Why autonomous driving will only reach its full potential in public transportation.
Admittedly, the title sounds more like a bad science fiction movie in which oversized, futuristic machines battle for world domination in the distant future. Of course, we don’t see the future as being that bleak and dramatic. When it comes to automated driving, however, we can no longer speak of a distant vision of the future. In the U.S. and China, driverless road vehicles are already in commercial operation; in Germany, too, the first Level 4 projects are being prepared and tested in public transportation. There are many indications that automated vehicles will become part of our future mobility. The important thing is how we use them!
The public debate is currently focused on robotaxis: driverless vehicles that can be ordered via an app and take passengers from door to door. That sounds convenient, modern, and marketable. However, this approach is not enough to bring about a sustainable mobility transition. An analysis of Waymo data from California published in May 2026 shows that during the first approximately 1,000 days of commercial robotaxi operation, the vehicles traveled just under 139 million kilometers. Passengers were transported for 54 percent of that distance; the rest were empty trips. [1]
When automated vehicles operate as privately managed robotaxi fleets in cities, they may perpetuate well-known problems associated with ride-hailing: empty trips, additional traffic, competition with existing public transportation, and low ridership consolidation.
An analysis of Waymo data shows an average occupancy rate of approximately 1.4 people per passenger trip over the entire study period. An earlier analysis of Uber and Lyft data from 2021 [2] also makes it clear that the willingness to share rides appears to be low and that shared ride-hailing services do not automatically lead to less traffic under real-world conditions. Wait times, detours, comfort expectations, and privacy concerns may lead users to continue preferring exclusive rides, resulting in a significant increase in vehicle-kilometers compared to the use of conventional modes of transportation.
Not every automated vehicle contributes to the mobility transition
Automated driving can strengthen public transportation. However, it can also generate additional motorized traffic. A non-integrated robotaxi service primarily responds to individual demand. It operates where there is a high willingness to pay and dense demand: in urban areas, on attractive routes, and during peak hours. This results in a lack of reasonable mobility options for rural areas, off-peak hours, or underserved regions.
An automated public transit system, on the other hand, serves a public purpose. It can be used as a feeder service to rail, to provide detailed access to neighborhoods, to connect rural areas, as a supplement during off-peak hours, or as a replacement for underutilized routes. It is integrated into fare structures, ticketing, passenger information, local transit planning, quality assurance, and monitoring. Only through this integration can automated driving in public transit be effective in terms of transportation.
The backbone remains the efficient public transit system
Even in an automated mobility system, traditional public transit remains indispensable. Commuter rail, subways, streetcars, and major bus routes can efficiently transport large numbers of passengers. Small automated vehicles cannot replace this capacity without creating new traffic problems. But how can automated shuttles, for example, strengthen public transit in areas where traditional route services reach their limits?
The answer lies in a hierarchical mobility system. An efficient scheduled bus and rail service forms the backbone. Automated public transit shuttles provide flexible supplementation. They fill gaps in the network, improve accessibility across the region, and enable mobility even in areas where frequent service is not economically feasible.
A recent study by DB Regio AG [3] confirms this logic. In the competitive scenario examined in the study, privately operated robotaxis operate separately from public transit. As a result, more than 80% of robotaxi trips are direct journeys, while only about 16% are trips to or from public transit. Precisely because the fare structure and the system are not integrated, combined use becomes less attractive.
Rural areas need a mobility guarantee
The difference is particularly evident in rural areas. There, the challenge lies in ensuring accessibility, participation, and essential public services. Many municipalities, neighborhoods, business districts, and educational and healthcare facilities can only be served to a limited extent by traditional scheduled bus services. Demand is spread out both geographically and over time; trips take place at various times of the day, and frequent service is often not financially viable [4].
Automated public transit shuttles can provide real added value here. They can consolidate smaller pockets of demand, serve flexible service areas, and act as feeders to trains, regional buses, or mobility stations. This does not replace public transit, but rather expands it.
Furthermore, the DB Regio study shows that a non-integrated robotaxi model does not solve this structural problem. In the competitive scenario, metropolitan areas and cities benefit the most; in rural areas, the degree to which the mobility guarantee is met remains significantly lower. Only in the public service scenario—that is, with an integrated public mobility system—is this imbalance resolved.
Integration Determines Whether There Are Benefits or Side Effects
This presents a key challenge for public authorities, transit agencies, and municipalities: Automated transportation systems must be integrated into existing planning and management tools at an early stage.
These include, in particular:
- Define operating areas and control functions,
- Integrate fare systems, sales, and passenger information,
- Prioritize quality over wait times, travel time, accessibility, and reliability,
- Ensure that data, the control center, technical supervision, and troubleshooting are covered by contract.
Current industry positions taken by the VDV [5] point in the same direction: Social value is created only with Level 4 vehicles in public transit—that is, in defined service areas, routes, or counties that can be publicly planned and operationally managed. At the same time, the VDV warns that, according to scientific scenarios, robotaxis used exclusively for private purposes could have negative traffic and environmental impacts; therefore, the integration of autonomous driving technology into public-service-oriented, shared public transit services is crucial.
Our Perspective as Technical Consultants
Automated driving in public transportation is not merely a matter of vehicles or technology, but rather a strategic task aimed at establishing and further developing a new intelligent transportation system (ITS).
Thanks to our international experience in the implementation and further development of intelligent transportation systems, we are thoroughly familiar with the necessary interplay between vehicles, infrastructure, data, front-end and back-end systems, technical and organizational interfaces, roles, institutions, financing, regulation, and operations within a scalable sociotechnical architecture.
We help public contracting authorities structure strategic issues early on and integrate transportation planning, technical system integration, operator roles, procurement logic, data requirements, and operational processes into a robust, manageable, and user-oriented mobility offering.
The future of transportation won't be better simply because it's driverless. It will be better if it is planned publicly, operated in an integrated manner, and made accessible to everyone.
Author: Erik Schaarschmidt
Sources:
[1] Abdelhalim, Awad (2026): Millions of Trips, “Waymo” Empty Miles: California’s First Thousand Days of Commercial Robotaxi Service, Findings, May 2026
[2] Schaller, Bruce (2021): Can Sharing a Ride Reduce Traffic? – Evidence from Uber and Lyft and Implications for Cities, published in *Transport Policy*, March 2021, pp. 1–10.
[3] DB Regio AG / ioki / KIT / DLR / Prognos (2026): Autonomous Driving—The Key to Tomorrow’s Mobility. Full version of the study.
[4] Borowski, M., Seisenberger, S., Kinigadner, J. (2025); Autonomous Driving to Improve the Quality of Public Transportation Services in Rural Areas: Where Do We Stand in Germany?, Journal of Mobility and Transportation, Issue 25 (2025), ISSN 2628-4154
[5] VDV (2025): The Public Transportation of the Future Will Be Autonomous – A Strategy for Promoting and Implementing Autonomous Driving in Public Transportation on Public Roads, Position Paper by the Association of German Transportation Companies (Verband Deutscher Verkehrsunternehmen e. V.), Cologne, March 2025
