Generalized Planning as Heuristic Search

TL;DR

This paper introduces a novel heuristic search approach for Generalized Planning, enabling the computation of algorithm-like plans that can generalize across multiple instances, by defining a new solution space and guiding search with heuristics.

Contribution

It presents the first native heuristic search method for GP, including a new solution space and evaluation functions, and implements the BFGP algorithm.

Findings

First heuristic search approach for GP.

Defines a solution space independent of instance size.

Develops heuristic functions guiding the search.

Abstract

Although heuristic search is one of the most successful approaches to classical planning, this planning paradigm does not apply straightforwardly to Generalized Planning (GP). Planning as heuristic search traditionally addresses the computation of sequential plans by searching in a grounded state-space. On the other hand GP aims at computing algorithm-like plans, that can branch and loop, and that generalize to a (possibly infinite) set of classical planning instances. This paper adapts the planning as heuristic search paradigm to the particularities of GP, and presents the first native heuristic search approach to GP. First, the paper defines a novel GP solution space that is independent of the number of planning instances in a GP problem, and the size of these instances. Second, the paper defines different evaluation and heuristic functions for guiding a combinatorial search in our GP solution space. Lastly the paper defines a GP algorithm, called Best-First Generalized Planning (BFGP), that implements a best-first search in the solution space guided by our evaluation/heuristic functions.