文章基本信息

标题：Online neural network compensator for uncertain nonlinear MIMO systems based on RISE feedback
本地全文：下载
作者：Behnaz Hadi ; Alireza Khosravi ; Pouria Sarhadi 等
期刊名称：International Journal of Mechatronics, Electrical and Computer Technology
印刷版ISSN：2305-0543
出版年度：2014
卷号：4
期号：12
页码：1108-1131
出版社：Austrian E-Journals of Universal Scientific Organization
摘要：The paper pro poses an adap tive cont roller to t rack time-vary ing t rajectory for non linear MIIMO syst ems in p resen ce o f un certainty and distu rban ces. The sugg ested co ntroller emplo ys an online adap tive Gau ssian Radial Bas is Function (RBF) net work to estimate n onlin ear functions tog ether with Ro bust Int egral of the Sign of t he Error (RISE) contro l strateg y. RISE t erm is in jected in cont rol s tructure in order to eliminate neu ral n etwo rk error and ext ernal dist urbances . Th e weight matrix an d Gaussian fu nctio ns u pdat e laws of adap tive RBF that are obtained of Lyap unov analysis tuned based on s tate erro r information in o nline manner. Hen ce, there are no n eed to choice of cent ers and width of Gauss ian funct ion t hat are ess ential for RBF des irable act ing. The closed -loo p stab ility is g uaranteed b y a Lyap uno v stabilit y an alysis. Finally , simulat ions on t wo link robo t man ipulator illustrate t he p erformance of the d eriv ed trackin g co ntrol tech niqu e
关键词：Ad aptiv e RBF n eural network; Lyapu nov analy sis; Ro bust Integral of th e Sig n of the ; Erro r (RISE); uncertainty