Invited Speakers

Professor of Computer Science and Chair of Perception for Intelligent Systems - Manager of the project KI.Fabrik - A lighthouse initiative  for Intelligent Manufacturing at the German Technology Museum at the  Technical University Munich (TUM), Germany 

• Title: Flexible, Self-Optimizing, Legible, and Protective – Autonomous Mobile Robots in Shared Environments as a Key Element in Future Intelligent Manufacturing

• Abstract:  For many applications of mobile robots with partial or full autonomy in industrial applications it is crucial that humans do *not* have to be banned from the operation space, i.e., that space is shared with humans. This is a cornerstone of the "KI.Fabrik Bayern" that I will introduce in the first part of this presentation. I will then talk about four characteristics that are important for the introduction of mobile robots in shared environments, in particular as flexible elements of intralogistics chains, and that provide complementary advantages:

— Flexible: Autonomous mobile robots in the vicinity of humans need to be efficient and safe. Among others, they need to know reliably where they are, which calls for accurate and robust localization algorithms and for approaches to introspection which allow the mobile robots to discover if their localization is not correct any longer.

— Self-Optimizing: The introduction of autonomous mobile robots in a given or only mildly modified industrial environment needs to be as simple as possible in order to reduce deployment time and cost. In this presentation, I will talk about ways to reduce the deployment time. In particular, ways to facilitate robots to blend into their environment smoothly based on their own learned understanding of the motion they observe in a new environment (learning implicit traffic rules).

— Legible: Legible, the interaction between humans and robots also has to be efficient and thus legible. Every point at which an autonomous vehicle meets a human or a human-driven vehicle is crucial in this respect. In my presentation, I will talk about different ways in which a robot can communicate its navigation intention to a human and, vice versa, can understand and predict the spatial intention of humans.

— Protective: Autonomous mobile robots can further contribute as part of heterogeneous sensor networks to regulatory monitoring of industrial sites, thus keeping exposure of human workers to healthy levels.

Inaugural Department Chair and Professor in the Robotics Department, University of Michigan, USA

• Title: Embedding natural and artificial intelligence in digital twins to improve manufacturing outcomes

• Abstract: As high-performance computing and high-speed networking become cost-effective and pervasive, there is an opportunity to leverage the huge amounts of data generated by manufacturing systems to improve performance.  Putting all of this data together to enable useful outcomes requires models, and developing appropriate models requires a combination of both natural and artificial intelligence. The knowledge from human subject-matter experts guides the selection of data and the type of models, and machine learning/artificial intelligence can be used to develop the models, as well as compute the estimates or predictions useful for decision making.

Recently, the concept of "digital twins" has been used to encapsulate these models.  We define a digital twin as a software entity that takes real-time data streaming from the factory floor, and uses this data together with models to create estimates or predictions that can inform decisions.  This talk will present a framework and examples for digital twins in manufacturing.

Senior Robotics Research Scientist at MathWorks in the Advanced Research & Technology Office,
Mathworks, USA

• Title: Breaking Barriers: Developing and Deploying Robotics Software for Intelligent Manufacturing in Modern Industry

• Abstract: In 1961, Unimate, the first industrial robot designed for manufacturing, was deployed on a General Motors assembly line. Surprisingly, the core functionalities of most industrial robots have seen modest evolution since then, even as the term "intelligent manufacturing" has been part of the industry lexicon for over a decade. From a software tool maker's perspective, what challenges slow down the widespread adoption of intelligent manufacturing solutions? This talk will explore how historical procedural norms, established arrangements, and repeatability requirements can hinder the rapid adoption of intelligent manufacturing research solutions. Examples will help demonstrate how Model-Based Design can help overcome these challenges to advance the field.

Research Scientist at the Munich Institute of Robotics and Machine Intelligence - MIRMI - at the 
Technical University of Munich (TUM), Germany

• Title: How Can We Understand if a Robot is Fit for Safe Physical Interactions?

• Abstract:  Multiple robot systems have freshly emerged o the market within the last years. While these systems incorporate multiple features they rearly state key performance indicators in order to assess their actual fitness to perform tasks.

In this talk the AI Robot Performance and Safety Center of the Munich Institute of Robotics and Machine Intelligence is introduced and the latest work on understanding a robot's fitness to perform physical interaction with the environment and fundamental research on generating injury protection databases required for real-world implementations of safe robotics.

