In the field of cognitive robotics, while commonsense knowledge suffices for virtual agents, embodied robots require grounded and semantically rich representations of their environments and physical embodiments when interacting with humans. Ontologies are essential for enabling explainable reasoning, especially during continuous knowledge updates. However, the manual construction of ontologies remains a bottleneck.
We present a preliminary approach for the automatic generation of robot semantic abstractions by transforming Unified Robot Description Format (URDF) models into populated ontologies. Although URDF files provide structural and kinematic descriptions, their identifiers often necessitate commonsense interpretation to recover meaningful semantics, a task at which Large Language Models (LLMs) excel.
Our pipeline leverages LLMs to infer semantic relationships by prompting them with concepts from an existing ontology, ensuring that the final classification aligns with the formal model. To improve reliability, the pipeline combines majority voting across multiple LLM queries along with syntactic and schema-level validation to ensure that generated outputs conform to the expected representation format and ontology constraints.
We evaluate the approach on multiple robot descriptions and discuss the generated abstractions. Initial results indicate that the proposed method effectively bridges the gap between low-level robot descriptions and the structured, grounded knowledge representations required for human-robot interaction.
Blogger's Review: This paper illustrates how to automate the construction of robot ontologies using LLMs. This innovative approach not only addresses the bottleneck of manual construction but also significantly enhances the semantic understanding of robots in human interaction, promising broad applications. The accuracy of semantic representation is crucial in robotics, and the introduction of LLMs marks a substantial technological leap.