SA-LIVO: Efficient LiDAR-Inertial-Visual Odometry with Subspace-Aware Degeneracy Handling
作者: Yinong Cao, Xin He, Yuwei Chen, Shijie Liu, Chunlai Li, Jianyu Wang
分类: cs.RO
发布日期: 2026-06-24
备注: 20 pages, 12 figures, 5 tables
💡 一句话要点
提出SA-LIVO以解决LiDAR-视觉惯性里程计的退化问题
🎯 匹配领域: 支柱三:空间感知与语义 (Perception & Semantics)
关键词: LiDAR 视觉惯性里程计 信息融合 鲁棒性 特征分解 自动驾驶 机器人导航
📋 核心要点
- 现有的LiDAR-视觉-惯性里程计在传感器失效时表现不佳,尤其是在几何约束不足或光照不良的情况下。
- SA-LIVO通过方向选择性融合和信息高效处理,针对退化方向进行优化,提升了系统的鲁棒性和准确性。
- 在29个序列的实验中,SA-LIVO在准确性和内存使用上均优于现有方法,显示出显著的性能提升。
📝 摘要(中文)
紧耦合的LiDAR-视觉-惯性里程计(LIVO)融合了精确的几何深度与互补的视觉测量,但其外部传感器面临独立的失效模式。现有的对策在模态层面上操作,未能解决联合信息矩阵的方向依赖结构。本文提出SA-LIVO,通过方向选择性融合和信息高效处理,克服了这些限制。SAIF框架对联合信息矩阵进行特征分解,并在每个特征方向应用线性夹紧软门,减弱退化方向的影响。实验结果表明,SA-LIVO在多个基准测试中表现出与最强基线相当的准确性,并在竞争系统发散的情况下保持漂移受限。
🔬 方法详解
问题定义:本论文旨在解决LiDAR-视觉-惯性里程计在传感器退化时的性能下降问题。现有方法在处理不同模态的退化时未能有效利用联合信息矩阵的方向特性,导致视觉信息在不必要的方向上分散。
核心思路:SA-LIVO的核心思想是通过方向选择性融合来优化信息处理,确保在LiDAR信息不足的情况下,视觉信息能够有效补充。通过特征分解和线性夹紧软门的应用,增强了系统在关键方向的表现。
技术框架:SA-LIVO采用Subspace-Aware Information Fusion(SAIF)框架,首先对联合信息矩阵进行特征分解,然后在每个特征方向上应用软门,最后在一个扩展卡尔曼滤波(InEKF)循环中联合优化LiDAR和视觉残差。
关键创新:SA-LIVO的主要创新在于引入了方向选择性融合机制,通过对信息矩阵的特征分解,能够有效抑制退化方向的影响,确保在关键方向上保留完整的观测信息。
关键设计:在实现中,视觉信息的光度雅可比矩阵在每次迭代前组装一次,并在后续迭代中重复使用,从而避免了传统迭代滤波器每次迭代的计算开销。
🖼️ 关键图片
📊 实验亮点
SA-LIVO在29个序列的实验中表现出与最强基线相当的准确性,且在竞争系统发散的情况下保持漂移受限。其在笔记本CPU上平均每帧处理时间为12.3毫秒,而在嵌入式ARM板上为26.8毫秒,峰值内存使用量降低了3.6-6.3倍。
🎯 应用场景
SA-LIVO的研究成果在自动驾驶、机器人导航和增强现实等领域具有广泛的应用潜力。通过提高LiDAR-视觉-惯性里程计的鲁棒性和准确性,该技术能够在复杂环境中提供更可靠的定位和地图构建能力,推动相关技术的进步与应用。
📄 摘要(原文)
Tightly coupled LiDAR-visual-inertial odometry (LIVO) fuses precise geometric depth with complementary visual measurements, yet its exteroceptive sensors face independent failure modes: LiDAR degenerates when scan geometry is under-constrained, while visual measurements degrade under adverse illumination or texture absence. Existing countermeasures, including binary degeneracy detection, covariance inflation, and scene-level quality gating, operate at the modality level and leave the direction-dependent structure of the joint information matrix unaddressed. Consequently, visual residuals enter pose directions where LiDAR is well-constrained, while in deficient directions visual compensation disperses across the full state space rather than concentrating where needed. We propose SA-LIVO, a LiDAR-inertial-visual odometry system addressing these limitations through direction-selective fusion and information-efficient processing. The Subspace-Aware Information Fusion (SAIF) framework eigendecomposes the joint LiDAR-visual information matrix and applies a linear-clamp soft gate per eigendirection, attenuating degenerate directions while preserving observable ones at full strength. LiDAR and visual residuals are then jointly optimized in one InEKF loop at a shared linearization point. Since visual information contributes only where LiDAR is deficient, photometric Jacobians are assembled once before the loop and reused across iterations, avoiding the per-iteration cost of conventional iterated filters. Experiments on 29 sequences from three benchmarks (HILTI'22, New College, Oxford Spires) and concurrent-degradation scenarios show accuracy competitive with the strongest baselines and bounded drift where competing systems diverge. SA-LIVO averages 12.3 ms per frame on a laptop CPU and 26.8 ms on an embedded ARM board without GPU, with 3.6-6.3x lower peak memory. The code will be open-sourced.