Intelligent Support for Industrial Processes Using Distributed Artificial Intelligence Technologies
Institution
Ceit Research Center
Role
Research collaborator (federated learning and agentic AI)
Date
Joined in November 2026
My contribution in this area is the implementation of federated learning and agentic AI for industrial environments. These are enabling technologies for a sovereign industrial AI foundry in the Basque Country, based on a federated architecture that supports the design, training, optimization, deployment and governance of advanced AI models directly in industrial environments (edge/fog), ensuring data sovereignty, traceability, multi-plant scalability under corporate control and compliance with the European regulatory framework.
Funded by: CDTI – Spain, national
Intelligent and cyber-secure platform for adaptive optimization in simulated operation of heterogeneous autonomous robots
Institution
Ceit Research Center
Role
Research collaborator
Date
Joined in November 2026
So far I have contributed the state of the art on federated learning applied to autonomous robots through adaptive optimization in the simulated operation of heterogeneous autonomous robots. I have also contributed to data-cybersecurity protocols for network monitoring and traffic management, and developed unsupervised and supervised machine-learning algorithms to detect network-traffic anomalies, including an intrusion detection system (IDS) based on spectral clustering that detects four attack types in industrial-robot network traffic. I co-authored the article “Anomaly Detection Using Machine Learning for Robotics Environments on 5G Networks” in the journal Technologies, which describes the deployed architecture and the anomaly-detection models with process data. In addition, I have developed AI-powered intrusion detection for MQTT-based IoT networks.
My contribution to this EU-funded project has been the development of software sensors for the water line of wastewater treatment plants (WWTPs), used to monitor ammonia concentrations in the aerobic tanks and in the anoxic tank. The software sensor also tracks the maximum nitrification rate. Specifically, an extended Kalman filter is derived that generalizes the Kalman filter to a much broader class of non-Gaussian distributions. The filter has several desirable properties, such as preserving statistical relationships in the prior state and converging to highly accurate observations.
Part of “Smart cybersecurity countermeasures for the network of the future” (CICERO).
Institution
Ceit Research Center
Role
Research collaborator
Date
Joined in June 2024
My contribution is the implementation of spectral clustering for anomaly detection in network traffic. The exponential growth of network traffic has increased network anomalies such as cyber-attacks, network failures and hardware malfunctions, and detecting them is critical to keep computer networks secure and stable. I also use multi-class support vector machines as a supervised algorithm to detect such anomalies early.
Funded by: IRONTEC, Internet y Sistemas sobre GNU/Linux – regional
My contribution, named IRONMAN-BRAIN, is a modular, intelligent auto-scaling system that uses reinforcement learning to enable proactive and adaptive scaling of virtualized and cloud-native network functions. IRONMAN-BRAIN uses a deep Q-network to make auto-scaling decisions within service function chains (SFCs) deployed across OpenStack and Kubernetes. Integration with Open Source MANO (OSM) handles lifecycle management, while SDN components such as OpenDaylight and Open vSwitch provide programmable control of the network infrastructure.
Past research projects
Funded by: Research and Postgraduate Coordination – IIT
The development of methods for cost-effective creation of information and control systems for mechanical and thermal energy equipment (in Czech)
Grant number
SGS10/252/OHK2/3T/12
Role
PhD student / collaborator
Research during my studies
PhD in Mechanical Engineering (2009–2012) Thesis: Lung Tumor Motion Prediction by Neural Networks. Czech Technical University in Prague (CTU), Faculty of Mechanical Engineering.
Visiting student for a research project (scientific research), 28 May – 28 July 2011 Project: Intelligent Prediction Method of Lung Tumor Motion for Highly Accurate Radiation Therapy. Tohoku University, Research Division on Advanced Information Technology, Cyberscience Center (Graduate School of Engineering & Graduate School of Biomedical Engineering).
Master’s thesis (2005–2007) Thesis: “Validation and Verification Automatic Module of Extracted Rules of a Neural-Symbolic Hybrid System”. National Center of Research and Technological Development (CENIDET), Department of Computer Sciences.
Summer of Scientific Research (28 June – 31 August 2004) Project: “Exploitation of Resources of a CMU Camera for its Start-up on a ROMIC Robot”. National Institute of Nuclear Research (ININ), Department of Applied Sciences.
