Research on compliant human-robot interaction based on admittance control strategy for shoulder rehabilitation exoskeleton with CGH self-alignment function

In wearable rehabilitation exoskeleton human-robot interaction, the flexibility of movement is crucial to achieve safe and effective rehabilitation. Based on our previously proposed shoulder rehabilitation exoskeleton with Center of Glenohumeral (CGH) self-alignment function, this paper focuses on compliance control based on admittance control strategy. Firstly, the CGH compliant shoulder exoskeleton system is briefly introduced, along with the proposed admittance control strategy and real-time compliance trajectory planning. Then, a joint simulation platform combining Adams and Simulink was built for verifying the accuracy of the contact force-based method for obtaining human-robot interaction torque. Lastly, corresponding to differentiated rehabilitation needs of stroke patients at different stages, the admittance control simulation model is utilized to verify the exoskeleton's horizontal plane circular trajectory motion and three-dimensional space linear trajectory motion. The simulation results show that the joint end closely tracks external forces in both plane and space trajectories, returning to the reference trajectory after their disappearance. With maximum offsets of only 0.025 m and 0.03 m, this verifies the adaptability and tracking effect of the admittance controller in human-robot interaction forces, improving patient participation and meeting different rehabilitation stages training needs.