Overview
What this challenge is about.
You receive 30 days of mission logs with task lists, time windows, and actual durations. Encode the planning problem with temporal PDDL (PDDL 2.1 durative actions) and solve with a temporal planner (e.g., OPTIC or POPF). Build a replay harness that simulates the planner under realistic battery + walking-time variance. Compare missed-window count + total slack vs. the current scheduler. Deliver a 3-page memo with deployment-readiness assessment.
The Brief
What you'll do, and what you'll demonstrate.
Cut the missed-time-window rate by 50 percent on 30 days of replayed missions with a temporal PDDL planner.
Earning criteria — what you'll demonstrate
- Model durative actions and time windows in PDDL 2.1
- Run a temporal planner on realistic operational data
- Simulate execution variance to stress-test planner robustness
- Translate planner results into a deployment-readiness assessment
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
Temporal-planner integration on real robot logs is high-leverage AI engineering work at any robotics startup.
This challenge sharpens
- temporal-planning
- pddl-modeling
- constraint-handling
Applied AI Scientist
Replaying real logs with realistic execution variance is the rigorous applied-AI methodology a research-driven robotics company expects.
This challenge sharpens
- simulation
- benchmarking
- temporal-planning
ML Researcher
Treating planner comparison as a controlled experiment with replay data is the ML researcher's contribution to a robotics product team.
This challenge sharpens
- benchmarking
- simulation
- temporal-planning