Keynotes

Over the past decade, the rapid advancements in machine learning have revolutionized various fields, significantly impacting our lives. In this talk, we will delve into the realm of medical applications and explore the challenges and opportunities associated with integrating these cutting-edge technologies into computer-assisted interventions. Our focus will be on surgical data science, holistic OR modeling, developing surgical world models, and fostering acceptance and trust in machine learning and robotic solutions within the medical domain, which often necessitates the path through Intelligence Amplification (IA). Augmented Reality allows us to leverage IA to augment human intelligence and expertise, ultimately paving the way for the seamless integration of Artificial Intelligence (AI) and robotics into clinical solutions.

Drawing from some groundbreaking research conducted at the Chair of Computer-Aided Medical Procedures at both TU Munich and Johns Hopkins Universities, I will present a series of novel techniques developed to address the unique demands of medical applications. Specifically, we will explore their practical implementations in diverse areas, including Robotic Ultrasound Imaging, Multimodal Data Analysis, Semantic Scene Graphs for Holistic Modeling of Surgical Domain, and finally the Surgical World Models. Furthermore, I will showcase compelling examples of how Augmented Reality solutions can serve as catalysts for embracing AI in computer-assisted surgery. In this talk, I will navigate at the intersection of machine learning, computer vision, intelligent user interfaces, and medical robotics to travel along the path from intelligence amplification and tele-manipulation to artificial intelligence and automation in computer assisted interventions.

Wearable robots are evolving from assistive devices into intelligent embodied systems capable of augmenting human capabilities and mediating physical collaboration between people and machines. In this context, soft wearable robots, or exosuits, represent a major departure from traditional rigid exoskeletons: they are lightweight, compliant, and intrinsically closer to the biomechanics of the human body. Yet their potential can only be fully exploited when their physical embodiment is coupled with intelligence — the ability to sense, interpret, and adapt to the user and to the surrounding environment.

This talk will present recent advances from my team over the past five years in the development of compact, robust, reliable, and efficient exosuits for human assistance, rehabilitation, and augmentation. Particular emphasis will be placed on how physiological signals, biomechanics, and movement patterns can be extracted from the wearer and transformed into control strategies that personalize the interaction between human and machine. This wearable intelligence enables different forms of augmentation: supporting the recovery of motor function after neurological damage, enhancing walking performance in older adults, and enabling a human operator wearing an exosuit to physically coordinate with a collaborative robot.

A central concept of the talk is Context-Aware Control, an approach that combines model-based control, human-in-the-loop sensing, artificial intelligence, and artificial vision to adapt robotic assistance to changing tasks, users, and environments. By integrating information from the body, the robot, and the external context, exosuits become more than actuated garments: they become embodied interfaces between human intention, physical assistance, and collaborative robotic systems.

The broader vision is that wearable embodied intelligence will play a central role in the future of human augmentation and collaborative robotics, creating systems that do not simply assist movement, but understand human state, adapt to human needs, and extend human capability in rehabilitation, daily life, and industrial environments.

Continuum robots have achieved remarkable success in applications such as minimally invasive surgery, where compliant bodies and human-in-the-loop teleoperation enable dexterous manipulation in confined environments. While this application-driven paradigm has demonstrated the unique strengths of continuum robots, it has also shaped the research questions the field has prioritized. As continuum robots are increasingly expected to operate with greater autonomy and interact more intelligently with their environments, advances in embodiment, modeling, state estimation, and control become essential.

In this keynote, I argue that intelligence in continuum robots should not be viewed solely as an algorithmic capability, but as an emergent property of the interplay between morphology, mechanics, and computation. I will present recent work on modular continuum robot embodiments with tailored stiffness profiles, demonstrating how mechanical design can encode functionality directly into the robot’s body. I will then discuss efficient physics-based simulation, showing how the accuracy of Goldstandard dynamic models can be closely approximated within the OpenCR-MuJoCo framework. Finally, I will examine how these advances provide the foundation for state estimation and localization for increasingly autonomous continuum robots.

By connecting robot embodiment, computational models, and estimation, this keynote highlights a shift from application-driven development toward establishing the fundamental principles that will enable the next generation of intelligent continuum robots.