Could a Co-Pilot be an Aircraft Co-Pilot
summary
Shows concrete failures and prototypes: MCAS killed single‑AoA sensor failure (189 and 157 fatalities cited), engine birdsurge and Hudson ditching where pilots used simulator‑trained heuristics like a 20–30 ft “flap bump” that an AI lacks, and a 2008 Heathrow fuel‑freeze go‑around saved by a split‑second flap decision. Describes engineering tools and deployments: Scade to Ada codegen, ECAM/auto‑land, Garmin‑style emergency airfield picker, Imperial College quantum inertial ISO container, camera‑based dragonfly vision and an autonomous taxi/takeoff demo that worked under pilot oversight.
AI-written summary · about
chapters (14)
- 0:00AI co-pilots in aviation
- 4:01Aircraft computer complexity
- 11:59Development cycles and safety
- 14:47Real vs AI co-pilots
- 19:21Past aviation incidents
- 27:13Emergency decisions
- 31:08Boeing 737 MAX MCAS
- 35:26Autoland system limits
- 40:29GPS jamming solutions
- 42:37AI automation paths
- 45:45Computer vision taxiing
- 48:10NLP for taxi assistance
- 51:00AI replace co-pilot?
- 53:30Post-flight reviews
related talks
16mThird-Person Piloting: Increasing Situational Awareness using a Spatially Coupled Second Drone
42mAutomated Driving - Taking the Human Factors Researcher out of the Loop
49mAI Composite in the Mayflower Autonomous navigation solution
26mComplex Autonomous UAV Task Execution and Decision-Making Using s(CASP)
46mPaparazzi, An Open Source UAV for Research
21mAutonomous driving and handling the handover
09mWIP: Hijacking Attacks on UAV Follow-Me Systems in Realistic Scenarios
57mReal-Time Decision Making in Hybrid Systems