下肢外骨骼助行机器人神经网络自适应滑模控制研究
Research on Neural Network Adaptive Sliding Mode Control of Lower Limb Exoskeleton Walking Aid Robot

穿戴式下肢外骨骼助行机器人是一种能够辅助人体进行仿人式步态运动的智能一体化设备，该系统需要良好的轨迹跟踪效果。为提高穿戴式下肢外骨骼助行机器人的轨迹跟踪精度及稳定性，基于拉格朗日法建立动力学模型，研究分析RBF神经网络自适应滑模控制算法。通过MATLAB-Simulink仿真，对比分析滑模算法和RBF神经网络自适应滑模控制算法，获得了两种算法下髋关节、膝关节的角位移及角速度控制曲线及力矩，并对比了两种算法的位置跟踪误差。实验结果表明，相比于滑模控制算法，RBF神经网络自适应滑模控制可以改善抖振情况，提高控制精度和抗干扰能力。
The wearable lower extremity exoskeleton walking aid robot is a kind of intelligent integrated device. It can assist the human body to perform human-like gait movement. The system needs a good track tracking effect. To make the tracking accuracy and stability of the system better, a dynamic model was established by Lagrangian method, and the RBF neural network adaptive sliding mode control algorithm was studied and analyzed. Through MATLAB-Simulink simulation, the sliding mode algorithm and RBF neural network adaptive sliding mode control algorithm are compared and analyzed. Through simulation, the angle and speed control curves of the hip and knee joints under the two algorithms are obtained. The torque curve is also obtained. On this basis, the position tracking errors of the two algorithms are compared. Experimental results show that RBF neural network adaptive sliding mode control can improve the chattering of sliding mode control, and improve the control accuracy and anti-interference ability of the system.