idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
idw - Informationsdienst
Wissenschaft
AI-based pedestrian traffic lights on the UNESCO best-of list.
Lemgo, 12 February 2026 – The project “KI4PED – AI-based optimisation of pedestrian crossing times through smart traffic lights” from the Fraunhofer institute in Lemgo has been selected by the International Research Centre on Artificial Intelligence (IRCAI) under the auspices of UNESCO for inclusion in the “IRCAI Top 100: Global AI for SDGs Index 2025” and was rated “Excellent” by the reviewers. This means that KI4PED is officially among the 100 AI innovations worldwide that make a particularly strong contribution to sustainability and the 17 Sustainable Development Goals of the United Nations.
Within the KI4PED project, an intelligent traffic light system for pedestrians was developed that detects pedestrians at an early stage, estimates their intention to cross and thus controls green phases according to actual demand. The aim is to significantly reduce waiting times while at the same time improving road safety.
Background: More safety and comfort at pedestrian crossings
Conventional pedestrian traffic lights often do not take into account the different walking speeds and mobility needs of various groups of people – such as older persons, children or people with mobility impairments. Moreover, it has been shown that excessively long waiting times at red lights increase the likelihood that people will cross the road illegally. This is where KI4PED comes in: by using intelligent, predictive control algorithms, green phases are adapted in such a way that pedestrians receive green more quickly and have sufficient time to cross the road safely. At the same time, the number of dangerous red light violations is to be reduced.
Technical approach: 3D LiDAR and artificial intelligence
For KI4PED, Fraunhofer IOSB-INA combined state-of-the-art 3D LiDAR sensors with AI-based data analysis:
• 3D LiDAR sensors contactlessly and anonymously record the surroundings at pedestrian crossings.
• AI models analyse movement patterns and predict whether a person intends to cross the road.
• On this basis, demand-driven control of the traffic lights is realised, without the need to push a button.
As part of the project, real traffic situations were recorded at selected traffic light installations in Lemgo and Bielefeld, used to train the AI, and extensive tests were carried out to examine the robustness of the system under different lighting and weather conditions. In addition, an integration concept was developed to enable connection to existing traffic light controllers, thus facilitating the retrofit of exis-ting installations.
Contribution to sustainability and UN goals
KI4PED contributes to several UN Sustainable Development Goals, including:
• SDG 10 “Reduced Inequality” by increasing road safety, particularly for vulnerable groups.
• SDG 11 “Sustainable Cities and Communities” by enabling safe, pedestrian-friendly mobility and more attractive urban spaces.
• SDG 13 “Climate Action” through the use of innovative AI technologies in traffic engineering.
With its inclusion in the IRCAI Top 100 Index, KI4PED is listed on IRCAI’s interna-tional project platform. In addition, exchange with other projects and potential partners worldwide is being promoted.
“With the KI4PED project, we demonstrate how AI can help in everyday life. By using so-called agentic AI – highly capable, context-sensitive AI systems – the technology understands people and adapts to them, rather than the other way round,” says Prof. Dr Jürgen Jasperneite, Director of Fraunhofer IOSB-INA.
Dr.-Ing. Dennis Sprute
Fraunhofer IOSB-INA
Institutsteil für industrielle Automation
Wissenschaftlicher Mitarbeiter Intelligente Sensorsysteme
Campusallee 1, 32657 Lemgo, Germany
Telefon +49 5261 942 90 – 63
E-Mail dennis.sprute@iosb-ina.fraunhofer.de
GDPR-compliant data collection: In the future, pedestrians could influence the switching of a traffi ...
Merkmale dieser Pressemitteilung:
Journalisten, Wirtschaftsvertreter, Wissenschaftler
Gesellschaft, Informationstechnik, Umwelt / Ökologie, Verkehr / Transport
überregional
Forschungs- / Wissenstransfer, Forschungsprojekte
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