Staff Research Scientist at Siemens Healthineers (Princeton NJ, USA)

• Title:  Intelligent Automation for Laboratory Diagnostics

• Abstract: The demand for fast and accurate in-vitro diagnostics is ever-increasing. Automation has provided a means to efficiently scale laboratory diagnostics, leading to complex systems that rely on hundreds of various components interacting with each other. What starts as a tube of blood goes through a unique set of operations on these systems that are specific to patient requirements – this calls for intelligent flexible automation that can maximize system throughput. However, safety and reliability are also key, as errors could mean redrawing a pediatric sample or delaying a critical medical decision in an emergency room.
  
  Everyone in the diagnostics business understands that improving speed and reliability doesn’t just help improve the profits of labs; it helps save lives. In this talk, I will briefly introduce in-vitro diagnostics and the intricate automation systems that touch millions of patients’ lives every year. I will detail how advancements in AI are being used to improve these operations and describe some of the adjacent open challenges in this space.

Assistant Professor and David Packard Jr. Faculty Fellow,  Santa Clara University (SCU), USA

• Title: Human-Centered Collaborative Robots in Manufacturing: From Understanding Speech Commands to Assessing Ergonomics  

• Abstract: Collaborative robots are becoming essential partners in modern manufacturing, enhancing productivity while working alongside human operators. However, effective human-robot collaboration requires advanced capabilities, from understanding natural language commands to ensuring human workers’ physical well-being.
  In this talk, I will present a robot learning framework that enables robots to interpret unstructured spoken instructions and autonomously generate sequences of actions — a crucial step toward intuitive, human-centered interaction on the factory floor. Additionally, I will discuss an intelligent robotic assistant designed to monitor and assess worker ergonomics and cognitive fatigue, aiming to detect signs of physical or mental strain during collaborative tasks. This dual approach — enhancing communication and promoting worker safety — highlights the transformative potential of collaborative robots in creating more adaptive, responsive, and human-friendly manufacturing environments.  

Robotics & Automation Technology Manager at the MTC (Manufacturing Technology Centre), UK

• Title:   Customisation Meets Efficiency: The Robofacturing Approach to Modern Manufacturing

• Abstract:  Over the past century, sequential design and assembly processes have facilitated the mass production of complex goods. While effective for mass manufacturing, these linear processes struggle with the efficient production of mass-customised goods with variable volumes. Robofacturing introduces a paradigm shift in design and manufacturing, optimising robotic technology from the outset rather than as an afterthought. Traditional workflows often constrain robotic production by embedding legacy inefficiencies rooted in product design. In conventional manufacturing, CAD tools provide designers with flexibility, but production engineers must modify designs for feasibility, limiting the role of robots to specific tasks. Robofacturing integrates robot-specific design tools with versatile robot cell configurations, imposing design constraints early to ensure designs are optimised for robotic manufacture. Utilising matrix-based production approaches, Robofacturing systems are fully reconfigurable and scalable, allowing rapid adaptation to changing production requirements and volumes. This flexibility enables the efficient production of customised goods, meeting diverse market demands without sacrificing efficiency or productivity. 

Leader of the  Sensing and Perception Systems Group  - National Institute of Standards and Technology, USA

• Title:  TBD

• Abstract:  TBD

ARM Institute - Advanced Robotics for Manufacturing,  Pittsburgh, USA

• Title:  Innovation to Transition – How the ARM Institute is Impacting Manufacturing through Robotics and Automation

• Abstract:  As one of nine DoD Manufacturing Innovation Institutes, the Advanced Robotics for Manufacturing (ARM) Institute’s mission is to accelerate the development and adoption of innovative robotics technologies that are the foundation of every advanced manufacturing activity today and in the future.  By doing so, we’ll make robotics, autonomy, and AI more assessable to manufactures, strengthen our economy and global competitiveness, and elevate our national security.

During this talk, you’ll learn how the ARM Institute has engaged is diverse consortium of partners to solve some of the most complex manufacturing problems with robotics and automation and transitioned the technology to the manufacturing floor.  From using AI/ML to increase the performance of defect detection systems, to improving path planning algorithms and programming time, to addressing high-mix-low-volume production, our projects address an array of challenges across manufacturing sectors.  Additionally, you’ll hear about where the ARM Institute envisions robotics and automation will in the not-so-distant future.