Autonomous and connected driving in public transport ahead of the next technological step
In the context of the "Shuttles&Co" pilot project, the research team investigated, among other things, the acceptance of automated and networked vehicles in public transportation and published the results in the Journal for Mobility and Transport . For this purpose, the scientists surveyed citizens in the Berlin district of Alt-Tegel about their public transport usage behavior. There, Berliner Verkehrsbetriebe (BVG) had deployed a highly automated minibus line with an attendant (SAE Level 3) on public roads for the first time between 2021 and 2022. The results of the survey show that there was no fundamental acceptance problem with the tested technology in this context. Nevertheless, the service was little used in everyday life.
How transferable are your results from Alt-Tegel to autonomous driving in public transport overall?
ARNDT: Unfortunately, there has been relatively little research into the acceptance of such bus projects in real operation, both in terms of autonomous driving and the associated transport services. There are only a few projects that query acceptance from (non-)users so explicitly, but the results are actually similar: one does not see critical factors of rejection. However, these are pilot projects on a smaller scale, otherwise it would probably look different. These are intentionally fringe projects, because the autonomous technology of the vehicles does not yet have the level of maturity to be able to drive stably. Euphoria prevails, there are many people interested in technology and perhaps not yet the mass of skeptics.
In the qualitative dialog forum that you did at the end of the study, there was also fundamentally negative feedback that doubted the point of the shuttles, for example.
ARNDT: You don't have to overstate that. The dialog forum took place within the framework of a qualitative approach and was intended to stimulate discussion in order to identify and sort out a broad range of advantages and disadvantages and patterns of argumentation. It was not intended to push content to the general public. It is also important to distinguish between the technology and the offer here. There was no fundamental rejection of autonomous technology, but some voiced criticism of the specific design of the pilot project.
LINKE-WITTICH: We actually wanted to hold such a dialog forum before the start of operations to draw attention to the project. Unfortunately, due to the pandemic, that didn't work out.
Have any new suggestions been made there?
SCHÄPER: Absolutely. Among other things, the discussion focused on a possible expansion of the service, for example, the operating hours or the spatial connections, for example, to consumer or leisure facilities. The functional expansion to on-demand transport was also discussed; after all, Shuttles&Co was a scheduled service. We want to implement the on-demand service in the follow-up projects. An advantage of the qualitative studies was also that mobility-impaired persons could participate. Some critical groups, on the other hand, could not be reached, for example with regard to the scarce parking space on site.
LINKE-WITTICH: I was a little surprised by the demand to be given an overall concept for the pilot project on our own doorstep. One might have assumed that the benefits of a last-mile shuttle would be obvious. But people wanted to know why the neighborhood needed the shuttles, how many parking spaces would be saved, or to what extent busy streets could be relieved by using shuttles.
The local public has overwhelmingly viewed the shuttles as a positive contribution to the quality of life in the neighborhood.
LINKE-WITTICH: Yes. There's a lot that plays into it, you've heard that from people as well. Land use, a bit more space, traffic calming in general. Even among the non-users, 47 percent expressed a positive opinion, i.e. almost half. Among users, the figure was as high as 75 percent. Overall, this is something that is positively evaluated. For a detailed understanding, the qualitative results are helpful.
Could acceptance increase if this form of bus service were part of a larger, new transportation concept for an entire neighborhood?
ARNDT: Yes, but that's purely speculative. If it's really a useful service that increases access, the answer will certainly be "yes," but then the service can also be driver-operated. But the advantage of autonomously driving buses is that you save on personnel costs. This enables a denser public transport service at a manageable cost, especially on the first and last mile or in the often forgotten urban-rural relationships. If technology makes it possible to hop right into a call-sharing cab in the surrounding area because it is just as inexpensive, then public transportation becomes very attractive there as well. However, we do not yet know whether there will be groups that will not use it as a matter of principle if it is offered on a broad basis.
Is the first and last mile the only application for autonomous driving in public transport? Many projects are designed for this at the moment.
