Study On Ship Localization Using Satellite-based AIS Signals

Automatic Identification System(AIS)is a navigation aid system designed to ensure maritime safety and maritime efficiency.The satellite-borne AIS can receive AIS signals from a wide range of vessels to achieve its all-weather,long-range,all-round regulation of ships on the sea,and to ensure the safety of the ships at sea.However,the normal operation of the AIS system needs to be carried out with the help of the Global Navigation Satellite System(GNSS).When the GNSS equipment fails,the ship will lose its position and the navigation safety will also be threatened.In this dissertation,aiming at the situation that the ship GNSS equipment is out of service or the signal is disturbed deliberately,a ship positioning method based on satellite AIS system is proposed.(1)The AIS signal of the target ship is monitored by satellite to construct the joint positioning model of the FOA/TDOA.Then,considering the fact that the ship position does not change much in the visible time with the satellite AIS,the predicted ship position based on Kalman Filter is regarded as a new measurement vector and is added to the joint positioning model of the FOA/TDOA.On the basis of the least square estimation,the method of grid search combined with Gauss Newton iteration is adopted to solve the locating problem.The simulation results show that in the same environment,the blind spot of the improved positioning model is less than the FOA/TDOA,and the positioning error of the improved model is usually much smaller than the FOA/TDOA.(2)In order to fit the actual motion state of the ship,the Interacting Multiple Model(IMM)algorithm is proposed to locate the ship.In addition,considering the robustness of the Extended Kalman Filter,the Expectation Maximization(EM)algorithm is introduced into the IMM to estimate the ship state under a single model.The experimental results show that the stability and accuracy of EM-IMM positioning are better than that of the IMM algorithm,regardless of whether the ship trajectory is approximately straight line or curve.(3)On the basis of AIS signals transmitted by satellite,a ship localization method combing TDOA/FDOA and IMM algorithm is proposed and a new probability updating method in IMM is designed.The simulation experiment analyzes the improved model positioning effect with or without reference to the historical ship position.The results show that the ship autonomous positioning with the improved model has better accuracy and reliability.