It's not always about the bigger picture. According to Nico Rosberg, the details are far more important for making progress. The ex-Formula 1 World Champion described one of the biggest problems left solving in automated driving at the 10th chassis.tech plus.
Keynote speaker Nico Rosberg at the 10th anniversary of the chassis.tech plus symposium.
Uli Regenscheit / ATZlive
Automated driving, vehicle dynamics and chassis development are intertwined. Nico Rosberg underlined the importance of vehicle dynamics in his keynote speech at the 10th chassis.tech plus symposium in Munich. "Solving motion sickness is the most urgent problem that we have to solve for automated driving in the future", said the Wiesbaden native. A lot of passengers already get nauseous when they're sitting in the back seat of a normal car. Even Rosberg's kids, if he turns too sharply. It would be terrible if passengers spilt coffee in an autonomously driving car just because the people mover is steering too abruptly. The solution he proposes for this problem is an active chassis that prepares the driver for upcoming manoeuvres by using data from the cloud. In chassis development, Rosberg and TRE rely heavily on simulation, i.e. virtual development using powerful models to depict reality. There's one thing that's very important to him, which he carried over from his Formula 1 career: "I'm asking developers to pay more attention to details, not just the bigger picture. Improving many details will sum up to a better end result in the long run."
Nico Rosberg is the co-owner of TRE GmbH in Neustadt an der Weinstraße, where 50 engineers have been driving chassis development for standard passenger cars for over 20 years. TRE is now associated with IAV. But for most, Rosberg is better known as the 2016 Formula 1 World Champion. He is also involved in electric mobility and the Formula E racing series. TRE developed the chassis for the Schaeffler Mover, which has four wheel hub motors and all-wheel steering. Rosberg still loves turning on the spot, like he did back in his F1 days: "So you can do a 360, that's what I love to do", he says.
10th chassis.tech
Before the keynote speech, Dr. Alexander Heintzel, Editor-in-Chief of the ATZ-MTZ Group, and Professor Peter E. Pfeffer, the conference chairman from HS München, welcomed 450 participants to the Bayerischer Hof hotel. The previous year's record number of applicants could not be broken, but it was higher than 2017, with a participation of 20 percent from the non-German-speaking sector. This anniversary chassis.tech plus is taking place in conjunction with partner TÜV Süd and the sponsors AVL and ZF for the tenth time, and has been taking place for 20 years. Pfeffer: "It's okay to look back on things after 20 years." The symposium was first held in 1999 as the "chassis.tech". It was meant to mark a shift from mechanics to mechatronics and electronics. Pfeffer believes that a similar shift is taking place now with the automation of driving and the electrification and digitisation of chassis development.
The first brake.tech was held as a "Mµ conference" in 1998 by Professor Eberhard Drechsel with TÜV Süd. The idea of also holding a chassis.tech event was inspired by the Mercedes A-Class Elk Test fiasco. The first conference was chaired by Prof. Bernhard Schick in 1999. The first tire.wheel.tech was held in 2004, adding tyre development to the conference series. This was followed by the steering.tech in 2008, which addressed steering. All four conferences were organised as a collaboration between TÜV Süd and ATZ magazine to form the overall event chassis.tech plus in 2008. This was the answer to the falling number of participants caused by the economic crisis. And this is the format that the chassis symposium in Munich still has nowadays.
Braess demands commitment to the chassis
On this historic occasion, Prof. Hans-Hermann Braess, Honorary Chairman of steering.tech and former BMW Board Member for Development, gave a welcoming address in which he invited the local engineers to "commit themselves to the chassis, as it will pay off". The inventor of Porsche's Weissach axle gave the visitors a historic outline. Back in the 1950s, designers were satisfied if they managed to adopt and implement old developments from the pre-war period. In the 1960s, new axle concepts and calculation methods arose. The massive expansion of the "Autobahn" network in Germany also contributed to this, as higher speeds and shorter travel times could now be achieved.
Braess: "I should also mention the objectification of driving behaviour based on theoretical and physical principles. Prior to this the only input flowing into car construction was reports from test drivers." Braess's first article in ATZ in 1965 on maintaining directional stability of the motor vehicle when driving straight and his other one in 1975, on determining the ideal negative scrub radius, also contributed to this. In the 80s and 90s, the simulation possibilities that computer technology offered were being used more and more. Electronics found their way into the chassis, which was often pioneering work. "The tasks today are much more comprehensive and complex", Braess noted, addressing the younger generation. For example when it comes to finding solutions for new vehicle concepts in the city that have a small turning radius. However, nowadays, much better and more powerful simulation and calculation methods are available for this purpose.
Eichler does not take sensor fusion lightly
In his keynote speech, Friedrich Eichler, Head of Chassis Development at Volkswagen, tried to find out how automated driving (AD) is changing the demands placed on classic chassis development. He thinks that the topic would become far more accessible if the relevant systems were more reliable and user-friendly for drivers and passengers. He believes that the real transition to automated driving hasn't been made yet, since even level 2 and 3 systems have been hard to homologate. Although the longitudinal and lateral guidance of the car on paved roads such as motorways has been mastered, driving off-road, where dirt can get on the sensors, is still a big challenge.
Another problem is traffic sign recognition: too many signs at a standard American traffic light have been proven to throw off the camera. Legislators in different regions also set different standards, which makes it difficult to manufacture one universal car. Sensor fusion is also a very important problem that needs solving. Although the vehicle already recognises the surroundings very well by camera, it is still impossible to combine these data with those of the radar precisely: "It's not easy, definitely not a walk in the park", says Eichler. It will also require large development teams as soon as laser data have to be processed.
The chassis.tech plus can still be visited tomorrow (26 June 2019) at the Bayerischer Hof. This will be the second day of the event. There will be numerous talks on chassis, steering, brakes and wheels/tyres.