Projects

DyNaMo

Safe and sustainable mobility in the city of tomorrow (DyNaMo)

AI assistance systems for cycling

Traffic accidents are one of the most common causes of death for children and young adults up to the age of 29 worldwide. The project "DyNaMo: Safe and sustainable mobility in the city of tomorrow" is committed to making cycling safer. To this end, it is pursuing a holistic approach:
To prevent traffic accidents in the long term and increase the safety of cyclists, DyNaMo is researching the basis for

• the implementation of suitable cycling infrastructure measures based on an AI-based analysis of road user behavior
• the development of training measures for cyclists, also based on an AI-based behavioral analysis
• the development of new, legally compliant AI-based assistance systems for cyclists and
• even better recording and anticipatory consideration of the behavior of cyclists in AI-based autonomous driving.

In addition, there are a number of other important points for cycling safety that are not the focus of DyNaMo. Such important points are e.g. reduction of motor vehicle speeds, bicycle helmets, visible clothing...

The project conducts research from the perspectives of computer science, law, traffic science and traffic psychology.

The aim of DyNaMo is to use AI in a holistic socio-technical approach to research a variety of methods and measures for a significant increase in road safety and thus sustainable mobility for the city of tomorrow, using cycling as an example.

The research objectives of the LOEWE focus area DyNaMo are (1) research into the AI-based recording of cyclists' driving behavior. The recording of this behaviour, e.g. raising the arm to turn, is carried out using vehicle and wearable-based sensors and is researched using various AI methods (e.g. transformers, hyperparameter optimization, transfer learning, sensor fusion). The data can then be used to research (2) effective infrastructure and training measures that make a substantial contribution to road safety for safe and sustainable mobility in the city of tomorrow. In order to address these goals, the expertise of computer science, traffic psychology, traffic planning and law must be combined.

Funding phase: January 2025 to December 2028

Contact: Prof. Dr.-Ing. Klaus David, University of Kassel, ComTec

Project partners: Departments of the University of Kassel

• Department of Public Law, IT Law & Environmental Law, Prof. Dr. Gerrit Hornung (LL.M.)
• Department of Cycling and Local Mobility, Prof. Dr. Angela Francke
• Department of Traffic Planning and Traffic Systems, Prof. Dr. Carsten Sommer
• Department of Intelligent Embedded Systems (IES), Prof. Dr. Bernhard Sick
• Department of Communication Technology (ComTec), Prof. Dr. Klaus David

Funding: State Offensive for the Development of Scientific and Economic Excellence (LOEWE), Hessian Ministry of Science and Research, Art and Culture

PharmaXR

The PharmaXR research project, funded by the German Federal Ministry of Education and Research, enables training using XR and AI technologies for employees in the pharmaceutical industry. PharmaXR uses the possibilities of eXtended Reality (XR) to design completely new digital training and learning scenarios for quality-critical value creation processes. To this end, the adaptability of companies is also considered at an organizational level. Sensor-based activity recognition with artificial intelligence should make it possible to digitally monitor the correct execution of employees' production steps and provide feedback in the event of errors. The research project has the potential to fundamentally change the way employees are trained in the pharmaceutical industry.

PharmaXR, funded as part of the "Industry 4.0 - Adaptability of companies in the value chain of tomorrow" initiative (In Wandel), is developing an advanced digital training environment. This makes it possible to provide employees with targeted training even before they work in real production facilities. PharmaXR offers considerable advantages, particularly when planning and setting up production processes for sensitive products such as vaccines.

One particularly noteworthy feature of PharmaXR is the ability to virtually enter a cleanroom. This feature allows employees to familiarize themselves with the specific requirements and protocols of these highly sensitive environments before actually working on site. This not only reduces training times, but also increases efficiency and safety in pharmaceutical production.

ComtTec is researching AI-based algorithms for activity recognition using wearables. Activities, i.e. work steps, are to be recognized via data recording with smartphones and smartwatches, for example. This makes it possible to ensure the quality of production processes during training using XR and AI technologies. This also provides the basis for a quality assistance and documentation system.

Funding phase: January 2025 to December 2027

Contact: Prof. Dr.-Ing. Klaus David, University of Kassel, ComTec

Project partners: PTS Training Service, FLAVIA IT-Management GmbH, Sigmund Lindner GmbH, University of Kassel, Wirtschaftsgenossenschaft deutscher Tierärzte eG, CUP Laboratorien Dr. Freitag GmbH as anticipated partner

Sponsor: Federal Ministry of Education and Research (BMBF)

DFG project Achieve

The DFG project"Architecturesfor VulnerableRoadUsers" (ACHIEVE) #516946933, is a cooperation between Prof. Klaus David from the Department of Communication Technology (ComTec) at the University of Kassel and Prof. Delphine Reinhardt from the Department of Computer Security and Privacy (CSP) at the University of Göttingen. The main objective of ACHIEVE is to investigate different architectures for cooperative systems with a focus on collision protection and privacy of pedestrians and cyclists. The performance of these architectures will be measured by quantitative evaluations.

There are different types of system architectures that will be investigated in the project. These include peer-to-peer architectures, client-server architectures and hybrid architectures. These differ, among other things, in the type of communication (direct, cellular or hybrid), the location of the calculation of the collision protection algorithms (on mobile devices, on-board units, servers or distributed) and the type of warnings.

The first step is to identify the various possible system architectures. Next, specially developed key performance parameters for collision detection and privacy compliance are defined. These parameters are used to evaluate and rank the architectures accordingly.

The evaluation is therefore based on quantitative measurements that are intended to accurately capture the performance of each architecture in terms of collision protection and privacy. This makes it possible to better understand the advantages and disadvantages of the different architectures and to make informed decisions for the further development of cooperative systems in the field of road traffic.

Funding phase: April 2023 to March 2026

Contact: Prof. Dr.-Ing. Klaus David, University of Kassel, ComTec

Project partner: Prof. Reinhardt

Sponsor: German Research Foundation

ZUKIPRO II

The follow-up project Zukunftszentrum für menschzentrierte KI in der Produktionsarbeit Hessen (ZUKIPRO II) will create a practice-oriented format for providing information, advice and training. To this end, participatory work and technology design will enable small and medium-sized enterprises to make better use of the potential of digital technologies in work and business processes. The ZuKIPro II project is intended to be a contact and transfer point for SMEs, sponsors, technology providers and social partners.

In ZUKIPRO II, ComTec offers workshop formats in the field of AI practice to give SMEs an easy introduction. ComTec will demonstrate how to create a suitable AI model from raw data. Through these formats, the Future Centre enables companies and their employees to identify and analyze the challenges in their own company, to independently initiate, design and implement learning and solution processes and to evaluate the results.

The overarching aim of the project is to create low-threshold access to the topic of AI for SMEs. To this end, SMEs are supported in the participatory, co-creative introduction and human-centered design of AI systems. Across modules, the Future Center informs and advises SMEs in line with their needs and enables them to implement concrete solutions.

Funding phase: 2023 - 2026

Contact: Prof. Dr.-Ing. Klaus David, University of Kassel, ComTec

Project partners: RWTH Aachen, Institute for Technology and Work, Technical University of Darmstadt, Chamber of Industry and Commerce Kassel-Marburg and Regional Management Nordhessen and BTQ-Bildungswerk.

Sponsors: Federal Ministry of Labor and Social Affairs, European Union