VDE: Let's talk about Marelli's research and development activities. What is currently in the pipeline or where are you currently looking very closely in terms of development?
Fetzer: Marelli develops, manufactures and sells electric motors. In addition, we are working on the integration of electric car transmissions, so-called e-axles, and are also active in the field of battery management systems, which are already being used in various vehicles. We are also involved in the development of other components, such as onboard chargers or DC/DC converters. We are also working on novel drive concepts that will fundamentally change battery management as well as DC/AC and AC/DC conversion. We expect this to result in significant advantages in terms of efficiency and packaging - for example, in cooperation with Transphorm, a company that focuses on the development and production of gallium nitride power components.
VDE: There are very many companies that are active in the fields of electric motors and inverters. You mentioned Bosch as one of the competitors. Tell us, what makes the Marelli approach so special?
Fetzer: Marelli was the first company in the world to start working with hairpins in 2008. High-performance electric motors were needed for the KERS systems, and Marelli supplied them. Marelli brought this hairpin technology into series production for the first time in 2019 - in a sports car. In this respect, we have the ability to bring high-end technologies from Formula 1 and thus new innovative approaches from motorsport to the mass market - see Hairpin - this is what sets Marelli apart and is our particular strength.
VDE: If you are currently a user of electromobility, you always have a bit of a problem with the charging infrastructure. Can I find a charging station that is compatible with my vehicle? Are there any plans from Marelli to address this issue or are they concentrating purely on the vehicle side?
Fetzer: We concentrate purely on the vehicle side. Of course, we support our customers and also committees with ideas, but we are not active operationally at component level or product development. This is a conscious decision because we want to focus on the OE market (original equipment).
VDE: Electromobility is certainly a megatrend, which is also partly due to decarbonization. Another important aspect is the topic of networking: autonomous driving, driver assistance systems, shared mobility, etc.; Is Marelli pursuing approaches here to enter the market or serve this field?
Fetzer: We are working on supporting fully automated driving (level 4) and, in the future, autonomous driving (level 5) with sensor technology and are cooperating here with Chinese manufacturers of radar sensors. Following on from this, we have further collaborations with startups in the field of lidar technology, and are also investing in and developing solutions ready for series production. In the process, we are also looking at the software for sensor evaluation or object classification that we want to produce.
With Automotive Lighting, Marelli is one of the world's leading manufacturers of lighting technology. We integrate the sensor technology in the front area of the vehicles. In the electric car, there is no longer any need for cooling at this point because it is no longer required. For car manufacturers such as Mercedes-Benz, the front area is currently becoming a styling element and is being expanded to include lighting functions such as Light Base, etc., into which we then integrate our sensors again.
VDE: When it comes to autonomous driving, experts argue about whether video sensors or lidar technology will prevail. Which side is Marelli on?
Fetzer: There are indeed many opinions. And Elon Musk (Tesla) probably has the most extreme view of all: According to him, eight video cameras are enough for fully automated driving. Companies like Waymo and Cruise, on the other hand, say that autonomous driving needs various sensors, including lidar (light detection and ranging), radar and ultrasonic sensors, as well as video cameras. Tesla models in the U.S. are reportedly being prepared for fully automated driving. However, radar sensors are not to be used for this. Mr. Musk must therefore already have evidence that what he is claiming can actually work. Otherwise, he would not act in this way.
When considering this, the use cases should always be kept in mind. Let's take Level 5, for example: driving without a steering wheel and without pedals. In other words, the vehicle drives autonomously. At the moment, I cannot conclusively judge which sensors are needed for this. To me, camera technology seems to be a major challenge: What do I do in fog, for example? Do I have to park the car? Several radar or lidar systems would certainly be helpful here.
Much also depends on the capabilities of the detection or processing software. What can be implemented in terms of AI (artificial intelligence) & machine learning in this context? And how much support is needed via edge computing - i.e. 5G/6G from the cloud. Waymo and Cruise have a clear strategy here: they don't need edge connectivity. But they do have an enormous amount of sensors and corresponding processing technology to be able to drive autonomously. The first demonstration projects in San Francisco prove that this can work. However, this is certainly not a vehicle for series production because it is not (yet) suitable for everyday use. And finally, in the next few years, the question will certainly arise as to how much user experience will be needed for the individual in automated driving?
Equipping a sports car with a fully automated driving system makes no sense. However, providing support for everyday driving by warning of dangerous situations, automatically braking the electric car, or intervening in the steering is much more likely to make sense. All of this increases safety on the road. It will be exciting to see in which applications fully automated, and autonomous driving will become established.