Expressway Speed Prediction Based on Electronic Toll Collection Data

Abstract

Expressway section speed can visually reflect the section operation condition, and accurate short time section speed prediction has a wide range of applications in path planning and traffic guidance. However, existing expressway speed prediction data have defects, such as sparse density and incomplete object challenges. Thus, this paper proposes a framework for a combined expressway traffic speed prediction model based on wavelet transform and spatial-temporal graph convolutional network (WSTGCN) of the Electronic Toll Collection (ETC) gantry transaction data. First, the framework pre-processes the ETC gantry transaction data to construct the section speeds. Then wavelet decomposition and single-branch reconstruction are performed on the section speed sequences, and the spatial features are captured by graph convolutional network (GCN) for each reconstructed single-branch sequence, and the temporal features are extracted by connecting the gated recurrent unit (GRU). The experiments use the ETC gantry transaction data of the expressway from Quanzhou to Xiamen. The results indicate that the WSTGCN model makes notable improvements compared to the model of the baseline for different prediction ranges.

Positioning method of expressway ETC gantry by multi‐source traffic data

“…Currently, ETC transaction data mining mainly includes statistics and prediction of travel time [3,4], traffic flow [5], driving speed [6,7] etc. Zou et al had done a lot of researches based on ETC data, which contains ETC abnormal data detection [8], travel time prediction [3], expressway maximum speed limit recognition [9], expressway speed prediction [7]. Wang et al [10] constructed a travel time dataset based on the time and place records of vehicles entering and leaving the expressway in ETC transaction data, and restored the historical trajectories of vehicles by data supplement algorithm, the characteristics of vehicle travel time in traffic accidents state and normal traffic state were used as training sets, and the K-nearest neighbour non-parametric regression was used for training and learning.…”

“…Since the research on ETC transaction data mining is in its initial stage, there are not many researches. Currently, ETC transaction data mining mainly includes statistics and prediction of travel time [3,4], traffic flow [5], driving speed [6,7] etc. Zou et al had done a lot of researches based on ETC data, which contains ETC abnormal data detection [8], travel time prediction [3], expressway maximum speed limit recognition [9], expressway speed prediction [7].…”

CSFPre: Expressway key sections based on CEEMDAN-STSGCN-FCM during the holidays for traffic flow prediction

“…ETC gantry data and toll data can record the driving conditions of each vehicle on the expressway. Therefore, using ETC gantry data and toll data, we can accurately estimate the location of ETC gantries [6], prediction of traffic flow [7], travel time [8], travel speed [9], and service area vehicle recognition and dwell time estimation [10]. Since the expressway traffic management department cannot fully grasp the changing trend of road status during holidays, travelers cannot timely understand the changes in road conditions, resulting in too concentrated vehicle travel and difficulty in effective traffic control and diversion, causing traffic congestion [11].…”

Multi-View Travel Time Prediction Based on Electronic Toll Collection Data

“…ETC data record most of the vehicles driving on the expressway, which basically reflects the traffic status of the expressway section [ 3 ]. Therefore, through ETC data, we can accurately obtain the road utilization rate, traffic rate, traffic speed, travel time, etc., which can help us effectively predict the traffic flow [ 4 ], travel time [ 5 ], and speed [ 6 ] of all vehicle types in each section of the expressway.…”