月球车是搭载探测任务的可移动多功能机器人. 月球车在实际地形行驶中, 从起点到目标点之间除了选择最优路径, 还应该将地形、障碍物等影响因素考虑进去. 地形的主要影响因素是陡坡方向和陡坡坡度, 其他因素归类为滑移, 这些在很大程度上增加了路径规划的长度和时间复杂度, 更影响了其安全性. 而传统蚁群算法只是单纯地寻求路径规划中的最优解, 存在收敛速度慢、时间复杂度高、寻优能力不平衡等问题, 且没有考虑滑移、地形等因素, 应用在月球车预测路径规划问题中极易陷入局部最优解. 提出了基于三维栅格地形环境下融合坡度、坡向的滑移预测改进蚁群算法路径规划; 通过设置相同的信息素启发因子和信息素挥发系数, 改变滑移预测地形参数, 得到了基于滑移预测的综合代价函数, 改进了传统蚁群算法; 分析了基于滑移预测的综合代价函数对改进蚁群算法路径长度、收敛速度、时间复杂度和迭代次数的影响. 最后利用实验仿真数据结果验证了本文改进后的蚁群算法在滑移预测路径规划问题中有更高的有效性.

The lunar rover is a multi-function, mobile robot equipped with a mission. Under real terrain driving conditions, in addition to selecting the optimal path from the start point to the target point, the robot should take into account the terrain, obstacles, and other influencing factors. The main influencing factors of the terrain are steep slope gradients and slope orientation; other factors are classified as slip. These greatly increase the length and time complexity of path planning as well as the overall safety of the robot. The traditional ant colony algorithm seeks the optimal solution in path planning, but it also encounters problems such as slow convergence speed, high time complexity, and unbalanced optimization. It does not consider factors such as slip and terrain when applied to lunar rover path prediction. It is easy to fall into a local optimal solution when dealing with path planning problems. This paper proposes an improved ant colony algorithm for path planning based on the slope gradient and slope orientation for 3D raster terrain. By applying a consistent pheromone heuristic factor and pheromone volatilization coefficient, changing the terrain parameters for slip prediction, and obtaining a comprehensive cost function based on slip prediction, the traditional ant colony algorithm is improved. The influence of the comprehensive cost function based on slip prediction on the path length, convergence speed, time complexity, and iteration number of the improved ant colony algorithm is analyzed. Finally, experimental simulation data is used to verify that the improved ant colony algorithm is more effective in addressing slip prediction path planning problems.