Interactive Humanoid Robot
Interactive Humanoid Robot
Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Interactive Humanoid Robot
Dr. Wilferd John Vaz1, Professor Sareddy Jaswanth Reddy 2, Student
Electronics & Communication Engineering Department Electronics & Communication Engineering Department
Nagarjuna College of Engineering &Technology Nagarjuna College of Engineering &Technology
Bengaluru, India Bengaluru, India
Sushmitha KR3, Student Swathi V4, Student
Electronics & Communication Engineering Department Electronics & Communication Engineering Department
Nagarjuna College of Engineering &Technology Nagarjuna College of Engineering &Technology
Bengaluru, India Bengaluru, India
Rashmi KR5, Student
Electronics & Communication Engineering Department
Nagarjuna College of Engineering &Technology
Bengaluru, India
Abstract:- Humanoid Robots are built to mimic automotive manufacturing company workers. Since they do
humans. The navigation problem of robot is solved by the repetitive jobs humans can be replaced with the
decomposing them into a series of small multi-objective humanoid robots for better accuracy and performance.
optimization problems (MOPs) with corresponding
local information, using multi-objective evolutionary II. RELATED WORK
algorithms (MOEAs).The humanoid robot can use its
control function to stimulate more number of responses First, Rainer Stiefelhagen, Hazım Kemal Ekenel,
to its environment and uses a computing speed of Christian F¨ugen, Petra Gieselmann, Hartwig Holzapfel,
milliseconds to anticipate and react to the movements Florian Kraft, Kai Nickel, Michael Voit, and Alex
done by workers at the workplace. For implementing Waibel[1]. (IEEE TRANSACTIONS ON ROBOTICS).
the robotic control and computer vision functions a They developed a natural multimodal humanoid robot
group of applications are developed in Raspberry Pi interaction system which spontaneously recognizes the
using Open CV modules. The Image processing and speech, process the multimodal dialogues and visual
controlling is done by the processor Raspberry Pi . perception of the person who is using the robot. It includes
Various sensors will be embedded to fetch the data from localization of the robot, tracking and identification of the
the environment or the user. Required information is user, identification of sign gestures shown and also
conveyed through LCD touchpad. recognition of a user’s head orientation. This project is
defined as human to human robot interaction than human to
In this paper we are mainly concentrating on robot interaction. Multiple algorithms were written for
SLAM technology for tracking the position of robot and recognition of various different types of speech or dialogues
to prepare a map of surroundings, which helps in and visual signs given by the user. They also presented
navigation of a person. several practical experiments on human-to-human robot
interaction through speech and gestures.
Keywords:- Raspberry Pi Microcontroller, Touch Pad,
Speaker, Accelerometer and Gyroscope Sensors. In Shu-Yin Chiang*, Yi-Quan Jiang, Hsin-Tieh Yang,
Chia-Chin WangandYu-ChenLee[2] Department of
I. INTRODUCTION Information and Telecommunications Engineering, Ming
Chuan University, Taiwan, IEEE. They developed a
A humanoid robot is a type of robot built to resemble intelligent multifunctional humanoid robot to provide a
human body. Humanoid robots generally have a head, two companion and entertainment to the user, They used Kinect
upper limbs, and two lower limbs, though some forms of depth image as the visual platform to gain the information
humanoid robots has only part of the body, for example, from the environment and to perform image processing.
from the upper part of waist. Few humanoids just have head Next, the results from the image processing are considered
designed to replicate human facial expressions through eyes in planning of robotic behavior. They used omnidirectional
and mouth. wheels, motors with high power, FPGA and ARM based
tools as motion system, Lower body consists of four
Humanoid robots are developed to perform human omnidirectional wheels in all the directions and upper body
based tasks such as personal assistance, through which they is human design. They used the information extracted from
are able to assist the old and sick and dangerous or dirty Kinect depth image stream to integrate the robot localization
works. Humanoids robots are also capable of doing some system and mapping of the real time environment and
procedurally repetitive jobs, such as receptionists and avoidance of obstacle. And also Kinect skeleton detection is
IJISRT20MAR773 www.ijisrt.com 1434
Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
used to recognize facial expressions so the task of intelligent information and also guiding visitors to reach their targets.
interaction is made easy. They placed a ubiquitous sensor network with six laser
range finders in a shopping mall. The system keep track og
In Takayuki Kanda, Member, IEEE, Dylan F. Glas, human positions and their behavior ,such as brisk walking,
Masahiro Shiomi, and Norihiro Hagita[4], Senior slow walking, wandering or standing, They collected
Member,IEEE. They built this robot mainly for providing human trajectories for a week, by applying cluster
service at shopping malls, To provide proper service to the technology to the collected information about the place and
consumers, This paper describe the series of abstraction people’s typical global behavior . This information helps
techniques for people’s trajectories, and techniques to use robot in targeting the people who needs service mainly who
the robot as social robot, which makes robot to proactively are slowly walking or stopping at the shop.
interact with the consumers and by providing sufficient
Fig 1:- System Architecture
Raspberry Pi processor for processing the images task codes. If the value matches with the s value in the code
captured by the camera module. Gyroscope and it will perform the task which was coded under that value
accelerometer for controlling the attitude of the robot. subset.
