Research area Intelligent Systems & AI

InnoSÜD: Dynamic innovation system in the Danube-Iller region

Project manager: Prof. Dr. rer.nat. Marianne von Schwerin
Project duration: 01.01.2018 - 31.12.2022
Funded by: Federal Government + State
Program name: Innovative University
Project description:
InnoSÜD is developing a dynamic innovation system in the Danube-Iller region, which has its strengths in the implementation of new transfer formats. Priority future tasks are addressed, which include the implementation fields of mobility, energy, health / biotechnology and transformation management.
THU represents the subject areas of energy and mobility. Both subject areas encompass areas that will have to change significantly in the near future, as the terms energy and transportation transition make clear. Electromobility and automated driving are shaping developments in the automotive sector, while renewable and decentralized energy systems are the future of energy supply.
By participating in this innovation project, THU aims to actively shape these significant technological upheavals at a regional level, pursuing an alternative approach in the field of knowledge and technology transfer. To this end, InnoSÜD is using specially developed transfer formats such as trialoge or innovation circles to bring regional business, science and society together. Real-world laboratories, open development environments (open labs and open products) as well as cooperative promotions and staff exchanges also help to share ideas for innovation and put them on a broad footing.

InnoTeach

Project manager: Prof. Dr. Marianne von Schwerin
Project duration: 01.01.2020 - 31.12.2021
Funded by: State - MWK
Project description:
In the sub-project at THU, the topic of AI is taken up, processed and used in cooperation with companies in projects in the Master's degree course in Intelligent Systems. Solution approaches for questions in the field of artificial intelligence from the economy are developed and prototypically implemented. The project aims to enable a broad range of students to come into contact with companies in the region. To this end, the transfer formats from the InnoSÜD (Innovative University) project are being further developed and are already being used in teaching.

Data literacy and data science for SMEs: qualification and further training (DLDS)

Project manager: Prof. Dr. Reinhold von Schwerin
Project duration: 01.05.2019 - 31.12.2021
Funding provider: State - MWK / EFRE
Program name: ESF
Project description:
The aim of the project is to expand the use of methods in the field of artificial intelligence and machine learning, especially in the SME sector. The areas of data engineering, analytics and deep learning are to be considered. The levels of awareness, literacy, practitioner and scientist will be addressed. In particular, research into the explainability of AI results will be used to create acceptance and a better understanding of AI solutions. Well-understood learning processes in AI systems are analyzed and the decision-making processes of machine learning are classified in order to provide a basis for understanding the results of AI algorithms.

Future City 3

Project manager: Prof. Dr. Michael Schlick
Project duration: 01.02.2020 - 31.12.2021
Funded by: Federal Government - BMBF
Program name: Zukunftsstadt
Project description:
Building on the gateway network of the city of Ulm created in the previous funding phase, a suitable sensor technology is to be developed and rolled out. Based on LoRaWAN wireless technology, a universal data collection and processing platform for data on air quality, temperature, humidity, bicycle mobility, WiFi scanners, acceleration sensors and GPS will be created. This data is to be analyzed, processed and further processed intelligently. Using suitable machine learning / AI methods, predictions can then be made about the use of the city center infrastructure as part of the project. This will be used to develop a concept for better utilization of sharing offers in public transport, in the combination of sharing and public transport and of cab rides.

City of the future 2

Project manager: Prof. Dr. Marianne von Schwerin
Project duration: 01.01.2018 - 31.12.2018
Funded by: Federal Government - BMBF
Program name: Zukunftsstadt
Project description:
The city is supported by various professors from Ulm University of Applied Sciences in the future areas of mobility, energy and networking as well as business, employment and work. The topics of electromobility and alternative transport concepts are being considered, as well as the energy transition and digital transformation. In this second phase of the City of the Future, the focus is on developing concepts that can be implemented.

Requirement-driven Optimization of System Concept with integrated Model Based Safety Analysis (SysRO)

Project manager: Prof. Dr. Rüdiger Lunde
Project duration: 01.11.2018 - 29.02.2020
Funded by: other public institution
Program name: smartflow
Project description:
The project focuses on the further development of the smartIflow approach for the automated creation of safety analysis artefacts based on component-oriented models. In particular, a new FMEA generation is to be developed and integrated into the existing workbench.
What is new about this is the type of model used as a basis, as the transition systems used here have not previously been used for this purpose.

