[1]赖少发,刘华军.机动目标跟踪支持向量回归学习新方法[J].南京理工大学学报(自然科学版),2017,41(02):264.[doi:10.14177/j.cnki.32-1397n.2017.41.02.019]
 Lai Shaofa,Liu Huajun.Novel approach in maneuvering target tracking based onsupport vector regression[J].Journal of Nanjing University of Science and Technology,2017,41(02):264.[doi:10.14177/j.cnki.32-1397n.2017.41.02.019]
点击复制

机动目标跟踪支持向量回归学习新方法()
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
41卷
期数:
2017年02期
页码:
264
栏目:
出版日期:
2017-04-30

文章信息/Info

Title:
Novel approach in maneuvering target tracking based onsupport vector regression
文章编号:
1005-9830(2017)02-0264-05
作者:
赖少发刘华军
南京理工大学 计算机科学与工程学院,江苏 南京 210094
Author(s):
Lai ShaofaLiu Huajun
School of Computer Science and Engineering,Nanjing University of Scienceand Technology,Nanjing 210094,China
关键词:
机动目标跟踪 支持向量回归 汽车毫米波雷达
Keywords:
maneuvering target tracking support vector regression automotive millimeter-wave radars
分类号:
TP301.6
DOI:
10.14177/j.cnki.32-1397n.2017.41.02.019
摘要:
针对强机动性车辆目标的运动建模、控制输入建模和噪声建模的不精确导致的汽车雷达目标跟踪滤波精度低的问题,该文提出了基于支持向量回归(SVR)的机动目标跟踪滤波新方法。在常加速度(CA)模型的基础上,对理论新息协方差与实际新息协方差残差的Frobenius范数在线学习,获得过程噪声协方差的自适应调节因子,实时调整运动模型。对汽车雷达目标跟踪系统的仿真实验表明,该文算法降低了汽车雷达目标跟踪滤波对车辆运动模型和噪声模型的依赖程度,在强机动目标跟踪滤波性能上优于CA模型,比Singer模型具有更强的机动适应性和更高的精度。
Abstract:
Aiming at the problem that the motion modeling,the control input modeling and the noise modeling of the intense maneuvering target is not so accurate that bringing out the low accuracy in the automotive radar target tracking task,a novel method based on the support vector regression(SVR)for the manuevering target tracking is proposed here.Through the online learning on the Frobenius norm of the residuals of theory innovation covarivance matrix and the actual innovation covariance maxtrix based on the constant accleration(CA)model,a scaling factor of the process noise covariance matrix is estimated to correct the previous motion model and the noise model in real time.Experiments indicate that the approach has the low dependency on the motion model and the noise model,is far superior to the constant acceleration model(CA),and it has higher accuracy and adaption than the Singer model.

参考文献/References:

