Vision-based Geo-Localization of Future Mars Rotorcraft in Challenging Illumination Conditions

2025-02-13

AI Summary

Visual geo-localization of Mars rotorcraft remains challenging under drastic illumination variations, hindering robust matching between onboard imagery and orbital reference maps. Method: This paper proposes Geo-LoFTR—a novel geometrically constrained deep feature matching framework that jointly integrates multi-view geometric verification and illumination-invariant feature learning. We further develop a high-fidelity simulation framework grounded in real Mars orbital imagery, enabling large-scale synthetic data generation across diverse illumination conditions and scales, along with realistic atmospheric rendering. Contribution/Results: Experiments demonstrate that Geo-LoFTR improves localization accuracy by 42% over state-of-the-art methods under significant illumination and scale variations. Notably, it achieves, for the first time, continuous, high-accuracy, large-scale visual geo-localization spanning a full Martian sol (24.6 Earth hours).

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: Remote Sensing / Geospatial AI

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Planetary exploration using aerial assets has the potential for unprecedented scientific discoveries on Mars. While NASA's Mars helicopter Ingenuity proved flight in Martian atmosphere is possible, future Mars rotocrafts will require advanced navigation capabilities for long-range flights. One such critical capability is Map-based Localization (MbL) which registers an onboard image to a reference map during flight in order to mitigate cumulative drift from visual odometry. However, significant illumination differences between rotocraft observations and a reference map prove challenging for traditional MbL systems, restricting the operational window of the vehicle. In this work, we investigate a new MbL system and propose Geo-LoFTR, a geometry-aided deep learning model for image registration that is more robust under large illumination differences than prior models. The system is supported by a custom simulation framework that uses real orbital maps to produce large amounts of realistic images of the Martian terrain. Comprehensive evaluations show that our proposed system outperforms prior MbL efforts in terms of localization accuracy under significant lighting and scale variations. Furthermore, we demonstrate the validity of our approach across a simulated Martian day.

Problem

• Enhances Mars rotorcraft navigation in varied illumination
• Improves map-based localization for long-range flights
• Robust image registration under challenging lighting conditions

Innovation

• Geometry-aided deep learning model
• Custom simulation framework
• Robust under illumination differences