Innovation Center
Fully Autonomous Driving

On the way to fully automated driving

The further development of driver assistance systems currently in series production in the direction of autonomous driving is a topic for the coming years. Engineers and information scientists participating in the development process face new challenges due to the associated increase in the complexity of systems internal to the vehicle. Autonomous driving systems are increasingly becoming distributed. Digital maps are generated in the back-end through crowd sourcing and provided to autonomous driving systems.

We accompany all manufacturers and suppliers on the way towards realising autonomous vehicles. Through our participation in research projects and cooperation with universities, we provide our customers with access to expert knowledge at the latest state of research.

Conceptual design and feasibility studies
Rapid prototyping (proof of concept)
Implementation and integration into the overall system
Testing and validation

Highly current, highly accurate maps

Autonomous vehicles rely to a large extent on highly current, highly accurate digital maps. These maps are generated in back-end systems, for example through crowd sourcing, and validated in real time. Maps can also be realised directly in the vehicle as self-learning maps.

C2C – Car2Car communication

Car2Car communication exchanges sensor information between vehicles. Based on this information, an environment model can be generated to support accident avoidance strategies among other things.

Simulation and virtualisation

More than 10 million travel kilometers are needed to ensure that autonomous vehicles are at least as safe as manually steered cars. Virtualisation is a possible solution – testing a (sub)system in a virtual environment. Simulations are used in early development phases and all the way to overall validation.

Focal points: data analysis and big data

Big data is the promise of gaining new information through the analysis of large data volumes. Combining technical expertise with IT competency is crucial in order to gain such insights. Building on these competencies, we realise complete solutions optimally tailored to the customer for the administration and analysis of all resulting technical data in the automotive lifecycle.

The goal is to provide our customers with data assistance systems that relieve technical experts of routine tasks and offer optimum support for technical data analysis. Visualisation concepts and UX analysis and design play a crucial role for interactive data analysis.

Approach and Proof of Concept

In collaboration with our customers, we analyze the problem and develop concepts for the solution. We validate these solutions as feasible in the proof-of-concept phase.

Realization

Realization of the complete solution using the actual methods and technologies as well as the commercially available tools (best in class).

Operation

We host the complete solution, i.e. the provision and operation of the infrastructure, the delivery of services as well as the performance of analyses and evaluations, the problem management and many other items. Also crucial in this process are the data protection, the secure and diligent data management on behalf of the client and the secure access to the data.

Project references

Virtual test process

The validation of driver assistance systems on actual roads is elaborate and not reproducible. Validation is therefore performed in a simulator on the basis of trials. The project scope encompassed setting up simulators, connecting the driver assistance systems and preparing tools (map conversion for example). Scenarios were prepared for conducting trials and support was provided for carrying out the trials.

ADTF, VTD, C++ and others

Questions? Leave us a message!

Your contact

Responsive image
Albert Habermann
Head of Innovation Center Fully Autonomous Driving
Munich/Garching