This paper describes the design of a variable structure PID controller using genetic algorithms for compliant motion control. This VSPID has a structure similar to that of a simple PID controller where the PD mode is used to speed up response, whereas the PI mode is applied to eliminate the steady state offset. Flexible changes of the controller modes and parameter values of the VSPID could therefore solve the conflict in nature between steady state accuracy and dynamic responsiveness. Due to the difficulties encountered in the application of traditional analytical and numerical methods for the design of variable structure systems, micro-genetic algorithm (uGA) is used for the determination of the different controller parameters. uGA is known to converge rapidly with optimal or near optimal results. In order to demonstrate the design procedure, the compliant motion control of a robot arm is used. Taking the integral of the absolute error as the performance measure it reveals that the proposed VSPID is better than the simple PID controller.
[1]
Darrell Whitley,et al.
A genetic algorithm tutorial
,
1994,
Statistics and Computing.
[2]
Yun Li,et al.
Genetic algorithm automated approach to the design of sliding mode control systems
,
1996
.
[3]
C.-L. Chen,et al.
Design and analysis of neural/fuzzy variable structural PID control systems
,
1996
.
[4]
Alberto Landi,et al.
ADVANCED VARIABLE STRUCTURE CONTROL OF HIGH PERFORMANCE DRIVES
,
1992
.
[5]
Yun Li,et al.
Genetic algorithms applied to fuzzy sliding mode controller design
,
1995
.
[6]
Jonathan L. Shapiro,et al.
Genetic Algorithms in Machine Learning
,
2001,
Machine Learning and Its Applications.
[7]
Yasuhiko Dote,et al.
Variable-structured robust controller by fuzzy logic for servomotors
,
1993,
IEEE Trans. Ind. Electron..