Abstract: For the problem that most of algorithms of path planning have low success rate and the time is long in trap space, an improved RRT algorithm is proposed. By combining the human and RRT algorithm, the virtual target point is set up through human, which guides the path searching to go out of trap space; the node expansion is optimized, which ensures the path searching is in the feasible region; the fast convergence strategy is designed, and the redundant nodes are deleted. Simulation verifies that the proposed method has a good effect in trap space path planning, which can quickly plan feasible path.

Key words: Unmanned Aerial Vehicle（UAV）, improved Rapidly Random-exploring Trees（RRT） algorithm, trap space, path planning

摘要： 针对大部分航迹规划算法在陷阱空间下，存在规划时间长、成功率低的问题，提出了一种改进RRT算法。通过将人与RRT算法相结合，由人设置虚拟目标点，引导航迹搜索走出陷阱空间；同时对节点扩展进行优化，保证航迹搜索在可行域内；并设置快速收敛策略，删除冗余节点，使航迹搜索速度加快。最后，通过仿真验证表明，该方法在陷阱空间规划中具有良好的效果，可快速规划可行航迹。

关键词: 无人机, 改进快速扩展随机树（RRT）算法, 陷阱空间, 航迹规划