DocumentCode :
2647597
Title :
A standard comparison of adaptive controllers to solve the cart pole problem
Author :
Randall, Marcus ; Thorne, Colin ; Wild, Clyde
Author_Institution :
Fac. of Bus. & Hotel Manage., Griffith Univ., Brisbane, Qld., Australia
fYear :
1994
fDate :
29 Nov-2 Dec 1994
Firstpage :
61
Lastpage :
65
Abstract :
The cart-pole problem is an example of an inherently unstable dynamical system and has been widely used to test adaptive controllers. While various researchers have applied a range of adaptive controllers to solve the problem, there has been little work done on comparing the results of these methods. To compound this, researchers have used differing configurations of the cart-pole system. This work compares three common techniques used to solve the cart-pole problem. The cart-pole controllers are assessed on the basis of balancing time and centring/oscillation ability for a standardised configuration. Comparative results are given
Keywords :
adaptive control; multivariable control systems; nonlinear control systems; nonlinear dynamical systems; oscillations; stability; BOXES scheme; adaptive controllers; adaptive heuristic critic; balancing time; cart-pole problem; centring ability; controller assessment; genetic algorithms; inherently unstable dynamical system; oscillation ability; standardised configuration; Adaptive control; Control systems; Genetic algorithms; Gold; Neurons; Postal services; Programmable control; Robots; System testing; Unsupervised learning;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Information Systems,1994. Proceedings of the 1994 Second Australian and New Zealand Conference on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
0-7803-2404-8
Type :
conf
DOI :
10.1109/ANZIIS.1994.396949
Filename :
396949
Link To Document :
بازگشت