Beyond Predefined Actions: Integrating Behavior Trees and Dynamic Movement Primitives for Robot Learning from Demonstration

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Domínguez, David Cáceres, Schaffernicht, Erik, Stoyanov, Todor
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916735319277568
author Domínguez, David Cáceres
Schaffernicht, Erik
Stoyanov, Todor
author_facet Domínguez, David Cáceres
Schaffernicht, Erik
Stoyanov, Todor
contents Interpretable policy representations like Behavior Trees (BTs) and Dynamic Motion Primitives (DMPs) enable robot skill transfer from human demonstrations, but each faces limitations: BTs require expert-crafted low-level actions, while DMPs lack high-level task logic. We address these limitations by integrating DMP controllers into a BT framework, jointly learning the BT structure and DMP actions from single demonstrations, thereby removing the need for predefined actions. Additionally, by combining BT decision logic with DMP motion generation, our method enhances policy interpretability, modularity, and adaptability for autonomous systems. Our approach readily affords both learning to replicate low-level motions and combining partial demonstrations into a coherent and easy-to-modify overall policy.
format Preprint
id arxiv_https___arxiv_org_abs_2505_08625
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beyond Predefined Actions: Integrating Behavior Trees and Dynamic Movement Primitives for Robot Learning from Demonstration
Domínguez, David Cáceres
Schaffernicht, Erik
Stoyanov, Todor
Robotics
Interpretable policy representations like Behavior Trees (BTs) and Dynamic Motion Primitives (DMPs) enable robot skill transfer from human demonstrations, but each faces limitations: BTs require expert-crafted low-level actions, while DMPs lack high-level task logic. We address these limitations by integrating DMP controllers into a BT framework, jointly learning the BT structure and DMP actions from single demonstrations, thereby removing the need for predefined actions. Additionally, by combining BT decision logic with DMP motion generation, our method enhances policy interpretability, modularity, and adaptability for autonomous systems. Our approach readily affords both learning to replicate low-level motions and combining partial demonstrations into a coherent and easy-to-modify overall policy.
title Beyond Predefined Actions: Integrating Behavior Trees and Dynamic Movement Primitives for Robot Learning from Demonstration
topic Robotics
url https://arxiv.org/abs/2505.08625