Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Planning as Goal Recognition: Deriving Heuristics from Intention Models - Extended Version

Giacomo Rosa, Jean Honorio, Nir Lipovetzky, Sebastian Sardina

Abstract

Classical planning aims to find a sequence of actions, a plan, that maps a starting state into one of the goal states. If a trajectory appears to be leading to the goal, should we prioritise exploring it? Seminal work in goal recognition (GR) has defined GR in terms of a classical planning problem, adopting classical solvers and heuristics to recognise plans. We come full circle, and study the adoption and properties of GR-derived heuristics for seeking solutions to classical planning problems. We propose a new framework for assessing goal intention, which informs a new class of efficiently-computable heuristics. As a proof of concept, we derive two such heuristics, and show that they can already yield improvements for top-scoring classical planners. Our work provides foundational knowledge for understanding and deriving probabilistic intention-based heuristics for planning.

Planning as Goal Recognition: Deriving Heuristics from Intention Models - Extended Version

Abstract

Classical planning aims to find a sequence of actions, a plan, that maps a starting state into one of the goal states. If a trajectory appears to be leading to the goal, should we prioritise exploring it? Seminal work in goal recognition (GR) has defined GR in terms of a classical planning problem, adopting classical solvers and heuristics to recognise plans. We come full circle, and study the adoption and properties of GR-derived heuristics for seeking solutions to classical planning problems. We propose a new framework for assessing goal intention, which informs a new class of efficiently-computable heuristics. As a proof of concept, we derive two such heuristics, and show that they can already yield improvements for top-scoring classical planners. Our work provides foundational knowledge for understanding and deriving probabilistic intention-based heuristics for planning.
Paper Structure (15 sections, 9 theorems, 14 equations, 2 tables)

This paper contains 15 sections, 9 theorems, 14 equations, 2 tables.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Planning Model Notation.
  4. Intention-Based Search
  5. IRPL Model
  6. Framework Properties
  7. Remarks.
  8. Uniform Regimes
  9. Estimating Measures of Goal Intention
  10. General Model of Divergence-based Likelihood
  11. Planning Heuristics
  12. Adding Goal-intentionality to BFWS
  13. Experimental Results
  14. Concluding Remarks
  15. Acknowledgments

Key Result

Lemma 1

Given non-empty $\hat{\mathcal{M}}$ and $\hat{\mathcal{M}}_G$, and any $w$, $\max_{O_e} P(G \mid O_e) \geq P(G \mid O_p)$.

Theorems & Definitions (20)

  • Claim 1
  • Claim 2
  • Lemma 1
  • proof : Proof sketch
  • Theorem 1
  • proof
  • Lemma 2
  • proof : Proof sketch
  • Lemma 3
  • proof
  • ...and 10 more