基于改进型Volterra 基函数网络的直接自适应逆控制方法

构造一种改进型Volterra基函数网络，其特点是结构简单，容易离线确定最佳网络结构和初始权值。通过利用该网络在线学习非线性系统的逆，构造了一种非线性系统的直接自适应逆控制策略，并从理论上证明了闭环系统跟踪误差一致最终有界。仿真结果表明该方法的鲁棒性能良好。

Direct Adaptive Inverse Control Strategy Based on Modified VPBF Neural Networks

A modified Volterra polynomial basis function (VPBF) neural network with simple structure is constructed. The structure and the initial parameters are easily determined. A direct adaptive inverse control strategy is proposed by on line learning the inverse of the controller plant directly by using modified VPBF neural networks. The closed loop control system is proved to be uniformly ultimately bounded (UUB). A simulation result shows that the strategy has good robustness.