Research On Unmanned Surface Vessel Path Planning Algorithm Based On ROS

With the deepening of marine exploitation in the world,unmanned ships are used widely in marine resources exploration,military operations and other fields.At the same time,researchers have put forward higher requirements for the performance of unmanned ships.Based on this background,unmanned ships are being unprecedented concerned about.Path planning technology has become the focus of research on unmanned ship technology,because it is one of the important criteria for judging the level of unmanned ship intelligence,which plans a global optimal path for unmanned ships and avoids obstacle safely in a dynamic environment.Based on the actual demand of unmanned ship navigation and the advantages of ROS(Robotic Operating System)in path planning,this paper proposes an overall scheme of ROS-based unmanned ship path planning.The core is to realize the unmanned ship global path planning based on improved particle swarm optimization algorithm and local path planning for the unmanned ship based on improved genetic algorithm.Meanwhile the ROS map server is used for environment modeling and coordinate transformation to realize the unmanned ship positioning,which realizing the unmanned ship path planning in the electronic chart environment.There are four kinds of contribution in this paper:(1)This academic paper firstly analyzes the present domestic and foreign research situation of unmanned ships as well as the structure and characteristics of ROS navigation framework.Combined with the advantages of ROS navigation framework,the overall scheme and implementation process of ROS-based unmanned ship path planning are designed.(2)Based on the principle analysis of path planning algorithm,in this paper,particle swarm optimization is used to implement global path planning in the overall plan of unmanned ship path planning.An improved strategy of random disturbance and adaptive control parameters is proposed,which greatly improves the global optimization ability of particle swarm optimization.Then the algorithm principle simulation experiment is carried out,which shows that the improved particle swarm optimization algorithm performs well in path optimization and path smoothing.Finally,the implementation process of the improved particle swarm optimization algorithm for unmanned ship global path planning in ROS is elaborated.(3)Based on the global path planning,this paper uses genetic algorithm to realize the local path planning in the overall plan of unmanned ship path planning.The fitness function and genetic operation of the traditional genetic algorithm are improved,the convergence speed of the algorithm is accelerated,and the algorithm principle simulation experiment is carried out.The results show that the improved genetic algorithm achieves the safe obstacle avoidance effect in the dynamic environment.Finally,the realization process of the improved genetic algorithm to realize the local path planning of unmanned ship in ROS is elaborated.(4)Based on QT design,a simulation verification platform was developed.The improved particle swarm optimization algorithm and improved genetic algorithm designed in this paper were combined with the ROS-based unmanned ship path planning scheme to carry out unmanned ship path planning experiments in the electronic chart environment.The experimental results have reached the requirements for unmanned ship path planning and obstacle avoidance,and verified the feasibility of the algorithm.