Kinematic Calculation and Simulation of Rehabilitation Robots under Neural Network Models

Limb damage after stroke can cause serious motor function damage, and early treatment and postoperative rehabilitation exercise are necessary. For patients with severe injuries, the recovery of their motor function requires a long process, and scientific and rigorous rehabilitation training will be beneficial for the recovery of their motor function. Traditional rehabilitation training mainly relies on the hands of rehabilitation physicians for one-on-one guidance, which not only reduces work efficiency but also increases the workload of physicians. With the continuous development of robotics technology, robot technology has been applied to rehabilitation treatment. It is an invisible rehabilitation robot that greatly simplifies traditional training methods, improves training efficiency, and accelerates the rehabilitation process by allowing rehabilitation physicians outside of the limbs to assist patients in completing rehabilitation training. The advantage of neural network models lies in their strong non-linear fitting ability and generalization ability, which can handle complex joint motion patterns and dynamic characteristics. By constructing a neural network model, the kinematic problems of rehabilitation robots can be transformed into a supervised learning or reinforcement learning training process, thereby achieving real-time prediction and control of parameters such as joint angle, position, and velocity. This calculation method based on neural network models can not only improve the kinematic performance of rehabilitation robots, but also adapt to the personalized needs of different users, providing more accurate and effective assistance for rehabilitation treatment. The experimental results have shown that during the movement process, the velocity curve of the joint gradually increases. The velocity in all directions decreases to 0 meters per second around 20 seconds. After 20-40 seconds, the velocity increases in the opposite direction and finally returns to 0 meters per second. The velocity fluctuation is smooth, which helps patients to perform rehabilitation exercises.