Abstract: Traffic congestion is a global problem and makes serious effect on the persistence development of city and living of human beings. In this paper, the flow, density and travel time of highway are combined into several new eigenvectors, and a new traffic condition identification method based on fuzzy logic is put forward, establishing the eigenvectors as input variables. Finally, the paper carries out demonstration on this method by using the traffic parameters collected on certain urban highway, which shows the process and output results of the algorithm. The result indicates that the effect of this method is better, and it can be a method for studying the urban highway traffic congestion recognition in our country.