Abstract Summary - Working Group - IEEE

earthblurtingAI and Robotics

Nov 14, 2013 (3 years and 6 months ago)

48 views


1



Project Proposal for


Automated
Smart
Robo
chair


By
-

Abhinav Garg
1
, Amit Kumar
2

and Satya Pal Singh
3


1, 2

B. Tech. II Year, Department of Electrical Engineering
, Madan Mohan Malviya Engineering College,
Gorakhpur, Uttar Pradesh, 273010, Indi
a

3
Departm
ent of Applied Sciences, Madan Mohan Malviya Engineering College, Gorakhpur, Uttar Pradesh,
273010, Indi
a

Mentor’s Email
address
for correspondence
-

(Dr. Satya Pal Singh
3
)

spsas@mmmec.net

/
singh.satyapal@hotmail.com


Co
ntents

1.

Introduction


2.

Current Status of the
W
ork


3: Our Objectives and the Layout of the Prototype


4
.
Function
of the prot
otype


5
.
Technical details


6.
Program Written
i
n Prototype

(C Language)

7
.
Summary
-
Realizations

i
n
Real S
cenario










2


1:
INTRODUCTION



It is a
common

problem faced b
y
disabled person
s

to

feel immobilised
because

of their
physical deficiency or
inability to move
in a normal fashion
.
If

there

is
a
partial disability due to any
limb malfunction
,

the person
feel
s

a certain degree of

restriction
to develop
as per best of his or her
capabilities.
The problem
s

arising from such physical or neurological disabilities are

particularly
observed

in case of patients of spinal injuries and stroke affected
people

i
n which whole segment of

the body, say below neck or so
, is rendered
to immobility

on account of the injuries in the
connecting nerves.


In this case the patient is unab
le to navigate through the arena i.e. workplace or
home, even in worst case
s ha
nds or legs may be partially or completely paralysed
. Sa
me
thing
applies to visually disabled persons.

T
hey are also helpless when it comes to mak
e

their way to
different locations of the same house or an academic campus
or inside a rehabilitation camp
wit
hout

fear of bumping into
obstacles
.


This situation
can be

dealt with artificial
supports and
enhancements ranging from prosthetic
limb to minute earplugs
.

A

variety of other products such as traditional wheel chairs, walking sticks
etc
.

are also availab
le.
But
somehow the inferiority

complex cr
eep
in
to

physically challenged

person
s

who most often feel isolated from the main stream of the spectrum of the people from the
most common to the most uncommon one
.

They
realiz
e

other people leading a life full of

luxuries
and
comforts.



But an

automated
smart
( intelligent or
say
programmed ) wheelchair can be of great help in
smoothing the lives of
such patients
,

if an aid
such as these intelligent wheelchairs are

available to
reduce their disabilities by

naviga
ting through the area specified

by electronic sensors and
developing a human
-
machine interface.

This can be envisioned as a blessing in disguise for people
fed up with the crutches and all prosthetics, and to whom moving independently is merely reduced
to
a dream. It is assumed that if proper

research

work is done in coming future then a successful

implement
ation

can

revolutionise
for a change and can be turn out to be a boon to the physically and
neurologically challenged persons
. It will reward a self sus
taining and content life to those who have
ambitions but always failed to achieve th
ose

due to lack of fitness
. Such smart automated wheelchair
have got good attention of the scientist

[1] [2] [3]

in last few decades with the manifold development
in inform
ation technology

and engineering
.



2.

Current Status of the Work

A group of researchers at MIT, USA is running a
project

sponsored by Microsoft, Nokia and
a Division of Rehabilitation of Government of USA

to enhance
a electrically

powered wheelchair
which

uses

sensors to perceive the wheelchair's surroundings
, a
human
-
machine interface as
to
interpret
speech into
commands, a
wireless device for room
-
level location determination

and
motor
-
control software

is used
to
control

the wheelchair's motion

in an opt
imized fashion
.

The robotic wheelchair learns the layout of its environment (hospital, rehabilitation center,
home, etc.) through a
pre
-
narrated, guided tour given by the user or the user's care
takers
.
Subsequently, the wheelchair can move to any previous
ly
-
named location under voice command

3

(e.g., "Take me to the cafeteria"

can carry the person to the cafeteria
). This technology is appropriate
for people who have lost mobility due to brain injury or
because of
the loss of limbs, but who retain
their
speec
h. The technology can also enhance safety for users who use ordinary joystick
-
controlled
powered wheelchairs, by preventing collis
ions with walls, fixed objects, furniture
and
with other
people
. Collisions are the most usual event in this kind of developme
nt in human
-
machine interface
which significantly affects the modeling, the cost of the projects and may require very advanced
improvements when proposing such models


[
4
] [
5
]
.