[1] Schubert R,Richter E,Wanielik G.Comparison and evaluation of advanced motion models for vehicle tracking[C]//11th International Conference on Information Fusion.Cologne,Germany:IEEE,2008:1-6.
[2]Zhai Guan,Meng Huadong,Wang Xiqin.A constant speed changing rate and constant turn rate model for maneuvering target tracking[J].Sensors(Basel),2014,14(3):5239-5253.
[3]Ozkan E.Lecture 3:Maneuvering target tracking[Z].Stockholm,Sweden:Linkoping University,2014.
[4]Li Xiaorong,Jilkov V P.Survey of maneuvering target tracking.Part I:Dynamic models[J].IEEE Transactions on Aerospace Electronic System,2003,39(4):1333-1364.
[5]Yuan Xianghui,Lian Feng,Han Chongzhao.Models and algorithms for tracking target with coordinated turn motion[J].Mathematical Problems in Engineering,2014,2014(2):1-10.
[6]张燕,柳超,李云鹏.基于改进Singer模型的机动目标跟踪方法[J].火控雷达技术,2015,44(3):37-40.
Zhang Yan,Liu Chao,Li Yunpeng.Maneuvering target tracking algorithm based on adaptive fading Kalman filter[J].Fire Control Radar Technology,2015,44(3):37-40.
[7]Bizup D F,Brown D E.Maneuver detection using the radar range rate measurement[J].IEEE Transactions on Areospace and Electronic System,2004,40(1):330-336.
[8]刘华军,赖少发.汽车毫米波雷达目标跟踪的快速平方根CKF算法[J].南京理工大学学报,2016,40(1):56-60.
Liu Huajun,Lai Shaofa.Fast square root CKF for automotive millimeter-waveradar target tracking[J].Journal of Nanjing University of Science and Technology,2016,40(1):56-60.
[9]Smola A J,Scholkopf B.A tutorial on support vector regression[J].Statistics Computing,2004,14(3):199-222.
[10]Liu Hui,Tian Hongqi,Li Yanfei,et al.Comparison of four Adaboost algorithm based artificial neural networks in wind speed predictions[J].Energy Conversion Management,2015,92(1):67-81.
[11]Chen Kuilin,Yu Jie.Short-term wind speed prediction using an unscented Kalman filter based state-space support vector regression approach[J].Applied Energy,2014,113(1):690-705.
[12]Zhang Yu,Dong Lian,Lai Lei,et al.Study of indoor positioning method based on combination of support vector regression and Kalman filtering[J].International Journal of Future Generation Communication and Networking,2016,9(3):201-214.
[13]Mehra R K.Approaches to adaptive filtering[J].IEEE Trans,Automatic Control,1972,17(5):693-698.

相似文献/References:

[1]陈建新,候建明,王 鑫,等.基于局部信息融合及支持向量回归集成的锂电池健康状态预测[J].南京理工大学学报(自然科学版),2018,42(01):48.[doi:10.14177/j.cnki.32-1397n.2018.42.01.007]
 Chen Jianxin,Hou Jianming,Wang Xin,et al.Prediction for state of health of lithium-ion batteriesby local informationfusion with ensemble support vector regression[J].Journal of Nanjing University of Science and Technology,2018,42(02):48.[doi:10.14177/j.cnki.32-1397n.2018.42.01.007]
[2]刘 皓,胡明昕,朱一亨,等.基于遗传算法和支持向量回归的锂电池健康状态预测[J].南京理工大学学报(自然科学版),2018,42(03):329.[doi:10.14177/j.cnki.32-1397n.2018.42.03.011]
 Liu Hao,Hu Mingxin,Zhu Yiheng,et al.Prediction for state of health of lithium-ion batteries by geneticalgorithm and support vector regression[J].Journal of Nanjing University of Science and Technology,2018,42(02):329.[doi:10.14177/j.cnki.32-1397n.2018.42.03.011]
[3]李龙刚,李立伟,杨玉新,等.基于改进灰狼优化与支持向量回归的 锂电池健康状态预测[J].南京理工大学学报(自然科学版),2020,44(02):154.[doi:10.14177/j.cnki.32-1397n.2020.44.02.005]
 Li Longgang,Li Liwei,Yang Yuxin,et al.Prediction for state of health of lithium-ion battery by improved grey wolf optimization and support vector regression[J].Journal of Nanjing University of Science and Technology,2020,44(02):154.[doi:10.14177/j.cnki.32-1397n.2020.44.02.005]

备注/Memo

备注/Memo:
收稿日期:2016-08-08 修回日期:2016-09-29
基金项目:国家“863”高技术研究计划资助项目(2015AA8106043); 国家自然科学基金(61402237; 61302156)
作者简介:赖少发(1991-),男,硕士生,主要研究方向:汽车毫米波雷达跟踪滤波,E-mail:213101255@seu.edu.cn; 通讯作者:刘华军(1978-),男,博士,副教授,主要研究方向:汽车毫米波雷达,智能汽车,人工智能等,E-mail:liuhj@njust.edu.cn。
引文格式:赖少发,刘华军.机动目标跟踪支持向量回归学习新方法[J].南京理工大学学报,2017,41(2):264-268.
投稿网址:http://zrxuebao.njust.edu.cn
更新日期/Last Update: 2017-04-30