In view of the increasing global consumption of resources and the finite nature of resources, industrial production in Lower Saxony, as well as worldwide, faces the challenge of overcoming the linear economy. This is essential in order to reduce greenhouse gas emissions and protect the ecosystem. Digitalization plays a crucial role in overcoming these challenges, as it enables and drives comprehensive change in the economy and society. 
The overarching theme of the ZL-DCE is to promote the transformation of society and the economy towards a digitized circular economy. This is achieved through the networking of digital and AI-supported circular service product and (retro) production systems. Real-world laboratories are used to test the approaches in a real environment with practice partners and users from society, using existing infrastructures such as intelligent collection and sharing stations, dismantling robots and repair services. Research and testing in the real-world laboratories stimulate new research impulses and contribute to the development of an open data pool that provides valuable input for AI-based approaches. By linking digitalization and the circular economy, sustainable, digital business models are promoted that contribute to an efficient and resource-conserving economy.

Funded by zukunft.niedersachsen, the joint science funding program of the Lower Saxony Ministry of Science and Culture and the Volkswagen Foundation

PROJECT PARTNERS: Clausthal University of Technology (IMW), Carl von Ossietzky University Oldenburg (VLBA), Fraunhofer Institute for Surface Engineering and Thin Films, Leibniz University Hannover (match) , OFFIS Institute for Information Technology (R&D) , Ostfalia University of Applied Sciences (IKAM & IVS), Braunschweig University of Technology (IWF & AIP)

Contact:
E-Mail: Dominique Briechle

Germany is ageing, which is increasing the need for outpatient care and rehabilitation services. At the same time, there is a significant shortage of skilled workers in these areas, which is expected to grow to almost 500,000 care workers by 2035, according to IW Köln. One promising solution is the use of assistance technologies. Studies show that collaboration between humans and robots is not only possible, but also improves efficiency and precision in many areas of work. The previously strict separation of human activities and robot actions, for example in industry, is changing towards close collaboration: robots provide active support by passing tools or holding objects. This flexible teamwork increases productivity and safety with less strain. Robots offer great potential to mitigate the shortage of skilled workers, particularly in body-related services such as care and therapy. In addition to technical and ethical issues relating to human-AI collaboration, these areas of application will be the focus of further research.

The Science Space Sustainable Human-AI Collaboration is funded by zukunft.niedersachsen, a funding program of the Lower Saxony Ministry of Science and Culture and the Volkswagen Foundation.

PROJECT PARTNERS: Clausthal University of Technology, Georg August University Göttingen, University Medical Center Göttingen, University of Applied Sciences and Arts

Contact:
E-Mail:Tobias Geger

The new 5G communication standard plays a crucial role for innovative applications, especially in the field of autonomous driving. This project aims to establish a 5G campus research network to meet the requirements of autonomous driving at level 3-5 without a safety driver on board and instead with a remote technical supervision (control center). According to the German law on autonomous driving, which came into force on July 28, 2021, the possibility has been created for the first time to implement autonomous driving at level 3 without a safety driver and with remote supervision (in the sense of a 5G remote control center in the project) in regular operation. For this 5G remote control center, a 5G network must be established that extends the required safety and security properties of the 5G radio standard in order to predict the quality of the radio network during teleoperation and protect the network from attacks.

In addition, autonomous vehicles should be able to cope with complex driving environments, such as intersections with poor visibility. 5G infra-structure sensor technology can support this by providing a replacement for the analog traffic mirror used by human drivers. Finally, a reliable AI-based driving system (5G fail-operational) with shadow mode must be developed and connected via 5G with the 5G infrastructure sensor technology and the 5G remote control center to form an integrated 5G-based mobility and logistics service system. An autonomous campus shuttle will be set up as a demonstration operation. The 5G campus research network will be used as a platform and test field for current and future research projects and will be available to the participating institutes and centers. The project results can serve as a model for the construction of autonomous shuttles in other 5G real-world laboratories and areas with 5G network coverage. The project was presented at the fourth advisory board meeting of the joint 5G real-world laboratory project in the Braunschweig-Wolfsburg mobility region and received unanimous positive support. This shows the excellent integration of the project into the 5G innovation region Braunschweig-Wolfsburg.

Contact:
E-Mail: Dr. Meng Zhang

Project website: Link

As part of its Future Concept 2030, Clausthal University of Technology has committed itself to the key topic of the circular economy. While degree-specific courses deal with selective aspects of the circular economy, there is a lack of a university-wide basic course for all students that teaches the essential concepts and basics of sustainability, environmental pollution, resource scarcity and the circular economy.

