Subteaming and Adaptive Formation Control for Coordinated Multi-Robot Navigation

2025-09-19

AI Summary

Maintaining fixed formations of multi-robot systems in complex, confined environments—such as narrow corridors—remains challenging due to dynamic obstacles and spatial constraints. To address this, we propose STAF, a hierarchical end-to-end adaptive formation navigation framework. STAF uniquely unifies three levels: (i) high-level topologically aware dynamic grouping via deep graph partitioning; (ii) mid-level cooperative sub-team navigation modeled using graph neural networks to capture inter-subteam interactions; and (iii) low-level reactive obstacle avoidance controlled through hierarchical reinforcement learning that jointly optimizes policies across all layers. Extensive simulations and real-robot experiments across diverse indoor and outdoor scenarios demonstrate that STAF significantly improves navigation success rate (+32.7%) and robustness, enables on-the-fly formation reconfiguration, and ensures seamless coordination. This work establishes a new paradigm for flexible, reliable multi-robot navigation in geometrically constrained spaces.

Technology Category

• Intelligent Robots: Multi-Robot Systems
• Multiagent Systems: Adversarial Agents
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Coordinated multi-robot navigation is essential for robots to operate as a team in diverse environments. During navigation, robot teams usually need to maintain specific formations, such as circular formations to protect human teammates at the center. However, in complex scenarios such as narrow corridors, rigidly preserving predefined formations can become infeasible. Therefore, robot teams must be capable of dynamically splitting into smaller subteams and adaptively controlling the subteams to navigate through such scenarios while preserving formations. To enable this capability, we introduce a novel method for SubTeaming and Adaptive Formation (STAF), which is built upon a unified hierarchical learning framework: (1) high-level deep graph cut for team splitting, (2) intermediate-level graph learning for facilitating coordinated navigation among subteams, and (3) low-level policy learning for controlling individual mobile robots to reach their goal positions while avoiding collisions. To evaluate STAF, we conducted extensive experiments in both indoor and outdoor environments using robotics simulations and physical robot teams. Experimental results show that STAF enables the novel capability for subteaming and adaptive formation control, and achieves promising performance in coordinated multi-robot navigation through challenging scenarios. More details are available on the project website: https://hcrlab.gitlab.io/project/STAF.

Problem

• Enabling robot teams to dynamically split into smaller subteams
• Adaptively controlling formations in complex navigation scenarios
• Maintaining coordinated navigation through challenging environments

Innovation

• Hierarchical deep graph cut for team splitting
• Graph learning for coordinated subteam navigation
• Policy learning for individual robot control