises
at the end of this chapter for more
suggestions.

Image Processing

One doesn’t always have to start with a

blank image to create a new one. You
can take existing images and transform
them in interesting ways. The basic
principles remain the same: you access
an individual pixel and its color value
and transform it in any way you like. In
this section we will do some simple
image transformations to illustrate this.
All the image transformations we will
learn will require the same
transformation on all pixel values of a
given image. You can of course choose
to select specific regions if you like.

We will use the sample images shown here to learn image transformations.
You can feel free to select one or more images from your own digital camera.

226
 Image Processing & Perception

In most of the examples we have used transformations on grayscale images.

You should feel free to use color images and experiment with them. You will
also notice that the bigger the size of the image, the longer it will take to
transform it. This is a direct consequence of the amount of computation you
have to do. For example, to transform a 500x500 image you will be operating
on 250,000 pixels. If each pixel transformation involves 10 operations, you
will be specifying 25 million operations! How long this will take depends on
the computer you are using. We will return to this in more detail later in the
chapter. For now, try and restrict yourself to small size images, say no bigger
than 300x300 pixels.

As we mentioned earlier, even the most rudimentary digital cameras these

days give you several megapixels of image resolution. So the first thing we
should learn to do is how to shrink a given image into a smaller size.

Shrinking & Enlarging

After Shrinking
(150x106, F=2)

Original Image (300x213)

Lets us write a program that will take an input image and shrink it by a given
factor, say F. For example, if the original image is 3000x3000 pixels and we
shrink it by a factor of 10, we would end up with an image of size 300x300

227
Chapter 9

pixels. This transformation boils down to selecting the value of each of the
pixel in the new image based on the old image. We can use the following rule
as our transformation:

New pixel at x, y is a copy of the old pixel at x*F, y*F

The program below prompts the user to pick an image, display it, and then ask
the user to specify the shrinking factor (which has to be larger than 1).

def main():

# read an image and display it

oldPic = makePicture(pickAFile())
show(myPic, "Before")

X = getWidth(oldPic)
Y = getHeight(oldPic)

# Input the shrink factor and computer size of new image

F = int(ask("Enter the shrink factor."))
newx = X/F
newy = Y/F

# create the new image

newPic = makePicture(newx, newy)

for x in range(newx):
for y in range(newy):
setPixel(newPic, x, y, getPixel(myPic, x*F, y*F))

show(newPic, "After")

Do This: Select one or more images and make sure they are in JPEG format
and are high resolution images (for our purposes, 400x400 pixels or larger).
Run the program on each image and observe the output for different values of
F. Modify the program to use the savePicture command to save the images
into a file for use in examples below.

228
 Image Processing & Perception

Enlarging is done in a complementary

manner using the rule:

New pixel at x, y is a copy of the old

pixel at x/F, y/F

The size of the new image will be F

times the size of the original image. 4-pixel neighborhood
When you enlarge an image from a
smaller image you are really generating
data where there wasn’t any. Therefore,
the resulting image is likely to be more
blurred than the original. This is similar
to so called upconverting DVD players
that claim to upconvert a movie to high
definition from a standard signal. If you
look closely, you will be able to see
flaws in the unconverted images.

Blurring & Sharpening

Blurred Image
You can blur an image by taking the
average color values of neighboring
pixels. The size of the neighborhood
determines the amount of blurring.
Below, we show how to blur each pixel
using a neighborhood of 4 (see picture
on right):

n1 = getPixel(oldPic, x, y)
n2 = getPixel(oldPic, x, y-1)
n3 = getPixel(oldPic, x-1, y)
n4 = getPixel(oldPic, x+1, y)
n5 = getPixel(oldPic, x, y+1)
Sharpened Image

229
Chapter 9

v = (getRed(n1)+…+getRed(n5))/5
setPixel(newPic, x, y, gray(v))

Notice in the above that n1 is the original pixel. You can use different sizes
and shapes of neighborhoods to create more elaborate blurring effects. The
next page shows the effect of using a larger, still square, neighborhood.

Instead of averaging the pixel values, if

you subtracted the pixel values you end up
sharpening the image. That is,

v = 5*getRed(n1) -
(getRed(n2)+…+getRed(n5))
setPixel(newPic, x, y, gray(v))

15‐pixel neighborhood
Do This: Write a complete program that
uses the above neighborhood averaging
formula to blur and sharpen images.
Depending on the size and shape of the
neighborhood you choose, the loop index
ranges will vary. For example, the loop for
the 4-pixel neighborhood will appear as
follows:

for x in range(1,X-1):
for y in range(1,Y-1):
# Average pixel values…

As the size of the neighborhood grows, Blurred Image

you will have to adjust the start and stop
values of the x and y variables in the loop above.

Negative & Embossing

Creating a negative of an image is obtained by inverting the pixel values. For

grayscale values this amounts to subtracting from 255. For color pixels you

230
 Image Processing & Perception

have to subtract from 255 for each of the RGB values. The rule for grayscale
is:

New pixel at x,y is 255-value of old pixel at x,y

Negative Emboss

You can specify the entire transformation using the loop:

for x in range(X):
for y in range(Y):
pixel = getPixel(oldPic, x, y)
v = MAX – getRed(pixel)
setPixel(newPic, x, y, gray(v))

Similarly, you can create an embossing or relief effect by subtracting from the
current pixels value, the value (or a fraction thereof) of a pixel two pixels
away:

pixel1 = getPixel(oldPic, x, y)
pixel2 = getPixel(oldPic, x+2, y+2)
v = getRed(pixel1) – getRed(pixel2)
setPixel(newPic, x, y, gray(v))

or

231
Chapter 9

v = getRed(pixel1) + (MAX/2 – getRed(pixel2))

Do This: Implement the two rules shown above. Try creating a negative of a
color image, as well as a grayscale image. Try different fractions for
embossing to see their effect.

Most likely, as you were reading and experimenting with the above
techniques, you were wondering about the kind of image processing done by
some of the popular photo-editing software programs. You may already be
familiar with or might have heard of Adobe Photoshop, Fireworks, GIMP, etc.
Most of these programs use techniques similar to the ones we have described
above and provide many more features and options for image manipulation.

Image Understanding & Robot Vision

The picture shown on right was taken from a

newspaper article in a local newspaper in
western New York. Read the accompanying
caption and see if you can identify each
individual in the picture. After reading the
caption, you will no doubt confirm that the
person in the front row center is Coach Kevin
Starr. It would be really hard to do identify
the people if the caption weren’t present. Still
you would be able to answer questions like,
how many people are in the picture. This is an "THE BEST - The All-WNY boys volleyball
example of the image understanding problem: team surrounds the coach of the year,
Frontier's Kevin Starr. Top row, from left,
Given an image, how can you determine, are Eden's Brian Heffernan, Lake Shore's
computationally, that there are seven people Tom Kait, Sweet Home's Greg
in the image, that three of them are kneeling Gegenfurtner, Kenmore West's George
and four are standing, etc. You would also not Beiter. Bottom row: from left, are Eden's
have any trouble drawing a box on the face of David Nagowski and Orchard Park's Rich
Bland."
each person in the image. In this section we
introduce you some basic techniques in image understanding and use them for
creating a visual perception system for the Scribbler robot. You are already

232
 Image Processing & Perception

familiar with the basic image processing commands. The same commands
will be used to do image understanding.

For a robot to perceive its environment visually it has to start by taking a

picture. Once the image is obtained perception of the contents of the image
are done computationally. Instead of taking on the image understanding task
shown above, we will start simple: how does a Scribbler recognize a ball?
Once it recognizes it, can it follow the ball wherever it goes?

The second problem becomes easy once we know how to recognize a ball. We
can write a simple control program like this:

while timeTemaining(T):
# take picture and locate the ball
…
if <ball is on the left of visual field>:
turnLeft(cruiseSpeed)
elif <ball is in the middle of visual field>:
forward(cruiseSpeed)
else:
turnRight(cruiseSpeed)

The main problem is locating the ball

in the image. Given the pixel
manipulation commands you have
learned so far, think about how we
might accomplish this?

The basic principles in image

understanding rely on starting at the
pixel level. Let us make things a little
more concrete. Take a look at the
image shown here on the right. The picture was taken by a Scribbler robot.
For our purposes we have exaggerated the scene by choosing a bright colored
(pink) ball. When you click your mouse anywhere on the image it will give
you the RGB values of that pixel. The picture shows the values of a pixel in
the center of the ball. Notice that the red pixel is 253 and the green pixel’s

233
Chapter 9

value is 66. By the way, true pink has RGB values (255,175,175). We can use
this information as a way of identifying the ball. Here is a simple image
transformation loop that turns all pixels in the image into white or black
pixels. If a pixel has a high red value and a lower green value it will be turned
white else it will be turned black.

for pixel in getPixels(p):

r, g, b = getRGB(pixel)
if r > 200 and g < 100:
setRGB(pixel, (255, 255, 255))
else:
setRGB(pixel, (0, 0, 0))

The resulting image is shown here. As

you can see, by just selecting the
appropriate values of individual colors
you can filter out the unwanted regions
and focus on just the colors you are
interested in. Even though the entire
ball has not been filtered, it is sufficient
for use in identifying its location. By
tweaking the RGB threshold values you
can further refine this filter. Further
processing will be required to identify
the ball’s location in the visual field
(which is the image).

We are interested in locating the ball in the left, right, or center of the image.
One way to accomplish this is to locate the x- coordinate of the center point
of the ball. One way to do this is to take the average of the x- locations of all
white pixels in the image:

def locateBall(picture):
tot_x = 0
count = 0

234
 Image Processing & Perception

for pixel in getPixels(p):

r, g, b = getRGB(pixel)
if r > 200 and g < 100:
tot_x = tot_x + getX(pixel)
count = count + 1
return tot_x/count

Alternately, you could also average the x- locations of all pink pixels without
using the filter. On the image shown, the average x- value returned was 123.
Given that the image is 256 pixels wide, 123 is just about in the center.
Below, we show the locations obtained for each of the images shown:

location = 41 location = 123 location = 215

Given that the image’s width is 256 pixels, you can divide the visual field into
three regions: left (0..85), center (86..170), and right (171, 255). The robot
control to follow the ball can then be defined as shown below:

while timeTemaining(T):
# take picture and locate the ball
pic = takePicture()
ballLocation = locateBall(pic)

if ballLocation <= 85:

turnLeft(cruiseSpeed)
elif ballLocation <= 170:
forward(cruiseSpeed)
else:
turnRight(cruiseSpeed)

Do This: Implement the above program to recognize a given object in the

picture. The object could be a box, a ball, a trash can, etc. Experiment with

235
Chapter 9

different colored objects. Can you use these techniques to recognize any
object? Say an animal? A person? Etc.

Blobs

Setting a blob on the ball. Result of takePicture("blob")

The kind of threshold filtering used above is commonly called blob filtering.
All pixels satisfying the selected color properties are to be identified and the
rest eliminated. This is such a common operation in image processing that
many cameras and camera hardware integrate blob filtering as a primitive.
The Scribbler also has this capability. Using this, you do not have to worry
about identifying specific color values for the filter threshold. All you have to
do is take a picture and then display it. Then using your mouse, define the
rectangular region that contains your desired blob. To do this, take a picture,
display it, and then click your mouse on the top left corner of the ball in the
image and then drag it to the bottom right corner before releasing it. You have
to make sure that you are connected to the robot because the selected blob
region is transferred back to the robot. Once that is defined, you will be able
to make use of the predefined blob filter by taking pictures using the
command:

pic = takePicture("blob")

236
 Image Processing & Perception

As soon as the picture is taken, the camera’s blob detector goes to work and
returns the resulting image (shown above). Now you can use this image as
before to control your robot. You will have to average the locations of all
black pixels.

Alternately, you could use the getBlob() function as follows:

onPixels, avgX, avgY = getBlob()

The function returns three values: the number of ‘on’ pixels (i.e. the pixels in
the image that were in the detected blob), and the average x- and y- location
of the blob pixels. This is a
better and much faster way of
doing blob recognition and you
will notice that your robot tracks
the ball much better.

Do This: Redo the ball tracking

example from above to use the
blob features introduced here.

Further computations on an
image can result in identifying
object boundaries. This is called
edge detection. These can be
further used to do object
recognition: Is there a car in the
picture? A face? Etc. Object
recognition requires the ability
to identify features that are
typical of an object. The domain
of image recognition and
computational vision is rich with techniques and applications like these. On
the right, we show the results of recognizing faces and identifying the
individuals in the picture. As researchers advance these techniques and as
computational capabilities of digital cameras increase, we are starting to see
237
Chapter 9

many of these incorporated into even the most basic digital cameras. Such
techniques can be used to automatically focus on the main subject and also
adjust exposure to get a perfect picture even for amateur photographers.

Summary

In this chapter, you have seen various Myro facilities that can be used for
image creation, processing and understanding. Most image creation,
processing, and understanding techniques involve pixel-level manipulations of
an image. However, many interesting effects as well as applications can be
developed using these simple ideas. Many of these features are increasingly
found in everyday devices like digital cameras, biometric identification
devices, and even automated mail sorting systems. Combined with robot
behaviors these facilities can be used for modeling smarter behaviors.

Myro review

getBlob()
Returns a triple of onPixels, avgX, avgY (the number of pixels that
matched the blob, the average x and y locations of blob pixels).
getHeight(<picture>)
getWidth(<picture>)
Returns the height and width of the <picture> object (in pixels).
getPixel(<picture>, x, y)
Returns the pixel object at x,y in the <picture>.
getPixels(<picture>)
When used in a loop, returns one pixel at a time from <picture>.
getRGB(pixel)
getRed(<pixel>)
getGreen(<pixel>)

238
 Image Processing & Perception

getBlue(<pixel>)
Returns the RGB values of the <pixel>.
makeColor(<red>, <green>, <blue>)
Creates a color object with the given <red>, <green>, and <blue> values (all
of which are in the range [0..255]).
makePicture(<file>)
makePicture(<width>, <height>)
makePicture(<width>, <height>, <color>)
Creates a picture object either by reading a picture from a <file>, or of the
given <width> and <height>. If <color> is not specified, the picture created
has a white background.
pickAColor()
Creates an interactive dialog window to select a color visually. Returns the
color object corresponding to the selected color.
pickAFile()
Creates an interactive dialog window that allows user to navigate to a folder
and select a file to open. Note: it cannot be used to create new files.
repaint()
repaint(<picture>)
Refreshes the displayed <picture>.
savePicture(<picture>, <file>)
savePicture(<picture list>, <gif file>)
Saves the <picture> in the specified file (a GIF or JPEG as determined by
the extension of the <file>: .gif or .jpg). <picture list> is saved as an
animated GIF file.
setColor(<pixel>, <color>)
setRed(<pixel>, <value>)
setGreen(<pixel), <value>)

239
Chapter 9

setBlue(<Pixel>, <value>)
Sets the color of <pixel> to specified <color> or <value>.
show(<picture>)
show(<picture>, <name>)
Displays the <picture> on the screen in a window named <name> (string).
takePicture()
takePicture(“gray”)
takePicture(“blob”)
Takes a picture from the Scribbler camera. It is a color picture by default, or
grayscale (“gray”), or a filtered image based on the defined blob (“blob”).
See chapter text for examples.

Python Review

There were no new Python features introduced in this chapter.

240
 Image Processing & Perception

Exercises

1. Instead of viewing each pixel in an image as x- and y- values in the

Cartesian coordinate system, imagine that it is in a polar coordinate system
with the origin in the center of the picture (See picture below).

Write a function called polar(x, y) that returns the polar coordinates (r, theta)
of the given point. Using this function write a transformation as follows:
for x in range (1, W):
for y in range(1, H):
r, theta = polar(x, y)
setPixel(myPic, x, y,
gray((((r+theta)%16)-8)*MAX/16+MAX/2))

Create a 400x400 picture using the above transformation. Start with a black
picture.

2. Based on the above exercise, try transforming an image (like a portrait). For
example use the rule: new x, y is a copy of original picture’s /2, theta.

241
Chapter 9

3. Write a transformation that flips an image clockwise (or counter clockwise)

by 90 degrees. Try it on a square image of your choice.

4. Write an image transformation that averages two images (same size) into a
single image.

5. Write a program that sews two or more images together to create a

panorama.

6. Use blob detection techniques described in this chapter to recognize Coke

and Pepsi cans. Write a program to hunt for a Coke or Pepsi can.

7. Can you expand on Exercise above to recognize small toy figurines of

Disney cartoon characters like Mickey, Minnie, Daisy, Goofy, Pluto, and
Donald?

