We have regular public meetups in our Budapest office to explore the potential of vehicle communication from different perspectives. This time, our guest was Péter Dalos, a mobility expert from BKK, the Hungarian capital’s public mobility management company. András Váradi, Research Director at Commsignia, took part in the discussion. Read our summary of this great conversation.
At BKK, as the mobility manager, planner, and developer of Budapest, we see public transport as the “backbone” of urban mobility. However, besides walking and cycling that are traditional “limbs“ connecting to this “backbone”, a large market for shared services is emerging and we do not consider them to be competitors – said Peter Dalos of the BKK. Our aim is to integrate them into the larger transport system in a way that will make their customers also passengers of public transport services.
In order to reduce traffic hazards and road dangers, data from sensors, passenger counts, route planning applications and safety risk assessment based on hotspot identification are collected through multiple channels – Peter Dalos added. The operational public road manager monitors vehicle traffic via CCTV, and pilot projects are underway as well to monitor pedestrian traffic and optimize traffic flow.
Until now, urban transport development has been car-centric, so we are now focusing on taking data collection on sustainable transport modes to a more professional level.
Commsignia develops vehicle-to-everything (V2X) technology for car manufacturers, and we also have a smart city business, supporting intersection connectivity. V2X provides data transfer between road users, so that what one car sees, knows or decides, other cars receive digitally, and they don’t have to locate, scan, predict or decide anything themselves – said Andras Varadi.
Vehicular communication is based on the idea that cars do not collide because they anonymously share their location, speed and direction. Connected vehicles can also be considered as moving sensors, and cities can use roadside devices to collect the shared information – Andras Varadi added. In the second generation of V2X, cars not only talk about themselves, but also about what they perceive with their onboard sensors, radars and cameras: whether they see a parking space, another car, a pedestrian.
Limited speed is the most important
Urban citizens are increasingly demanding a more humane environment. Less dangerous or perceived to be so, less noisy, without speeding vehicles in their neighborhoods. Shared micromobility service providers are pioneering in this respect, as they can remotely and automatically control the speed of their own vehicles by so called geofenced zones defined by BKK. For example, a virtual fence around a pedestrian zone forces all micromobility vehicles to slow down or not enter the zone with acceleration at all. Modern solutions really help to ensure that rules that used to be enforced with a lot of communication, monitoring and sanctions can be enforced automatically in the near future.
And the question also arises:
Shared e-scooters have this by default. Why don’t we have this kind of on-board speed limit on shared cars or public utility vehicles?
If we are serious about Vision Zero and want to minimize the number of injury conflicts, the key is to ensure that large and heavy vehicles do not travel faster than 30 km/h in the city – Peter Dalos underlined.
The EU is trying to enforce speed limits through automatic cruise control, and I think it will soon be mandatory – replied Andras Varadi. This means that cars will have to be aware of the speed limit. This information can be distributed digitally, so that road signs don’t not have to be recognized by a camera, which can be challenging at night and in poor visibility conditions. If the speed information is posted as a digital message with certificates, there is no room for error, these software are highly reliable.
If only a few cars start obeying the rules, other cars and their drivers will follow, a whole wave of cars can be kept under control.
Andras Varadi mentioned that Commsignia has also experimented with using artificial intelligence to detect situations that could escalate into collisions before they happen. “In such cases, we don’t control traffic, just tweak the existing rules: change the traffic light rules, insert a speed limit.”
BKK’s strategic aim is to return space to its original urban functions in actively used areas that have been flooded by cars. “To do so, we need to pay close attention to the behavior of the people at intersections, and understanding this will lead to much higher quality urban public spaces. If such a development is initiated, fewer people will unnecessarily burden the city with their car use.”
MOL Bubi is a bike-sharing service managed by the BKK, and its backend software sees the bikes that have accumulated due to sporting events for instance – said Peter Dalos. The dispatcher knows if there’s room for that many bikes, and there’s no need to call logistics, because two hours later, people will take away most of them. If the city had 200,000 shared bicycles instead of 2,000, it would make sense to use artificial intelligence to deal with such situations.
However, reducing the number of parking spaces, if the number of cars remains the same, leads to heavy traffic – adds Andras Varadi. V2X can help in the way mentioned earlier, when cars use their sensors to scan the streets for available parking spaces. Connected vehicles know whether it makes sense to drive into a particular street to look for a spot. A more sophisticated solution is automated valet parking: cars automatically park in a pre-designated space and automatically pull out when we want to move on. For now this only works in closed environments, and will be available on public roads after further developments.
Some urban citizens use technology to hack their environment: they set up non-existent virtual speed cameras in Waze because they are fed up with speeding on their street, and they see that it makes drivers go slower – said Péter Dalos. In many cases, however, these same solutions allow users to harm others. While some people avoid traffic jams by using apps, those living in quiet residential areas are not happy about the increased car traffic.
Apps are certainly influencing transport if there is no better alternative, says András Váradi.
V2X communication is a great way to put cities back in the driving seat.
The standardized language of V2X needs to be understood by everyone in traffic, and it’s used on a frequency dedicated to road safety.
V2X messages generated by cities will be displayed by cars: first only signs, suggestions, later full routes. Commsignia has research in which we propose complete routes to cars. Initially this only works with ambulances and first responders, but if a city’s infrastructure allows it, dynamically organized self-driving fleets can be driven from one point to another.
Successful cities have a road safety strategy to reduce the danger on the roads and streets. Traffic safety is determined by three factors: human behavior, infrastructure and vehicles. Until the vast majority of vehicles and objects in the city are self-driving, the built environment will remain determining how vehicles driven by people, cyclists and pedestrians travel – Peter Dalos stressed.
Once algorithms can describe the behavior of people and vehicles in cities, no anomaly can ever be considered an accident
After all, an accident is an event made up of coincidences, but here we are talking about events that we know will happen. As part of BKK’s strategy, we do not consider injuries and deaths as accidents, but we believe that they are preventable through specific interventions – said Peter Dalos. These are not technology-dependent, but technology can help a lot: it can tell us where the first places are that we need to address.
This obviously requires penetration – added András Váradi – to get vehicle communication in as many cars as possible. Once this is in place, the system can reconfigure itself within a second in response to dangerous situations. If, for example, a V2X application detects emergency braking from three different cars, it’s trivial that something has happened, traffic needs to be slowed down, which can then be checked again by a human operator and modified a few seconds or minutes later.
In the second generation of V2X, when vehicles also share sensor data, it will be enough for every third car to communicate with each other to detect every other car, pedestrian and cyclist on the road. We are trying to map everything that could cause a safety-critical situation, even a piece of rock falling on the road, because it’s important information for the vehicles.