Design and implementation of low-cost intelligent wheelchair controller for quadriplegias and paralysis patient

Abstract

Wheelchair controller selection is very important for users or patients with disabilities like quadriplegia, paralyzed, amputee and elderly. The comfort, simplicity, and safe use, directly affect their social efficiency and quality of life. In this paper, the design and test of a new low-cost intelligent wheelchair controller for users with disabilities has been proposed. It consists of two main controllers, which are the voice controller and the head tilt controller. The voice controller consists of two voice recognition modules that use dynamic time warping DTW and hidden Markov model HMM to implement the voice recognition. This controller has been tested with three global languages English, German, and Chinese at two different noise levels 42 and 72 dB. The voice controller was tested with a motion command and can be used for other parameters like light, sound alarm, etc. The second controller is the head tilt controller. It uses a sophisticated orientation detection module BNO055 to detect and track the user's head orientation and the wheelchair orientation basing on Euler angles. It has an auto-calibrated algorithm to calibrate the control commands and adjust the speed of the wheelchair motors in case of ascending or descending ramp or when passing non-straight roads. The head tilt controller was tested successfully for indoor and outdoor applications.