The Smart Mobility business unit has many years of experience in the areas of environmental perception, localization, and artificial intelligence for highly automated mobility systems. Our expertise ranges from autonomous shuttles in local public transit to cooperative driving functions for connected vehicle fleets and technologies for automated rail operations. The goal of our research is to make mobility safer, more efficient, and more sustainable.
A particular focus of our work is on AI-based perception systems. Fraunhofer FOKUS covers the entire development chain: from data acquisition and automated annotation, through the training of neural networks, to the validation and verification of the developed AI models. Using our own tools and methods, we create robust solutions specifically tailored to the requirements of complex mobility and rail systems.
For automated and digitized rail operations, we develop intelligent systems that enable rail vehicles to reliably sense their surroundings and precisely locate objects and people. By fusing sensor data from laser scanners and cameras, our solutions detect tracks, signaling systems, platforms, people, vehicles, animals, and other static and dynamic objects in real time. In this way, we lay the foundation for safe automation functions and the operation of automated trains even outside of strictly controlled test environments.
These technologies were successfully further developed in the Berlin Digital Rail Operations (BerDiBa) project. This led to the development of AI methods for the reliable detection of objects in the track environment, as well as powerful methods for generating high-quality training data for rail applications. Together with partners from industry and research, Fraunhofer FOKUS thus made an important contribution to the digitalization and automation of rail transport.
In addition, we develop digital twins for rail vehicles and railway infrastructure. By linking vehicle, infrastructure, environmental, and sensor data, we create virtual representations of railway operations that enable continuous monitoring, predictive analytics, and support for automated operations and maintenance processes. These digital twins enable the remote control of trains from the control center. Digital twins thus form a central foundation for efficient, resilient, and sustainable rail operations.
With our expertise in artificial intelligence, sensor fusion, localization technologies, and digital twins, we support rail companies and infrastructure operators in developing safe, high-performance, and trustworthy solutions for the automated rail transport of the future.
Fraunhofer Institute for Open Communication Systems