Physically Ground Commonsense Knowledge for Articulated Object Manipulation with Analytic Concepts

Jianhua Sun, Jiude Wei, Yuxuan Li, Cewu Lu

Keywords: Articulated Object Manipulation, Robotics, Neural Symbolic, Physical Concepts

TL;DR: We propose analytic concepts as knowledge representation in physical form and a pipeline to ground semantic-level knowledge inferred by LLMs in the physical world through concepts, providing concrete guidance for articulated object manipulation.

Abstract: We human rely on a wide range of commonsense knowledge to interact with an extensive number and categories of objects in the physical world. Likewise, such commonsense knowledge is also crucial for robots to successfully develop generalized object manipulation skills. While recent advancements in Multi-modal Large Language Models (MLLMs) have showcased their impressive capabilities in acquiring commonsense knowledge and conducting commonsense reasoning, effectively grounding this semantic-level knowledge produced by MLLMs to the physical world to thoroughly guide robots in generalized articulated object manipulation remains a challenge that has not been sufficiently addressed. To this end, we introduce analytic concepts, procedurally defined upon mathematical symbolism that can be directly computed and simulated by machines. By leveraging the analytic concepts as a bridge between the semantic-level knowledge inferred by LLMs and the physical world where real robots operate, we are able to figure out the knowledge of object structure and functionality with physics-informed representations, and then use the physically grounded knowledge to instruct robot control policies for generalized, interpretable and accurate articulated object manipulation. Extensive experiments in both simulation and real-world environments demonstrate the superiority of our approach. Please refer to the appendix for more details, and our codes will be made publicly available.

Primary Area: applications to robotics, autonomy, planning

Submission Number: 1361