242
Image Processing & Perception

243
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Artificial Intelligence
David: Martin is Mommy and Henry's real son. After I find the Blue Fairy then I can
go home. Mommy will love a real boy. The Blue Fairy will make me into one.
Gigolo Joe: Is Blue Fairy Mecha, Orga, man or woman?
David: Woman.
Gigolo Joe: Woman? I know women! They sometimes ask for me by name. I know
all about women. About as much as there is to know. No two are ever alike, And
after they've met me, no two are ever the same. And I know where most of them can
be found.
David: Where?
Gigolo Joe: Rouge City. Across the Delaware.

Dialog between two Artificial Intelligence entities: Gigolo Joe (played by Jude Law)
and David (played by Haley Joel Osment) in the movie, Artificial Intelligence,
Directed by Steven Speilberg, Warner Bros., 2001.

Opposite page: A.I. Artificial Intelligence

From the movie poster. Warner Bros., 2001.

245
Chapter 10

The Question of Intelligence

The quest for the understanding of intelligence probably forms the oldest and
yet to be fully understood human inquiry. With the advent of computers and
robots the question of whether robots and computers can be as intelligent as
humans has driven the scientific pursuits in the field of Artificial Intelligence
(AI). Whether a computer can be intelligent was lucidly discussed by
Professor Alan Turing in 1950. To illustrate the issues underlying machine
intelligence, Turing devised a thought experiment in the form of an imitation
game. It is played with three people, a man, a woman, and an interrogator.
They are all in separate rooms and interact with each other by typing text into
a computer (much like the way people interact with each other over IM or
other instant messaging services). The interrogator's task is to identify which
person is a man (or woman). To make the game interesting, either player can
try and be deceptive in giving their answers. Turing argues that a computer
should be considered intelligent if it could be made to play the role of either
player in the game without giving itself away. This test of intelligence has
come to be called the Turing Test and has generated much activity in the
community of AI researchers (see Exercises). The dialog shown above, from
the movie Artificial Intelligence, depicts an aspect of the test of intelligence
designed by Alan Turing. Based on the exchange between Gigolo Joe and
David, can you conclude that they are both intelligent? Human?

After over five decades of AI research the field has matured and evolved in
many ways. For one, the focus on intelligence is no longer limited to humans:
insects and other forms of animals with varying degrees and kinds of
intelligence have been the subject of study within AI. There has also been a
fruitful exchange of ideas and models between AI scientists, biologists,
psychologists, cognitive scientists, neuroscientists, linguists and philosophers.
You saw examples of such an influence in the models of Braitenberg vehicles
introduced earlier. Given the diversity of researchers involved in AI there has
also been an evolution of what AI itself is really about. We will return to this
later in the chapter. First, we will give you a few examples of models that
could be considered intelligent that are commonly used by many AI scientists.

246
 Artificial Intelligence

Language Understanding

One aspect of intelligence acknowledged by many people is the use of

language. People communicate with each other using a language. There are
many (several thousand) languages in use on this planet. Such languages are
called natural languages. Many interesting theories have been put forward
about the origins of language itself. An interesting question to consider is:
Can people communicate with computers using human (natural) languages?
In other words, can a computer be made to understand language? Think about
that for a few moments.

To make the question of language understanding more concrete, think of your

Scribbler robot. So far, you have controlled the behavior of the robot by
writing Python programs for it. Is it possible to make the Scribbler understand
English so that you could interact with it? What would an interaction with
Scribbler look like? Obviously, you would not expect to have a conversation
with the Scribbler about the dinner you ate last night. However, it would
probably make sense to ask it to move in a certain way. Or to ask whether it is
seeing an obstacle ahead.

Do this: Write down a series of short 1-word commands like: forward,

right, left, stop, etc. Create a vocabulary of commands and then write a
program that inputs a command at a time interprets it and makes the Scribbler
carry it out. For example:

You: forward
Scribbler: starts moving forward…
You: right
Scribbler starts turning right…
You: stop
…

Experiment with the behavior of the robot based on these commands and
think about the proper interpretation that may make its behavior more natural.

247
Chapter 10

You will find yourself making several assumptions about interpretation of

even the simplest commands in the exercise above. For example, what
happens when after you command the Scribbler to move forward, you ask it to
turn right? Should the Scribbler stop going forward or should it stop and then
start turning?

Decisions like these also give deep insights into our own abilities of
understanding language. You can also see that, as in the case of visual
perception, processing of language (or text) begins at a very primitive level:
words. If the input is speech, the basic units are electrical signals, perhaps
coming from a microphone. Just like processing individual pixels to try and
understand the contents of an image, one has to start at a low level of
representation for beginning to understand language.

Researchers working in the field of computational linguistics (or natural

language understanding) have proposed many theories of language
processing that can form the basis of a computational model for a Scribbler to
understand a small subset of the English language. In this section, we will
examine one such model which is based on the processing of syntax and
semantics of language interaction. Imagine, interacting with the Scribbler
using the following set of sentences:

You: do you see a wall?

Scribbler: No

You: Beep whenever you see a wall.

You: Turn right whenever you see a wall to your left.
You: Turn left whenever you see a wall to your right.
You: Move for 60 seconds.

[The Scribbler robot moves around for 60 seconds turning

whenever it sees a wall. It also beeps whenever it sees a
wall.]

Earlier, you have written Python programs that perform similar behaviors.
However, now imagine interacting with the robot in the fashion described.
From a physical perspective, imagine that you are sitting in front of a

248
 Artificial Intelligence

computer, and you have a Bluetooth connection to the robot. The first
question then becomes: Are you actually speaking or typing the above
commands? From an AI perspective, both modalities are possible: You could
be sitting in front of the computer and speaking into a microphone; or you
could be typing those commands on the keyboard. In the first instance, you
would need a speech understanding capability. Today, you can obtain
software (commercial as well as freeware) that will enable you to do this.
Some of these systems are capable of distinguishing accents, intonations, male
or female voices etc. Indeed, speech and spoken language understanding is a
fascinating field of study that combines knowledge from linguistics, signal
processing, phonology, etc.

You can imagine that the end result of speaking into a computer is a piece of
text that transcribes what you said. So, the question posed to the Scribbler
above: Do you see a wall? will have to be processed and then transcribed into
text. Once you have the text, that is, a string “Do you see a wall?” it can
be further processed or analyzed to understand the meaning or the content of
the text. The field of computational linguistics provides many ways of
syntactic parsing, analyzing, and extracting meaning from texts. Researchers
in AI itself have developed ways of representing knowledge in a computer
using symbolic notations (e.g. formal logic). In the end, the analysis of the
text will result in a getIR() or getObstacle() command to the Scribbler
robot and will produce the response shown above.

Our goal of bringing up the above scenario here is to illustrate to you various
dimensions of AI research that can involve people from many different
disciplines. These days, it is entirely possible even for you to design and build
computer programs or systems that are capable of interacting with robots
using language.

Game Playing

In the early history of AI, scientists posed several challenging tasks which if
performed by computers could be used as a way of demonstrating the
feasibility of machine intelligence. It was common practice to think of games

249
Chapter 10

in this realm. For example, if a computer could play a game, like chess, or
checkers, at the same level or better than humans we would be convinced into
thinking that it was indeed feasible to think of a computer as a possible
candidate for machine intelligence. Some of the earliest demonstrations of AI
research included attempts at computer models for playing various games.
Checkers and chess seemed to be the most popular choices, but researchers
have indulged themselves into examining computer models of many popular
games: Poker, Bridge, Scrabble, Backgammon, etc.

Two‐person zero‐sum games: Chess, Tic Tac Toe, Konane

In many games, it is now possible for computer models to play at the highest
levels of human performance. In Chess, for example, even though the earliest
programs handily beat novices in the 1960's, it wasn't until 1996 when an
IBM computer Chess program, named Deep Blue, beat the world champion
Gary Kasparov at a tournament-level game, though Kasparov did manage to
win the match 4-2. A year later, in New York, Deep Blue beat Kasparov in a 6
game match representing the very first time a computer defeated the best
human player in a classical style game of Chess. While these
accomplishments are worthy of praise it also now clear that the quest for
machine intelligence is not necessarily answered by computer game playing.
This has resulted in much progress in game playing systems and game playing
technology which is now a multi-billion dollar industry.

It turns out that in many Chess-like games the general strategy for a computer
to play the game is very similar. Such games are classified as two-person
zero-sum games: two people/computers play against each other and the result
of the game is either a win for one player and loss for the other, or it is a draw.

250
 Artificial Intelligence

In many such games, the basic strategy for making the next move is simple:
look at all the possible moves I have and for each of them all the possible
moves the other player might have and so on until the very end. Then, trace
back from wins (or draws) and make the next move based on those desirable
outcomes. You can see this even in simple games like Tic Tac Toe where it is
easy to mentally look ahead future moves and then make more informed
decisions about what to do next. The best way to understand this is to actually
write a program that plays the game.

Tic Tac Toe

Also known as Noughts and Crosses or Hugs and Kisses, Tic Tac Toe is a
popular children’s game (see description on right). We will develop a program
that can be used to play this game against a person. Almost any board game
can be programmed using the basic loop shown below:

def play():
# Initialize board
board = makeBoard()

# set who moves first/next: X always moves first

player = 'X'

# Display the initial board

display(board) Tic Tac Toe Rules

# The game loop Two players, O and X, take

while (not gameOver(board)): turns filling the squares in a
move(board, player)
display(board) 3×3 grid. X always going first.
player = opponent(player) The player who succeeds in
placing three pieces in a
# game over, show outcome
winningPiece = winner(board) horizontal, vertical or diagonal
row wins the game.
if winningPiece != 'Tie':
print winningPiece, "won."
else:
print "It is a tie."

251
Chapter 10

The function above can be used to play a round of any two-person board
game. The variable player is the player (or piece) whose move is next. We
are already using the Tic Tac Toe piece ‘X’ in the function above. Six basic
functions (shown highlighted above) make up the basic building blocks of the
game. For Tic Tac Toe, they can be defined as:

1. makeBoard(): Returns a fresh new board representing the start of the

game. For Tic Tac Toe, this function will return an empty board
representing the nine squares.
2. displayBoard(board): Displays the board on the screen for the user
to see. The display can be as simple or elaborate as you wish. It is
good to start with the easiest one you can write. Later you can make it
fancier.
3. opponent(player): Returns the opponent of the current player/piece.
In Tic Tac Toe, if the player is X, it will return an O, and vice versa.
4. move(board, player): Updates the board by making one move for
the player. If the player is the user, it will input the move from the
user. If the player is the computer, it will decide how to make the best
move. This is where the smarts will come in.
5. gameOver(board): Returns True if there are no more moves left to be
made, False otherwise.
6. winner(board): Examines the board and returns the winning piece or
that the game is not yet over, or that it is a tie. In Tic Tac Toe, it will
return either an X, O, a blank (representing game is not over yet), or a
TIE.

We will need a few more functions to complete the game but these six form
the basic core. We will write them first.

The game itself consists of nine squares that start out being empty. Players
choose a game piece: an ‘X’ or an ‘O’. We will assume that ‘X’ always goes
first. The first thing to do then is to design a representation of the game board.
We will use the following simple representation:

board = [' ',' ',' ',' ',' ',' ',' ',' ',' ']

252
 Artificial Intelligence

This is a list of 9 1-character strings. Note that we are using this linear
representation of the board instead of a 2-dimensional one. However, as you
will see, this representation makes it easier to do many manipulations for the
game. During play, the board can be displayed in its natural format. Below,
we show two functions: one creates a fresh new board each time it is called;
and one displays it:

def makeBoard():
# A 3x3 board is represented as a list of 9 elements.
# We will use the following numbering to locate a square
# 0 | 1 | 2
# ---|---|---
# 3 | 4 | 5
# ---|---|---
# 6 | 7 | 8

return [' ',' ',' ',' ',' ',' ',' ',' ',' ']

def display(board):
for i in range(0, 9, 3):
if i > 0:
print '---|---|---'
print " %c | %c | %c "%(board[i],board[i+1],board[i+2])
print

One advantage of writing the display function as shown is that it gives us a

quick way of creating and displaying the game. Later, when you are done, you
can write a fancier version that displays the game graphically (see Exercises).
With the above functions, we can easily create a fresh new board and display
it as follows:

board = makeBoard()
display(board)

To determine the opponent of a given piece is simple enough:

253
Chapter 10

def opponent(player):
if player == "X":
return "O"
else:
return "X"

Next, we have to write the function move. It first determines whose move it is.
If it is the user’s move, it will input the move from the user. Otherwise, it will
determine the best move for the computer. Then it will actually make the
move. As a first design, we will let move make a random choice out of the
possible moves for the computer. Later, we will make a more informed
decision. The choice function is available in the random library module:

from random import choice

You = ‘X’
Me = ‘O’
def move(board, player):

if player == You: # user's move?

square = input("Enter your move: ") – 1
else: # my turn
# player is the computer, make a random choice
square = choice(possibleMoves(board, player))

# place player's piece on the chosen square

applyMove(board, player, square)

We have set the global variables You and Me to specific pieces. This is a
simplification for now. Later you can come back and rewrite them so that for
each game, the user gets to select their piece. In Tic Tac Toe, X always moves
first. So by making the user’s piece X, we have made an assumption that X
always goes first. Again, later we can come back and modify this (See
Exercises). Also notice that we are not doing any error checking in user input
to ensure that a legal move was input (see exercises).

The user inputs their move by entering a number from 1..9 where the square
numbering is as shown below. This is slightly different from our internal
numbering of squares and more natural for people. In the move function

254
 Artificial Intelligence

above, we subtract 1 from the input number so it maps to the proper square in
our internal scheme.

1 | 2 | 3
---|---|---
2 | 5 | 6
---|---|---
7 | 8 | 9

Again, we have simplified the interface for now. The exercises suggest how to
improve on this design.

The move function defined above requires two additional functions (shown
highlighted). These are also core functions in any board-based game and are
described below:

7. possibleMoves(board, player): Returns a list of possible moves

for the given player.
8. applyMove(board, player, square): Given a specific square and
a player/piece, this function actually applies the move on the board. In
Tic Tac Toe all one has to do is actually place the piece in the given
square. In other games, like Chess or Checkers, there may be pieces
that get removed.

In Tic Tac Toe all empty squares are possible places where a player can
move. Below, we write a function that, given a board, returns a list of all the
possible locations where a piece can be placed:

def possibleMoves(board):
return [i for i in range(9) if board[i] == ' ']

We are using list comprehensions (a Python feature) above to create the list of
possible moves. You may be more comfortable writing the following version
instead:

255
Chapter 10

def possibleMoves(board):
moves = []
for i in range(9):
if board[i] == ' ':
moves.append(i)
return moves

To complete the game playing program, we need to write two more functions
defined above. The winner function examines the board and determines who
won. It returns the winning piece (an 'X', 'O') or the string 'Tie'. In case
the game is not yet over, it returns a ' '. Below, we first define all the
winning positions in Tic Tac Toe, based on our board representation. Next,
we define the function itself to examine the board.

# These square triples represent wins (three in a row).

WINS = [[0, 1, 2],[3, 4, 5],[6, 7, 8], # the rows

[0, 3, 6],[1, 4, 7],[2, 5, 8], # the columns
[0, 4, 8],[2, 4, 6]] # diagonals

def winner(board):
for win in WINS:
posWinner = board[win[0]]
if (posWinner != ' ' and
posWinner == board[win[1]] and
posWinner == board[win[2]]):
return posWinner

# No winner yet, are there empty squares left?

for i in range(9):
if board[i] == ' ':
return ' '

# The board is full and no one has three in a row

return 'Tie'

256
 Artificial Intelligence

Last, the gameOver function can be written either by relying on the fact that
winner retruns a ' ' when the game is not yet over. Alternately, we can
write it using possibleMoves as follows:

def gameOver(board):
return len(possibleMoves(board)) == 0

Now, we have all the ingredients for a program to play a game of Tic Tac
Toe. The version above is simplified in many ways, and yet it captures the
essential elements of playing a round of Tic Tac Toe.

Do This: Implement the program above and play a few rounds of Tic Tac
Toe. When you play against the computer, you are anticipating possible
moves down the road and then playing your own moves with those in mind.

You probably had no problems beating the computer at the game. That is
because the computer is just picking its moves by random from a set of
possible moves:

# player is the computer, make a random choice

square = choice(possibleMoves(board, player))

This is the line in move where we can put some intelligence into the program.
However, the basic question that needs to be asked is: which of the possible
moves is the best one for me? Think about this for yourself a little. In Tic Tac
Toe, you play in two modes: defensively so as not to lose after the next move,
or offensively to try and win. If a win is imminent for you, you will of course
make that move, but if it isn’t (and neither is a loss) how do you select your
move? We will try and generalize this idea next in a way that will also be
applicable to other board games.

