First On-Orbit Demonstration of a Geospatial Foundation Model

TL;DR

This paper introduces compact, resource-efficient geospatial foundation models based on Vision Transformers, demonstrating their successful deployment and inference onboard the International Space Station for Earth observation tasks.

Contribution

It presents novel compressed GeoFMs that maintain performance while enabling onboard deployment, validated through real space environment testing.

Findings

Model compression reduces size significantly while preserving accuracy.

Domain adaptation enhances performance in operational conditions.

Successful on-orbit inference demonstrates practical deployment feasibility.

Abstract

Geospatial foundation models (GeoFMs) promise broad generalisation capacity for Earth observation (EO) tasks, particularly under data-limited conditions. However, their large size poses a barrier to deployment on resource-constrained space hardware. To address this, we present compact variants of a Vision Transformer (ViT)-based GeoFM that preserve downstream task performance while enabling onboard execution. Evaluation across five downstream tasks and validation in two representative flight environments show that model compression and domain adaptation are critical to reducing size and resource demands while maintaining high performance under operational conditions. We further demonstrate reliable on-orbit inference with the IMAGIN-e payload aboard the International Space Station. These results establish a pathway from large GeoFMs to flight-ready, resource-efficient deployments, expanding the feasibility of onboard AI for EO missions.