雷达
计算机科学
校准
弹道
坐标系
极高频率
地理坐标转换
实时计算
遥感
计算机视觉
地理
电信
数学
统计
物理
天文
作者
Changlong Zhang,Jimin Wei,Jingang Dai,Shibo Qu,Xianning She,Zetao Wang
出处
期刊:IEEE Intelligent Transportation Systems Magazine
[Institute of Electrical and Electronics Engineers]
日期:2022-12-08
卷期号:15 (3): 117-131
被引量:4
标识
DOI:10.1109/mits.2022.3224151
摘要
Roadside millimeter-wave radar is one of the most important sensors in roadside perception systems, which have been widely used in intelligent traffic systems. It is important to find an effective calibration procedure for roadside radar to obtain the World Geodetic System-1984 (WGS-84) coordinates of detected targets. However, it is difficult to calibrate roadside millimeter-wave radar safely on public roads without road closures. To solve this problem, this article proposes a roadside millimeter-wave radar calibration method based on connected vehicle technology. First, all the trajectory points of vehicles detected by radar, including the connected vehicle, are collected, and the WGS-84 coordinates of the connected vehicle are obtained by vehicle-to-everything (V2X) communication with the roadside unit. Then, the trajectory points of vehicles are clustered with the density-based spatial clustering of applications with noise (DBSCAN) method, and the trajectory of the connected vehicle in the radar coordinate system is identified with the velocity-matching method automatically. The data pairs of the connected vehicle in the radar coordinate system and WGS-84 coordinate system are obtained with time synchronization. Finally, the calibration parameters are solved with five different types of optimization methods. We verify these methods on the beltway in Changsha. The results show that the improved pseudo inverse method (IPIM) and PIM achieve better performance than the extrinsic calibration method (ECM), rotate coordinate method (RCM), and improved RCM, and they can meet lane-level positioning requirements for V2X applications. This study provides an economical and effective way to solve the calibration problem for roadside millimeter-wave radar in an engineering project.
科研通智能强力驱动
Strongly Powered by AbleSci AI