How can autonomous mobility be deployed in Belgium so that public and market objectives can be combined?

Some reflections based on the EIT Masterclass 08/2024 organised by ITS.be and Espaces-Mobilités on AV driving in San Francisco

The Masterclass on AVs in San Francisco last August was an eye opener for those who experienced it. Many of the participants confirmed it. For me, the experience with Waymo's robotic taxis was most valuable. Above all, I remember that this is a mature technology and concept, that has got about a city wide coverage, that these taxis are easy to get (app based), that you get a very smooth ride, you especially experience a safe feeling: looking through the eyes of the AV (200 m range, also around the corner). So I came back with the feeling: would love to have them in our country too.

In Belgium we have been facing tough mobility problems for several decades, perhaps autonomous mobility is the golden solution? Consider the traffic congestion that has been increasing year on year since Covid, the federal planning agency gives anything but favourable perspectives for the future on this issue. Autonomous mobility is also often looked at in relation to road safety, more than 80,000 fatalities on Belgian roads over the past 50 years, and in more than 90% of cases road users played a role in those crashes. But there are many more problem areas to mention, think of transport poverty, parking imbalance in urban neighbourhoods....

But isn't there also a danger: that the enthusiasm and success of autonomous vehicles create a new car euphoria, similar to the modernism at the time of EXPO58? In the late 1950s and the mass motorisation that followed, there was not only the super dominance that the car claimed in the mobility pattern, but that pattern permeated the entire organisation of society. Not at least in the organisation of the public domain and more broadly urbanism - understand often demolition of streets and even neighbourhoods such as the Brussels North Quarter. Will city governments and citizens only be bystanders to increasing car traffic and related patterns of urbanisation?

A one-on-one transfer of the developments of autonomous mobility in the US and in Europe, in our country does not seem possible to me, certainly not desirable.

In the EU there is a different culture to deal with technological and other innovations. In San Francisco we saw how the introduction of the robot taxi was not only a business-driven development, the role of the governments is limited to that of facilitator of business use cases. Apart from the recognition of the vehicles on the federal level, only the state of California played a role in the field, that is to say they determine the procedural rules and monitor them in a concrete use case for the development from pilot to full-scale deployment. But with the robot taxis it is Waymo that determines how many taxis drive, in which streets they do or do not. City government and citizen organizations are only bystanders.

Confront this with our approach to innovation: inspired by the DG Research and Innovation of the EU, we often refer to the quadruple helix model in which there is interaction and cooperation between business, research and public authorities and citizens.

Regarding the question of what the impact of robot taxis on mobility can be, the question arises as to what the impact is on the existing modal split, which modal shift will occur after their introduction and deployment.

According to figures from the University of San Francisco, in San Francisco more than half of the customers of robot taxis came from the classic rideshare, Uber and Lyft, a quarter shifted from walking and another quarter from public transport. Cycling does not have a relevant share in the modal split and shift. However, on top of the modal shift, there is a quarter of newly induced traffic.

These modal split figures are very far removed from the mobility patterns in our country where about 70% of individual journeys are made by private car, as a driver or as a passenger. The remaining journeys are made by public transport or active modes. Cycling is particularly important in cities - especially Flemish cities: a share of 30% or more is not exceptional. In larger cities, the car has a share that is less than 50%.

The tension between the mobility patterns in our country and the US is clearly reflected in the residential school traffic. The design challenge in American school environments is often: how do you create sufficient space for drop-off and pick-up places for cars for children who are either brought or picked up by parents or by taxi services. Waymo in SF picks up on this in that they see a niche in this residential school traffic for their services and are marketing around it. The design challenge for school environments is very different in our country. In Flanders, on average no less than half of the pupils come to school on foot or by bike. How to make the school environment - and also the school routes - suitable and safe is central to the mobility policy of many municipalities. Do we think it is desirable that AVs would break this pattern, and that school environments would take the place of independently moving children? Because this is more than just a mobility issue, it is also about happiness and physical activity, for example. And that does not only apply to children, but also to seniors and adults… .

Mobility in urban and rural environments in our country is different, in particular the role of the car in rural environments is very dominant. Is there therefore room for the application of robot taxis in the countryside? In fact, ‘countryside ‘ has become a peripheral phenomenon in Belgium, limited to some parts near the Dutch and French border areas in Flanders and the Ardennes region in Wallonia.

Check the concept of the ‘Daily Urban System'. The area of functional cities - covering the range of the large part of daily commute to urban nodes covers a large part of the country: our country became what urbanists sometimes call a horizontal metropolis. The only mobility backbone apart from car infrastructure is the railway system. Protecting this system from cannibalization by robot taxi systems seems to me an important precondition. The transport capacity of collective rail transport will also greatly exceed that of autonomous transport given the ratio to conventional car traffic and can avoid the need for additional highway capacity.

Robot taxis outside city centres can also have an important spatial effect: as mentioned, the daily urban systems cover a large part of the surface of our country and causes urbanisation pressure. It’s important to know that the experience of time in a robot taxi will be different than in a regular car - it will no longer be the case that one can see travel time as a pure cost. People will therefore be prepared to commute over longer distances, and so the spatial footprint of the urban systems, say urbanisation pressure, will become much larger.

Under condition of strong land use governance regarding additional land occupation, there is -under additional mobility conditions-? clearly room for robot taxi applications in urban sprawl areas. Car dependency is a fact in there. The most desirable scenario described in the literature on this issue is to orient this taxi system as much as possible towards rail and bus and to create intermodal nodes for exchange between AV and PT. Apart from this, attention should go to protect school environments and routes. And of course? - more in general there is a need to restrict car traffic to social and environmental street capacities in residential areas and town centres. AV deployment with robot taxi systems, also as flex public transport systems could give added value to the mobility system in these areas.

In order to comply with sustainable mobility planning, which has become a European standard, the use of AV's in urban centres will have to contribute to the modal shift away from individual car use. In addition to sustainable mobility, urban policy is often focused on the human city, tailored to active mobility - including cyclists: the ‘carless drivers’. A city where space for people is safeguarded or created. This means that in the urban fringe there is a need for intermodal nodes to accommodate car traffic from urban sprawl areas and to provide transfers to shuttles and other forms of city-friendly transport. This also requires systems of Urban Vehicle Access Regulations (UVAR), with which many European cities have built up experience. Spatial planning has models at its disposal to respond to this: Transit Oriented Development (TOD) and the more recent concept of the 15-minute city.

To conclude, I would like to return to traffic safety. Systems that are centrally managed (such as robot taxis and shuttle services) offer the opportunity to make a huge leap in this area. Something else is the individual car that falls under the responsibility of the individual owner. We have noticed that major questions need to be asked about level 3 and the highly developed level 2, call it level 2+. The problem becomes even worse if owners can choose features that refer to unsafe traffic behaviour: systematically exceeding speed limits or opting for an aggressive driving style, regardless of the environment (see e.g. Tesla). I have already discussed the differences in application in urban core areas and suburban areas in detail. What I have not yet mentioned is the need for adapted street and road design. All steps can still be taken in that area. What has been developed in the US can serve as inspiration, but no more than that.

Dirk Lauwers

This article refers to the presentation I gave on 26th September 2024 in Brussels during the ITS.be annual conference as an introduction to Knowledge Sharing Session on Autonomous Mobility

#AutonomousMobility #driverlesscars #carlessdrivers #DailyUrbanSystem #TOD #15MinuteCity #TrafficSafety #ModalShift #SustainableMobility #Level3 #Robottaxi #shutlle #P&R #EITMobility