Towards Explaining Uncertainty Estimates in Point Cloud Registration

2024-12-29

AI Summary

Probabilistic ICP suffers from poor interpretability of its uncertainty estimates in complex scenarios—such as sensor noise, occlusion, and environmental ambiguity—hindering diagnostic analysis and trustworthiness. To address this, this paper introduces kernel SHAP (SHapley Additive exPlanations) to point cloud registration for the first time, establishing an attribution-based uncertainty explanation framework. Our method quantitatively decomposes the marginal contributions of individual uncertainty sources—including noise intensity, occlusion ratio, and geometric ambiguity—to registration failure, producing human-interpretable, ranked attributions. Experiments demonstrate that the framework accurately identifies dominant failure factors, significantly improving both the explainability and diagnostic efficiency of registration failures. This work establishes the first game-theoretic, attribution-driven paradigm for uncertainty explanation in robotic perception, advancing trustworthy autonomous decision-making.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Probabilistic Programming
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Security and Privacy: Data transparency and provenance
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Iterative Closest Point (ICP) is a commonly used algorithm to estimate transformation between two point clouds. The key idea of this work is to leverage recent advances in explainable AI for probabilistic ICP methods that provide uncertainty estimates. Concretely, we propose a method that can explain why a probabilistic ICP method produced a particular output. Our method is based on kernel SHAP (SHapley Additive exPlanations). With this, we assign an importance value to common sources of uncertainty in ICP such as sensor noise, occlusion, and ambiguous environments. The results of the experiment show that this explanation method can reasonably explain the uncertainty sources, providing a step towards robots that know when and why they failed in a human interpretable manner

Problem

• ICP algorithm
• 3D point cloud alignment
• reliability assessment

Innovation

• Interpretable AI
• Kernel SHAP
• ICP Algorithm Evaluation