Thermal-LiDAR Fusion for Robust Tunnel Localization in GNSS-Denied and Low-Visibility Conditions

TL;DR

This work targets reliable localization in GNSS-denied, low-visibility environments such as tunnels by fusing LiDAR odometry with thermal-camera odometry through a loosely-coupled Extended Kalman Filter. It combines GenZ-ICP-based LiDAR odometry with LDSO-based thermal odometry under a constant-acceleration motion model, represented by the state $x=[x,y,v,\dot v,\psi,\dot\psi,\ddot\psi]^T$, and uses pseudo-measurements with EKF updates to maintain accurate pose despite sensor outages. Key findings show that the fusion approach sustains localization where single-sensor methods deteriorate, though long-term drift remains without loop-closure and IMU data; the framework is modular and extendable for cyber-physical systems in constrained environments. The work demonstrates practical potential for autonomous vehicles and inspection robots operating in GNSS-denied and perceptually poor settings.

Abstract

Despite significant progress in autonomous navigation, a critical gap remains in ensuring reliable localization in hazardous environments such as tunnels, urban disaster zones, and underground structures. Tunnels present a uniquely difficult scenario: they are not only prone to GNSS signal loss, but also provide little features for visual localization due to their repetitive walls and poor lighting. These conditions degrade conventional vision-based and LiDAR-based systems, which rely on distinguishable environmental features. To address this, we propose a novel sensor fusion framework that integrates a thermal camera with a LiDAR to enable robust localization in tunnels and other perceptually degraded environments. The thermal camera provides resilience in low-light or smoke conditions, while the LiDAR delivers precise depth perception and structural awareness. By combining these sensors, our framework ensures continuous and accurate localization across diverse and dynamic environments. We use an Extended Kalman Filter (EKF) to fuse multi-sensor inputs, and leverages visual odometry and SLAM (Simultaneous Localization and Mapping) techniques to process the sensor data, enabling robust motion estimation and mapping even in GNSS-denied environments. This fusion of sensor modalities not only enhances system resilience but also provides a scalable solution for cyber-physical systems in connected and autonomous vehicles (CAVs). To validate the framework, we conduct tests in a tunnel environment, simulating sensor degradation and visibility challenges. The results demonstrate that our method sustains accurate localization where standard approaches deteriorate due to the tunnels featureless geometry. The frameworks versatility makes it a promising solution for autonomous vehicles, inspection robots, and other cyber-physical systems operating in constrained, perceptually poor environments.

Figures (4)

• Figure 1: Top--down 2D view of the map used as the simulation environment.
• Figure 2: The position estimate of the kalman filter is shown in addition the GenZ-ICP Odometry is shown to highlight the influence of the thermal camera odometry. Lying underneath is the ground truth trajectory shown.
• Figure 3: Estimated angle of the extended Kalman filter, and the LiDAR odometry, as well as the ground truth.
• Figure 4: Position error between the ground truth and the estimated trajectory.