SLAM in Low-Light Environments: Project Report

📄 arXiv: 2607.17699v1 📥 PDF

作者: Oleh Basystyi, Anna Stasyshyn, Oleksandr Kosovan, Yaroslav Prytula

分类: cs.RO, cs.CV

发布日期: 2026-07-20

备注: Conducted as part of the certification program "Off-Road Visual Navigation: Development and Evaluation of Systems in Challenging Environments'' at the Faculty of Applied Sciences, Ukrainian Catholic University, in collaboration with the UCU UGV Club


💡 一句话要点

提出低光照环境下SLAM系统的性能评估与改进方案

🎯 匹配领域: 支柱三:空间感知与语义 (Perception & Semantics) 支柱六:视频提取与匹配 (Video Extraction)

关键词: SLAM 低光照 特征提取 惯性融合 全局优化 机器人导航 性能评估

📋 核心要点

  1. 现有SLAM方法在低光照环境中面临特征提取和匹配困难,导致定位精度下降。
  2. 本文通过对六种SLAM系统进行基准测试,探索在低光照条件下的有效解决方案。
  3. 实验结果显示,只有结合惯性传感器和全局优化的SLAM系统才能在低光环境中保持稳定跟踪。

📝 摘要(中文)

同时定位与地图构建(SLAM)是机器人领域的基本问题,能够支持自主操作。然而,在低光照环境中,特征提取和匹配受到传感器噪声和运动模糊的影响,现有的RGB相机在黑暗中表现不佳。本文对六种SLAM系统进行了基准测试,结果表明,只有在结合惯性融合和全局优化时,RGB-only SLAM才能保持稳定的低光跟踪。Kimera-VIO系统在所有序列中表现最佳,而ORB-SLAM3和DSO等经典单目管道在困难的低光序列中则失败或偏离。未来的研究可能需要针对低光环境的学习前端或回归到互补传感。

🔬 方法详解

问题定义:本文旨在解决低光照环境下SLAM系统的性能问题,现有方法在特征提取和匹配方面存在显著不足,导致定位精度下降。

核心思路:论文通过对六种SLAM系统进行基准测试,分析其在低光照条件下的表现,提出结合惯性融合与全局优化的策略,以提高RGB-only SLAM的稳定性。

技术框架:整体架构包括特征提取、特征匹配、状态估计和优化四个主要模块,采用不同的SLAM算法进行比较,涵盖特征基础、直接、滤波和学习基础的方法。

关键创新:最重要的创新在于通过系统性比较不同SLAM算法在低光照条件下的表现,发现只有在结合惯性传感器和全局优化时,RGB-only SLAM才能保持稳定跟踪。

关键设计:实验中使用了五个不同难度和光照条件的LaMARia序列,评估了绝对和相对位姿误差以及控制点召回率,特别关注了Kimera-VIO、DPVO和DPV-SLAM等系统的表现。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示,Kimera-VIO系统在所有五个序列中均能完成跟踪,且相对位姿误差最低,而DPVO和DPV-SLAM在低光条件下也能保持跟踪,但绝对误差约为100米。经典的ORB-SLAM3和DSO在大多数困难和低光序列中则失败或偏离。

🎯 应用场景

该研究的潜在应用领域包括自动驾驶、无人机导航和机器人探索等需要在低光照环境中进行自主定位和地图构建的场景。通过改进SLAM系统的性能,可以显著提升这些应用的可靠性和实用性,推动相关技术的发展。

📄 摘要(原文)

Simultaneous localization and mapping (SLAM) is one of the fundamental problems in robotics, as it enables autonomous operations in real-world scenarios. Under low illumination, reduced contrast, sensor noise, and motion blur degrade both feature extraction and feature matching, while compensating with LiDAR, depth, or thermal sensors raises cost, power draw, and integration complexity. Existing benchmarks remain dominated by well-lit indoor or daylight sequences, leaving open how far SLAM with standard RGB cameras can be pushed in the dark. We benchmark six systems spanning the feature-based, direct, filter-based, and learning-based paradigms - ORB-SLAM3, DSO, Kimera-VIO, OpenVINS, DPVO, and DPV-SLAM - on five LaMARia sequences of varying difficulty and illumination, reporting absolute and relative pose error alongside control-point recall. Kimera-VIO is the only system to track all five sequences to completion, combining the lowest relative pose error with steadily growing absolute error due to the absence of loop closure; DPVO and DPV-SLAM never lose tracking but incur absolute errors of roughly 100 m under low light; and the classical monocular pipelines (ORB-SLAM3, DSO) together with the filter-based OpenVINS fail outright or diverge on most of the harder and low-light sequences. The results suggest that RGB-only SLAM maintains stable low-light tracking only when both inertial fusion and global optimization are present. Closing the remaining gap will likely require low-light-specific learned front-ends or a return to complementary sensing.