Optimal Feedback Control for Continuous-Time Systems via ADP
In this chapter, we focus on the design of controllers for continuous-time systems via the ADP approach. Although many ADP methods have been proposed for continuous-time systems, a suitable framework in which the optimal controller can be designed for a class of general unknown continuous-time systems still has not been developed. Therefore, in the first part of this chapter, we develop a new scheme to design the optimal robust tracking controllers for unknown general continuous-time nonlinear systems. The merit of the present method is that we require only the availability of input/output data instead of an exact system model. The obtained control input can be guaranteed to be close to the optimal control input within a small bound. In the second part of this chapter, a novel ADP-based robust neural network controller is developed for a class of continuous-time nonaffine nonlinear systems, which is the first attempt to extend the ADP approach to continuous-time nonaffine nonlinear systems. Numerical simulations have shown that the present methods are effective and can be used for a quite wide class of continuous-time nonlinear systems.
- 4.Cui LL, Luo YH, Zhang HG (2011) Adaptive critic design based robust neural network control for a class of continuous-time nonaffine nonlinear system. In: Proceedings of international conference on modelling, identification and control, Shanghai, pp 26–29 Google Scholar
- 7.Khalil HK (2002) Nonlinear system. Prentice Hall, Englewood Cliffs Google Scholar
- 11.Lewis FL, Jagannathan S, Yesildirek A (1999) Neural network control of robot manipulators and nonlinear systems. Taylor & Francis, London Google Scholar