Compiling Temporal Numeric Planning into Discrete PDDL+: Extended Version

2026-03-12 · Source: Artificial Intelligence · Field: Technology & Digital — Artificial Intelligence & Machine Learning · Depth: Advanced, quick

Summary

A new practical compilation method has been developed to translate temporal planning problems with durative actions, as defined in PDDL 2.1, into the PDDL+ modeling language. This compilation fully captures the original semantics and operates under the assumption that actions do not self-overlap. The method is polynomial in complexity, preserves the plan length within a constant factor, and has demonstrated practical utility for solving complex temporal numeric planning problems. This addresses a long-standing gap in the literature since PDDL+ was introduced, as no practical compilation for this purpose had been previously presented.

Key takeaway

For AI Researchers and Planning Scientists working with temporal numeric problems, this compilation offers a robust method to leverage PDDL+ solvers for PDDL 2.1 durative action models. You should consider integrating this approach to tackle complex planning scenarios, potentially improving solution efficiency and expanding the range of solvable problems by utilizing the advanced features of PDDL+.

Key insights

A practical, polynomial compilation translates PDDL 2.1 durative actions into PDDL+ while preserving semantics.

Principles

• Non-self-overlapping actions simplify temporal compilation.
• Polynomial compilation can be practically relevant.

Method

The method compiles PDDL 2.1 durative actions into PDDL+ by fully capturing semantics, assuming non-self-overlapping actions, and maintaining plan length within a constant factor.

In practice

• Solve hard temporal numeric problems.
• Bridge PDDL 2.1 and PDDL+ models.

Topics

• Temporal Planning
• PDDL+
• Durative Actions
• Planning Compilation
• Numeric Planning

Best for: AI Researcher, AI Scientist, Research Scientist

Related on AIssential

• PO-PDDL: Learning Symbolic POMDPs from Visual Demonstrations for Robot Planning Under Uncertainty — PO-PDDL learns symbolic POMDPs from visual demonstrations for robust robot planning under uncertainty.
• Exploiting Search in Symbolic Numeric Planning with Patterns — Dynamic pattern recomputation and refinement in symbolic numeric planning guides search more efficiently.
• Google Goes to EXTREMES: In-Context Symbolic AI — LLMs can perform symbolic reasoning with high accuracy when rigorously constrained by in-context PDDL definitions.
• Learning Visual Spatial Planning from Symbolic State via Modality-Gap-Aware Self-Distillation — Visual spatial planning benefits from self-distillation that bridges the perception-reasoning modality gap.
• Semantic Partial Grounding via LLMs — LLMs can semantically prune PDDL planning tasks to accelerate grounding and improve plan quality.
• VLANeXt: The Design Recipes Behind Vision-Language-Action Robots — VLANeXt provides design recipes for integrating vision, language, and action in robots.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.