Towards Explaining Uncertainty Estimates in Point Cloud Registration
This work addresses the need for interpretable uncertainty explanations in robotics, particularly for failure analysis, but is incremental as it applies existing explainable AI techniques to a specific domain.
The paper tackled the problem of explaining uncertainty estimates in point cloud registration by proposing a method based on kernel SHAP to assign importance values to sources like sensor noise and occlusion, with results showing it can reasonably explain these uncertainty sources.
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