SAFELIGHT

European cities have a big challenge to achieve Vision Zero and safety for vulnerable urban road users. 

Fatalities due to accidents are increasing in cities due to the use of new micromobility vehicles (scooters, bicycles, etc.) and their interaction with private and delivery vehicles. 

Some of the accidents are caused by users’ negligence but an important part of the accidents with vulnerable users involved could be avoided by improving the infrastructure. In particular, separating the circulation of users with different vehicle sizes, signalling correctly the crossing places and keeping the crossing areas well illuminated, at night, at dawn and at dusk. 

The evaluation of illumination conditions in cities at night has been performed both visually and using manual luxmeters. This is not an effective way to analyse visibility in all types of conditions, and it is impossible to apply across the entirety of cities, simply. Doing the analysis place by place, intersection by intersection , and covering different weather conditions would be impossible manually. 

An automatic way to  measure visibility in cities would be to check the most hazardous points, analyse and apply the most efficient improvements to the lighting when necessary. 

This project consists of the development and validation of a computer vision-based solution that will evaluate the lighting of cities automatically: that is, the ‘Autonomous Urban Inspector for Lighting’. Special care will be taken to ensure that these automated evaluations can easily be integrated in the usual urban maintenance operations. 

The host cities will use this ‘Autonomous Inspector for Lighting’ to discover critical locations regarding illumination, and Flash Lighting will recommend the best solutions for each location. A second automated inspection will be performed after the implementation of the lighting modifications, to validate the improvement for VRU visibility in the affected locations, with the goal that the ‘Autonomous Urban Inspector for Lighting’ is a useful tool for the  reduction of traffic accidents.