Speaker and mic with audio driver setup to interact with the
user. LCD touchpad is used to display the required visual It involves 4 steps
information. 1. Transition of the value
2. Receiving the value by serial communication
Initially the processor is set into active receiving mode 3. Comparing the values with the pre-specified threshold
and it will wake with commands and do the task and it value.
sends the data to the microcontroller. In the microcontroller 4. Executing task when there is a match.
the received value is compared with all the values in the
IJISRT20MAR773 www.ijisrt.com 1435
Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig 2:- Data flow diagram
III. WORKING or recognizes a familiar landmark, he or she can figure out
where they are in relation to it. If the person does not finds
Initially, when the robot is turn on it manually asks or recognizes landmarks, he or she will be considered as
the person to choose an option like navigate, introduce etc lost. However, if the person observe the environment and
where these options are provided in the touch pad of the landmarks, the person will recognize the place and begin to
robot. As per the person’s choice the robot will do its build an imaginary map of that place. The person may have
function. If a person chooses option1, it describes about the to navigate or roam around this certain environment several
college and if a person uses chooses option2, the robot times before becoming familiar with a previously unknown
navigates the person, where this navigation works on place. In a similar way, a humanoid robot built with the
SLAM technology. SLAM technology is the technique help of SLAM technology tries to build map of an
behind robot localization and mapping. The robot plots an unknown environment while figuring out where it is at.
imaginary line in an area and at the same time, it also After building a map it tries to figure out its own position
figures out its own position located in the place. The in which location it is by scanning the surroundings with
process of SLAM technology uses a complex group of the help of camera attached to it. After observing the
computations, algorithms and inputs from sensors to surroundings it calibrates and comes to conclusion whether
navigate around a previously unknown environment or to it is in right place or not. This humanoid robot also uses
revise a map of a previously known environment. SLAM is gyroscope and accelerometer sensor where these sensors
just similar to that of a person trying to find his or her route maintains the orientation and angle of the robot in which it
around a new or an unknown place. First, the person looks is moving. These sensors are connected to raspberry pi
for familiar markers or sign boards. Once the person finds where it takes the input from the sensors and maintains the
IJISRT20MAR773 www.ijisrt.com 1436
Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
position and angle of the robot. It also as an Ultrasonic
sensor, where these sensors are used to avoid the obstacles
during movement of the robot. And the motors are driven
by motor drivers which drives the robot for movement of
robot based on the input taken by various sensors connected
to it. And for this autonomous robot battery pack is used as
the robot is in movement with respect to its function and
battery distortion for each component is controlled by
raspberry pi board.
IV. CONCLUSION
The Proposed System “Interactive Humanoid Robot
Removes” Is Mainly Built To Mimic Humans Who Does
Repeated Task Such As Receptionists, Automotive Line
Workers, Humanoid Robot Can Be Placed At Institutions
And Working Areas To Guide And Navigate The Visitors
And Also To Knowledge Them With Some Basic
Information About The Respected Place.
REFERENCES
[1]. Rainer Stiefelhagen, Hazım Kemal Ekenel, Christian
F¨ugen, Petra Gieselmann, Hartwig Holzapfel,
Florian Kraft, Kai Nickel, Michael Voit, and Alex
Waibel(IEEE TRANSACTIONS ON ROBOTICS)
[2]. Shu-Yin Chiang*, Yi-Quan Jiang, Hsin-Tieh Yang,
Chia-Chin WangandYu-ChenLee Department of
Information and Telecommunications Engineering,
Ming Chuan University, Taiwan, IEEE
[3]. Engineering(ICCSCE). Takayuki Kanda, Member,
IEEE, Dylan F. Glas, Masahiro Shiomi, and Norihiro
Hagita, Senior Member, IEEE
[4]. Ki-Baek Lee, Member, IEEE Hyun Myung, Member,
IEEE and Jong-Hwan Kim, Fellow, IEEE
