Identifying and correcting the errors of vehicle trajectories from roadside millimetre‐wave radars

Abstract

Millimetre‐wave (MMW) radars have been increasingly deployed along the roadside highways to collect vehicle trajectory data, which are valuable for traffic safety analyses and traffic control decisions. Nevertheless, possible errors of the trajectories resulting from roadside MMW radars have not been well documented. This study scrutinizes roadside MMW radar data and the concurrent video ground truth to identify five typical errors of vehicle trajectories, including different vehicles tagged with the same ID, different trajectories produced by the same vehicle, ghost vehicles, erroneous vehicle classification, and vehicle location drift. Data cleaning methods were developed to identify and correct these errors. The performance of the developed methods was evaluated empirically in a case study based on the data of five radars, which were installed at different locations along the highway and had different headings. The thresholds of the developed method were calibrated using one radar data and then applied to the data of the five radars. The results showed that the data qualities of all radars are greatly improved by the developed methods, demonstrating that the developed methods are generally applicable for cleaning vehicle trajectory data from roadside MMW radars.

Analysis of Perception Accuracy of Roadside Millimeter-Wave Radar for Traffic Risk Assessment and Early Warning Systems

“…The task of traffic risk assessment and early warning systems is first to ide high-risk traffic environment or behavior and then send the warning message to vant parties. For microscopic level risk assessment, multiple indicators have be However, due to the measuring defects of MMW radars, challenges such as low angular resolution and poor elevation measurement have limited their measurement accuracy, leading to data quality issues such as vehicle positioning errors [8,9]. Positioning errors regarding vehicles on the roads may lead to the misperception of the vehicle motion and interaction behavior, potentially causing inaccurate decision making or fatal crashes in transportation systems [10].…”

“…The RSU performs vehicle matching and sends back the message, if it exists. However, due to the measuring defects of MMW radars, challenges such as gular resolution and poor elevation measurement have limited their measureme racy, leading to data quality issues such as vehicle positioning errors [8,9]. Pos errors regarding vehicles on the roads may lead to the misperception of the vehicle and interaction behavior, potentially causing inaccurate decision making or fatal in transportation systems [10].…”

Two step vehicle trajectory reconstruction strategy based on an improved outlier detection and data smoothing algorithm

“…Perhaps the best-known vehicle trajectory dataset is the Next Generation Simulation Program (NGSIM) dataset [3] , which is also widely used in vehicle trajectory reconstruction, vehicle lane change, and performance evaluation of traffic flow operation, etc. [4] [5][6] [7][10] [11] [12][13] [14] .…”

Automated Lane-Level Road Geometry Estimation Using Microscopic Trajectory Data

“…LiDAR can also provide point cloud data containing road information for geometric information extraction [13]. MMW radars, which ofer better tracking capability and speed accuracy compared to video cameras [14], have gained increasing popularity for real-time trafc data collection [3]; for these sensors, as geometric information cannot be directly recorded, studies have attempted to estimate geometric information from extracted trajectories. For instance, roadside MMW radar can extract road information through trajectory clustering and ftting [15].…”