Abstract--In this paper iGoggle guided wheel chair is developed to assist the patients who are paralyzed below the neck. This iris controlled wheelchair makes use of eye movement for the control of wheelchair. Infra-red sensors are used to detect the iris movement and control the movements of wheelchair. Infra-red sensor senses the movement of eye ball and control the movement of wheel chair in different directions. The performance of iGoggle wheel chair is evaluated by measuring the eye movements.

Index Terms iGoggle, iris movement, IR sensor, wheelchair.

I. INTRODUCTIONBiomedical engineering is the application of engineering

principles to medical science. It helps the people in the field of medicine to diagnose and implement treatment in a better way to cure the disease. It also helps the patients with a better life. This paper deals about controlling a wheelchair using iris movement detection.

In bio signal based wheelchair, EEG (Electroencephalography EMG (Electromyography) and EOG (Electrooculography) are used to control the wheelchair [9].EMG signal is obtained from the action potential on the levator scapulae muscle (LSM). If the person is paralytic below the neck, this cannot be used. EOG is a technique of measuring resting potential of retina[12] [18]. The drawback of EOG includes need of frequent calibration, the noise generated between the electrodes’ contacts and the skin, the metabolic state of tissues and the difficulties arising due to muscle artifacts [10] [18]. Wheelchair can be guided by voice commands given by the user [14]. In most of the paralytic disabilities speech recognition is difficult. Vision based wheelchair can use EOG or a camera to capture the signals to control the movements of wheelchair [11].

People having multiple disabilities can control eye movements. Vision is one of our most valued senses and during the course of each day our eyes are constantly moving. Eyes are organs that detect light and convert it into electrical impulses in neurons. The eye is a complex optical system which collects light from the surrounding environment, regulates its intensity through a diaphragm, focuses it through an adjustable assembly of lenses to form an image, converts this image into a set of electrical signals, and transmits thesesignals to the brain through complex neural pathways that connect the eye via the optic nerve to the visual cortexand other areas of the brain. Light waves from an object (such

Lekha Das Associate Professor, Department of Electronics EnggK.J.Somaiya College of Engineering, Mumbai, IndiaEmail: lekha2005@gmail.com

as a tree) enter the eye first through the cornea, which is the clear dome at the front of the eye. The light then progresses through the pupil, the circular opening in the center of the colored iris. Fluctuations in incoming light change the size of the eyes pupil. When the light entering the eye is bright enough, the pupil will constrict (get smaller), due to the pupillary light response. Initially, the light waves are bent or converged first by the cornea, and then further by the crystalline lens (located immediately.behind the iris and the pupil), to a nodal point (N) located immediately behind the back surface of the lens. At that point, the image becomes reversed (turned backwards) and inverted (turned upside-down). The light continues through the vitreous humor, the clear gel that makes up about 80% of the eyes volume, and then, ideally, backs to a clear focus on the retina, behind the vitreous. The small central area of the retina is the macula,which provides the best vision of any location in the retina. Ifthe eye is considered to be a type of camera (albeit, an extremely complex one), the retina is equivalent to the film inside of the camera, registering the tiny photons of light interacting with it. Within the layers of the retina, light impulses are changed into electrical signals. Then they are sent through the optic nerve, along the visual pathway, to the occipital cortex at the posterior (back) of the brain. Here, the electrical signals are interpreted or seen by the brain as a visual image.There are three antagonistic muscle pairs, which relax and contract in order to induce eye movement attached to the globe of the eye. These pairs of muscles are responsible for horizontal, vertical and torsional (clockwise and counter clockwise) movement [18].

Fig.1. Eye muscles

The human input is the electronic signals produced by moving eyes. There are different ways to measure the signals produced by eyeball movements [1]. This system uses an iGoggle (Intelligent Goggle). The iGoggle has sensors [3]fitted on it.

