A magnetorheological damper-based prosthetic knee (MRPK) and sliding mode tracking control method for an MRPK-based lower limb prosthesis

Abstract

Based on a two-bar linkage and a magnetorheological damper (MRD) with a double-ended structure and shearing operation mode of the magnetorheological fluid, an MRD-based prosthetic knee (MRPK) is realized. Utilizing the developed MRPK, an MRD-based lower limb prosthesis (MRLLP) is developed, modeled, and simulated in this paper, to analyse the effects of hysteresis of the integrated MRD on the swing angle of the shank of the MRLLP. Based on this, a sliding mode tracking control (SMTC) method for controlling the swing angle of the shank of the MRLLP is proposed to suppress hysteresis, along with a robustness analysis. Utilizing the SMTC method, co-simulations on controlling the swing angle of the shank of the MRLLP are carried out in ADAMS and Simulink. The simulation results show that the root mean square error (RMSE) of the swing angle of the shank of the MRLLP produced by the SMTC method is 80% less than that from the computed torque plus PD (CT+PD) control method. Therefore, the SMTC method is effective in suppressing hysteresis of the MRD. Furthermore, when the MRLLP is disturbed, the RMSE of the swing angle of the shank of the MRLLP produced by the SMTC method is 67% less than that from the CT+PD control method. Therefore, the SMTC method has strong robustness to random disturbance. A rapid control prototype of the MRLLP system and a corresponding experimental test system are established. On the established experimental test system, experiments are carried out on control of the swing angle of the shank of the MRLLP via the SMTC method. The results are compared with those from the ON/OFF and the CT+PD control methods. The experimental results show that the MRPK has controllable joint torque, and can be used to imitate the natural swing of a human knee joint. Additionally, the RMSE of the controlled swing angle of the shank of the MRLLP produced by the SMTC method is 34% less than that produced by the CT+PD control method and is 37% less than that from the ON/OFF control method.