Application of a Neural Network Predictor / Controller to Dynamic Positioning of Offshore Structures

This paper describes an auto-regressive moving average (ARMA) functional-link neural network predictor/controller for dynamic positioning of offshore structures. The ARMA neural network predictor acquires the knowledge about the system through an on-line training using a small number of samples of the latest system status measured on board of the structure. The trained ARMA functional-link neural network is used with an optimal controller to control the output of the system. The accuracy and robustness of the ARMA predictor are demonstrated through the numerical simulations of two ship maneuvers. The neural network predictor/controller is applied to the dynamic positioning (station-keeping) of a ship in a uniform current with and without external environmental disturbances. The results of the numerical simulations are very satisfactory. Click here to review the complete paper > Click here to return to the session directory >

[1]  Yoh-Han Pao,et al.  Neural-net computing for real-time control of a ship's dynamic positioning at sea , 1993 .

[2]  Duo Li,et al.  DYNAMIC POSITIONING OF SHIPS USING A PLANNED NEURAL NETWORK CONTROLLER , 1995 .

[3]  Pao,et al.  NEURAL-NET COMPUTING FOR DYNAMIC POSITIONING OF VESSELS AT SEA , 1992 .

[4]  P.A. DeBitetto.  Fuzzy logic for depth control of unmanned undersea vehicles , 1994, Proceedings of IEEE Symposium on Autonomous Underwater Vehicle Technology (AUV'94).

[5]  Tamaki Ura,et al.  A quick adaptation method in a neural network based control system for AUVs , 1994, Proceedings of IEEE Symposium on Autonomous Underwater Vehicle Technology (AUV'94).

[6]  Yusong Cao,et al.  An initial assessment of fuzzy logic vessel path control , 1994, Proceedings of IEEE Symposium on Autonomous Underwater Vehicle Technology (AUV'94).

[7]  Robert Sutton,et al.  Approaches to Fuzzy Autopilot Design Optimization , 1997 .