ARNDT: Yes. But BVG does not want to limit itself to that: In the current projects, we are also looking for larger vehicles. Autonomous driving is not intended to be an isolated solution for the first and last mile, but it is a good entry point: We have pent-up demand there, it can be presented well, and at the moment there are more smaller automated vehicle types. There are very few really large buses on offer at the moment, and no Level 4 vehicles at all. It also makes sense to design demand-oriented services with autonomous technology: With conventional buses, the driver is constantly at work, while with on-demand services, he stands around from time to time, just like with cabs. Autonomous technology makes this more efficient, because it costs almost nothing to stand around.
LINKE-WITTICH: I was surprised that our survey found that a majority could not imagine the technology on large buses. I'm curious to see whether this assessment will be confirmed and endure in the future.
ARNDT: That's a psychological aspect. It is crucial to convey the reliability and safety of the technology, that even without a driver, help will be provided quickly and effectively in the event of a malfunction. Driverless people movers at airports are used en masse, so people already trust the technology.
City traffic is the ultimate discipline, because anything can happen there.
LINKE-WITTICH: That was also a positive finding, that it worked quite well on the narrow roads and with the special road conditions.
One of their findings was curiosity as a short-term motivation for use.
LINKE-WITTICH: Exactly. The vast majority, almost 80 percent, used the shuttle out of curiosity or for leisure - but really only once - and not for daily shopping or for the way to work
Do you think on a high-demand line or ten years from now, the results would change?
ARNDT: You're speculating on two independent variables: Whether it's a large operation that serves a real demand, and the reliability of the vehicles. That will be higher in ten years.
LINKE-WITTICH: Definitely. The hope is, a positive development in terms of regularity of use in larger follow-on projects.
And at the current state of the art? Could habituation effects set in after a while?
LINKE-WITTICH: There can certainly be familiarization effects, but not with the current state of technology. We have tested the beginnings of autonomous driving at Level 3, i.e. with an accompanying person on board. This will not be possible in the long term due to the lack of personnel. We will soon be testing Level 4, in which the vehicle will drive autonomously without an attendant. This will then be another new technology with teething problems.
They also asked how people felt about the widespread introduction of such autonomous shuttles in public transport.
Comparison of agreement among respondents by telephone (first and second pillar) and directly after use (third pillar) © Department of Mobility and Space, Center for Technology and Society, Technical University of Berlin
ARNDT: The response is basically positive, and you can imagine that for minibuses. If the vehicles become larger in the future, we will have to ask the question again. In addition, there will no longer be any drivers in the follow-up projects. The vehicles will then be teleoperated, so to speak, and can be supported virtually from a distance in certain situations. For the people on the bus, however, it looks like unaccompanied driving, with corresponding effects on their sense of safety.
LINKE-WITTICH: I think our figures are very meaningful in terms of the group of users. In the case of non-users, this is not a representative result for Berlin or Germany as a whole, but relates only to the area of operation with the two postal codes.
ARNDT: But this graph also shows that the popularity is greater when people already have experience with autonomous technology. Then you can see that the shuttle is not a monster and not unmanageable, but a way to fill gaps in the public transport system.
Which result surprised you the most and where are there still unanswered questions?
SCHÄPER: It emerged from the dialog that interest in active, participatory citizen involvement is not necessarily high. There is a high level of trust in the provider BVG as a public institution, so the respondents felt that not quite so much should be done there. In addition, it was difficult to identify negative opinions in the qualitative survey: For the dialogue forum, we did street canvassing and specifically approached people in the neighborhood. Many had noticed the shuttle but did not have a deep understanding of it. Some of these people then gave brief negative assessments, found it annoying or criticized it for taking away parking spaces. Unfortunately, they did not take the opportunity to voice their opinions in dialogue. We plan to collect such opinions in more detail in follow-up projects.
LINKE-WITTICH: Maybe that's where you see the tension: On the one hand, people want an overall concept and explanations, but on the other hand they don't really want to participate. That's why we need good concepts for low-threshold participation and with all the relevant information so that everyone feels involved.
LINKE-WITTICH: I'm very curious to see whether passengers will have the same positive feeling about public transportation without vehicle assistance.
ARNDT: In the current KIS'M project, we are working with teleoperated shuttles, which has the same psychological effect as driverless vehicles. I'm curious about that, too.
LINKE-WITTICH: The next project, NoWeL4 , has already been started, where Level 4 will really be available. But we can't say when it will start operating yet.
Thank you very much for the interesting interview!