An Accurate HDDL Domain Learning Algorithm from Partial and Noisy Observations

TL;DR

This paper introduces HierAMLSI, a novel grammar induction-based algorithm that accurately learns HTN planning domain models, including actions and methods, from noisy and partial observations, addressing a key challenge in automated planning.

Contribution

HierAMLSI is the first approach capable of learning both actions and HTN methods with high accuracy from incomplete and noisy data.

Findings

Achieves high accuracy in learning action models from noisy data.

Successfully learns HTN methods with partial observations.

Outperforms existing methods in noisy environments.

Abstract

The Hierarchical Task Network ({\sf HTN}) formalism is very expressive and used to express a wide variety of planning problems. In contrast to the classical {\sf STRIPS} formalism in which only the action model needs to be specified, the {\sf HTN} formalism requires to specify, in addition, the tasks of the problem and their decomposition into subtasks, called {\sf HTN} methods. For this reason, hand-encoding {\sf HTN} problems is considered more difficult and more error-prone by experts than classical planning problem. To tackle this problem, we propose a new approach (HierAMLSI) based on grammar induction to acquire {\sf HTN} planning domain knowledge, by learning action models and {\sf HTN} methods with their preconditions. Unlike other approaches, HierAMLSI is able to learn both actions and methods with noisy and partial inputs observation with a high level or accuracy.