That’s a huge topic, of course, but there’s always what I call some kind of linearity in technological progress, one step leads to the next, and so on. And then there are qualitative leaps that lead to industry-wide, general discussions. We’re experiencing one such leap in public transport at the moment with autonomous vehicles, which has been the subject of debate for nearly three years now. If we follow this development in the media, the initial optimism has become somewhat dampened of late – and this is typical of technical debates. Following the euphoria at the start, come the technical difficulties, social opposition, legal issues, etc. I like to point out the example of flat TV screens: we dreamed about TVs hanging on the wall in the 80s, but it took 30 years to become reality.

The question is, what happens when this automation of public transport I just mentioned becomes reality. It always takes longer than you think, or may wish, but it will happen. The development of autonomous vehicles is taking place on two levels, or two paths: the first being heavily promoted by the automobile industry and the second that of public transportation. Large-scale public transportation is and will remain a necessity in cities, unless you are able to change the urban structure, which is conceivable, but obviously more difficult and would take longer. Nonetheless, mobility and urban structures are inter-related. Take the changes over the past year due to Covid-19, for example, where most people no longer travel to work, which impacts mobility. And this is a very interesting aspect that needs to be discussed, especially in terms of future investment in transportation.

Yes, and it works. But when discussing mobility, it’s not just about urbanity and city spaces: we mustn’t forget rural areas where automation (such as self-driving shuttles) has a number of advantages. Not everyone can or wants to drive a car, and it’s the responsibility of public transport authorities to enable these people to travel, too.

Another example of digitalization here is the transportation of goods. I think it makes sense for rural public transport agencies to deliver goods and the first model projects are already running. Passenger transportation has clearly defined peak times, corresponding to people’s work days, and in the interim periods you could move goods using the same vehicles.

That’s right. It’s a question of profitability, but the continued increase in parcels due to online shopping is already obvious. These are all aspects that need to be brought together, but are currently still separate. On the one hand, you have the logistics business, and, on the other, public transport operators: two things that are not really linked.

A good question. I believe they are the prerequisite for this interlinking. If you like, these two modes – public transport and the transportation of goods – are already digitalized. In general, passengers can see the current state of the transport network in real time on their mobile phones. This problem has been solved. And the same is true of logistics services: customers can track their parcels in real time. The technology is already in place. So the two could be combined, but the industry sectors are separate. It’s not a technical problem, but an organizational one. Logistics is a private sector, while public transportation is usually in the hands of municipal or regional authorities. So it’s a question of how to cross the borders – and one very viable option is a digital interface.

I believe the technology is already available, and data transfer is no longer a problem. Generating more bandwidth is a technical problem, but one that will be solved. Rather, the issue is a legal, or a social one – namely, who owns the data? That’s the current debate, a debate which is far from being closed. Do I have access to my own data? Who may evaluate my data? Can companies use the data for their business models? Then there’s the issue of data security: who guarantees that my data is not manipulated?

And these issues are also linked to mobility. When vehicles are automated, there is always the possibility of remote access. So there has to be protection against hacking. But these problems are well known and will be solved. In larger networks this will take longer, but the time will come. It may take 5, 10, or 15 years, as these are administrative matters, but automated vehicles will be implemented.

Remember that when we started in the 90s, automated passenger counting was relatively new and unknown. Now it’s state of the art. Today we may assume that all larger, modern transport agencies are using this technology. Some, as in German-speaking countries, use it to distribute revenue among associated transportation networks, while in the USA it’s used to provide the Federal Transit Administration with statistics. But if we peer into the future…

… there is a transformation from a purely statistical use to real-time processing. This can be current load information, which is used by the operator to control the fleet. This is another new development during the pandemic, as transport authorities need to keep load numbers below a certain limit, but also for passengers so they can decide to wait for another vehicle if the next one is too full. 

What we’re also seeing is that public transport is becoming more individual, as the data is increasingly being used for prognoses, adding real-time data to statistical values so as to estimate passenger numbers several stations down the line, for example, or for notifying passengers waiting on the platform in advance of which cars are less crowded. For operators, this could mean substantial savings in terms of efficiency, reducing the number of trains needed. Additionally, those passengers with mobility challenges would be able to board a car or bus knowing that there are free seats. While there’re already some interesting pilot models in operation, there is a whole field of opportunity for real-time passenger data.

If we take this idea a bit further, I see the automatic passenger counting system as a human-machine interface. An APC “sees” that there is a human and relays this information to the machine. This may be in the area around the door and signals to the operator not to close it, for example, heightening passenger comfort and safety, which is an increasingly important aspect as automation progresses.

Take autonomous shuttle buses as an example. What happens when someone starts misbehaving and there are only passengers on board, no driver or person of authority? This weakens the subjective feeling of safety. Or, as another example, what if you’re running toward the bus and wave to the driver to wait. How would an autonomous vehicle react? By waving back? These questions, of course, are irrelevant for the developers of self-driving cars, but extremely relevant to public transportation agencies, of course.

I see it as both. As an engineer, it is exciting because it’s difficult ­– it’s a challenge, but one that can be overcome. This is a very interesting task for companies like ours, at the interface of people and machines. We see these developments and the questions raised and I believe we can solve them. And we will have to, if autonomous vehicles are to be accepted. Because they will have to react to people, to collaborate with them.

That’s right, and it’s aligned with the trend in the western world to focus on the individual. It’s a debate we have with our Asian colleagues: China is going in a different direction. At the moment we all sit in a row on the subway, but there are design models that place passengers in single armchairs – even in mass transportation. Comfort is more than just an app, and we at iris play a small role in this as our technology is part of vehicle design.

Well, it’s still all a question of security. Digitalization creates its own problem by allowing data to be manipulated. This applies to the area of mobility as well. A few years back a prototype autonomous car was hacked, and the brakes applied remotely, and someone could use a networked bus as the entry point to access the wider network. As the manufacturer of APC systems we’re aware of this.

But APC systems also open up new areas of application and one challenge could be not only to count passengers, but to see where they are going and whether the vehicle should wait for them, for example. The technology we have available today must be seen as the starting point for future opportunities.

Personally, and once again I see automation here as the answer, I’d like to see more services on the edge of cities, like Berlin. At the moment, this area is unprofitable, so it is under-serviced. If you arrive late in the evening at the end of the urban railway (S-Bahn) line, for example, you have to take a taxi as there are no more buses running, and it will cost you much more than the 2.80 euros for the public transport ride. I believe autonomous buses would be a good solution, not only for Berlin, but for any city that has clearly laid-out residential areas on the outskirts.

Correct. Otherwise, people are not interested and take the car from the start. So the public transport authority has to compete with individual modes of transport, as well as such models as TNCs (like Uber and Lyft) and car sharing. Again, ITPS and APCs have a role to play in shortening the journey time and increasing overall passenger comfort.