Lights Out: A Nighttime UAV Localization Framework Using Thermal Imagery and Semantic 3D Maps

2026-04-28

AI Summary

This work addresses the challenge of nighttime UAV localization in GNSS-denied environments, where large modality gaps between thermal imagery and daytime RGB maps hinder appearance-based matching. To overcome this, the authors propose a semantics-driven re-projection localization method that operates without relying on visual appearance. The approach aligns segmented nighttime thermal images with a pre-built daytime semantic 3D map within a shared semantic space, enhanced by a confusion-aware weighting scheme and a symmetric bidirectional re-projection objective to improve robustness in semantically ambiguous regions. Evaluated on a real-world nighttime flight trajectory spanning 6.5 kilometers, the method achieves a 2D RMSE of 2.18 meters after bias correction, with a median error of only 1.52 meters, demonstrating its effectiveness and novelty.

Abstract

Reliable backup localization for unmanned aerial vehicles (UAVs) operating in GNSS-denied nighttime conditions remains an open challenge due to the severe modality gap between daytime RGB maps and nighttime thermal imagery. This work presents a semantic reprojection framework for map-relative nighttime UAV localization by aligning segmented thermal observations with a globally referenced, semantically labeled 3D map constructed from daytime RGB data. Rather than relying on appearance-based correspondence, localization is formulated in a shared semantic domain and solved via a symmetric bidirectional reprojection objective with confusion-aware weighting to improve robustness under segmentation uncertainty. The approach is evaluated offline across 6.5 km of nighttime, real-world UAV flight trajectories in urban and semi-structured environments. Relative to RTK GNSS ground truth, the system achieves a bias-corrected RMSE2D of 2.18 m and a median RMSE2D of 1.52 m. Results show that localization performance is strongly correlated with the availability of semantic edge evidence and that large-error events are spatially localized to semantically ambiguous areas rather than uniformly distributed. These findings indicate that semantic reprojection offers a promising pathway toward globally referenced nighttime UAV localization using thermal imagery alone.

Problem

• UAV localization
• GNSS-denied
• thermal imagery
• semantic 3D maps
• nighttime navigation

Innovation

• semantic reprojection
• thermal imagery
• nighttime UAV localization
• semantic 3D map
• GNSS-denied navigation