Let us delegate the responsibility of finding the best move to a function:

bestMove that will return the best move from among a set of possible moves.
That is, we can change the line in move as follows:

257
Chapter 10

# player is the computer, make a random choice

square = bestMove(board, player, possibleMoves(board, player))

If bestMove were to make a random choice (as above) we could write it as:

def bestMove(board, player, moves):

return choice(moves)

Imagine that you decided to go watch two players playing a game of Chess (or
Tic Tac Toe). However, you get there in the middle of the game. You walk
into the room, look at the board, and are able to evaluate how each player is
doing without knowing anything about how the board came to be. For
example, take a look at the Tic Tac Toe boards:

X | O | X O | X | X X | O | X X | |
---|---|--- ---|---|--- ---|---|--- ---|---|---
| O | | | O | X | O | |
---|---|--- ---|---|--- ---|---|--- ---|---|---
O | X | X X | | O | X | | |

1 2 3 4

In all the cases above, the computer is playing O and it is O’s turn next.
Examine each board and think about what the best move for O would be.

In case 1, the computer has to play defensively and place an O in square 6 to

avoid losing the game. In case 2, it can be on the offensive and recognize that
it will win the game by placing an O in square 5. In case 3, it has to play to
avoid losing. In the last case, the board is wide open. Of the eight possible
choices, is there a best move? From experience, you are probably going to
place the O in the center square (square 5). Why? Let us elaborate this by
looking ahead (see Figure on next page).

Think about how we could quantify each of the above board positions with
respect to O’s chances of winning. For each board, count the number of
possible winning positions still remaining for O and compare those for X. For
example, in the first board, O has 5 possible winning positions, and X has 6.

258
 Artificial Intelligence

We could say that O is 5 – 6 = -1 has one less possibility than X. If you did
this for the second and the third boards, you will see that both X and O are
even in the number of possible remaining winning positions (5 each).
However, in the fourth case, O has 5 possible winning positions remaining,
and X has only 4. Try and work out the remaining board positions to confirm

the scores shown above. Clearly, you can see that placing O in square 5 is the
best of all the possibilities. Try and elaborate the 5 possible moves for O in
case 2 above. You will find that one of the five leads to a win for O. Similarly,
when you elaborate the two options for O in case 1, you will notice that one
leads to a tie and the other to a loss for O. We can capture all these situations
in a function that takes a look at the board and returns a number representing a
score in favor of O: the higher the value returned, the more desirable the
board.

∞,
∞,
, /

That is, a win for the computer receives a high score, a loss receives a low
score. Otherwise, it is the difference in the number of open wins remaining for
each. We can capture this in Python as follows:

259
Chapter 10

INFINITY = 10
def evaluate(board):
# if board is a win for player, return INFINITY
piece = winner(board)
if piece == Me:
return INFINITY
elif piece == You:
return –INFINITY
else:
return openWins(board,Me) - openWins(board,You)

We define INFINITY as 10, a large enough number relative to the other values
that might be returned by evaluate. openWins looks at each winning triple
on the board and counts the number of openings for the given player.

def openWins(board, player):

possWins = 0
for position in WINS:
n = 0
for i in range(3):
if (board[position[i]] == player) or
(board[position[i]] == ' '):
n += 1
if n == 3: possWins += 1
return possWins

Do This: Implement the two functions above and then test them. Create
several board positions (use the ones from examples above) and confirm that
the board is being evaluated to the correct values.

Next, we can rewrite bestMove to take advantage of evaluate:

def bestMove(board, player, moves):

scores = []
for move in moves:
b = copy(board)
applyMove(b, player, move)
scores.append(evaluate(b))

return moves[scores.index(max(scores))]

260
 Artificial Intelligence

Notice how it takes each possible move, creates a new board with that move,
and then evaluates the score of the resulting board. Finally, it returns the move
with the highest score as the best move. Modify your program from above to
use this version of bestMove and play the game several times. You will notice
that there is a significant improvement in the computer’s playing ability. Can
you measure it in some way? (See Exercises).

The above rewrite of bestMove will make the program play significantly
better but there is still more room for improvement. In most board games
good players are able to mentally picture the game several moves ahead. In
many games, like Chess, certain recognizable situations lead to well
determined outcomes and so a great part of playing a successful game also
relies on the ability to recognize those situations.

Looking ahead several moves in a systematic manner is something computers

are quite capable of doing and hence anyone (even you!) can turn them into
fairly good players. The challenge lies in the number of moves you can look
ahead and in the limited capacity, if time to make the next move is limited,
how to choose among the best available options? These decisions lend
interesting character to computer game programs and continue to be a
constant source of fascination for many people. Lets us look at how your Tic
Tac Toe program can easily look at all the possible moves all the way to the
end of game in determining its next move (which, in most situations leads to a
draw, given the simplicity of the game).

When you look ahead a few moves, you take into account that your opponent
is going to try and beat you at every move. Your program, in trying to select
the best move, can look ahead at the opponent’s moves and take that into
consideration when choosing its best move. In fact, it can go further, all the
way to the end. The evaluate function we wrote above can be used
effectively to evaluate future board situations by assuming that when it is the
computer’s move, it will always try to pick the move that promises the highest
score in the future. However, when it examines the opponent’s moves, it has
to assume that the opponent is going to make the move that is worst for the
computer. In other words, when looking ahead, the computer is going to

261
Chapter 10

maximize its possible score while the opponent is going to minimize the
computer’s chances to win. This can be captured in the following:

def lookahead(board, player, MAX, level):

moves = possibleMoves(board)
if level <= 0 or len(moves)==0: # limit of look ahead
return evaluate(board)

if MAX: # computer’s move

V = -INFINITY
for m in moves:
b = copy(board)
b[m] = player
V =max(V,lookahead(b,opponent(player),1-MAX,level-1))
else: # opponent’s move
V = INFINITY
for m in moves:
b = copy(board)
b[m] = player
V =min(V,lookahead(b,opponent(player),1-MAX,level-1))
return V

The lookahead function defined above takes the current board, the player
whose turn it is, whether the player is the computer (one trying to maximize
its outcome) or the computer (one trying to minimize the computer’s
outcomes), and the levels still to look ahead, and computes a score based on
examining all the moves going forward to the limit of look ahead. In the
above function when MAX is 1 it represents the computer and 0 represents its
opponent. Thus, depending on the value of MAX, the evaluation is minimized
or maximized accordingly. Each time it looks ahead further, the level is
reduced by 1. The final value returned by lookahead can be used by
bestMove as follows:

LEVEL = 9
def bestMove(board, player, moves):

scores = []
for m in moves:

262
 Artificial Intelligence

b = copy(board)
b[m] = player
scores.append(lookahead(b,opponent(player),0,LEVEL-1)

return moves[scores.index(max(scores))]

As before, the move with the highest value is considered the best move. We
have set the value of LEVEL above at 9 (i.e. look 9 moves ahead!) implying
that each time it will look as far as the end of the game before making the
decision. There can only be a maximum of 9 moves in a Tic Tac Toe game.
The quality of the computer’s decision maker can in fact be adjusted by
lowering the value of LEVEL. At LEVEL = 1, it will be equivalent to the
version we wrote earlier that only used the evaluate function.

How many levels ahead one looks in a game like this can depend on the game
itself. Can you guess how many board situations the computer will have to
look at in doing a look ahead at LEVEL = 9 after the user’s first move? It will
be 40,320 different board situations! Why? Additionally, by the time it is the
computer’s second move, it will only need to look at 720 board positions.
This is because, in Tic Tac Toe, as the board gets filled, there are fewer
possible moves remaining. In fact, by the time it is the computer’s third move,
it only needs to look at a total of 24 boards. And, if the computer makes it to
its fourth move, it will only have to look at two possible moves. Thus, an
exhaustive search for all the possible board positions until the end of the game
each time the computer has to make a move it will be examining a total of
41,066 board positions. However, if you consider a typical game of Chess, in
which each player makes an average of 32 moves and the number of feasible
moves available at any time averages around 10, you would soon realize that
the computer would have to examine something of the order of 10 board
positions before making a move! This, even for the fastest computers
available today, will take several gazillion years! More on that later. But, to
play an interesting two-person zero-sum game, it is not essential to look so far
ahead. You can vary the amount of look ahead by adjusting the value of
LEVEL in such programs.

263
Chapter 10

Do This: Implement lookahead as described above and compare how well

the computer plays against you. Try and vary the levels from 1, 2, …, to see if
there is any improvement in the computer’s play. Would you consider this
program intelligent?

The exercises at the end of the chapter will guide you in transforming the
above program into a more robust and even efficient game playing program
for Tic Tac Toe. However, study the program structure carefully and you will
be able to use the same strategy, including much of the core of the program, to
play many other two-person board games.

Smarter Paper Scissors Rock

In Chapter 7, you saw an example of a program that played the game of Paper
Scissors Rock against a human user. In that version, the program’s choice
strategy for picking an object was completely random. We reproduce that
section of the program here:

…
items = ["Paper", "Scissors", "Rock"]
…
# Computer makes a selection
myChoice = items[randint(0, 2)]
…

In the above program segment, myChoice is the program’s choice. As you can
see, the program uses a random number to select its object. That is, the
likelihood of picking any of the three objects is 0.33 or 33%. The game and
winning strategies for this game have been extensively studied. Some
strategies rely on detecting patterns in human choice behavior. Even though
we may not realize it there are patterns in our seemingly random behavior.
Computer programs can easily track such behavior patterns by keeping long
histories of player’s choices, detect them, and then design strategies to beat
those patterns. This has been shown to work quite effectively. It involves
recording player’s choices and searching through them. Another strategy is to

264
 Artificial Intelligence

study human choice statistics in this game. Before we present you with some
data, do the exercise suggested below:

Do This: Play the game against a few people, Play several dozen rounds.
Record the choices made by each player (just write a P/S/R in two columns).
Once done, compute the percentages of each object picked. Now read on.

It turns out that most casual human players are more prone towards picking
Rock than Paper or Scissors. In fact, various analyses suggest that 36% of the
time people tend to pick Rock, 30% Paper, and 34% Scissors. This suggests
that RPS is not merely a game of chance there is room for some strategies at
winning. Believe it or not, there are world championships of RPS held each
year. Even a simple game like this has numerous possibilities. We can use
some of this information, for instance, to make our program smarter or better
adept at playing the game. All we have to do is instead of using a fair 33%
chance of selecting each object we can skew the chances of selection based on
people’s preferences. Thus, if 36% of the time people tend to pick Rock, it
would be better for our program to pick Paper 36% of the time since Paper
beats Rock. Similarly, our program should pick Scissors 30% of the time to
match the chance of beating Paper, and pick Rock 34% of the time to match
the chances of beating Paper. We can bias the random number generator using
these percentages as follows:

First generate a random number in the range 0..99

If the number generated is in the range 0..29, select Scissors (30%)
If the number generated is in the range 30..63, select Rock (34%)
If the number generated is in the range 64..99, select Paper (36%)

The above strategy of biasing the random selection can be implemented as

follows:

265
Chapter 10

def mySelection():

# First generate a random number in the range 0..99

n = randrange(0, 100)

# If the n is in range 0..29, select Scissors

if n <= 29:
return "Scissors"
elif n <= 63:
# if n in range 30..63, select Rock
return "Rock"
else:
return "Paper"

Do This: Modify your RPS program from Chapter 7 to use this strategy. Play
the game several times. Does it perform much better that the previous
version? You will have to test this by collecting data from both versions
against several people (make sure they are novices!).

Another strategy that people use is based upon the following observation:

After many rounds, people tend to make the move that would have beaten
their own previous move.

Say a player picks Paper. Their next pick will be Scissors. A computer
program or a player playing against this player should then pick Rock to beat
Scissors. Note that the relationship between the choices is cyclical. Paper
beats Rock, Rock beats Scissors, and Scissors beat Paper. Therefore, since the
player’s previous move was Paper, your program can pick Rock in
anticipation of the player’s pick of Scissors. Try to think over this carefully
and make sure your head is not spinning by the end of it. If a player can spot
this they can use this as a winning strategy. We will leave the implementation
of this strategy as an exercise. The exercises also suggest another strategy.

The point of the above examples is that using strategies in your programs you
can make your programs smarter or more intelligent. Deliberately, we have
started to use the term intelligence a little more loosely than what Alan Turing

266
 Artificial Intelligence

implied in his famous essay. Many people would argue that these programs
are not intelligent in the ultimate sense of the word. We agree. However,
writing smarter programs is a natural activity. If the programs incorporate
strategies or heuristics that people would use when they are doing the same
activity, then the programs have some form of artificial intelligence in them.
Even if the strategy used by the program is nothing like what people would
use, but it would make the program smarter or better, we would call it
artificial intelligence. Many people would disagree with this latter claim. To
some, the quest for figuring out intelligence is limited to the understanding of
intelligence in humans (and other animals). In AI both points of view are quite
prevalent and make for some passionate debates among scholars.

Discussion

The very idea of considering a computer as an intelligent device has its

foundations in the general purpose nature of computers. By changing the
program the same computer can be made to behave in many different ways.
At the core of it a computer is just a symbol manipulator: manipulating
encodings for numbers, or letters, or images, etc. It is postulated that the
human brain is also a symbol manipulator. The foundations of AI lie in the
fact that most intelligent systems are physical symbol systems and since a
computer is a general purpose symbol manipulator, it can be used for studying
or simulating intelligence.

267
Chapter 10

Myro Review

No new Myro functions were introduced in the Chapter. In the completed

version there will be some neural network functions described here.

Python review

choice(LIST)
Returns a randomly selected element from LIST. This function is to be
imported from the random module.

List Comprehensions
A short and elegant way to construct lists. See Python documentation for
more information.

268
 Artificial Intelligence

Exercises

1. Read Alan Turing’s paper Computing Machinery and Intelligence. You can
easily find a copy of it by searching on the web.

2. Do a web search for “Searle Chinese Room argument” to locate

Philosopher John Searle’s arguments that no matter how intelligent a
computer or a program gets, it will never have a “mind”.

3. Rewrite display for Tic Tac Toe game to display the board graphically.

4. Design a language of one-word English commands for the Scribbler. Write

a program to input one command at a time, interpret it, and then execute the
command on the Scribbler.

5. Extend the language from Exercise 4 to include queries (e.g. wall?) and
then modify your program to incorporate such queries.

6. Do a survey of speech understanding systems.

7. Do a survey of computational linguistics.

8. In the Tic Tac Toe program designed in this Chapter, we assumed that the
user always plays an X. Modify your program so that it gives the user a choice
at the beginning of the game. Further, at the end of each game, the pieces are
swapped.

9. In the function move defined for Tic Tac Toe, the program accepts whatever
the user inputs for their move. Try the program and instead of entering a valid
move, enter your name instead. What happens? Such an errors might be easily
detected since it will halt the program’s execution. However, next try entering
a number from 1-9 using a spare position that is already occupied in the
board. What happens? Modify the function to accept only correct moves. If

269
Chapter 10

the user enters an incorrect move, the program should point that out and give
the user another chance.

10. The function gameOver can make use of the winner function to make its
decision in the Tic Tac Toe program. Rewrite gameOver to do this.

11. One way to measure how one strategy compares against another is to play
it against another strategy over and over again recording the number of wins,
losses, and draws. Modify your Tic Tac Toe or RPS program to substitute the
second strategy for the user (instead of taking input from the user, it uses a
function that implements the second strategy. Add statements to play the
game many times (say 1000) and record the wins, losses, and draws. You may
also want to suppress all board output since the game is being played
completely inside the program. Do a comparison of all the strategies discussed
in this Chapter and how they compare against each other.

270
Artificial Intelligence

271
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Computers &
Computation
Home computers are being called upon to perform many new functions, including the
consumption of homework formerly eaten by the dog.
Doug Larson

Computer Science is no more about computers than astronomy is about telescopes.

Edsger W. Dijkstra

What is the central core of computer science?My answer is simple -it is the art of
programming a computer. It is the art of designing efficient and elegant methods of
getting a computer to solve problems, theoretical or practical, small or large, simple
or complex. It is the art of translating this design into an effective and accurate
computer program.
C.A.R. Hoare
Opposite page: The XO Laptop
Photo courtesy of OLPC Project (www.olpc.org)

273
Chapter 11

Today, there are more computers than people on most college campuses in the
United States. The laptop computer shown on the previous page is the XO
laptop developed by the One Laptop Per Child Project (OLPC). It is a low
cost computer designed for kids. The OLPC project aims to reach out to over
2 billion children all over the world. They aim to get the XO computers in
their hands to help in education. Their mission is to bring about a radical
change in the global education system by way of computers. Of course a
project with such a grandiose mission is not without controversy. It has had a
bumpy ride since its inception. The technological issues were the least of their
problems. They have had to convince governments (i.e. politicians) to buy-in
into the mission. Several countries have agreed to buy-in and then reneged for
all kinds of reasons. The laptops are not available for open purchase within
the United States which has led to other socio-political controversies.
Regardless, in the first year of its release the project aims to have over 1
million XO’s in the hands of children in many developing countries. The
thing about technology that has always been true is that a good idea is
infective. Other players have emerged who are now developing ultra-low cost
computers for kids. It will not be long before other competitive products will
be available to all. The bigger questions are: What kind of change will this
bring about in the world? How will these computers be used in teaching
elementary and middle school children? Etc. You may also be wondering if
your Scribbler robot can be controlled by the XO. It can.

