Completed visiting research period
The University of Arizona visit focused on reinforcement learning for sensor tasking in space surveillance and tracking.
Academic path ↗Research / Area of inquiry
AI for space applications
How can learning-based methods support observation decisions?
← All research areasConceptual illustration / not research data · Earth imagery: NASA/NOAA
The scientific focus
From question to method.
A 2026 conference contribution examines meta-reinforcement learning for conflict-aware sensor scheduling. The completed University of Arizona visiting period focused on reinforcement learning for SST sensor tasking.
Methods and themes
- Machine learning for SST workflows
- Reinforcement and meta-reinforcement learning
- Human-reviewed, LLM-enabled scientific workflows as an emerging direction
LLM-enabled workflow support is described as an area of interest; no validated LLM result is claimed here. A separate manuscript remains outside the public archive.