让点云配准的不确定性可解释，还能自动发现错误来源。
Human-Interpretable Uncertainty Explanations for Point Cloud Registration
- 用高斯过程归因法量化并解释配准误差来源。
- 在真实机器人实验中表现优于现有方法，且效率高。
- 适合需要鲁棒感知的机器人与自动驾驶场景。
本文针对点云配准问题，解决传统方法如ICP在传感器噪声、位姿估计误差及遮挡导致的部分重叠等不确定性下的失效问题。提出新型方法GP-CA，不仅能量化注册不确定性，还可通过归因分析将不确定性定位到已知误差源。借助主动学习，该方法能从真实数据中发现新的不确定性来源。我们在三个公开数据集和一次真实机器人实验中验证了GP-CA的有效性。大量消融实验证明了设计合理性。相比其他先进方法，GP-CA在运行时间、样本效率（主动学习）和精度上均表现更优。真实实验清晰展示了其实际应用价值。视频演示表明，GP-CA支持有效的故障恢复行为，显著提升机器人感知鲁棒性。
原文摘要 · Abstract (English)
In this paper, we address the point cloud registration problem, where well-known methods like ICP fail under uncertainty arising from sensor noise, pose-estimation errors, and partial overlap due to occlusion. We develop a novel approach, Gaussian Process Concept Attribution (GP-CA), which not only quantifies registration uncertainty but also explains it by attributing uncertainty to well-known sources of errors in registration problems. Our approach leverages active learning to discover new uncertainty sources in the wild by querying informative instances. We validate GP-CA on three publicly available datasets and in our real-world robot experiment. Extensive ablations substantiate our design choices. Our approach outperforms other state-of-the-art methods in terms of runtime, high sample-efficiency with active learning, and high accuracy. Our real-world experiment clearly demonstrates its applicability. Our video also demonstrates that GP-CA enables effective failure-recovery behaviors, yielding more robust robotic perception.
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。