From Dialogue to Execution: Mixture-of-Agents Assisted Interactive Planning for Behavior Tree-Based Long-Horizon Robot Execution

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Suzuki, Kanata, Hori, Kazuki, Miyoshi, Haruka, Kurita, Shuhei, Ogata, Tetsuya
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915825861001216
author Suzuki, Kanata
Hori, Kazuki
Miyoshi, Haruka
Kurita, Shuhei
Ogata, Tetsuya
author_facet Suzuki, Kanata
Hori, Kazuki
Miyoshi, Haruka
Kurita, Shuhei
Ogata, Tetsuya
contents Interactive task planning with large language models (LLMs) enables robots to generate high-level action plans from natural language instructions. However, in long-horizon tasks, such approaches often require many questions, increasing user burden. Moreover, flat plan representations become difficult to manage as task complexity grows. We propose a framework that integrates Mixture-of-Agents (MoA)-based proxy answering into interactive planning and generates Behavior Trees (BTs) for structured long-term execution. The MoA consists of multiple LLM-based expert agents that answer general or domain-specific questions when possible, reducing unnecessary human intervention. The resulting BT hierarchically represents task logic and enables retry mechanisms and dynamic switching among multiple robot policies. Experiments on cocktail-making tasks show that the proposed method reduces human response requirements by approximately 27% while maintaining structural and semantic similarity to fully human-answered BTs. Real-robot experiments on a smoothie-making task further demonstrate successful long-horizon execution with adaptive policy switching and recovery from action failures. These results indicate that MoA-assisted interactive planning improves dialogue efficiency while preserving execution quality in real-world robotic tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2603_01113
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Dialogue to Execution: Mixture-of-Agents Assisted Interactive Planning for Behavior Tree-Based Long-Horizon Robot Execution
Suzuki, Kanata
Hori, Kazuki
Miyoshi, Haruka
Kurita, Shuhei
Ogata, Tetsuya
Robotics
Interactive task planning with large language models (LLMs) enables robots to generate high-level action plans from natural language instructions. However, in long-horizon tasks, such approaches often require many questions, increasing user burden. Moreover, flat plan representations become difficult to manage as task complexity grows. We propose a framework that integrates Mixture-of-Agents (MoA)-based proxy answering into interactive planning and generates Behavior Trees (BTs) for structured long-term execution. The MoA consists of multiple LLM-based expert agents that answer general or domain-specific questions when possible, reducing unnecessary human intervention. The resulting BT hierarchically represents task logic and enables retry mechanisms and dynamic switching among multiple robot policies. Experiments on cocktail-making tasks show that the proposed method reduces human response requirements by approximately 27% while maintaining structural and semantic similarity to fully human-answered BTs. Real-robot experiments on a smoothie-making task further demonstrate successful long-horizon execution with adaptive policy switching and recovery from action failures. These results indicate that MoA-assisted interactive planning improves dialogue efficiency while preserving execution quality in real-world robotic tasks.
title From Dialogue to Execution: Mixture-of-Agents Assisted Interactive Planning for Behavior Tree-Based Long-Horizon Robot Execution
topic Robotics
url https://arxiv.org/abs/2603.01113