This presentation discusses controlling a wheelchair using electroencephalography (EEG) signals from the brain. The methodology involves collecting EEG data from nine subjects performing five mental tasks, extracting features from the EEG data using wavelet packet transform, and classifying the tasks using a radial basis function neural network. The classified mental tasks are then mapped to control directions for a wheelchair prototype. Experimental results showed 100% accuracy in classifying the five mental tasks of movement imagination, arithmetic tasks, geometric figure rotation, and relaxation. This BCI system has potential applications for assisting people with disabilities to control wheelchairs and other devices.

1. A Presentation on
Controlling Wheelchair
Using Electroencephalogram
Presented By:
Abu Shams Md. Shazid Reaj
Student ID: 120914
Electronics and Communication Engineering Discipline
Khulna University, Khulna-9208.
E-mail: s.reaj@yahoo.com

2.  Controlling Wheelchair Using EEG
24.11.2016 2
Contents:
 INTRODUCTION
 METHODOLOGY
 Subje cts
 EEG Data Acq uisitio n
 Expe rim e nt Paradig m
 Fe ature Extractio n
 Classifie r
 Hardware im ple m e ntatio n
 RESULT AND DISCUSSION
 CONCLUTION

3. 3
 Controlling Wheelchair Using EEG
INTRODUCTION:
 There are number of people who were physically
challenged such as fully paralised but only there mind
work properly.
 To improve their lifestyle this work aims at developing a
wheelchair system that moves in accordance with EEG
signal.
 To achieve this goal Wavelet packet transform (WPT) ,
Radial basis function neural network (RBFNN), Brain
computer interface (BCI) functions are needed implement.
24.11.2016

4. 4
 The electrical activity of the brain can be monitored in
real– time using electrodes, which are placed on the scalp
in a process known as electroencephalography (EEG).
 Radial Basis Function Neural Network was used to
classify the pre defined movements such as rest, forward,
backward, left and right of the wheelchair.
 BCI system enables the user to communicate with their
external surroundings using the brain’s electrical activity
measured as EEG.
 Wavelet Packet Transform (WPT) method was used for
feature extraction of mental tasks from eight channel EEG
signals.
 Controlling Wheelchair Using EEG
24.11.2016

5. 5
 Controlling Wheelchair Using EEG
METHODOLOGY >Subje cts:
 Nine right-handed healthy male subjects of age (mean: 23yr) having
no sign of any motor- neuron diseases were selected for this experiment.
 EEG data was collected after taking written consent for participation.
SL. NO. Subject Age Education
1 Subject 1 22 BE
2 Subject 2 21 BE
3 Subject 3 23 M.TECH
4 Subject 4 27 BE
5 Subject 5 23 BE
6 Subject 6 22 BE
7 Subject 7 27 M.TECH
8 Subject 8 22 BE
9 Subject 9 22 BE
TABLE I: CLINICAL CHARACTERISTICS OF SUBJECTS
24.11.2016

6. 6
 Controlling Wheelchair Using EEG
METHODOLOGY > EEG Data Acq uisitio n:
 Subjects were done for five
mental tasks for five days.
 Data was recorded for 10
sec during each task.
 Each task was repeated five
times per session per day.
 EEG was recorded using
eight standard positions C3,
C4, P3, P4, O1 O2, and F3,
F4 by placing gold electrodes
on scalp.
Figure1:- Montage for present study
24.11.2016

7. 7
 Controlling Wheelchair Using EEG
METHODOLOGY > EEG Data Re co rding Te chniq ue s :
24.11.2016

8. 8
 Controlling Wheelchair Using EEG
METHODOLOGY > Expe rim e nt Paradig m :
 The subject was comfortably lie down in a relaxed
position when checking the status of alpha waves.
 The EEG was recorded for 50 sec, collecting five session
of 10sec epoch each for the relaxed state.
 Five session of 10sec epoch for five mental task were
recorded, each with a time gap of 5 minutes.
Figure 2: Timing of the Protocol
24.11.2016

9. 9
 The following mental tasks were used to record the EEG
data.
 Movement Imagination: Plan movement of the right hand.
 Geometric Figure Rotation: see a complex 3D object
(30sec) removed the object visualize the object being
rotated about an axis.
 Arithmetic Task
 trivial calculation: multiply 2 by 3
 nontrivial calculation: multiply 49 by 78
 Relaxed: Relax with eyes closed. No mental or physical
task to be performed at this stage.
METHODOLOGY > Expe rim e nt Paradig m :
 Controlling Wheelchair Using EEG
24.11.2016

10. 10
 Controlling Wheelchair Using EEG
METHODOLOGY > Fe ature Extractio n :
24.11.2016
 The frequency spectrum of the signal was first analyzed
through Fast Fourier Transform (FFT) method.
 The FFT plots of signals from all the electrode pairs were
observed for the alpha frequency range(8-13Hz).
 Maximum average change in EEG amplitude as show in
Fig3.

