Control Strategy and Experimental Research of Cable-Driven Lower Limb Rehabilitation Robot

The passive training with fixed trajectory is more suitable for the initial rehabilitation training of patients with no muscle strength in the affected limb. In order to meet the needs of patients’ rehabilitation training in different rehabilitation stages and improve the active participation and rehabilitation training effect of patients, a fuzzy sliding mode variable admittance (FSMVA) controller based on safety evaluation and supervision is proposed for the cable-driven lower limb rehabilitation robot (CDLR) in this paper. The FSMVA controller consists of an inner loop fuzzy sliding mode controller, an outer loop variable admittance controller, and a safety evaluation and supervision module. Through the safety evaluation index of the CDLR system and the comprehensive evaluation of patients’ rehabilitation effect given by rehabilitation physiotherapists, a real-time adjustment algorithm of variable admittance controller parameters is designed, which can realize the real-time switching of training modes and adjustment of variable admittance controller parameters in active training mode and passive training mode. The trajectory tracking experiments with no admittance, fixed admittance, and variable admittance based on safety evaluation and supervision were carried out on the experimental platform. The experimental results show that the proposed FSMVA control strategy has high position control accuracy in the active training mode. In addition, the training intensity and the active participation of patients can be increased according to the patient’s exercise ability. In the passive training mode, according to the training needs of patients, the amount of the trajectory adjustment can be increased to improve the comfort of the training process and the safety of patients. It lays a foundation for the further study of the evaluation system of the rehabilitation training process and the experiment of human-machine interaction compliance.