Wheelchair is guided by voice commands full documentation

School of Applied technological sciences
Department of Mechatronics engineering
MotoChair: A Fully Motorized Voice-Operated
Wheelchair
Graduation Project I Report
by:
Ala’a Dweikat
Majd Al Yazori
Sa’ed AlDaraghmeh
Supervisor by:
Dr. Aiman Al-Share’, PhD
Dr. Nathir Rawashdeh, PhD
Submitted in partial fulfillment of
the requirements for the degree of
BACHELOR OF SCIENCE
in
Mechatronics Engineering
at
The German Jordanian University
Amman, Jordan
Summer 2012

MotoChair: A Fully Motorized Voice-Operated Wheelchair 2012
2
MotoChair: A Fully Motorized Voice-Operated Wheelchair
A Project by
Ala’a Dweikat
Majd Al Yazori
Sa’ed AlDaraghmeh
Supervised by:
_______________________,Head of Mechatronics Engineering Dept.
Dr. Nathir A. Rawashdeh
_______________________,Mechatronics Engineering Dept.
Dr. Aiman Alshare
Evaluated by:
_______________________,Dean, School of Applied Technical Sciences
Dr. Ziyad N.Masoud
_______________________,Mechatronics Engineering Dept.
Dr. Mohammad A. Nazzal
___________________________
Date

3
Acknowledgment
First and foremost, all our gratitude is due to Allah the Almighty for guiding us
through, and aiding us in completing this work successfully.
We would like to express our sincere gratitude to our supervisors Dr. Aiman
AlShare' and Dr. Nathir Rawashdeh, for their help, support and empowerment
throughout this project. We would also like to deeply thank and appreciate the
Mechatronics department at the German Jordanian University, for providing us
with great assistance and insight.
Finally, we would like to thank our families and friends for their love, support
and encouragement.
A. Dweikat
M. Al Yazori
S. AlDaraghmeh

4
Abstract
This project aims to design, integrate, program, interface and test a fully
motorized, voice-operated wheelchair. A regular standard wheelchair was used
as the main skeleton that has been on modified to meet this project’s goals. In
this project, the procedure of Mechatronic systems design was followed to
assure the quality of the final product i.e. the MotoChair. The project has
consisted of the following parts: Hardware, software, interface and testing.
Finally, a working prototype in addition to a complete comprehensive
documentation will be submitted prior to a project defence session at the
German Jordanian University.
Keywords:
Mechatronics engineering, Interface, Assistive technology, Mobility aids,
Powerchair, Motorized wheelchair, Voice recognition, Disabled, Battery
powered wheelchair.

5

6
Table of Contents
2.1.1 Advantages of DC series motors.......................................................................................... 11
2.1.2 Series Motor Operation .................................................................................................. 11
2.1.3 Reversing the Rotation.................................................................................................... 12
2.2 Power and Torque calculations .......................................................................................... 13
2.3 Gear Box Calculations:....................................................................................................... 15
2.4 Extra pulley calculations: ................................................................................................... 15
Gears.................................................................................................................................. 15
Belts................................................................................................................................... 16
Chapter 3: Electrical Design......................................................................................................... 21
3.1 Drive circuit...................................................................................................................... 21
3.2 Pulse width modulation PWM............................................................................................ 27
3.2.1 PWM Duty cycle andfrequency ................................................................................... 27
3.2.2 Averaging the PWMmicrocontroller output ................................................................. 28
3.3 Voice recognition kit.......................................................................................................... 28
3.3.1 What is voice recognition?........................................................................................... 28
3.3.2 Features of the HM2007 voice recognition kit.............................................................. 29
3.4 The Battery....................................................................................................................... 30
3.4.1 Wheelchair batteries overview .................................................................................... 30
Chapter 4: Conceptual Design...................................................................................................... 33
4.1 Functional Block diagram of the system.............................................................................. 33
4.2 Joystick commands flow chart............................................................................................ 34
4.3 Voice commands flow chart............................................................................................... 35
4.4 Component list.................................................................................................................. 36
Chapter 5: Control Method ......................................................................................................... 37
Chapter 6: Software and Interfacing ............................................................................................ 38
6.1 Microcontroller selection and programming....................................................................... 38
6.2 Voice recognition kit programming..................................................................................... 40
Programming.......................................................................................................................... 40
Testing Recognition................................................................................................................. 40
Chapter 7: Testing and Performance............................................................................................ 43
7.1 Actual speed measurement ............................................................................................... 43
7.2 accuracy of the wheelchair's voice commands at a quite environment................................. 44

7
7.3 accuracy of the wheelchair's voice commands at a noisy environment................................. 45
Chapter 8: Challenges and Difficulties.......................................................................................... 48
Chapter 9: Further development................................................................................................. 49
Possible additions to this project:................................................................................................ 49
References................................................................................................................................. 50
Appendix.................................................................................................................................... 51
Components' information and datasheets................................................................................ 51
200V N-Channel MOSFET..................................................................................................... 51
BJT (2N3055). ......................................................................................................................... 51
SERIES MOTOR WITH GEAR BOX .............................................................................................. 52
PIC 16F877A microcontroller................................................................................................... 52
Battery:.................................................................................................................................. 53
HM2007 speech recognition:................................................................................................... 54
Voltage Regulator:.................................................................................................................. 55
Datasheet from Ibn Sina Center for Medical Equipment and Services......................................... 56
Software Code........................................................................................................................ 57
Extra Pictures ......................................................................................................................... 62

