Research On Path Planning Of Unmanned Ships Based On ROS

In recent years,with the rapid development of artificial intelligence and other technologies,unmanned ships have become an important research direction for the shipping industry and shipbuilding industry.As one of the key technologies for self-navigation of unmanned ships,path planning has a reliable and effective path planning system is the key to achieving self-navigation of unmanned ships.In response to the problems existing in the self-navigation of unmanned ships,this article proposes a mixed path planning algorithm for the integration of the global optimization algorithm and the local optimization algorithm,and has verified it in the simulation platform based on ROS and Gazebo.The main work content of this article is as follows:(1)In the global path planning,problems such as the low efficiency of the traditional ant colony algorithm and the length of the path length are not effective,and the path length is too long.By introducing a new inspirational function,the problem that is easy to fall into a very small value.At the same time,the pheromone renewal rules and pheromone volatilization factors are improved,and the convergence of traditional ant colony algorithms is optimized,and the convergence is accelerated.The simulation results show that the improvement of the ant colony algorithm has a significant improvement in the optimization of the path length of the traditional ant colony algorithm and the efficiency of path search.(2)In the local path planning,in response to the problem of self-touching of unmanned surface boats in the inner river crowded water area,this article proposes a local path planning method based on improving the dynamic window algorithm.First of all,simplifying the predictive trajectory of obstacles and improving the algorithm avoidance efficiency;then,combined with the rules of the People’s Republic of China to introduce new rules evaluation functions,and establish a non-human boat avoidance action,avoiding angle,resuscitation timing opportunity Model.The simulation results show that the improvement of the dynamic window method is significantly better than the traditional dynamic window method in terms of observing the rules of the People’s Republic of China,while observing the rules of the People’s Republic of China.(3)In order to cope with the navigation requirements of the crowded water area of the river,the global optimization algorithm and the local optimization algorithm fusion were proposed,and the superiority of the hybrid path planning algorithm was verified through simulation.And for the shortcomings of traditional Matlab simulation experiments and poor transplantability,the ROS-based unmanned ship simulation platform is set up in the Gazebo and RVIZ environment to make it more in line with physics.Finally,the overall,local,and hybrid path planning methods proposed in this article are configured in the simulation platform,and their feasibility and reliability are tested in the environment of the inner river squeezing water.