车载毫米波雷达目标跟踪及运动状态分类方法
详细信息 查看官网全文
- 英文论文题名：Development of Target Tracking and Motion State Classification Algorithm for Vehicle-Borne Millimeter Wave Radar
- 论文作者：王竣 ; 王林 ; 王成浩 ; 张可科 ; 江丽君 ; 王大志 ; 刘军勇
- 英文论文作者：Wang Jun ; Wang Lin ; Wang Chenghao ; Zhang Keke ; Ji Lijun ; Wang Dazhi ; Liu Junyong ; SAIC Motor Technical Center
- 年：2016
- 作者机构：上汽技术中心;
- 论文关键词：先进驾驶辅助系统 ; 车载雷达 ; 目标跟踪 ; 运动状态分类
- 英文论文关键词：advanced driver assistance systems ; vehicle-borne radar ; target tracking ; motion state classification
- 会议召开时间：2016-10-26
- 会议录名称：2016中国汽车工程学会年会论文集
- 英文会议录名称：Proceedings of 2016 SAE-China Congress
- 语种：中文
- 分类号：U463.67
- 学会代码：QCGC
- 会议名称：2016中国汽车工程学会年会暨展览会
- 会议地点：中国上海
- 主办单位：中国汽车工程学会
- 学会名称：中国汽车工程学会
- 页数：4
- 文件大小：610k
- 原文格式：O
- 会议级别：全国
摘要
目前自适应巡航控制系统等驾驶辅助系统大多采用毫米波雷达探测前方车辆和障碍物,其中目标跟踪及其运动状态分类作为前方车辆和障碍物识别的前提基础,其跟踪性能及分类准确性直接关系到驾驶辅助系统有效目标识别的有效性。本文基于对汽车前方目标运动特点和车载雷达信息检测机理分析,考虑了地面车辆运动以地表平面上二维运动为主、机动性小、跟踪坐标系随动的特点,建立了基于车载雷达运动坐标系的前方目标的运动状态模型,并采用卡尔曼滤波算法实现了前方目标的跟踪估计。此外,基于对地面静动目标运动状态及转移机理的分析,本文将目标分为未分类目标、静止目标、同向运动目标、反向运动目标和起停目标五类,同时建立了目标运动状态的转移状态机模型实现目标运动状态分类。实车实验结果表明,本文提出的方法能够实现对雷达前方目标的快速准确跟踪,且长时间跟踪过程中滤波收敛稳定,并能通过分类算法实现对雷达目标的准确分类。
Millimeter wave radar is used to detect obstacles in front of ego vehicle by most Advanced Driver Assistance Systems(ADAS) such as Adaptive Cruise Control system(ACC).Because target tracking and motion state classification are the base of obstacles detection,so the performance of Advanced Driver Assistance Systems is directly influenced by it.With the analysis of preceding object motion feature and vehicle- borne radar measure principle,a novel target motion model has been established in the vehicle- borne radar coordinate system,meanwhile peculiarities of ground vehicle motion are taken into account in the model,such as the motion of ground vehicle is plane motion mostly,less mobility and tracking coordinate system is moving.The target tracking algorithm is developed based on Kalman filter.With extracting the features of object motion and analyzing the transfer mechanism between the object motion states,the object motion states have been classified into five types,including unclassified object,stationary object,moving object,oncoming object and stopped object.Then a suitable state machine is modeled for the object motion states.Finally,road tests have been carried out to verify the algorithm proposed in the paper.The results prove that this target tracking method provides fine estimation accuracy,better filter convergence and stability.And the classification algorithm can recognize the motion state of objects correctly.
