Optimal Control for a Modified Bouc-Wen Model in a Magnetorheological Fluid Master Robot

Abstract

Magnetorheological fluid (MRF) clutch is a kind of passive actuator with advantages, like fast response and low inertia. However, due to the complex natural property, it is difficult to model such devices accurately, and this leads to degraded performance. In this letter, a steady-state model and transient model are combined to describe the MRF clutch. In the steady-state part, hysteresis phenomenon is accurately depicted by a modified Bouc-Wen model and a polynomial. In the transient part, a 1st-order system is applied to express the dynamic response. Model predictive control (MPC) is applied to enhances the response time and alleviate resulting hysteresis distortion of the clutch. Two experiments were implemented to verify the effectiveness of the design. The first reveals the high precision and response speed of the clutch, based on the proposed model and MPC. It shows that the average model error and settling time are improved to be less than 1 Nmm and 20 ms respectively. The second confirms the effectiveness of MPC in alleviating hysteresis distortion, and the minor errors prove the accuracy of proposed model even with multi-frequency commands. Thus, the proposed model and MPC are adequate for promoting the model accuracy and response time of the MRF clutch based force feedback devices.