Darting-out Target Detection with NLOS Signals for Vehicle MIMO mmWave Radar

Detection and tracking of crossing objects are cru-cial in automated driving to ensure the safety of pedestrians and drivers. However, in urban scenarios, occlusions not only make it difficult to detect the crossing target in non line of sight (NLOS) area, but also make the crossing target in line o...

Full description

Saved in:
Bibliographic Details
Published in2023 IEEE Radar Conference (RadarConf23) pp. 1 - 6
Main Authors Shen, Yuanjie, Zhang, Minglong, Wu, Yulin, Cui, Guolong, Guo, Shisheng
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2023
Subjects
Clustering algorithms
Darting-out target
Detection and tracking
Heuristic algorithms
mmWave radar
NLOS area and LOS area around parked vehicle
Object detection
Point cloud compression
Radar detection
Signal processing algorithms
Target tracking
Online AccessGet full text

Cover

More Information
Summary:Detection and tracking of crossing objects are cru-cial in automated driving to ensure the safety of pedestrians and drivers. However, in urban scenarios, occlusions not only make it difficult to detect the crossing target in non line of sight (NLOS) area, but also make the crossing target in line of sight (LOS) area susceptible to multipath signals. In this paper, a method is proposed to effectively detect the darting-out target occluded by parked vehicle during the entire movement for millimeter wave (mmWave) radar. Two typical situations of electromagnetic (EM) propagation are studied when the target darts out from NLOS area to LOS area. The ground reflection EM signal is used to detection and tracking when the darting-out target is located in NLOS area. The reflecting plane of parked vehicle is estimated to assist in eliminating interference caused by NLOS signals when the darting-out target is located in LOS area. Experimental results ultimately validate the effectiveness of the proposed method.
DOI:10.1109/RadarConf2351548.2023.10149684