Feedback-error-learning neural network for the automatic maneuvering system of a ship

Recently optimal control theory has been applied to the control of the maneuvering motion of a ship. But there are some problems for practical use. So the author tries here to apply the feedback-error-learning neural network technique proposed by Kawato et al. (1987) to the automatic maneuvering system of a ship. The characteristics of the control system studied here are that the system has the inverse dynamics model of the controlled object and it is composed of the feedforward control loop besides the feedback control loop. The inverse dynamics model is considered as a neural model, and it is refined with learning. Here the control system for the follow-up control to the desired value and for the compensation of the influence from the disturbance is studied by computer simulation. It is recognized that the system has a self-tuning ability and a good controllability and that by adding the proportional term to the learning equation, the learning speed is hastened remarkably. This control system is expected to be put to practical use