Natural Language as Polices: Reasoning for Coordinate-Level Embodied Control with LLMs

arXiv:2403.13801v1 Announce Type: cross
Abstract: We demonstrate experimental results with LLMs that address robotics action planning problems. Recently, LLMs have been applied in robotics action planning, particularly using a code generation approach that converts complex high-level instructions into mid-level policy codes. In contrast, our approach acquires text descriptions of the task and scene objects, then formulates action planning through natural language reasoning, and outputs coordinate level control commands, thus reducing the necessity for intermediate representation code as policies. Our approach …

abstract arxiv code code generation contrast control cs.ai cs.cl cs.ro embodied experimental language llms natural natural language planning policy reasoning results robotics text type