8
Chapter 1: Introduction
This Document contains information about the project “MotoChair: A
Motorized Voice-Operated Wheelchair”. This project is done as a graduation
prerequisite for the Bachelor of Science in Mechatronics engineering at the
1.1 Project Description
This project aims to design, integrate, program, interface and test a fully
motorized, voice-operated wheelchair. A regular standard wheelchair was used
as the main skeleton that has been on modified to meet this project’s goals. In
this project, the procedure of Mechatronic systems design was followed to
assure the quality of the final product i.e. the MotoChair. The project has
consisted of the following parts: Hardware, software, interface and testing.
Finally, a working prototype in addition to a complete comprehensive
documentation will be submitted prior to a project defence session at the
Keywords:
Mechatronics engineering, Interface, Assistive technology, Mobility aids,
Powerchair, Motorized wheelchair, Voice recognition, Disabled, Battery
powered wheelchair.
1.2 Objectives
This projectaims to:
1. Design, integrate, program, interface and test a fully motorized, voice-
operated wheelchair.
2. Apply the major engineering skills and use the engineering knowledge
acquired fromthe study of Mechatronics engineering.
3. Appreciate the importance of coordinated team work.

9
4. Integratebetween the engineering knowledgeand the needs of
everyday life.
5. Know that any technology gains its importance fromthe value it adds to
our lives, and fromits role in solving a problemor satisfying a need.
1.3 Team
This project was conducted by the following Mechatronics Engineering
students at the German Jordanian University as a prerequisite to their
graduation, under the supervision and guidance of Dr. Aiman AlShare’, PhD &
Dr. Nathir Rawashdeh, PhD.
1. Ala’a Dweikat
2. Majd Al-Yazori
3. Sa’ed AlDaraghmeh
1.4 The theory behind this project
Assistive Technology (AT), that refers to hardware and software solutions for
persons with physical, cognitive or sensory disabilities, can help people to have
a more productive and pleasant lives. There are several physical
disabilities/conditions which require the use of a wheelchair including brain
injury, stroke, fractures, amputation, pulmonary disease, neurological
disorders, musculoskeletal diseases/injuries and spinal cord injuries. In such
cases the use of a wheelchair can bring an enhanced independence that will
increase the user’s quality of life. However some of the impairments cause
severe difficulties on the use of wheelchair manual or electric. [1]
In a survey aimed to collect information from patients concerning the
usefulness of new electric wheelchairs. The study concluded that 9 to 10 % of
patients who use power chairs and who received appropriate training “find it
extremely difficult or impossible to use the wheelchair for activities of daily

10
living”. Some of the pointed reasons aredifficulty on controlling the wheelchair
with a joystick, uncomfortable and inappropriate interface for the disability
(because users with severe motor impairments are unable to operate the
joystick smoothly).
For elderly patients arthritis is one of the major reasons for wheelchair use.
The repeated usage of joysticks and continuous wrist movements can be very
painful for an arthritic patient, and may result in reinforced difficulties. For the
referred groups of users a voice based interface is highly encouraging because
it represents a natural and simple way of controlling the device. For other
types of disabilities different types of interfacing devices can be used.
The first known application of a voice interface for controlling a wheelchair
was published by Staton et al.. The voice interface provided basic commands
for wheelchair movement and no real world experiences were performed. In
2002 the RoboChair have been developed, with three operating modes:
intelligent obstacle avoidance, collision detection, and contour following. It
uses mixed voice and joystick inputs and a fuzzy control system. A third-party
speech recognition system was used. The VOIC is a Voice Operated Intelligent
Wheelchair. The voice recognition uses a neural network for pattern detection
with a self organizing architecture. The systemwas evaluated on two scenarios
with distinct users and SNRs (Signal to Noise Ratio). The announced error rate
for the 5 words vocabulary, in Slovenian, was 1.8% for a quiet environment
and 6.4% with heavy background noise (no SNR was specified). [1]

11
Chapter 2: Mechanical Design
2.1 Series DC motors
In this project two 12V Series DC motors were used. Series motors are
commonly used as traction motors in many applications, as they offer high
starting torque, are robust, have a simple design and are relatively low cost.
Most of their applications are of an industrial nature, such as conveyors, but
are also common in road-going electric vehicles. DC series motors are an ideal
choice for battery-operated equipment over AC motors, as they don’t require
the use of expensive complicated inverter circuitry to convert the DC voltage
to an AC voltage required by the motor.
2.1.1 Advantages of DC series motors
• Hugestarting torque
• Simple Construction
• Designing is easy
• Maintenance is easy
• Costeffective
2.1.2 Series Motor Operation
Operation of the series motor is easy to understand. In Figure 2.1 you can see
that the field winding is connected in series with the armature winding. This
means that power will be applied to one end of the series field winding and to
one end of the armature winding (connected at the brush).
When voltage is applied, current begins to flow from negative power supply
terminals through the series winding and armature winding. The armature is
not rotating when voltage is first applied, and the only resistance in this circuit
will be provided by the large conductors used in the armature and field
windings. Since these conductors are so large, they will have a small amount of
resistance. This causes the motor to draw a large amount of current from the

12
power supply. When the large current begins to flow through the field and
armature windings, it causes a strong magnetic field to be built. Since the
currentis so large, it will cause the coils to reach saturation, which will produce
the strongest magnetic field possible. [7]
2.1.3 Reversing the Rotation
The direction of rotation of a series motor can be changed by changing the
polarity of either the armatureor field winding. It is important to remember
that if you simply changed the polarity of the applied voltage, you would be
changing the polarity of both field and armaturewindings and the motor's
rotation would remain the same.
Figure 2.1: DC series motor connected to forward and reverse motor starter.
Thomas E. Kissell. Industrial Electronics, Second Edition. © 2000.

