CaPo: Cooperative Plan Optimization for Efficient Embodied Multi-Agent Cooperation

TL;DR

CaPo introduces a two-phase cooperative plan optimization framework for LLM-based embodied agents, significantly improving long-term strategic cooperation and task efficiency in complex multi-agent scenarios.

Contribution

The paper proposes CaPo, a novel method with meta-plan generation and progress-adaptive execution, enhancing cooperation among embodied agents over prior approaches.

Findings

Achieves higher task completion rates in multi-agent tasks.

Reduces redundant actions through dynamic plan adjustments.

Outperforms state-of-the-art methods in efficiency and success rate.

Abstract

In this work, we address the cooperation problem among large language model (LLM) based embodied agents, where agents must cooperate to achieve a common goal. Previous methods often execute actions extemporaneously and incoherently, without long-term strategic and cooperative planning, leading to redundant steps, failures, and even serious repercussions in complex tasks like search-and-rescue missions where discussion and cooperative plan are crucial. To solve this issue, we propose Cooperative Plan Optimization (CaPo) to enhance the cooperation efficiency of LLM-based embodied agents. Inspired by human cooperation schemes, CaPo improves cooperation efficiency with two phases: 1) meta-plan generation, and 2) progress-adaptive meta-plan and execution. In the first phase, all agents analyze the task, discuss, and cooperatively create a meta-plan that decomposes the task into subtasks with detailed steps, ensuring a long-term strategic and coherent plan for efficient coordination. In the second phase, agents execute tasks according to the meta-plan and dynamically adjust it based on their latest progress (e.g., discovering a target object) through multi-turn discussions. This progress-based adaptation eliminates redundant actions, improving the overall cooperation efficiency of agents. Experimental results on the ThreeDworld Multi-Agent Transport and Communicative Watch-And-Help tasks demonstrate that CaPo achieves much higher task completion rate and efficiency compared with state-of-the-arts.The code is released at https://github.com/jliu4ai/CaPo.