Research On Multi Vessel Dynamic Obstacle Avoidance Method In Narrow Water

In recent years,with the development of intelligent shipping,intelligent ship has been widely concerned in the shipping industry,mainly in the military and civil fields.Automatic collision avoidance technology is one of the key technologies of intelligent ship,which can not only measure its intelligent level,but also the premise of its efficient and safe completion of various navigation tasks.At present,there are still many defects in automatic collision avoidance technology,especially the problem of real-time avoidance of multiple ships in narrow waters.This thesis focuses on the research of multi vessel dynamic obstacle avoidance method for intelligent ships in narrow waters,analyzed the dynamic obstacle avoidance requirements of intelligent ships in narrow waters,proposed the Adaptive Speed-weight Dynamic Window Algorithm for ships,built the Automatic Collision Avoidance System for intelligent ship,and carried out real ship experiments to verify it.The main work of this thesis includes:(1)The collision avoidance characteristics constraints of intelligent ships in narrow water environment are studied,and the relative position parameters of intelligent ships and target ships are derived.The collision risk degree of intelligent ships is calculated based on the weighted method.By analyzing the situation of collision avoidance in narrow water,comparing the advantages and disadvantages of each obstacle avoidance algorithm,the Dynamic Window Algorithm is chosen as the core algorithm of collision avoidance decision.(2)Combining the environmental characteristics of narrow water and the dynamic characteristics of intelligent ships,the two-dimensional motion model of ships is constructed.At the same time,according to the actual navigation requirements of the ship,speed objective function and heading objective function in the evaluation function are derived and calculated,and the adaptive speed weight is introduced,and the Dynamic Window Algorithm based on the adaptive speed weight is proposed,The problem of collision avoidance failure due to fixed speed weight and too close to obstacles in the process of collision avoidance is solved.The simulation results show that the proposed algorithm can change the speed according to the distance from the obstacles,and has better trafficability and security.At the same time,the simulation ofmulti ship dynamic obstacle avoidance in complex environment verifies the effectiveness of the improved algorithm.(3)Building the Intelligent Ship Automatic Collision Avoidance System.Firstly,according to the requirements of intelligent ship in narrow waters,the whole scheme of automatic collision avoidance system is designed.Then,coordinate transformation and multithreading are used to realize the fusion of multi-sensor in time and space,and the data after fusion is displayed and transmitted by Rviz visualization tool.A ship positioning sensing system based on ROS(Robot Operating System)is built.Finally,the man-machine interface control platform is developed to realize the digital display of sensor data and control parameters,then the ship control system is built.(4)Ship control characteristics test and collision avoidance test in narrow waters.According to the content of the previous study and the built system,the ship control characteristics test,single ship and multi ship collision avoidance test are carried out.The results show that the dynamic window method based on adaptive speed weight can effectively avoid dynamic obstacles,and the avoidance trajectory is smoother,which verifies the reliability of the experimental platform and the effectiveness of the improved collision avoidance algorithm proposed in this thesis,and provides a solution for multi ship dynamic obstacle avoidance of intelligent ships in narrow waters.