A

global positioning system (GPS) receiver
is
being used by the group
for navig
ation and remote controlling purpose
which
includes a braille
keyboard and a voice synthesizer. With access to only the current geo
-
stationary satellites, a device
such as the TORMES cannot give the precise information
which
a visually impaired person need
s to
locate and travel to a destination without running into obstacles. This is because satellite signals can
be blocked by tall buildings and are rarely accurate to within 30 meters

[
6
]
.

The present
ly one group at Institute of Computer Science, Greece is
developing

wheelchair
for people with special needs, which is capable of navigating semi
-
autonomously within its
workspace

[7]
.

This system is expected to prove useful to people with impaired mobility and limited
fine motor control of the upper extremities
. Among the implemented
behavio
u
rs

of this robotic
system are the avoidance of obstacles, the motion in the middle of the free space and the following of
a moving target specified by the user (e.g., a person walking in front of the wheelchair). The
wheelch
air is equipped with
sonars
,

which are used for distance measurement in
preselected critical
directions

and wit
h a panoramic camera with a 360
0

field of view, which is used for following a
moving target. After suitably processing the colo
u
r sequence of the

panoramic images using the
colo
u
r histogram of the desired target, the orientation of the target with respect to the wheelchair is
determined, while its distance is determined by the sonars. Moreover, they are as simplified as
possible to minimize impleme
ntation requirements. An experimental prototype has been developed
at ICS

FORTH, based on a commercially
-
available wheelchair. The sensors, the computing power
and the electronics needed for the implementation of the navigation
behaviours

and of the user
i
nterfaces (touch screen, voice commands)
are
developed as add
-
on modules and
are

integrated with
the wheelchair
.

Camera
based human
-
machine interface
which can scan images of the face
expression
are also used by some of the groups [
8
].

A wheelchair robot d
eveloped by scientists at Coimbra Uni
versity already has a prototype
“capable of navigating without colliding with obstacles by commanded human voice”, as professor
Urbano Nunes
the team leader

of the Electro
-
technical Engineering Department responsible fo
r the
project

states
.
S
imple voice commands, like “forward”, “backwards”, “to the right” and “to the left”
can give great autonomy to the elderly, the infra
-
red sensors allow mobility without accidents, with

no collisions with obstacles, narrow passages or

doors [
9
].


3
.

Our objectives and the Layout of the Prototype


In this project we have made an effort to develop a prototype wheelchair which works via the
human
-
machine interface. A programmed microcontroller
reads the present location of the
wheelchair
and the input command for the destination and
then
give
s

necessary wireless message to
move the robochair to
move to the
desired location.

We have
worked on the
conce
pt of automation

using which
a disabled person is taken to the desired location autonomous
ly through the command
given by user

(that can be remote when work is further developed)
. In this project
not
only the
disable
person
can be

saved from
giving directions to navigate through the house

but also taken care
of
his well being by

IR
sensor inter
face at various level and positions. This project aims at designing
a framework
in
home, in which the commands and their execution is controlled
through a
microcontroller
.

This facility of single user interface can be further extended to multiple users as
in

4

the case of educational campus which
can turn out to

be a breakthrough in itself. Thus it
can

enable
multiple users to facilitate the benefits of the automatic driving facility
.



Layout of the prot
otype


The basic structure of the prototype developed t
o
ach
ieve

the concept of mapping and guiding is
shown in figure given below. The structure

consists
of wooden floor, partitions, doors, windows
, and
number of blocks (
8 in
our present study

case) which
signifies the different part of a house. Each part
has

a sensor embedded in the floor which indicates the presence of robochair in that particular area.
As the robochair navigates through the house
and
it utilises the sensor’s signal to register its
presence.
Whole structure can be summarised as:




This layout

is the most simplified way of demonstrating the concept of placement of


sensors and their utilisation for the mapping of the whole area.



In the layout below sensors are represented by black circles, placed at the door of each


block or r
oom of house.



Other sensors and devices used for further enhancement are not shown for the time being


to avoid complexity and to reveal the genuine idea. i.e. Precise navigation



Each block has particular value which is given by the sensor planted
on the floor.



