Number of the records: 1
Fully probabilistic control design in an adaptive critic framework
- 1.
SYSNO ASEP 0364820 Document Type J - Journal Article R&D Document Type Journal Article Subsidiary J Článek ve WOS Title Fully probabilistic control design in an adaptive critic framework Author(s) Herzallah, R. (JO)
Kárný, Miroslav (UTIA-B) RID, ORCIDSource Title Neural Networks. - : Elsevier - ISSN 0893-6080
Roč. 24, č. 10 (2011), s. 1128-1135Number of pages 8 s. Language eng - English Country GB - United Kingdom Keywords Stochastic control design ; Fully probabilistic design ; Adaptive control ; Adaptive critic Subject RIV BC - Control Systems Theory R&D Projects GA102/08/0567 GA ČR - Czech Science Foundation (CSF) CEZ AV0Z10750506 - UTIA-B (2005-2011) UT WOS 000297000300012 EID SCOPUS 80054779788 DOI 10.1016/j.neunet.2011.06.006 Annotation Optimal stochastic controller pushes the closed-loop behavior as close as possible to the desired one. The fully probabilistic design (FPD) uses probabilistic description of the desired closed loop and minimizes Kullback–Leibler divergence of the closed-loop description to the desired one. Practical exploitation of the fully probabilistic design control theory continues to be hindered by the computational complexities involved in numerically solving the associated stochastic dynamic programming problem; in particular, very hard multivariate integration and an approximate interpolation of the involved multivariate functions. This paper proposes a new fully probabilistic control algorithm that uses the adaptive critic methods to circumvent the need for explicitly evaluating the optimal value function, thereby dramatically reducing computational requirements. This is a main contribution of this paper. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2012
Number of the records: 1