Gaussian Mixture Model-Based Speed Estimation and Vehicle Classification Using Single-Loop Measurements

9 Traffic speed and length-based vehicle classification data are critical inputs for traffic operations, 10 pavement design and maintenance, and transportation planning. However, they cannot be 11 measured directly by single-loop detectors, the most widely deployed type of traffic sensor in the 12 existing roadway infrastructure. In this study, a Gaussian Mixture Model (GMM)-based 13 approach is developed to estimate more accurate traffic speeds and classified vehicle volumes 14 using single-loop outputs. The estimation procedure consists of multiple iterations of parameter 15 correction and validation. After the GMM is established to empirically model vehicle on-times 16 measured by single-loop detectors, the optimal solution can be initially sought to separate length17 based vehicle volume data. Based on the on-time of the separated short vehicles from the GMM, 18 an iterative process will be conducted to improve traffic speed and classified volume estimation 19 until the estimation results become statistically stable and converge. This method is 20 straightforward and computationally efficient. The effectiveness of the proposed approach was 21 examined using data collected from several loop stations on Interstate-90 in the Seattle area. The 22 traffic volume data for three vehicle classes are categorized based on the proposed method. The 23 test results show the proposed GMM approach outperforms the previous models, including 24 conventional constant g-factor method, sequence method, and moving median method, and 25 produces more reliable, accurate estimates of traffic speeds and classified vehicle volumes under 26 various traffic conditions. 27