A pair of IR sensors are fitted on the sides of the frame for the detection of right and left eye movements [7]. A pair of sensors at the center senses the signal when looking straight. The same pair is used for the detection of eye blink. The iGoggle detects the eye movements, which are used to guide the movement of the wheelchair.

iGoggle (Intelligent) Guided Wheel Chair

Lekha Das

978-1-4673-4866-9/13/$31.00 ©2013 IEEE

International conference on Communication and Signal Processing, April 3-5, 2013, India

582

II. WHEELCHAIR GUIDING MECHANISM The wheelchair control has two modules, the sensor module

and the wheelchair control module. These modules are interfaced using a microcontroller. The sensor module consists of iGoggle, signal conditioning circuit, direction guiding LEDs and level obstacle sensors. The IR sensor on the right and left of right eye and left eye respectively detects the movement of eye towards right and left. The IR pair on the middle of the eyes detects eye blink as well as straight gaze. Reflective type IR sensors are used.Reflective sensors incorporate an infrared emitter and photo detector adjacent to each other. When an object is in the sensing area, the emitted light is reflected back towards the photo detector, the amount of light energy reaching the detector increases. This change in light energy or photocurrent is similarly used an input signal in the application.

Fig.2. Block diagram of iGoggle guided wheelchair

Fig. 3 iGoggle

The gaze on right side directs the guiding mechanism to set the LED on right and left side gaze set the LED onleft. The middle LED is on when the vision is straight. It also detects valid eye blink (of 1sec duration) with which the wheelchair can go in start or stop mode[2].

Fig.4. Sensor modules with direction guiding LEDs

The collapsible direction guiding mechanism is equipped with three different colored LEDs located at 8-10 inches away from the headgear assisting the patient to fix the location for eye movements. This improves smooth turning movement mechanism. At start the Light Emitting Diodes (LEDs) (both green and red) are blinking. When iris is fixed on right side green LED on right is ON and other two are OFF. When fixed on left side green LED on left is ON and other two are OFF.

Two level sensors LS1 and LS2 to detect the floor leveling as shown in the table 1. As soon as any of these obstacles occurs the wheelchair comes to an emergency halt. An alarm is also provided to indicate the emergency halt, the alarm can be sound or a vibration mode if the subject is hearing impaired.

Fig.5. Signal conditioning of sensor.

TABLE I LEVEL SENSOR TRUTH TABLE.

LS2 LS1 Level Status of Floor 1 1 Leveled Floor 1 0 Up (Staircase) 0 1 Down (Slope) 0 0 Don’t Care

III. WHEELCHAIR MOVEMENT MECHANISM Sensor module keeps on checking left/right movement of

the eyes. This will keep on sending incremental pulse to the wheelchair control module. With respect to that the wheelchair will take ten-degree incremental turn for each incremental pulse (received from the sensor module). The anti turn movement will start when the subject looks straight [15]. Differential mode of control is used for the movement of the wheels. The front wheel drive gives a lower top speed than rear wheel drive chairs, but offers a good turning capability. The drive wheels are in front of the center of gravity while the rear wheels are casters. Two dc motors having a voltage rating of 12V each power the wheelchair[4]. A single-stage speed reducing gear is integrated with each motor. The major constraint on motor operation is thermal dissipation.

Heat Dissipated = I2 * R terminal

The torque the motor produces solely determines the current through a motor. Current and torque are related by the torque constant of the motor. Current through motor = torque produced / torque

The control mode for motors is PWM. It is a powerful technique for controlling analog circuits with a microprocessor's digital outputs. PWM digitally encodes the analogue signals. PWM signal is digital since at any instant of

Pre Amplifier

Notch Filter 100Hz

LPF 3Hz

Analog to TTL compatible

Microcontroller

583

time the dc supply is completely on or completely off. PWM cannot give full speed. So it cannot be used for fast controlling of speed. However, fast movement is not advisable here as a measure of safety precaution. When both motors are having PWM 50% each both motor move straight. . PWM duty cycle of Left Motor and Right Motor decides the turning mechanism. PWM duty cycle control techniques enable greater efficiency of the DC motor. Differential mode of control is used. For every 10% change in the PWM the motor turns 10°. After reaching maximum pulse width it starts with anti-turn mechanism to make the wheels straight. Then depending on eyeball position it moves [8].

