EFLUX: Elastic Multi-Robot Formation Navigation and Adaptation with Agentic LLMs

2026-07-13

AI Summary

This work addresses the challenges of deadlock and suboptimal trajectories commonly encountered by multi-robot teams operating in narrow or cluttered environments, often due to decoupled behaviors or reliance on handcrafted rules. The authors propose a geometry-aware LLM agent framework that, for the first time, leverages large language models to jointly coordinate continuous formation deformations—such as scaling and shearing—and discrete reconfigurations, including splitting and merging. By integrating structured scene representations with a closed-loop generate–verify–refine mechanism, the system autonomously produces executable waypoints. The approach introduces geometry-driven elastic navigation, enabling adaptive coordination without manual rule design. Extensive simulations and hardware experiments demonstrate that the method significantly reduces deadlock occurrences and navigation failures, achieving safe, continuous, and efficient multi-robot formation coordination in constrained spaces.

Abstract

Multi-robot teams operating in confined or cluttered environments must adapt both their formation geometry and group topology to navigate through complex obstacles. This adaptation requires two complementary behaviors: deformation, where the team continuously reshapes its geometry while remaining connected, and reconfiguration, where robots split into subgroups or merge back into a single formation. Existing methods often model these behaviors independently, connect them through handcrafted rules, or lack explicit geometric criteria for determining when each behavior should be invoked. However, challenging environments may require online changes in formation shape, connectivity, and effective team composition, making decoupled or rule-based approaches prone to suboptimal trajectories and deadlock. We propose EFLUX, a geometry-grounded LLM agentic framework for automatic and elastic multi-robot formation navigation. EFLUX extracts a structured scene representation and uses an LLM to reason jointly over both deformation actions, such as scaling and shearing, and reconfiguration actions, such as splitting and merging. These strategies are then translated into executable per-robot waypoints through a closed-loop generation, verification, and correction pipeline. Simulation and hardware experiments show that EFLUX enables safe, continuous, and elastic formation navigation in constrained environments, reducing deadlock and navigation failures compared with baselines while maintaining coherent multi-robot coordination.

Problem

• multi-robot formation
• formation deformation
• formation reconfiguration
• geometric adaptation
• navigation in cluttered environments

Innovation

• elastic formation navigation
• agentic LLMs
• multi-robot reconfiguration
• geometry-grounded reasoning
• closed-loop trajectory generation