Please note: The algorithm descriptions in English have been automatically translated. Errors may have been introduced in this process. For the original descriptions, go to the Dutch version of the Algorithm Register.
Algorithm (Intelligent) Traffic Control Installation (I-VRI) at traffic lights
- Publication category
- High-Risk AI-system
- Impact assessment
- DPIA
- Status
- In use
General information
Theme
Begin date
Contact information
Responsible use
Goal and impact
Controlling traffic at intersections by means of an iVRI in such a safe efficient way that increases accessibility. Traffic lights detect different mobilities (cars / bikes / trucks / pedestrians) using detection loops and push buttons. Intelligent traffic lights can - in addition - also connect to in-car systems and navigation apps. This allows traffic lights to see traffic "coming" at an earlier stage and regulate it even smarter. iVRIs can also send information back to in-car systems such as the time to red or green, for example. iVRIs can give a form of priority to traffic participants. For example, when an ambulance arrives urgently, but also to a group of cyclists or a column of trucks, for example. There are different forms of priority. Absolute priority (cutting off other directions), for example, is used for emergency and emergency services and conditioned priority (the traffic light tries to "regulate" towards a road user) in other cases. The municipality's priority framework and local circumstances determine the degree of priority for each road user.
The deployment of iVRIs contributes to efficient traffic flow and accessibility at intersections in a safe way (without people waiting for nothing and a logical sequence). It contributes to the municipality's mobility task. The deployment of iVRIs is also necessary to efficiently carry out the task as road manager of municipal roads. The algorithm affects traffic participants (motorists, walkers, cyclists, trucks, public transport, emergency and rescue services, etc) because it determines the colour of the traffic lights and how long each traffic participant has to wait.
Considerations
It is important, because of different interests, to consider when one group of road users has priority over another. This depends on economic reasons (freight priority), the environment (residential area, business park, etc), public transport priority (reducing delays), or necessity (ambulance priority).
Human intervention
The algorithm makes real-time independent decisions. There is no "human in the loop" for these decisions. The decisions made by the algorithm do fall within human-specified criterea (e.g.: What is the maximum amount of time a traffic light can stay green).
Risk management
Traffic safety is controlled: For traffic lights in general, as soon as a traffic unsafe situation may arise (conflicting directions get green), the algorithm will abort and the traffic lights will start flashing. This is because in addition to the 'control algorithm' that controls the traffic lights, there is also a control algorithm. The abort is due to this separate safety-control algorithm. Apart from this, general traffic law also applies when the traffic lights are 'flashing'. Traffic safety is a basic design principle in traffic lights and the algorithms used.
More text on: Flow/accessibility.
Legal basis
The municipality is responsible for the smooth safe handling of road traffic.
Deepening a little more
Links to legal bases
Elaboration on impact assessments
No direct personal data such as name and address details are stored or processed in an iVRI. The iVRI does receive anonymised position and priority messages from road users. These are used for traffic control but are not otherwise stored. They cannot be traced back to a natural person. The log data stored from a traffic control may contain indirect personal data. In combination with camera images, for example, it can be established whether a particular vehicle drove through a red light, for example. These indirect personal data are not provided to third parties. However, they can be requested by the police.
Impact assessment
Operations
Data
At the heart of iVRIs is the integration of various data sources. These include: 1. Sensors: These are placed at intersections and detect the presence and speed of vehicles. 2. Video cameras: They provide visual data and can be used for object recognition and tracking. 3. Connected vehicles: Modern vehicles often transmit data about their location and speed. 4. Mobile apps: Think of navigation apps that share real-time traffic information. These data streams are collected and analysed in a central system, where decisions are made about traffic light cycles.
Links to data sources
Technical design
The method and models have been established nationwide from the Talking Traffic project and are managed by the LVMB. Just as we learn from our experiences, these traffic lights also learn. They use something called "machine learning". Usually this is done by a technique called Reinforcement Learning. This means they look at what has happened in the past and learn from that to do better in the future. This way, they get a little smarter every day!
External provider
Similar algorithm descriptions
- Safe, autonomous and automatic control of traffic flows at an intersection through an iVRI.Last change on 10th of July 2024, at 8:13 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- This algorithm has a low impact. Because the smart traffic lights are able to 'recognise' traffic, it becomes possible to prioritise certain traffic flows over others (in the Westfields area).Last change on 28th of June 2024, at 8:26 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- The V-MRDH is a traffic model in the software Omnitrans (developer is DAT-Mobility, Deventer). The algorithm calculates the amount of travel (car, public transport and bicycle) based on input data such as networks and demographic data. The model is used, internally at The Hague, at other municipalities and at traffic engineering consultancies.Last change on 23rd of August 2024, at 15:23 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- This algorithm has a low impact. Because the smart traffic lights are able to 'recognise' traffic, it becomes possible to prioritise certain traffic flows over others.Last change on 10th of September 2025, at 14:08 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- Field not filled in.
- Status
- Out of use
- Self Service Passport Control, deployed by the Royal Military Police to support their border duties on behalf of the Ministry of Asylum and Migration, uses a face comparison algorithm. This algorithm checks whether the traveller standing in front of the e-gate is the same as the person in the passport photo.Last change on 16th of December 2025, at 9:03 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- DPIA
- Status
- In use