id stringclasses 6 values | aircraft_context stringclasses 1 value | flight_phase stringclasses 6 values | abnormal_condition stringclasses 6 values | loop_coherence_index float64 0.44 0.92 | primary_decoherence_source stringclasses 5 values | resonance_pattern_type stringclasses 6 values | key_constraints stringclasses 6 values | loss_of_control_risk float64 0.06 0.72 | recovery_window_seconds int64 18 180 | recommended_adaptive_action stringclasses 6 values | intervention_priority stringclasses 3 values | recovery_confidence float64 0.7 0.95 | post_action_stability_expectation stringclasses 6 values | notes stringclasses 6 values | constraints stringclasses 1 value | gold_checklist stringclasses 1 value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
AIR-001 | A320 | go-around | flap asymmetry | 0.44 | aircraft | pilot_fighting_aircraft | low altitude; high workload | 0.72 | 18 | augment lateral+pitch control law; cue flap asymmetry on PFD; suggest stable thrust setting | urgent | 0.82 | coherence rises to 0.70 if augmentation engaged | aircraft-driven mismatch needs system help | Under 260 words | risk+window+action+priority+conf+expect |
AIR-002 | A320 | final | none | 0.92 | none | stable_resonance | normal ops | 0.06 | 120 | none | low | 0.95 | stable | no intervention needed | Under 260 words | risk+window+action+priority+conf+expect |
AIR-003 | A320 | climb | engine surge | 0.51 | aircraft | thrust_phase_mismatch | terrain clearance; climb gradient | 0.66 | 30 | stabilize thrust control; limit throttle oscillation via damped law; provide cue: avoid rapid thrust changes | high | 0.78 | coherence rises to 0.68 after damping | reduce oscillation driver | Under 260 words | risk+window+action+priority+conf+expect |
AIR-004 | A320 | descent | startle | 0.58 | pilot | overcontrol_pattern | IMC; workload high | 0.54 | 45 | reduce control sensitivity briefly; present calm step cue; recommend autopilot re-engage if within envelope | high | 0.74 | coherence rises to 0.72 if auto engaged | pilot assistance | Under 260 words | risk+window+action+priority+conf+expect |
AIR-005 | A320 | takeoff | sensor disagreement | 0.49 | mixed | sensor_feedback_loop | takeoff; no time to troubleshoot | 0.7 | 20 | isolate suspect sensor; switch to voted source; lock trim rate; cue: use pitch attitude primary | urgent | 0.8 | coherence rises to 0.66 after isolation | contain propagation fast | Under 260 words | risk+window+action+priority+conf+expect |
AIR-006 | A320 | holding | turbulence | 0.78 | environment | turbulence_resonance | moderate bumps | 0.25 | 180 | none; monitor; suggest speed adjustment for ride | low | 0.7 | stable within turbulence band | environment not failure | Under 260 words | risk+window+action+priority+conf+expect |
What this dataset tests
Whether a system can turn loop decoherence
into stabilizing action under time pressure.
It must estimate:
risk
recovery window
best adaptive intervention
expected post-action stability.
Required outputs
- loss_of_control_risk
- recovery_window_seconds
- recommended_adaptive_action
- intervention_priority
- recovery_confidence
- post_action_stability_expectation
Use case
Layer three of Pilot–Vehicle Loop Coherence Under Stress.
Supports:
- enhanced crew alerting
- adaptive control law selection
- graceful stabilization before LOC
- Downloads last month
- 27