Gögelein, David; von Schwerin, Marianne; Herbort, Volker:
PV system installation assessment based on power measurement for balcony power plant applications,
in: IEEE Journal of Photovoltaics, vol. 14, no. 4, IEEE Xplore, 2024, pages 13 (571 - 582).
DOI: 10.1109/JPHOTOV.2024.3384914
ISSN: 2156-3381

Krenmayr, Lucas; von Schwerin, Reinhold; Schaudt, Daniel; Riedel, Pascal; Hafner, Alexander:
DilatedToothSegNet: Tooth Segmentation Network on 3D Dental Meshes Through Increasing Receptive Vision,
in: Journal of Imaging Informatics in Medicine Imaging, vol. 37, Springer, 2024, pages 1846-1862.
DOI: 10.1007/s10278-024-01061-6
ISSN: 2948-2933

Riedel, Pascal; Belkilani, Kaouther; Reichert, Manfred; Heilscher, Gerd; von Schwerin, Reinhold:
Enhancing PV feed-in power forecasting through federated learning with differential privacy using LSTM and GRU,
in: Energy and AI, Vol. 18, Science Direct, Elsevier, 2024, pages 100452 (17 pages).
DOI: 10.1016/j.egyai.2024.100452
ISSN: 2666-5468

Riedel, Pascal; Schick, Lukas; von Schwerin, Reinhold; Reichert, Manfred; Schaudt, Daniel; Hafner, Alexander:
Comparative analysis of open-source federated learning frameworks - a literature-based survey and review,
in: International Journal of Machine Learning and Cybernetics, Vol. 15, Springer Nature, 2024, pages 5257-5278.
DOI: 10.1007/s13042-024-02234-z
ISSN: 1868-8071

von Schwerin, Reinhold; Schaudt, Daniel; Hafner, Alexander:
Umsetzung von KI-Transferprojekten - Praxisbericht zu Risiken und Herausforderungen,
in: Informatik 2024, Lecture Notes in Informatics (LNI), Klein et. al. (ed.), GI Gesellschaft für Informatik, 2024, pages 1751-1756.
DOI: 10.18420/inf2024_152
ISBN: 978-3-88579-746-3
ISSN: 1617-5468

Goldstein, Markus:
Special Issue on Unsupervised Anomaly Detection,
in: Applied Sciences, 2023 13(10):5916., MDPI, Switzerland, 2023, pages 3.
DOI: 10.3390/app13105916 ISSN: 2076-3417

Hwang, Young-Seok; Um, Jung-Sup; Pradhan, Biswajeet; Choudhury, Tanupriya; Schlüter, Stephan:
How does ChatGBT evaluate the value of spatial information in the 4th industrial revolution?,
in: Spatial Information Research, Springer, 2023, pages 8.
DOI: 10.1007/s41324-023-00567-5
ISSN: 2366-3286 (Print), eISSN: 2366-3294

Krenmayr, Lucas; Goldstein, Markus:
Explainable Outlier Detection using Feature Ranking for k-Nearest Neighbors, Gaussian Mixture Model and Autoencoders,
in: Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - ICPRAM 2023, INSTICC, SciTePress, 2023, pages 245-253.
DOI: 10.5220/0011631900003411
ISSN: 2184-4313, ISBN: 978-989-758-626-2

Riedel, Pascal; Singh, Gaurav; von Schwerin, Reinhold; Reichert, Manfred; Hafner, Alexander; Schaudt, Daniel:
Performance Analysis of Federated Learning Algorithms for Multilingual Protest News Detection using Pre-trained DistilBERT and BERT,
in: IEEE Access, IEEE Publishing Operations, 2023, pages 14.
DOI: 10.1109/ACCESS.2023.3334910
ISSN: 2169-3536

Riedel, Pascal; von Schwerin, Reinhold; Schaudt, Daniel; Hafner, Alexander; Späte, Christian:
ResNetFed: Federated Deep Learning Architecture for Privacy-Preserving Pneumonia Detection from COVID-19 Chest Radiographs,
in: Journal of Healthcare Informatics Research. 7, Springer Nature, 2023, pages 203-224.
DOI: 10.1007/s41666-023-00132-7
ISSN: 2509-4971