You have been writing programs to control your robot through a computer.
Along the way you have seen many different ways to control a robot and also
ways of organizing your Python programs. Essentially you have been
engaging in the activity commonly understood as computer programming or
computer-based problem solving. We have deliberately refrained from using
those terms because the general perception of these conjures up images of
geeks with pocket-protectors and complex mathematical equations that seem
to be avoidable or out of reach for most people. Hopefully you have
discovered by now that solving problems using a computer can be quite
exciting and engaging. Your robot may have been the real reason that
motivated you into this activity and that was by design. We are confessing to
you that robots were used to attract you to pay some attention to computing.

274
 Computers & Computation

You have to agree, if you are reading this, that the ploy worked! But
remember that the reason you are holding a robot of your own in your hand is
also because of computers. Your robot itself is a computer. Whether it was a
ploy or not you have assimilated many key ideas in computing and computer
science. In this chapter, we will make some of these ideas more explicit and
also give you a flavor for what computer science is really all about. As
Dijkstra puts it, computer science is no more about computers than astronomy
is about telescopes.

Computers are dumb

Say you are hosting an international exchange student in your home. Soon
after her arrival you teach her the virtues of PB&J (Peanut Butter & Jelly)
sandwiches. After keenly listening to you, her mouth starts to water and she
politely asks you if you can share the recipe with her. You write it down on a
piece of paper and hand it to her.

Do This: Go ahead; write down the recipe to make a PB&J sandwich.

Seriously, do try to write it down. We insist!

OK, now you have a recipe for making PB&J sandwiches.

Do This: Go ahead, use your recipe from above to make yourself a PB&J
sandwich. Try to follow the instructions exactly as literally as you can. If you
successfully managed to make a PB&J sandwich, congratulations! Go ahead
and enjoy it. Do you think your friend will be able to follow your recipe and
enjoy PB&J sandwiches?

You have to agree, writing a recipe for PB&J sandwiches seemed trivial at
first, but when you sit down to write it you have no choice but to question
several assumptions: does she know what peanut butter is? Should I
recommend a specific brand? Ditto for jelly? By the way, did you forget to
mention it was grape jelly? What kind of bread to use? Will it be pre-sliced?
If not, you need a knife for slicing the loaf? Did you specify how thick the

275
Chapter 11

slices should be? Should she use the same knife for spreading the peanut
butter and the jelly? Etc. The thing is, at such a level of detail you can go on
and on...does your friend know how to use a knife to spread butter or jelly on
a slice? Suddenly, a seemingly trivial task becomes a daunting exercise. In
reality, in writing down your recipe, you make several assumptions: she
knows what a sandwich is, it involves slices of bread, spreading things on the
slices, and slapping them together. There, you have a sandwich!

Think of the number of recipes that have been published in cookbooks all
over the world. Most good cookbook authors start with a dish, write down a
recipe, try it a number of times and refine it in different ways. Prior to
publication, the recipe is tested by others. After all, the recipe is for others to
follow and recreate a dish. The recipe is revised and adjusted based on
feedback by recipe testers. The assumption that cookbook authors make is that
you will have the competence to follow the recipe and recreate the dish to
your satisfaction. It may never result in the same dish that the author prepared.
But it will give you a base to improvise upon. Wouldn't it be nice if there was
an exact way of following a recipe so that you end up with the exact same
result as the cookbook author every time you made that dish? Would it? That
may depend on your own tastes and preferences. Just for fun, here is a recipe
for some yummy Saffron Chicken Kabobs.

Saffron Chicken Kabobs

Ingredients

1 lb boneless chicken breast, cubed into 1-2 inch pieces

1 medium onion, sliced
1 tsp saffron threads
1 lime
1 tbsp olive oil
Salt and black pepper to taste

276
 Computers & Computation

Preparation

1. Mix the chicken and onions in a non-reactive bowl.

2. With your fingers crush and add saffron threads.
3. Add the juice of the lime, olive oil, and salt and pepper.
4. Marinade in the refrigerator for at least 30 min (or overnight).
5. Preheat a grill (or oven to 400 degrees).
6. Skewer kabobs, discarding the onion slices. Or place everything in a
lined baking sheet if using oven.
7. Grill/bake for 12-15 min until done.

In cooking recipes, like the ones above, you can assume many things: they
will be used by people (like you and me); they will be able to follow them;
perhaps even improvise. For instance, in the recipe above, we do not specify
that one will need a knife to cut the chicken, onions, or the lime; or that you
will need a grill or an oven; etc. Most recipes assume that you will be able to
interpret and follow the recipe as written.

Computer programs are also like recipes, to some extent. Think of the
program you wrote for choreographing a robot dance, for instance. We have
reproduced the version from Chapter 3 here:

# File: dance.py
# Purpose: A simple dance routine
# First import myro and connect to the robot

from myro import *

initialize("com5")

# Define the new functions...

def yoyo(speed, waitTime):

forward(speed, waitTime)
backward(speed, waitTime)
stop()

def wiggle(speed, waitTime):

motors(-speed, speed)

277
Chapter 11

wait(waitTime)
motors(speed, -speed)
wait(waitTime)
stop()

# The main dance program

def main():
print "Running the dance routine..."
yoyo(0.5, 0.5)
wiggle(0.5, 0.5)
yoyo(1, 1)
wiggle(1, 1)
print "...Done"

main()

In many ways, this program above is like a recipe:

To do a robot dance

Ingredients

1 function yoyo for the robot to go back and forth at a given speed
1 function wiggle that enables the robot to wiggle at a given speed

Preparation

1. yoyo at speed 0.5, wait 0.5

2. wiggle at speed 0.5, wait 0.5
3. yoyo at speed 1, wait 1
4. wiggle at speed 1, wait 1

Further, you could similarly specify the steps involved in doing the yoyo and
wiggle motions as a recipe. This may seem like a trivial example, but it
makes two very important points: a computer program is like a recipe in that it
lays out the list of ingredients and a method or steps for accomplishing the
given task; and, like a recipe, its ingredients and the steps require careful pre-

278
 Computers & Computation

planning and thought. Importantly, computer programs are different from

recipes in one aspect: they are designed to be followed by a computer!

A computer is a dumb device designed to follow instructions/recipes. We will

save the technical details of how a computer does what it does for a later
course. But it is almost common knowledge that everything inside is
represented as 0's and 1's. Starting from 0's and 1's one can design encoding
schemes to represent numbers, letters of the alphabet, documents, images,
movies, music, etc. and whatever other abstract entities you would like to
manipulate using a computer. A computer program is ultimately also
represented as a sequence of 0's and 1's and it is in this form that most
computers like to follow recipes. However limiting or degenerate this might
sound it is the key to the power of computers. Especially when you realize
that it is this simplification that enables a computer to manipulate hundreds of
millions of pieces of information every second. The price we have to pay for
all this power is that we have to specify our recipes as computer programs in a
rather formal and precise manner. So much so that there is no room for
improvisation: no pinch of salt vagaries, as in cooking recipes, is acceptable.
This is where programming languages come in. Computer scientists specify
their computational recipes using programming languages. You have been
using the programming language Python to write your robot programs. Other
examples of programming languages are Java, C++ (pron.: sea plus plus), C#
(pron.: sea sharp), etc. There are well over 2000 programming languages in
existence!

Do This: Can you find out how many programming languages there are?
What are the ten most commonly used programming languages?

Prior to the existence of programming languages computers were

programmed using long sequences of 0's and 1's. Needless to say it drove
several people crazy! Programming languages, like Python, enable a friendlier
way for programmers to write programs. Programming languages provide
easy access to encodings that represent the kinds of things we, humans, relate
to. For example, the Python statement:

meaningOfLife = 42

279
Chapter 11

is a command for the computer to associate the value, 42 with the name
meaningOfLife. This way, we can ask the computer to check that it is indeed
42:

if meaningOfLife == 42:
speak("Eureka!")
else:
speak("What do we do now?")

Once again, it would be good to remind you that the choice of the name,
meaningOfLife, doesn't really mean that we are talking about the meaning of
life. We could as well have called it timbuktoo, as in:

timbuktoo = 42

You see, computers are truly dumb!

It is really up to us, the programmer, to ensure that we use our names

consistently and choose them, in the first place, carefully. But, by creating a
language like Python, we have created a formal notation so that when
translated into 0's and 1's each statement will mean only one thing, no other
interpretations. This makes them different from a cooking recipe.

Robot goes to buy fresh eggs

Recipes, however, form a good conceptual basis for starting to think about a
program to solve a problem. Say, you have in mind to make your favorite
Apple Strudel. You know you will need apples. Perhaps it is the apple season
that prompted the thought in the first place. You will also need pastry. But
when you get down to it, you will need that recipe you got from your
grandma.

Whenever we are asked to solve a problem using a computer, we begin by

laying out a rough plan for solving the problem. That is, sketch out a strategy.
This is further refined into specific steps, perhaps even some variables are

280
 Computers & Computation

identified and named, etc. Once you convince yourself that you have a way of
solving the problem, what you have is an algorithm

The idea of an algorithm is central to computer science so we will spend some

time here developing this notion. Perhaps the best way to relate to it is by an
example. Assume that a robot goes into a
grocery store to buy a dozen fresh eggs.
Assuming it is capable of doing this, how
will it ensure that it has selected the freshest
eggs available?

Your personal robot is probably not up to

this kind of task but imagine that it was.
Better yet, leave the mechanics aside, let us
figure out how you would go and buy the freshest eggs. Well, you would
somehow need to know what today's date is. Assume it is September 15, 2007
(why this date? it'll become clear soon!). Now you also know that egg cartons
typically carry a freshness date on them. In fact, USDA (the United States
Department of Agriculture) offers voluntary, no cost, certification programs
for egg farms. An egg farmer can volunteer to
participate in USDA's egg certification Egg Carton Labeling
program whereby the USDA does regular
inspections and also provides help in
categorizing eggs by various sizes. For
example, eggs are generally classified as
Grade AA, Grade A, or Grade B. Most grocery
stores carry Grade A eggs. They can also come
in various sizes: Extra Large, Large, Small,
etc. What is more interesting is that the carton
labeling system has some very useful
information encoded on it.

Every USDA certified egg carton has at least

three pieces of information (see picture on the
right): a "sell by" date (or a "use by date" or a

281
Chapter 11

"best by" date), a code identifying the specific farm the eggs came from, and a
date on which the eggs were packed in that carton. Most people buy eggs by
looking at the "sell by" date or the "best by" date. However the freshness
information is really encoded in the packed on date. To make things more
confusing, this date is encoded as the day of the year.

For example, take a look at the top carton shown on the previous page. Its
"sell by" date is October 4. "P1107" is the farm code. This carton was packed
on the 248th day of the year. Further, USDA requires that all certified eggs be
packed within 7 days of being laid. Thus, the eggs in the top carton were laid
somewhere between day 241 and day 248 of 2007. What dates correspond to
those dates?

Next, look at the bottom carton. Those eggs have a later "sell by" date
(October 18) but an earlier packed date: 233. That is those eggs were laid
somewhere between day 226 and day 233 of 2007.

Which eggs are fresher?

Even though the "sell by" date on the second carton is two weeks later, the
first carton contains fresher eggs. In fact, the eggs in the upper carton were
laid at least two weeks later!

The packed on date is encoded as a 3-digit number. Thus eggs packed on

January 1 will be labeled: 001; eggs packed on December 31, 2007 will be
labeled: 365.

Do This: Go to the USDA web site (www.usda.gov) and see if you can find
out which farm the two eggs cartons came from.

For a robot, the problem of buying the freshest eggs becomes that of figuring
out, given a packed on date, what the date was when the eggs were packed?

Fasten your seatbelts, we are about to embark on a unique computational

voyage...

282
 Computers & Computation

Designing an algorithm The Etymology of Algorithm

So far, we have narrowed the problem The word algorithm, an anagram of

down to the following specifications: logarithm, is believed to have been
derived from Al‐Khowarizmi, a
Input mathematician who lived from 780‐
3-digit packed on date encoding 850 AD. His full name was Abu
Ja’far Muḥammad ibn Mūsā al‐
Output
Khwārizmī, (Mohammad, father of
Date the eggs were packed
Jafar, son of Moses, a
Khwarizmian). Much of the
For example, if the packed on date was
encoded as 248, what will be the actual mathematical knowledge of
date? medieval Europe was derived from
Latin translations of his works.
Well, that depends. It could be
September 4 or September 5 depending
on whether the year was a leap year or
not. Thus, it turns out, that the problem
above also requires that we know
which year we were talking about.
Working out one or two sample
problems is always a good idea because
it helps identify missing information
that may be critical to solving the
problem. Given that we do need to
know the year, we can ask the user to
enter that at the same time the 3-digit
code is entered. The problem
specification then becomes:
In 1983, The Soviet Union issued
Input the stamp shown above in honor of
3-digit packed on date encoding his 1200th anniversary.
Current year

283
Chapter 11

Output
Date the eggs were packed

Example:
Input: 248, 2007
Output: The eggs were packed on September 5, 2007

Any ideas as to how you would solve this problem? It always helps to try and
do it yourself, with pencil and paper. Take the example above, and see how
you would arrive at the output date. While you are working it out, try to write
down your problem solving process. Your algorithm or recipe will be very
similar.

Suppose we are trying to decode the input 248, 2007. If you were to do this by
hand, using a pen and paper, the process might go something like this:

The date is not in January because it has 31 days and 248 is

much larger than 31.
Lets us subtract 31 out of 248: 248 - 31 = 217

217 is also larger than 28, the number of days in February,

2007.
So, let us subtract 28 from 217: 217 - 28 = 189

189 is larger than 31, the number of days in March.

Subtract 31 from 189: 189 - 31 = 158

158 is larger than 30, the number of days in April.

So: 158 - 30 = 128

128 is larger than 31, the number of days in May.

Hence: 128 - 31 = 97

97 is larger than 30, the number of days in June.

97 - 30 = 67

67 is larger than 31, the number of days in July.

67 - 31 = 36

284
 Computers & Computation

36 is larger than the number of days in August (31).

36 - 31 = 5

5 is smaller than the number of days in September.

Therefore it must be the 5th day of September.

The answer is: 248th day of 2007 is September 5, 2007.

That was obviously too repetitious and tedious. But that is where computers
come in. Take a look at the process above and see if there is a pattern to the
steps performed. Sometimes, it is helpful to try another example.

Do This: Suppose the input day and year are: 56, 2007. What is the date?

When you look at the sample computations you have performed, you will see
many patterns. Identifying these is the key to designing an algorithm.
Sometimes, in order to make this easier, it is helpful to identify or name the
key pieces of information being manipulated. Generally, this begins with the
inputs and outputs identified in the problem specification. For example, in this
problem, the inputs are: day of the year, current year. Begin by assigning
these values to specific variable names. That is, let us assign the name day to
represent the day of the year (248 in this example), and year as the name to
store the current year (2007). Notice that we didn't choose to name any of
these variables timbuktu or meaningOfLife!

Also, notice that you have to repeatedly subtract the number of days in a
month, starting from January. Let us assign a variable named, month to keep
track of the month under consideration.

Next, you can substitute the names day and year in the sample computation:

Input:
day = 248
year = 2007

# Start by considering January

month = 1

285
Chapter 11

The date is not in month = 1 because it has 31 days and 248 is

much larger than 31.
day = day - 31

# next month
month = 2

day (= 217) is also larger than 28, the # of days in month = 2

day = day - 28

# next month
month = 3
day (= 189) is larger than 31, the # of days in month = 3.
day = day - 31

# next month
month = 4
day (= 158) is larger than 30, the # of days in month = 4.
day = day - 30

# next month
month = 5
day (= 128) is larger than 31, the # of days in month = 5.
day = day - 31

# next month
month = 6
day (= 97) is larger than 30, the # of days in month = 6.
day = day = 30

# next month
month = 7
day (= 67) is larger than 31, the # of days in month = 7.
day = day - 31

# next month
month = 8
day (= 36) is larger than the # of days in month = 8.
day = day - 31

# next month
month = 9
day (= 5) is smaller than the # of days in month = 9.

286
 Computers & Computation

Therefore it must be the 5th day of September.

The answer is: 9/5/2007

Notice now how repetitious the above process is. The repetition can be
expressed more concisely as shown below:

Input:
day
year

# start with month = 1, for January

month = 1
repeat
if day is less than number of days in month
day = day – number of days in month
# next month
month = month + 1
else
done

Output: day/month/year

It is now starting to look like a recipe or an algorithm. Go ahead and try it

with the sample inputs from above and ensure that you get correct results.
Additionally, make sure that this algorithm will work for boundary cases: 001,
365.

