Ismael Etxeberria‐Agiriano

About

Ismael Etxeberria-Agiriano is a researcher whose work spans robotics, Cyber-Physical Systems (CPS), and multi-agent reinforcement learning, with a particular focus on building robust, modular architectures for intelligent and autonomous systems. His most influential contribution, a modular CPS architecture design based on ROS and Docker (2016, 37 citations), established practical frameworks for integrating open-source tools into scalable robotic systems. Complementing this, his research on multi-robot coordination through Distributed Round-Robin Q-Learning (2015, 22 citations) formally proved convergence properties of a novel MARL algorithm, addressing fundamental challenges such as the curse of dimensionality and environment non-stationarity. His work on reinforcement learning with safe veto policies (2015, 13 citations) further advanced safe and adaptive control of complex linked robotic systems. Beyond robotics intelligence, Etxeberria-Agiriano has made consistent contributions to industrial middleware and communication standards, including guidelines for high-performance factory automation using DDS (2013, 10 citations). More recently, he has explored deep learning-based robot localization using natural environmental features, reflecting an evolving interest in vision-driven autonomy. Across his career, his work bridges theoretical rigor with practical engineering, making him a valuable contributor to both academic research and applied robotics education.

Research Focus

Key Achievements

Top Papers

1. 1
   A modular CPS architecture design based on ROS and Docker

   37 citations · 2016
2. 2
   Learning Multirobot Hose Transportation and Deployment by Distributed Round-Robin Q-Learning

   22 citations · 2015
3. 3
   Reinforcement Learning endowed with safe veto policies to learn the control of Linked-Multicomponent Robotic Systems

   13 citations · 2015
4. 4
   Designing High Performance Factory Automation Applications on Top of DDS

   10 citations · 2013
5. 5
   Building a CPS as an Educational Challenge

   6 citations · 2014
6. 6
   Towards a Generic Architecture for Building Modular CPS as Applied to Mobile Robotics

   4 citations · 2016
7. 7
   Efficient Robot Localization Through Deep Learning-Based Natural Fiduciary Pattern Recognition

   2 citations · 2025
8. 8
   Flexible, modular, standard, free and affordable model for CPS control applied to mobile robotics

   2 citations · 2015
9. 9
   A CORBA Wrapper for Applications with Multiple Robots

   2 citations · 2011
10. 10
    ISO26262 SEooC Compliance of a ROS Based Architecture

    2 citations · 2017

Key Collaborators

Contact & Links