6 papers
Adaptive Band Selection for Hyperspectral Classification with Spatially Disjoint Evaluation
Ikram El-Hajri, Ouassim Karrakchou, Alejandro Mousist
Hyperspectral band selection methods based on differentiable selectors can be sensitive to initialization and to extracting a final discrete subset, while prescribed band counts li…
Beyond detection: cooperative multi-agent reasoning for rapid onboard EO crisis response
Alejandro D. Mousist, Pedro Delgado de Robles MartÃn, Raquel Lladró Climent +1
Rapid identification of hazardous events is essential for next-generation Earth Observation (EO) missions supporting disaster response. However, current monitoring pipelines remain…
Confidence-gated training for efficient early-exit neural networks
Saad Mokssit, Ouassim Karrakchou, Alejandro Mousist +1
Early-exit neural networks reduce inference cost by enabling confident predictions at intermediate layers. However, joint training often leads to gradient interference, with deeper…
First On-Orbit Demonstration of a Geospatial Foundation Model
Andrew Du, Roberto Del Prete, Alejandro Mousist +6
Geospatial foundation models (GeoFMs) promise broad generalisation capacity for Earth observation (EO) tasks, particularly under data-limited conditions. However, their large size…
ASTREA: Introducing Agentic Intelligence for Orbital Thermal Autonomy
Alejandro D. Mousist
This paper presents ASTREA, the first agentic system executed on flight-heritage hardware (TRL 9) for autonomous spacecraft operations, with on-orbit operation aboard the Internati…
Real-Time Blind Defocus Deblurring for Earth Observation: The IMAGIN-e Mission Approach
Alejandro D. Mousist
This work addresses mechanical defocus in Earth observation images from the IMAGIN-e mission aboard the ISS, proposing a blind deblurring approach adapted to space-based edge compu…