Research On Data Driven Adaptive Berthing Method And Technology

With the rise of e-navigation strategy and the vigorous development of maritime transportation,the research on intelligent ships has attracted more and more scholars’ attention,and the research on automatic berthing of ships is one of the most difficult and complex problems in intelligent ships.According to statistics,over 70% of ship berthing accidents are related to the bad marine technology of the navigators in the port.Therefore,the study focus on automatic berthing technology of ships has profound practical significance.During the process of ship berthing,especially when the ship is close to the berth,and the speed is low,there are complex situations such as bank effects,shallow water effect,etc.,therefore it is difficult to use the model-based motion control algorithms to achieve precise control of ships.Moreover,the allowable error is small,and the single application of global positioning system（GPS）data cannot meet the positioning accuracy of automatic berthing during the process of ship berthing,therefore a kind of hardware architecture of real-time dynamic data collection is designed,and the method of direct motion control based on the feedback data is used to realize the automatic berthing of ships.In this method,the integrated control system of the experimental waterjet propulsion unmanned boat is established at first,and the real-time distance information of the ship’s different positions relative to the berth,bank are obtained by arranging microwave radar from an array at different positions on the ship.Also combined with the real-time position information of the ship,the real-time position and attitude of the ship relative to the berth are obtained by calculating the geometry method,and the parameters of the ship are obtained as well.The mathematical model of ship motion is established,and the mapping relationship of various parameters are verified by designed experiments.Then,according to the berthing rules and the precise definition of the berthing state of the ship,an automatic berthing method composed of the upper-level scheduling algorithm and the lower-level control algorithm is designed.The upper-level scheduling algorithm adopts the logic inference table based on the expert system,and combines the real-time data of the ship to calculate the command speed and course of the ship.The lower-level uses the model free adaptive control based on the real-time data for achieving the maneuverability of ship.Finally,the real ship experiments are designed to verify the above ship scheduling algorithm and ship motion control algorithm.By analyzing the experimental trajectory,which is verified the model free adaptive controller has good control effect,and the designed ship scheduling algorithm can realize the ship’s automatic berthing function.The experimental results show that the method is feasible and practical.