Research And Implementation Of Unmanned Ship Autonomous Obstacles Avoidance Based On Improved Artificial Potential Field

In recent years,with the rapid development of new concepts and technologies such as the internet of things,big data,cloud computing,and artificial intelligence,the level of intelligence of ships has been continuously improved,and the development of ships has gradually developed toward unmanned driving.As an intelligent ship that can achieve complete autonomous navigation,the unmanned ship not only gets rid of the restrictions of the traditional ship industry on the crew,but also reduces the ship collision accident caused by human factors,improves the safety of shipping and promotes the development of the whole shipping industry,that has great significance.Under this background,the development of unmanned ships has been highly valued by various countries.How to plan a reasonable path for unmanned ships sailing on the water according to effective means,and how to deal with unmanned ships the problem of autonomous obstacle avoidance in emergencies has become the key research direction in the development of unmanned ships.Based on the needs of practical problems,this thesis proposes an autonomous ship avoidance autonomous obstacle avoidance algorithm to deal with the problem of autonomous obstacle avoidance in the surrounding environment when the unmanned ship is sailing.The main research work completed in this thesis is as follows:(1)Studying the data model and data structure of the S57 international standard electronic chart,extracting the original chart data to establish a rasterized navigation environment model,and considering the maneuverability of the unmanned ship and the standard of the IMO route.The method of unmanned ship navigation global route planning is proposed.Through the electronic sea chart display and information system route detection function,the safety detection of the generated route is completed,and a feasible route is planned for the navigation of the unmanned ship.(2)For the problem that the target ship detection position information in the unmanned ship collision avoidance system is not accurate and the information is incomplete,the kalman filter principle and multi-sensor information fusion technology are analyzed in detail,and the distributed kalman fusion structure is adopted.The AIS and the target ship information data acquired by the radar are fused,and the mutual compensation between the different sensor data improves the accuracy and integrity of the unmanned ship's acquisition of the target ship information,and enhances the environmental awareness of the unmanned ship.(3)Combining the actual ship navigation process,the principle and process of unmanned ship collision avoidance based on artificial potential field are analyzed.By improving the repulsion function and gravity function in the potential field,the obstacle identification and ship are added.The self-returning of the ship and the establishment of the PID+MMG model simulate the ship's steering obstacle avoidance process,and construct an unmanned ship autonomous obstacle avoidance algorithm based on the improved artificial potential field.(4)The realization of the unmanned ship autonomous obstacle avoidance system is completed on the embedded ARM platform.