论文部分内容阅读
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 robot neural network (NN) controller is studied. A control scheme composed of neural network controller and supervisory controller is proposed. The controller design method and NN learning adaptive law are given. Based on Lyapunov method, The stability of control system and the convergence of NN parameters. The simulation results show that the control scheme has good robustness and parameter convergence, which proves the effectiveness of the controller.