Research On Complete Coverage Path Planning For Unmanned Surface Vessel

As an important research direction of robot path planning,complete coverage path planning(CCPP)has been widely used in daily life and industrial automation field.With the increasing application scenarios,the solution method of complete coverage path planning need to update urgently.At present,the main problem of CCPP is the global optimization ability of the mainstream solution method,while some intelligent clustering algorithms have a large amount of calculation and slow convergence speed.According to the above problems,by the method of optimization of grid environment modeling,this paper solve CCPP based on improved genetic algorithm.Considering the influences of water we made a local optimization to ensure that the actual operation of Unmanned Surface Vessel(USV)can navigate according to the optimal path.The details are as follows:(1)The raster environment map was established by using the boundary coordinates of the original area,which explicitly expressed information such as obstacles,boundaries,raster state and raster relationship,etc.,providing a concise and efficient coding basis for the following research on path planning,and a new idea for the establishment of environmental model;(2)With genetic algorithm(GA)as the basic algorithm framework,and on the analysis of the advantages and disadvantages of GA for its existence and fixed before the evolution of crossover and mutation probability problem such as the loss of elite individual late,improvement are as follows: the first is to constraint of chromosome coding way,provisions of adjacent genes corresponds to the grid must be continuous adjacent grids in the chromosomes,to reduce the number of invalid solution,improve operation efficiency;The second improvement is to adopt the adaptive fitness function to avoid that large fitness value in the early evolutionary stage caused by the fixed fitness function.The third improvement is to adopt the adaptive crossover and mutation probability to avoid the situation that the small crossover and mutation probability in the initial stage and the large crossover and mutation probability in the later stage.The fourth improvement is to add the explosion operator in the fireworks algorithm and conduct an explosion search near the optimal solution,comparing with the optimal solution of the original population to avoid falling into local convergence.(3)This paper analysis the influence of the hydrostatic water and water of the navigation path,and proposes a simple optimization model based on decomposition of speed,through the model output USV’s velocity and direction to ensure USV navigate in a certain velocity of water flow under the influence of the shortest path.This paper mainly studies the environment map modeling,the solution of the CCPP problem based on the improved genetic algorithm and the local path optimization considering the influence of water flow.An improved genetic algorithm for solving the full-area coverage path planning is proposed,and the superiority of the algorithm compared with other algorithms is verified by simulation experiments.