Multi-Track Path Planning of Outdoor Scanning Robot in Unknown Scene

Abstract

As the digital age continues to evolve, automated scanning of outdoor environments ushers in challenges. There are currently two main problems with outdoor mobile robot scanning: firstly, the scanning path is too long; secondly, it is easy to collide with obstacles, and this thesis will focus on the above two problems. Firstly, a topological structure graph is constructed for the outdoor scene, and the set of accessible points is obtained. The high exploration value region is first explored by the greedy algorithm, and then the ant colony algorithm is used for path planning of the general exploration value region. Secondly, we make algorithmic improvements to the ant colony algorithm by adopting a multi-trajectory path planning algorithm that allows decision makers to obtain multiple solutions, proposing a negative feedback ant colony algorithm, and introducing guidance pheromones and alert pheromones to enable the ant colony algorithm to explore with a comprehensive consideration of the effects brought about by the environment. Finally, for the algorithm proposed in this thesis, we conducted simulation experiments using point cloud map and raster method respectively, our proposed algorithm has better performance on the map, and the improved ant colony algorithm can improve the scanning range by about 4% than the original algorithm when moving the same distance.