摘要
随着ADAS系统在汽车领域的普及,基于角毫米波雷达的ADAS系统由于其成本低、环境适应能力强被广泛应用。其中,使用角毫米波雷达的盲区监测系统能够有效辅助驾驶员对车辆周围环境的感知。根据24GHz角毫米波雷达的特性,使用2个角毫米波雷达对驾驶员盲区进行辅助监控,建立基于角毫米波雷达的盲区监测系统。而毫米波雷达输出目标存在一定的误检,文章使用角毫米波雷达连续5帧数据,建立反馈目标值运动模型,对目标位置数据进行更新,使用K-means算法对检测目标数据进行聚类,使用聚类结果判断检测目标是否真实存在,以消除毫米波雷达的误检,从而实现角毫米波雷达的目标筛选。
With the popularity of ADAS system in the field of automobile,ADAS system based on the angular radar is widely used because of its low cost and strong adaptability to the environment.Among them,the blind area monitoring system using the angular radar can effectively assist the driver to perceive the surrounding environment of the vehicle.According to the characteristics of the 24 GHz angular radar,two angular radar are used to assist the driver's blind area monitoring,and a blind area monitoring system based on the angular radar is established.However,The output target of angular radar has some false detection.In this paper,use the five consecutive frames of angular radar data to establish the motion model of the feedback target value,update the target position data,use k-means algorithm to cluster the detected target data,and use the clustering results to judge whether the detected target is real,so as to eliminate the false detection of angular radar,so as to realize the target selection of angular radar.
作者
郭蓬
何佳
刘修知
张正奇
唐风敏
Guo Peng;He Jia;Liu Xiuzhi;Zhang Zhengqi;Tang Fengmin(China Automotive Technology&Research Center Co.Ltd Automotive Engineering Research Institute,Tianjin 300300)
出处
《汽车实用技术》
2020年第11期26-28,共3页
Automobile Applied Technology
基金
中汽中心重点科研项目(18200111)。