Thirty days hath September

We can refine the algorithm above further: one thing we have left unspecified
above is the computation of the number of days in a month. This information
has to be made explicit for a computer to be able to follow the recipe. So, how
do we compute the number of days in a month? The answer may seem simple.
Many of you may remember the following poem:

287
Chapter 11

Thirty days hath September

April, June, and November
All the rest have thirty-one
Except for February alone
Which hath twenty-eight days clear
And twenty-nine in each leap year

From a design perspective, we can assume that we have an ingredient, a

function in this case, called daysInMonth that, given a month and a year will
compute and return the number of days in the month. That is, we can refine
our algorithm above to the following:

Ingredients:
1 function daysInMotnh(m, y): returns the number of days in
month, m in year y.

Input:
day
year

# start with month = 1, for January

month = 1
repeat
if day is less than number of days in month
day = day - daysInMonth(month, year)
# next month
month = month + 1
else
done

Output: day/month/year

Now, we do have to solve the secondary problem:

Input
month, M
year, Y

288
 Computers & Computation

Output
Number of days in month, M in year, Y

On the surface this seems easy, the poem above specifies that April, June,
September, and November have 30 days, and the rest, with the exception of
February have 31. February has 28 or 29 days depending upon whether it falls
in a leap year or not. Thus, we easily elaborate a recipe or an algorithm for
this as follows:

Input:
m, y

if m is April (4), June(6), September(9), or November (11)

days = 30
else if m is February
if y is a leap year
days = 29
else
days = 28
else
(m is January, March, May, July, August, October, December)
days = 31

Output:
days

This still leaves out one more detail: how do we tell if y is a leap year?

First, try and answer the question, what is a leap year?

Again, we can refine the algorithm above by assuming that we have another
ingredient, a function: leapYear, that determines if a given year is a leap year
or not. Then we can write the algorithm above as:

Ingredients:
1 function leapYear(y)
returns True if y is a leap year, false otherwise

289
Chapter 11

Input:
m, y

if m is April (4), June(6), September(9), or November (11)

days = 30
else if m is February
if leapYear(y)
days = 29
else
days = 28
else
(m is January, March, May, July, August, October, December)
days = 31

Output:
days

Most of us have been taught that a leap year is a year that is divisible by 4.
That is the year 2007 is not a leap year, since 2007 is not divisible by 4, but
2008 is a leap year, since it is divisible by 4.

Do This: How do you determine if something is divisible by 4? Try your

solution on the year 1996, 2000, 1900, 2006.

Leap Years: Papal Bull

To design a recipe or an algorithm that determines if a number corresponding

to a year is a leap year or not is straightforward if you accept the definition
from the last section. Thus, we can write:

Input
y, a year

Output
True if y is a leap year, false otherwise

290
 Computers & Computation

Method
if y is divisible by 4
it is a leap year, or True
else
it is not a leap year, or False

However, this is not the complete story. The western calendar that we follow
is called the Gregorian Calendar which was adopted in 1582 by a Papal Bull
issued by Pope Gregory XIII. The Gregorian Calendar defines a leap year, by
adding an extra day, every fourth year. However, there is a 100-year
correction applied to it that makes the situation a little more complicated:
Century years are not leap years except when they are divisible by 400. That
is the years 1700, 1800, 1900, 2100 are not leap years even though they are
divisible by 4. However, the years 1600, 2000, 2400 are leap years. For more
information on this, see the exercises at the end of the chapter. Our algorithm
for determining if a year is a leap year can be refined as shown below:

Input
y, a year

if y is divisible by 400
it is a leap year, or True
else if y is divisible by 100
it is not a leap year, or False
else if y is divisible by 4
it is a leap year, or True
else
it is not a leap year, or False

Finally, we have managed to design all the algorithms or recipes required to

solve the problem. You may have noticed that we used some familiar
constructs to elaborate our recipes or algorithms. Next, let us take a quick
look at the essential constructs that are used in expressing algorithms.

291
Chapter 11

Essential components of an algorithm

Computer scientists express solutions to problems in terms of algorithms,

which are basically more detailed recipes. Algorithms can be used to express
any solution and yet are comprised of some very basic elements:

1. Algorithms are step-by-step recipes that clearly identify the inputs and
outputs
2. Algorithms name the entities that are manipulated or used: variables,
functions, etc.
3. Steps in the algorithm are followed in the order they are written (from
top to bottom)
4. Some steps can specify decisions (if-then) over the choice of some
steps
5. Some steps can specify repetitions (loops) of steps
6. All of the above can be combined in any fashion.

Computer scientists claim that solutions/algorithms to any problem can be

expressed using the above constructs. You do not need any more! This is a
powerful idea and it is what makes computers so versatile. From a larger
perspective, if this is true, then these can be used as tools for thinking about
any problem in the universe. We will return to this later in the chapter.

Programming Languages

Additionally, as you have seen earlier, in writing Python programs,

programming languages (Python, for example) provide formal ways of
specifying the essential components of algorithms. For example, the Python
language provides a way for you to associate values to variables that you
name, it provides a sequential way of encoding the steps, it provides the if-
then conditional statements, and also provides the while-loop and for-loop
constructs for expressing repetitions. Python also provides means for defining
functions and also ways of organizing groups of related functions into
libraries or modules which you can import and use as needed. As an example,

292
 Computers & Computation

we provide below, the Python program that encodes the leapYear algorithm
shown above:

def leapYear(y):
'''Returns true if y is a leap year, false otherwise.'''

if y %400 == 0:
return True
elif y % 100 == 0:
return False
elif y % 4 == 0:
return True
else:
return False

The same algorithm, when expressed in C++ (or Java) will look like this:

bool leapYear(int y) {
// Returns true if y is a leap year, false otherwise.
if (y % 400 == 0)
return true
else if (y % 100 == 0)
return false
else if (y % 4 == 0)
return true
else
return false
}

As you can see, there are definite syntactic variations among programming
languages. But, at least in the above examples, the coding of the same
algorithm looks very similar. Just to give a different flavor, here is the same
function expressed in the programming language CommonLisp.

(defun leapYear (y)

(cond
((zerop (mod y 400)) t)
((zerop (mod y 100)) nil)
((zerop (mod y 4)) t)
(t nil)))

293
Chapter 11

Again, this may look weird, but it is still expressing the same algorithm.

What is more interesting is that given an algorithm, there can be many ways
to encode it, even in the same programming language. For example, here is
another way to write the leapYear function in Python:

def leapYear(y):
'''Returns true if y is a leap year, false otherwise.'''

if ((y % 4 == 0) and (y % 100 != 0)) or (y % 400 == 0):

return True
else:
return False

Again, this is the same exact algorithm. However, it combines all the tests into
a single condition: y is divisible by 4 or by 400 but not by 400. The same
condition can be used to write an even more succinct version:

def leapYear(y):
'''Returns true if y is a leap year, false otherwise.'''

return ((y % 4 == 0) and (y % 100 != 0)) or (y % 400 == 0)

That is, return whatever the result is (True/False) of the test for y being a
leap year. In a way, expressing algorithms into a program is much like
expressing a thought or a set of ideas in a paragraph or a narrative. There can
be many ways of encoding an algorithm in a programming language. Some
seem more natural, and some more poetic, or both, and, like in writing, some
can be downright obfuscated. As in good writing, good programming ability
comes from practice and, more importantly, learning from reading well
written programs.

From algorithms to a working program

To be able to solve the fresh eggs problem, you have to encode all the
algorithms into Python functions and then put them together as a working
program. Below, we present one version:

294
 Computers & Computation

# File: fresheggs.py

def leapYear(y):
'''Returns true if y is a leap year, false otherwise.'''

return ((y % 4 == 0) and (y % 100 != 0)) or (y % 400 == 0)

def daysInMonth(m, y):

'''Returns the number of days in month, m (1-12)
in year, y.'''

if (m == 4) or (m == 6) or (m == 9) or (m == 11):
return 30
elif m == 2:
if leapYear(y):
return 29
else:
return 28
else:
return 31

def main():
'''Given a day of the year (e.g. 248, 2007),
convert it to the date (i.e. 9/5/2007)'''

#Input: day, year

day, year = input("Enter the day, year: ")

# start with month = 1, for January

month = 1

while day > daysInMonth(month, year):

day = day - daysInMonth(month, year)

# next month
month = month + 1

# done, Output: month/day/year

print "The date is: %1d/%1d/%4d" % (month, day, year)

main()

295
Chapter 11

If you save this program in a file, fresheggs.py, you will be able to run it
and test it for various dates. Go ahead and do this. Here are some sample
outputs:

Enter the day, year: 248, 2007

The date is: 9/5/2007

>>> main()
Enter the day, year: 12, 2007
The date is: 1/12/2007

>>> main()
Enter the day, year: 248, 2008
The date is: 9/4/2008

>>> main()
Enter the day, year: 365, 2007
The date is: 12/31/2007

>>> main()
Enter the day, year: 31, 2007
The date is: 1/31/2007

All seems to be good. Notice how we tested the program for different input
values to confirm that our program is producing correct results. It is very
important to test your program for a varied set of input, taking care to include
all the boundary conditions: first and last day of the year, month, etc. Testing
programs is a fine art in itself and several books have been written about the
topic. One has to ensure that all possible inputs are tested to ensure that the
behavior of the program is acceptable and correct. You did this with your
robot programs by repeatedly running the program and observing the robot's
behavior. Same applies to computation.

Testing and Error Checking

What happens, if the above program receives inputs that are outside the
range? What if the user enters the values backwards (e.g. 2007, 248 instead of
248, 2007)? What if the user enters her name instead (e.g. Paris, Hilton)? Now

296
 Computers & Computation

is the time to try all this out. Go ahead and run the program and observe its
behavior on some of these inputs.

Ensuring that a program provides acceptable results for all inputs is critical in
most applications. While there is no way to avoid what happens when a user
enters his name instead of entering a day and a year, you should still be able
to safeguard your programs from such situations. For example:

>>> main()
Enter the day, year: 400, 2007
That corresponds to the date: 14/4/2007

Obviously, we do not have a month numbered 14!

The thing that comes to rescue here is the realization that, it is your program
and the computer will only carry out what you have expressed in the program.
That is, you can include error checking facilities in your program to account
for such conditions. In this case, any input value for day that is outside the
range 1..365 (or 1..366 for leap years) will not be acceptable. Additionally,
you can also ensure that the program only accepts years greater than 1582 for
the second input value. Here is the modified program (we'll only show the
main function):

def main():
'''Given a day of the year (e.g. 248, 2007), convert
it to the date (i.e. 9/5/2007)'''

#Input: day, year

day, year = input("Enter the day, year: ")

# Validate input values...

if year <= 1582:
print "I'm sorry. You must enter a valid year
(one after 1582). Please try again."
return
if day < 1:
print "I'm sorry. You must enter a valid day
(1..365/366). Please try again."

297
Chapter 11

return
if leapYear(year):
if day > 366:
print "I'm sorry. You must enter a valid day
(1..365/366). Please try again."
return
elif day > 365:
print "I'm sorry. You must enter a valid day
(1..365/366). Please try again."
return

# input values are safe, proceed...

# start with month = 1, for January
month = 1

while day > daysInMonth(month, year):

day = day - daysInMonth(month, year)

# next month
month = month + 1

# done, Output: month/day/year

print "The date is: %1d/%1d/%4d" % (month, day, year)

main()

Here are the results of some more tests on the above program.

Enter the day, year: 248, 2007

The date is: 9/5/2007

>>> main()
Enter the day, year: 0, 2007
I'm sorry. You must enter a valid day (1..365/366). Please try
again.

>>> main()
Enter the day, year: 366, 2007
I'm sorry. You must enter a valid day (1..365/366). Please try
again.

>>> main()
Enter the day, year: 400, 2007

298
 Computers & Computation

I'm sorry. You must enter a valid day (1..365/366). Please try
again.

>>> main()
Enter the day, year: 248, 1492
I'm sorry. You must enter a valid year (one after 1582).
Please try again.

>>> main()
Enter the day, year: 366, 2008
The date is: 12/31/2008

Starting from a problem description it is a long and carefully planned journey

that involves the development of the algorithm, the encoding of the algorithm
in a program, and finally testing and improving the program. In the end you
are rewarded not just by a useful program, you have also honed your general
problem solving skills. Programming forces you to anticipate unexpected
situations and to account for them prior to encountering them which itself can
be a wonderful life lesson.

Modules to organize components

Often, in the course of designing a program, you end up designing

components or functions that can be used in many other situations. For
example, in the problem above, we wrote functions leapYear and
daysInMonth to assist in solving the problem. You will no doubt agree that
there are many situations where these two functions could come in handy (see
Exercises below). Python provides the module facility to help organize related
useful functions into a single file that you can then use over and over
whenever they are needed. For example, you can take the definitions of the
two functions and put them separately in a file called, calendar.py. Then,
you can import these functions whenever you need them. You have used the
Python import statement to import functionality from several different
modules: myro, random, etc. Well, now you know how to create your own.
Once you create the calendar.py file, you can import it in the fresheggs.py
program as shown below:

299
Chapter 11

from calendar import *

def main():
'''Given a day of the year (e.g. 248, 2007), convert
it to the date (i.e. 9/5/2007)'''

#Input: day, year

day, year = input("Enter the day, year: ")

# Validate input values...

if year <= 1582:
print "I'm sorry. You must enter a valid year
(one after 1582). Please try again."
return
if day < 1:
print "I'm sorry. You must enter a valid day
(1..365/366). Please try again."
return
if leapYear(year):
if day > 366:
print "I'm sorry. You must enter a valid day
(1..365/366). Please try again."
return
elif day > 365:
print "I'm sorry. You must enter a valid day
(1..365/366). Please try again."
return
# input values are safe, proceed...
# start with month = 1, for January
month = 1

while day > daysInMonth(month, year):

day = day - daysInMonth(month, year)

# next month
month = month + 1

# done, Output: month/day/year

print "The date is: %1d/%1d/%4d" % (month, day, year)

main()

300
 Computers & Computation

It may have also occurred to you by Homer contemplates P = NP?

now that for any given problem
there may be many different
solutions or algorithms. In the
presence of several alternative
algorithms how do you decide
which one to choose? Computer
Scientists have made it their primary
business to develop, analyze, and
classify algorithms to help make
these decisions. The decision could
be based on ease of programming,
efficiency, or the number of The second equation on the right is
resources it takes for a given Euler’s Equation.
algorithm. This has also led
computer scientists to create a classification of problems: from easy to hard,
but in a more formal sense. Some of the hardest open questions in the realm of
problems and computing lie in this domain of research. We will not go into
details here, but these questions have even shown up in several popular TV
shows.We will attempt to give you a flavor of this next.

Space & Time Complexity

Let us start with another problem: You have to travel from Washington State
to Washington DC in the United States of America. To make things
interesting, lets us add a restriction that you can only travel through states
whose names begin with the letters in the word “woman”. That is, it is OK to
go from Washington to Oregon since both “W” and “O” are in the word
“woman” but it is not OK to go from Washington to California. Is this
feasible? If it is, how many ways are possible? Which one goes through the
least/most number of states? Etc.

If you are thinking about how to solve this, you have to rely on your
geographic knowledge of the United States. Alternately, you can Google a

301
Chapter 11

state map and then figure out a solution. But, in doing so, you have stumbled
upon the two key ingredients of computing: data & algorithm.

Data gives you a representation of the relevant information which, in this case
is a way of knowing which states adjoin which other states. The algorithm
gives you a way of finding a solution: Start in Washington, look at its
neighbors. Pick a neighbor whose name satisfies the constraint, then look at
that state’s neighbors, and so on. Additionally, an algorithm also forces you to
make certain choices: if there is more than one eligible neighbor, which one
do you pick? What if you end up in a dead end, how do you go back to the
choices you left behind to explore alternative paths? Etc. Depending on these
decisions, you are likely to end up with many different algorithms: one that
may suggest exploring all alternatives simultaneously; or one that forces you
to choose, only to return to other choices in case of failure. Each of these will
further impact on the amount of data you will need to store to keep track of
your progress.

Developing computer programs to solve any problem requires one to design

data representations and to choose among a set of alternative algorithms.
Computer scientists characterize choices of data representations and
algorithms abstractly in terms of the computer resources of space and time
needed to implement those choices. Solutions vary in terms of the amount of
space (or computer memory) and time (seconds, minutes, hours, days, years)
required on a computer. Here are some examples:

• To compute the product of two numbers requires constant time: to a

computer there is no difference between multiplying 5 by 2 or
5,564,198 by 9,342,100. These are called constant time algorithms.
• To find a number in a list of N unordered numbers takes time
proportional to N, especially if the number you are looking for is not in
there. These are called linear time algorithms.
• To find a number in a list of N ordered numbers (say, in ascending
order) requires at most log time. How? Such algorithms are called
logarithmic time algorithms.

