Trucks are an essential element in freight movements, transporting 73% of freight tonnage among all modes. However, they are also associated with severe adverse impacts on roadway congestion, safety, and air pollution. Truck speed by truck body types has been considered as an indicator of traffic conditions and roadway emissions. Even though vehicle speed estimation has been researched for decades, there exists a gap in estimating truck speeds particularly at the individual vehicle level. The wide diversity of vehicle lengths associated with trucks makes it especially challenging to estimate truck speeds from conventional inductive loop detector data. This paper presents a new speed estimation model which uses detailed vehicle signature data from single inductive loop sensors equipped with advanced detectors to provide accurate truck speed estimates. This model uses new inductive signature features that show a strong correlation with truck speed. A modified feature weighting K-means algorithm was used to cluster vehicle length related features into 16 specific groups. Individual vehicle speed regression models were then developed within each cluster. Finally, a multi-layer perceptron neural network model was used to assign single loop signatures to the pre-determined speed related clusters. The new model delivered promising estimation results on both a truck-focused dataset and a general traffic dataset.

The Federal Highway Administration (FHWA) vehicle classification scheme is designed to serve various transportation needs such as pavement design, emission estimation, and transportation planning. Many transportation agencies rely on Weigh-In-Motion and Automatic Vehicle Classification sites to collect these essential vehicle classification counts. However, the spatial coverage of these detection sites across the highway network is limited by high installation and maintenance costs. One cost-effective approach has been the use of single inductive loop sensors as an alternative to obtaining FHWA vehicle classification data. However, most data sets used to develop such models are skewed since many classes associated with larger truck configurations are less commonly observed in the roadway network. This makes it more difficult to accurately classify under-represented classes, even though many of these minority classes may have disproportionately adverse effects on pavement infrastructure and the environment. Therefore, previous models have been unable to adequately classify under-represented classes, and the overall performance of the models is often masked by excellent classification accuracy of majority classes, such as passenger vehicles and five-axle tractor-trailers. To resolve the challenge of imbalanced data sets in the FHWA vehicle classification, this paper constructed a bootstrap aggregating deep neural network model on a truck-focused data set using single inductive loop signatures. The proposed method significantly improved the model performance on several truck classes, especially minority classes such as Classes 7 and 11 which were overlooked in previous research. The model was tested on a distinct data set obtained from four spatially independent sites and achieved an accuracy of 0.87 and an average F1 score of 0.72.

No abstract