ADRA Forum WS

This workshop will explore the design and deployment of trustworthy, multi-agent AI and robotics systems to strengthen global resilience, autonomy, and security. 

As AI, Data, and Robotics (ADR) increasingly converge, decentralized multi-agent ecosystems and particularly in human-AI/robot teaming, which offers significant potential for managing complexity and ensuring continuity in critical domains such as infrastructure, disaster response, defense, and healthcare.

Bringing together leaders from academia, industry, and government, this session will:

-   Unpack ethical, explainable architectures for decentralized multi-agent systems with human oversight.

-    Explore frameworks for trust, safety, and robustness in autonomous human-robot collaboration.

-    Showcase real-world cases applying MAS and robotics for security and resilience.

-    Define actionable strategies for responsible innovation, regulation, and capacity-building.

Participants will engage in keynotes, expert panels, and interactive ideation sessions to co-develop a Global Resilience, Autonomy & Safety Roadmap. This will prioritize pilot projects, ethical guidelines, and skills frameworks to scale trustworthy ADR systems. The goal is to align technical excellence with societal values, enabling resilient autonomy through secure and responsible multi-agent intelligence.

Chair: 

Biography

Assia Belbachir is a Senior Researcher at NORCE Norwegian Research Centre, specializing in autonomous systems, digital twins, and multi-agent/robotics. With a PhD in Artificial Intelligence from INPT–LAAS‑CNRS (France), she has worked across Europe on advanced driver-assistance systems, multi-robot cooperation, and drone technologies. 

Her research focuses on intelligent task planning, reinforcement learning, and real-time decision-making for cooperative autonomous systems.

Presentation title: Learning to See the World from Above with unsupervised multiple interpretations consensus

If machines can interpret the same scene in multiple ways, how do we decide which ‘view of reality’ is the most reliable—and what does that mean for decision-making at scale?

Prof. Marius Leordeanu (PhD, Carnegie Mellon, 2009) is Senior Scientist at NORCE and Full Professor at the Polytechnic University of Bucharest. His research on computer vision and AI has earned over 8,500 citations (h-index 38), top prizes from the Romanian Government (2024) and the Romanian Academy (2014), and major grants including a Google Research Award. He serves as Associate Editor for TPAMI and Area Chair for top AI conferences. His book Unsupervised Learning in Space and Time (Springer, 2020) is a best-seller in the field.

Presentation title: Human-Robot Collaboration: Ensuring Robustness & Trust

How can we build human-robot teams that remain robust and trustworthy in unpredictable, 

high-stress environments where failure is not an option?

Dr. Amit Kumar Pandey, PhD (Robotics & AI), is a globally recognized scientist, innovator, and leader in Robotics and AI, committed to advancing technology for humanity. He is President of Socients AI and Robotics, France, and has held key leadership roles including CTO at Rovial Space, and President, CSO, and CTO at Hanson Robotics, creators of Sophia. Formerly, he was Chief Scientist at SoftBank Robotics Europe, behind Pepper and Nao. Co-founder of GRASS and GIRAI, he has 60+ publications, 100+ talks, and multiple honors, including Top 3 PhDs in Robotics in Europe and global innovation leadership awards.

Presentation title: Integrated Safety and Security Considerations for Critical Industrial Infrastructures

When critical infrastructures face both cyber and physical risks, how do we design AI systems that are not just secure but resilient against cascading failures?

Dr. Wissam Mallouli is CTO of Montimage (France), specializing in cybersecurity, digital twins, and AI for smart manufacturing. He leads Work Package 2 in the COGNIMAN project, applying digital twin technology to enhance flexible production. He holds a Telecommunications Engineering degree (INT) and a PhD in cybersecurity (Telecom & Management SudParis). His expertise spans cyber defense, risk management, and industrial applications. Author of numerous publications, he actively contributes to European research initiatives, driving innovation in cybersecurity and digital transformation for industry.

Presentation title: Ethics and AI governance in the European Union

As artificial intelligence continues to evolve at a rapid pace, how can we ensure that its development 

remains rooted in human values, democratic principles, and accountability? Could you tell us something about the European model?

Dr. Valentina Pagnanelli is a lawyer, Privacy Officer, and TÜV-certified consultant on data governance and compliance. She holds a PhD in Public Law – Data Protection (University of Florence) and has conducted research on GDPR’s impact on public institutions. She also completed advanced studies in data protection and smart cities at Politecnico of Milan. 

 Author of multiple publications, she lectures and trains on law and emerging technologies, particularly in AI regulation and digital governance. She advises companies and governments, speaks at international conferences, and serves on EU project ethics committees.