Uncertainty-Aware Lidar Place Recognition in Novel Environments

TL;DR

This paper introduces a new benchmark for uncertainty-aware lidar place recognition, demonstrating that ensemble methods improve performance in novel environments despite higher computational costs.

Contribution

It presents the first comprehensive benchmark for uncertainty-aware lidar place recognition across multiple datasets and techniques, highlighting ensemble methods as the most effective.

Findings

Ensemble methods outperform other uncertainty estimation techniques.

Uncertainty-aware models improve recognition reliability in new environments.

The benchmark facilitates future research in this area.

Abstract

State-of-the-art lidar place recognition models exhibit unreliable performance when tested on environments different from their training dataset, which limits their use in complex and evolving environments. To address this issue, we investigate the task of uncertainty-aware lidar place recognition, where each predicted place must have an associated uncertainty that can be used to identify and reject incorrect predictions. We introduce a novel evaluation protocol and present the first comprehensive benchmark for this task, testing across five uncertainty estimation techniques and three large-scale datasets. Our results show that an Ensembles approach is the highest performing technique, consistently improving the performance of lidar place recognition and uncertainty estimation in novel environments, though it incurs a computational cost. Code is publicly available at https://github.com/csiro-robotics/Uncertainty-LPR.