AI helps improve road safety in cities

AI can greatly contribute to increased road safety in cities and help municipalities make informed decisions that improve safety and comfort for vulnerable road users. This is the conclusion reached by researchers from the School of Engineering (JTH) at Jönköping University in a collaboration with Jönköping Municipality and the technology company Viscando.

The Vinnova-funded project DAIMOND (Data and AI for Decision-Making Support in Traffic Infrastructure Development) ran between November 2023 and November 2024 with the aim of increasing road safety in cities by using artificial intelligence (AI) and data-driven methods.

The project aimed to improve the traffic environment and increase the safety of the most vulnerable road users, such as children and the elderly, thus creating safer conditions for all vulnerable road users. By collecting data on behaviour and interactions of road users using smart 3D sensors, and analysing it using advanced AI methods, the project has supported more informed, proactive and objective decision making in traffic infrastructure development, while at the same time increasing knowledge and expertise in using AI tools in Jönköping municipality.

“The AI Lab (JAIL) at JTH contributes with technical expertise in artificial intelligence and works closely with Jönköping Municipality to understand and identify how AI can be applied to real challenges faced by municipalities,” says He Tan, Assistant Professor in Computer Science at JTH and one of those involved in the project.

Increasing safety for children and the elderly

The project has helped develop a decision support prototype that analyses the behaviour of road users at different types of pedestrian crossings.

The prototype can be used to improve safety at intersections where vehicles and pedestrians meet. The prototype allows Jönköping Municipality to assess and understand safety risks for children on their way to school, and to introduce optimal safety interventions to make the road safer for children walking or cycling to school.

“The prototype is also used to increase safety at pedestrian crossings by analysing the behaviour of drivers and pedestrians. It is very important to take into account the elderly and people with disabilities,” says Carlos Trischler, traffic engineer and project manager at the city planning office.

Through the project, Jönköping Municipality has gained further insight into how artificial intelligence contributes to more qualitative analyses for increased quality in planning, design and follow-up of projects linked to traffic infrastructure, as well as cost efficiency through better traffic flow management.

“In retrospect, we can see that the project has directly and positively impacted the safety and well-being of children in the community, especially through improvements in traffic planning around school zones,” says He Tan.

Future opportunities

The project has helped to create a safer and more efficient traffic environment in Jönköping Municipality, and in the future, other Swedish municipalities may also benefit from the knowledge of AI and data-driven methods.

“JTH is working on research that is very important. They have been able to demonstrate the great impact that AI can have in urban planning and specifically traffic planning,” says Carlos Trischler.

For Jönköping Municipality, the project has helped raise awareness of AI issues. It has also opened up the possibility of using AI in more and broader applications in traffic management and traffic planning.

“Working with road safety is the most important task in my work. More specifically, trying to increase road safety for children and the elderly who are the most vulnerable road users. We put a lot of focus on working on that,” says Carlos Trischler.

Impact on public sector activities

He Tan believes that the DAIMOND project has been very meaningful for JTH and that the collaboration has strengthened the school’s role as a key partner in regional innovation, especially in smart cities.

“It has given us an opportunity to see how AI can go beyond theoretical models and directly impact public sector operations. Municipalities often face complex challenges with limited resources, and the application of AI has the potential to transform decision-making processes, improve efficiency and make data-driven decisions. Being part of this project means having a direct role in that transformation where AI is not just a buzzword but a practical tool that can deliver real solutions for society,” says He Tan.

The results of the project will be presented on 15 January when Jönköping Municipality organizes an afternoon on artificial intelligence for employees and managers.