Volker Gloeckle, Senior Vice President of Integrated Automation at SICK AG.

At 54 years old, Volker Gloeckle brings a wealth of experience to the company. After completing an apprenticeship and further training as a technician in electronic engineering, Volker Gloeckle began a professional career as a Field Engineer in the industrial environment at Siemens AG.

Since joining SICK AG in 1995, Volker Gloeckle has held several key positions, including Project Manager in Corporate Logistics, Head of Logistics Maintenance, and Technical Industry Manager for Intralogistics.

Later, Volker Gloeckle served as Senior Vice President Technical Industry Competence & Innovation before taking on the current role as Senior Vice President Integrated Automation.

Federico will introduce the challenges of setting safety requirements for dynamically stable robots, and determining validation criteria, for complex behaviors. When used in environments designed for humans, legged robots can display extraordinary mobility but must also face difficult constraints. On top of it, the simultaneous coordination of limbs for locomotion, balancing and manipulation must be combined with perception abilities at the forefront of the state of the art. Humanoids are poised to represent the next generation of enablers in manufacturing, undertaking flexible tasks in shared space. How does the industrial domain build trust in these types of robots?

Federico Vicentini is Head of Product Safety at Boston Dynamics helping foster robot safety technology, risk-based methodology, and compliance for legged robots, humanoids and mobile manipulators. Previously, he was a Researcher at the Italian National Research Council with interests in human-robot collaboration. After a decade of serving as an Expert in national and international safety standardization committees, he's been recently appointed Convenor of ISO TC 299 WG 12 for Safety of Dynamically Stable industrial mobile robots, those that need an active control of stability. He holds a PhD in Mechanical Engineering from Politecnico di Milano (Italy) and co-authored 60+ scholarly publications and technical reviews.

Trust is all we need. This project is part of a transformation journey - from conventional, compliance-driven safety to dynamic, adaptive, and intelligent safety aligned with end-users’ operational objectives. It aims to establish a trustworthiness framework for intelligent automation solutions by integrating safety, OT security, and AI/ML to enable runtime cyber-physical risk evaluation.

Robert Puto holds an M.Sc. in Electronic Engineering from Politecnico di Torino, Italy, where he conducted research on Uncertainty Propagation in Bayesian Expert Systems. He also earned a Diploma in International Business Management from the China-Europe International Business School (CEIBS) in Shanghai. With over 30 years of professional experience, including 22 years in international leadership roles, Robert combines technical expertise with strategic business acumen. His hands-on leadership has been pivotal in advancing comprehensive technical services for emerging technologies such as Photovoltaics (PV), Energy Storage Systems, Industrial Machinery, Robotics, and Transportation. Robert pioneered and led the "Adaptive Safety & Security System (AS3)" – a global project and winner of TÜV SÜD’s 2023 Innovation Award. This patented solution enables runtime risk evaluation at the system level, with pilot projects currently underway.

Driven by his vision for transformative safety in the era of AI and quantum computing, Robert has established a cross-disciplinary team of experts in safety, industrial cybersecurity, robotics, and AI. Their mission is to develop a trustworthiness framework for intelligent automation solutions that integrates safety, OT security, and AI/ML, aligned with end-users’ operational objectives.

Autonomous and robotic systems rely on several technologies that cannot currently be fully verified or certified due to their complexity and inherent uncertainties. These technologies—including AI-based functions for perception, planning, and control—offer high performance but low assurance levels. One solution is to implement runtime safety monitors designed to keep the system within acceptable operational states, even in the presence of failures or adversarial situations. This presentation will explore the main challenges and current solutions for designing such monitors for autonomous systems.

Jérémie Guiochet is a professor of computer science at the University of Toulouse. His research is conducted at LAAS-CNRS. He received a Ph.D. in computer science and robotics in 2003 from INSA Toulouse, France. He teaches courses in Computer Science, Industrial Computing, Dependability, and Risk Management. His research, carried out within the TRUST group (Trustworthy Systems) at LAAS-CNRS, focuses on safety assessment, fault removal, and fault tolerance in safety-critical autonomous systems.

The rapid evolution of reconfigurable manufacturing systems (RMS) demands a paradigm shift in functional safety, moving beyond static compliance toward dynamic, adaptive assurance. Traditional certification processes, which often rely on manual, fixed-configuration assessments, can hinder the agility and economic benefits that RMS offer.

This session explores how to reconcile functional safety standards with the inherent dynamism of modern factories by leveraging Digital Twins for virtual reconfiguration and automated risk assessment. It examines AI-generated virtual scenarios to identify both advantages and weaknesses, with a specific focus on establishing quantitative risk levels and exposure data through systematic empirical evidence. It will propose factorial experimental designs on physical testbeds, to continuously refine the digital twin’s fidelity and assess viability.

