Key points are not available for this paper at this time.
Recent studies have shown that a homogenous urban road network exists a well-defined macroscopic fundamental diagram (MFD), which can be used for perimeter control conveniently. Most of the existing perimeter control strategies are model-based control methods, leading to the result that the control effect may be not good enough if the traffic model is not accurate. In this paper, a novel data-driven strategy called model free adaptive control (MFAC) method is proposed for multi-region perimeter control in order to get rid of the dependency of the aforementioned model-based MFD-based perimeter control methods. Due to the fact that the multi-region urban traffic system (MRUTS) is a complex interconnected system, a decentralized estimation and decentralized MFAC (DED-MFAC) method is utilized to deal with the strong-coupled characteristic of the traffic system. In this framework, MFD is used to determine the desired accumulations in each region and generate the throughput data of the urban traffic system, since the acquisition of trip completion flow is more difficult than accumulations. In addition, route choice is also integrated in the proposed policy to further improve the performance of the urban traffic system. A key advantage of the proposed approach is that it can only use the traffic data instead of the traffic model for real-time perimeter control. The effectiveness of the proposed perimeter control scheme is tested in simulation for a multi-region system, and the results show that it is superior to some other commonly used perimeter control methods.
Lei et al. (Tue,) studied this question.