PASCAL

Pro-Active Smart Controller for Traffic Lights
(BMFTR, 2022-2025)

Urban areas are particularly affected by changes in mobility. New mobility offers as well as changing private and professional mobility behavior lead to new challenges in coping with the constantly increasing traffic volume. According to a recent study by the European Court of Auditors, road traffic is one of the main causes of air pollution and greenhouse gas emissions in urban areas, with societal costs of around 270 billion euros per year across Europe. One fundamental approach to tackling the increased volume of traffic is to digitize the transport infrastructure. The data collected from road users can then be used to analyze traffic flow. This makes it possible to liquefy traffic at intersections, which in turn means a reduction in CO2 emissions.

The goal of the joint project PASCAL (Proaktiver Smart Controller für Ampelanlagen) was to develop AI methods for the pro-active control of traffic lights for urban traffic optimization and to test them in the urban V2X test field Magdeburg. The test field was set up in preliminary work by Thorsis Technologies in cooperation with the city of Magdeburg and is used to collect and analyze traffic data in real time. To achieve the set goal, the project team relied on the research and application of new AI-based methods (supervised learning, reinforcement learning, graph-based learning) for the optimization of traffic light switching cycles for traffic flow. The existing test field provided the data basis for the development of the proactive smart controller for traffic lights. In addition to the traffic data of the test field, traffic information (e.g. road works, events, congestion or weather) and simulation data (traffic, communication and emission) was used. On the one hand, the simulation data was used to compress the traffic data for training the AI models and, on the other hand, to evaluate determined traffic light switching cycles. The developed system was integrated into the existing urban V2X test field in Magdeburg and extensively tested. The prototype developed in the project forms the basis for widespread use for urban traffic flow optimization and follow-up projects.

official project website

final project report

Project Partners