A simulation study on autonomous wheelchair control with visual feedback for following straight lines

Scope of the research is modeling, simulation and control design of an autonomous wheelchair with control purposes. Currently available electric wheelchairs fail in satisfying the needs of disabled people in many cases such as neuromuscular disabilities and reflex disabilities. Additionally, wheelchair outdoor navigation encounters high risk and serious challenges. The study aims to solve mobility problem of disabled people by adding visual information technology and efficient motion control and navigation design to standard electric wheelchair. Vision has proved to be the most versatile sensor for robotic applications, since it passively provides accurate and non-contact measurements of the environment with low cost. In this research, first the mathematical model of wheelchair dynamics is built in such a way that it describes the main characteristics and dynamical behavior of system. Then, vanishing point detection algorithm is applied. A computer simulation model is developed to observe the results of the study. Using the results of simulation tests, control algorithm is developed for efficient motor control. Linear displacement, position control, direction control and velocity of wheelchair are presented in the computer simulation. Motion control and tracking via visual feedback is achieved using a feedback control system with visual data collected from outdoor environment. The results show that the desired goals are achieved. Key Words: Visual servoing, Smart wheelchair, Assistive robotics.