CLIMB: Language-Guided Continual Learning for Task Planning with Iterative Model Building

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Byrnes, Walker, Bogdanovic, Miroslav, Balakirsky, Avi, Balakirsky, Stephen, Garg, Animesh
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910654634393600
author Byrnes, Walker
Bogdanovic, Miroslav
Balakirsky, Avi
Balakirsky, Stephen
Garg, Animesh
author_facet Byrnes, Walker
Bogdanovic, Miroslav
Balakirsky, Avi
Balakirsky, Stephen
Garg, Animesh
contents Intelligent and reliable task planning is a core capability for generalized robotics, requiring a descriptive domain representation that sufficiently models all object and state information for the scene. We present CLIMB, a continual learning framework for robot task planning that leverages foundation models and execution feedback to guide domain model construction. CLIMB can build a model from a natural language description, learn non-obvious predicates while solving tasks, and store that information for future problems. We demonstrate the ability of CLIMB to improve performance in common planning environments compared to baseline methods. We also develop the BlocksWorld++ domain, a simulated environment with an easily usable real counterpart, together with a curriculum of tasks with progressing difficulty for evaluating continual learning. Additional details and demonstrations for this system can be found at https://plan-with-climb.github.io/ .
format Preprint
id arxiv_https___arxiv_org_abs_2410_13756
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CLIMB: Language-Guided Continual Learning for Task Planning with Iterative Model Building
Byrnes, Walker
Bogdanovic, Miroslav
Balakirsky, Avi
Balakirsky, Stephen
Garg, Animesh
Robotics
Artificial Intelligence
Machine Learning
Intelligent and reliable task planning is a core capability for generalized robotics, requiring a descriptive domain representation that sufficiently models all object and state information for the scene. We present CLIMB, a continual learning framework for robot task planning that leverages foundation models and execution feedback to guide domain model construction. CLIMB can build a model from a natural language description, learn non-obvious predicates while solving tasks, and store that information for future problems. We demonstrate the ability of CLIMB to improve performance in common planning environments compared to baseline methods. We also develop the BlocksWorld++ domain, a simulated environment with an easily usable real counterpart, together with a curriculum of tasks with progressing difficulty for evaluating continual learning. Additional details and demonstrations for this system can be found at https://plan-with-climb.github.io/ .
title CLIMB: Language-Guided Continual Learning for Task Planning with Iterative Model Building
topic Robotics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2410.13756