13
The figure bellow demonstrates a schematic of the series DC motor:
Figure 2.2: Series DC motor schematic
2.2 Power and Torque calculations
The Rotational Speed of each of the available motors = 4000 rpm.
The relation between the rotational speed and power is given by the equation:
𝑇𝑜𝑟𝑞𝑢𝑒 =
𝑂𝑢𝑡𝑝𝑢𝑡 𝑃𝑜𝑤𝑒𝑟
𝑅𝑜𝑡𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑆𝑝𝑒𝑒𝑑
𝐸 𝑎 = 𝑉𝑡 − 𝐼 𝑎( 𝑅 𝑎 + 𝑅𝑓)
𝐸 𝑎 = 12𝑉 − 12.5 × (0.05Ω + 0.02Ω) = 11.125 𝑉
Where:
𝐸 𝑎 is the armature voltage (Volt).
𝑉𝑡 is the total voltage (12V).
𝐼𝑎 is the armature current (Ampere).

14
𝑅 𝑎 is the armature resistance (Ohm).
𝑅𝑓 is the field resistance (Ohm).
𝑷 𝒅 = 𝐼 𝑎 × 𝐸 𝑎
𝑷 𝒅 = 12.5 𝐴𝑚𝑝 × 11.125 𝑉 = 139.063𝑊
Where:
𝑷 𝒅 is the Electrical power developed inside the motor.
𝜏 𝑒 =
𝑃𝑑
𝜔 𝑚𝑜𝑡𝑜𝑟
=
139.063
2 × 𝜋 ×
4000
60
= 0.33198 𝑁. 𝑚
Where:
𝜏 𝑒 is the electrical torque developed inside the motor (N.m).
𝜔 𝑚𝑜𝑡𝑜𝑟 is the rotational speed of the motor (r.p.m).
𝜔 𝑚𝑜𝑡𝑜𝑟 = 2 × 𝜋 ×
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑜𝑡𝑎𝑡𝑖𝑜𝑛𝑠 𝑝𝑒𝑟 𝑚𝑖𝑛𝑢𝑡𝑒
60
𝑆𝑒𝑐𝑜𝑛𝑑
𝑚𝑖𝑛𝑢𝑡𝑒
𝑃𝑜𝑢𝑡 = 𝑃𝑑 − 𝑃𝑟
Where:
𝑃𝑜𝑢𝑡 is the mechanical output power of the motor (Watt).
𝑃𝑟 is the rotational power losses (%).
𝜏 𝑠ℎ𝑎𝑓𝑡 =
𝑃𝑑 − 𝑃𝑟
= 𝜏 𝑒 × 0.9 = 0.3 𝑁. 𝑚
Where:
𝜏 𝑠ℎ𝑎𝑓𝑡 is the mechanical torque on the shaft.

15
2.3 Gear Box Calculations:
By using a gear box with a step down ratio = 0.08 we find that:
𝜔𝑔𝑒𝑎𝑟 = 0.08 × 𝜔 𝑚𝑜𝑡𝑜𝑟 = 0.08 × 4000 = 320 𝑟𝑝𝑚
Where:
𝜔𝑔𝑒𝑎𝑟 is the rotational speed of the gear (r.p.m).
𝜔 𝑚𝑜𝑡𝑜𝑟 is the rotational speed of the motor (r.p.m).
And:
𝜏 𝑔𝑒𝑎𝑟 =
𝑃𝑜𝑢𝑡
=
125.666
2 × 𝜋 ×
320
60
= 3.75 𝑁. 𝑚
Where:
𝜏 𝑔𝑒𝑎𝑟 is the torque on the gear.
2.4 Extra pulley calculations:
Gears and belts transmit rotary motion from one shaft to another, often
changing speed and torque in the process. Gear sets are generally used where
the two shafts are close together. Belts and pulleys, or sheaves, on the other
hand, link shafts that are farther apart. [15]
Gears
A pair of gears reduces speed in proportion to the relative number of teeth.
The gear on a motor shaft is typically smaller and has fewer teeth than the one
on the machine shaft. The speed ratio is R = NL/NS, where NL = number of

16
teeth in large gear, and NS = number of teeth in small gear. If the large gear
has 40 teeth and the small one 20, the speed ratio is 2:1, and the machine
shaft turns once for two turns of the motor shaft. The speed is cut in half, and
torque is doubled.
Belts
A belt decreases speed in proportion to the diameters of its pulleys or
sheaves. The pulley on a motor shaft is typically smaller than the one on the
machine shaft. The speed ratio is R = DL /DS, where DL = diameter of large
pulley, and DS = diameter of small pulley.
Thus, if the diameters of the two pulleys are 40 in. and 20 in., the speed ratio is
2:1. As before, the speed is cut in half, but the torque is doubled.
The figure bellows shows the Gear drive and the belt drive that transmitters
the motion to the wheel through the pulley that is attached to it.
Figure 2.3: Gear drive and Belt drive.
Penton Media Inc. © 2012

17
Taking into consideration that the Power output from motor is reserved; it will
be transmitted to the pulley with small losses in tension in the belt.
𝐷 𝑔𝑒𝑎𝑟 = 5 𝑐𝑚
𝐷 𝑝𝑢𝑙𝑙𝑦 = 20 𝑐𝑚
Where:
𝐷𝑔𝑒𝑎𝑟 is the diameter of the small pulley attached to the gearbox.
𝐷 𝑝𝑢𝑙𝑙𝑦 is the diameter of the large pulley attached to the wheel.
To find the rotational speed of the large pulley:
𝐷1 𝑛1 = 𝐷2 𝑛2
0.05 𝑚 × 320 𝑟𝑝𝑚 = 0.2 𝑚 × 𝑛2 − −→ 𝑛2 = 80 𝑟𝑝𝑚
Where:
𝐷1 is the diameter of the small pulley attached to the gearbox.
𝐷2 is the diameter of the large pulley attached to the wheel.
𝑛1 is the rotational speed of the small pulley attached to the gearbox.
𝑛2 is the rotational speed of the large pulley attached to the wheel.
From the power equation:
𝑃2 = 𝑃1 × (1 − 𝑃𝑟)
𝑃2 = 125.666 × (1 − 0.10) = 113.099 𝑊
Where:
𝑃1 is the power of the small pulley attached to the gearbox (Watt).

