AUTHOR=Chalvatzaki Georgia , Younes Ali , Nandha Daljeet , Le An Thai , Ribeiro Leonardo F. R. , Gurevych Iryna TITLE=Learning to reason over scene graphs: a case study of finetuning GPT-2 into a robot language model for grounded task planning JOURNAL=Frontiers in Robotics and AI VOLUME=10 YEAR=2023 URL=https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2023.1221739 DOI=10.3389/frobt.2023.1221739 ISSN=2296-9144 ABSTRACT=

Long-horizon task planning is essential for the development of intelligent assistive and service robots. In this work, we investigate the applicability of a smaller class of large language models (LLMs), specifically GPT-2, in robotic task planning by learning to decompose tasks into subgoal specifications for a planner to execute sequentially. Our method grounds the input of the LLM on the domain that is represented as a scene graph, enabling it to translate human requests into executable robot plans, thereby learning to reason over long-horizon tasks, as encountered in the ALFRED benchmark. We compare our approach with classical planning and baseline methods to examine the applicability and generalizability of LLM-based planners. Our findings suggest that the knowledge stored in an LLM can be effectively grounded to perform long-horizon task planning, demonstrating the promising potential for the future application of neuro-symbolic planning methods in robotics.