A New Approach to 3D ICP Covariance Estimation

TL;DR

This paper introduces a novel method for estimating the 3D uncertainty of ICP in mobile robotics, accounting for initialization errors, sensor noise, and convergence issues, validated on real datasets.

Contribution

It presents a new approach that models ICP covariance considering initialization and sensor biases, improving uncertainty estimation accuracy.

Findings

Predicts consistent uncertainty with real data

Outperforms previous methods in various environments

Accounts for biases and convergence issues in ICP

Abstract

In mobile robotics, scan matching of point clouds using Iterative Closest Point (ICP) allows estimating sensor displacements. It may prove important to assess the associated uncertainty about the obtained rigid transformation, especially for sensor fusion purposes. In this paper we propose a novel approach to 3D uncertainty of ICP that accounts for all the sources of error as listed in Censi's pioneering work [1], namely wrong convergence, underconstrained situations, and sensor noise. Our approach builds on two facts. First, the uncertainty about the ICP's output fully depends on the initialization accuracy. Thus speaking of the covariance of ICP makes sense only in relation to the initialization uncertainty, which generally stems from odometry errors. We capture this using the unscented transform, which also reflects correlations between initial and final uncertainties. Then, assuming white sensor noise leads to overoptimism as ICP is biased owing to e.g. calibration biases, which we account for. Our solution is tested on publicly available real data ranging from structured to unstructured environments, where our algorithm predicts consistent results with actual uncertainty, and compares favorably to previous methods.