302
 Computers & Computation

• To transform a NxN pixel camera image by manipulating all its pixels

takes time proportional to time. These are called quadratic
algorithms.
• To find a path from a state to another, in a map of N states, given
certain constraints, can take time proportional to where b is the
average number of neighbors of each state and d is the number of
states that make up the solution. In general, many problems fall into
the category where N is the size of the problem. These are called
polynomial time algorithms.
• There are also several unsolvable problems in the world.

In Chapter 9, when doing image transformations, we restricted ourselves to

fairly small sized images. You may have noticed that the larger the image, the
longer it takes for the transformation. This has to do with the speed of the
computer you are using as well as the speed of the programming language
implementation. For example, a computer running at 4GHz speed is capable
of doing approximately 1 billion arithmetic operations in one second (or
10 operations/second). A program written in the C programming language
might be able to give you a speed of ½ billion operations per second on the
same computer. This is due to extra operations required to implement the C
language and additional tasks the operating system is carrying out in the
background. Python programs run approximately 20 times slower that C
programs. That is, a Python program running on the same computer might
give you 25 million operations per second at best. A typical transformation,
say computing the negative, of an image of size WxH pixels would require the
following loop:

for x in range(W)
for y in range(H):
pixel = getPixel(myPic, x, y)
r, g, b = getRGB(pixel)
setRGB(pixel, (255-r,255-g,255-b))

If the image is 1000x1000 pixels (i.e. W=1000 and H=1000), each of the three
statements in the loop is executed 1 million times. Each of those statements in
turn requires an average of 8-16 low-level computing operations: the actual

303
Chapter 11

calculations, plus calculations to locate pixels in memory, etc. Thus, the

transformation above would require over 24-48 million operations. As you
can imagine, it will take a few seconds to complete that task.

In the computing industry, computing speeds are classified based on official

benchmark computations that calculate speeds in terms of the number of
floating-point operations per second, or flops. A typical laptop or a desktop
these days is capable of delivering speeds between ½ to 1 Gflops (Giga flops).
The world’s fastest computer can deliver computing speeds as fast as 500
Tflops (Terra flops). That is, it is about a million times faster (and costs many
millions to make as well). However, if you stop and think about the Chess
playing program we mentioned in Chapter 10 that would require
approximately 10 operations before making a single move, even the world’s
fastest computer is going to take gazillion years to finish that computation!
Such a problem would be considered uncomputable.

Computer scientists have developed an elaborate vocabulary for discussing

problems and classifying them as solvable or unsolvable, computable or
uncomputable, based on whether there are known models to solve a given
problem and whether the models are solvable, computable, etc. There are also
hierarchies of problem solutions, from simple to hard; constant time to
polynomial time and longer; and equivalence classes implying the same
algorithm can solve all problems in a an equivalence class, etc. It is indeed
amazing to conceptualize an algorithm that is capable of solving many
unrelated problems! For example, the algorithms that optimize shipping and
delivery routes can also be used in determining protein folding structures for
DNA molecules. This is what makes computer science intellectually
interesting and exciting.

Summary

In this chapter we have tied together many fundamental ideas in computing

and computer science. While our journey started in Chapter 1 with playing
with personal robots, we have, in the process, acquired a wealth of
fundamental concepts in computing. As you can see, computing is a rich,

304
 Computers & Computation

diverse, and deeply intellectual discipline of study that has implications for all
aspects of our lives. We started this chapter by pointing out that there are now
more computers on a typical college campus in the United States than the
number of people. It probably will not be long before there are more
computers than people on this entire planet. Yet, the idea of an algorithm that
is central to computing is barely understood by most users despite its simple
and intuitive constituents. Many computer scientists believe that we are still
sitting at the dawn of the age of algorithm and that there are much bigger
intellectual and societal benefits yet to be realized, especially if more people
were aware of these ideas.

Myro review

No new Myro features were introduced in this chapter.

Python Review

The only new Python feature introduced in this chapter was the creation of
modules. Every program you create can be used as a library module from
which you can import useful facilities.

Exercises

1. To compute the number of days in a month, we used the following:

def daysInMonth(m, y):
'''Returns the #of days in month, m (1-12) in year, y.'''

if (m == 4) or (m == 6) or (m == 9) or (m == 11):
return 30
elif m == 2:
if leapYear(y):
return 29
else:
return 28
else:
return 31

305
Chapter 11

You can further simplify the writing of the condition in the first if-statement
by using lists:

if m in [4, 6, 9, 11]:
return 30
…

Rewrite the function to use the above condition.

2. Define a function called valid(day, month, year) so that it returns true

if the day, month, and year conform to a valid date as defined in this chapter.
Use the function to rewrite the program as follows:

def main():
'''Given a day of the year (e.g. 248, 2007), convert
it to the date (i.e. 9/5/2007)'''

#Input: day, year

day, year = input("Enter the day, year: ")

# Validate input values...

if not valid(day, month, year):
print “Please enter a valid date.”

# input values are safe, proceed...

# start with month = 1, for January
month = 1

while day > daysInMonth(month, year):

day = day - daysInMonth(month, year)

# next month
month = month + 1

# done, Output: month/day/year

print "The date is: %1d/%1d/%4d" % (month, day, year)

main()

306
 Computers & Computation

Rewrite the program as shown above to use the new function. Besides
developing a correct algorithm it is also important to write programs in a way
that makes them more readable for you.

3. Find out the how fast your computer is by noting the clock speed of the
CPU. Based on that estimate how many arithmetic operations it will be able to
perform in 1 second. Write a program to see how many operations it actually
performs in one second.

4. Do a web search for “Chazelle age of algorithm”. You will be rewarded

with a link to an essay written by Prof. Chazelle. Read the essay and write a
short commentary on it.

5. What is the fastest computer in use today? Can you find out? How fast is it
compared to the computer you use? 100 times? 1000 time? 100,000 times?

6. What is/was the “Y2K” problem?

307
Opposite page: Mars Rover.
Photo courtesy of NASA/JPL­Caltech
 Fast, Cheap &
Out of Control
We suggest that within a few years it will be possible at
modest cost to invade a planet with millions of tiny robots…
With imagination and nerve we can invade the whole solar system.
From a paper titled,
Fast, cheap and out of control: A robot invasion of the solar system,
Rodney Brooks and Anita M. Flynn,
Journal of the British Interplanetary Society, Volume 42, pp 478-485, 1989.

Opposite page: Orb Swarms

Used with permission, Copyright Phil Spitler (www.orbswarm.com)

309
Chapter 12

Nearly two decades ago, Brooks and Flynn laid out a radical vision for the
future that, on July 4, 1997, resulted in successful landing of Sojourner rover
on the surface of Mars. A few years later, in 2004, two rovers Spirit and
Opportunity arrived on Mars to explore the planet. In 2002, in his book, Flesh
and Machines, a more reflective Rodney Brooks confesses to his exuberance
in making the comments in his paper with Anita Flynn and describes the
rovers as planetary ambassadors. The vision put forward by Brooks and Flynn
was to replace a single large 1000 kilogram rover by 100 smaller 1 kilogram
rovers to explore planetary surfaces. The smaller robots would move much
faster and could be built much cheaper and mass produced. They will also
offer much needed redundancy in case of occasional robot failure which in the
case of a single larger robot would lead to total mission failure. Moreover,
each individual robot would be autonomous (i.e. out of human control)
operating on its own agenda as defined it its control program.

It is debatable whether Brooks actually coined the phrase, fast, cheap and out
of control. But the credit for using it in the title of their landmark paper
followed by the deployment of three rovers on Mars led to its widespread use
in popular culture. The phrase became the title of a 1997 documentary movie
by Errol Morris in which Brooks himself was featured. In his 2002 book,
Brooks recounts how a scientist at NASA’s Jet Propulsion Laboratory
lambasted the idea of fast, cheap, and autonomous rovers. While the original
dream of 100 small, autonomous robots remains to be realized, today one
would find it hard to argue against the idea. We have used Brooks’ phrase in
the title of this chapter to suggest that less than twenty years from the coining
of the phrase we are in the thick of exploring the full potential of small, cheap,
personal robots. Given that several million personal robots have already been
sold, we could conclude that our planet itself has been invaded by them.
Brooks says that, “the robotics revolution is in its nascent stage, set to burst
over us in the early part of the twenty-first century.” In this chapter we present
several examples of the ways robots are becoming a part of our everyday
lives.

310
 Fast, Cheap & Out of Control

Robots are mechanisms guided by automated control

We accepted the above definition for robots in the beginning of this text.
Using the Scribbler robot we have also learned much about its mechanisms:
sensors and motors; and how to control them via Python programs. In the
course of this journey we have learned valuable lessons in building different
kinds of robot brains, how to design insect-like behaviors, create sounds,
images, and also ventured briefly into the realm of Artificial Intelligence.
Computing lies at the heart of defining automated control. Any robot, no
matter how small or large, has sensing and motor mechanisms and its
behavior can be programmed to enable it to act autonomously in its
environment. It is important to keep this in mind as we explore the various
dimensions of the robotics spectrum.

The range of applications for robots is nearly as diverse as the world we live
in. In fact, it is limited only by our own imagination. Throughout this text we
have presented examples of robots in several domains: planetary rovers,
vacuum cleaners, explorers in hazardous environments, toys and
entertainment, education, medical surgery, manufacturing, etc. Below, we
present additional interesting examples of the use of robot technology. It will
be important, as we explore some of these examples, to try and use your
understanding from this text to figure out what mechanisms are in use in each
case. Try to think about possible other areas where similar mechanisms could
be out to use.

Toys

Simple robotic toys are everywhere.

Some actually look and behave like
robots but most of them use
computing and robotics technology in
simple and innovative ways. Take for
example the Tengu, designed by
Crispin Jones (tengutengutengu.com).
Tengu plugs into your computer’s

311
Chapter 12

USB port. It is capable of displaying

over a dozen mouth shapes. It is
designed to react to sound and music in
its environment. Facial/mouth
expressions change depending upon the
sounds heard. If you sing a song or if
there is music playing, it will appear to
lip-sync to it.

Desktop robotic toys are increasing in

popularity for several reasons. The
enabling technology is the presence of a
computer with a USB port. The USB ports are
unique in that in addition to providing channels
for exchange of data (like we do in
communicating with the Scribbler) they can also
provide power. Many toys, like Tengu, use the
USB port only for the power. All the controls
are built into the Tengu unit.

Some desktop toys can run independent of a

computer (all the controls are present in the unit
itself) and yet do not require any batteries: they
run on solar power. The Flip Flap flowers and plants made
by the TOMY Company (www.tomy.com) are good
examples. These toys incorporate very simple mechanisms:
solar sensing and power generation coupled with small
motors that are activated by the electric current. These are
simple, yet clever, and entertaining Braitenberg creatures.
The TOMY Company has a whole line of products that
employ these ideas.

There are also plenty of battery operated desktop robotic

toys. The Facebank designed by Takada “eats” a coin when
you flash it in front of its eyes. It runs on batteries and its
mechanism includes IR sensors, similar to the ones on the
312
 Fast, Cheap & Out of Control

Scribbler and a motor that drives or pushes the “skin” of the face from behind.

Most electronic educational toys employ programmed automated control

mechanisms. An interesting educational toy is the Tag Reading Pen made by
Leapfrog (leapfrog.com/tag). A child can use the pen to point on words or text
on specially made story books and the pen
speaks out the word or the entire sentence.
Designed for pre-school kids who are just
getting interested in reading, such a toy can
enhance a child’s reading and pronunciation
abilities. The pen has an optical reader and a
speaker coupled with a memory that records
a child’s reading patterns. Parents can plug
the pen into a computer to download and
track their child’s progress.

Robots need not be constructed out of

digital mechanisms. One can also create
control mechanisms using analog circuitry. Why do you make this thing?
The picture on the right shows devices Because there are not nearly
called Thingamapoops (bleeplabs.com). enough beeping, zapping,
What do they do? They can be used to bixxerfouping,
produce or synthesize crazy sounding
anthropomorphic synthesizer
beeps. The beeps can also be sent as inputs
to standard musical instruments for creating monsters in the world.
From: FAQ on bleeplabs.com
even more bizarre sound effects.

Art

Robots have been actively used in creating all kinds of art and robots
themselves have been the subject of art. There are several organizations
worldwide that are devoted to creating art with robots and robotic devices. In
this book you have also experimented with the Scribbler drawings. A couple
of nice examples of robots creating artwork are illustrated in the works of
Swiss engineers Jürg Lehni and Uli Franke who created the Hektor graffiti
drawing robot (hector.ch) at the School of Art in Lousanne, and Zefrank
313
Chapter 12

(zefrank.com) who has created two versions of a

robot he calls Scribbler which is different from the
robot you have.

Both Hektor and Scribbler create new drawings

based on an existing drawing. A drawing is first
created. The robot (program) reads the drawing
and then embellishes a new drawing based on the
one input. Hektor is mounted on the wall and has a spray can that moves on a
system of robot controlled pulleys. The graffiti shown here on left was painted
by Hektor. You can visit Lehni and Franke’s web site to view movies of the
robot in action. Scribbler uses basic sketches as the basis for creating
drawings. In the picture shown
here, three sketches are shown
that were created based on the
one in the top left corner. The
Scribbler concept is interesting in
that anyone can go and use it via
the web to create drawings in a
web browser. The creators have
also constructed a physical robot
that makes actual drawings.

Do This: Write a program that

samples or scans an image and
creates a graphics drawing based
on that. Read the details provided on the Scribbler web site and use it to create
some sketches. Watch the process of drawing and think about how you might
create something similar.

Show me the way

Global positioning systems (GPS) have been in use for many years now. More
recently, small portable GPS devices have become available for use in
consumer cars. A GPS enables you to enter a destination or a point of interest
314
 Fast, Cheap & Out of Control

and then it plots a route for you to take. It provides real-time map-based turn-
by-turn guidance. The technology used in these devices uses mechanisms that
involve satellite signals and street maps. These devices also form the core of
autopilot mechanisms in an airplane. An airplane operating in autopilot mode
can be considered a robot by our definition. Over 90% take-offs and landings
of commercial flights these days are done using autopilot systems. You may
have also read about robotic surveillance drones that are used by the army to
spy on enemy territories. Unmanned guided missiles use similar devices. The
technology also exists today to create autopilot systems in cars. That is, it is
possible for your car to drive itself to wherever you’d like it to go. In some of
these applications, the questions of technology
deployment become more social and ethical: Do we
really want these or not? What are the implications?

Affective & Social Robots

In Chapter 10 we mentioned that one of the challenges

of AI research is to understand and/or artificially create
human-level embodied intelligence. Many AI
researchers work on advancing the state of
understanding human intelligence while many others
are working on building smarter, more intelligent
models of behaviors. At the same time the field of
robotics itself is rapidly moving in the direction of more
capable, agile, and human-like robots. A good, fun way
to find convergence of these advances can be seen in
the goals of the RoboCup (www.robotcup.org). The
RoboCup organization is focusing on soccer playing robots as a test bed for
AI and robotics. They hold yearly robot soccer playing competitions that
include, besides two-legged humanoids, four-legged and wheeled robot soccer
players.

Besides, soccer playing robots, another area of AI and robotics research that is
gathering momentum is Human-Robot Interaction (HRI). As the name
suggests it is an area of research that is studying models of interaction
315
Chapter 12

between humans and robots. Given that we already have millions of robots
amongst us, it is important to recognize the need for creating friendlier ways
of interacting with robots. While we
are of the opinion that every citizen of
this planet should be well versed in
programming and computation we also
recognize that we are nowhere near
that goal. As we mentioned several
times earlier soon there will be more
computers than people on this planet.
Perhaps robots will also follow?
Nevertheless the need for friendlier
interactions with computers has always
been recognized. With the rapid iCat is a research platform for studying
increase in the number of robot-based human‐robot interaction topics. The
applications it would be even more robot is 38 cm tall and is equipped with
imperative for robots, especially given 13 servos that control different parts of
their physical presence, to have the face, such as the eyebrows, eyes,
socially relevant behavior traits. If not eyelids, mouth and head position. With
for any other reason, it would make it this setup iCat can generate many
easier for people to accept them into different facial expressions ‐ happy,
our society and many levels. surprise, angry, sad ‐ that are needed
to create social human‐robot
Within HRI researchers are studying interaction dialogues.
models of emotion, gesture
recognition, and other social
From: research.philips.com
modalities. Emotive robots have been
studied in many applications that range from toys to medical therapy (we
mentioned the Paro seal robot in Chapter 1). Given the interest and recent
advances in the area of HRI a new field of research has emerged: social
robotics. Social robotics studies models that take into account social norms of
behavior relevant to the environment and/or application of the robot.
Emotions can play an important role in this area of research. Within the
relatively small community of researchers there is much debate about whether
social robotics requires one to have a physical robot. Simulated agents acting
socially are acceptable to some researchers but not to others. Take for
316
 Fast, Cheap & Out of Control

example the simulated robotic agent

Ananova (www.ananova.com). Ananova was
designed to deliver news over the web just as
a news anchor might on TV. While it is not a
physically embodied robot, it has the
simulated morphology of a human and is
capable of using the same models of emotion
and expression that are designed for
embodied robots. In fact, given the physical
limitations of robots, the simulation is far
more realistic (see picture on right).
Ananova News reader
We do not want to leave you with a picture (www.ananova.com)
of the iCat as a representative image of
physical robots capable of emotional expression. In the late 1990’s Cynthia
Breazeal and Brian Scassellati developed a sociable robot, Kismet, to explore
expressions and basic human-robot interactions. Both Breazeal and Scassellati
are former students of Rodney Brooks and have been active in HRI research.
Breazeal’s current efforts in affective computing are based on the robot, Nexi
(see picture below) which was developed at MIT in collaboration with
University of Massachusetts and other corporate partners.

