基于RWNN补偿的下肢外骨骼滑模控制

Sliding Mode Control for Lower Limb Exoskeleton Based on RWNN Compensation

针对下肢外骨骼系统精确动力学模型难以得到,且易受干扰等不确定性因素影响,提出一种基于递归小波神经网络(recurrent wavelet neural network,RWNN)补偿的滑模控制方法。结合拉格朗日原理和气动肌肉驱动特性,建立外骨骼系统模型,并将模型分为结构参数已知的标称部分和结构参数未知的不确定部分;对于标称部分,采用滑模控制方法进行控制,对于不确定部分,采用递归小波神经网络进行逼近;根据Lyapunov稳定性原理,证明了闭环控制系统的稳定性。搭建实验平台进行验证,结果表明外骨骼系统能够较好地跟踪期望轨迹,验证了所提控制方法的有效性。

A recurrent wavelet neural network(RWNN)compensation based sliding mode control method is proposed in view of the difficulty in obtaining accurate dynamic model of lower limb exoskeleton system and the recurrent wavelet neural network(RWNN)compensation.Combined with Lagrange principle and pneumatic muscle driving characteristics,the model of lower limb exoskeleton is established,and the model is divided into the nominal part with known-structural parameters and the uncertain part with unknown-structural parameters.The sliding mode control method is used to control the nominal part,and the recursive wavelet neural network is used to approximate the uncertain part.According to the Lyapunov stability principle,the stability of the closed-loop control system is proved.The experimental results show that the exoskeleton system can track the desired trajectory well,which verifies the effectiveness of the proposed control method.