Background · Automated Planning

Automated Planning, at Scale

Automated planning finds an ordered sequence of actions that transforms an initial state into a goal state.

📥
Initial State + Goalspecified in PDDL
→
⚙️
Plannersearch + heuristic
→
📋
Planaction sequence

A vast ecosystem

Many domains, planners, search algorithms & heuristics — each combination differs in quality, cost & optimality.

Rich empirical data

The International Planning Competition (IPC) yields performance results across many domains.

But unstructured

This metadata is rarely captured in a structured, reasoning-friendly form — the gap we address.