Interactive Control And Experiment On Upper Limb Rehabilitation Robot Based On Data Driven

Stroke is a serious disease that severely affects patients’ normal life.Rehabilitation after treatment is extremely important for stroke patients.Currently,with a large population in China,the number of stroke patients is increasing day by day,and rehabilitation activities after treatment consume a large amount of medical resources.Therefore,rehabilitation robots have gradually become an effective way to alleviate the shortage of medical resources.The control methods of current rehabilitation robots are relatively single,and the implemented functions are relatively simple.Therefore,it is necessary to design an upper limb rehabilitation robot that can incorporate patient motion data for effective rehabilitation treatment for stroke patients.This paper established a dynamic model,proposed a data-driven control method,and conducted experimental research on the upper limb rehabilitation robot platform.The specific work completed is as follows:(1)A method for collecting human upper limb joint angle data using an optical motion capture system instead of an inertial sensor was proposed to improve data collection accuracy,specifically targeting the problem of increased time error in upper limb motion data collected with inertial sensors.(2)A dynamic model,including external disturbance terms and human-robot interaction force terms,was established based on the Lagrangian method,and combined with the upper limb rehabilitation robot experimental platform,laying the foundation for a data-driven dynamic model of the upper limb rehabilitation robot.(3)A data-driven control method for the upper limb rehabilitation robot was proposed by combining human upper limb motion data to address the human-robot rehabilitation training matching problem.The Lyapunov principle was used to prove the stability of the system.(4)An upper limb rehabilitation robot experimental platform was built,and simulation and actual platform experiments were completed.The experimental results show that the proposed data-driven control method can overcome external disturbances and model errors,improve the practicality and applicability of the rehabilitation robot,meet the requirements of upper limb rehabilitation training.