STADT:up — Pioneers in complex spaces
How STADT:up is advancing automated driving in urban areas.
STADT:up (Solutions and Technologies for Automated Driving in Town: an urban mobility project) is a research and development project in which Bosch Research has joined forces with 21 other companies from the automotive industry and research institutions. Since January 1, 2023, they have been working together to develop concepts and pilot applications for fully automated driving in urban areas, with a particular focus on complex traffic situations. AI-based methods will play an important role in achieving this ambitious goal. Bosch Research is not only a project partner in STADT:up, but also hosted the kick-off event at the beginning of March at the research campus in Renningen.
The future of mobility is autonomous — but there is still a long way to go. Before traffic can be fully automated, many problems need to be solved and hurdles overcome. Particularly complex situations in urban areas are a challenge for automated vehicles — and it is precisely this path to the urban mobility of the future that the STADT:up project aims to pave.
"Implementing automated mobility into the dynamic urban system is a challenging task. In STADT:up, we are using data-driven models and the latest AI methods to develop and demonstrate automated driving functions that can handle even complex urban traffic scenarios. With vehicle manufacturers, automotive suppliers, and research institutions, STADT:up brings together extensive know-how and expertise, and we look forward to working with our colleagues," explains project coordinator Dr. Lutz Bürkle from Bosch Research.
In order to tackle the ambitious range of tasks, the project work is divided into five subprojects — Perspectives on Urban Mobility, Human Factors, Environment and Context, Situation Analysis and Planning, and Automated Driving. Bosch Research is significantly involved in 4 of the 5 STADT:up subprojects.
Environment and context
Sensors such as camera systems, lidar or radar and special algorithms are for automated vehicles what eyes and brains are for humans. This subproject investigates how they can be trained to reliably recognize and interpret challenging urban scenarios even under difficult conditions. Bosch Research is focusing on 360° object recognition and tracking based on multimodal data using deep learning and modern network architectures.
Situation analysis and planning
The enhanced perception of the environment provided by data fusion allows the project participants to focus intensively on interaction and cooperation issues. This includes various decisions that the automated vehicle must make in urban traffic — for example, aborting a merging maneuver if other road users are not clearing a gap. In this subproject, Bosch Research wants to develop a planner that can handle complex situations. A hybrid AI approach (AI= Artificial Intelligence) is to serve as a basis — in other words, a combination of rule-based and AI models.
In this subproject, STADT:up is investigating how humans and automated vehicles can interact and communicate with each other. Bosch Research wants to use data from driver observation and environment detection to determine the driver's state with the help of an algorithm. This will ensure that the driver remains appropriately involved in the traffic situation during automated driving.
In this final subproject, the individual subprojects come together to form an overall picture: The project partners involved will make automated driving in urban areas a tangible experience — both in simulators and with test vehicles in real traffic. Here, Bosch Research intends to implement AD software (AD = Automated Driving) that can cover challenging inner-city use cases. On a reference route through Renningen and Malmsheim, the application clusters "complex intersections" and "merging and avoiding obstacles" are to be tested in real traffic and thus made tangible.
The total budget of the project is around 62.8 million euros, of which 33.5 million euros are provided by the German Federal Ministry of Economics and Climate Protection. The project duration is three years. Bosch Research is proud to be part of this project community and to coordinate the project under the leadership of Dr. Lutz Bürkle.