Coverage path planning considering the cell number and starting position
The coverage tasks of mobile robots are evolving towards large-scale and intelligent di-rections,demanding urgent requirements for the coverage efficiency and environmental adaptabil-ity of coverage path planning.To address the inadequate adaptability of the traditional boustro-phedon cellular decomposition in complex maps and to improve coverage efficiency,a complete coverage path planning method is proposed.Firstly,based on the boustrophedon cellular decom-position method,a strategy of area-descending traversal and monotone polygon judgment was proposed to merge cells,reducing approximately half of the cell quantity.Finally,by establis-hing the mapping relationship between the starting and ending positions of cells,a genetic algo-rithm was employed to optimize the selection of cells'starting positions and the global access or-der.The research results show that:1)when processing maps with dimensions of 1 300 pixels,the algorithm in this paper can obtain calculation results within 10 s,and compared with the boustrophedon method,neural network method,and contour line method,the growth rate of calculation time is smaller with the increase of map area;2)compared with the boustrophedon method,contour line method,neural network method,and energy-optimal method,the total op-eration time of robots in this paper is reduced by 5.4％to 47.0％,and the ineffective operation time is reduced by 5.8％to 29.2％;3)the average coverage rate of this algorithm on 1 800 test maps reaches 99.91％;4)the tests further confirm the significant coverage efficiency advantages of the algorithm proposed in this paper.
coverage path planningcellular decomposition methodgenetic algorithmmonoton-ic polygonstarting position optimization
万方数据
维普
