【摘要】 在实际情况下，月面任务的规模和地形随着所选区域或工况而有所不同，因此需要一种更灵活和高效的算法做路径规划任务。为满足月球空间大规模、复杂地形的需求，本文针对月面机器人设计了一种混合尺度蚁群规划方法。该算法将实际的月面图像网格化处理成栅格图，并对其进行路径规划算法建模，由此模拟真实月面任务的路径规划。利用混合尺度的方法优化经典的蚁群规划算法用以适应不同的任务。此外，还采用路径平滑方法减小了路径中转向角度。最后通过几种典型场景验证了优化算法的效率和可行性。更多还原

【Abstract】 In the real-world situation, the lunar missions’ scale and terrain are different according to various operational regions or worksheets, which requests a more flexible and efficient algorithm to generate task paths. A multi-scale ant colony planning method for the lunar robot is designed to meet the requirements of large scale and complex terrain in lunar space. In the algorithm, the actual lunar surface image is meshed into a gird map, the path planning algorithm is modeled on it, and then the actual path is projected to the original lunar surface and mission. The classical ant colony planning algorithm is rewritten utilizing a multi-scale method to address the diverse task problem.Moreover, the path smoothness is also considered to reduce the magnitude of the steering angle. Finally, several typical conditions to verify the efficiency and feasibility of the proposed algorithm are presented.更多还原