Millimeter wave radar is used to detect obstacles in front of ego vehicle by most Advanced Driver Assistance Systems(ADAS) such as Adaptive Cruise Control system(ACC).Because target tracking and motion state classification are the base of obstacles detection,so the performance of Advanced Driver Assistance Systems is directly influenced by it.With the analysis of preceding object motion feature and vehicle- borne radar measure principle,a novel target motion model has been established in the vehicle- borne radar coordinate system,meanwhile peculiarities of ground vehicle motion are taken into account in the model,such as the motion of ground vehicle is plane motion mostly,less mobility and tracking coordinate system is moving.The target tracking algorithm is developed based on Kalman filter.With extracting the features of object motion and analyzing the transfer mechanism between the object motion states,the object motion states have been classified into five types,including unclassified object,stationary object,moving object,oncoming object and stopped object.Then a suitable state machine is modeled for the object motion states.Finally,road tests have been carried out to verify the algorithm proposed in the paper.The results prove that this target tracking method provides fine estimation accuracy,better filter convergence and stability.And the classification algorithm can recognize the motion state of objects correctly.
引文
[1]Xiao L,Gao F.A comprehensive review of the development of adaptive cruise control systems[J].Vehicle System Dynamics.2010,48(10):1167-1192.
[2]Kawakubo,S.Tokoro,Y.Yamada,et al.Electronically-Scanning Millimeter-Wave RADAR for Forward Objects Detection[J].SAE 2004-01-1122.
[3]张景波.刘昭度.齐志权.马岳峰.汽车自适应巡航控制系统的发展[J].车辆与动力技术.2003(2):44-49.
[4]侯德藻.汽车纵向主动避撞系统的研究[D].北京:清华大学,2004.
[5]刘大学,孙振平,戴斌.多线激光雷达越野环境障碍检测[J].计算机仿真,2009,07:159-163.
[6]Montemerlo,M.et al.Junior:The Stanford entry in the Urban Challenge[J].Journal of Field Robotics,2008(25),569-597.
[7]Broggi,A.,Cerri,P.,Ghidoni,S.,et al..A New Approach to Urban Pedestrian Detection for Automatic Braking[J].IEEE Transactions on Intelligent Transportation Systems 2009,10,594-605.
[8]Blackman S S.Multiple hypothesis tracking for multiple target tracking[J].Aerospace and Electronic Systems Magazine,IEEE,2004,19(1):5-18.
[9]ISO/DIS 15622.Road vehicles-Adaptive Cruise Control Systems-Performanoe requirements and test procedures[S].1999.
[10]Polychronopoulos A,Amditis A,Floudas N,et al.Integrated object and road border tracking using 77 GHz automotive radars[J].IEE Proceedings-Radar,Sonar and Navigation,2004,151(6):375-381.
[11]Nath R K,Deb D S K.Vehicle Detection Based on Video for Traffic Surveillance on road[J].International Journal of Computer Science&Emerging Technologies,2012,3(4).
[2]Kawakubo,S.Tokoro,Y.Yamada,et al.Electronically-Scanning Millimeter-Wave RADAR for Forward Objects Detection[J].SAE 2004-01-1122.
[3]张景波.刘昭度.齐志权.马岳峰.汽车自适应巡航控制系统的发展[J].车辆与动力技术.2003(2):44-49.
[4]侯德藻.汽车纵向主动避撞系统的研究[D].北京:清华大学,2004.
[5]刘大学,孙振平,戴斌.多线激光雷达越野环境障碍检测[J].计算机仿真,2009,07:159-163.
[6]Montemerlo,M.et al.Junior:The Stanford entry in the Urban Challenge[J].Journal of Field Robotics,2008(25),569-597.
[7]Broggi,A.,Cerri,P.,Ghidoni,S.,et al..A New Approach to Urban Pedestrian Detection for Automatic Braking[J].IEEE Transactions on Intelligent Transportation Systems 2009,10,594-605.
[8]Blackman S S.Multiple hypothesis tracking for multiple target tracking[J].Aerospace and Electronic Systems Magazine,IEEE,2004,19(1):5-18.
[9]ISO/DIS 15622.Road vehicles-Adaptive Cruise Control Systems-Performanoe requirements and test procedures[S].1999.
[10]Polychronopoulos A,Amditis A,Floudas N,et al.Integrated object and road border tracking using 77 GHz automotive radars[J].IEE Proceedings-Radar,Sonar and Navigation,2004,151(6):375-381.
[11]Nath R K,Deb D S K.Vehicle Detection Based on Video for Traffic Surveillance on road[J].International Journal of Computer Science&Emerging Technologies,2012,3(4).