Active Disturbance Rejection Control With Backstepping for Decoupling Control of Hydraulic Driven Lower Limb Exoskeleton Robot

摘要

The hydraulic driven lower limb exoskeleton robot (HDLLER), a power-assisted exercise device, can help wearers to improve their load-bearing capacity and reduce fatigue. However, the HDLLER is a nonlinear system and has strong coupling between the joints and parameter ingestion problems, which will affect its trajectory tracking performance. This article focuses on an effective decoupling control algorithm. First, the strongly coupled nonlinear system of the HDLLER is converted from the strict-feedback form into a canonical form through feedback linearization, which can avoid the differential explosion problem of the backstepping method. To reconstruct the state variables and nonlinear term of the transformed system, the extended state observer (ESO) which is a foremost component of active disturbance rejection control (ADRC) is adopted. Then, the backstepping controller based on ESO is designed and its stability is proved by Lyapunov's theorem. Finally, contrast experiments are implemented to confirm the correctness of the proposed algorithm, which indicates the presented approach can better suppress the coupling to improve tracking performance.