18
𝑃2 is the power of the large pulley attached to the wheel (Watt).
𝑃𝑟 is the Frictional power losses in the belt (assumed 10%).
Thus:
𝜏 𝑝𝑢𝑙𝑙𝑒𝑦 =
𝑃𝑜𝑢𝑡
=
113.099
2 × 𝜋 ×
80
60
= 13.50 𝑁. 𝑚
Where:
𝜏 𝑝𝑢𝑙𝑙𝑒𝑦 is the torque on the large pulley attached to the wheel.
𝑉𝑒𝑙𝑜𝑐𝑖𝑡𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑏𝑒𝑙𝑡 ( 𝑉𝑏𝑒𝑙𝑡 ) =
𝜋 × 𝑛1 × 𝐷1
60
=
𝜋 × 320 × 0.05
60
𝑉𝑏𝑒𝑙𝑡 = 0.8377 𝑚
𝑠⁄
Where:
𝑉𝑏𝑒𝑙𝑡 is the theoretical velocity of the belt.
Assuming the max load = 100 kg:
𝜏 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑 = 𝜏𝑙𝑜𝑠𝑠𝑒𝑠 + ( 𝐽 ×∝ )
Where:
𝜏𝑙𝑜𝑠𝑠𝑒𝑠 is the torque losses (N.m).
J is the second moment of area (kg.m2
).
∝ is the rotational acceleration.
𝐽 = 𝑚𝑎𝑠𝑠 × 𝑟2
= 100𝑘𝑔 × 0.3𝑚2
= 9 𝑘𝑔. 𝑚2

19
Where:
𝑟 is the radius of the Wheel (It's diameter = 0.6 m).
The figure bellow shows the speed profile of the wheelchair. The speed profile
is obtained from the assumption that the wheelchair will accelerate from 0 to
0.5 m/s speed in 1 second. Then it will continue its motion in a constant speed
equals to 0.5 m/s.
Figure 2.4: Speed profile of the wheelchair
We want the wheelchair toreacha constant speed of (0.5 m/s) during (1
second):
Fromprofile speed we determine the angular acceleration and then the
required torqueas follows:
𝛼 =
𝑎
𝑟
Where:
0
0.1
0.2
0.3
0.4
0.5
0.6
01234
Speed Profileof the WheelChair
Speed (m/s)

20
𝛼 is the angular acceleration of the Wheel ( 𝑟𝑎𝑑
𝑠2⁄ ).
𝑎 is the linear acceleration of the wheel ( 𝑚
𝑠2⁄ ).
𝑎 =
0.5
𝑚
𝑠
1 𝑠
= 0.5 𝑚
𝑠2⁄
𝛼 =
0.5
𝑚
𝑠2
0.3 𝑚
= 1.666 𝑟𝑎𝑑
𝑠⁄
𝜏 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑 = 0.8 𝑁. 𝑚 + 9
𝑘𝑔
𝑚2
× 1.666
𝑟𝑎𝑑
𝑠2
= 15 𝑁. 𝑚
Because we have 2 motors (one for each wheel) we need a total torque of:
27.0 N.m. This value is based on the extreme conditions of operation, i.e. the
belt losses are10% and the Load = the max load = 100 Kg at the required speed
profile.

21
Chapter 3: Electrical Design
The electrical design of this project consisted of motors' drive circuit, battery,
Pulse width modulation, the voice recognition kit and the connections
between them, in addition to the connection with the joystick and
microphone.
3.1 Drive circuit
Drive circuit is a circuit used for the purpose of controlling another circuit of
component, in order to control and regulate their operation. Motor drives
have been developed to offer power supply for the motors and isolate the
electronic components, such as the ICs, from electrical problems.
One popular type of motor drive circuits is the H-Bridge (sometimes called: the
Full Bridge). It has been named that because it looks like the letter H when
viewed on the discrete schematic. An H-Bridge is an electronic circuit that
allows the voltage to be applied on the load in either direction. It is used to
allow DC motors to operate in two opposite directions i.e. forward and
Backward. The direction of rotation of a series motor can be changed by
changing the polarity of either the armature or field winding.
The figure bellow shows a schematic of a simple H-Bridge:
Figure 3.1: simple H-Bridge

22
© www.mcmanis.com/chuck/robotics/tutorial/h-bridge/
The four sides of the H-Bridge represent four switches that are activated in
pairs; (1) High left and Low right, (2) High right and low left. This will result in
controlling the current flow direction and therefore the direction of rotation of
the motor. In this project two H-Bridges were used; one for each motor.
Four BJT transistors were used as switches for the H-Bridge. They take their
signals from the microcontroller through resistors and transmit it to the four
relays that control the power supply of the motors, thus deterring the
direction of rotation of the motors and consequently the direction of the
wheelchair's motion.
The figure bellow demonstrates the connection of the drive circuit:
Figure 3.2: Motors' drive circuit schematic.