Schaudt, Daniel; Späte, Christian; von Schwerin, Reinhold; Reichert, Manfred; von Schwerin, Marianne; Beer, Meinrad; Kloth, Christopher:
ACritical Assessment of Generative Models for Synthetic Data Augmentation on Limited Pneumonia X-ray Data,
in: Bioengineering - Artificial Intelligence (AI) for Medical Image Processing, Vol.12, Is. 10, pages 24.
DOI: 10.3390/bioengineering10121421
ISSN: 2306-5354

Schaudt, Daniel; von Schwerin, Reinhold; Hafner, Alexander; Riedel, Pascal; Reichert, Manfred; von Schwerin, Marianne; Beer, Meinrad; Kloth, Christopher:
Augmentation strategies for an imbalanced learning problem on a novel COVID-19 severity dataset,
in: Scientific Reports (Sci Rep) 13, Article number: 18299, 2023, Pages 16.
DOI: doi.org/10.1038/s41598-023-45532-2
ISSN: 2045-2322

Schaudt Daniel; von Schwerin, Reinhold; Hafner, Alexander; Riedel, Pascal; Späte, Christian; Reichert, Manfred; Hinteregger, Andreas; Beer, Meinrad; Kloth, Christopher:
Leveraging Human Expert Image Annotations to Improve Pneumonia Differentiation through Human Knowledge Distillation,
in: Nature - Scientific Reports 13, Article number: 9203, 2023, Pages 13.
DOI: 10.1038/s41598-023-36148-7
ISSN: 2045-2322

von Schwerin, Reinhold; Hafner, Alexander:
DASU - Transferzentrum für Digitalisierung, Analytics & Data Science Ulm - Intelligente Transferstrategien zur Unterstützung der digitalen Transformation,
in: Lecture Notes in Informatics (LNI) - Proceedings Volume 337, GI - Gesellschaft für Informatik, 2023, pages 1751-1762.
DOI: 10.18420/inf2023_178
ISBN: 978-3-88579-731-9, ISSN: 1617-5468

Choi, Jeonghoon; Suh, Dongjun; Otto, Marc-Oliver:
Boosted Stacking Ensemble Machine Learning Method for Wafer Map Pattern Classification,
in: Computers, Materials & Continua 74(2), Tech Science Press, 2022, pages 2945-2966.
DOI: 10.32604/cmc.2023.033417
ISSN: 1546-2218 (print), 1546-2226 (online)

Hafner, Alexander; Schaudt, Daniel; Riedel, Pascal; von Schwerin, Reinhold:
Vertrauen als (notwendige) Optimierungsgröße für ML-Modelle im produktiven Umfeld,
in: 3. Innovationskongress Ulm | Neu-Ulm Datasience to go, von Schwerin, Marianne; Reichert, Manfred; Urban, Karsten (ed.), OPARU, 2022, pages 47-55.
DOI: 10.18725/OPARU-44168
ISBN: 978-3-9820843-3-6

Hwang, Young-Seok; Schlüter, Stephan; Park, Seong-Il; Um, Jung-Sup:
Comparative Evaluation for Tracking the Capability of Solar Cell Malfunction Caused by Soil Debris between UAV Video versus Photo-Mosaic,
in: Remote Sensing, 14(5), 1220, MDPI, 2022, pages 15.
DOI: 10.3390/rs14051220
ISSN: 2072-4292

Hwang, Young-Seok; Schlüter, Stephan; Um, Jung-Sup:
Spatial Cross-Correlation of GOSAT CO2 Concentration with Repeated Heat Wave-Induced Photosynthetic Inhibition in Europe from 2009 to 2017,
in: Remote Sensing, 14(18), 4536, MDPI, 2022, pages 13.
DOI: 10.3390/rs14184536
ISSN: 2072-4292

Kojić, Milena; Schlüter, Stephan; Mitić, Petar; Hanić, Aida:
Economy-environment nexus in developed European countries: Evidence from multifractal and wavelet analysis,
in: Chaos, Solitons & Fractals Volume 160, Science Direct, Elsevier, 2022, pages 15.
DOI: doi.org/10.1016/j.chaos.2022.112189
ISSN: 0960-0179 / eISSN: 1873-2887

