Research On Safe Path Planning Of Surface Unmanned Vessel Based On Improved Ant Colony Algorithm And Dynamic Window Approach

The 21 st century is the century of the ocean,and the status of the ocean in national economic development and geopolitics has increased significantly.Surface unmanned vessels(USV),as a class of important marine operation platforms,can work under various harsh conditions all day long,greatly reducing labor costs as well as potential risks,and are receiving more and more extensive attention from countries and industries around the world.The actual marine environment is extremely complex,and most of the hidden and dynamic unknown obstacles cannot be detected and known in advance,which may produce uncertain dynamic threats to the safety of USV navigation missions,and if the USV is not capable of taking timely security measures to deal with these unexpected situations,it will lead to serious accidents,affecting the quality and execution efficiency of the USV missions,and even directly leading to mission failure.Therefore,in the information and decision layer of the USV,reliable and safe path planning is a prerequisite for achieving autonomous navigation,and path planning specifically refers to the object to find a path to avoid obstacles to reach the target point from the starting point under certain constraints,and this path is required to achieve some performance index optimal,and path planning is used in many fields such as traffic planning,workshop operations,robotics,UAV,etc.Path planning is widely used in many fields such as traffic planning,workshop operations,robotics,and UAV,so how to implement an efficient path planning method is a hot research direction.In this paper,we analyze the path planning tasks of surface unmanned vessels under different working conditions,and propose a hierarchical global path planning method and a local path planning method,which are finally integrated.This paper firstly analyzes the research background and significance of the topic,introduces the current situation of USV at home and abroad,and then describes the research status of ant colony algorithm and existing path planning algorithms and spatial modeling methods,and after combining the relevant knowledge of graph theory,the spatial model is established by raster method.At present,most of the path planning algorithms based on intelligent optimization algorithms have problems such as slow convergence speed and poor path security.As a heuristic optimization algorithm for solving combinatorial optimization problems,the ant colony algorithm is widely used in many combinatorial optimization problems such as traveler’s problem,vehicle pathfinding problem and path planning because of its positive feedback property,good parallelism and strong robustness.In this paper,we propose a global path planning method based on the improved ant colony algorithm,which accelerates the convergence of the algorithm by initializing the pheromone unequal allocation before the start of the algorithm,adjusts the search strategy of the subsequent ants after the deadlock of the predecessor ants,and improves the pheromone update method.The simulation experiments of the improved algorithm in different raster environments have verified that the improved algorithm has better performance in terms of optimal path length,deadlock number,convergence iteration number and so on.After that,in order to accomplish the local path planning task of surface unmanned vessels,this paper proposes a local path planning algorithm based on the improved dynamic window approach.This algorithm designs a new comprehensive evaluation function,which solves the problems that may be encountered in the practical application of the traditional dynamic window approach and improves the collision avoidance performance of the dynamic window approach when facing unknown obstacles.Simulation experiments are conducted in different environments,and the experimental results show that the improved algorithm proposed in this paper can complete local collision avoidance more safely and reliably compared with the traditional dynamic window approach.Finally,based on the two planning methods proposed above,this paper proposes a hierarchical fusion path planning algorithm for USV.This fusion algorithm can use the global known information to get the global optimal path while coping well with the dynamic obstacles in the position to avoid collisions,which can better accomplish the path planning task in the actual situation with superior performance,improve the safety of USV during the mission and guarantee the efficiency of the subsequent execution.