摘要
针对无人机三维航迹规划的实时性问题,提出了基于快速扩展随机树的三维航迹规划方法。该算法能够根据当前环境快速有效搜索规划空间,通过随机采样点将搜索导向空白区域,使三维航迹规划能够用于实时航迹规划。通过引入航迹距离约束,搜索树将沿着路径距离最短的近似最优航迹的方向进行扩展,克服了基本快速扩展随机树方法随机性强,只能快速获得可行航迹,无法获得较优航迹的缺点。在搜索过程中无人机的航迹约束条件和地形信息得到了充分利用,使算法生成的航迹能够自动回避地形和威胁,同时满足无人机的动力学约束。通过生成的虚拟数字地图对算法进行了仿真验证,仿真结果表明该方法能够快速有效地规划出满意的无人机三维航迹。
To satisfy the real-time requirement of path planning in three dimensions for unmanned aerial vehicle, a path planning algorithm based on rapidly-exploring random tree is proposed. By random sampling point in configura-tion space, the search will be guided to empty area, thus the algorithm can search the high-dimension space quickly and efficiently according to the current environment, which can be used in real-time path planner. By introducing the path length constraint, the search tree will explore along the direction of the near optimal path. The proposed al-gorithm overcomes the disadvantage of basic RRT algorithm that only to quickly get feasible path, unable to obtain near optimal path. During the search process, the path constraints of UAV and the terrain information are fully uti-lized, so that the path generated by the algorithm can avoid terrain and threat automatically, and meet the dynamic constraints of UAV. Simulations for the algorithm are made on a generated virtual digital map. Simulation results demonstrated that this proposed method can complete path planning mission in three dimensions quickly and effec-tively.
出处
《西北工业大学学报》
EI
CAS
CSCD
北大核心
2016年第4期564-570,共7页
Journal of Northwestern Polytechnical University
基金
航空科学基金(20135184007)资助
关键词
无人机
快速扩展随机树
实时性
地形回避
三维航迹
unmanned aerial vehicle （UAV）
rapidly-exploring random tree
real-time
terrain avoidance
3D-path
作者简介
尹高扬（1987-），海军航空工程学院博士研究生，主要从事导航、制导与控制的研究。