Brian Scassellati is
developing a humanoid
robot at Yale University
that is roughly the size
of a 1-year old child.
The robot is being used
to explore more
fundamental
developmental tasks A new experimental robot from the MIT Media Lab can slant its
like hand-eye eyebrows in anger, or raise them in surprise, and show a wide
coordination. It is also assortment of facial expressions to communicate with people
being used to enhance in human‐centric terms. Called Nexi, it is aimed at a range of
applications for personal robots and human‐robot teamwork.
the diagnosis of autism From: David Chandler, MIT News Office, April 2008.
in children. Brian and
317
Chapter 12

his colleagues postulate that the social cues that a robot needs to detect and
learn are the very cues that are deficient in autistics children. They also think
that such robots can be used to create functional models of autistic behavior.

Autonomous robots are already in use delivering daily mail in large offices.
You can easily imagine the same technology being used for building roaming
vending machines. One can already configure an office espresso maker to
make coffee based on each individual’s preferences: the machine senses you
(or devices you wear) which transmit your preferences to the machine and it
gets to work. The possibilities are endless. We conclude here with another
novel application: Here is problem description (from: www.koert.com/work):
In the morning paper, I can read the weather report as well as the stock
quotes. But when I look out of my window I only get a weather update and no
stock exchange info. Could someone please fix this bug in my environmental
system? Thanks!

The solution: Data Fountain

Designed by Koert van

Minsvoort, Charles Mignot and
their colleagues, the Data
Fountain uses the environment to
display real time data. They
designed the system to map
currency rates to the height of the
fountains. They call this idea
information decoration. You do
not have to restrict yourself to
computer screens and displays for visualizing information. The actual
realization of the concept employs robotics or device control technology that
obtains real time currency data from the web and then transforms it into the
actuation of fountain jets.

318
 Fast, Cheap & Out of Control

Summary

Robots have been used to clean up hazardous substances, removal of mines,

and even unclog deep sewage systems in large cities. They are increasingly
finding applications in unusual situations: exploring small passageways in
ancient pyramids, medical surgery, and even farming. Researchers are
developing intelligent agricultural vehicles to monitor crops, plant diseases,
and general growing conditions. While it may be acceptable and even
desirable to have a robot that vacuums your house, or even clean up streets,
one has to pause and wonder if the same technology can be used for
destructive purposes. Invariably, like it is with any new technology, there are
benefits as well as potential for misuse. Social and ethical concerns have to be
taken into consideration to make sound decisions in all circumstances.

Robots are getting fast, small, cheap, and autonomous. However, it is only in
the sense of autonomy that makes them desirable to be out of control. We may
not yet have sent hundreds of small robots to another plant but we surely seem
to be surrounded by millions of them here on earth. The picture of orb-like
robots on a landscape comes from the Orb Swarm project (orbswarm.com).
They are exploring using swarms of these robots coordinating and exploring
and interacting with their environment similar to the vision of Brooks and
Flynn. Whether we are successful in sending armies of orb-like robots to
another planet still remains to be seen. Though one thing is for sure: it makes
learning computing a whole lot more fun with a personal robot!

319
Chapter 12

320
 Python
Overview
Chapter 1

Start Python.pyw
This is the icon you double-click on to start a Python Shell (IDLE).

>>>
The Python prompt. This is where you type in a Python command.

Note: All commands you type (including the Myro commands listed above)
are essentially Python commands. Later, in this section we will list those
commands that are a part of the Python language.

Chapter 2

def <FUNCTION NAME>(<PARAMETERS>):

<SOMETHING>
...
<SOMETHING>
Defines a new function named <FUNCTION NAME>. A function name should
always begin with a letter and can be followed by any sequence of letters,
numbers, or underscores (_), and not contain any spaces. Try to choose names
that appropriately describe the function being defined.

321
Python Overview

Chapter 3

Values
Values in Python can be numbers (integers or floating point numbers) or
strings. Each type of value can be used in an expression by itself or using a
combination of operations defined for that type (for example, +, -, *, /, % for
numbers). Strings are considered sequences of characters (or letters).

Names
A name in Python must begin with either an alphabetic letter (a-z or A-Z) or
the underscore (i.e. _) and can be followed by any sequence of letters, digits,
or underscore letters.

input(<prompt string>)

This function prints out <prompt string> in the IDLE window and waits for
the user to enter a Python expression. The expression is evaluated and its
result is returned as a value of the input function.

from myro import *

initialize("comX")

<any other imports>

<function definitions>
def main():
<do something>
<do something>
...

main()
This is the basic structure of a robot control program in Python. Without the
first two lines, it is the basic structure of all Python programs.

print <expression1>, <expression2>, ...

Prints out the result of all the expressions on the screen (in the IDLE
window). Zero or more expressions can be specified. When no expression is
specified, it prints out an empty line.

322
 Python Overview

<variable name> = <expression>

This is how Python assigns values to variables. The value generated by
<expression> will become the new value of <variable name>.

range(10)
Generates a sequence, a list, of numbers from 0..9. There are other, more
general, versions of this function. These are shown below.

range(n1, n2)
Generates a list of numbers starting from n1…(n2-1). For example,
range(5, 10) will generate the list of numbers [5, 6, 7, 8, 9].

range(n1, n2, step)

Generates a list of numbers starting from n1…(n2-1) in steps of step. For
example, range(5, 10, 2) will generate the list of numbers [5, 7, 9].

Repetition

for <variable> in <sequence>:

<do something>
<do something>
...

while timeRemaining(<seconds>):
<do something>
<do something>
...

while True:
<do something>
<do something>
...

These are different ways of doing repetition in Python. The first version will
assign <variable> successive values in <sequence> and carry out the body
once for each such value. The second version will carry out the body for
<seconds> amount of time. timeRemaining is a Myro function (see above).
The last version specifies an un-ending repetition.

323
Python Overview

Chapter 4

True, False
These are Boolean or logical values in Python. Python also defines True as 1
and False as 0 and they can be used interchangeably.

<, <=, >, >=, ==, !=

These are relational operations in Python. They can be used to compare
values. See text for details on these operations.

and, or not
These are logical operations. They can be used to combine any expression that
yields Boolean values.

random()
Returns a random number between 0.0 and 1.0. This function is a part of the
random library in Python.

randint(A, B)
Returns a random number in the range A (inclusive) and B (exclusive). This
function is a part of the random library in Python.

Chapter 5

if <CONDITION>:
<statement-1>
...
<statement-N>
If the condition evaluates to True, all the statements are performed. Otherwise,
all the statements are skipped.

return <expression>
Can be used inside any function to return the result of the function.

<string>.split()
Splits <string> into a list.

324
 Python Overview

urlopen(<URL>)
Establishes a stream connection with the <URL>. This function is to be
imported from the Python module urlopen.
<stream>.read()
Reads the entire contents of the <stream> as a string.

Lists:
[] is an empty list.

<list>[i]
Returns the ith element in the <list>. Indexing starts from 0.
<value> in <list>
Returns True if <value> is in the <list>, False otherwise.

<list1> + <list2>
Concatenates <list1> and <list2>.

len(<list>)
Returns the number of elements in a list.
range(N)
Returns a list of numbers from 0..N
range(N1, N2, N3)
Returns a list of numbers starting from N1 and less than N3 incrementing by
N3.

<list>.sort()
Sorts the <list> in ascending order.
<list>.append(<value>)
Appends the <value> at the end of <list>.
<list>.reverse()
Reverses the elements in the list.

325
Python Overview

Chapter 6

The if-statement in Python has the following forms:

if <condition>:
<this>

if <condition>:
<this>
else:
<that>

if <condition-1>:
<this>
elif <condition-2>:
<that>
elif <condition-3>:
<something else>
...
...
else:
<other>

The conditions can be any expression that results in a True, False, 1, or 0

value. Review Chapter 4 for details on writing conditional expressions.

Chapter 7

The math library module provides several useful mathematics functions.

Some of the commonly used functions are listed below:

ceil(x) Returns the ceiling of x as a float, the smallest integer value greater
than or equal to x.

floor(x) Returns the floor of x as a float, the largest integer value less than
or equal to x.

exp(x) Returns e**x.

326
 Python Overview

log(x[, base]) Returns the logarithm of x to the given base. If the base is
not specified, return the natural logarithm of x (i.e., the logarithm to base e).

log10(x) Returns the base-10 logarithm of x.

pow(x, y) Returns x**y.

sqrt(x) Returns the square root of x.

Trigonometric functions

acos(x) Returns the arc cosine of x, in radians.

asin(x) Returns the arc sine of x, in radians.

atan(x) Returns the arc tangent of x, in radians.

cos(x) Returns the cosine of x radians.

sin(x) Returns the sine of x radians.

tan(x) Returns the tangent of x radians.

degrees(x) Converts angle x from radians to degrees.

radians(x) Converts angle x from degrees to radians.

The module also defines two mathematical constants:

pi The mathematical constant pi.

e The mathematical constant e.

327
Python Overview

Chapter 8

In this chapter we presented informal scope rules for names in Python

programs. While these can get fairly complicated, for our purposes you need
to know the distinction between a local name that is local within the scope of
a function versus a global name defined outside of the function. The text
ordering defines what is accessible.

Chapter 9 & 10

There were no new Python features introduced in this chapter.

Chapter 11

The only new Python feature introduced in this chapter was the creation of
modules. Every program you create can be used as a library module from
which you can import useful facilities.

328
 Myro
Overview
Below is a chapter by chapter summary of all the Myro features introduced in
this text. For a more comprehensive listing of all the Myro features you
should consult the Myro Reference Manual.

Chapter 1

from myro import *

This command imports all the robot commands available in the Myro library.
We will use this whenever we intend to write programs that use the robot.

initialize(<PORT NAME>)
init(<PORT NAME>)
This command establishes a wireless communication connection with the
robot. <PORT NAME> is determined at the time you configured your software
during installation. It is typically the word com followed by a number. For
example, "com5". The double quotes (") are essential and required.

beep(<TIME>, <FREQUENCY>)
Makes the robot beep for <TIME> seconds at frequency specified by
<FREQUENCY>.

getName()
Returns the name of the robot.

329
Myro Overview

setName(<NEW NAME>)
Sets the name of the robot to <NEW NAME>. The new name should be enclosed
in double quotes, no spaces, and not more than 16 characters long. For
example: setName("Bender").

gamepad()
Enables manual control of several robot functions and can be used to move
the robot around.

Chapter 2

backward(SPEED)
Move backwards at SPEED (value in the range -1.0…1.0).

backward(SPEED,SECONDS)
Move backwards at SPEED (value in the range -1.0…1.0) for a time given in
SECONDS, then stop.

forward(SPEED)
Move forward at SPEED (value in the range -1.0..1.0).

forward(SPEED,TIME)
Move forward at SPEED (value in the range -1.0…1.0) for a time given in
seconds, then stop.

motors(LEFT,RIGHT)
Turn the left motor at LEFT speed and right motor at RIGHT speed (value in the
range -1.0…1.0).

move(TRANSLATE, ROTATE)
Move at the TRANSLATE and ROTATE speeds (value in the range -1.0…1.0).

rotate(SPEED)
Rotates at SPEED (value in the range -1.0…1.0). Negative values rotate right
(clockwise) and positive values rotate left (counter-clockwise).

330
 Myro Overview

stop()
Stops the robot.

translate(SPEED)
Move in a straight line at SPEED (value in the range -1.0…1.0). Negative
values specify backward movement and positive values specify forward
movement.

turnLeft(SPEED)
Turn left at SPEED (value in the range -1.0…1.0)

turnLeft(SPEED,SECONDS)
Turn left at SPEED (value in the range -1.0..1.0) for a time given in seconds,
then stops.

turnRight(SPEED)
Turn right at SPEED (value in the range -1.0..1.0)

turnRight(SPEED,SECONDS)
Turn right at SPEED (value in the range -1.0..1.0) for a time given in seconds,
then stops.

wait(TIME)
Pause for the given amount of TIME seconds. TIME can be a decimal number.

Chapter 3

speak(<something>)
The computer converts the text in <something> to speech and speaks it out.
<something> is also simultaneously printed on the screen. Speech generation
is done synchronously. That is, anything following the speak command is
done only after the entire thing is spoken.

speak(<something>, 0)
The computer converts the text in <something> to speech and speaks it out.
<something> is also simultaneously printed on the screen. Speech generation

331
Myro Overview

is done asynchronously. That is, execution of subsequent commands can be

done prior to the text being spoken.

timeRemaining(<seconds>)
This is used to specify timed repetitions in a while-loop (see below).

Chapter 4
randomNumber()
Returns a random number in the range 0.0 and 1.0. This is an alternative Myro
function that works just like the random function from the Python random library
(see below).

askQuestion(MESSAGE-STRING)
A dialog window with MESSAGE-STRING is displayed with choices: 'Yes' and
'No'. Returns 'Yes' or 'No' depending on what the user selects.

askQuestion(MESSAGE-STRING, LIST-OF-OPTIONS)
A dialog window with MESSAGE-STRING is displayed with choices indicated in
LIST-OF-OPTIONS. Returns option string depending on what the user selects.

currentTime()
The current time, in seconds from an arbitrary starting point in time, many years
ago.

getStall()
Returns True if the robot is stalled when trying to move, False otherwise.

getBattery()
Returns the current battery power level (in volts). It can be a number between 0
and 9 with 0 indication no power and 9 being the highest. There are also LED power
indicators present on the robot. The robot behavior becomes erratic when batteries
run low. It is then time to replace all batteries.

Chapter 5

getBright()
Returns a list containing the three values of all light sensors.

332
 Myro Overview

getBright(<POSITION>)
Returns the current value in the <POSITION> light sensor. <POSITION> can
either be one of 'left', 'center', 'right' or one of the numbers 0, 1, 2.

getGamepad(<device>)
getGamepadNow(<device>)
Returns the values indicating the status of the specified <device>. <device>
can be "axis" or "button". The getGamepad function waits for an event
before returning values. getGamepadNow immediately returns the current
status of the device.

getIR()
Returns a list containing the two values of all IR sesnors.
getIR(<POSITION>)
Returns the current value in the <POSITION> IR sensor. <POSITION> can
either be one of 'left' or 'right' or one of the numbers 0, 1.

getLight()
Returns a list containing the three values of all light sensors.
getLight(<POSITION>)
Returns the current value in the <POSITION> light sensor. <POSITION> can
either be one of 'left', 'center', 'right' or one of the numbers 0, 1, 2. The
positions 0, 1, and 2 correspond to the left, center, and right sensors.

getObstacle()
Returns a list containing the two values of all IR sesnors.
getObstacle(<POSITION>)
Returns the current value in the <POSITION> IR sensor. <POSITION> can
either be one of 'left', ‘center’, or 'right' or one of the numbers 0, 1, or 2.

333
Myro Overview

savePicture(<picture>, <file>)
savePicture([<picture1>, <picture2>, …], <file>)
Saves the picture in the file specified. The extension of the file should be
“.gif” or “.jpg”. If the first parameter is a list of pictures, the file name
should have an extension “.gif” and an animated GIF file is created using
the pictures provided.

senses()
Displays Scribbler’s sensor values in a window. The display is updated every
second.

show(<picture>)
Displays the picture in a window. You can click the left mouse anywhere in
the window to display the (x, y) and (r, g, b) values of the point in the
window’s status bar.

takePicture()
takePicture(“color”)
TakePicture(“gray”)
Takes a picture and returns a picture object. When no parameters are
specified, the picture is in color.

Chapter 6 & 7

No new Myro features were introduced in these chapters.

Chapter 8

GraphWin()
GraphWin(<title>, <width>, <height>)
Returns a graphics window object. It creates a graphics window with title,
<title> and dimensions <width> x <height>. If no parameters are specified,
the window created is 200x200 pixels.

<window>.close()
Closes the displayed graphics window <window>.

