基于RBF网络的模型参考自适应控制

Model Reference Adaptive Control Based on RBF Network

针对一类非线性动态系统给出了一种基于RBF(径向基函数)神经网络的模型参考自适应控制算法,控制器的结构中使用RBF网络来动态的补偿系统的非线性性。基于Lyapnuov稳定性理论,给出了控制器参数的调整机制——σ-modification-type修正律,并根据神经网络的逼近误差给出了控制误差的估计,控制误差渐近收敛于0附近的一个紧集。仿真实例说明了所给出的算法切实可行。

The model reference adaptive control that based on RBF neural networks is proposed for a class of nonlinear dynamical systems. The controller algorithm employs a radial basis function network to compensate the nonlinearities adaptively. A stable controller-parameter adiustment mechanism based on the Lyapunov theory is constructed with a-modification-type updating law. The evaluation of control error in terms of the neural network learning error is studied. The control error converges asymptotically to a neighborhood of zero, whose size is evaluated and depends on the approximation error of the neural network. Simulation results showing the feasibility and performance of the proposed approach are given.