Abstract: Individuals with cerebral palsy and muscular dystrophy often lack fi ne motor control of their fi ngers which makes it diffi cult 
to control traditional powered wheelchairs using a joystick. Studies have shown the use of surface electromyography to steer 
powered wheelchairs or automobiles either through simulations or gaming controllers. However, these studies signifi cantly 
lack issues with real world scenarios such as user’s safety, real-time control, and effi ciency of the controller mechanism. 
The purpose of this study was to design, evaluate, and implement a hybrid human–machine interface system for a powered 
wheelchair that can detect human intent based on artifi cial neural network trained hand gesture recognition and navigate 
a powered wheelchair without colliding with objects around the path. Scaled Conjugate Gradient (SCG), Bayesian Regularization (BR), and Levenberg Marquart (LM) supervised artifi cial neural networks were trained in offl ine testing on eight 
participants without disability followed by online testing using the classifi er with highest accuracy. Bayesian Regularization 
architecture showed highest accuracy at 98.4% across all participants and hidden layers. All participants successfully completed the path in an average of 5 min and 50 s, touching an average of 22.1% of the obstacles. The proposed hybrid system 
can be implemented to assist people with neuromuscular disabilities in near future.

Key words: Electromyography, Artifi cial neural network, Hybrid control, Powered wheelchair;Assistive technology, Hand gesture recognition