The digital twin is the foundation of the highway of the future

Scientists around the world are urgently conducting research into the vehicle traffic system of the future. After all, connected and autonomous driving has the potential to make mobility significantly safer and more reliable. Studies that examine better ways to provide real-time information to drivers, assistance systems, self-driving vehicles and road infrastructure operators are serving as the foundation for this research. With the aim of safeguarding interaction between the participating systems in the near future, fortiss is collaborating with research and industry partners as part of the Providential++ consortium project to establish the required digital infrastructure.

Making mobility safer and more comfortable

Apart from are working on improved safety, the participating project partners from research and industry are also creating new opportunities for more targeted traffic management and optimization, such as traffic congestion avoidance. To achieve these goals, the consortium partners – TUM (project lead as of 2020), Cognition Factory, Deutsche Telekom, Elektrobit, Intel and Valeo, plus Huawei and Rohde & Schwarz, two non-members involved in the project – are carrying out development activities and field tests on the A9 digital test field highway and extending into urban areas. These activities include researching and testing information flows in highly-automated vehicles in concert with the participating communications and road infrastructures.

The next set of project milestones were introduced in detail during the initial consortium meeting in December 2020. These include expanding the A9 test route an additional two kilometers and employing 59 new area scan cameras, radar sensors and lidar sensors (a radar-like method for measuring distance and speed) on a section of the test highway between Munich and Garching. The project team will begin to capture data in urban areas in the spring of 2021.

Video: Providentia – Safe, connected driving on the digitally-enhanced highway.

After working with its partners within the Providentia precursor project to reach the milestone sensor data from the highway in real-time, fortiss is now in a position to offer a complete representation – or digital twin - of the actual traffic environment. The results of the first three years of research clearly illustrate that in principle, the digital twin is a viable tool in traffic situations.

The follow-up project Providentia++ is now focused on significantly fine-tuning the digital highway twin, but most of all making it more robust. Because it can deliver targeted information that vehicles are unable to capture with their own sensors, the digital twin developed by the project team will play a crucial role in future assistance systems. 

fortiss: making the digital twin more reliable

And it’s precisely at this central point where fortiss comes into play. After leading the consortium project from 2017 till the end of 2019, the Research Institute of the Free State of Bavaria is now contributing significantly to making the digital twin as reliable as possible. fortiss scientist Annkathrin Krämmer is thus switching her focus to ensuring that the system provides solid results and predictions even in poor visibility conditions, such as at night, or in challenging traffic situations. Three issues in particular are at the forefront here:  merging the data, including tracking, movement predictions and development of a real-time platform. In order to determine the current performance of the system for tracking vehicles on the road, the German Aerospace Center (DLR) is enhancing it with data from the air (see the article below A birds-eye view: more precision for the digital twin). The goal is to establish a ground truth that serves as a basis for all further developments.

An extensive evaluation based on this data has already demonstrated that the Providentia system is highly precise and detects vehicles. The goal now is to provide additional robustness, thus allowing this capability to be transferred to more complex traffic scenarios. Particularly when there is high vehicle density, the idea is to enable active occultation management, which means vehicles can be precisely tracked even when briefly concealed. A neural network-based movement forecast will make it possible to determine future trajectories with the highest degree of probability.

In order to handle hardware outages, fortiss is relying on a real-time platform tasked with “application migration”, which involves relocating applications to other hardware if the primary hardware fails. fortiss will continue to perfect this approach during 2021 in order to algorithmically improve the fortuna test vehicle and the Providentia system, as well as to develop a demonstrator for real-time applications.

Scientist Annkathrin Krämer supervises the Providentia++ project at fortiss.

Target in 2021: the fortuna test vehicle is to be algorithmically improved.

Radars on the A9 autobahn test field record traffic.

  Marketing & press

Your contact

Marketing & press