Localization Meets Uncertainty: Uncertainty-Aware Multi-Modal Localization
This work addresses the need for more reliable localization in real-world robot navigation, though it is incremental as it builds on existing multi-modal localization methods by adding an uncertainty rejection strategy.
The paper tackles the problem of unreliable robot localization in complex indoor environments by proposing an uncertainty-aware method that filters out unreliable pose predictions using aleatoric and epistemic uncertainties, resulting in significant reductions in mean position error by up to 69.4% and orientation error by up to 73.3% with stricter thresholds.
Reliable localization is critical for robot navigation in complex indoor environments. In this paper, we propose an uncertainty-aware localization method that enhances the reliability of localization outputs without modifying the prediction model itself. This study introduces a percentile-based rejection strategy that filters out unreliable 3-DoF pose predictions based on aleatoric and epistemic uncertainties the network estimates. We apply this approach to a multi-modal end-to-end localization that fuses RGB images and 2D LiDAR data, and we evaluate it across three real-world datasets collected using a commercialized serving robot. Experimental results show that applying stricter uncertainty thresholds consistently improves pose accuracy. Specifically, the mean position error is reduced by 41.0%, 56.7%, and 69.4%, and the mean orientation error by 55.6%, 65.7%, and 73.3%, when applying 90%, 80%, and 70% thresholds, respectively. Furthermore, the rejection strategy effectively removes extreme outliers, resulting in better alignment with ground truth trajectories. To the best of our knowledge, this is the first study to quantitatively demonstrate the benefits of percentile-based uncertainty rejection in multi-modal end-to-end localization tasks. Our approach provides a practical means to enhance the reliability and accuracy of localization systems in real-world deployments.