Research On Real-time Collision Risk Index For Inland Vessel-buoy Based On AIS

As an important marker of water transportation,inland navigation buoys directly identify dangerous waters on the navigation channel and provide navigation services for the vessels.However,accidents in which the vessel crashing the navigation buoy result in the loss of navigational aid function,which brought safety hazards to the navigation of the subsequent ship.Although the navigation status can be monitored to a certain extent by the construction of the “digital waterway”,the monitoring is still limited to the detection alarm after the navigation buoy is hit.It is not able to prejudge the collision accident at present.With the fully and effectively utilization of the abundant data resources provided by the construction of “digital waterway”,and the research on the anti-collision technology of navigation buoy,the vessel-buoy collision accident rate can be well reduced,the maintenance work of the navigation channel can be well improved,and finally we can ensure the safety of inland navigation.The navigation mark telemetry remote control platform and the Automatic Identification System(AIS)provide a large number of real-time dynamic information of the vessel and navigation buoy for the navigation management part.Based on the research results of Vessel-Vessel active collision avoidance and Vessel-Bridge active collision avoidance,and AIS information,the thesis carries out the prediction research on Vessel-Buoy collision risk,and provide theoretical support for the active collision avoidance technology of ship navigation mark.The main research works of the thesis are as follows:⑴ With the current and historical real-time dynamic information of the AIS are inputs,two inland ship track prediction algorithms are proposed,which includes: the vessel track prediction algorithm based on Cubic Spline Innovation-Kalman Filter(CS-IKF),and the vessel track prediction algorithm based on the joint Kalman filter of IKF and CS-IKF.The validity of each algorithm is verified by comparison experiments;⑵ Through the introduction of the inland river ship track prediction algorithm,the traditional Distance at Closest Point of Approach(DCPA)/ Time at Closest Point of Approach(TCPA)Model is improved to make it more suitable for the narrowing channel in inland waterway traffic and the frequent direction changes of the vessels;⑶ By introducing the improved DCPA/TCPA model,the Kunzi model originally applied to Vessel-Bridge collision risk analysis is improved,then a Vessl-Buoy collision risk algorithm based on improved Kunz model is proposed.The effectiveness and reliability of the proposed algorithm are verified by simulations and comparison experiments.