Space Operations: before the button is pressed

I am a space engineer working at the intersection of AI, satellite operations, and validation tooling.

At DLR, I am studying how satellite operations teams can structure the decision step before commanding, re-tasking, deferring, rerouting, prioritizing, or blocking an action.

The focus is not autonomous approval.

The question is whether operators would benefit from a clearer pre-command decision layer: operator intent, AI/tool suggestions, deterministic safety gates, planning constraints, evidence, approval state, and replayable rationale in one place.

My related work includes an AI decision-assurance layer for satellite operations, where model-generated proposals are translated into bounded constellation decisions such as feasible plan or cannot proceed, with reasons and approval state.

At SmallSat Europe, I would like to meet satellite operators, ground-segment teams, autonomy/software teams, mission planners, and organizations working on AI-assisted operations.

I am especially interested in shadow-mode studies using historical or redacted operations cases: cases where a recommendation looked plausible, but needed clearer evidence, safety gates, or human approval before action.