TABLE II RELATIONS BETWEEN PWM AND DIRECTION Left Motor Right Motor Direction

10% 90% 40° Right 20% 80% 30° Right 30% 70% 20° Right 40% 60% 10° Right 50% 50% Straight 60% 40% 10° Left 70% 30% 20° Left 80% 20% 30° Left 90% 10% 40° Left

The driver circuit is having power transistor 2N3055. As hFEof 2N3055 is small, Darlington pair is used with SL100B.To completely isolate the microcontroller circuit from the interface circuit opto couplers (also known as opto isolators) circuit is needed. The opto coupler 4N33 which has hFE of 500 (Current Transfer Ratio) and maximum collector current of 125mA is used. The internal infrared LED transfer the infrared LED light intensity to the photo transistor. Based on this infrared LED light intensity the phototransistor will be turned ON or OFF, giving more current to drive this infrared LED will affect more current to flow on the phototransistor collector. This effect is known as the current transfer ratio (CTR). The 100% CTR means that all the current flow on the infrared LED will be transferred 100% to the phototransistor collector. Each time the DC motor turns ON or OFF, an instantly drop of voltage power outage may occur in the batteries. The capacitor, like the battery, can be continuously trickle charged until power delivery is needed. Immediately upon power outage, two 100μF capacitors connected in parallel with the battery will deliver the back-up power to the DC motor [5]. Once operating, the battery can deliver uninterrupted power to the DC motor and the capacitors can be idled. Note that the two DC motors should either be rotating clockwise together or counter clockwise together, or not rotating at all. It is impossible for one to be rotating clockwise while to other is rotating counter clockwise. But it is possible for one to be rotating in a certain direction while the other is not rotating at all: in this case the wheelchair should be steering either to the right or left [6]. No reverse movement of the wheel is used, as the user is handicapped. So H Bridge is not used. A protection diode 1N4001 is connected in parallel with the motor.

Fig.6. Interfacing of complete sensor module with driver

The output from iGoggle is connected to signal conditioning circuit to remove noise. The three LEDs shown are the direction guiding LEDs.

Fig.7 Microconntroller interface block diagram

The output from the sensor module (left eye movement, straight, right eye movement and blink) are given to the port of microcontroller [13]. Based on the signal from sensor, corresponding PWM is generated and given to the motors.

IV. EXPERIMENTAL RESULTS The simulation of the complete circuit is performed using

proteus 7.1. When the switch corresponding to the right sensor right sensor is closed the right motor rotates. The corresponding direction guiding LED turns ON. The left motor rotates when the switch identified as left sensor is ON. The direction guide LED on left side turns ON indicating left movement. There are eight sensors in the simulated version. They are to indicate whether there is a level up or down in all four directions. If level change happens the wheel chair stops [17]. A vibrating motor is attached to indicate the same.

Table III EXPERIMENTAL RESULTS OF IRIS MOVEMENTS Sr. No Operation Voltage Level

1 Blink 4V 2 Left Movement 0.5V 3 Right Movement 0.5V 4 Ambient Noise at 100Hz 0.8V

The table 3 shows the readings of iris movement measurements on a CRO. The voltage levels measured are without signal conditioning circuits.

584

Fig.8. Left Eye Left Sensor before signal conditioning

Fig.9. Left Eye Left Sensor after signal conditioning

The fig.9 shows the output of the sampling circuit when looked in the left direction. As this is not taken at the same time it is not the output of the signal conditioning circuit when fig.8 is given as the input to it. Errors due to ambient light are also there which has to be minimized. This can be done by protecting the sensors from ambient light.

Fig.10. Eye Blink before signal conditioning

The average length of a natural blink is 100-400 milliseconds. Eye Blink duration of 1 second is considered as a valid blink for starting and stopping of the wheel chair. In the figure more than one blink occurred.