Sensor is connected wirelessly to the microcontroller board placed on the robochair.



Robochair consist of AVR development board, DC motors, batteries and buttons to give
command to particular location



Robochair is capable of moving through th
e home and transmit and receive

signals from
the sensor transmitter.



A common corridor is provided in the
work

to facilitate the easier navigation of
robochair.



In practical situations although, complex mapping may be done with sensor as landmark
at
differ
ent locations.



Robochair is equipped with IR sensors so that it can detect the presence of any obstacle
and avoid collision from the object

though currently not a part of this project.



Further
improvements

are
provided in the
real

environment
which
can be
done
by using
multiple sensors at different places which takes input
signals
at various levels. For
example switching on the fan or AC as the person enters, opening and closing of windows
can be
guided
and controlled
through the programmed directions accor
ding to the
presence or absence of the person at a particular location

in the premises
.



Security aspect can be taken care of by providing laser detection
based counters
at
window and doors, preventing entry of any unwanted element

without much changing the

project cost
and can be integrated as a part of the local module which can communicate
the adverse conditions to a global or remote controller or administrator via GPS.



This concept can be further generalized to i
ntegrat
e

the
concept of mobile robotics
. I
t can
be
facilitate
d

by
relatives
or friends
present outside

as

well who can

control the

robochair
to guide it to a safe
place
and

can
also
watch the proceedings from a webcam through
internet.



Voice command and face image scanners can be used for human
-
ma
chine

interface
.


5





Actual robochair being used in the prototype










































Atmega 16 microcontroller

L239D motor driver

AVR development board

Base of robocha
ir

Wheels connected to motors

ROBOCHAIR

DINING ROOM


BEDROOM

BATHROOM

MUSIC ROOM

LIBRARY

TV ROOM

GUEST ROOM

KITCHEN


It will navigate through the house through the corridor and guide to
the place as desired by the user



6





4
.
Function
of the prototype


The key

idea
of the whole project is :


Ma
pping

of location
in the controller and its
efficient
ut
ilisation by u
ser with ease



Basically the whole map of the space is fed into the
microcontroller

through the coordinates
assigned to a particu
lar sensor as shown in
above
figure
, thus each sensor carry a specific value,

i.e.
the address of the location in which it
is planted.
Micro
control
ler

takes into account the sensor
specification of the present location and process the command to navigate though desired location.
Taking this in overview, now we have a particular
allocated
value of each place and we can utilise
this value for the purpose of navigation
. The command given by the user is processed into suitable
signals which are transmitted through wireless channels to the central control
. Micro
control
ler

detects the present location and uses the pre
-
coded map to wo
rk out the way to go to the locatio
n by
giving output direc
tly to motors of the wheelchair.



In this way the whole area is mapped through the sensors and the presence of robochair is indicated
runtime. i.e. while moving it continues to register its locati
on so that decision can be taken by the
controller taking in view of the users command to direct to the particular location within the house
and the area mapped.


The exact algorithm of the working of prototype is explained with the help of block diagram
as
shown below



















C
urrent
location

Command
given by user
to targe
t
location

Microcontroller
detects
the
current
location and process
command from user
to target location.

Output to motors
of the robochair

which drives the
disabled person
to the desired
destination






Target location

(gives
feedback)

(Completes
feedback)


7

The steps involved in the whole process is summarised as below



First of all

the sensor1 detects the present location of the robochair in the house. It
transmit
s

the signal to microcontroller through wireless channel
.



User
give
s

command to microcontroller
to
the desired location
mounted on the robochair
.



Microcontroller receives the command of desired location and points out the target
sensor.



It works out the distance bet
ween the current sensor and the

target sensor.



Direct
ion is given to the motors to drive particular distance
as worked out by the
programmed
microcontroller

by calculating the distance between two sensors under
consideration



The confirmation is received by the microcontroller through the sensors placed at ta
rget
location
. A confirmation stops the robochair when destination condition is met
.



Motors specification
and powers

s
uppl
y features are

to be selected taking care of
the
weight of the person subjected under use
.


