Research On Lower Limb Exoskeleton Robot Based On The Flexible Tendon Sheath Actuator

With the growing problem of aging population in current society and the increasing number of people with motor disabilities,lower limb exoskeleton robots that can be used for rehabilitation training have shown profound research value in the field of medical rehabilitation.The existing design and research of rigid lower limb exoskeletons generally have problems such as large motion inertia,inaccurate recognition of human motion intention,and poor control performance.To address these problems,this paper designs a lower limb exoskeleton robot based on the flexible drive of tendon sheath,and adopts the Radials Basis Function Neural Network(RBFNN)sliding mode control algorithm to precisely control the system in order to realize the application of lower limb exoskeleton in human rehabilitation training.The main research work of this paper is as follows.1.This thesis also aimed to resolve the slack and hysteresis issues inherent in the tendon sheath transmission system through comprehensive static and dynamic model analyses.Elastic devices were embedded to mitigate these problems,and a modeling analysis of the flexible tendon sheath transmission system was conducted.To provide a thorough understanding of the lower limb exoskeleton robot,the Lagrange method was employed for dynamics analysis.Additionally,the modeling analysis of the lower limb exoskeleton robot system,which relied on the flexible tendon sheath transmission system,was completed by integrating the frictional resistance analysis of the lower limb exoskeleton human-machine coupling system.2.This thesis investigates the control strategy for lower limb exoskeleton rehabilitation,focusing on the development of a sliding mode control algorithm for the lower limb exoskeleton robot.The algorithm is designed based on the characteristics of a radial basis function neural network,and its stability is rigorously analyzed using Lyapunovs method.To improve the accuracy of human motion intention recognition for rehabilitation training,this thesis is to investigate the human lower limb motion mechanism and develop a reliable data acquisition system.To achieve real-time data acquisition,an inertial measurement unit(IMU)based device was designed specifically for human lower limb motion.The experimental design focused on the study of human walking gait.The data collected from the hip,knee,and ankle joints were analyzed,and the results were compared to the established analysis of human lower limb motion mechanism.The findings successfully validated the feasibility of the experimental device and established a solid foundation for the subsequent study of human motion intention recognition.3.In order to validate the studied exoskeleton robot and the control algorithm,the software and hardware platform of the lower limb exoskeleton robot control system was built,and the lower limb exoskeleton robot control system was designed,mainly including the main controller module,sensor module,communication module and power module.Finally,the lower limb exoskeleton was subjected to unassisted upright walking and assisted crutch walking wearing rehabilitation training experiments.The experimental results show that the position tracking errors of the hip and knee joints of the exoskeleton on the desired trajectory curve are between-2 and 2°,and the root mean square error of the tracking error is less than5 and the average percentage error is less than 10%,which proves that the lower limb exoskeleton robot has good trajectory tracking performance and can meet the demand of practical rehabilitation training.