Urban delivery traffic has grown 71% over the last 20 years and now accounts for 20-30% of vehicle kilometres in cities and over 16% of urban air pollution. European cities are responding by encouraging cargo bike logistics, yet current traffic monitoring systems cannot distinguish between cargo bikes and leisure cyclists. This leaves local governments without the data needed to plan, evaluate and regulate sustainable urban logistics.
Telraam is a citizen-powered traffic monitoring platform compliant with the General Data Protection Regulation (GDPR), with over 3,000 sensors deployed across more than 20 countries. As part of this project, Telraam will develop and validate a proprietary edge-AI classifier that detects and counts cargo bikes and delivery vans using its next-generation S3 sensor. The model is trained on a hybrid dataset that combines real-world imagery and large volumes of AI-generated synthetic video, and runs entirely on-device, enabling remote deployment to compatible sensors across the existing network without new physical infrastructure.
Mechelen City Council (Belgium) is the committed end client. From April to October 2026, Telraam will deploy the new model on 6-10 sensors across Mechelen, including locations near schools and logistics hubs, to validate its accuracy under a range of weather and lighting conditions and to provide the city with continuous, granular data on cycle-logistics flows. The data will inform the council’s mobility planning and low-emission zone (LEZ) regulations.
By the end of the project (TRL 9), the cargo bike detection service will be commercially launched as a premium tier within Telraam’s data-as-a-service (DaaS) offering, targeting public authorities, logistics operators, real estate developers and the cargo bike ecosystem. A parallel research and development (R&D) track will explore the detection of bicycles carrying children, demonstrating the AI model’s flexibility and de-risking future features.
Project start:
Project end:
Budget:
Countries
Urban delivery traffic has grown 71% in 20 years, accounting for 30% of vehicle kilometres and 16% of air pollution, yet cargo bikes remain largely unmeasured.
Telraam’s new edge-AI classifier automatically detects and counts cargo bikes and delivery vans via existing citizen-deployed sensors, generating continuous granular freight data for cities.
By October 2026, a TRL-9 cargo bike detection service will launch on Telraam’s platform, validated in Mechelen and remotely deployable on Telraam’s new S3 sensor.
