Physical-Based Hysteresis Modeling and Compensation for Magnetorheological Actuator With Asymmetric and Rate-Dependent Behaviors

Magnetorheological actuator (MRA) offers advantages such as controllability, low consumption, and inherent safety. However, its practical application is hindered by asymmetric and rate-dependent hysteresis, which degrades torque estimation accuracy and control complexity. Existing hysteresis models often rely on universal frameworks or introduce additional dynamic elements to capture rate-dependent behavior, leading to increased parameter complexity and limited physical interpretability. To address these issues, this paper proposes a novel physics-based hysteresis model that integrates the mechanical and electromagnetic properties of MRA to characterize asymmetric and rate-dependent hysteresis. Unlike conventional models, the structure of the proposed model is derived directly from the actuator’s physical principles, enabling more accurate torque estimation with few parameters. Furthermore, a self-rate-dependent operator is introduced to describe rate-dependent hysteresis without the need for extra dynamic terms, resulting in a more compact and efficient model formulation. Theoretical analysis and simulations validate the structural characteristics of the proposed model. In addition, an inverse model is developed and incorporated into a torque compensation strategy, effectively improving torque tracking performance. Numerous experiments demonstrate the feasibility and effectiveness of the proposed model in both hysteresis compensation and torque control.