Mitić, Petar; Kojić, Milena; Hanić, Aida; Schlüter, Stephan:
Environment and Economy Interactions in the Western Balkans: Current Situation and Prospects,
in: Lecture Notes in Networks and Systems (LNNS, Vol. 529), Tufek-Memišević, T.; Arslanagić-Kalajdžić, M.; Ademović, N., Springer Cham, 2022, pages 3-21.
DOI: 10.1007/978-3-031-17767-5_1
ISBN: 978-3-031-17766-8, ISSN: 2367-3370

Riedel, Pascal; Schaudt, Daniel; Hafner, Alexander; von Schwerin, Reinhold:
Datenschutzkonformer Einsatz des maschinellen Lernens bei Patientendaten anhand einer föderalen Lernstrategie,
in: 3. Innovationskongress Ulm | Neu-Ulm - Datasience to go, von Schwerin, Marianne; Reichert, Manfred; Urban, Karsten, OPARU, 2022, pages 2-13.
DOI: 10.18725/OPARU-44168
ISBN: 978-3-9820843-3-6

Schlüter, Stephan; Jung, Sejung; von Döllen, Andreas; Lee, Wonhee:
An Alternative to Index-Based Gas Sourcing Using Neural Networks,
in: Energies 2022, 15(13), MDPI, 2022, pages 11.
DOI: 10.3390/en15134708
ISSN: 1996-1073

Schlüter, Stephan; Menz, Fabian; Kojić, Milena; Mitić, Petar; Hanić, Aida:
ANovel Approach to Generate Hourly Photovoltaic Power Scenarios,
in: Sustainability 2022, 14(8), MDPI, 2022, pages 17.
DOI: 10.3390/su14084617
ISSN: 2071-1050

Strahnen, Manfred; Kessler, Philipp:
Investigation of a Deep-Learning based Brain-Computer Interface with Respect to a Continuous Control Application,
in: IEEE Access, 2022, Volume 10, IEEE, 2022, Pages 12.
DOI: 10.1109/ACCESS.2022.3228164
ISSN: 2169-3536

von Schwerin, Marianne:
Bias and Fairness in AI,
in: 3rd Innovation Congress Ulm / Neu-Ulm 2022 - Data Science to go, von Schwerin, Marianne; Reichert, Manfred; Urban, Karsten (ed.), OPARU, 2022, pages 14-29.
DOI: 10.18725/OPARU-44168
ISBN: 978-3-9820843-3-6

von Schwerin, Marianne:
Trusted AI, Can standardization and certification create trust in artificial intelligence?
in: Embedded Software Engineering Kongress, 2022, ELEKTRONIKPRAXIS, Vogel Communications Group GmbH & Co. KG, 2022, pages 293 - 298.
ISBN: 978-3-8343-6305-3

Hwang, Youngseok; Schlüter, Stephan; Choudhury, Tanupriya; Um, Jung-Sup:
Comparative Evaluation of Top-Down GOSAT XCO2 vs. Bottom-Up National Reports in the European Countries,
in: Sustainability 2021, 13(12), MDPI, MDPI, 2021, pages 15.
DOI: 10.3390/su13126700
ISSN: 2071-1050

Hwang, Young-Seok; Schlüter, Stephan; Park, Seong-Il; Um, Jung-Sup:
Comparative Evaluation of Mapping Accuracy between UAV Video versus Photo Mosaic for the Scattered Urban Photovoltaic Panel,
in: Remote Sensing 2021, 13(14), MDPI, MDPI, 2021, pages 11.
DOI: 10.3390/rs13142745
ISSN: 2072-4292

Hwang,Young-Seok; Schlüter, Stephan; Lee, Jung-Joo; Um, Jung-Sup:
Evaluating the Correlation between Thermal Signatures of UAV Video Stream versus Photomosaic for Urban Rooftop Solar Panels,
in: Remote Sensing 2021, 13(23), MDPI, MDPI, 2021, pages 15.
DOI: 10.3390/rs13234770
ISSN: 2072-4292

Liebermann, Simon; Hwang, YongSeok; Um, Jung-Sup; Schlüter, Stephan:
Performance Evaluation of Neural Network-Based Short-Term Solar Irradiation Forecasts,
in: Energies 2021, 14(11), 3030, MDPI, MDPI, 2021, pages 21.
DOI: 10.3390/en14113030
ISSN: 1996-1073

von Döllen, Andreas; Hwang, YoungSeok; Schlüter, Stephan:
The Future is Colorful - An Analysis of the CO2 Bow Wave and Why Green Hydrogen Can't do it Alone,
in: Energies 2021, 14(18), 5720, MDPI, MDPI, 2021, pages 21.
DOI: 10.3390/en14185720
ISSN: 1996-1073

