Cyber-Physical Systems and Systems Engineering

Impact: Research Topic/Theme (at group level)

Description of impact

Cyber-physical systems (CPSs) integrate sensing, computation, control and networking into physical objects and infrastructure, connecting them to the Internet and to each other. CST is a leader in generating the fundamental knowledge and tools to make CPSs a reality. These advances hold the potential to reshape our world with more responsive, precise, reliable, and efficient systems, enabling a revolution of "smart" devices and systems in the domains of:
Manufacturing: high-tech mechatronic equipment (wafer scanners, electron microscopes, wire bonders, etc.),
Agriculture: Precision farming, and food processing,
Mobility and transportation: Automated and cooperative cars,
Smart cities and large-scale civil infrastructures: Tunnels, bridges, water locks, etc.
Personalized Healthcare: Smart medical devices delivering hyperthermia treatments for cancer care.

In advancing CPS technologies, the demand for systematic integration of the different design disciplines (mechanics, electronics, software sciences) and the innovation across application domains are main drivers to accelerate fundamental research to keep pace. New concepts emerging from artificial intelligence and machine learning, and the use of digital twins for monitoring, analysis and design of the CPS of the future create new research avenues with major societal implications.

The CPS research in CST aims to develop the core research needed to engineer these complex CPS, many of which may also require provable behaviors. By abstracting from the particulars of specific systems and application domains, our CPS research strives to reveal cross-cutting, fundamental scientific and engineering principles that underpin the integration of cyber and physical elements across all our application domains. The methods are being developed in close co-operation with leading industries such as Thermo-Fisher Scientific, Avular, ASMPT, NXP, TNO Automotive, ASML, Ford, Canon, etc.

Hybrid Systems and Control
The mathematical modelling of CPS needs both discrete and continuous model ingredients, as CPS switch between discrete operation modes (often in the form of automata or finite state machines) and decisions, next to requiring physical models often in the form of differential or difference equations. This leads to an overall hybrid system description. The CST group provides important fundamental developments in hybrid systems that directly connect to the essential challenges in CPS applications.

Large-scale and Networked Systems
Networked systems are complex dynamical systems composed of many simple subsystems interacting through physical connections or one or more communication media. Both control systems operating over (resource-constrained) communication networks and multi-agent systems (and their combination) are at the heart of this research line. Networked systems arise as natural models in many areas of engineering and sciences, such as autonomous unmanned vehicles, cooperative robotics, smart grids, biological networks, internet-of-things and many more, and taming their complexity in analysis and (distributed) controller design are important open questions.

Digital twinning and Learning
Asimov, reinforcement learning, ADP

Control over Communications event-triggered and self-triggered control

Supervisory Control is concerned with well-defined integration of the behavior of (networked) system components with an emphasis of safe and productive overall system behavior. The developed methods find their roots in discrete event systems and supervisory control theory in support of a synthesis-based engineering approach that provides ample guarantees on the desired properties of such supervisory controllers.

Systems Engineering aims to develop quantitative methods for the analysis, design and implementation of (embedded) mechanical engineering systems exhibiting concurrent behavior, with particular focus on the manufacturing high tech industry. The objectives are to generate theory, to develop techniques, and to build computational tools. The basis is mechanical engineering science, in which we use formal methods from computer science and methods from mathematics. Model-based engineering design methods and tools are developed on this scientific basis. These model-based engineering methods and tools are validated in selected industrial applications.
Category of impactResearch Topic/Theme (at group level)