Research on Traffic Parameter Measurement Methods for Intelligent Transportation Systems

The effectiveness and accuracy of traffic parameter measurement is a key means to improve the intelligence level of intelligent transportation system. In this paper, the spatio-temporal data of traffic flow on GY expressway in a city is selected as research data to analyze the spatio-temporal correlation of traffic flow data. And in this way, the GCN-BiLSTM model is constructed, using the advantages of the GCN algorithm and BiLSTM algorithm to capture the potential information in the time series and improve the prediction accuracy, which is used to predict the traffic flow parameters of the highway in each lane section. The spatio-temporal correlation coefficient values of the characterization parameters flow, speed, and occupancy are mostly greater than 0.7, which has a strong correlation. The results of the constructed GCN-BiLSTM model on MSE, MAE and MAXRE are 1.027, 1.606 and 0.511 respectively, which are smaller than the other comparative methods, and there is a GCN-BiLSTM model that can more accurately show the situation of traffic parameters, and better serve for the management and control of the intelligent transportation system.