Bowoo Kim, Dongjun Suh, Marc-Oliver Otto, Jeung-Soo Huh:
ANovel Hybrid Spatio-Temporal Forecasting of Multisite Solar Photovoltaic Generation,
in: Remote Sensing 2021, 13(13), 2605, Special Issue Remote Sensing for Smart Renewable Cities, MDPI, MDPI, 2021, pages 20.
DOI: 10.3390/rs13132605
ISSN: 2072-4292

Minjeong Sim, Dongjun Suh, Marc-Oliver Otto:
Multi-Objective Particle Swarm Optimization-Based Decision Support Model for Integrating Renewable Energy Systems in a Korean Campus Building,
in: Sustainability 2021, 13(15): 8660., Special Issue Human-Technology Interaction Sustainable Data Use for Environmental Decision Making;, MDPI, MDPI, 2021, pages 18.
DOI: 10.3390/su13158660
ISSN: 2071-1050

YoungSeok Hwang, Jong Wook Roh, Dongjun Suh, Marc-Oliver Otto, Stephan Schlueter, Tanupriya Choudhury, Jeung-Soo Huh & Jung-Sup Um:
No evidence for global decrease in CO2 concentration during the first wave of COVID-19 pandemic,
in: Environmental Monitoring and Assessment 193, Article number: 751, 2021, Springer, Springer Link, 2021, Pages 15.
DOI: 10.1007/s10661-021-09541-w
ISSN: 0167-6369 / eISSN: 1573-2959

Akharath, Philipp; Altkrüger, Jaqueline; Sahota, Harkiran; Herbort, Volker; te Heesen, Henrik:
Modeling of a Photovoltaic Fault Detection Approach Considering Machine Learning,
in: INFORMATIK 2021, Lecture Notes in Informatics (LNI) - Proceedings Series of the Gesellschaft für Informatik (GI) Volume P-314, Gesellschaft für Informatik e.V., 2021, pages 251-267.
DOI: 10.18420/informatik2021-021
ISBN: 978-3-88579-708-1 / ISSN 1617-5468

Akharath, Philipp; Altkrüger, Jaqueline; Sahota, Harkiran; Herbort, Volker; te Heesen, Henrik:
Modeling a PV Fault Detection Approach with Regards to Machine Learning,
in: EU PVSec Conference Proceedings, 38th European Photovoltaic Solar Energy Conference and Exhibition, , EU PVSEC Proceedings, 2021, pages 1234 - 1237.
DOI: 10.4229/EUPVSEC20212021-5CV.2.20
ISBN: 3-936338-78-7 / ISSN 2196-100X

Müller, Christian; Lunde, Rüdiger; Hönig, Philipp:
Generation of a Failure Mode and Effects Analysis with smartflow,
in: Proceedings of the 30th European Safety and Reliability Conference (ESREL2020), Venice, Italy, 2020, (ed.), 2020, pages 8.
DOI: 10.3850/978-981-14-8593-0, ISBN: 978-981-14-8593-0

Kreuzer, David; Munz, Michael; Schlüter, Stephan:
Short-term temperature forecasts using a convolutional neural network - An application to different weather stations in Germany,
in: Machine Learning with Applications, (ed.), Elsevier, 2020, pages 26.
DOI: doi.org/10.1016/j.mlwa.2020.100007

Hwang, YoungSeok; Um, Jung-Sup; Hwang, JunHwa; Schlüter, Stephan:
Evaluating the Causal Relations between the Kaya Identity Index and ODIAC-Based Fossil Fuel CO2 Flux,
in: Energies, MDPI, 2020, pages 20.
DOI: doi.org/10.3390/en13226009

Hwang, YoungSeok; Um, Jung-Sup; Schlüter, Stephan:
Evaluating the Mutual Relationship between IPAT/Kaya Identity Index and ODIAC-Based GOSAT Fossil-Fuel CO2 Flux: Potential and Constraints in Utilizing Decomposed Variables,
in: International Journal of Environmental Research and Public Health 2020, Vol.17,Is.16, MDPI,2020, Pages 18.
DOI: doi.org/10.3390/ijerph17165976, ISSN: 1661-7827 / 1660-4601 (eISSN)