Design And Control Strategy Of Exoskeleton System For Upper Limb Rehabilitation Training

Upper limb hemiparesis resulting from stroke is a prevalent condition that significantly impacts the quality of life of patients.To effectively address this issue,upper limb rehabilitation robots,which replace therapists for repetitive training,have become increasingly prominent in this research field.However,two issues remain unresolved.Firstly,current exoskeleton designs are primarily based on simplified representations of the upper limb anatomy,leading to poor adaptability to the human body and cumbersome control algorithms,even after structural optimization.Secondly,current upper limb rehabilitation exoskeletons mainly rely on physiological,force,and motion signals generated by the upper limb to infer human motion intentions,with little attention paid to compensatory abnormal movements in other body parts during rehabilitation training,which may pose a risk to patients.These issues significantly impact the safety and effectiveness of upper limb rehabilitation training for hemiparetic patients.Therefore,this study aims to address these two issues to improve the effectiveness of upper limb rehabilitation training for hemiparetic patients.Firstly,an in-depth analysis of the upper limb rehabilitation needs and its anatomical and physiological characteristics is presented,based on the upper limb rehabilitation movement mechanism.Subsequently,this dissertation proposes employing a rope-driven mechanism to explore this problem and focus on the challenges of human-robot adaptability that arise during the human-robot interaction process.This approach will provide a foundation for the subsequent design of the upper limb exoskeleton body.Subsequently,building on the previous research foundation,this study identified the optimal type of upper limb exoskeleton design.Using performance goals such as range of motion,endpoint stiffness,and flexibility as the basis,the shoulder mechanism parameters were optimized using multi-objective methods.Additionally,based on the human-robot motion chain theory,the number of degrees of freedom for the exoskeleton was determined using this approach.A comprehensive analysis of the upper limb exoskeleton system was conducted.Utilizing spinor theory to analyze the forward and inverse kinematics of the exoskeleton unit,this study provided a thorough examination of the upper limb rehabilitation training trajectory based on exoskeleton units.Furthermore,this study focused on analyzing the motion control algorithm of the adaptive mechanism unit from the perspective of human-robot adaptability.Through validation of the motion control algorithm of the adaptive mechanism unit based on exoskeleton unit trajectory-driven experimental results,this research laid the groundwork for the motion control of the exoskeleton system in subsequent studies to enhance human-robot adaptability.Finally,this dissertation addresses the prevalent issue of compensatory abnormal movements in upper limb rehabilitation training by proposing a method that utilizes a seat as a platform for multiple sensors.These sensors gather data such as force,displacement,and s EMG produced by the subject’s movement.Subsequently,various machine learning algorithms are employed to train a four-class classification for abnormal movements,from which the best model is selected as the detection method for irregular motion.Upon confirming the feasibility of this approach,the dissertation integrates the abnormal movement detection technique with the motion assistance provided by the exoskeleton system,resulting in an abnormal movement suppression strategy.The efficacy and practicality of this strategy are then validated through a series of experiments.This study comprehensively evaluated the safety and effectiveness of upper limb rehabilitation exoskeleton training from three distinct viewpoints: execution mechanism,motion perception,and decision-making in terms of behavior.Through enhancing the humanrobot adaptability between the exoskeleton and the upper limb and addressing compensatory abnormal movements resulting from upper limb rehabilitation training,this study significantly improves the scientific,safety,and effectiveness of upper limb rehabilitation training.The findings of this study proffer a practical and innovative solution for the development of upper limb rehabilitation exoskeletons in the domain of hemiparetic rehabilitation.