Coordinated Intelligent Traffic Lights using Uppaal Stratego

Automatic decision making in traffic signal controllers, semi-automated assistance to drivers, accident detection and response, anti-collision measures in autonomous driving etc., are relatively new applications in Intelligent Transport Systems (ITS). Recent developments in radar and sensor technology coupled with algorithmic and software advances brings ITS closer to realization. In this paper, we extend the work of Ericksen et. al., “Uppaal Stratego for Intelligent Traffic Lights”, in Proc. of the 12th Int. Conf. on ITS European Congress, 2017, France where they use a tool called UPPAAL STRATEGO to synthesize traffic light timing strategies through statistical model checking and machine learning. While Ericksen et.al. consider a single traffic light controller at an isolated intersection, we consider coordination between the controllers at two traffic intersections by providing a “green wave” in the heavily congested direction which reduces the overall waiting time of cars and queue length. Our experimental results show a significant improvement over uncoordinated isolated traffic light controllers in terms of the waiting time of cars and providing a new functionality of the controller such as giving a green wave.

[1]  Chun-Gui Li,et al.  Traffic signal control based on genetic neural network algorithm , 2010, 2010 International Conference on Intelligent Computing and Integrated Systems.

[2]  J. Hellendoorn,et al.  Multi-agent coordination of traffic control instruments , 2008, 2008 First International Conference on Infrastructure Systems and Services: Building Networks for a Brighter Future (INFRA).

[3]  Peter Corcoran,et al.  Traffic Light Control Using Deep Policy-Gradient and Value-Function Based Reinforcement Learning , 2017, ArXiv.

[4]  Marco Muñiz,et al.  Uppaal Stratego for Intelligent Traffic Lights , 2017 .

[5]  Kim G. Larsen,et al.  A Tutorial on Uppaal , 2004, SFM.

[6]  Chao Peng,et al.  Cooperative Traffic Light Control Based on Semi-real-time Processing , 2015 .

[7]  Daniel Krajzewicz,et al.  Recent Development and Applications of SUMO - Simulation of Urban MObility , 2012 .

[8]  Kim G. Larsen,et al.  Uppaal Stratego , 2015, TACAS.

[9]  Kim G. Larsen,et al.  Efficient On-the-Fly Algorithms for the Analysis of Timed Games , 2005, CONCUR.

[10]  Kim G. Larsen,et al.  Safe and Optimal Adaptive Cruise Control , 2015, Correct System Design.

[11]  Kim G. Larsen,et al.  Online and Compositional Learning of Controllers with Application to Floor Heating , 2016, TACAS.

[12]  Lei Liu,et al.  Intelligent traffic light control using distributed multi-agent Q learning , 2017, 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC).

[13]  Wolfgang Niebel,et al.  Advantage of cooperative traffic light control algorithms , 2017 .