The aim is toward a certifiable, data-driven safety process that bridges the critical sim-to-real gap to verify assumption of use, unlocking agile, safe, and certifiable manufacturing for Industry 5.0 and beyond.

Dr. Giulio Corradi AMD Fellow - is AMD’s Industrial Vision & Healthcare Principal Architect in the AECG business unit. He is based in Munich, Germany and has 30 years of experience of management, software engineering and development of ASICs and FPGA in industrial automation, robotics and medical systems specifically in the field of control, communication, machine intelligence, DSP algorithms, functional safety, ultrasound and MRI equipment. From 2000-2005 he headed the Train Communication IEC61375 Conformance Test standard, and he has been a member of CENELEC WG14 for railway functional safety. Giulio led the safety certification of AMD´s FPGA tools compliant to IEC61508, ISO26262, and ISO13849. His recent work centers on making sure AMD succeeds in business with robotics and AI at the edge within real-time, mixed-criticality safe systems. He is a regular keynote speaker at IEEE industrial and real-time conferences, ROSCon, and other leading conferences and events.

Industrial applications of robotics are faced with increasing demands for flexibility and adaptability. This can arise, for example, from small lot sizes and numerous product variants as well as from dynamic and changing operating environments, as may occur in logistics operations. Numerous approaches enable such accommodation of the functional control of fixed and mobile robots. Classical implementations of safety of machinery, on the other hand, are characterized by limited flexibility and comparatively simple logical operations. While this meets the need for verifiable implementation of safety functions, it may fall short of fully supporting the desired adaptability of future robotic installations.

Trends are apparent that seek to augment present-day automation of manufacturing and logistics operations with increasing aspects of autonomy. Since this introduces indeterminism to the functional behavior of the system, it can drive the safety concept from a pseudo-deterministic behavior towards more probabilistic approaches. A brief review of present best practices to implement safety for machinery is given as a basis. Then, crisp definitions for some key terms are suggested and the above paradigm shift from automation to autonomy is discussed using these. The objective is to convey the sense that safety of machinery continues to evolve with technological advances and to propose unsolved questions for discussion.

Björn Matthias is the ABB Corporate Research Fellow for Robotic Automation.

After studying physics at the California Institute of Technology (Caltech) and obtaining his PhD in experimental particle physics at Yale University in 1990, Björn spent 3 1/2 years as a post-doc at the University of Heidelberg, doing research on spectroscopy of exotic atoms and on rare decays. In 1994, Björn joined ABB Corporate Research in Germany, working in 3D computational design and simulation of power electrical systems and of electrostatic paint application. Since 2001, his research focused on industrial robotics, including robot safety, human-robot collaboration, and robotic small parts assembly. He participated in the development of the YuMi and GoFa robot families of ABB. Björn is active in national and international committees for industrial robotics standardization since 2009. The introduction of risk reduction best practices for collaborative applications was developed during this period. In the European context, he served on the Board of Directors of euRobotics aisbl from 2012 to 2015 and continues to contribute to the road mapping work for robotics and AI in both euRobotics and ADRA topic groups. Björn has co-authored over 100 scientific publications and contributed to more than 100 patent applications.

This presentation addresses the specific challenges associated with the safety and security of AI-based systems, with a particular focus on verification and validation (V&V) methods. Drawing on current research and development activities within the Safe AI Engineering project, the talk highlights various methodological and technical approaches aimed at adapting and extending conventional V&V practices to meet the unique demands of AI-based systems. Special emphasis is placed on the integration of safety-critical requirements into the development process, as well as on tools and procedures that help ensure the reliability and trustworthiness of AI components throughout their entire lifecycle.

Prof. Dr. Frank Köster began his academic career in 1989 studying computer science with a minor in psychology at the Carl of Ossietzky University of Oldenburg, where he later earned his doctorate and habilitation. After working as a research assistant, he joined the German Aerospace Center (DLR) in 2007, where he served as group leader, department head, and business development manager at the Institute of Transportation Systems. Since late 2020, he has been the founding director of the DLR Institute for AI Safety and Security, focusing on safety-critical AI systems, distributed data infrastructures, and ethical aspects of AI. He also holds a professorship at the University of Oldenburg and contributes to the standardization of test infrastructures for automated vehicle systems. Additionally, since 2020, he has been a domain ambassador for mobility in the GAIA-X project.

Currently, Frank Köster is one of the coordinators of the Safe AI Engineering project (safety of AI-based perception) and also one of the coordinators of the RoX project (digital ecosystems for AI-based robotics) – both project focus on advanced AI-based applications and are funded by the BMWE. Since some years, Frank Köster is member of the ASAM Board of Directors.