+49 234 32 10186
david.gilgen@pknrw.de
Do you have any questions about the Department or need further information? Please contact us.
Research
The research focus of the department lies on systems characterized by the interaction of engineering and computer science. These systems interact with their environment and adapt to it autonomously (they are adaptive); they are capable of dealing with unexpected and unforeseen situations in dynamic environments (they are robust); they anticipate the effects of different influences based on experience (they are predictive); and they take into account the diverse behavior of users (they are user-friendly). This general characterization defines the department’s overall research activities.
Concept of Research
Positioning within the scientific landscape shows that Cyber-Physical Systems (CPS), Cyber-Physical Production Systems (CPPS), and intelligent automation require research performance in strongly interdisciplinary areas. These include cognition and perception, sensor technologies, self-X behavior, autonomic principles, image processing and information fusion, machine learning and computational intelligence, network technologies and communication models, digital infrastructures, as well as security and safety.
Intelligent technical systems that interact autonomously with their environment rely on machine learning (ML), data analysis, efficient sensor systems, and intelligent networking through embedded software (edge computing). Products, mobile platforms, robots, or vehicles collect data and use it to optimize their behavior. Production systems increasingly use data to respond more agilely to market developments and customer needs, and to manufacture products in an optimal and resource-efficient way. Digital twins model individual modules or entire plants, with machine learning playing a central role. The successful application of lightweight, localized ML approaches in products and production systems close to the data source is particularly significant.
However, the technologies available in current industrial practice often fail to meet many of these requirements: data acquisition and analysis are typically carried out by experts, while real-time capability is crucial in technical systems. The physical interrelations of technical systems require explicit integration of domain knowledge. High data dimensionality, system heterogeneity and dynamics, implementation in distributed hardware topologies (edge), and the need for easily maintainable and reliable components represent major challenges. High-speed communication (including 5G) is a necessary prerequisite for distributed autonomous systems.
Developing these technologies to a competitive level of maturity requires substantial research effort, in order to:
- realize adaptive, resource-optimized, safe, and user-friendly production and process engineering systems in the spirit of Industry 4.0, and
- establish new, data-based services.
Machine learning enables concepts such as predictive maintenance or product individualization, and thus holds the potential to generate added value at all stages of corporate processes through the extraction of knowledge from digital data. Instead of centralized black-box approaches, decentralized and lightweight ML methods are crucial—those that are directly embedded in products and production systems and that seamlessly integrate engineering models.
At the same time, the use of ML technologies in agile production processes presents manufacturers of technical systems with new challenges: novel product development and engineering approaches are needed—approaches that allow space for adaptivity.
Cyber Physical Systems
Cyber-Physical Systems are physical systems with inherent partial intelligence provided by embedded software. They collect data via sensors and influence both the system and its environment through actuators. They analyze and store data, interact actively or reactively with the physical and virtual worlds, and are interconnected via digital communication systems—both with each other and within global networks.
Particularly relevant topics in this research area include: machine learning, networking and integration technologies, communication engineering, network architecture, embedded resource-constrained systems, internet technologies, and multimedia communication.
Disciplinary Positioning
The department defines itself as an interdisciplinary research environment that fosters and supports collaboration across established academic boundaries. Disciplinarily, it is particularly rooted in the fields of mechanical engineering, computer science, electrical engineering, production engineering, and mechanical systems engineering.
Activities
The department offers the structured doctoral program ‘Cyber Physical Systems’ and regularly organizes colloquia, specialized academic events, and a lecture series. The organization of an annual conference for early-career researchers is currently being planned. A summer school with Tel Aviv University (TAU, Israel) took place in September 2022 in both North Rhine-Westphalia and Israel.
The department maintains a high level of networking—whether through facilitating contact persons for cooperative doctorates, joint research proposals, exchanges on emerging research topics, or connecting with practice partners.
In addition to national collaborations, the department also maintains numerous international partnerships, currently with a particular focus on Europe, the United States, and South Africa.
