论文部分内容阅读
提出了一种新的小脑模(Cerebellar Model Articulation Controller,CMAC)神经网络标称补偿控制器.采用二阶扩展B样条CMAC网络平滑逼近机器人标称模型,消除了常规神经网络控制对输入的严格假设.为了确保系统闭环的全局稳定性,采用Lyapunov直接法设计网络权值的更新律,并引入非线性反馈项完全抵消补偿的残留项.未知的CMAC逼近误差和系统随机干扰,通过一个简洁的鲁棒自适应律估计.最后,针对两自由度机器人的仿真实例验证了所提算法的有效性.
A new nominal compensation controller for Cerebellar Model Articulation Controller (CMAC) neural network is proposed.A second-order extended B-spline CMAC network is used to approximate the robot nominal model smoothly, which eliminates the strict control of the conventional neural network control Hypothesis.In order to ensure the global stability of the closed-loop system, the Lyapunov direct method is used to design the updating law of network weights, and the nonlinear feedback term is introduced to completely cancel the compensated residual term. The unknown CMAC approximation error and system random disturbance are obtained by a simple Robust adaptive law estimation.Finally, the simulation of a two-degree-of-freedom robot verifies the effectiveness of the proposed algorithm.