On the intelligent control of a rotary crane, neural network and fuzzy logic approaches

This paper deals with the determination of the optimal control strategies of a rotary crane. Such system is represented by a sixth order nonlinear mathematical model. The proposed idea is based on the decomposition of the system into two interconnected subsystems which are linear with respect to their state and control vectors. The nonlinearities are located in the interconnexion terms. In this context, local feedback gains can be computed by solving a nonlinear Riccati equation, whereas, the local gains depend on the state vector of the other subsystem. Within this approach, a fuzzy logic system and a neural network have been constructed in order to identify these gains. Simulation results show the high performances of the proposals.