Unmanned vehicle off-road path-planning method with comprehensive constraints on multiple environmental factors
Qingyun Liu,Xiong You,Xin Zhang,Lianrui Xu,Jiwei Zuo,Jia Li
DOI: https://doi.org/10.1080/17538947.2024.2408453
IF: 4.606
2024-10-03
International Journal of Digital Earth
Abstract:In complex off-road environments, different slopes, slope directions, and land types differently affect unmanned vehicle trafficability. Thus, we propose an unmanned vehicle off-road path-planning method with comprehensive constraints on multiple environmental factors. First, the influence of slope, aspect, and surface-coverage type on unmanned vehicle drivability was quantitatively evaluated, and the slope and land type influence layers were formed. Concurrently, a goal guidance layer was formed using the artificial potential field method. Second, the slope influence, land type influence, and goal guidance layers were used to quantitatively evaluate each grid's pass cost value. Finally, the traditional A* algorithm, A* algorithm considering the impassable area, A* algorithm with comprehensive constraints, and Dijkstra algorithm with comprehensive constraints were used for path planning. The results show that the path slopes planned by A* algorithm with comprehensive constraints and Dijkstra algorithm with comprehensive constraints were smaller than those planned by A* algorithm and A* algorithm considering the impassable area; particularly, the length of the uphill path was shorter and the stability was better. The relevant research results provide methods and technical guidance for obtaining the shortest length, shortest time, and most stable and passable path in off-road environments.
geography, physical,remote sensing