Human-friendly motion control of power-assisted wheelchair

Abstract

Wheelchairs which have enhanced locomotion of people with muscular weakness or paralysis still can be improved using various assistive technologies, e.g., electric wheelchairs have been developed and widely used for enhancement of the maneuverability and safety of people. The power-assisted wheelchair (PAW) is a relatively new type of a wheelchair, which is equipped with torsion sensors to measure force applied on rims by a human and provides the assistive torque based on the measured force. With this propelling process, the user still can involve with the propulsion of the wheelchair while the assistive torque can relieve the user’s propulsion effort. This PAW is a case of human-machine interface and requires an adequate control of assistive torques to interact with a human without confliction. To address this issue, human-friendly motion control of the PAW is introduced in this chapter, which consists of the accurate observation of the physical states of the PAW and the disturbance rejection control based on the observed states to support safe and comfortable operation of the PAW. Several physical states of the PAW are defined as the operation states, which are associated with the motion of the PAW and the environmental condition, and they are observed utilizing multiple motion sensors (i.e., encoders, gyroscopes, and accelerometers) and an algorithm to fuse them. Then, control algorithms to remove the effect of undesired disturbance on the PAW without perturbing human’s operation are developed in a human-friendly way utilizing the observed operation states. The effectiveness of this human-friendly motion control of PAW is verified through various experiments using a commercialized PAW.