Practical stability issues in CMAC neural network control systems

被引:59
作者
Chen, FC
Chang, CH
机构
[1] Department of Control Engineering, National Chiao Tung University, Hsinchu
关键词
D O I
10.1109/87.481771
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The cerebellar model articulation controller (CMAC) neural network is a practical tool for improving existing nonlinear control systems. A typical simulation study is used to clearly demonstrate that the CMAC can effectively reduce tracking error, but can also destabilize a control system which is otherwise stable. Then quantitative studies are presented to search for the cause of instability in the CMAC control system. Based on these studies, methods are discussed to improve system stability, Experimental results on controlling a real world system are provided to support the findings in simulations.
引用
收藏
页码:86 / 91
页数:6
相关论文
共 9 条
[1]  
Albus J. S., 1975, J DYNAMIC SYSTEMS ME, V97, P228
[2]  
Albus J. S., 1975, J DYNAMIC SYSTEMS ME, V97, P220
[3]  
[Anonymous], IEEE CONTROL SYS APR
[4]   CMAC-BASED ADAPTIVE CRITIC SELF-LEARNING CONTROL [J].
LIN, CS ;
KIM, H .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1991, 2 (05) :530-533
[5]   REAL-TIME NEURAL-NETWORK CONTROL OF A BIPED WALKING ROBOT [J].
MILLER, WT .
IEEE CONTROL SYSTEMS MAGAZINE, 1994, 14 (01) :41-48
[6]   APPLICATION OF A GENERAL LEARNING ALGORITHM TO THE CONTROL OF ROBOTIC MANIPULATORS [J].
MILLER, WT ;
GLANZ, FH ;
KRAFT, LG .
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH, 1987, 6 (02) :84-98
[7]   REAL-TIME DYNAMIC CONTROL OF AN INDUSTRIAL MANIPULATOR USING A NEURAL-NETWORK-BASED LEARNING CONTROLLER [J].
MILLER, WT ;
HEWES, RP ;
GLANZ, FH ;
KRAFT, LG .
IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION, 1990, 6 (01) :1-9
[8]  
Slotine J-JE., 1991, APPLIED NONLINEAR CO
[9]   LEARNING CONVERGENCE IN THE CEREBELLAR MODEL ARTICULATION CONTROLLER [J].
WONG, YF ;
SIDERIS, A .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (01) :115-121