Intelligent Traffic SensorsCopyright: © ISAC
Problem / Motivation
The goal formulated by the EU "Vision Zero" aims at reducing the number of persons killed in road traffic to zero. The current methods of road safety work are no longer target-oriented in this context. A good example of this is that a road safety analysis for an intersection area, for example, is based on an examination of the accidents that have occurred. People first have to die or be seriously injured in order to be able to locate and eliminate a danger zone. There is a great need for a data-based, preventative analysis of safety on our roadways.
Intensive research on preventive methods of safety analysis has been carried out at the Institute of Roads over many projects, and a concrete software solution has been developed in the process. Through the precise and fully automated recording of the trajectories (driving patterns) of all road users, a whole range of use cases can be covered that could previously only be realized inadequately or not at all. These include:
- Preventive safety analyses of critical areas of the road infrastructure (so far only possible via analysis of accidents that have already occurred).
- Validation of measures on the infrastructure (speed reduction, reduction of critical interactions)
- Active influencing of road users based on the recorded traffic situation
Via the demonstrator funded by Innovation Sprint, the advantages of the system will be demonstrated, especially in conflict analysis. Due to the financing and operation independent of projects, there is the possibility of independent operation and further development. For this purpose, four sensors including evaluation units were installed at the corner of Mies-van-der-Rohe Str. / Seffenter Weg (a known accident hotspot). The collected data is used for further software development and for test operations.
Copyright: © ISAC
Security and Data Privacy
Thermographic cameras are used. These have the advantage of recording at all times of day and many weather conditions.
Also, no personal data (faces, license plates, ...) can be captured.
Exemplary evaluations are to be carried out and published soon.