Humanoid control of lower limb exoskeleton robot based on human gait data with sliding mode neural network

Lower limb exoskeleton robots offer an effective treatment for patients with lower extremity dysfunction. In order to improve the rehabilitation training effect based on the human motion mechanism, this paper proposes a humanoid sliding mode neural network controller based on the human gait. A humanoid model is constructed based on the human mechanism, and the parameterised gait trajectory is used as target to design the humanoid control system for robots. Considering the imprecision of the robot dynamics model, the neural network is adopted to compensate for the uncertain part of the model and improve the model accuracy. Moreover, the sliding mode control in the system improves the response speed, tracking performance, and stability of the control system. The Lyapunov stability analysis proves the stability of the control system theoretically. Meanwhile, an evaluation method using the similarity function is improved based on joint angle, velocity, and acceleration to evaluate the comfort of humans in rehabilitation training more reasonably. Finally, to verify the effectiveness of the proposed method, simulations are carried out based on experimental data. The results show that the control system could accurately track the target trajectory, of which the robot is highly similar to the human.