Joint angle adaptive coordination control of a serial parallel lower limb rehabilitation exoskeleton

A serial-parallel lower limb rehabilitation exoskeleton can achieve a kinematic function similar to that of the human lower limb. Based on this bionic structure, joint-angle coordination is needed for model-based control. Because the kinematics of the parallel mechanism is difficult to solve, real-time attitude control of the task space cannot be realized. Moreover, owing to structural errors and parameter uncertainty, accurate modeling cannot be realized, which causes difficulties in system control. Therefore, based on existing modeling methods, adaptive control of a serial-parallel lower limb rehabilitation exoskeleton is proposed. The sensor of the parallel mechanism can realize accurate perception of the attitude, and the controller is designed to solve the control problem of the task space of the parallel component. A radial basis function neural network adaptive controller is used to compensate for the uncertainty and external disturbance of the model. Experiments were conducted to verify the effectiveness of the dynamic modeling and control system.