Transferability Through Cooperative Competitions

Rodrigo Serra; Carlos Azevedo; André Silva; Kevin Alcedo; Quentin Rouxel; Peter So; Alejandro Suarez; Alin Albu-Schäeffer

cs.ROMar 29, 2026

Transferability Through Cooperative Competitions

Authors: Rodrigo Serra, Carlos Azevedo, André Silva, Kevin Alcedo, Quentin Rouxel, Peter So, Alejandro Suarez, Alin Albu-Schäeffer,

Organizations: Institute for Systems and Robotics / LARSyS, Instituto Superior Técnico, University of Lisbon · Laboratory for Automatics and Systems, Instituto Pedro Nunes · INRIA, CNRS, Université de Lorraine, Nancy, France · Munich Institute of Robotics and Machine Intelligence, TUM, Germany ·

Abstract

This paper presents a novel framework for cooperative robotics competitions (coopetitions) that promote the transferability and composability of robotics modules, including software, hardware, and data, across heterogeneous robotic systems. The framework is designed to incentivize collaboration between teams through structured task design, shared infrastructure, and a royalty-based scoring system. As a case study, the paper details the implementation and outcomes of the first euROBIN Coopetition, held under the European Robotics and AI Network (euROBIN), which featured fifteen robotic platforms competing across Industrial, Service, and Outdoor domains. The study highlights the practical challenges of achieving module reuse in real-world scenarios, particularly in terms of integration complexity and system compatibility. It also examines participant performance, integration behavior, and team feedback to assess the effectiveness of the framework. The paper concludes with lessons learned and recommendations for future coopetitions, including improveme

Figures & tables

Fig. 1: Group photo of the euROBIN Coopetition held in Nancy, France, in November 2024.

Fig. 2: Coopetition design comprised of two stages, and tracked by a coopetition service tailored to its needs.

Team	Institution	Robot / End Effector / Sensors
TUM-MIRMI	TUM	Franka FR3 / Franka Hand / Realsense D435i
HCR Team	J. Stefan Inst.	Franka FR3 / Franka Hand / Realsense D435
Tecnalia Flexbotics	Tecnalia	Nextage / Pneumatic EE / Stereo Vision
Fraunhofer IPA	Fraunhofer IPA	UR5e / WSG 50-110 / Realsense D435
OSCAR	CEA	UR10e / Robotiq 2F-140 / Realsense D435, Phoxi M, FT Sensor

TABLE I: IRL teams, institutions, and robots.

Fig. 4: Teams participating in the euROBIN Coopetition across the Industrial (left), Service (center), and Outdoor Robots Leagues (right images). Image credits: © Inria / Photos B. Fourrier and Justine Galet.

Team	Institution	Robot
Alter-Ego	University of Pisa	Alter-Ego
INRIA	INRIA	Inria Tiago Dual
SUSR Team	ISIR, Sorbonne Université	TIAGo
DLR	DLR	Rollin’ Justin
KIT	H2T	ARMAR-7
SocRob@Home	ISR (IST)	TIAGo

TABLE II: Teams and robot platforms participating in the SRL (top), and task-relevant kitchen, location, and object constraints for the Multi-Functional Service Robot Challenge (bottom).

Team	Institution	Robot
USE	GRVC Robotics Lab	Dual Arm Aerial Manipulation Robot & LiCAS A1
RSL	RSL ETH Zurich	Transformer Robot & Leva
IIT	Istituto Italiano Di Tecnologia	Centauro

TABLE III: ORL teams, institutions, and robots.

Fig. 5: Module Transferability Social Graphs during the Marketplace (Pre-Event) and On-Site Coopetition (Post-Event) phases. The nodes represent the teams, and the edges indicate the direction of module integration. For example, “INRIA → IIT” indicates that INRIA developed a module, which was then integrated by IIT. The node sizes are proportional to the amount of royalties the team receives, with larger nodes indicating more royalties. The node color corresponds to the league the team participated in. The edge color represents the general category of the modules as illustrated in the image.

Robotics

Robotics studies the design and control of machines that perceive, move, and act in physical environments.

See all 747 papers

Cooperative-Competitive Team Play of Real-World Craft Robots
Feb 24, 2026Rui Zhao, Xihui Li, Yizheng Zhang +6Cooperative MARL Multi-Robot Coordination
The authors are with Tencent Robotics X Laboratory, Shenzhen, Guangdong, China. · The author is with Tsinghua University, Shenzhen, Guangdong, China.
Coopetition-Gym v1: A Formally Grounded Platform for Mixed-Motive Multi-Agent Reinforcement Learning under Strategic Coopetition
May 3, 2026Vik Pant, Eric YuRL Benchmarks Game Theory
Faculty of Information University of Toronto · Faculty of Information and Department of Computer Science University of Toronto
Learning Altruistic Collaboration in Heterogeneous Multi-Team Systems
May 20, 2026Riwa Karam, Ruoyu Lin, Brooks A. Butler +1Multi-Agent Collaboration Multi-Robot Coordination
Samueli School of Engineering, University of California, Irvine, Irvine, CA, 92697, USA. · University of North Carolina at Chapel Hill, Chapel Hill, NC, 27599, USA.

Feb 24, 2026J/K move · Enter open · S save

Cooperative-Competitive Team Play of Real-World Craft Robots

Rui Zhao, Xihui Li, Yizheng Zhang, Yuzhen Liu+5

The authors are with Tencent Robotics X Laboratory, Shenzhen, Guangdong, China. · The author is with Tsinghua University, Shenzhen, Guangdong, China.

Cooperative MARL Multi-Robot Coordination Sim-to-Real Transfer Distributed RL

Open paper AskPDF arXiv

Multi-agent deep Reinforcement Learning (RL) has made significant progress in developing intelligent game-playing agents in recent years. However, the efficient training of collective robots using multi-agent RL and the transfer of learned policies to real-world applications remain open research questions. In this work, we first develop a comprehensive robotic system, including simulation, distributed learning framework, and physical robot components. We then propose and evaluate reinforcement learning techniques designed for efficient training of cooperative and competitive policies on this platform. To address the challenges of multi-agent sim-to-real transfer, we introduce Out of Distribution State Initialization (OODSI) to mitigate the impact of the sim-to-real gap. In the experiments, OODSI improves the Sim2Real performance by 20%. We demonstrate the effectiveness of our approach through experiments with a multi-robot car competitive game and a cooperative task in real-world settings.

Abstract

Robotics

Explore similar work

Cooperative-Competitive Team Play of Real-World Craft Robots

Coopetition-Gym v1: A Formally Grounded Platform for Mixed-Motive Multi-Agent Reinforcement Learning under Strategic Coopetition

Learning Altruistic Collaboration in Heterogeneous Multi-Team Systems