23
The table bellow shows the direction of rotation of the Wheelchair based on
the direction of rotation of the two motors.
Left motorRight motorDirection
ON ( CW )ON ( CW )Forward
ON ( CCW )ON ( CCW )Backward
ON ( CW )OFFRight
OFFON ( CW )Left
Table 3.1: Wheelchair motion
Where:
CW: Clockwise
CCW: Counter Clockwise
The table bellow shows the direction of rotation of the motors based on the
state of the transistors i.e. on – off.
Direction
Left motor Right motor
Transistor 1 Transistor 2 Transistor 3 Transistor 4
Non 0 0 0 0
Forward 1 0 1 0
Reverse 0 1 0 1
Right 1 0 0 0
Left 0 0 1 0
Table 3.1: the direction of rotation of the motors based on the state of the
transistors

24
Drive circuit simulation:
In order to verify our design prior to the circuit's actual construction a
computer simulation was conducted using the software: Proteus VSM version
7.6.
Proteus Virtual System Modeling (VSM) combines mixed mode SPICE circuit
simulation, animated components and microprocessor models to facilitate co-
simulation of complete microcontroller based designs. The simulation takes
place in real time (or near enough to it): a 1GMHz Pentium III can simulate a
basic 8051 system clocking at over 12MHz. Proteus VSM also provides
extensive debugging facilities including breakpoints, single stepping and
variable display for both assembly code and high level language source. [12]
The following figure shows the simulation:

25
Figure 3.3: Proteus VSM simulation

26
The pictures bellow show the actual testing circuit built for this projectand the
motors relay box:
The relay box
The drive circuit

27
3.2 Pulse width modulationPWM
Pulse width modulation is used to control the average power that goes a load
in a motor. A digital signal is used to generate an analogue output that is
intended to go to the motor. It operates by changing the average voltage level
through generating a constant frequency signal in which the pulse width is
manipulated or changed.
For example, if a the normal digital signal (5V = on , 0V = off) is to be changed
to an average of 2.5V, the signal 5V will be on for half of the time. As shown in
the figure bellow:
Figure : PWM digital output and average output
© http://www.best-microcontroller-projects.com/pwm-pic.html
3.2.1 PWM Duty cycle and frequency
The Duty cycle refers to the percentage of time that a signal is high or low. For
example, in the upper graph of above figure, the signal is 10% of the time high
so the averageis low. Whereas in the lower graph, the signal is 90% of the time
high so the average is high. In this project the duty cycles 100% and 50% to
represent to speeds modes High speed and low speed.
The importance of the frequency of the PWM depends on the driven device; it
should be high (KHz) if the target is to create a DC signal. The magnitude of it
depends on the desired output, the amount of tolerated error and the use.

28
3.2.2 Averaging the PWM microcontroller output
We need to average the PWM signal by converting it to a useful analogue
signal so that we can benefit from it. This is done by using a capacitor as a low
pass filter, in which the output frequency will decrease as the input frequency
increases.
3.3 Voice recognitionkit
In this projectthe HM2007 voicerecognition kit was used.
3.3.1 What is voice recognition?
Voice recognition is "the technology by which sounds, words or phrases
spoken by humans are converted into electrical signals and these signals are
transformed into coding patterns to which meaning has been assigned. While
the concept could more generally be called "sound recognition", but the focus
is on the human voice because we most often and most naturally use our
voices to communicate our ideas to others in our immediate surroundings. In
the context of a virtual environment, the user would presumably gain the
greatest feeling of immersion, or being part of the simulation, if they could use
their most common form of communication, the voice. The difficulty in using
voice as an input to a computer simulation lies in the fundamental differences
between human speech and the more traditional forms of computer input.
While computer programs are commonly designed to produce a precise and
well-defined response upon receiving the proper (and equally precise) input,
the human voice and spoken words are anything but precise. Each human
voice is different, and identical words can have different meanings if spoken
with different inflections or in different contexts. Several approaches have
been tried, with varying degrees of success, to overcome these difficulties. [13]

29
3.3.2 Features of the HM 2007 voice recognitionkit
 Single chip voice recognition CMOS LSI.
 Speaker –dependent isolates-word recognition system.
 External 64k SRAMcan be connected directly.
 Maximum 40 words can be recognized for one chip.
 Maximum 1.92 sec of words can be recognized.
 Two controlmodes are supported: manualand CPU mode.
 Responsetime: less than 300ms
 A microphonecan be connected directly.
 5V single power supply.
The figure bellow shows a schematic of the HM2007 voicerecognition kit:
Figure : HM2007 voice recognition kit
© http://www.imagesco.com/speech/sr07schematic.JPG

30
3.4 The Battery
3.4.1 Wheelchair batteries overview
[http://www.apparelyzed.com/wheelchair/electric-wheelchairs.html]
There are three different battery types that can be used in electric
wheelchairs. These are "Wet", "Gel", and the newer "AGM (Absorbed Glass
Mat)" types. Their properties are listed below:
Wet Batteries
Wet batteries use the chemical reaction between lead and sulphuric acid to
create electrical energy. As the batteries need filling with distilled water, they
do have a higher maintenance rate, but are lighter than Gel or AGM batteries.
Wet batteries are also prone to leakage, something which is important if you
intend to put your wheelchair in the hold of an aircraft.
 Positive Aspects
Cheaper.
Less vulnerable to overcharging.
Great performancewith carefulmaintenance.
Lighter per Ah (AmpereHour) compared to most Gel or AGMs.
 Negative Aspects
require maintenance.
Battery acid can leak, causing corrosion and damage to chair and wiring.
Not approved for airline travel.
High rate of self-dischargewhen left sitting (6-7% per month).
Gel Batteries
Gel batteries contain a mixture of sulphuric acid, fumed silica, pure water, and
phosphoric acid, which forms a thixotropic gel. As there is no liquid in the
battery, they do not leak or require maintenance like wet batteries.
No maintenance.
Cannot leak.

