Task

The task was to comprehensively and accurately monitor traffic at a two-level roundabout in a European city. The primary goal was the differentiation and tracking of various vehicle types and participants—such as cars, trucks, motorcycles, and pedestrians. Data collected should identify potentially dangerous situations early and support intelligent traffic management approaches.

Challenge

Complex roundabouts pose significant safety risks due to high traffic volumes, diverse participants, and unpredictable driving behaviors. Conventional traffic monitoring methods often fall short of precisely and continuously capturing, classifying, and tracking all participants. A reliable automated solution was crucial to minimize accidents and enhance traffic flow efficiency.

Solution

SALTIR implemented an advanced computer vision system leveraging specialized AI models to distinctly identify, classify, and track vehicles and pedestrians continuously. This solution enables real-time identification of critical situations and provides robust data for optimizing traffic management strategies. The system has been successfully deployed in a busy, real-world roundabout environment, delivering consistent and actionable insights.

Results

The deployment of AI-based traffic monitoring significantly improved safety by identifying potentially hazardous situations proactively. The generated data also facilitated targeted optimizations of traffic flow, reducing congestion and enhancing overall traffic efficiency. The practical operation demonstrates substantial benefits and confirms the solution’s effectiveness and real-world viability.