Digitalization enables development methods of a completely new Dimension

You’re Chief Digital Officer at IAV – what path did you take to get here?
My interest in automation goes back a long time, not least because I studied automation and control engineering. I have been working intensively with digitalization for around six years now. At IAV, I want to lead the way in digital change and bring the latest research findings to series development. To lay the foundations for this, I founded the Research Center at IAV. In early 2018, a joint research lab was set up in Kaiserslautern in collaboration with the German Research Center for Artificial Intelligence (DFKI). One year later, I was given responsibility for our Digital Lab. About a year ago I was appointed CDO.

What happens in the Digital Lab at IAV?
In the Digital Lab, we take care of the transfer of know-how. We are a node and network for key technologies of digitalization and new ways of working. The Digital Lab works for the whole of IAV and drives the company’s digitalization forward. Our aim is to combine knowledge from the various disciplines – from autonomous systems to the use of the latest AI technologies – with know-how from the various fields. This is where domain meets technology. This mission is driven forward by our innovation process that involves all IAV colleagues and enables cross-divisional networking. Through innovation campaigns and subsequent pitch sessions, colleagues can submit and implement their own ideas. A jury from the technical areas decides on which topics will be worked on further. In test sprints, our employees then develop minimum viable products (MVP) and simple demonstrators in a very short time. This results in products and services that we develop further with our customers, and strategies that continually improve our processes. Our colleagues then take the newly acquired knowledge from the Digital Lab back to their specialist departments – a constant cycle of know-how transfer.

Another important component is the exchange with our research network, such as the Research Lab, which we operate together with DFKI, and Cyber Valley in southern Germany, of which we are a founding member. It is one of our core competencies to reliably develop new solutions to production maturity. By bridging the gap to basic research, we ensure that new technologies and methods find their way quickly and efficiently from research, development and the first prototypes into production vehicles.

In which digitalization technologies do you see the greatest potential for the automotive industry?
An extremely large amount of data is generated in the automotive sector. There is a lot of potential there, but the systematic evaluation of this data is a major challenge. We must focus on smart information gathering. Here, it is a case of generating the maximum information content from the existing data so that we can permanently reduce redundant measurements.

I also see a big lever in virtualization: The Digital Twin, for example, opens up undreamt-of possibilities for better safeguarding of systems compared to testing. In the field of predictive health monitoring, methods for detecting anomalies in measurement series from different control units have become established. This enables us to react to errors at an early stage. On the one hand, we can minimize the time needed for error correction and on the other hand we can reconfigure the system based on the quantitative estimation of the error. This way, the system behavior remains in the optimum despite an error. I also see many opportunities in the use of reinforcement learning, which allows models to be developed that independently find an optimal solution for the respective task within a target system. Even the system reaction flows into the optimization process and the model adapts itself until it reaches the desired behavior.

In addition, technologies such as Safety Supervisor help to secure neural networks in control units. Only those who also know and master the weaknesses of the AI are able to exploit the potential.

Every day, we deal with increasingly complex systems, higher quality requirements, increasingly documentation effort and shorter development times. Digitalization helps everyone in the industry to meet and master these challenges effectively.

How digital is IAV already positioned?
We have come a long way, but for us it is always a matter of finding new or even better solutions. As a leading development partner, we cannot accept standstill and have therefore worked successfully on core technologies in the past. For instance, we are excellently positioned in the field of predictive maintenance, in safeguarding neural networks (Safety Supervisor) and in control and regulation with AI (Reinforcement Learning). The task now is to deploy these technologies across the board at IAV from administration to engineering. To achieve this, we need to extend our mindset, expand knowledge management and create tools that all employees can use – not just individual specialists. Our customers will also benefit enormously from this.

Where do you want to move the company in the coming years?
It is important to me that IAV is digitalized as a whole – from technology (i.e. our divisions that work directly for our customers) through IT to administration. We are also working on adding new business models to our portfolio: In the future, IAV is to distinguish itself even more as a tech provider without losing our position in terms of implementation expertise. We want to continue to bring the latest developments from research into series production. However, customers should turn to us with all questions relating to digitalization and the use of new technologies. To this end, we want to offer a broad range of modules from which we develop individually tailored technical solutions for every problem and every domain.

How do you see the future?
With a high degree of optimism! We can rely on excellent employees in all areas. They have the know-how to drive digitalization forward. At the same time, we have the best development methods and customers who entrust us with their most demanding and success-critical topics and projects. There can be no better basis for us as a development partner. And this trust ensures that our expertise is increasingly appreciated and in demand beyond the automotive industry. And this is not an end in itself, because in a digital and networked world we need universal, cross-system solutions – solutions that know no industrial sector boundaries. And that is what we already offer our customers today.

The article was published in automotion 03/2020, the automotive engineering magazine of IAV. Here you can order the automotion free of charge.