Neural-fuzzy control of truck backer-upper system using a clustering method

Truck backer-upper problem is a typical benchmark for many control methods in nonlinear system identification. In this paper, first, the traditional fuzzy control system is studied for the truck backer-upper problem, and then the fuzzy control system based on a hybrid clustering method and neural network is presented. The clustering method is proposed to construct an initial fuzzy model to determine the number of fuzzy rules from the intuitionistic-desired trajectories. Neural network is used to train the parameters of the constructed fuzzy model (neural-fuzzy system). Compared with traditional fuzzy system, this neural-fuzzy controller demonstrates advantages not only on the control performance but also on its convenience and feasibility.