31
Operate better than wet batteries in low temperatures.
Less gas released when charging than wet batteries.
Approved for air travel.
Longer life cycle than wet batteries
Expensive.
More weight per Ah than wet batteries.
Susceptible to overcharging.
AGM Batteries
AGM batteries have an absorbent glass mat sandwiched between the plates,
saturated with acid electrolyte, but with none free to spill. This “sandwich”
allows uniform distribution of the electrolyte over the plates, and reduces the
chance of battery damage caused by vibration and jarring.
No maintenance.
Can’t spill or leak.
Shock resistant.
Minimal gasses released when charging.
Low self-dischargerate (3% per month at 77’F).
Approved for air travel.
Highest cost.
Susceptible to overcharging.
3.4.2 The battery used in this project
In this project a regular car battery (ACDelco 46B24L) was used as a proof of
concept; Due the availability of this type, its price and the lack of special
electrical wheelchairs' batteries in the market. However, using a car battery is
not a good choice for this application becauseit is a starting (Cranking) battery;
which is designed to deliver quick bursts of energy for short periods of time
and is not design for long operation periods. Whereas such application

32
requires a Deep Cycle battery; which has less instant energy, but greater long-
term energy delivery. [18]
These are some typical (minimum - maximum) typical expectations for
batteries if used in deep cycle service. There are so many variables, such as
depth of discharge, maintenance, temperature, how often and how deep
cycled, etc. that it is almost impossible to give a fixed number. [19]
 Starting: 3-12 months
 Golf cart: 2-7 years
 AGM deep cycle: 4-8 years
 Gelled deep cycle: 2-5 years
 Deep cycle (L-16 type etc): 4-8 years
 Industrialdeep cycle (Crown and Rolls 4KS series): 10-20+years.
 NiFe (alkaline): 5-35 years
 NiCad: 1-20 years

33
Chapter 4: Conceptual Design
4.1 Functional Block diagram of the system

34
4.2 Joystickcommands flowchart

35
4.3 Voice commands flowchart

36
4.4 Component list
 Regular Wheelchair (Adult size).
 Voice recognition kit.
 Drive circuit
o Microcontroller.
o Relays.
o Transistors.
o Resistors.
 Voltage regulator to get 5v for microcontroller.
 Battery 12v 40A.
 2 motors 12.5 A 12V 1/8 HP + A gearbox ( to increase torque and decrease
speed of the motor)
 2 Belts ( to connect pulley with motors )
 2 pulleys ( to increase torque and decrease speed of motor )
 Joystick

37
Chapter 5: Control Method
In this project an open-loop control approach was used in order to reduce the
system's complexity. We assumed that the user will control the wheelchair as
they see suitable and then adapt to the operation of the system. Although
some control on the wheels rotation must be added at least, but this might be
solved by manipulating the Pulse width modulation and adding a
compensation factor (multiplier) to the software code to make the weaker
wheel rotate faster than the other. This will help in overcoming the differences
of the speeds of rotation in the wheels.

38
Chapter 6: Software and Interfacing
6.1 Microcontroller selectionandprogramming
A microcontroller is a small computer built for the purpose of executing
specific tasks. It is manufactured on a single small integrated circuit (IC) and
used mainly in products that require giving the user some degree of control on
the operation. Microcontrollers are designed for embedded applications not
like the microprocessors or the other general purpose applications.
Microcontroller have many advantages that make it the number one candidate
to be used in automatically controlled devices such as remote controls, power
tools, office machines, automobile engine control systems and other
embedded systems.
Advantages of microcontrollers:
 Microcontrollers are cost effective and very cheap to replace while
microprocessors are 10 times more expensive.
 They require less power.
 All-in-one: they usually contain a CPU, ROM, RAM and I/O ports.
The above mentioned reasons made it very suitable for us to choose the
controller of the wheelchair to be a microcontroller (PIC 16F877A). The code
was programmed using MicroBasic programming language and then tested
and simulated in the German Jordanian University's Microcontroller and
Microprocessor laboratory, before being installed on the PIC microcontroller.
*Check the Appendixfor more information about the microcontroller used and
to see the software code.

39
The following Pictures demonstrate the PIC16F877A microcontroller and its
programmer:
Figure 6.1: PIC 16F877A microcontroller
Figure 6.2: PIC programmer

40
6.2 Voice recognitionkit programming
When the kit is turned on the HM2007 checks the static RAM periodically. In
case that everything is ok, the screen displays "00" and the red light goes on. It
is now ready for commands.
Programming
Programming the circuit starts by pressing (on the keypad) the word number
that is desired to save a sound command on it. The circuit can save up to 40
words. When the number(s) is pressed on the keypad, the red light goes off
and the number will be displayed on the screen. The next step is to press "#" to
start saving the sound command. Then the command can be spoken in a clear
way. Finally the LED should blink momentarily to show that the command has
been saved. The process continues until the desired number of commands
(given that it should be within the memory range of the kit) is saved.
Testing Recognition
The circuit is continually listening. Repeat a trained word into the microphone.
The number of the word should be displayed on the digital display. For
instance if the word "directory" was trained as word number 25. Saying the
word "directory" into the microphone will cause the number 25 to be
displayed.
Programming our commands:
In our project we saved each command two times (one in a quite environment
and the other in a noisier one) to assure that the system will be able to
recognize the commands in different situations without the need to implement
some sophisticated noise reduction or elimination techniques.
However, this method is not highly reliable and was used as a proof of concept
and will be altered in any further development ventures.

41
The table bellow shows the chosen numbers and the command saved on them:
Command numbers
Forward 1 2
Backward 3 4
Right 5 6
Left 7 10
Stop 11 12
Table 6.1: Voice kit commands
The following figures show a Picture and a schematic of the keypad, in addition
to Pictures of the real voice kit:
Figure 6.3: Voice Kit's Keypad
© 2007 Images SI, Inc.

