Overview
What this challenge is about.
You receive a stylized warehouse map (aisle graph), 30 sample shifts of pick tasks, and the current heuristic's outputs. Write a PDDL domain + problem generator, solve with at least two planners (e.g., Fast Downward + LAMA, plus a satisficing alternative), and compare on plan cost (total robot-minutes), solve time, and aisle-congestion violations. Deliver a 4-page memo with a clear recommendation: keep the heuristic, adopt the planner, or build a hybrid.
The Brief
What you'll do, and what you'll demonstrate.
Decide whether a classical PDDL planner beats the hand-coded pick-route heuristic on plan cost and solve-time under realistic shift loads.
Earning criteria — what you'll demonstrate
- Model a real operational problem in PDDL
- Apply state-space search and heuristic-guided planning in practice
- Benchmark planner trade-offs (cost vs. time) honestly
- Translate planning results into a written engineering recommendation
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
AI Engineer
Modeling an operational problem in PDDL and benchmarking planners is exactly the AI engineering work at any robotics or scheduling-AI company.
This challenge sharpens
- pddl-modeling
- classical-planning
- domain-modeling
Applied AI Scientist
Comparing search-based methods on a real operational benchmark and writing the recommendation memo is core applied AI scientist work.
This challenge sharpens
- state-space-search
- benchmarking
- pddl-modeling
ML Researcher
Treating planner choice as a rigorous experiment is the methodological discipline ML researchers bring to symbolic AI projects.
This challenge sharpens
- state-space-search
- benchmarking
- classical-planning