Title :
Fuzzy neural control of systems with unknown dynamic using Q-learning strategies
Author :
Kwok, D.P. ; Deng, Z.D. ; Li, C.K. ; Leung, T.P. ; Sun, Z.-Q. ; Wong, J.C.K.
Author_Institution :
Hong Kong Polytechnic University, China
Abstract :
In this paper an efficient Q-learning paradigm implemented on a fuzzy CMAC network is proposed. The fuzzy CMAC network topological architecture is described. The continuous states of the system are partitioned into a number of fuzzy boxes. With the proposed fuzzy CMAC the Q-values of agents in the fired fuzzy boxes are evaluated and the control actions with maximum Q-values can be derived. The proposed hybrid adaptive and learning type of Fuzzy Neural control system based on the Q-learning is applied to the control of a pH-neutralization process.
Keywords :
adaptive control; cerebellar model arithmetic computers; fuzzy control; fuzzy neural nets; learning (artificial intelligence); neurocontrollers; pH control; step response; Q-learning strategies; fired fuzzy boxes; fuzzy CMAC network; fuzzy neural control; hybrid adaptive control; hybrid control architecture; learning type control; network topological architecture; pH neutralization process; reinforcement learning system; systems with unknown dynamic; Adaptive control; Control systems; Fuzzy control; Fuzzy neural networks; Fuzzy reasoning; Fuzzy systems; Neural networks; Programmable control; State-space methods; Sun;
Conference_Titel :
Fuzzy Systems, 2003. FUZZ '03. The 12th IEEE International Conference on
Print_ISBN :
0-7803-7810-5
DOI :
10.1109/FUZZ.2003.1209411
