30 May 2022

Drones already deliver defibrillators and medicine in sparsely populated areas, but how should unmanned traffic operate in large cities? With the assistance of AI and visualisation, the UTM City project will show us how.

Jonas Lundberg, Linköping University, and Jesper Tordenlid, Combitech. Photo credit: Peter Karlsson
Jonas Lundberg, Linköping University, and Jesper Tordenlid, Combitech. Photo credit: Peter Karlsson

Drones that replace heavy trucks in city centres, transport goods right to the shop door, pick up and drop off deliveries and even fly us out to the airport, above and past the rush-hour traffic jams. This may sound like pie-in-the-sky, but it can be reality in the not-too-distant future. Several projects are under way – in the Swedish region of Östergötland, Dubai and Brazil.

For this to become reality – for planning, governance and follow-up of the traffic – AI is required. A central problem for the researchers is that the automation and the artificial intelligence must be both verifiable and controllable, both for society and for those monitoring the autonomous systems.

“We currently use visualisation and simulation so that future operators and societal actors can take part in the discussions. They can create traffic scenarios, and discuss traffic situations, problems and opportunities”, says Jonas Lundberg, professor at LiU’s Division of Media and Information Technology, and director of the UTM City project.

Started with UTM 50

The two-year project UTM 50 started in 2017, in collaboration with the Swedish Air Navigation Service Provider (LFV). UTM stands for Unmanned aircraft system Traffic Possible drone traffic in Linköping. Possible drone traffic in Linköping. Management. The aim was to find out what drones could be used for and how they could interact with other airborne traffic, especially in the proximity of airports. The twin cities of Norrköping and Linköping were used as models. A simulator and a visualisation environment were created for drone traffic between the towns – UTM City.

The research generated huge interest both at home and globally, and has continued in a number of new projects within the framework of UTM City. As before, the research is conducted in close collaboration with LVF and Saab, and is funded by Vinnova, Swedish Transport Administration, and by stakeholders of the respective projects.

“UTM City will provide answers to two different questions, one of more general interest, which will clarify for decision-makers and others how drone traffic could be and work in the future. For this work, our simulator is invaluable. The second question is a bit narrower, but just as important, as it concerns how we can control drone traffic above heavily built-up areas and next to large airports”, says Jonas Lundberg.

The work with the AI architecture is headed by associate professor Magnus Bång at LiU’s Department of Computer and Information Science.

“UTM City creates huge volumes of traffic data which must be processed in a short time. Decisions on the traffic are essentially made in real time by our system”, he confirms.

City traffic

Traffic above heavily built-up areas places strict demands on safety and control.

Hotspots for possible drone traffic in Dubai.Hotspots for possible drone traffic in Dubai. “If we can solve the existing challenges of using drones in cities, we simultaneously solve how it can work optimally in the countryside. Today there are strict regulations for flying over heavily built-up areas – and it’s these areas that will have the greatest need in the future. We need to get the balance right”, says Billy Josefsson from LFV research, innovation and digitalisation section, who is heading the project for LFV.

The UTM City project is the core of several sub-projects and collaborations. The biggest is probably UTM Explore for CORUS, where a large number of Europe’s aviation authorities collaborate, funding from the EU’s Horizon 2020 programme.

In this two-year project, a large-scale demonstrator will be developed. It will visualise how drone traffic between two cities, in this case Linköping and Norrköping, can function and be controlled.

“A demo is planned for early October of this year”, says Jonas Lundberg.

Billy Josefsson is happy with the collaboration:

Billy Josefsson.Billy Josefsson. “Together we conduct frontline research in the field, thanks to our ability to combine knowledge and expertise from the industrial sector and state authorities with the excellent research at Linköping University. There is no better way to conduct this type of complicated research”, he concludes.

Now it is important for the actors to get society into the loop; the public and decision-makers must understand how drone traffic can look and work, and which societal challenges it can solve.

“We have a fantastic interactive visualisation environment here in Norrköping; it can help societal actors and authorities understand what drones can achieve”, says Billy Josefsson.

Other sub-projects of UTM City

UTM Dubai expo.UTM Dubai expo. UTM City Dubai: An intermediate result from UTM Explore was presented at a workshop at EXPO2020 in Dubai, generating a lot of interest. It is about and exploring the possibilities of using autonomous drones in large cities. Special challenges are posed by airports, hotels, logistics centres etc. Safety must be 100%. A simulator will be developed, which will suggest the best routes for the drones’ package deliveries. The project is funded by Vinnova, and is managed by, in addition to Saab and the Swedish Civil Aviation Administration, the University of Sharjah and Dubai Airports Engineering Projects, DAEP.

Last Mile Delivery Sevices (LMDS): This project will study how drone traffic can take over the final leg of more and more transports, i.e. from a transport hub or airport to the customer. The project is part of Autonomous Airport Test Bed, funded by industry and Vinnova.

UTM Brazil: The aim is to find ways to come by different types of large and small unmanned vehicles controlled from the ground and autonomous vehicles. The project is in the initial phase.

Read more

https://ivis.itn.liu.se/research/projects/utm_city.html

Read more about UTM50

Latest news from LiU

Server room and data on black background.

Machine Psychology – a bridge to general AI

AI that is as intelligent as humans may become possible thanks to psychological learning models, combined with certain types of AI. This is the conclusion of Robert Johansson, who in his dissertation has developed the concept of Machine Psychology.

Research for a sustainable future awarded almost SEK 20 million grant

An unexpected collaboration between materials science and behavioural science. The development of better and more useful services to tackle climate change. Two projects at LiU are to receive support from the Marianne and Marcus Wallenberg Foundation.

Innovative idea for more effective cancer treatments rewarded

Lisa Menacher has been awarded the 2024 Christer Gilén Scholarship in statistics and machine learning for her master’s thesis. She utilised machine learning in an effort to make the selection of cancer treatments more effective.