BEVIO: Efficient Bird�s-Eye-View Based Sparse-Update Visual-Inertial Odometry for Lunar Day-Night Navigation
Mohit Singh, Shehryar Khattak, Ashish Goel, Michael Paton, Kostas Alexis, Issa Nesnas
AI summary
Problem
Planetary rovers face extreme computational and energy constraints, making high-frequency visual updates for Visual-Inertial Odometry (VIO) impractical, especially during challenging day-night transitions and self-illumination conditions.
Approach
The authors propose a scene-adaptive Bird’s-Eye View (BEV) feature-matching framework that warps camera images to a ground-aligned perspective, stabilizing feature appearance and enabling robust matching across large inter-frame baselines.
Key results
- Enables reliable VIO at visual update rates as low as 0.25 Hz
- Maintains robust feature matching across large inter-frame baselines (~1 m)
- Validated through high-fidelity lunar simulations and real-world day/night traversals
- Reduces onboard computational and energy requirements for state estimation
Why it matters
This method makes long-term, high-speed lunar exploration feasible by drastically lowering the onboard compute and power requirements for state estimation.
Abstract
Visual–Inertial Odometry (VIO) provides smooth, high-rate state estimates and has been widely used for robotic navigation in both terrestrial and planetary applications. How- ever, its performance is typically dependent on the frequency of visual updates, which is a challenge for planetary rovers operating under extreme resource constraints and low frame rates. This work investigates enabling reliable VIO with very sparse visual updates for lunar rover applications, addressing both day and night-time operations where feature associations become especially difficult under self-illumination conditions. We propose a Bird’s Eye View (BEV)–based image matching scheme that remains robust to larger inter-frame motions and more reliable feature matching despite significant visual appearance changes. We extensively evaluate our proposed approach, BEVIO, through high-fidelity photorealistic lunar and real-time robotic experiments conducted using a half-scale lunar rover, in a long-term day–night deployment at Plaster City, CA, USA. The results demonstrate that our method enables reliable day and nighttime self-illuminated traverses at visual update rates as low as 0.25 Hz, underscoring its suitability for navigation on power- and compute-limited lunar rovers.