基于蚁群算法的四旋翼航迹规划

由于四旋翼无人机(UAV)自身的特点和其复杂的飞行环境,考虑到全球定位系统(GPS)定位的精度,在环境模型方面,建立了一个基于高程图的三维环境模型,减小了碰到障碍物的概率。在规划算法方面,大部分现有的路径规划算法只能规划二维平面路径,而一般的三维规划算法,大多数运算算法复杂,需要很大的存储空间,同时难以进行全局路径规划。该蚁群算法具有分布式计算、群体智能等优势,在路径规划上有很大潜力。但在应用基本三维蚁群算法进行航迹搜索时,两平面直接相连容易使航迹直接穿过障碍物,并且搜索出的航迹节点较多,适应度值过大。针对这两个问题对算法做出了改进,提出了变主方向搜索策略和简化航迹策略。仿真实验证明改进后的蚁群算法能够很好地避开障碍物,减小了路径长度,提高了搜索效率。
Given a four-rotor unmanned aerial vehicle’s characteristics and complex flight environment, as well as the accuracy of the global positioning system in the environment model, the establishment of a 3D environment model based on elevation maps has reduced the probability of encountering obstacles. In terms of planning algorithms, most of the existing path planning algorithms can only plan 2D paths. Numerous 3D planning algorithms have complex computations and require much storage space. A global path is also difficult to plan. The advantages of the ant colony algorithm include distributed computing and swarm intelligence. Moreover, this algorithm has great potential in path planning. However, when the fundamental ant colony algorithm is used in a 3D track search, the two directly connected planes easily track straight through obstacles. The track then includes more nodes, and the fitness value becomes too large. The algorithm was improved to address these issues by proposing the strategy of converting the main direction to search and the simplified track strategy. Ant simulation results showed that the improved algorithm could avoid obstacles, reduce path length, and improve search efficiency