针对路径规划中的狭窄通道，提出一种智能采样方法提升效率
A Novel Narrow Region Detector for Sampling-Based Planners' Efficiency: Match Based Passage Identifier
- 基于占用栅格图识别狭窄区域，动态增加该区域采样密度
- 在仿真与真实环境中均显著缩短规划时间，减少里程碑数量
- 适合需要高效路径规划的机器人、无人机等自主系统使用
自主技术已广泛应用于移动机器人、机械臂和无人机等多种场景，其核心任务之一是路径规划。现有方法主要分为概率型与确定型两类，其中概率型方法在狭窄通道环境中普遍存在性能下降问题。本文提出一种新型采样器，通过占用栅格图确定性识别狭窄通道区域，并在这些区域增加采样密度。算法代码已开源。在三类测试环境（特定仿真、随机仿真与真实环境）中进行基准评估，结果表明，相比基线采样器，本方法在规划时间与里程碑数量上均有显著优化，验证了其有效性与实用性。
原文摘要 · Abstract (English)
Autonomous technology, which has become widespread today, appears in many different configurations such as mobile robots, manipulators, and drones. One of the most important tasks of these vehicles during autonomous operations is path planning. In the literature, path planners are generally divided into two categories: probabilistic and deterministic methods. In the analysis of probabilistic methods, the common problem of almost all methods is observed in narrow passage environments. In this paper, a novel sampler is proposed that deterministically identifies narrow passage environments using occupancy grid maps and accordingly increases the amount of sampling in these regions. The codes of the algorithm is provided as open source. To evaluate the performance of the algorithm, benchmark studies are conducted in three distinct categories: specific and random simulation environments, and a real-world environment. As a result, it is observed that our algorithm provides higher performance in planning time and number of milestones compared to the baseline samplers.
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。