42
Figure 6.4: The voice kit, the display and the keypad
Figure 6.5: Our Voice circuit connected to the drive circuit

43
Chapter 7: Testing and Performance
7.1 Actual speed measurement
The actual speed of the wheelchair was measured in a flat surfacewith a user
weight = 80kg. Severaltrials were conducted in which the distancecrossed was
4m, and the times were recorded. The averagetime of the trials resulted in 7s
per 5 meters. Therefore the theoretical and experimental speeds will be:
(1)Theoretical speed(𝑽𝒕𝒉𝒆𝒐𝒓𝒊𝒕𝒊𝒄𝒂𝒍 ):
𝜋 × 𝑛1×𝐷1
60
=
𝜋 × 320𝑟𝑝𝑚 × 0.05𝑚
60
= 𝟎. 𝟖𝟑𝟕𝟕 𝒎
𝒔⁄
(2)Experimental Speed(𝑽 𝒆𝒙𝒑𝒆𝒓𝒊𝒎𝒆𝒏𝒕𝒂𝒍 ):
Weight of the Testing person = 80 kg.
Distance: 5 meters
Time: 7 seconds
𝑉𝑒𝑥𝑝𝑒𝑟𝑖𝑚𝑒𝑛𝑡𝑎𝑙 =
5𝑚
7𝑠
= 𝟎. 𝟕𝟏𝟒𝟑 𝒎
𝒔⁄

44
7.2 accuracy ofthe wheelchair's voice commands at a quite
environment
A series of 10 trials were conducted in a quite environment (indoors with
minimum noise), and the responses of the system were observed to test the
percentage error of the voice recognition in a quite environment. The results
were as follows:
Trials Forward Reverse Right Left
Done Error Done Error Done Error Done Error
Trial 1
Trial 2
Trial 3
Trial 4
Trial 5
Trial 6
Trial 7

45
Trial 8
Trial 9
Trial 10
error 10% 10% 20% 20%
7.3 accuracy ofthe wheelchair's voice commands at a noisy
environment
A series of 10 trials were conducted in a noisy environment (outdoors with
cars'and surroundings'noise), and the responses of the system were observed
to test the percentage error of the voice recognition in a noisy environment.
The results were as follows:
Trials Forward Reverse Right Left
Done Error Done Error Done Error Done Error
Trial 1
Trial 2
Trial 3

46
Trial 4
Trial 5
Trial 6
Trial 7
Trial 8
Trial 9
Trial 10
error 40% 50% 60% 60%

47
7.4 Battery life
Due to the fact that the battery used in this project is a regular car starting
battery, we could not obtained any scientifically reliable data on its life time
per charge; because its operational life was short and we needed to recharge it
many times during the testing phase of this project. However, we have
contacted some commercial electric wheelchairs suppliers in the market and
obtained some data regarding the battery life time per charge (in terms of the
distance crossed) of the regular power wheelchairs.
The battery specifications for a commercial electric wheelchair obtained from
Ibn Sina Center for Medical Equipment and Services –Amman, Jordan:
Battery : 26AH X 12V X 2pcs. (option: 40AH)
Charger : 24V, 4AMP (Automatic Type)
Battery Weight : 20 Kgs (10Kgs X 2pcs.)
Continues travelling distance: 21 Km (Option 40AH batteries: 33Km)
*The data above are intended to illustrate the normal battery life per charge
for one of the available commercial electric wheelchair, and may not apply to
the case of this project due to differences in other factors.
To see the complete datasheet, refer to the Appendix.

48
Chapter 8: Challenges and Difficulties
Throughout the making of this project we encountered several challenges
difficulties. Some of them we were able to manage, but others we were not.
The major challenge was to find and obtain the required components from the
Jordanian market especially the motors and the battery.
Also we encountered a great challenge in finding the suitable pulleys and fixing
them to the wheels without affecting the motion of the wheels. Several trials
were done but none was successful. In order to overcome this obstacle, we
had to design a new mechanism for the wheels in which a new shaft was made
using a turning machine, then two sets of bearings were added to the shaft
outside the wheel. Thus we were able to fix the pulleys with the wheels
without affecting the motion of the latter ones.
The battery was another challenge and it very hard to find a suitable battery
for the electric wheelchair application, in an affordable price. Batteries of this
kind can be found at the medical equipment supplier but with very expensive
prices. Therefore we used a regular battery for the purpose of proving the
concept.
Finally, we faced several difficulties in programming and debugging the
software. The major one was that at the beginning after the code was
simulated and verified, it did not work in the real application. We tried
countless times to fix this problem and changed several PIC microcontrollers
but this did not solve the problem. After a long period of trials we found that
the voltage regulator was corrupted (the microcontroller was not getting the
required 5V) and needed to be replaced!

49
Chapter 9: Further development
After working on this project, we found that the electric wheelchair market in
Jordan is far away from saturation and it still needs a lot of contributions in this
area. This project can be –with a small effort- commercially worthwhile.
The major area of necessary improvement is the noise elimination in order to
get the best voice operation functionality. Many sophisticated methods can be
used such as Real-time adaptive noise cancellation for automatic speech
recognition, but an excellent quality microphone with noise isolation may be
sufficient.
Also a well reliable control method is very essential to this project and the
safety of its user. A PID controller may be added in addition to torque sensors
and speed feedback to ensure the perfect operation of the system in all
environments especially starting, stopping and breaking at inclined surfaces.
Possible additions tothis project:
After completing the major milestones of this project, one or more of the
following feature might be added to it.
1. Emergency stop system.
2. User Bio-Feedback.
3. Foldable multi-purposetable.
4. GPS emergency system(GPS/E911).
5. Solar battery charging.

