arXiv 2309.080862023-09-15cs.ROcs.AIcs.CVcs.LG

用于可靠 LiDAR SLAM 的快速且准确的深度闭环与重定位

Fast and Accurate Deep Loop Closing and Relocalization for Reliable LiDAR SLAM

College of Intelligence Science and Technology, National University of Defense Technology（国防科技大学智能科学与技术学院）；National Innovation Institution of Defense Technology（国防科技国家创新研究院）

Chenghao Shi, Xieyuanli Chen, Junhao Xiao, Bin Dai, Huimin Lu

备注：: 20 pages 10 figures 7 tables

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出统一框架 LCR-Net，通过特征提取和位姿感知注意力机制精确估计 LiDAR 扫描相似度与 6-DoF 位姿，实现快速准确的深度闭环与重定位，并集成至 SLAM 系统提升在线可靠性。

摘要：

闭环与重定位是通过解决位姿估计漂移和退化来建立可靠且鲁棒的长期 SLAM 的关键技术。本文首先在统一框架内对闭环与重定位进行了数学表述。然后，我们提出了一种新颖的多头网络 LCR-Net，以有效处理这两项任务。它利用新颖的特征提取和位姿感知注意力机制，精确估计成对 LiDAR 扫描之间的相似度和 6-DoF 位姿。最后，我们将 LCR-Net 集成到 SLAM 系统中，在室外驾驶环境中实现了鲁棒且准确的在线 LiDAR SLAM。我们通过源自闭环与重定位的三个设置对 LCR-Net 进行了全面评估，包括候选检索、闭环点云配准以及使用多个数据集的连续重定位。结果表明，LCR-Net 在所有三项任务中均表现出色，超越了现有最先进方法并展现出卓越的泛化能力。值得注意的是，我们的 LCR-Net 在未使用耗时的鲁棒位姿估计器的情况下优于基线方法，使其适用于在线 SLAM 应用。据我们所知，LCR-Net 的集成产生了首个具备深度闭环与重定位能力的 LiDAR SLAM。我们的方法实现将开源。

英文摘要：

Loop closing and relocalization are crucial techniques to establish reliable and robust long-term SLAM by addressing pose estimation drift and degeneration. This article begins by formulating loop closing and relocalization within a unified framework. Then, we propose a novel multi-head network LCR-Net to tackle both tasks effectively. It exploits novel feature extraction and pose-aware attention mechanism to precisely estimate similarities and 6-DoF poses between pairs of LiDAR scans. In the end, we integrate our LCR-Net into a SLAM system and achieve robust and accurate online LiDAR SLAM in outdoor driving environments. We thoroughly evaluate our LCR-Net through three setups derived from loop closing and relocalization, including candidate retrieval, closed-loop point cloud registration, and continuous relocalization using multiple datasets. The results demonstrate that LCR-Net excels in all three tasks, surpassing the state-of-the-art methods and exhibiting a remarkable generalization ability. Notably, our LCR-Net outperforms baseline methods without using a time-consuming robust pose estimator, rendering it suitable for online SLAM applications. To our best knowledge, the integration of LCR-Net yields the first LiDAR SLAM with the capability of deep loop closing and relocalization. The implementation of our methods will be made open-source.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。