基于连续波毫米波雷达的车辆开门防撞预警技术  

Vehicle door open collision warning technology based on continuous wave millimeter-wave radar

作　　者：刘贵如[1] 周鸣争[1] 陈效华[2] 王陆林[2] 王海[2] 郭贤生[3] 

机构地区：[1]安徽工程大学计算机与信息学院,安徽芜湖241000 [2]奇瑞汽车股份有限公司前瞻技术科学院,安徽芜湖241006 [3]电子科技大学电子工程学院,四川成都611731

出　　处：《江苏大学学报（自然科学版）》2015年第6期634-640,共7页Journal of Jiangsu University：Natural Science Edition

基　　金：国家自然科学基金资助项目(91120307);安徽省自然科学基金资助面上项目(1408085MF124);江苏省六大人才高峰项目(2014-DZXX-040)

摘　　要：为了解决驾乘人员突然开车门导致的车门和后方接近目标发生碰撞的问题,提出了一种开门预警技术,通过低成本的连续波毫米波雷达实时检测接近车辆后方报警区域的运动目标,采用巴特沃斯4阶带通模拟滤波器和数字滤波器组,结合对消和加权方法对中频信号进行处理,通过降低中频数字信号副瓣功率的方法提高远距离目标的检测概率.同时采用单元最大最小平均恒虚警检测算法(CMMA-CFAR),根据噪声强度实时调整识别门限,保持较高的目标检测概率.该技术在奇瑞G5车上进行了标定和测试,针对3种典型接近目标自行车、摩托车和轿车,平均预警率分别为95.25%,95.50%,96.37%.结果表明:该技术能够在各种天气和时间,实时检测进入车辆后方报警区域的动态目标,并进行有效碰撞预警.In order to solve the problem that the rear closing target was collided due to the abrupt door opening by drivers and passengers, a technology of door open warning was proposed. The low-cost continuous wave millimeter-wave radar sensor was used to really detect the closing targets in the rear of vehicle. The intermediate frequency signal was processed with fourth-order butterwort band-pass filter and digital filter bank by cancellation and weighting methods. The far target detection rate was improved by reducing the digital signal sidelobe power. The cell maximum and minimum average-CFAR （CMMA- CFAR） algorithm was proposed to maintain high detection rate by adjusting threshold in time based on the noise intensity. The system was calibrated and tested on the Chery G5 car. For three representative closing targets of bicycles, motorcycles and motor vehicles,the early average warning rate reach 95.25%, 95.50% and 96.37%, respectively. The results show that the technology can really detect targets moving into the behind warning area of vehicle and give collision warning to driver and passenger effectively.

关 键 词：汽车工程 开门防撞预警 连续波毫米波雷达 车辆检测 目标检测 

分 类 号：U491.6[交通运输工程—交通运输规划与管理]