Jump to content

Swarm intelligence on the road – the uma app from Wolfsburg AG

UMA App wird beim Autofahren verwendet. (Copyright: Ingo Bartels)

Avoid traffic jams, and arrive on time without any stress – these are the most important objectives for anyone travelling on Germany’s roads, especially commuters. In order to achieve these goals, Wolfsburg AG has launched the family of UMA (Urban Mobility Assistance) apps. But drivers aren’t the only ones who benefit from it...

A conversation with Dr Gerrit Schrödel, Head of Mobility Research at Wolfsburg AG:

Mr Schrödel, how would you explain to my grandma what the UMA app actually does?

First and foremost, it navigates! (Laughs) The app will take you from A to B via the quickest route. You could object and say that other navigation systems claim to do the same thing. The difference with the UMA app is that everyone who uses the app becomes part of a swarm. Like fish or birds. These swarms move in motion patterns that they effortlessly coordinate with each other. Our app simulates that, and our algorithms know the start and end points of all journeys currently made via the app – all of which are anonymised, of course. Based on this information, the app navigates drivers so that all traffic routes are used in an optimal manner. And every driver reaches his or her destination as quickly as possible accordingly. For example, taking the Autobahn isn’t always the best choice. I discovered that again this morning on my daily commute from Braunschweig to Wolfsburg. Today, for example, the UMA app routed me through the villages because there was a traffic jam on the Autobahn. I now use the app every day and am always pleased when I arrive at the exact minute that the app has predicted.

But GoogleMaps also shows me which streets are busy, and then directs me to the shortest route, right?

Yes and no. Google will give you the recommended route that is the fastest at the time of the navigation request. By contrast, the UMA app monitors the traffic situation over the entire trip in thirty-second intervals and actively reroutes you in the event of unplanned delays. Although we also source traffic data from the outside, we are constantly supplementing the data, which is provided to us by our own swarm in a sense.

Swarm intelligence requires the input of many individuals. How big does the swarm have to be to yield a real benefit?

It works really well if at least five per cent of active road users use the app. There are an estimated 80,000 commuters in and around Wolfsburg every day. We’re heading in the right direction if we have at least 4,000 active app users. The app has now been downloaded more than 6,000 times, but what is even more important is whether the users then turn on the app every day.

Who else benefits from the UMA navigation apart from the users?

Municipalities do for sure. On the one hand, they have the opportunity to ensure that their roads are utilised in a more uniform manner thanks to an intelligent control system and achieve a smooth flow of traffic on all roads. Let’s say there is a marathon or a demonstration. If the municipality is then able to enter the key data relating to such events in a networked system, you can prevent a great deal of traffic chaos with little effort. Or you can keep unnecessary traffic away from schools between seven and eight o’clock in the morning. The UMA technology uses a gentle approach to traffic control to achieve a better traffic situation on the whole.

Does this mean you also plan to export your technology to other cities?

Yes, precisely. And we’ve recently been able to celebrate a success: UMA was launched in May 2017 in Dresden, in cooperation with VW Sachsen and the city of Dresden.

Does that mean that navigating with UMA is no longer a just a research project, but a real application already?

It’s definitely both. The number of downloads is constantly increasing, and we have very good ratings in the AppStore (4.5 stars on average). But at the same time, the entire field of mobility is an extremely exciting research topic for us, which deals with far-reaching questions. The UMA navigation system doesn’t just take you from A to B; we also experiment with precise routing in terms of parking, so that the search for a parking space becomes less stressful. And we’ve also developed UMA Mitfahren, a prototype for carpools which has been used during a friendly user test with employees of Volkswagen AG so far and will be available to everyone in the near future. Our topics are also important in the field of ​​autonomous driving: which system should be used to subsequently control autonomous vehicles? It could well be a swarm working as equals. Or maybe, if I am in a hurry, I will be able to buy a fast lane for the left lane on the Autobahn in future. These are all questions that are not only technically but also politically and socially relevant.

Who actually had the idea for UMA?

UMA was created in 2014 as part of a strategy workshop for the mobility research we conduct. Everyone was talking about the topic of intelligent mobility then, and we got into some brainstorming. We were inspired by the Internet, where millions of data packages travel and reach their destination in a variety of ways. We quickly came upon the idea that it should be possible to apply this to traffic somehow! No sooner said than done. Now, two years later, there are already several UMA apps which are designed to improve mobility.

Dr Schrödel, thank you for talking to us!