Haptic Interface Design for a Novel Wheelchair Simulator using Linear Time-Varying MPC Framework

This paper deals with the design of a new haptic interface to drive a wheelchair locomotion simulator in real-time. The proposed haptic interface considers a reference model of a manual wheelchair (MWC) with a coupled straight line, turn and slope/cross-slope dynamics. This model allows both to reproduce the wheel-ground contact resistances in real-time and to estimate the kinematic motion states of the wheelchair. Then the theoretical concept of the model predictive control (MPC) scheme for linear time-varying (LTV) systems is exploited for the design of the haptic controller in order to handle the reference tracking problem. This controller is based on an LTV model of the MWC system coupled with the ergometer rollers of the simulator using an online identification of its dynamics. This formulation allows us to deal with nonlinearities and variations of the contact friction between the MWC wheels and the simulator rollers. The stability of the simulator haptic interface is demonstrated using the Lyapunov stability tools. Finally, the performance and effectiveness of the proposed haptic interface are evaluated by experimental tests on the PSCHITT-PMR simulator platform using standardized locomotion scenarios.