5:
Technical Details


This project utilise
s the foundation of autonomous robotics. i.e. microcontroller to make the

use of

sensor mapping
of the whole area specified. To
uch sensors are planted on the floor of the

area which
is to be mapped

























MICROCONTRO
LER (Atmega16)

To take input
form sensors and
give output as per
the programming

Arra
y of
infrared
sensors

1

2

3

4

Comman
d module

L239D
motor
control
IC

Left
motor
oro

right
motor
oro


8



(a)
AVR Atme
ga16 microcontroller

Atmel's AVR® microcontrollers have a RISC core running single cycle

instructions

and a well
-
defined I/O
structure that limits the need for external

c
omponents. Internal oscillators, timers, UART, SPI, pull
-
up
resistors, pulse

width mo
dulation, ADC, analog comparator and watch
-
dog timers are

some of

the features
you will find in AVR devices.
AVR instructions are tuned to decrease the size of the program whether the
code

is

written in C or Assembly. With on
-
chip in
-
system programmable Fla
sh and

EEPROM, the AVR is
a perfect choice in order to optimize cost and get product to

the market quickly.

















(b) Motor Control Using L239d

L293 series of chips are power H
-
bridge motor drivers. The L293 D have been used for several years i
n
many MIT robot controller boards, such as the 6.270 Rev 2.1 board and miniboard. The L293D chip is 16
-
pins dip packages, has, two h
-
bridge drivers. An H bridge is typically capable of running one DC motor
bidirectional (forward, backwards,

off), or two s
eparate motors unidirectional

(forward, off). Thus a L293
D

chip can run two motors bidirectional, or 4

motors

unidirectional.

L293D chip

take logic
-
level inputs to
direct the H
-
bridges, and have a separate pin for the motor

supply (which is often higher th
an the standard
5V logic supply).


(c) Circuit Diagram of Infrared Sensor

This IR sensor circuit is designed using 1 IR LED and 1 Photodiode. LM 358 is a opAmp and

is used as
voltage comparator mode. This circuit works on the reflection criteria. IR LED an
d

Photodiode is placed
adjac
ent

to each other. When no IR light falls on the photo diode the

resistance of the diode falls in the range
oh Mega ohms or approximately infinity. At this

condition the voltage at the pin 3 of IC LM358 is about
zero volt. And t
he voltage at PIN 2 of

IC LM358 can be set
in between 5volt to zero volt. H
ere we have set
the voltage is about 2.5

volt using Variable resister R7. at this point of time the output of the LM358 is low.

When any reflecting surface comes near to IR LED and
photo diode pair the reflected light of

IR LED falls
on the photodiode which rapidly decreases the resistance of the Photodiode and

photodiode starts
conducting. At this point of time the voltage at PIN 3 of IC LM358 is

approximately 5 volt
, w
hich exceeds

9

the voltage level at pin 2 of LM358 and the output at pin

1 goes high.The sensitivity of the IR sensor can be
controlled by variable resister R7.








6.
Program
Written
i
n Prototype

(C Language)

#define F_CPU 12000000UL //set your clock speed

#incl
ude <avr/io.h>



// including directory files

#include <util/delay.h>



int move_backward=0b01110100; // variable to define movements

int right_turn=0b10110100;

int fwd=0b10111000;

int stop = 0b00000000;

int left_turn=0b01111000;

int nt1_on = 0b10000000
; //variable defined to take input

int nt1_off = 0b00000000;

int nt2_on = 0b01000000;

int nt2_off= 0b00000000;

int nt3_on= 0b00100000;

int nt3_off = 0b00000000;

int sen1_on = 0b00000000; // variable defined to take sensor input

int sen1_off= 0b10000000;

in
t sen2_on = 0b00000000;

int sen2_off = 0b01000000;


10

int sen3_on = 0b00000000;

int sen3_off= 0b00100000;


int main (void)

{

DDRD = 0xFF; //Output port

DDRC =0x00;

// input port for command by user

DDRA = 0x00; //input port for IR sensor

int nt1=0;

int nt2 =
0;

int nt3 =0;

int sen1=0;

int sen2=0;

int sen3=0;

while(1)

{

sen1=(PINA & 0b10000000); // asisigning value to sensor variable

sen2=(PINA & 0b01000000);

sen3=(PINA & 0b00100000);

nt1=(PINC & 0b10000000); // assigning value to command variable

nt2=(PINC &
0b01000000);

nt3=(PINC & 0b00100000);


if(nt1==nt1_on)// detecting the command

{

if(sen1==sen1_on)//decision making from the input of sensors

{

PORTD=stop;

}

else if(sen1==sen1_off)

PORTD=fwd;


11

}


if(nt2==nt2_on)

{

if(sen2==sen2_on)

PORTD=stop;

else if (sen
2==sen2_off)

PORTD=fwd;

}

if(nt3==nt3_on)

{

if(sen3==sen3_on)

PORTD=stop;

else if (sen3==sen3_off)

PORTD=fwd;

}

}

}


7.

