Path planning of coal mine rescue robot based on blocked grid map
SHAO Xiaoqiang
LIU Mingqian
MA Bo
LI Hao
LYU Zhiyue
HAN Zehui
Abstract：To address the issues of coal mine rescue robots' path planning algorithms taking too long,producing too many redundant points,and easily falling into deadlock in complex underground environments,a grid map blocking algorithm based on type matching is proposed.This algorithm iteratively blocks passable nodes that do not require exploration in the grid map.The blocking process matches the grid map with a subgraph type composed of defined grid nodes and their neighbors.Firstly,blockable and non-blockable grid types are defined based on the pathfinding characteristics of the path planning algorithm.Then,the model is built according to these types,assign-ing weight and bias to each type.Finally,each type of subgraph is matched with the initial grid map through a two-dimensional convolu-tion operation to block nodes that do not require expansion.Blocking the input raster map before using the path planning algorithm.Block-ing nodes does not disconnect the minimum cost path in the original raster map.Experimental results show that this algorithm can be ap-plied to various grid environment maps.In real underground coal mine grid maps,compared to using the path planning algorithm alone,the proposed algorithm combined with the A* algorithm reduces total path planning time by 60.0％and the number of expansion nodes by 60.4％.When combined with the ant colony algorithm,total time is reduced by 55.8％and the number of iterations by 53.7％.The pro-posed algorithm significantly reduces path planning time,solves the deadlock issue,has clear advantages in complex environment maps,and ensures timely accident rescue.
Keywords：coal mine rescue robotgrid mapblocking grid mapA* algorithmant colony optimization
Publication Date：2025-07-31
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：13( 249-261 )