»Automated driving can save lives«

The trend toward automated driving drives the automotive industry just as much as car drivers. In an interview, Berthold Hellenthal, Head of the Progressive Semiconductor Program at Audi, explains how far the technology already is, how he assesses the acceptance of customers and what challenges the industry currently faces. His forecast: Automated driving will save lives – if the development of the quality and reliability of electronic components keeps pace with the rapid development in vehicle electronics. Around 400 international experts exchanged ideas and thoughts on the topic at the ESREF 2016 symposium from 19-22 September in Halle (Saale).

automatisiertes Fahren autonom fahrerlos Interview
© Audi AG
Berthold Hellenthal heads the Progressive Semiconductor Program at Audi.

That automated driving will become accepted is considered a foregone conclusion. What is your forecast, Mr Hellenthal: When will the car driver who steers, brakes and keeps an eye on traffic themselves be extinct?

I don’t believe that will ever happen. Automated driving can make an enormous contribution to more safety, comfort and efficiency in road traffic. But I also don’t begrudge drivers in future the fun that can be had as a pilot, with your hands on the steering wheel and gearstick and feet on the pedals. When automated driving actually becomes established is decided not only by technological development but also be the demand of customers.

When do you expect the technology to be ready?

I think that in ten years time we will see many cars on the road that drive by automated means, which organize themselves and communicate with other components of the digital world, for example, the home, multi-storey car park, repair garage or with traffic management systems and other cars.

You are one of the keynote speakers at the European Symposium on Reliability of Electron Devices, Failure Physics and Analysis (ESREF) from 19-22 September 2016 in Halle. Why are such conferences important for Audi?

The three most important innovation drivers in the automobile industry are electrification, networking and automated driving – in all three areas we depend on efficient and reliable electronic components, for which new requirements are set all the time. We used to have significantly longer innovation cycles and more time to develop the reliability of electronic components to the necessary degree before they were used in a car. Today many premium manufacturers have become early adopters of electronic components, for example, because customers expect the same performance from the infotainment system in their car as they are get with their tablet or smartphone. We therefore have to act faster – and along the entire value-adding chain – in order to ensure the necessary quality and life. At the ESREF 2016, around 400 international experts come together; each brings their know-how and perspective. Only through this exchange between the leading experts from the automobile industry, semiconductor production and science can we pave the way quickly for reliable automotive electronics of the future.

Tesla has recently caused negative headlines: A vehicle of the US manufacturer caused a fatality in autopilot mode. Tesla has now admitted other accidents involving automated driving cars. Doesn’t that show that the technology is not yet fully developed?

We at Audi make a clear distinction between two systems: assistive and piloting systems. According to the international classification for autonomous driving there are 5 levels. The assistance systems currently in our portfolio are on level 2 – partly automated driving with combined longitudinal and lateral driving of the vehicle. The driver always has the responsibility and is at all times fully responsible for the situation. We envisage a gradual development from the car completely controlled by the driver, as in level 1, to driving without any human intervention, as in level 5. To achieve this we should take the necessary time to find safe solutions for the further levels. Current legislation in Germany also only allows temporary automated driving, which can be overridden or switched off by the driver at any time.

What about car drivers’ acceptance in your view?

Ultimately, one can look at autonomous driving as a combination of several assistive systems, which is why I am very confident about it. Technologies such as parking assistance, emergency braking functions or traffic jam assistance are already in use, and car drivers very much appreciate this technical support. However, automated driving increases the importance of the reliability of electronic components. Sensors and chips must function faultlessly, without error, under all conditions and on redundant levels, for the entire life of the car. This is a challenge for us as a manufacturer and for our suppliers. This is why it is so important for approaches to solutions and new ideas to be discussed at a symposium such as the ESREF 2016. I am certain that we will be able to find viable solutions here. We mustn’t forget: more than 90 percent of all car accidents are caused by human error. Automated driving can prevent a large number of these accidents and thus save lives. I am convinced that when automated, networked driving is used extensively accident figures will fall significantly.

In which areas do the greatest challenges for the manufacturers currently lie?

On the technical side the trend to miniaturisation in the semiconductor industry continues. We have now reached an area less than 10 nanometres. The smaller and more complex the electronic components are, the more difficult it naturally becomes to track down defects and weakspots. Here the suitable failure diagnostics instruments and methods have to be developed so that we can ensure optimum quality control in the development phase. We also need innovative solutions for the data volume produced by automated driving. An example: In one of our research vehicles, which is equipped with an adaptive parking assist system and processes the data from seven installed cameras in parallel, the quantity of data is already 350 megabyte per second. The most important challenges of microelectronics in the automotive sector include new materials such as gallium nitride, which can also enable improved performance in smaller components. With such materials we must also understand, down to the very smallest detail, how they perform in use, even under extreme conditions. All this requires a large amount of research.

Finally, a personal question: Have you already dared to sit in a car without driver yourself?

Yes, I have. By the way, it was also in Halle, where the ESREF 2016 will take place. We presented a research vehicle for automated driving at a joint event of the Fraunhofer IMWS and the National Academy of Sciences Leopoldina. I had no queasy feeling when I got in. But I must admit: The view of an empty driver’s seat in a moving car is an unusual experience.