Kilometer-Scale GNSS-Denied UAV Navigation via Heightmap Gradients: A Winning System from the SPRIN-D Challenge

2025-10-01

AI Summary

In GNSS-denied, loop-closure-free environments with limited embedded computational resources, UAVs suffer severe odometry drift during long-range (9 km), low-altitude (<25 m AGL) autonomous navigation. Method: This paper proposes a lightweight elevation gradient matching localization correction method: real-time local elevation maps are constructed from onboard LiDAR; template matching in the gradient domain aligns these maps with prior georeferenced digital elevation data; and tightly coupled visual-inertial odometry is fused via clustered particle filtering. The entire pipeline runs exclusively on CPU without requiring loop closure detection. Contribution/Results: Evaluated in the SPRIN-D Challenge, the system achieved beyond-line-of-sight, GNSS-free autonomous flight across complex terrains—including urban, forested, and open areas—while maintaining kilometer-scale real-time positioning accuracy. This significantly enhances robustness and practicality for long-distance navigation in unknown environments.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Local Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Reliable long-range flight of unmanned aerial vehicles (UAVs) in GNSS-denied environments is challenging: integrating odometry leads to drift, loop closures are unavailable in previously unseen areas and embedded platforms provide limited computational power. We present a fully onboard UAV system developed for the SPRIN-D Funke Fully Autonomous Flight Challenge, which required 9 km long-range waypoint navigation below 25 m AGL (Above Ground Level) without GNSS or prior dense mapping. The system integrates perception, mapping, planning, and control with a lightweight drift-correction method that matches LiDAR-derived local heightmaps to a prior geo-data heightmap via gradient-template matching and fuses the evidence with odometry in a clustered particle filter. Deployed during the competition, the system executed kilometer-scale flights across urban, forest, and open-field terrain and reduced drift substantially relative to raw odometry, while running in real time on CPU-only hardware. We describe the system architecture, the localization pipeline, and the competition evaluation, and we report practical insights from field deployment that inform the design of GNSS-denied UAV autonomy.

Problem

• Enables long-range UAV navigation without GNSS signals
• Corrects odometry drift using heightmap gradient matching
• Achieves autonomous low-altitude flight in diverse terrains

Innovation

• LiDAR heightmap gradient-template matching for localization
• Clustered particle filter fusing geo-data with odometry
• Real-time CPU-only onboard navigation in GNSS-denied environments