An interview with Professor Dr.-Ing. Markus Stöckner, Steinbeis Entrepreneur at the Steinbeis Transfer Center Infrastructure Management in Transportation (IMV)
Autonomous driving is about a lot more than just vehicles that operate autonomously. The transportation infrastructure also has to meet new requirements. Professor Dr.-Ing. Markus Stöckner, an expert at the Steinbeis Transfer Center for Infrastructure Management in Transportation, spoke to TRANSFER about tasks that have to be taken on and the challenges that have to be overcome.
Hello Professor Stöckner. What demands does autonomous driving place on transportation infrastructure?
I think first you have to determine what the term autonomous driving actually means. Do we mean semi-autonomous driving, or do we really mean fully automated driving in complex project situations and/or complex driving scenarios? On a fundamental level, the more demands placed on automatic driving, the more demands are placed on the infrastructure. There are three important aspects to this: the road space with its particular geometry, “roadside equipment” – things like signs and light signals – and then the technology for transmitting information using mobile networks. The road space must be clearly defined and recognizable. This can be achieved for example with road markings, which already provide important orientation when finding your way around during conventional driving. You’ve probably been driving along a section of highway yourself when the lines suddenly disappeared. Depending on how fast you’re traveling, it doesn’t take long to lose your bearings. It’s different with autonomous driving because the visibility requirements regarding road markings are even higher. The same applies to recognizing and locating traffic signs. Vehicles also need digital models of road spaces to gain their bearings. For example, they’re fitted with rotating lasers to generate high-precision models for comparison and orientation purposes. But the problem comes with costs, not just in terms of initial outlays but also for maintenance so they continue to function over time. The inevitable consequence of this is that asset management systems will need to be developed.
It would be impossible to imagine driving without traffic lights or road signs. What will it be like for self-driving vehicles? Will “analog” roadside equipment be replaced by other systems for autonomous driving?
It will also be impossible to imagine driving in the future without such equipment. For the moment, the existing equipment is enough for the more straightforward systems. Vehicles with so-called driver assistance systems are already able to read signs and speed limits or help passengers maintain the right distance from vehicles in front of them. But if you think beyond the horizon, you realize that it makes sense to get things like traffic light systems to interact with vehicles. For example, it would make sense for switching sequences or traffic information to be shared so autonomous vehicles can react themselves. But if you look at the investment costs this would entail, in financial terms this is a gargantuan task.
Self-driving cars pose new challenges in terms of road safety. How important will it be to develop road networks strategically?
Public debate regarding road safety will already be a big challenge. You just need to look at the current accident statistics. In recent years, there have been around 3,300 deaths per annum on the roads in Germany. Justifiably, public scrutiny focuses on accidents involving people riding bikes when vehicles turn right, people driving too fast in downtown areas, or the occasional accident that’s already happened with semi-autonomous vehicles. But that doesn’t change the task we face of making the roads safe, which will mean significantly reducing the number of deaths and injuries – although actually they should be avoided altogether. These are fundamental ethical issues we face in society, and with self-driving vehicles we will need to discuss them again. What will happen if we have mixed traffic on the roads – non-autonomous vehicles alongside autonomous vehicles? Are we willing to accept self-driving vehicles having accidents? Sure, one could do something like reserving certain road sections for autonomous vehicles in the strategic planning of road networks. But I think we have to look at highly complex systems not just from the standpoint of technology, but also from a societal perspective before we make the answers too simple for ourselves.
In the future, the transport infrastructure should not just make it possible to use autonomous vehicles, but also make things sustainable and environmentally friendly. What’s the best way to achieve this?
That’s a complex question, and there’s no straightforward answer to it. A colleague of mine at Karlsruhe University of Applied Sciences, Christoph Hupfer, once said in an interview that “it’s no use if we’re standing in traffic being energy-efficient.” It’s a short statement, but there’s many a truth in it. Just look at the traffic in a city like Stuttgart in the morning or evening rush hour. The roads are bursting at the seams and the public transportation network is also approaching breaking point. But you’re no better off sitting in traffic in a self-driving vehicle. So we need to go one step further and see infrastructure as a complex system that can be optimized with different subsystems. We’re already achieving this by dovetailing different municipal mobility options – using bicycles and pedestrian routes, or different sharing projects like car sharing. You have to set up the system to take economic, environmental, and social factors into account. That’s our task, and beyond the usual kind of partisan thinking, there’s still some room to find a happy medium because we’re going to have keep on this journey to work up new mobility concepts. Self-driving vehicles can contribute to the overall system within the transportation infrastructure by optimizing traffic flows, while at the same time we do what we can to reduce the number of accidents. That will solve some, although not all of our problems. But we’ll need to have all these problems fully solved for the transportation infrastructure to function properly in the future. So in those terms, self-driving vehicles do make an important contribution, even if we still have many challenges to solve – but we will solve them.
Prof. Dr.-Ing. Markus Stöckner (author)
Steinbeis Transfer Center Infrastructure Management in Transportation (IMV) (Bruchsal)