Keynote Speakers

This plenary lecture presents embodied artificial intelligence as a framework for understanding how intelligent systems operate when coupled with real bodies, sensors, energy constraints, infrastructures, and institutional environments. The lecture begins from the premise that intelligence in the real world cannot be reduced to disembodied computation alone. Instead, it emerges through the interaction of perception, state estimation, control, decision-making, and action under uncertainty. In this view, embodied AI is not only a topic within robotics, but a broader systems perspective that connects engineering, medicine, autonomy, and governance.
The first part of the lecture introduces the theoretical foundations of embodied AI through concepts from control theory, signal processing, and estimation. It emphasizes that every embodied intelligent system is a closed-loop system operating in an uncertain environment, where sensing is partial, decisionmaking is constrained, and action reshapes the world that is being observed. Particular attention is given to the distinction between reversible and irreversible dynamics, to the role of measurement in shaping controllability, and to the importance of designing systems that remain robust when models are incomplete and environments are nonstationary.

The second part focuses on robotics as the most visible and mature domain of embodied AI. The lecture discusses how autonomous aerial, surface, underwater, and ground systems combine perception, control, navigation, and energy management in real time. It then extends this perspective to heterogeneous autonomy and swarms, in which intelligence is distributed across multiple agents with varying sensing, communication, and mobility capabilities. In these systems, coordination emerges not from a single centralized controller but from local interactions, consensus, adaptive behaviors, and networked control, making them especially relevant for maritime operations, distributed sensing, and adaptive ocean observation.

The third part moves from robotics toward medical and mobile diagnostics, showing that embodied AI is equally relevant when intelligence is embedded in healthcare devices, workflows, and institutional structures. The lecture examines how AI in clinical environments must operate under strict constraints of safety, accountability, limited resources, and human trust. It argues that in such settings, the control loop extends far beyond the device itself, influencing hospital processes, decision pathways, and the generation of future data. This makes embodied AI a useful framework not only for smart machines, but also for understanding healthcare systems as dynamic, feedback-driven environments.

The final part broadens the discussion toward ethics, responsibility, and future directions. It considers how control architectures also exist in institutions, finance, and policy, and how these structures shape the conditions under which intelligent systems are developed and deployed. The lecture concludes by outlining future research directions in integrated embodied AI theory, robust autonomy, clinically grounded intelligent systems, and socio-technical system design. Its central message is that intelligence should be studied not as isolated computation, but as action embedded in physical, social, and institutional reality, and that the next generation of AI must therefore be designed with equal attention to technical performance, human context, and long-term systemic consequences.

Speaker Biography

Dr. Vladimir Djapic is a research associate at the Institute for Artificial Intelligence Research and Development of Serbia, where he leads the Belgrade office and works on embodied AI, autonomous systems, and real-world deployment of intelligent technologies.

He holds B.S., M.S., and Ph.D. degrees in Electrical Engineering from the University of California, with academic training in machine intelligence, robotics, control, and intelligent systems. Over more than two decades of professional work in the United States and Europe, he has built a career that connects rigorous autonomy research with operational deployment in demanding environments.

Dr. Djapic has held senior scientific and technical leadership roles in maritime autonomy, defense-oriented robotics, and intelligent transportation. His experience includes leading multidisciplinary teams in the development and field deployment of aerial, surface, and underwater autonomous systems for cooperative missions, maritime security, navigation, sensing, and complex realworld operations. He has also contributed to major sea trials, multi-robot coordination programs, and international research initiatives focused on autonomy in challenging environments.

His technical expertise spans localization and mapping, GNSS-denied navigation, multi-sensor fusion, simultaneous localization and mapping, behavior-based control, distributed autonomy, and coordinated control across heterogeneous robotic platforms. He has worked with a wide range of sensing modalities, including IMU, LiDAR, SONAR, stereo vision, UWB, and signals of opportunity, and has consistently focused on systems that can perceive uncertain environments, make robust decisions, and act safely outside the laboratory.

In industry, Dr. Djapic held a senior leadership role in autonomous driving, where he helped build high-definition mapping and multi-sensor localization technologies for self-driving vehicles and supported development of geofenced electric mobility platforms. His work bridges academic research, government and defense programs, and applied engineering, giving him an unusual combination of theoretical depth and deployment experience.

Since returning to Serbia, he has remained active in research leadership and innovation, contributing to projects in trustworthy AI, privacy-preserving localization, embodied AI for healthcare and smart cities, intelligent sensing, and autonomous mobility. He has authored more than sixty scientific publications, contributed to international conferences and editorial activities, and is recognized for linking control theory, perception, navigation, and AI into capable embodied systems that operate in the real world.

With a career spanning underwater robotics, autonomous maritime systems, self-driving vehicles, and emerging AI applications, Dr. Djapic brings a broad international perspective and a strong record of interdisciplinary innovation at the intersection of engineering, autonomy, and intelligent systems.