Traffic Density Estimation and Congestion Identification Based on Switched Decentralized State Observer for Large-Scale Urban Freeway Network

In order to estimate traffic flow densities and detect the traffic congestion for the large-scale freeway network, by using the limited known data, a switched decentralized state observer design strategy is proposed. First of all, the road network can be modeled by means of the distributed method, the whole network is divided into several subsystem, and thus a distributed piecewise affine system (DPWAS) model is deduced. Secondly, based on the established model, a switching type of decentralized observer is considered, the road densities of each subsystem is allowed estimated by the local observer, and by comparing the error between the estimated densities and the known congested ones, the congestion case can be further detected. There is no interference between adjacent observers. Finally, the experiment results of Jingtong freeway shown that the proposed method is feasible.