一种稳定的机器人神经网络控制器研究

Research on a Stable Neural Network Controller for Robot Manipulators

研究一种稳定的机器人神经网络(NN)控制器,提出了由神经网络控制器和监督控制器构成的控制方案,给出了控制器的设计方法及NN学习自适应律,并基于Lyapunov方法证明了控制系统的稳定性和NN参数收敛性。仿真结果表明该控制方案具有良好的鲁棒性和参数收敛性,从而证明控制器的有效性。

A stable neural network(NN) controller of robot manipulators is proposed. This controller is composed of a NN controller and a supervisory controller, the supervisory controller is introduced in order to assure the convergence of parameters and the stability of system. The design method of the controller and the adaptive learning rule of NN also are given. On the basis of the Lyapunov method, the system stability and the parameters convergence are proved. This controller can be utilized to all kinds of trajectory tracking control problems of robot manipulators. The simulation results demonstrate the feasibility of the proposed controller.