An Upper Limb Rehabilitation Training System Based On Improved FABRIK Algorithm And Virtual Reality

As China’s population ages,the demand for rehabilitation is growing rapidly,but there are not enough rehabilitation therapists in practice.In recent years,robot-assisted therapy has great potential in rehabilitation treatment.In this paper,a rehabilitation training system that integrates a rehabilitation exoskeleton robot with virtual reality is proposed and implemented.This rehabilitation training system enables personalized upper limb rehabilitation training.The physician can set the end motion trajectory of the rehabilitation exoskeleton robot at any time,and the system performs real-time robot inverse kinematic solving to calculate the angle of each joint of the rehabilitation exoskeleton robot,and the self-designed rehabilitation exoskeleton robot uses this as a command signal to control the rehabilitation exoskeleton robot to drive the patient’s upper limb for rehabilitation training according to the set trajectory.In order to achieve efficient and immersive human-machine interaction and improve the motivation of patients’ rehabilitation training,the paper also developed a virtual reality visual feedback game to assist patients’ rehabilitation training.The research content of the paper mainly includes:The paper designs a rehabilitation exoskeleton robot,which mainly analyzes the mechanical structure,workspace and control system of the upper limb rehabilitation exoskeleton robot.The upper limb rehabilitation exoskeleton robot consists of a main part and several moving parts,using aluminum alloy as the main material.The design includes the mechanism of shoulder joint,elbow joint and wrist joint.The workspace was analyzed by Monte Carlo method,and the results showed that the workspace of the rehabilitation exoskeleton robot can meet the upper limb motor rehabilitation tasks of hemiplegic patients.The control system adopts position,speed and current triple closed-loop PI control,combined with trajectory planning to realize real-time trajectory control of joint motion.In order to achieve real-time adjustment of the rehabilitation training trajectory,the system solves the set end motion trajectory of the rehabilitation exoskeleton robot in real time.In this paper,an improved FABRIK algorithm is proposed,which abstracts the parallelogram structure of the upper arm and the right-angle structure of the lower arm of the rehabilitation exoskeleton robot to form two abstract linkages.The pose of the abstract linkage is calculated iteratively using the improved FABRIK algorithm for the abstract linkage.The Euler angles of the abstract linkages are then calculated to obtain the required angles for each joint of the rehabilitation exoskeleton robot.The improved FABRIK algorithm has the advantage of high accuracy and fast solution speed in inverse kinematics solution,which is suitable for scenarios where the motion trajectory changes in real time.The joint angle values can be calculated within 3.6ms in the simulation experiments of the following task,and the average value of the distance between the end position and the target position is 0.064 mm.In order to allow patients to interact with the robot more naturally and flexibly,as well as to receive timely feedback and improve patients’ motivation for rehabilitation training,the paper also developed a virtual reality visual feedback game to assist patients’ rehabilitation training.A cognitive rehabilitation game was designed to assist patients in their rehabilitation training.This game was designed with cognitive rehabilitation levels,where common animals were made into silhouettes and patients were asked to guess what animals they were based on the silhouettes.In this way,not only can the patient’s cognitive ability be improved,but also the patient can be induced to make choices for upper limb movements.In this study,we propose a rehabilitation training system that combines a rehabilitation exoskeleton robot,a virtual reality scenario,and an improved FABRIK algorithm to provide patients with a personalized and immersive upper extremity rehabilitation training experience.We designed the mechanical structure,workspace,and control system of the rehabilitation exoskeleton robot,and used the improved FABRIK algorithm to achieve real-time rehabilitation trajectory adjustment.In addition,we have developed a cognitive rehabilitation game that allows patients to play games while performing rehabilitation training,improving their cognitive abilities and rehabilitation outcomes.This study brings new ideas for robot-assisted rehabilitation.