Title :
A nonlinear receding horizon controller based on connectionist models
Author :
Sbarbaro, D. ; Hunt, K.J.
Author_Institution :
Dept. of Mech. Eng., Glasgow Univ., UK
Abstract :
The authors focus on the development of connectionist architectures and learning algorithms for adaptive control of nonlinear systems. In particular, they pursue the receding horizon approach to optimal control. The control structure consists of two networks: one models the plan and provides prediction data for optimization, and the other is trained as an approximate plant inverse. Simulation results indicating the feasibility of the proposed approach are presented
Keywords :
adaptive control; learning (artificial intelligence); nonlinear control systems; optimal control; adaptive control; approximate plant inverse; connectionist models; learning algorithms; neural nets; nonlinear receding horizon controller; nonlinear systems; optimal control; Adaptive control; Control system analysis; Control system synthesis; Control theory; Cost function; Functional analysis; Inverse problems; Mechanical engineering; Nonlinear control systems; Nonlinear systems; Optimal control; Predictive models;
Conference_Titel :
Decision and Control, 1991., Proceedings of the 30th IEEE Conference on
Conference_Location :
Brighton
Print_ISBN :
0-7803-0450-0
DOI :
10.1109/CDC.1991.261281
