Neural network control of wheelchairs using telemetric head movement

Powered wheelchairs are traditionally used by people with insufficient upper body strength and dexterity to operate a manual wheelchair. However, the operation of these wheelchairs can still be a difficult and stressful task. Head movement is a natural form of pointing and can be used to directly replace the joystick whilst still allowing for similar control. Through the use of artificial intelligence, a trainable wheelchair controller can be designed which provides an alternative control method with improved posture, ease of use and attractiveness. A computer simulation of this head controlled wheelchair has been successfully designed and tested. It consists of a motion detector for head motion measurement, a telemetry system for the elimination of wiring, and neural networks to provide the system with the ability to be trained for each individual operator irrespective of their disability