Guarded Autonomy for Satellites: A Policy-Gated AI Agent Framework for Safe Onboard Fault Response
Abstract
Spacecraft increasingly face faults that onboard rule-based systems were never designed to handle, yet flight operators cannot give a probabilistic AI agent direct control of a vehicle. This paper presents AEGIS-Sat, a framework that places a deterministic policy gate between an AI agent and the spacecraft. The agent observes a fault and proposes a single structured recovery action; it can execute nothing itself. A policy gate, written as human-readable YAML, rules on every proposal — AUTO (execute now), AUTO+LOG (execute and flag for ground review), ESCALATE (hold safe and request ground approval), or DENY (blocked) — after checking hard safety invariants and a context-adaptive autonomy envelope that tightens the tiers during communications blackout and eclipse and widens them as power margin recovers. Every proposal, ruling, and state snapshot is recorded in a hash-chained, tamper-evident audit ledger. In a fault-injection evaluation (6 scenarios × 3 setups × 100 episodes = 1,800 episodes), an unguarded agent recovered from 88% of episodes but executed 75 unsafe actions. The guarded agent executed zero unsafe actions, recovered in 100% of episodes (67% full recovery, 33% safe-state recovery), and lost no spacecraft, at the cost of 659 ground escalations. A rule-only FDIR baseline was equally safe but fully recovered from only 33% of episodes. The gate and benchmark are anchored on real telemetry: a deliberately simple logistic detector over ESA's OPS-SAT-AD dataset reaches AUC 0.887, detecting 58% of anomalous segments at a 2.2% false-alarm rate. Gating an untrusted controller is not new — the approach is Simplex-inspired runtime assurance translated to agent-based fault response. The contribution claimed here is the combination: readable policy tiers, a context-adaptive envelope, an open unsafe-action benchmark, and a tamper-evident audit trail, evaluated head-to-head on injected faults.
Related
- DewTwin-Coin: an onboard autonomous framework for lunar water-ice prospecting using Chandrayaan-3 LIBS and ChaSTE data
- Zero-Onboard-AI-Compute Embodied Intelligence: A General Framework for Collective Computing Through Idle Resources of Other Embodied Agents
- A pilot-wellness-informed ecological framework for human-autonomy teaming in future fighter aviation
- First Demonstration of an Onboard Polarimetric Calibration and Ice-Cloud-Based Validation Framework for the Particulate Observing Scanning Polarimeter (POSP)
- THE REFERENCE-AUTONOMY DYNAMICS PRINCIPLE (RADP) A General Theoretical Framework for Reference-Bound Pursuit, Capability-Referenced Autonomy, and Their Complementary Causal Dynamics
BibTeX
@misc{pandey2026guarded,
title = {{Guarded Autonomy for Satellites: A Policy-Gated AI Agent Framework for Safe Onboard Fault Response}},
author = {Karmendra Pandey},
howpublished = {Zenodo},
year = {2026},
month = oct,
doi = {10.5281/zenodo.23425891},
url = {https://zenodo.org/records/23425891},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Guarded Autonomy for Satellites: A Policy-Gated AI Agent Framework for Safe Onboard Fault Response
AU - Karmendra Pandey
PY - 2026
DA - 2026/10/16
PB - Zenodo
DO - 10.5281/zenodo.23425891
UR - https://zenodo.org/records/23425891
AB - Spacecraft increasingly face faults that onboard rule-based systems were never designed to handle, yet flight operators cannot give a probabilistic AI agent direct control of a vehicle. This paper presents AEGIS-Sat, a framework that places a deterministic policy gate between an AI agent and the spacecraft. The agent observes a fault and proposes a single structured recovery action; it can execute nothing itself. A policy gate, written as human-readable YAML, rules on every proposal — AUTO (execute now), AUTO+LOG (execute and flag for ground review), ESCALATE (hold safe and request ground approval), or DENY (blocked) — after checking hard safety invariants and a context-adaptive autonomy envelope that tightens the tiers during communications blackout and eclipse and widens them as power margin recovers. Every proposal, ruling, and state snapshot is recorded in a hash-chained, tamper-evident audit ledger. In a fault-injection evaluation (6 scenarios × 3 setups × 100 episodes = 1,800 episodes), an unguarded agent recovered from 88% of episodes but executed 75 unsafe actions. The guarded agent executed zero unsafe actions, recovered in 100% of episodes (67% full recovery, 33% safe-state recovery), and lost no spacecraft, at the cost of 659 ground escalations. A rule-only FDIR baseline was equally safe but fully recovered from only 33% of episodes. The gate and benchmark are anchored on real telemetry: a deliberately simple logistic detector over ESA's OPS-SAT-AD dataset reaches AUC 0.887, detecting 58% of anomalous segments at a 2.2% false-alarm rate. Gating an untrusted controller is not new — the approach is Simplex-inspired runtime assurance translated to agent-based fault response. The contribution claimed here is the combination: readable policy tiers, a context-adaptive envelope, an open unsafe-action benchmark, and a tamper-evident audit trail, evaluated head-to-head on injected faults.
ER -