Automated Planning Domain Inference for Task and Motion Planning

2024-10-21

arXiv.org

AI Summary

Existing Task and Motion Planning (TAMP) approaches rely on manually engineered planning domains, making it difficult to automatically learn action preconditions and effects from few demonstration trajectories—thus limiting generalization and deployment efficiency for complex tasks. This paper proposes an end-to-end framework that, for the first time, enables zero-shot generation of compact, executable planning domains from only a handful of trajectory demonstrations. Our method employs a deep neural network to predict action components and integrates symbolic search for verifiable domain refinement, seamlessly embedding the learned domain into a TAMP system. By jointly optimizing learning-based generalization and logical executability, our approach significantly outperforms behavior cloning baselines across multi-task benchmarks—improving planning success rates, cross-task generalization, and reducing both test-time computational overhead and data requirements.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Multiagent Systems: Multiagent Planning

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Task and motion planning (TAMP) frameworks address long and complex planning problems by integrating high-level task planners with low-level motion planners. However, existing TAMP methods rely heavily on the manual design of planning domains that specify the preconditions and postconditions of all high-level actions. This paper proposes a method to automate planning domain inference from a handful of test-time trajectory demonstrations, reducing the reliance on human design. Our approach incorporates a deep learning-based estimator that predicts the appropriate components of a domain for a new task and a search algorithm that refines this prediction, reducing the size and ensuring the utility of the inferred domain. Our method is able to generate new domains from minimal demonstrations at test time, enabling robots to handle complex tasks more efficiently. We demonstrate that our approach outperforms behavior cloning baselines, which directly imitate planner behavior, in terms of planning performance and generalization across a variety of tasks. Additionally, our method reduces computational costs and data amount requirements at test time for inferring new planning domains.

Problem

• Task and Motion Planning
• Machine Learning
• Robotics

Innovation

• Automated Task and Motion Planning (TAMP)
• Deep Learning Rule Prediction
• Efficient Task Completion