A wavelet- and neural network-based voice system for a smart wheelchair control

Power wheelchairs provide unique mobility for the disabled and the elderly with motor impairments. However, it is extremely difficult for people suffering from severe motor impairments, such as tremors, to control standard power wheelchairs. In this paper, we present a design of an automated powered wheelchair system that integrates the latest technologies to assist users with motor disability in moving around and sending help messages to four distinct destinations using SMS messages. A user can move the wheelchair using one of the three techniques integrated in the wheelchair: a joystick, direction buttons, or voice. Moreover, the speed of the wheelchair can be controlled using two buttons (slow and fast). For the recognition of the voice commands, the system uses wavelets and neural networks for feature extraction and classification, respectively.