50
References
[1] Coelho, L & Braga, D. zGo: A Voice Operated Wheelchair with Biosignal Monitoring for
Home Environments. Proceedings of the 2th International Conference on Software
DevelopmentforEnhancingAccessibility and Fighting Info-Exclusion, June 3-5, Porto Salvo,
Portugal. 2009. http://www.danielabraga.com/PDF/DSAI09_3.4.p84.pdf
[2] Mano, M et al.. Wheelchair for physically disabled people with voice, ultrasonic and
infrared sensor control. Aotonomous Robotics, 2, 203-224. 1995.
[3] http://en.wikipedia.org/wiki/Motorized_wheelchair
[4] http://www.nskelectronics.com/files/hm2007_voice_recog_kit.pdf
[5] http://www.imagesco.com/articles/hm2007/SpeechRecognitionTutorial02.html
[6] http://zone.ni.com/devzone/cda/ph/p/id/53
[7] http://www.lmphotonics.com/DCSpeed/series_dc.htm
[8] http://www.rehab.research.va.gov/
[9] http://electric-wheelchairs-pro.com/
[10] http://www.robotroom.com/HBridge.html
[11] http://www.mcmanis.com/chuck/robotics/tutorial/h-bridge/
[12] http://www.labcenter.com/index.cfm
[13] http://www.hitl.washington.edu/scivw/EVE/I.D.2.d.VoiceRecognition.html
[14] http://www.engineersgarage.com/microcontroller/
[15] http://www.microchip.com/pagehandler/en-us/family/16bit/architecture/home.html

51
Appendix
Components' informationand datasheets
200V N-Channel MOSFET
IRF640B
Features:
 18A, 200V, RDS(on) = 0.18Ω @VGS = 10 V
 Fast switching
BJT (2N3055).
Ic max VcE max hFE min Ptot max Category
(typical use)
15 A 60 V 20 117 W General
purpose, high
power

52
SERIES MOTOR WITH GEAR BOX
The name plate:
Vn (V) In (A) n1 Before gear
box
n1 After gear
box
Rg no.
12 12.5 4000 rpm 320 rpm 11AIM186002
PIC 16F877A microcontroller.
Pin diagram:

53
Feature for PIC 16F877A:
Key feature Pic 16F877A
Operating frequency DC – 20 MHz
Resets ( and delays) POR, BOR (PWRT,OST)
FLASH program memory (14-bit
words)
8K
Data memory (bytes) 368
EEPROM Data memory (bytes) 256
interrupts 15
I/O ports Ports A,B,C,D,E
Timers 3
Capture/Compare/PWM modules 2
Serial communication MSSP,USART
Parallel communication PSP
10-Bit Analog-to-digital module 8 input channels
Analog comparators 2
Instruction set 35 instructions
packages 40-pin PDIP
Battery:

54
HM2007 speech recognition:
Features for HM2007:
 Single chip voice recognition CMOS LSI.
 Speaker –dependent isolates-word recognition system.
 External 64k SRAM can be connected directly.
 Maximum 40 words can be recognized for one chip.
 Maximum 1.92 sec of words can be recognized.
 Two control mode is supported: manual and CPU mode.
 Response time: less than 300ms
 A microphone can be connected directly.
 5V single power supply.

55
Voltage Regulator:
Electrical Characteristics (KA7805)
(Refer to test circuit, 0°C < TJ < 125°C, IO = 500mA, VI =10V, CI= 0.33μF, CO=0.1μF,
unless otherwise specified)

56
Datasheet from IbnSina Center for Medical Equipment and Services

57
Software Code
Programwheelchair
dim j as byte
dim v as byte
main:
trisb = $ff 'define portb as input
trisd = $ff 'define portd as input
trisc = $00 'define portc as output
pwm_init (5000) 'initialize the PWMwith a desired frequency in
Hz
pwm_start 'starts PWM
if portd.3 = 1 then 'check the selector button if pressed or no
delay_ms (300)
goto joy
else
delay_ms (300)
goto voice
end if
joy:
while true
j = portd 'the four direction of joystick and the selector
button(selector d3)
select casej

58
case $18 'left
pwm_change_duty(90) 'changes PWM duty ratio
portc.4 = 0 'output to transistor1
case $28 'right
portc.4 = 1
portc.5 = 0
portc.6 = 0
portc.7 = 0
case $48 'forward
portc.4 = 1
portc.5 = 0
portc.6 = 1
portc.7 = 0
case $88 'backward
portc.4 = 0
portc.5 = 1
portc.6 = 0
portc.7 = 1
case else 'if we not press any button
portc.4 = 0
portc.5 = 0
portc.6 = 0
portc.7 = 0

59
end select
delay_ms(200)
if portd.3 = 0 then 'Confirmation of the selector case
goto voice
end if
wend
voice:
while true
v = portb 'the output of SRC
select case v
case $81,$82 'forward
portc.4 = 1
portc.5 = 0
portc.6 = 1
portc.7 = 0
delay_ms(1000)
case $83,$84 'backward
portc.4 = 0
portc.5 = 1
portc.6 = 0
portc.7 = 1
delay_ms(1000)
case $85,$86 'right

60
portc.4 = 1
portc.5 = 0
portc.6 = 0
portc.7 = 0
delay_ms(1000)
case $87,$90 'left
portc.4 = 0
portc.5 = 0
portc.6 = 1
portc.7 = 0
delay_ms(1000)
case $93,$94 'speed
portc.4 = 1
portc.5 = 0
portc.6 = 1
portc.7 = 0
delay_ms(1000)
case $91,$92 'stop
portc.4 = 0
portc.5 = 0
portc.6 = 0
portc.7 = 0
delay_ms(500)
case else 'for any disturbance
portc.4 = 0
portc.5 = 0
portc.6 = 0
portc.7 = 0

61
end select
if portd.3=1 then 'Confirmation of the selector case
goto joy
end if
wend
end.

62
ExtraPictures

63

64

Wheelchair is guided by voice commands full documentation

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (20)

Similar to Wheelchair is guided by voice commands full documentation

Similar to Wheelchair is guided by voice commands full documentation (20)

More from Majd Khaleel

More from Majd Khaleel (9)

Recently uploaded

Recently uploaded (20)

Wheelchair is guided by voice commands full documentation