Optimization of Traffic Lights Using Quantum Behaving Particles
By William Kerr, Benjamin Rubin
Particle swarm optimization (PSO) is an optimization algorithm that mimics the flocking behavior of birds, ants, and other animals in order to discover the minimum of a cost function. We implemented and improved quantum particle swarm optimization (QPSO), a variation of PSO representing each particle as a quantum wave-function. We tested various potential functions to determine the optimal function for our purposes. This gave both a more efficient and more reliable algorithm. For our project we applied this algorithm to minimize wait times at local traffic lights, inputting data into the traffic simulator SUMO and using the lengths of light cycles as the variables.