Strong bike traffic is one of the most important pillars to achieve a green change in mobility. But what does it take to get more people on their bikes? How can cities become more bicycle-friendly? Researchers at the Nuremberg University of Applied Sciences used virtual reality (VR) to investigate how urban development can make cycling subjectively safer and more attractive. We talked about this with one of them, Daniela Ullmann.
Daniela, for this research you built a virtual twin of a road in the Bavarian city of Fürth. Around 100 test persons used a bicycle simulator to “ride” along it in virtual reality, under various conditions. What was the aim?
Our focus was on the subjective safety perception of cyclists. Traffic planning often concentrates on objective safety factors, i.e., technology comes first. We wanted to focus on people, their individual needs, and perceptions. After all, it is the cyclists themselves who must accept the infrastructure. To manage the transition toward a more sustainable mobility system, we’ll need such a citizen-centric view.
How was the digital twin built?
First, we collected lots of traffic data as we wanted to have a realistic model of the road section. The city of Fürth provided us with the corresponding data and site plans and supported us with the traffic survey.
We built two models: In a 3D model, we modelled everything static, i.e. buildings and all objects in the road space such as trees or benches. In addition, we created a traffic simulation in PTV Vissim.
By overlaying the two models, we created the VR environment. PTV Vissim generates the vehicles and road users and inserts them into the 3D model. Our participants rode the bicycle simulator in this digital environment. Thus, they were able to dive into the virtual traffic situation with almost all their senses.
What happened then? Which scenarios did you look at in the simulation?
To analyze influencing factors in cycling, we looked at different road layout as well as traffic and urban planning parameters.
There were three forms of road layouts: First, the cyclist was guided on the roadway without special cycling infrastructure, i.e. in mixed traffic with motorized traffic. The second scenario was a marked cycle lane on the roadway. The third form of guidance was a physically separated cycle path.
For the traffic parameters, we looked at, for example, what a change in car traffic volume would do. We reduced the traffic volume in PTV Vissim. This is a great advantage of the simulation as this would not be possible in a real survey on the road. Other parameters were speed limits of 30 instead of 50 km/h as well as parking at the edge of the road.
And the urban development changes?
We introduced various elements. For example, a nature strip of trees and bushes at the edge of the roadway. Also, amenity areas, i.e. lively zones with restaurants, recreational areas, and many pedestrians. The third scenario was a change in road marking: a red colored cycling lane. So, there were a total of 20 different scenarios in combination. Each test person cycled through three different scenes.
What were your most important results?
Regarding the road layout, our expectations were confirmed, and our results are in line with the current state of research: the participants felt safest on the cycle path that is physically separated from motorized traffic. The cycle lane was ranked second and the lane in mixed traffic was subjectively rated as the worst.
This illustrates how subjective and objective safety can differ. At least at intersections, cyclists are seen better by car drivers if the are on the roadway, which is why riding in mixed traffic is – at least objectively – safer than riding on a separated cycle path.
The scenery with the nature stripe was the most effective. The vegetation significantly increased the attractiveness of the route and the subjective safety. What was also very interesting was that we asked the participants how long they had been cycling on the section of road. Wherever there were green strips – regardless of the type of road layout – the travel time was estimated to be shorter than without green. This indicates that an attractive design of the road space motivates people to cycle.
Did leisure zones also increase the attractiveness?
Yes, but the perception of safety rather decreased because there were more pedestrians in the streets.
It was also interesting that speed of 30 km/h only resulted in a subjective increase in safety in mixed traffic. Cyclists on the cycle lane and separated cycle path did not perceive the speed changes as much. In contrast, participants without a special cycling infrastructure felt significantly safer at 30 km/h. This means: If cyclists must be guided on the road due to spatial conditions, a speed limit can be a good way to make the route subjectively safer.
Another lean and simple measure to promote cycling is the coloring of the cycle lane. This is comparatively inexpensive and does not require any construction measures. In our user study, red coloring significantly increased the subjective perception of safety.
Where do you go from here with your study?
Our results were recently published in the Journal of Urban Mobility. We also handed the report over to the city of Fürth, of course. It was important for us to generate knowledge and provide an impetus for further research on the topic.
We are also continuing our work. One goal, for example, is to make the bicycle simulator even more realistic. It is important that the urban planning aspects are further illuminated. There is a lot of talk about safety in the sense of avoiding accidents – which is of course very important. But it is just as important to consider what is missing to bring more people on their bicycles. What makes people use this means of transport? The design of the street and the urban environment can make a significant contribution here. Further research is very important here.
2 thoughts on “Making cycling more attractive with VR”
Hi Stefanie I’m trying to create a VR simulation for cyclist using Unity and Vissim. I was wondering how did you implement your solution, could you share more about the software used and any documentation if possible? Thank you very much!
Hi Derwim, the studiy was done by Daniela Ullmann from the Nuremberg University of Applied Sciences: https://www.th-nuernberg.de/en/