Asymptotic tracking by a reinforcement learning-based adaptive critic controller  

作　　者：Shubhendu BHASIN Nitin SHARMA Parag PATRE Warren DIXON 

机构地区：[1]Department of Mechanical and Aerospace Engineering,University of Florida [2]Department of Physiology,University of Alberta [3]NASA Langley Research Center

出　　处：《控制理论与应用（英文版）》2011年第3期400-409,共10页

基　　金：supported by the National Science Foundation (No.0901491)

摘　　要：Adaptive critic(AC) based controllers are typically discrete and/or yield a uniformly ultimately bounded stability result because of the presence of disturbances and unknown approximation errors.A continuous-time AC controller is developed that yields asymptotic tracking of a class of uncertain nonlinear systems with bounded disturbances.The proposed AC-based controller consists of two neural networks(NNs)-an action NN,also called the actor,which approximates the plant dynamics and generates appropriate control actions;and a critic NN,which evaluates the performance of the actor based on some performance index.The reinforcement signal from the critic is used to develop a composite weight tuning law for the action NN based on Lyapunov stability analysis.A recently developed robust feedback technique,robust integral of the sign of the error(RISE),is used in conjunction with the feedforward action neural network to yield a semiglobal asymptotic result.Experimental results are provided that illustrate the performance of the developed controller.Adaptive critic(AC) based controllers are typically discrete and/or yield a uniformly ultimately bounded stability result because of the presence of disturbances and unknown approximation errors.A continuous-time AC controller is developed that yields asymptotic tracking of a class of uncertain nonlinear systems with bounded disturbances.The proposed AC-based controller consists of two neural networks(NNs)-an action NN,also called the actor,which approximates the plant dynamics and generates appropriate control actions;and a critic NN,which evaluates the performance of the actor based on some performance index.The reinforcement signal from the critic is used to develop a composite weight tuning law for the action NN based on Lyapunov stability analysis.A recently developed robust feedback technique,robust integral of the sign of the error(RISE),is used in conjunction with the feedforward action neural network to yield a semiglobal asymptotic result.Experimental results are provided that illustrate the performance of the developed controller.

关 键 词：Adaptive critic Reinforcement learning Neural network-based control 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]