Design of an analytic constrained predictive controller using neural networks

Abstract

Abstract This paper shows how the solution of the standard predictive control problem can be recast as a continuous function of the state, the reference signal, the noise and the disturbances, and hence can be approximated arbitrarily closely by a feed-forward neural network. The existence of such a continuous mapping eliminates the need for linear independency of the active constraints, and therefore the resulting analytic constrained predictive controller will combine constraint handling with speed while being applicable to fast and complex control systems with many constraints. The effectiveness of the proposed controller design methodology is shown for a simulation example of an elevator model and for a real-time laboratory inverted pendulum system. Keywords: Model predictive controlAnalytic implementationNeural networks Acknowledgements This research was partially supported by programme POCTI, FCT, Ministério da Ciência e da Tecnologia, Portugal. Ton van den Boom was born in Bergen op Zoom (The Netherlands). He received his MSc and PhD degrees in Electrical Engineering from the Eindhoven University of Technology, The Netherlands, in 1988 and 1993, respectively. Currently, he is an Associate Professor at the Delft University of Technology, The Netherlands. His research interests are in the areas of linear and nonlinear model predictive control, robust control, modelling of uncertainty for robust control, discrete event systems and hybrid systems. Miguel Ayala Botto obtained his degree in Mechanical Engineering from Instituto Superior Técnico (IST), Technical University of Lisbon, in 1989, and received his MSc and PhD degrees in Mechanical Engineering from the same university in 1992 and 1996, respectively. He is currently an Associate Professor at IST and co-author of several international journal papers. His current research interests include control of hybrid systems, neural networks modelling and identification for robust control, and control of flexible robotic manipulators. Peter Hoekstra was born in Arnhem, The Netherlands, in 1976. He received the BSc (with Honours) and MSc degrees in electrical engineering from Delft University of Technology, Delft, The Netherlands, in 1995 and 2000, respectively. Currently, he works as a software designer for motion control applications at Philips Applied Technologies.