Model predictive control using Adaptive Task Allocation for optimization of networks with traffic lights

This paper presents the implementation of a controller based on Adaptive Task Allocation Algorithm (ATAA) for a traffic network with signalized intersections, where it is required to decrease waiting times of vehicles at every intersection. The main contributions of this work are the implementation of ATAA controller under a Model Predictive Control (MPC) scheme and the use of an estimator which does not require an explicit model of the system to predict its future behavior. Two case studies are presented and implementation results are compared for scenarios using both ATAA with open loop controller and the MPC proposed. For both case studies improvements up to 31.3% were obtained for vehicles' mean delay time with the MPC approach in contrast to a network with no traffic light control.