Research On Ship Path Planning Based On The ACO-DWA Fusion Algorithm

Under the increasing popularization rate of intelligent shipping and various environmental protection and green development policies,countries have put forward future development plans of intelligent shipping.The realization of intelligent shipping requires a complete intelligent system,including the integration of information and communication technology,artificial intelligence,big data technology,Internet of Things,cloud computing and other technologies.Although some technologies have made breakthroughs in some fields,there are still huge challenges.The study of ship’s path planning is one of the key technologies to realize ship’s intelligent shipping.Aiming at the study of ship’s path planning,this paper proposes a method that combines the improved Double Ant Colony Optimization algorithm(DACO,Double Ant Colony Optimization)and the improved dynamic window method(DWA,Dynamic Window Approach)Combined with the ship’s propulsion characteristics,an optimal safety path is planned for the ship that conforms to the ship’s kinematics constraints and propulsion characteristics.The specific research content of this paper is as follows.(1)Design ship path planning method based on ACO-DWA fusion algorithm,namely FA algorithm.Firstly,in view of the traditional ant colony algorithm’s slow iterative convergence and falling into local optimal solutions,this paper classified ant colonies into exploration ant and optimization ant,and improved pheromone updating strategy and probability transfer rules.Secondly,by adding global oriented evaluation factor and improving target orientation evaluation factor into the evaluation function of DWA algorithm,some shortcomings can be improved.Then,double ant colony algorithm is used for global path planning,and the transfer probability of multi-factor comprehensive heuristic factors is optimized to obtain the global optimal path.On this basis,DWA local path planning is carried out,and the global path planning combined with ant colony algorithm and ship propulsion characteristic factors are introduced for comprehensive evaluation of the trajectory,making the trajectory better and in line with ship propulsion characteristics.(2)Build simulation environment model,design ship kinematics model and maneuvering dynamics model,and then carry out path planning simulation experiment in static environment and dynamic obstacle environment respectively.Compare the simulation results of improved DWA algorithm,mixed path planning algorithm and FA algorithm in this paper.The generated trajectory graph,yaw Angle time graph,yaw speed time graph and ship speed time graph of the three are analyzed.The advantages and reliability of FA algorithm in this paper are verified through the above comparative analysis,which provides a theoretical basis for the subsequent ROS simulation and ship test.(3)In ROS system,the URDF model of the algorithm simulator and the Marine simulation environment model were built,the motion model of the algorithm simulator was established,the ROS navigation architecture was optimized,and the improved fusion algorithm in this paper was implanted into the navigation package to conduct ship simulation experiments,so as to further verify the feasibility of the fusion path planning algorithm in this paper.Then,the hardware and software and system structure of the model ship based on ROS algorithm simulator were designed,and the navigation test of the model ship’s path planning was carried out in the real water environment.The analysis of the test results showed that the FA algorithm in this paper had certain feasibility and effectiveness.