Research On Control System Of Upper Limb Exoskeleton Rehabilitation Robot Based On ROS

The number of new stroke patients in China is increasing year by year due to changing dietary habits and increasing aging,and the reduced motor function of the limbs in stroke patients greatly affects their quality of life.The existing treatment method is physician-assisted rehabilitation,which is based on the experience of the physician,but as the number of patients increases and the training process becomes more labor-intensive,the number of physicians can no longer meet the training needs of many patients.The emergence of the upper limb exoskeleton rehabilitation robot has solved this contradiction to a certain extent.This study investigates the control system of an upper limb exoskeleton rehabilitation robot based on the ROS system with the starting point of securing the training effect of rehabilitation training and increasing the experimental verification of patients before actual training.The designed control system can intuitively obtain the real-time motion status of the rehabilitation robot,design and reproduce the rehabilitation movements,and record and analyze the operation results so that customized rehabilitation training planning can be performed for each patient.An adaptive iterative learning control algorithm is introduced to improve the robot trajectory tracking effect for the existence of non-repetitive interference in the robot motion process.The joint simulation using ROS and Matlab completes the information interaction between ROS and Matlab,allowing real-time control of the robot model in ROS through Matlab,and laying the foundation for the verification of the algorithm developed based on Matlab in the ROS system.Finally,the planning of the rehabilitation movements of the joints cooperating with each other was completed.The experimental results showed that the model could accurately complete the designed training movements and run smoothly,which illustrated that the designed control system could accomplish the training planning task of the rehabilitation robot.The main work and research contents are as follows:(1)The research progress of upper limb exoskeleton rehabilitation robots at home and abroad was analyzed and organized,and the needs and technical differences of rehabilitation robots were sorted out.Stroke patients were analyzed and summarized according to their disease development,and different training tools were applied to different stages of disease development.Based on the principles of rehabilitation medicine and the human body size of Chinese adults,the movement pattern of each part of the rehabilitation robot was confirmed and the structural rationality of the designed rehabilitation robot was verified.The hardware of the rehabilitation robot control system was selected and designed,and a complete rehabilitation robot control system was built.The improved DH modeling method was applied to perform forward and inverse kinematic analysis and motion space solution of the established robot model,and the results were verified by using Matlab.(2)Focused on analyzing the system framework of ROS system,the communication mechanism of topics and services,Rviz and Gazebo visualization tools and motion planning core Moveit! and completed the design of the rehabilitation robot software control system based on Moveit! The robot description file urdf was constructed using sw2 urdf in Solidworks,and the model file was verified.The configuration of the robot model Moveit!was completed using the Moveit! configuration assistant,and the rehabilitation robot simulation platform was built.The feasibility of using this system for the simulation control of the rehabilitation robot was verified by trajectory planning control based on Rviz and Gazebo.(3)Research on adaptive iterative learning control algorithms was carried out.An adaptive iterative learning control method with gain self-adjustment containing a robust control term is designed for the non-repetitive disturbances that exist during the robot model motion.Based on the Lyapunov stability criterion,the convergence and stability analysis of the designed controller is performed by establishing the Lyapunov function.The algorithm is designed and experimented by Matlab,and the experimental results show that the designed adaptive iterative learning control algorithm has obvious advantages compared with PD control and iterative learning control algorithm.(4)The joint simulation of ROS system and Matlab was completed.The communication principle of ROS system and Matlab was introduced,and the communication construction of ROS system and Matlab under topic-based communication mechanism was completed.According to the planning of the rehabilitation physician,the rehabilitation robot in the simulation platform smoothly completed the rehabilitation movements of the affected limb such as forward extension and abduction,demonstrating that the designed rehabilitation robot simulation platform has the ability to perform rehabilitation training planning,laying a solid foundation for the construction of a real upper limb exoskeleton rehabilitation robot,and providing an intuitive way to reproduce the algorithm developed based on Matlab.