ZF’s Autonomous Systems Head Takes a Pragmatic View of the ‘Levels’ Playing Field

Well aware of the hype that surrounds autonomous vehicles, ZF’s AV lead knows cost will be crucial and customer persuasion needed.

September 1, 2019

Stuart Birch

“We are constantly asking this question: Who is the customer likely to buy an SAE Level 4 autonomous vehicle in the consumer passenger market?” This is one of the key queries that drives the thinking of Torsten Gollewski, Head of Autonomous Mobility Systems at Tier 1 global technology supplier ZF, a company that is playing a crucial role in enabling the next generation of mobility.

A technology pragmatist, Gollewski (below) is well aware of the hype that surrounds autonomous vehicles (AVs). There’s a commonly held belief outside the automotive industry that all vehicles will soon be automated and electric. “But regarding ADAS and AV passenger cars, you have to convince the customer to buy the system about which they may have great expectations,” Gollewski said. Cost will be crucial, and the higher the SAE Level, the more customer persuasion will be needed, he added.

“If you go from Level 2 to Level 3, it must deal with many complex traffic situations in every geographical area of the world. However, the driver may have dozed, suddenly needs to awake and take instant and correct decisions. To begin with, we have to decide how to wake that driver. It would not be possible to limit distribution of Level 3 because that would also limit market size.” But Level 3 is technically possible and available in premium cars in certain use cases within three to five years depending on model cycles, assured Gollewski.

As for Level 4, he confirmed that ZF has “everything in its portfolio” and in series production to create a Level 4 sensor set (up to 40), with solid-state lidar an important aspect. But as so often in today’s engineering capability, the fact that it can be done does not mean that it should be. There are significant peripheral aspects that must be addressed for upper level AVs, with motion sickness high on ZF’s agenda. “Reaching 99.8% autonomous driving capability may be possible with a huge effort. This is what Silicon Valley companies aim to do. But to achieve 100%? Rather difficult.”

Tech duo: ZF and Airbus

To support its wide-ranging AV R&D programs, ZF is collaborating with Airbus Defence and Space. On-board automotive systems will be complemented by the aerospace company’s satellite-derived information systems, to provide reliable end-to-end solutions for autonomous driving and self-positioning vehicles. These permit extreme accuracy via Airbus’s Ground Control Points (GCP) satellite imagery expertise to support the fusion of sensor data, including lidar and radar. GCPs validate the accuracy of mobile mapping data.

But this is not a silver bullet, warned Gollewski, underling the market’s need: “Let us assume you are an airline pilot; you will have the aviation equivalent of an SAE Level 2-plus system: The flying task is with the aircraft, the responsibility is with the pilot.” Compared to Levels 3 and 4, Level 2-plus provides a significant advantage and reduces cost/complexity of more advanced solutions, with the possibility of being marketed with “huge opportunities” in the passenger car sphere. “Level 2-plus is a good answer,” Gollewski noted.

Attitudes towards personal transport for the present generations and those that will follow represent a significant element of Gollewski’s lateral thinking about practical autonomous levels. “How can we create new ways of urban mobility? In my view it cannot be the answer to further increase vehicle numbers in overcrowded cities – that would not be the younger generation’s idea of city living in the future. Their environmental attitude and awareness has increased; there should be social change.”

However, looking back to his OEM background (senior positions at Audi, including Head of Electronic Systems Car Safety; CEO Automotive Safety Technologies GmbH), he said the key traditional factors of successful car sales included “emotional design, driving behavior, and driving feel”, together with many safety facets. But with higher level AVs, these would become less or entirely irrelevant. Yet Gollewski cites many of today’s teenagers as budding car enthusiasts, unlikely to show emotional interest in higher level AVs.

“So Level 2-plus with a pure electric or hybrid powertrain would meet their aesthetic/emotional expectations for driving pleasure outside cities,” he said, unlike characterless, workhorse Level 5 solutions for inner city mobility, and possibly for on-demand taxis. “Level 5 passenger cars for the city would operate in dedicated areas, be interconnected and flow control would be in place,” Gollewski said. “With some cities around the world now banning, or considering banning, today’s cars from their centers, it could be a game changer.”

Confusion and the public

What’s the timing of all this? There is considerable confusion among the buying public, while individual OEMs, academics and governments have their own take on the implementation calendar. Targets are discussed – perhaps 2040 or 2050 for volume arrival of AVs – and modified as politicians bring short-term dimensions to the picture.

Gollewski’s pragmatism prevents him from providing specific figures, with the subject of autonomous driving as complex as the traffic situations it will manage. “For a long time we will have a driver as the back-up solution,” Gollewski said. “Ten to 15 years at least to cover the complex scenarios.” But for dedicated lanes, the time would be much earlier and some automated transit systems such as 2getthere (since March this year part of the ZF Group), are already in place in several parts of the world.

It is 107 years since the first “autonomous” aircraft flew courtesy of Lawrence Sperry’s inventiveness, its basic autopilot sufficient for the aircraft to be flown with hands/feet off the controls. But Sperry was a “pilot in command”. In today’s world of advanced technology, two humans still remain on the flight deck of passenger aircraft. To adapt Gollewski’s aviation summation: The driving task is with the vehicle, the responsibility with the driver. At least until that 0.2% gap is closed and SAE Level 5 becomes a reality.