Multi-Robot Open Adaptive Teaming Across Unseen Environments, Partners, and Scales
2026-07-06 • Robotics
RoboticsArtificial Intelligence
AI summaryⓘ
The authors study how teams of robots can work together smoothly even when their members or environments keep changing unexpectedly. They introduce a new way to model robot teamwork using something called a hypergraphic-form game, which helps robots understand group cooperation beyond just pairs. Based on this, they create an algorithm named HOLA that trains robots to handle different partners and places, not just fixed setups. Their method worked better than others in tests with multiple drones and robotic dogs, and the trained robot teams performed well in real-world experiments without extra adjustments.
multi-robot teamingopen adaptive systemshypergraph gamescoordinationgame theoryreinforcement learningmulti-agent systemstransfer learningdrone roboticsquadruped robots
Authors
Yang Li, Feng Xue, Fan Mo, Yunhao Liu, Jianhong Wang, Ying Wen, Qingrui Zhang, Shaoshuai Mou, Wei Pan
Abstract
Deploying robot teams in the real world requires simultaneous adaptation to unseen environments, unknown partners, and varying team sizes, yet existing approaches often address these challenges in isolation under the closed-world assumption of fixed teammates. We formalize this as open adaptive multi-robot teaming and propose a hypergraphic-form game formulation that captures team-level cooperative relationships beyond pairwise interactions, providing a principled foundation for coordination structure inference when team composition changes dynamically within episodes. Unlike graph neural network architectures, this is a game-theoretic construct for modeling strategic interactions and payoff structures among agents. Building on this formulation, we develop the Hypergraphic Open-ended Learning Algorithm (HOLA), which progressively expands partner and environment diversity during training rather than optimizing for fixed configurations. Evaluated on cooperative pursuit with multi-drone and multi-quadruped platforms, HOLA outperforms all baselines across all three adaptability dimensions. Learned policies transfer directly to physical hardware without fine-tuning, with successful deployments on Crazyflie and Zsibot L1 platforms confirming robust real-world coordination in novel environments with unseen teammates.