Chapter 2
Optimization and adaptation of
traffic lights
Contributions to traffic management such as synchronizing traffic lights on a pri-
mary road have been an improvement. Those synchronizations are called green-
waves: a car that gets green at the first traffic light will get green at the next
traffic light when the car is approaching it (when driving at given speed). This
green-wave is pre-calculated for a given travel speed. When the traffic density is
high, the given speed cannot be reached and the aim of the green-wave cannot be
fulfilled.
Some research is done on optimizing traffic by letting the traffic light con-
troller learn the best actions. Wiering et al proposed to use reinforcement learning
algorithms. Those algorithms learn the expected waiting times of cars for red and
green lights at each intersection, and sets the traffic lights to green for maximal
gain configuration. [WVVK04]
Other researchers are focussing on adaptation of traffic light controllers. Traf-
fic demands varies during the day and the traffic lights should sense the changes
in traffic flow and adapt to it. [Ger05, DH05]