11. 11
 Controlling Wheelchair Using EEG
METHODOLOGY > Fe ature Extractio n :
Figure 3: Maximum Average change in Amplitude of PSD
 By applying Wavelet packet transform on the original signal
wavelet coefficients in the (8-13Hz) frequency band at the 5th
level node (5, 3) were obtained. Twenty one coefficients have
been obtained from one second of EEG data.
24.11.2016

12. 1224.11.2016
 Controlling Wheelchair Using EEG
METHODOLOGY > Classifie r :
 For classification, Radial Basis Function Neural Network
(RBFNN) classifier was employed.
 A two layer network was implemented with 21 input vectors.
 Using RBFNN the five mental tasks were classified, as
shown in Tables II.
Tasks Accuracy
%
classifications
Movement Imagery 100 00100
Trivial Multiplication 100 01000
Geometric Figure Rotation 100 00010
Nontrivial Multiplication 100 10000
Relax 100 00001
TABLE II. CLASSIFICATION OF FIVE MENTAL TASKS

13. 13
METHODOLOGY > Hardware im ple m e ntatio n :
 Controlling Wheelchair Using EEG
Figure 4: Conceptual block diagram of the wheelchair controlled by EEG signals
24.11.2016

14. 14
 Controlling Wheelchair Using EEG
METHODOLOGY > Hardware im ple m e ntatio n :
 The motor driver required 3 bit of data. The output of
classifier was mapped into 3 bit as shown in table3.
 Using parallel port, Motor driver IC (IC L293) was interfaced
with computer as shown in Fig 5 for the wheelchair controller.
Figure 5: Circuit Diagram for wheelchair controller
24.11.2016

15. 15
 Controlling Wheelchair Using EEG
 In the circuit, P1 acts to enable the chip and combination of
P2 and P3 were used to control direction of wheelchair.
 The truth table for the above logic is shown in Table III with
polarities of motor of M1and M2.
METHODOLOGY > Hardware im ple m e ntatio n :
P1 P2 P3 M1
+ -
M2
+ -
TASKS
1 0 0 0 1 1 0 LEFT (L)
1 1 0 1 0 1 0 FORWARD(F)
1 0 1 0 1 0 1 BACKWARD(B)
1 1 1 1 0 0 1 RIGHT (R)
0 x x -- -- -- -- STOP(S)
TABLE III. TRUTH TABLE OF HARDWARE DESIGN
24.11.2016

16. 16
 Controlling Wheelchair Using EEG
METHODOLOGY > Hardware im ple m e ntatio n :
Figure 6: State diagram for Wheelchair Movement
 For movement imagery task, the output of parallel port would
be [1 0 0]. Due to opposite polarities, M2 motor would move
forward and M1motor backward which would be lead to left
movement of the wheelchair.
 Similarly, it works for others tasks.
24.11.2016

17. 17
 Controlling Wheelchair Using EEG
RESULT ANDDISCUSSION:
 The subjects were asked to mentally drive the wheelchair
From the starting point to a goal by executing the five different
Mental tasks.
Figure 7(a-d): Top view of random path
24.11.2016

18. 18
 Controlling Wheelchair Using EEG
RESULT ANDDISCUSSION:
 To complete task from staring point to goal, the subject performed
sequence of the mental tasks as shown in Table VI.
 They are named by Movement Imagery (MI), Trivial Multiplication (TM),
Geometrical Figure Rotation (GFR), Non Trivial Multiplication (NTM) and
Relax (R) to control direction of the power wheelchair.
Path a TM/
Forward
GFR/
Left
GFR/
Left
GFR/
Left
R/ Stop
Path b TM/
Forward
MI/
Right
MI/
Right
MI/
Right
R/ Stop
Path c TM/
Forward
GFR/
Left
GFR/
Left
GFR/
Left
R/ Stop
Path d TM/
Forward
MI/
Right
GFR/
Left
GFR/
Left
R/ Stop
TABLE VI. MATRIX OF MENTAL TASKS AND DIRECTION OF WHEELCHAIR
24.11.2016

19. 19
 Controlling Wheelchair Using EEG
PRACTICAL APPLICATION:
 Sports applications
 Therapy applications
 Neuroscience research applications
 Games applications
24.11.2016

20. 20
 Controlling Wheelchair Using EEG
CONCLUTION:
 This experiment is an attempt to control direction of wheel
chair via brain signals.
 To differentiate five mental tasks, wavelet packet transform
was employed for feature extraction and Radial basis function
neural network was used for classification.
 The experimental result showed 100% accuracy.
 This kind of system can also be used in a variety of
applications like–
􀂾 Environment control units (ECU’S)
􀂾 Helping disable people to directly interact with hand
held devices such as cell phones.
􀂾 Dealing with hazardous material/chemical at laboratories.
24.11.2016

21. 21
 Controlling Wheelchair Using EEG
References:
Controlling Wheelchair Using Electroencephalogram
“International Journal of Computer Science and Information Security” Vol. 8, No.2, 2010
Vijay Khare
Dept. of Electronics and Communication, Engineering
Jaypee Institute of Information Technology
Nioda, India.
Email : vijay.khare@jiit.ac.in
Jayashree Santhosh
Computer ServicesCentre
Indian Institute of Technology,
Delhi, India.
Email : jayashree@cc.iitd.ac.in
Sneh Anand
Centre for Biomedical Engineering Centre
Indian Institute of Technology
Delhi, India.
Email : sneh@iitd.ernet.in
Manvir Bhatia
Department of Sleep Medicine,
Sir Ganga Ram Hospital,
New Delhi, India.
Email : manvirbhatia1@yahoo.com
24.11.2016

22. Thanks
22
 Controlling Wheelchair Using EEG
24.11.2016

23. Questions ?Questions ?
 Controlling Wheelchair Using EEG
24.11.2016