Controlling Signalized Intersections using Machine Learning

Andreas Berre Eriksen, Harry Lahrmann, Kim Guldstrand Larsen, Jakob Haahr Taankvist

Research output: Contribution to journalConference article in JournalResearchpeer-review

2 Citations (Scopus)
78 Downloads (Pure)


Signalized intersections are the capacity-determining points on roads in cities, and the signal settings are usually based on very primitive algorithms which cause road users to experience a lot of unnecessary delay The work presented in this paper, show the effect of deploying a controller based on the optimization software Uppaal Stratego in four signalized intersections on the same road segment, the controller is fully distributed meaning there is no direct coordination between the intersections. The controller is tested against the controller deployed in the intersections today. The controllers have been tested using the micro simulation program VISSIM. The simulation shows that in comparison with the existing controller, this controller provides a reduction of between 30% and 50% in average delays, queues and number of stops. The fuel consumption and total travel time of cars in the coordinated section are reduced by about 20% in the simulation study. All these reductions are achieved without making the situation worse for the side roads.

Original languageEnglish
JournalTransportation Research Procedia
Pages (from-to)987-997
Number of pages11
Publication statusPublished - 2020
EventWorld Conference on Transport Research - WCTR 2019 - Mumbai, India
Duration: 26 May 201931 May 2019


ConferenceWorld Conference on Transport Research - WCTR 2019


  • Signalized intersections
  • Optimization
  • Machine learning
  • Reinforcement learning
  • Model checking
  • Uppaal Stratego


Dive into the research topics of 'Controlling Signalized Intersections using Machine Learning'. Together they form a unique fingerprint.

Cite this