Perimeter Control for Two-region Urban Traffic System Based on Model Free Adaptive Predictive Control with Constraints

Abstract

Abstract Recent studies have shown that macroscopic urban traffic control, especially perimeter control, plays an important role in the field of urban traffic control. In this paper, a novel data driven control method named model free adaptive predictive control with constraints (C-MFAPC) is utilized for perimeter control for two-region urban traffic system. In this strategy, the advantages of model free adaptive control (MFAC) and model predictive control (MPC) are combined. That is, the controller can be designed by merely using the I/O data of the system, while the system output sequence can be predicted without the system model. Moreover, the constraints of perimeter control input and the urban traffic system’s output are both considered. In this framework, macroscopic fundamental diagram (MFD) is utilized to determine the desired vehicle accumulations in each region and generate the output data of the urban traffic system. The effectiveness of the proposed method is tested via simulation, and the result shows that it works better than MPC strategy, which is commonly used for perimeter control.