Distributed optimization for traffic-emission control in urban road networks via cooperative game approach.

In this study, we introduce a cooperative game-based distributed optimization strategy for green-time traffic signals to balance traffic-emission reduction and computational efficiency. Utilizing a macroscopic link traffic-flow model for the road network and a microscopic vehicle-emission model for traffic emissions, we apply a cooperative game framework to enable communication and coordination among traffic subnetworks. The optimization problem is decomposed into subproblems using the augmented Lagrangian alternating-direction inexact Newton (ALADIN) algorithm, thus enabling effective collaboration among subnetworks. Our findings reveal that this cooperative approach, which leverages the ALADIN algorithm, closely approximates the optimal solution achieved via centralized control, thereby significantly enhancing the computational efficiency while preserving the performance. Through simulation experiments, during both peak and off-peak hours, our approach reduces average computation time by over 48.58% compared to centralized methods. Additionally, our distributed control strategy outperforms fixed-time control, reducing traffic emissions by at least 3.3% in both scenarios.