Genetic programming based output regulation via optimal inversion for nonlinear systems

In this paper, we deal with an output regulation for nonlinear systems. Especially, when it comes to the output feedback control problem for nonlinear systems, it would be difficult to solve. From a practical point of view, it is an important issue to design a feedback controller using only output information. We introduce a certain kind of optimal inverse system for the output feedback control design. The optimal inverse system is a new concept. With the optimal inverse system, we obtain feedback controllers using only output information. In the proposed method, Genetic Programming (GP) also plays a key role. We utilize the GP's emergent ability for the control design. However, it is known that GP requires a lot of computation for evaluation of the individuals. The high computation load is one of the obstacles in the GP based control design. We adopt the so-called efficient GP, which is suitably modified for control design methods. Finally, some simulation results are given to demonstrate the effectiveness of the proposed control design.