Robotics

R&N: ch 25
based on material from Jean-
Claude Latombe, Daphne Koller,
Stuart Russell
Agent
sensors

?
environment
agent

effectors

Robots ⇒ Physical sensors and effectors

Senso
Sensorss
Sensors that tell the robot position/change of
joints: odometers, speedometers, etc.
Force sensing. Enables compliant motion--
robot just maintains contact with object
(video: compliant)
Sonar. Send out sound waves and measure
how long it takes for it to be reflected back
back.
Good for obstacle avoidance.
Vision systems
Effecto
Effectorss
Converts software commands into
physical motion
Typically electrical motors or
hydraulic/pneumatic
yd au c/p eu at c cy cylinders
de s
Two main types of effectors:
„ locomotion
„ manipulation
Locomotion
Legs!
„ traditional (video: honda human)
„ Other types
Š Statically stable locomotion: can pause at any stage
during its gate without falling
Š Dynamically stable locomotion: stable only as long as it
k
keeps movingi (video:
( id hopper)
h )
Still, wheeled or tread locomotion like Shakey
is still most practical for typical environments
Other methods: reconfigurable robots, fish
robots, snake-like robots. (video: mod-robot)
Manip lation
Manipulation
Manipulation of objects
Typical manipulators allow for:
„ Prismatic motion (linear movement)
„ Rotary motion (around a fixed hub)
Robot hands go from complex anthromorphic
models to simpler ones that are just graspers
„ ((video:
id manipulation)
i l ti )
„ (video: heart surgery)
P oblems in Robotics
Problems
Localization and Mapping
Motion planning
Locali ation Where
Localization: Whe e Am I?
Use probabilistic inference: compute current location
and orientation (pose) given observations

At-1 At-1 At-1

Xt-1 Xt-1 Xt-1

Zt-1 Zt-1 Zt-1

Motion Planning
Simplest task that a robot needs to
accomplish
Two aspects:
„ Finding a path robot should follow
„ Adjusting motors to follow that path
Goal: move robot from one
configuration to another
Config ation space
Configuration
Describe robot’s configuration using a set of
real numbers
Flatland -- robot in 2D -- how to describe?
Degrees of freedom: a robot has k degrees of
freedom if it can be described fully by a set of
k real numbers
„ e.g. robot arm (slide)
Want minimum-dimension parameterization
Set of all possible configurations of the robot
i the
in th k-dimensional
k di i l space is
i called
ll d the
th
configuration space of the robot.
E ample
Example
workspace for 2-D robot that can only
translate, not rotate
configuration space describes legal
configurations
„ free-space
„ obstacles
Configuration
f space d
depends
d on how
h
big robot is—need reference point
Path planning
Goal: move the robot from an initial
configuration to a goal position
path must be contained entirely in free space
assumptions:
„ robot can follow any path (as long as avoids
obstacles)
„ dynamics are completely reliable
„ obstacles known in advance
„ obstacles don’t
don t move
Ass mption #1
Assumption
robot can follow any path
what about a car?
degrees of freedom vs. controllable
degrees of freedom
„ holonomic (same)
„ nonholonomic
„ (video: holonomic)
Motion planning
reduces to problem of finding a path
from an initial state to a goal in robot’s
configuration space
whyy iss tthiss hard?
ad
Reformulate as discrete
sea ch
search
finely discretized grid
cell decomposition: decompose the space into
large cells where each cell is simple, motion
planning in each cell is trivial
roadmap
d (skeletonization)
( k l ) methods:
h d come up
with a set of major “landmarks” in the space
and a set of roads between them
Iss es in Sea
Issues ch
Search
Complete
Optimality
Computational Complexity
Motion planning algorithms
algo ithms
grid
cell decomposition
„ exact
„ approximate
roadmap (skeletonization) methods:
„ visibility graphs
„ randomized path planning
Robotics
Robotics: S mma
Summary
We’ve just seen a brief introduction…
Issues:
„ sensors, effectors
„ Locomotion, manipulation
Some problems:
„ Localization
„ Motion Planning
Lots more!!