Research On Collision Avoidance Method Of USV In Dynamic Obstacle Environment

Nowadays,with the rapid development of population and industry,land resources are becoming increasingly scarce.People pay more attention to the development and utilization of marine resources.As a new type of intelligent equipment at sea,unmanned surface vessel(USV)has become a research hotspot of scholars at home and abroad.USVs in both military and civilian areas have broad prospects for development.USVs can avoid obstacles independently and effectively,which is the precondition of its successful mission.In this paper,the collision avoidance method of UAV in dynamic obstacle environment is studied in depth.The main contents of this paper are as follows:Firstly,the related theories of USV collision avoidance are studied,and the specific collision avoidance process is analyzed.Combining with maritime rules,the constraints in the process of USV collision avoidance and the collision avoidance behavior in different encounters are discussed.Combining the space and time collision hazards,the collision hazards model of USV is established to judge whether the collision hazards will occur with the obstacles in real time,and the correctness of the collision hazards model is verified by the results of an example.The three-degree-of-freedom motion model of USV is established,and its maneuverability is tested by direct navigation and rotary test,which lays the foundation for subsequent design of collision avoidance control law of USV.Secondly,a hierarchical collision avoidance strategy is designed for the known static obstacles and unknown dynamic obstacles in the USV navigation area.An improved PSO algorithm with adaptive dynamic weights is used to realize the global planning of collision avoidance path for the shortest path in the environment of static known obstacles.In the environment with dynamic obstacles,the improved artificial potential field method is used to solve the possible local extremum and GNRON phenomena through the construction of gravity function and repulsion function,and realize the local dynamic collision avoidance.The collision avoidance strategy of hierarchical structure combines global planning collision avoidance with local dynamic collision avoidance coordination,and obtains a collision avoidance safety path in complex environment.The simulation of collision avoidance in different encounter situations and complex environments proves that the hierarchical collision avoidance strategy has good collision avoidance effect in various environments.Finally,aiming at the insufficiency that the improved artificial potential field method used in dynamic obstacle environment does not consider the dynamic constraints of UAV,a backstepping collision avoidance control algorithm based on the improved dynamic window method is designed.Based on the guidance-control framework,the improved dynamic window method is used to obtain the safe path in the dynamic obstacle environment.Then the backstepping method is used to design the collision avoidance control law for the dynamic characteristics of USV.Finally,the avoidance of dynamic obstacles in complex environment is realized.The improved dynamic window method integrates the set of velocity obstacles into the velocity vector space of the dynamic window method,which improves the efficiency of the dynamic window method in evaluating and optimizing the prediction trajectory.Aiming at the under-driving characteristics of USV,the Lyapunov function is designed to design the backstepping-turning controller and the backsteping-speed controller respectively to realize the real-time tracking of the collision avoidance path.The simulation results verify that the improved dynamic window planning algorithm and anti-step collision avoidance control law can realize automatic avoidance of dynamic obstacles in complex environments.