Traffic Safety in Future Cities by Using a Safety Approach Based on AI and Wireless Communications

Conference: Mobilkommunikation - 25. ITG-Fachtagung
11/03/2021 - 11/04/2021 at Osnabrück

Proceedings: ITG-Fb. 299: Mobilkommunikation – Technologien und Anwendungen

Pages: 6Language: englishTyp: PDF

Authors:
Koenig, Immanuel; Bachmann, Marek; Bieshaar, Maarten; Schindler, Stephan; Lambrecht, Franz; David, Klaus; Sick, Bernhard; Hornung, Gerrit; Sommer, Carsten; Francke, Angela (University of Kassel, Germany)

Abstract:
In Germany, about 1 000 VRUs (vulnerable road users) are killed in road accidents, and over 100 000 are injured per year. Worldwide 350 000 are killed per year (2016). In this article, we present our vision of active warnings of impending collisions and dangerous hotspots to increase the safety of VRUs in future cities. Hotspots are detected by wireless devices and exchanged between the devices via 5G so warnings can be given. The following competencies are required for this vision to become true: Computer science/AI (artificial intelligence), legal sciences, and traffic planning. Computer Science / AI is necessary to analyze the data from cameras and wearables from VRUs to give active warnings to road users and VRUs. The data is also analyzed with AI algorithms to detect irregular patterns such as accidents or dangerous situations to identify the location of dangerous hotspots. The traffic planners then use hotspot locations to implement safety precautions. Legal aspects are also to be considered as early as during the design of the AI algorithms.