334
 Myro Overview

<window>.setBackground(<color>)
Sets the background color of the window to be the specified color. <color>
can be a named color (Google: color names list), or a new color created using
the color_rgb command (see below)

color_rgb(<red>, <green>, <blue>)

Creates a new color using the specified <red>, <green>, and <blue> values.
The values can be in the range 0..255.

Point(<x>, <y>)
Creates a point object at (<x>, <y>) location in the window.

<point>.getX()
<point>.getY()
Returns the x and y coordinates of the point object <point>.

Line(<start point>, <end point>)

Creates a line object starting at <start point> and ending at <end point>.

Circle(<center point>, <radius>)

Creates a circle object centered at <center point> with radius <radius>
pixels.

Rectangle(<point1>, <point2>)
Creates a rectangle object with opposite corners located at <point1> and
<point2>.

Oval(<point1>, <point2>)
Creates an oval object in the bounding box defined by the corner points
<point1> and <point2>.

Polygon(<point1>, <point2>, <point3>,…)

Polygon([<point1>, <point2>, …])
Creates a polygon with the given points as vertices.

335
Myro Overview

Text(<anchor point>, <string>)

Creates a text anchored (bottom-left corner of text) at <anchor point>. The
text itself is defined by <string>.

Image(<centerPoint>, <file name>)

Creates an image centered at <center point> from the image file <file
name>. The image can be in GIF, JPEG, or PNG format.

All of the graphics objects respond to the following commands:

<object>.draw(<window>)
Draws the <object> in the specified graphics window <window>.

<object>.undraw()
Undraws <object>.

<object>.getCenter()
Returns the center point of the <object>.

<object>.setOutline(<color>)
<object>.setFill(<color>)
Sets the outline and the fill color of the <object> to the specified <color>.

<object>.setWidth(<pixels>)
Sets the thickness of the outline of the <object> to <pixels>.

<object>.move(<dx>, <dy>)
Moves the object <dx>, <dy> from its current position.

The following sound-related functions were presented in this chapter.

beep(<seconds>, <frequency>)
beep(<seconds>, <f1>, <f2>)
Makes the robot beep for <seconds> time at frequency specified. You can
either specify a single frequency <frequency> or a mix of two: <f1> and
<f2>.

336
 Myro Overview

<robot/computer object>.beep(<seconds>, <frequency>)

<robot/computer object>.beep(<seconds>, <f1>, <f2>)
Makes the robot or computer beep for <seconds> time at frequency specified.
You can either specify a single frequency <frequency> or a mix of two: <f1>
and <f2>.

robot.playSong(<song>)
Plays the <song> on the robot.

readSong(<filename>)
Reads a song file from <filename>.

song2text(song)
Converts a <song> to text format.

makeSong(<text>)
text2song(<text>)
Converts <text> to a song format.

Chapter 9

getHeight(<picture>)
getWidth(<picture>)
Returns the height and width of the <picture> object (in pixels).

getPixel(<picture>, x, y)
Returns the pixel object at x,y in the <picture>.

getPixels(<picture>)
When used in a loop, returns one pixel at a time from <picture>.

getRGB(pixel)
getRed(<pixel>)
getGreen(<pixel>)
getBlue(<pixel>)
Returns the RGB values of the <pixel>.

337
Myro Overview

makeColor(<red>, <green>, <blue>)

Creates a color object with the given <red>, <green>, and <blue> values (all
of which are in the range [0..255]).

makePicture(<file>)
makePicture(<width>, <height>)
makePicture(<width>, <height>, <color>)
Creates a picture object either by reading a picture from a <file>, or of the
given <width> and <height>. If <color> is not specified, the picture created
has a white background.

pickAColor()
Creates an interactive dialog window to select a color visually. Returns the
color object corresponding to the selected color.

pickAFile()
Creates an interactive dialog window that allows user to navigate to a folder
and select a file to open. Note: it cannot be used to create new files.

repaint()
repaint(<picture>)
Refreshes the displayed <picture>.

savePicture(<picture>, <file>)
savePicture(<picture list>, <gif file>)
Saves the <picture> in the specified file (a GIF or JPEG as determined by
the extension of the <file>: .gif or .jpg). <picture list> is saved as an
animated GIF file.

setColor(<pixel>, <color>)
setRed(<pixel>, <value>)
setGreen(<pixel), <value>)
setBlue(<Pixel>, <value>)
Sets the color of <pixel> to specified <color> or <value>.
show(<picture>)
show(<picture>, <name>)
Displays the <picture> on the screen in a window named <name> (string).
338
 Myro Overview

takePicture()
takePicture(“gray”)
takePicture(“blob”)
Takes a picture from the Scribbler camera. It is a color picture by default, or
grayscale (“gray”), or a filtered image based on the defined blob (“blob”). See
chapter text for examples.

Chapter 10
There were no new Myro features introduced in this chapter. Actually, when
the chapter is complete it will have Myro primitives for neural nets/conx
described here.

Chapter 11 & 12

No new Myro features were introduced in this chapter.

339
Myro Overview

Index
A Bluetooth 6, 8
AAAI 18 Bluetooth logo 18, 40
ABC song 68 blur image 229
abstraction 36 Boolean 85
acos 183, 326 boundary conditions 296
Aesop 43 Braitenberg, Valentino 130, 153
Aggressive, Vehicle#2 135‐6 Braitenberg Vehicles 130‐
Aibo 4 Breazeal, Cynthia 18, 317
algorithm 281, 292, 302, 305 Brooks, Rodney 309‐10, 317
Al Khwarizmi 283 Burglar Alarm 145
Alive, vehicle#1 131‐5
Anananova 317 C
and 85, 86, 92, 324 C6
Animated GIF 220 C++ 293
AntiqueWhite 189 CamelCase 68
append 111, 125, 325 case sensitive 48
Apple Strudel 280 ceil 164, 182, 326
Artificial Intelligence 4, 88, 245‐271 Celsius 67
movie 3, 245 characters 52
Asimov’s Laws 17 Chazelle, Brian 307
asin 183, 326 Chess 250
askQuestion 78, 79, 91, 137, 332 Chicken Kabobs 276‐7
assignment 50, 66, 323 Chinese Room 270
atan 183, 326 Ching‐Chong‐Cha 177
Avoiding obstacles 148‐151 choice 257, 269
Circle 192, 210, 335
B Clash, The 129
backward 24, 38, 330 close 189, 209, 334
battery disposal 89 Cockroach 153
beep 12, 16, 201‐205, 211, 329, 336 color_rgb 194, 210, 335
Behavior‐based control 158 CommonLisp 293
Blobs 236 comments 32
blob filtering 236 computable 304

340
 Myro Overview

Computational Linguistics 248 Euclid v

Computer Science 273, 275 Error Checking 296‐9
conditions 84, 87 Euro 67
constant time algorithms 302
Corral Exiting 151 F
cos 183, 326 Face Bank 312
Coward, Vehicle#2 135‐6 False 63, 84, 86, 91, 324
currentTime 81‐83, 91, 332 Farenheit 67
Firefox 44
D floating point numbers 51
Dar es Salaam 110, 112 floor 164, 182, 326
data 302 Flip Flap 312
Data Fountain 318 Flops 304
dead reckoning 73 Fluke Dongle 8, 9, 125
decision making 176 Flying Circus 29
Deep Blue 250 Flynn, Anita M. 309‐10
def 27, 39, 118, 321 Follower 146
degrees 183, 326 for 60, 66, 83, 323
Delaware River 5 formal languages 14
Dijkstra, Edsger W. 273, 275 forward 24, 38, 330
Direct Control 159 fractals 213
draw 190, 211, 336 Franke, Uli 313
Franklin Institute 117
E from 34, 166, 322
e 183, 326 function 27, 34
edge detection 237
Elvis 85 G
emboss image 230 Gamepad controller 14, 114‐16
Energy Problem 55 gamepad 14, 16, 108, 330
enlarge image 227‐8 Game playing 251‐267
exp 139, 164, 182, 326 Geier, Sven 187
Explorer 43 getBattery 88, 89, 91, 332
Explorer, Vehicle#3 138 getBlob 237
expression 50 getBlue 238, 337
equal temperament 204 getBright 104‐5, 122, 332
Ericsson 18 getGamepad 115‐16, 122, 333

341
Myro Overview

getGamepadNow 115‐16, 122, 333 Hoare, C. A. R. 273

getCenter 192, 211, 336 Hogg, David 153
getGreen 238, 337 Hugs & Kisses 251
getHeight 218, 238, 337 Human‐robot interaction 315‐16
getIR 106‐7, 122, 333
getLight 103, 122, 333 I
getName 13, 16, 329 iCat robot 316
getObstacle 107‐8, 123, 333 IDLE 10, 27, 29, 36, 62
getP1 191 Idle, Eric 29
getP2 192 if‐statement 120, 123, 142, 154. 324,
getPixel 222, 238, 337 326
getPixels 223, 238, 337 image 216
getRed 238, 337 Image 201, 210, 336
getRGB 223, 238, 337 image processing 226
getStall 87, 91, 332 image understanding 232
getX 190, 210 Imitation Game 246
getY 190, 210 import 166, 322
getWidth 218, 238, 337 in 110, 124, 325
GIF 217 Indecisive 141
global name 196‐7, 328 init 16, 329
Google 3 initialize 11, 12, 16, 329
Gore, Al 68 input 56, 65, 113, 322
Gormson, Harald B. 18 integers 51
GPS 314‐15 internet 5
Grade A eggs 281 interoceptors 72
GraphWin 188, 209, 334 invocation, function 28
Gray Poupon 47, 113 iPhone 85
grayscale images 102 iRobot 2, 59
Gregorian Calendar 291
J
H jalapeno 72
Hallway Cruiser 145 Jankenpon 177
Hektor robot 314 Java 6
Hertz (Hz) 201 Joe, Gigolo 245
hi‐fidelity 202 Joel, Billy 187
HiLo game 184 Jones, Crispin 311

342
 Myro Overview

Jones, Mick 129 loop index variable 60

JPEG 217 Love, Vehicle#3 138
Julia Sets 213 Lousanne 314

K M
Kasparov, Gary 250 main 65, 322
Kismet robot 317 makeColor 222, 224, 239, 338
Kitaoka, Akiyoshi 215 makePicture 219, 221‐, 239, 338
Koch Snowflakes 213 Mandelbrot Sets 213
Konane 250 Mars Rover 1, 2, 4
Martin, Fred 153
L math library 139‐40, 164‐6, 326
Ladybug 129 Maze solver 151
Larson, Doug 273 meaning of life 280
LavenderBlush 189 Measuring Device 145‐6
Law, Jude 245 Media Player 43
Leap Frog 313 Megapixel 217
leap year 291‐5 Mignot, Charles 318
Learning 267 Minimax algorithm 262
LED 89 Minsvoort, Koert van 318
LEGO Mindstorms 3 MIT Media Lab 153, 317
len 110, 124, 325 mixed case 68
Lenhi, Jurg 313 module 31, 33, 34, 299
Light following 146‐8 Monty Python 6, 29
Line 191, 210, 335 Morris, Errol 310
linear time algorithms 302 Moscow 110, 112
List comprehensions 255 motors 23, 38, 330
lists 60, 109‐13, 325 move (robot) 25, 38, 330
Loan calculator 167‐76 move (graphics object) 198, 211, 336
local name 196‐7, 328 musical scale 204
localtime 94 Myers, Mike 157
log 165, 183, 326 Myro 6, 8, 13, 329
log10 165, 183, 326 myro Song Format 206
logarithmic time algorithms 302
logical operations 85 N
loop 60 Names 48, 49, 65, 163, 195‐8, 322

343
Myro Overview

NASA JPL 1, 3, 310 pickAFile 239, 338

Nash, Johnny 97 pixels 102, 216‐7
Natural Language Understanding 248 playSong 208, 211, 337
natural languages 14, 247 Pleo 4, 20‐22
Naughts & Crosses 251 PNG 217
negative image 230 Point 190, 210, 335
New York 85, 110, 112 Polar coordinates 241
Nexi robot 317 Polka 206
Nike viii Polygon 210, 335
not 85, 86, 92, 324 polynomial time algorithms 303
notes 204 Pope Gregory XIII 291
numbers 50 pow 165, 183, 326
Powers, Austin 157
O print 46, 65, 322
octave 204 Programming 6, 274
Ok Corral 151 Programming Language 6, 44, 279, 292
OLPC Project 22, 273‐4 Proprioception 72
Opportunity robot 1, 2, 17, 310 proximity sensor 73
or 85, 86, 92, 324 Python 6, 44, 47
Orb Swarms 309, 319 Python Shell 10, 11
Osment, Haley Joel 71, 245‐6
Oval 210, 335 Q
quadratic algorithms 303
P
Papal Bull 290‐1 R
Paper Scissors Rock 177‐82, 264‐7 radians 183, 326
parameters 30, 195‐7 random 77, 92, 93, 178, 324
Paranoid 141 randomNumber 91, 93, 332
Paris 85 randint 77, 92, 324
Paro robot 23 range 60, 66, 112, 124‐5, 325
Pathfinder 17 Reactive behaviors 143‐152
PB&J 265 Reactive control 159
Pennsylvania 5 read 117, 124, 325
Philadelphia 117 readSong 208, 211, 337
pi 183, 326 Rectangle 210, 335
pickAColor 222, 239, 338 Refrigerator Detective 145

344
 Myro Overview

repaint 222, 239, 338 Sensor Fusion 159

repetition 60, 66, 176 setBackground 189, 209, 335
Resnick, Mitchel 153 setBlue 239, 338
return 119, 124, 324 setColor 223, 239, 338
return values 163‐4 setFill 193, 211, 336
reverse 111, 125, 325 setGreen 239, 338
RGB 102, 193, 216, 223 setName 13, 16, 330
RoboCup 315 setOutline 193, 211, 336
Robot, definition 3, 4 setPixel 225
Robot Hall of Fame 18 setRed 239, 338
Robot Vision 232 setRGB 223
Rochambeau 177 setWidth 193, 211, 336
Rock Paper Scissors 177‐82, 264‐7 sharpen image 229
Roomba 2, 3, 22, 59 shrink image 227‐8
rotate 25, 38, 330 shrinking factor 228‐9
Rotating Snakes 214 show 100, 123, 216, 218, 240, 334, 338
Royal Mail 5 Shyamalan, M. Night 71, 72
Run Module 46 Sierpinski Triangles 213
runic alphabet 18 Simpson, Homer 42
sin 183, 326
S Sixth Sense movie 71, 72
Saab 167 Snicker’s moment 72
savePicture 100, 123, 218‐, 239, 334, Sojourner 17, 310
338 solvable 304
scale 204 song2text 212, 337
Scassellati, Brian 316 SONY 4
Science Magazine 153 sort 111, 125, 325
scope 195 Social Robotics 316
Scribbler 6, 22 Soviet Union 283
Scribbler 208 Space Complexity 301‐5
Scribbler drawings 313‐14 speak 47, 64, 125, 331
Scribbler sensors 99 Spitler, Phil 309
Sear, Cole 71, 72 split 113, 124, 324
Searle, John 270 Spirit robot 1, 2, 17, 310
sequential execution 176 Spielberg, Steven 3, 245
senses 99, 108, 123, 334 sqrt 164, 183, 326

345
Myro Overview

Squyres, Steve 1 U
Start Python 9, 16, 321 UGOBE Inc. 21
strings 50, 52 uncomputable 304
stop 25, 38, 331 undraw 211, 336
Subsumption Architecture 160 Unicode 85
Sullivan, Jon 71 Unimation 4
syntax error 35 unsolvable 304
urlopen 117, 124, 325 urllib 117
USDA 281‐2
V
T Values 50, 65
Tag reading pen 313 variable 50
takePicture 100‐2, 123, 216‐, 240, Victoria Crater, Mars 3
334, 339
Takada 312 W
tan 183, 326 wait 29, 39, 322, 331
Tengu 311‐12 Wales 58
Testing 296‐9 Wall Detector 145
text 200, 210, 336 Washington DC 301
Terra flops 304 Washington state 301
Thingamapoops 313 while 62, 66, 83, 87, 323
Tic Tac Toe 250‐264 WhiteSmoke 189
time 94 Wikipedia 18, 29
Space Complexity 301‐5 Wong, Yingshun 157
timeRemaining 62, 64, 332 world population 53, 64, 89
Timid 141 world wide web 5
TOMY Company 312
Toyota Prius 110 X
Traffic Lights 157 XO Laptop 273‐4
translate 25, 38, 331
True 63, 84, 86, 91, 324 Y
Tumbleweed robot 2 Y2K Problem 307
Turing, Alan 246, 270
Turing Test 246 Z
turnLeft 24, 38‐39, 331 Zamboni 58, 92
turnRight 24, 39, 331 Zefrank 313‐14

346
Myro Overview

347
Myro Overview

348