Since the Limits to Growth (LTG) course is open to all students at Clausthal University of Technology - and in the future also to other universities, e.g. Georg August University from SS2023 as part of the university-wide soft skills modules or Ostfalia University as part of the joint Digital Technologies degree program - an adapted teaching format and appropriate technical support is required. The plan is to implement the LTG course as a Massive Open Online Course (MOOC), which scales traditional forms of knowledge transfer (slides, recordings, etc.) via an IT-driven infrastructure and adds further technical features such as forums, quizzes and other forms of (semi-) automated tasks or tasks to be assessed via peer feedback. In addition, corresponding MOOC platforms offer participants interaction and networking opportunities as well as the possibility of asynchronous learning, in which students study according to their own schedule. They are given access to all teaching and learning materials, which are discussed in regular meetings - followed by optional question-and-answer sessions to clarify any questions. All teaching and learning materials, recordings, homework, tutorials, etc. are published under an open source license (CC-BY-SA-4.0) and are therefore freely available as teaching units that can be directly integrated or as a basis for further courses.

Contact:
Prof. Dr. Benjamin Leiding benjamin.leiding@tu-clausthal.de

Project website: https://www.sorec-greentech.de/

Closing material and substance cycles is an integral part of the circular economy. Many products (especially electronic devices) are becoming increasingly complex in terms of their structure and the raw materials used. In order to recover raw materials, the material flows have to be shredded into ever finer grain sizes. As a rule, dry sorting processes in the fine particle size range achieve good sorting results with lower throughputs. From a certain increase in throughput, the quality of the sorting results decreases significantly. In order to work economically, sorting machines are operated as close as possible to the "tipping point" between good quality and maximum throughput, and downtimes are reduced through regular maintenance. Fluctuations in throughput lead to the optimum operating point being undershot or exceeded, and in the worst case scenario, the fractions produced have to be costly re-sorted.
The SoRec project is dedicated to the digitalization of sorting processes for fine-grained, metal-containing material flows in the recycling industry. By installing optical detection systems, the sorting process is monitored and data such as mass throughput, particle size and impurities are obtained. This data is used to regulate and control the sorting process. In addition, they are combined with other machine parameters and sensors to create a comprehensive data image in the form of a digital twin. In this way, sorting and recycling systems can be operated semi-automatically and with minimal manpower. Economic benefits result from the maximized throughput of the system, reduced downtimes and higher sorting quality. From a circular economy perspective, the recovery rate of already extracted raw materials for circular use and the closing of material and substance cycles increases.

Contact:
Prof. Dr. Benjamin Leiding benjamin.leiding@tu-clausthal.de

In the area of underground management, previous simulation models have considered individual aspects such as geology, cavity geometry, weather technology/flow mechanics and geomechanics separately. However, a more comprehensive model coupling enables a more holistic analysis of the system interrelationships and the interdependencies between the models. As part of the MOVIE project, a hybrid model is being developed using artificial intelligence in order to optimize the computing time for these simulation models and enable interactive use in the virtual laboratory. A retrospective simulation approach is used in which the forecasts based on historical data are compared with current measurements from the real underground laboratory. The resulting difference is used to fine-tune the model.

Project partners: Freiberg University of Mining and Technology - Institute of Mining and Special Civil Engineering, Research and Training Mine (FLB) "Reiche Zeche"; fibrisTerre Systems GmbH (Berlin), Terranigma Solutions GmbH (Aachen) and other partners at Clausthal University of Technology: IGMR-Geomatics, IGMR- Underground Mining, IGMR-Geomechanics, IGMR-Geology

Project website: https://www.igmr.tu-clausthal.de/department-of-geo-engineering/geomatics-for-underground-systems/forschung/movie

Contact person: Dhananjeyan Jeyaraj

This project is part of the CircularLIB research training group on the cycle-oriented production of lithium-ion batteries. In this program, the three universities TU Braunschweig, TU Clausthal and Leibnitz University Hannover are cooperating with the Fraunhofer Institute for Surface Engineering and Thin Films IST.

The aim is to research a hybrid AI-based modeling approach for lithium-ion batteries. Recently, machine learning has also been increasingly used in the context of traditional engineering disciplines. Deep learning approaches in particular offer fast and relatively accurate modeling approaches based on learning specific tasks from examples. This is of particular interest in the context of high-dimensional complex systems where the underlying physics is not fully known or the computational effort required for simulation using conventional numerical methods is high. Another aspect that promotes the use of deep learning approaches is the need for fast models that can be used in iterative tasks such as optimization and control. However, data-driven models are usually black-box approaches that are developed based on data only and do not explicitly incorporate physical knowledge or constraints into the model development. This can result in such models lacking robustness and accuracy, especially with limited training data.

Recent advances in the field of physically informed machine learning have led to a number of approaches that are well suited for solving various types of partial differential equations (PDEs). This project aims to develop models based on both data and available physical knowledge for lithium-ion batteries using physically informed machine learning techniques that allow a hybrid AI-based modeling approach.

Contact person: Hamidreza Eivazi Kourabbaslou