Research On Multi-task Based Autonomous Path Planning Method For Unmanned Boats

As a new type of small intelligent body with strong autonomy and small size,the unmanned boat can accomplish various tasks independently,and now it has become an effective tool to realize efficient and convenient water monitoring and safety inspection.Path planning is one of the key technologies in the field of unmanned boat motion control,which can ensure the smooth execution of multiple tasks by unmanned boats.This paper focuses on the single-boat obstacle avoidance task and multi-boat roundup task,and focuses on the path planning problems under different tasks such as single-boat global obstacle avoidance,single-boat local obstacle avoidance and multi-boat roundup,improves the relevant control algorithms,and verifies the effectiveness of the proposed algorithms based on the simulation experiment platform.The main work of the paper is as follows:(1)In order to study the unmanned boat multi-mission path planning problem,the group built an unmanned boat multi-mission path planning simulation platform that can simulate the real lake water environment,described the development architecture and tools of the platform,the platform simulation of unmanned boat obstacle avoidance and siege path planning algorithm research required to obtain the water environment,unmanned boat parameter settings,task settings display and algorithm implementation and other functional design,analyzed the system module The system modules are analyzed,and the platform interface visualization design and operation methods are given.(2)To solve the single-boat global obstacle avoidance problem,a cosine pinch angle cost minimization(Mcosh(m))heuristic function and redundant node improved A~* algorithm are proposed to verify the algorithm accuracy performance and optimization performance.The simulation results show that the improved algorithm has a faster search speed,an 8%improvement in algorithm time performance,and a 40% reduction in path redundancy points compared with the algorithm before the improvement,and finally the improved algorithm is validated in the simulation platform.(3)In order to solve the single-boat local obstacle avoidance problem,the modified repulsive field function is proposed for the unreachable target problem of the artificial potential field algorithm;for the problem that the algorithm is easy to fall into local minima,the simulated annealing algorithm is proposed to be integrated with the artificial potential field method to avoid the algorithm falling into local minima by adding random target points.The simulation results show that the improved algorithm can effectively solve the problems of unreachable targets and local minima.(4)To solve the multi-boat cooperative encircling problem,we firstly model the multiboat cooperative encircling and propose the Cooperative dynamic encircling formation(CDEF)for the problem that the static formation is prone to the escape of the enemy ship;secondly,we estimate the distance and speed of the enemy ship by using the speed and orientation information of the unmanned ship itself,and design the motion target encircling control law with the anti-collision mechanism.Secondly,the distance and speed of the enemy ship are estimated by using the speed and orientation information of the unmanned boat itself,and the motion target encircling control law of the anti-collision mechanism is designed.The simulation results show that compared with the static formation,the CDEF formation improves the success rate of three unmanned boats by 36% and four unmanned boats by 25%,and finally the improved strategy is validated in the simulation platform.