Abstract: Abstract: To address the problems of low exploration efficiency, key information exchange under limited communication bandwidth in the current Multiple Unmanned Aerial Vehicle (Multi-UAV) system when exploring large-scale complex environments, a fast and fully autonomous exploration method for Multi-UAV system in large-scale complex environments was proposed, which included a fast and hierarchical exploration strategy and a lightweight large-scale environment modeling method. First, closed viewpoints were generated in the front-end trajectory planning to drive the UAV to explore unknown environments. Then, the smooth, continuous, and time-optimal trajectory optimization was transformed into a convex optimization problem in the back-end. Besides that, for large-scale environments, a random mapping method was used for lightweight mapping and map data exchange. In simulation, the proposed method was compared with three existing approaches: FUEL(Fast Unmanned aerial vehicle ExpLoration), FBE(Frontier-Based Exploration), NBVP(Next Best View Planner). The results showed that the performance of the proposed method in terms of exploration time is improved by 14.4%, 43.9% and 47.7%, respectively. The lightweight mapping method used 28.3% and 22.4% less data than the Bayesian method and the polyhedron method. Combined with the results of the exploration experiment in terms of speed and data volume, the proposed method can perform fast and fully autonomous exploration in large-scale complex environments.

Key words: Keywords: Multi-UAV System, Autonomous Exploration, Large-Scale Environment, Trajectory Optimization, Exploration Strategy

摘要： 摘 要: 针对当前多无人机系统在探索大范围复杂环境时存在探索效率低下和在通信带宽受限下的信息交换问题，提出了一种适用于在大范围复杂环境下无人机的快速全自主探索方法，包括一种快速、分层探索策略和轻量级大规模环境建模方法。首先，在前端轨迹规划部分生成闭式视点以驱动无人机进行未知探索。其次，在后端将平滑、连续和时间最优的轨迹优化转化为一个凸优化问题进行了系统建模。同时在环境表征方面，使用随机映射的方法进行轻量化建图和地图数据交互。在仿真中，与使用增量边界信息和分层规划的快速探索方法FUEL（Fast Unmanned aerial vehicle ExpLoration）、基于边界的快速探索方法FBE（Frontier-Based Exploration）以及基于下一个最佳视点的探索方法NBVP（Next Best View Planner）进行了对比实验。结果表明，提出的方法在探索时间方面的性能分别提高了14.4% 、43.9%、47.7%。轻量化建图方法在数据量上比贝叶斯（Bayesian）方法和多面体（Polyhedron）方法少28.3%和22.4%。结合探索实验在速度和数据量方面的结果，提出的方法可以高效地在大规模复杂环境下进行快速全自主探索。

关键词: 关键词: 多无人机系统, 自主探索, 大范围环境, 轨迹优化, 探索策略