Robust control using second order derivative of universal learning network-for system parameter perturbation
Machiko Ohbayashi, Kotaro Hirasawa, Masayuki Hashimoto, Junichi Murata · 2002
In this paper, we propose a robust control method using a universal learning network (ULN) and the second order derivative of ULN. The proposed method can realize more robustness than the commonly used neural network. The robust control considered here is defined as follows: even though the system parameter variables in a nonlinear function of the system at control stage change from those at learning, the control system is able to reduce its influence to the system and can control the system in a preferable way as in the case of no variation. In order to realize such robust control, a new term concerning the variation is added to a usual criterion function, and control parameter variables are adjusted so as to minimize the above mentioned criterion function using the second order derivative of the criterion function with respect to the parameters. Finally it is shown that the controller constructed by the proposed method works in an effective way through a simulation study of a nonlinear crane system.