A ship motion control system design based on AND-OR fuzzy neural networks

This paper proposes a new AND-OR fuzzy neural networks (AND-OR FNN) based on the AND and the OR neurons, which are made up of T norm and S norm and have the ability to reduce the input space. Zadeh's operators are employed in order to infer the symbolic expression of every layer. A piecewise hybrid optimization approach is proposed to train the fuzzy neural networks to avoid the shortcoming of gradient algorithm, including the slow convergence, disturbance of local minimum and high dimension. The optimization procedure consists of two phases, first, the blueprint of network is reduced by GA (genetic algorithm) and PA (pruning algorithm); the second phase, the parameters are refined by ACO (ant colony optimization). An example of a 100 thousand ton oil tanker is made as the simulating plant, which includes three aspects, track control, course keeping and parameter changing. Simulation results show that the method proposed is effective.