Adaptive Genetic Algorithm Observer and its Application to a Trailer Truck Control System

In this paper, we present an adaptive observer for nonlinear systems using a genetic algorithm (GA). It is considered that the GA has a superior performance for solution searching of multivariable systems. We utilize the GA searching ability to determine unobtainable variables in the control system. We extended the role of the GA observer to that of a parameter estimator for unknown systems and constructed an adaptive GA observer. The adaptive GA observer has a searching ability to estimate unobtainable variables and parameters simultaneously. In this paper, we use the estimated values of the system variables and parameters to design a neurocontroller (NC) using a GA. We applied the adaptive GA observer and NC design method to the backward movement control of a trailer truck. A simulation shows the effectiveness of the proposed method