不确定机械手的自适应神经滑模控制

Neiral Network Based Adaptive Sliding Model Control of Uncertain Manipulators

针对不确定机械手的跟踪控制 ,提出了一种基于神经网络的自适应鲁棒控制器。该控制方案利用一个 Radial basis function神经网络逼近系统非线性不确定性 ,然后 ,通过一个滑模控制项消除网络逼近误差和外部干扰的影响 。

An adaptive sliding model control scheme based on neural networks is proposed according to the tracking control of uncertain robot-manipulators is this paper. The control scheme uses the nonlinear uncertainty of a RBF neural network approximation system, and then eliminates the effects of network approcximation errors and external disturbances with the help of a sliding model controller, which can guarantee the stability of a closed loop system and the asymptotic convergence of system tracking errors.