Abstract:

Expressway traffic state is divided into unimpeded, slight smooth and congested sub-states by fuzzy C-means clustering method. Combining expressway second-order macroscopic traffic flow model with particle filter algorithm, the traffic state parameters are estimated. By clustering the estimation values of the parameters to one of the three kinds of traffic sub-states, the corresponding traffic state estimation is obtained. Based on the traffic state estimation and the characteristics of each traffic sub-state, an expressway joint control model is established, which combines variable speed-limit control and on-ramp control. The traffic data of Changchun East-expressway is used to validate the model. Results show that the total time spend of the expressway is reduced by 4.55% and the traffic flow density is also reduced. The traffic congestion is eased to certain extent.

Key words: engineering of communications and transportation system, expressway, traffic state estimation, variable speed-limit control, on-ramp control