RecipeMasterLLM: Revisiting RoboEarth in the Era of Large Language Models

2025-12-19

AI Summary

To address the bottlenecks of manual effort, low efficiency, and poor scalability in knowledge graph construction for cloud robotics, this paper proposes the first automated framework that deeply integrates large language models (LLMs) into the RoboEarth knowledge acquisition pipeline. Methodologically, it combines domain-specific fine-tuning with retrieval-augmented generation (RAG) to enable end-to-end translation of natural-language instructions into OWL action ontologies compliant with RoboEarth standards, and constructs standardized RDF/OWL knowledge graphs supporting cross-platform knowledge sharing. Key contributions include: (1) the first LLM-driven paradigm for automatic action ontology generation, overcoming limitations of traditional manual modeling; (2) zero-shot action planning across diverse robotic tasks; and (3) significant improvements in action ontology accuracy and generalization, with knowledge graph construction efficiency enhanced by two orders of magnitude over manual approaches.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

RoboEarth was a pioneering initiative in cloud robotics, establishing a foundational framework for robots to share and exchange knowledge about actions, objects, and environments through a standardized knowledge graph. Initially, this knowledge was predominantly hand-crafted by engineers using RDF triples within OWL Ontologies, with updates, such as changes in an object's pose, being asserted by the robot's control and perception routines. However, with the advent and rapid development of Large Language Models (LLMs), we believe that the process of knowledge acquisition can be significantly automated. To this end, we propose RecipeMasterLLM, a high-level planner, that generates OWL action ontologies based on a standardized knowledge graph in response to user prompts. This architecture leverages a fine-tuned LLM specifically trained to understand and produce action descriptions consistent with the RoboEarth standardized knowledge graph. Moreover, during the Retrieval-Augmented Generation (RAG) phase, environmental knowledge is supplied to the LLM to enhance its contextual understanding and improve the accuracy of the generated action descriptions.

Problem

• Automates knowledge acquisition for robots using LLMs
• Generates OWL action ontologies from user prompts
• Enhances action description accuracy with environmental context

Innovation

• LLM generates OWL ontologies from user prompts
• Fine-tuned LLM aligns with RoboEarth knowledge graph
• RAG integrates environmental knowledge for accuracy