1. The Airbus A320's diversity approach to redundant flight computers involves:
Correct. Software diversity — different teams, different languages — dramatically reduces the probability that both systems contain the same software bug, addressing common-cause software failure.
The A320 uses software diversity: different programming teams using different languages, so software bugs are unlikely to appear identically in both systems.
2. How many drones did Intel fly in a coordinated autonomous formation during the PyeongChang 2018 Winter Olympics closing ceremony?
Correct. 1,218 Intel Shooting Star drones flew with decentralized collision avoidance — no human monitored individual drone behavior during the 8-minute show.
Not quite. Intel flew 1,218 drones in the PyeongChang 2018 closing ceremony, with decentralized autonomous collision avoidance.
3. The CFPB's 2022 guidance most directly establishes what AI governance principle?
Correct. The CFPB closed the "black box" exemption — lenders cannot cite algorithmic complexity to avoid providing specific adverse action reasons. Interpretability is a legal compliance requirement.
The CFPB guidance established that model complexity is not a defense against adverse action notice requirements — interpretability is legally mandated in consumer credit AI.
4. Model Predictive Control (MPC) in the planning stage generates trajectories by:
Correct. MPC optimizes over a finite prediction horizon — balancing objectives like travel time, comfort, and safety margins — then executes the first portion of the plan before re-optimizing with updated information.
MPC optimizes a cost function over a rolling time horizon. Review Lesson 3's Planning section.
5. The Extended Kalman Filter (EKF) is used in autonomous drone navigation to do what?
Correct. The EKF is a sensor fusion algorithm — it combines noisy measurements from multiple sources (IMU, GPS, optical flow, etc.) into a single best-estimate state.
Not quite. The Extended Kalman Filter fuses data from multiple sensors to produce a single, more reliable state estimate than any individual sensor can provide.
6. What does "6-DoF pose estimation" mean in robotic manipulation, and why does it matter for grasping?
Correct. 6-DoF means the full position (three translational) plus orientation (three rotational) — everything a robot arm needs to know about an object's pose to approach and grasp it precisely.
Not quite. 6-DoF pose estimation means determining all six dimensions of an object's position and orientation in 3D space — three position coordinates plus three rotation angles — which is what a robot needs to plan a successful grasp.
7. New Jersey's 2017 bail reform requirement that judges document departures from algorithmic recommendations is an example of:
Correct. The New Jersey system creates accountability in both directions: the algorithm's recommendations are on record, and judicial departures must be explained — a bidirectional audit trail preserving both algorithmic and human accountability.
The documentation requirement creates a bidirectional audit trail — both algorithmic recommendations and human departures are documented. Neither the AI nor the judge can act without a record.
8. The Facebook/Myanmar case demonstrates the "scale-oversight gap" because:
Correct. The scale-oversight gap: Facebook's system operated at a scale and in a linguistic context that its oversight infrastructure could not handle — no Burmese-language moderators until late 2015, inadequate capacity through 2017.
The scale-oversight gap: Facebook deployed at scale in a context where oversight infrastructure (Burmese-language capacity, moderator staffing) was wholly inadequate — not malicious design, but structural oversight failure.
9. The Astrée static analyzer was used to verify the absence of runtime errors in flight control software for which aircraft?
Correct. Astrée, developed at INRIA using abstract interpretation, was applied to primary flight control software for the A340 and A380.
Astrée was applied to Airbus A340 and A380 primary flight control software — an early industrial application of formal static analysis at scale.
10. Amazon's Sparrow robot uses what primary sensor combination to identify and grasp items from mixed bins?
Correct. Sparrow combines RGB-D perception with force-sensing grasping and a neural recognition network trained on millions of product images.
Not quite. Sparrow uses RGB-D cameras (color plus depth), a vacuum gripper with force feedback, and a neural network trained on product images.
11. What is the primary challenge with "long-tail events" for autonomous vehicle development?
Correct. Long-tail events are fundamentally challenging because the universe of unusual scenarios is open-ended — and simulation, the primary mitigation tool, cannot capture the full diversity of real-world stochasticity.
The challenge is the unbounded space of possible unusual scenarios combined with simulation's inherent limits. Review Lesson 4's Unsolved Problems section.
12. In James Reason's Swiss Cheese Model applied to autonomous systems, what condition must be met for an accident to occur?
Correct. The Swiss Cheese Model emphasizes systemic multi-layer failure alignment — not single points of failure — as the mechanism of accident causation.
The Swiss Cheese Model holds that accidents require simultaneous alignment of vulnerabilities across multiple independent layers — a systems perspective on failure causation.
13. Edsger Dijkstra's famous observation about software testing states that testing can:
Correct. This foundational observation motivates formal verification — mathematical proof rather than empirical testing — for safety-critical systems.
Dijkstra's principle: testing shows presence of bugs, never absence. This is why formal verification exists.
14. Which accident most directly caused the FAA to mandate explicit MCAS training and dual-sensor input requirements?
Correct.
The 737 MAX crashes — Lion Air 610 (2018) and Ethiopian Airlines 302 (2019) — directly resulted in MCAS-specific regulatory requirements.
15. In the 2018 Uber Tempe fatality, how many seconds before impact did the perception system first detect Elaine Herzberg?
Correct. The NTSB found the system detected the pedestrian 5.6 seconds before impact but failed to stabilize classification, and emergency braking had been disabled.
The NTSB documented 5.6 seconds of detection time before impact — sufficient for emergency braking had it not been disabled.
16. In the Uber ATG fatal crash (2018), the safety driver failed to exercise oversight because:
Correct. The NTSB found the safety driver was watching a phone-streamed video and that the monitoring system had no automated alert for driver inattention — the oversight expected but not technically ensured.
The NTSB documented that the safety driver was watching Hulu and that no automated alert system existed to detect or respond to driver inattention — oversight was expected but not technically guaranteed.
17. What is an Operational Design Domain (ODD)?
Correct. ODD defines the envelope of conditions a system is designed for. A Level 4 system operating outside its ODD is effectively unguided — the system is not certified for those conditions.
ODD is the set of conditions — not just geography — within which the system is designed to operate safely. Review Lesson 2.
18. Google's RT-2 (Robotics Transformer 2, 2023) demonstrated what key capability for robot manipulation?
Correct. RT-2 showed that internet-scale vision-language pre-training could be fine-tuned for physical robot control, enabling manipulation of novel objects without task-specific robot training data.
Not quite. RT-2's advance was using internet-scale vision-language pre-training — fine-tuned for robot arm control — to manipulate novel objects described in natural language without any specific robot training data for those objects.
19. LIME and SHAP are best described as:
Correct. Both are model-agnostic post-hoc explanation methods — they approximate local decision behavior for specific predictions without modifying the underlying model.
LIME and SHAP are post-hoc explanation methods — model-agnostic approaches that generate approximations of what drove a specific prediction.
20. Which company became the first to achieve regulatory approval for a Level 3 automated driving system in production vehicles?
Correct. Mercedes-Benz received approval for Drive Pilot in Germany in December 2021 and in Nevada in January 2023 — the first Level 3 system approved in production vehicles, with Mercedes accepting legal liability while engaged.
Mercedes-Benz was first with Drive Pilot — approved in Germany in 2021 and Nevada in 2023.