For over 40 years, the Bertrandt Group has been offering development solutions for the international automotive and aviation industries as well as the mechanical and plant engineering, energy, medical technology and electrical industries in Europe, China and the USA. A total of around 13,000 employees at over 50 locations stand for in-depth expertise, future-oriented project solutions and a high level of customer orientation. The main customers include the large manufacturers and numerous system suppliers.
Jochen Schwenninger, Lead Expert in the “ADAS / AD Software and Function” team at Bertrandt Technologie GmbH Regensburg
Coordination of ongoing projects
Trend scouting and promotion of innovations
Technology consulting
In the area of software for autonomous vehicles in particular, artificial intelligence or machine learning is not an option, but has become THE standard tool for many tasks. Examples of this are the interpretation of sensor data for observing the vehicle environment, the optimization of longitudinal and lateral guidance of vehicles or the prediction of the behavior of other road users. With the increasing networking of vehicles, the aspect of big data is also playing an increasingly important role in the development of new functions.
There is currently a collaboration with UR in the modeling of e-machine controls. We give students an insight into our practical tasks through internships, working students activities and theses.
Through the cooperation with OTH and UR in different departments, there is, on the one hand, the opportunity for technical exchange and, o the other hand, the chance to come into contact with interested students at an early stage. Joint funding projects have also already been initiated.
The automated analysis of data will have a major impact on daily life in the future, even if it is not always superficially present. It is therefore important to create an awareness of the technology and its possibilities, to carry out educational work in order to reduce potential reservations, but aso to be able to objectively assess opportunities and risks.
The topics of environmental observation, especially with the help of LiDAR sensor, are an active research area in which we are also active. A second large topic block is the optimal control of electric machines, which have improved behavior under transient loads thanks to new approaches.
A very specific topic is the detection and classification of road users in LiDAR point clouds. You can get a first impression of the topic on the KITTI benchmark page.
I studied electrical engineering at the University of Ulm and came into contact with the topic of Music Information Retrieval (similarities of songs, automatical pplaylist generation, extraction of instruments,…) during an internship. After graduation, I worked in the speech recognition department for Fraunhofer IAIS for 6 years, where I saw the transition from “classic” speech recognition to the use of neural networks. During this time, I was able to learn a lot about related topics such as text mining or image processing from colleagues. In the last 6.5 years I have used and continued this experience in the development of new functions in the automotive environment. In addition to further training in work, private interest was always a component.
As long as the topics remain factual, dealing with them is easy. What bothers me is that sometimes the topic of AI is either fejected completely and not rationally, depending on the counterpart, or is presented as the only way to happiness. I find both points of view a shame.
I am fascinated by the utopia of being able to automate repetitive or potentially dangerous work and thus being able to offer people freedom for creative activities. In these areas, machines will be inferior to humans for a while.
We are currently working on a project application together with Stadtwerke Regensburg, Continental, AVL, UR and Regensburg e-mobility cluster. The high density of exciting companies in the Regensburg area ensures that interesting topics can always be worked on together. And for a few months now, we have been involved in the AIR (Applied Artificial Intelligence in Regensburg) initiative, which also aims to promote and increase the visibility of AI topics.
From my point of view, AI is a tool that should be presented to students. AI is not the right tool for every task, but students should be able to make this decision correctly. For this they need practical experience in order to be able to assess the possibilities and consequences.
At the moment I am still an interested onlooker and I am pleased that the topic of AI is being promoted through the public relations project.
One of the great dangers of AI is that training high-quality models very often requires both large amounts of high-quality data and enormous computing resources. As a result, there will probably be a concentration on a few global players who then cover entire subject areas in an oligopoly. How can initiatives such as “Knowledge creates² data” control against this? Thank you for taking the time to answer these questions. We wish you a nice day!