Predictive ship control using a fuzzy-neural autopilot

This source preferred by Martyn Polkinghorne

This data was imported from Scopus:

Authors: Richter, R., Burns, R.S., Polkinghorne, M.N. and Nurse, P.

Journal: Ship Control Systems Symposium, Proceedings

Volume: 1

Pages: 161-172

Intelligent methods of control which have been employed in an attempt to maintain optimal marine autopilot performance have all been retrospective in nature, thereby allowing performance levels to deteriorate before remedial action can be subsequently applied. Of significantly more interest would be a system which is capable of anticipating such performance deterioration prior to its occurrence so that corrective action may be applied in expectation of events by combining aspects of both modelling and control. Advancing from the classical approach to modelling the dynamic behaviour of rigid bodies by expressing behaviour as a set of simultaneous differential equations using calculated hydrodynamic coefficients, or by the application of a series of pseudo-random binary sequence (PRBS) to the real system, a novel alternative to the state variable representation of a ship in three degrees of freedom is demonstrated employing an artificial neural network approach. Using this enhanced model, it is therefore possible for the neural network model to predict the performance of the ship, and for this information to then be channelled to an intelligent control device, with any necessary rudder changes to optimize a rulebase of a fuzzy logic controller then being calculated in an anticipatory mode of operation.

This data was imported from Web of Science (Lite):

Authors: Richter, R., Burns, R.S., Polkinghorne, M.N. and Nurse, P.

Journal: ELEVENTH SHIP CONTROL SYSTEMS SYMPOSIUM, VOL 1

Pages: 161-172

The data on this page was last updated at 05:10 on February 17, 2020.