对比三种无人机路径规划算法，实测A*最优，PSO适合密集环境。
Comparative Analysis of UAV Path Planning Algorithms for Efficient Navigation in Urban 3D Environments
- 在三维城市环境中测试A*、RRT*和PSO算法性能
- A*计算效率与路径质量均最优，PSO擅长狭窄转弯场景
- RRT*随机搜索平衡性好，适应所有实验条件
无人机在复杂城市3D环境中面临路径规划与避障的核心挑战。近年来发展出多种路径规划算法，虽已解决部分问题，但仍存在诸多局限。本文在障碍物密集的城市三维环境中，对A*、RRT*和粒子群优化（PSO）三种主流算法进行系统评估。设计三组实验，每组包含两个场景，分别考察不同地图规模、飞行高度及障碍物密度与尺寸的影响。结果表明：A*算法在计算效率与路径质量上表现最佳；PSO特别适用于弯道密集与障碍物繁多的环境；RRT*凭借其随机搜索策略，在所有实验中均展现出良好平衡性与鲁棒性。
原文摘要 · Abstract (English)
The most crucial challenges for UAVs are planning paths and avoiding obstacles in their way. In recent years, a wide variety of path-planning algorithms have been developed. These algorithms have successfully solved path-planning problems; however, they suffer from multiple challenges and limitations. To test the effectiveness and efficiency of three widely used algorithms, namely A*, RRT*, and Particle Swarm Optimization (PSO), this paper conducts extensive experiments in 3D urban city environments cluttered with obstacles. Three experiments were designed with two scenarios each to test the aforementioned algorithms. These experiments consider different city map sizes, different altitudes, and varying obstacle densities and sizes in the environment. According to the experimental results, the A* algorithm outperforms the others in both computation efficiency and path quality. PSO is especially suitable for tight turns and dense environments, and RRT* offers a balance and works well across all experiments due to its randomized approach to finding solutions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。