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

TL;DR

This paper introduces a framework combining Mixture-of-Agents with Behavior Trees to enhance long-horizon robot task planning, reducing human input and improving execution robustness in complex tasks.

Contribution

It presents a novel integration of MoA-based proxy answering with Behavior Tree generation for structured, efficient, and adaptive robot task execution.

Findings

Reduces human response requirement by ~27%.

Maintains high structural and semantic similarity to fully human-answered plans.

Demonstrates successful real-robot long-horizon task execution with adaptive switching.

Abstract

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.