Fig.11. Eye Blink after signal conditioning

Fig.12. Right Eye Right Sensor before signal conditioning

Fig.13. Right Eye Right Sensor after signal conditioning

Fig.13 gives the signal while looking towards right from the right side sensor after signal conditioning. The extra wave in the figure is due to an eye blink happened while taking measurements.

REFERENCE [1] R. Barea, L. Boquete, M. Mazo and E. López."Guidance of wheel chair

using electrooculography".Proceedings of the third IMAC International Multiconference on circuits, Systems, Communications and Computers (CSCC'99). Greece. July 1999.

[2] M. S. Reddy, A.Sammaiah, B. Narsimha and K. S. Rao International Conference on Multimedia and Signal Processing, 2011.

[3] P.Britto,Indumathi.J,SudeshSivarasu, and Lazar Mathew, ”Automation of Wheel chair Using Ultrasonic and Body Kinematics”, NCCI 2010 National Conference on Computational Instrumentation CSIO Chandigarh, INDIA, 19- 20 March 2010 .

[4] M.A.Fkirin, S.Badawy, A.S.El-SherbenyDriving a DC Motor by Numerically Manipulated Eye Signal Captured by EOG The Online Journal on Electronics and Electrical Engineering. (OJEEE) Vol. (2) – No.(1).

[5] Dr. Osman Yildirim Design Consideration of a Low Cost Battery-Operated Wheelchair, Vol.10 Issue 4 (Ver 1.0), September 2010 Global Journal of Researches in Engineering.

[6] Holly A. Yanco, Wheelesley: A Robotic Wheelchair System: Indoor Navigation and User Interface.

[7] Chern Sheng Lin, ChienWaHo,Wen Chen Chen, Chuang ChienChiu,MauShiunYeh: “Powered Wheelchair Controlled by eye tracking system.”,OpticaApplicata, Volume XXXVI,No.2-3,2006.

[8] Meng-Hui Hsu, Hsueh-Yu Chen, Jen-Yu Liu and Chien-Liang Chen, Dual-Purpose Wheelchair Mechanism Designs Proceedings of the International Multi-Conference of Engineers and Computer Scientists 2009 Vol II IMECS 2009, March 18 - 20, 2009, Hong Kong.

[9] Lai Wei KuiYuan,andHuosheng Hu Use of Forehead Bio-signals for Controlling an Intelligent Wheelchair, Proceedings of the 2008 IEEE International Conference on Robotics and Biomimetics Bangkok, Thailand, February 21 - 26, 2009.

[10] Poonam S. Gajwani, Sharda A. Chhabria, Eye Motion Tracking For Wheelchair Control, International Journal of Information Technology and Knowledge Management July-December 2010, Volume 2, No. 2, pp. 185-187.

[11] DjokoPurwanto , Ronny Mardiyanto · Kohei Arai, Electric wheelchair control with gaze direction and eye blinking.

[12] Rafael Barea, Luciano Boquete, Manuel Mazo and Elena López, System for Assisted Mobility Using Eye Movements Based on

585

Electrooculography, IEEE Transactions On Neural Systems and Rehabilitation Engineering, Vol. 10, No. 4, December 2002.

[13] Jan Axelson, The Microcontroller Idea Book Circuits, Programs, & Applications featuring the 8052-BASIC Microcontroller.

[14] Manual. M, Francisco. J. R, Jose. L.Z, Jesus. U, Juan. C. G, Wheelchair for physically people with voice, ultrasonic, IR sensor control, Kluer Academic Publication, Boston.

[15] TorstenFelzer, Martin Ernst, Bruno Strah and Rainer NordmannAccessibility Research at the Department of Mechatronics atDarmstadt University of Technology, Sigaccess News Letter, Issue 88,June 2007.i.

[16] Timothy Bourke, Development of a robotic wheelchair, November 2001. [17] S. Yathunanthan , L. U. R. Chandrasena, A. Umakanthan, V. Vasuki and

S. R. Munasinghe, Controlling a Wheelchair by Use of EOG Signal. [18] Available on line: www.medicin.mcgill.ca/phisio/vlab/other_exps/EOG/

sccc ades_n.htm..

586