Estimation of Traffic Density on Urban Freeways

Abstract Traffic density is one of the most important parameters to identify traffic state, but it is very difficult to be acquired in real world. This paper has proposed a method to estimate traffic density using the output (traffic flow and time mean speed) of fixed traffic detectors. The method firstly converts time mean speed to space mean speed by introducing the coefficient of variation and standard error of space speed, and then, acquires traffic density through the relationship between traffic flow, density, and space mean speed in traffic flow theory. Furthermore, the paper has indicated that traffic density is insensitive to the coefficient of variation of space speed. In the end, the method has been applied to Shanghai inner elevated ring freeway and the traffic density can be reasonably estimated. In addition, fitting curves among traffic flow, density, and space mean speed are acquired. The results are similar to classic traffic flow theory. The method can estimate traffic density directly on the basis of outputs of fixed detectors, especially for urban freeways with lots of traffic detectors, which are laid with mutually short distance. It is proved that the method is very simple and applicable to urban freeways.