Summary
-

Realizations

i
n
Real

Scenario

This prototype
model
is developed considering ideal situation, like a single person resi
ding in
his house
in absence of his

family members or occasional visitors. Thus the imperfections
may
arise as sensors
can
respond to the presence
of external persons
as instead of robochair
or pre
-
defined (programmed) trajectories
may change from time to time bec
ause of presence of different shape obstacles
as shown in following figure
which makes

it difficult to ascertain the actual position of robochair
.

A complete collapse of the system is
also possible because of communication lapses for very short to long int
ervals specially
,

when navigation is
done using GPS.

There are certain other aspect
s

that may arise in the practical implementation, these
problems
and their poten
tial solutions are listed below
-



As for the problem of other person’s presence, a row of
phot
o
-
detectors

can be plant
ed along with
the touch sensors.

LASER Emitting Diodes with well ascertained frequencies

can be
fixed on the
robochair, so as to confirm whether the signal is originated by robochair and not any other object.



This prototype consider
s a single corridor, but the map of a house

as shown in following figure
,
multiplex or any rehabilitation centre

is
more complex
, with turns and multiple pathways leading to
various locations. So in
real practice
sensor has to be placed at each corner or t
urn to confirm that

12

path followed by the robochair to guide towards a desired location
in a

shortest and precise

manner
.

An adaptive planning is required when working in large scale environment.



Such local but complete modules can be more cost effective an
d can
have

necessary communications

with

a remote or global user via GPRS.


Obstacle detection is possible through the intelligent interfacing of IR sensors with the process
or

being
carried out within the microcontroller so that it can change its way to av
oid the obstacles.







Acknowledgements:
We express our deep acknowledgements to Sri B. K. Bajpai, ITRC, Madan Mohan
Malviya Engineering College, Gorakhpur, Uttar Pradesh, 273010 for proving computer facility for this
work.








13


R
E
FERENCES
-


1:

Cyrus

Farivar reports

-

http://www.engadget.com/2006/09/22/intelligent
-
wheelchair
-
helps
-
you
-


avoid
-
hazards/Intelligent wheelchair helps you avoid hazards

.



2:
Holly A. Yanco

Wheelesley, a Robotic Wheelchair System: Indoor Navigation and User


Interface
,
MIT Artificial Intelligence Laboratory

545 Technology Square, Room 705

Cambridge,


MA
02139
-

holly@ai.mit.ed

.





3
:

Sarangi
P. Pari!&, Rahul Rao', Sang
-
Hack
Jungt

Vijay
Kumar
James
P
.
Ostrowski
,
Camillo J.
Taylor,


Human Robot Interaction and Usability Studies for a Smart Wheelchair
,
Proceedings
of
the 20
03


IEEE/RSJ

Intl. Conference on Intelligent Robots and Systems

Las Vegas, Nevada . October 2003
.



4:
Tao Lu Kui Yuan Wei Zou Huosheng Hu
,
Smooth Path Planning for Intelligent Wheelchair



Based on Human
-
Machi
ne Interaction
,
Proceedings of the
2006 IEEE
International Conference


on Information Acquisition August 20
-

23, 2006, Weihai, Shandong, China
.



5:
Envisioning a Radar
-
Based

Automatic Road Transportation System
,
IEEE Intelli
gent Systems,


May
-
June 2001, p.75

{77).


6
:
http://rvsn.csail.mit.edu/wheelchair

See
-
Prof
.

Nick Roy,
Prof
.

Seth Teller
,

Dr. Bryan Reimer
,


Prof. Mike Mason,
Yoni

Batt
at , Sachi Hem
achandra,
Javier Velez
,
Mish Madsen
.


7:
A semi
-
autonomous robotic wheelchair
:

Institute of Computer Science, Foundation for Research



and Technology


Hellas (FORTH), Vasilika Vouton, P.O. Box 1385, 71110 Heraklion, Crete,


Greece;

E
-
mail: argyr
os@ics.forth.gr

.


8
.
Jin Sun Ju, Yunhee Shin and Eun Yi Kim
,
Vision based interface system for hands free control of


an intelligent wheelchair, Journal of NeuroEngineering and Rehabilitation,
2009
,
6
:33, p. 1
-
17.


9
.

www.isr.uc.pt/~urbano/Robchair.htm