Skip to main content
conference lunch move company map contacts lindholmen lindholmen 2 travel info


Intelligent and self-learning traffic control with 3D & AI

The project aims to show how new types of sensors and traffic management models, combined with AI, can contribute to improved accessibility and safety in signal controlled intersections. An optimization that includes pedestrians and cyclists and not just motor vehicles.
Photo by Thomas Winkler on Unsplash

The project proposes the development of intelligent sensors and innovative methods for traffic control through computer vision and machine learning methods. The focus is on complex intersections in major cities where motor vehicles, cyclists, pedestrians and others, in the city, existing transport solutions exist simultaneously and affect each other's accessibility and road safety.

In the project, we intend to demonstrate adaptive and self-learning traffic control that continuously diagnoses the intersection's function. The purpose is to:

1. Further develop Viscando's 3D & AI solution for use for traffic control at intersections. We intend to combine traffic control detection with detailed traffic data for traffic flows and continuous diagnosis of intersection function. Among other things, we intend to develop methods for multi-camera object tracking and, based on AI, develop new types of traffic control parameters which, in addition to presence detection, also include accessibility and road safety for pedestrians, cyclists and motor vehicles.

2. Swarco Sverige AB is investigating how new types of parameters can be handled in today's control equipment and identify software development in future control devices.

3. In Uppsala, work is in progress on the smart cities of the future. This project fits well with the wishes for more efficient and safer traffic in the city. Uppsala participates in the requirements and choice of demo locations.

Project period:
2020-03-01 - 2021-06-30

Project partners:
Viscando, Swarco, Uppsala kommun

Amritpal Singh,