Learning to reason over scene graphs: a case study of finetuning GPT-2 into a robot language model for grounded task planning

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 specifi...

Full description

Bibliographic Details
Main Authors: Georgia Chalvatzaki, Ali Younes, Daljeet Nandha, An Thai Le, Leonardo F. R. Ribeiro, Iryna Gurevych
Format: Article
Language:English
Published: Frontiers Media S.A. 2023-08-01
Series:Frontiers in Robotics and AI
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/frobt.2023.1221739/full
Description
Summary: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.
ISSN:2296-9144