CTF定位策略下基于多特征的智能机器人目标跟踪

摘要/Abstract

摘要：

针对复杂环境下机器人目标跟踪问题，提出由粗到精定位策略下基于多特征的智能机器人目标跟踪方法. 该方法首先利用射频识别系统实现目标粗定位，然后采用自适应模板匹配算法、改进核函数的连续自适应均值飘移算法及扩展卡尔曼滤波算法提取目标头肩形状、衣服颜色与运动特征，实现精确定位. 最后根据人机运动状态设计基于模糊规则的智能调速控制器，实时自动调整机器人的基准线速度与转弯增益，以稳定跟随运动目标. 实验结果表明,该方法能有效保持人机之间的安全距离，对遮挡、相近颜色背景干扰及目标突然转弯的跟踪问题有较强的鲁棒性.

关键词: 智能机器人, 目标跟踪, 由粗到精定位策略, 多特征

Abstract:

To realize a human tracking task in a cluttered environment, a method of multifeature based human tracking under the CTF(coarsetofine) locating strategy was proposed. The proposed method located the target from a RFID system coarsely. Then, the silhouette of the headshoulder, the cloth color and motion feature were extracted to locate the target accurately by using the processing techniques including adaptive template matching algorithm, improved Camshift and Extended Kalman Filter. At last, an intelligent gear shift controller based on fuzzy rules considering the motion state of the target and the robot was utilized to drive the robot. The experimental results show that the presented method can keep the robot in a suitable distance from the target and handle the problem of occlusion, a sudden turn, and complicated background.

Key words：

Key words: intelligent robot, target tracking, coarse-to-fine (CTF) locating strategy, multifeature

中图分类号:

TP 391.4

贾松敏1,王爽1,王丽佳1,2,李秀智1. CTF定位策略下基于多特征的智能机器人目标跟踪[J]. 上海交通大学学报（自然版）, 2014, 48(07): 1039-1045.

JIA Songmin1,WANG Shuang1,WANG Lijia1,2,LI Xiuzhi1. Human Tracking Based on MultiFeature for Intelligent Robot Under the CTF Locating Strategy[J]. Journal of Shanghai Jiaotong University, 2014, 48(07): 1039-1045.

参考文献

［1］Yin Shimin, Na JinHee, Choi JinYoung. Hierarchical Kalmanparticle filter with adaptation to motion changes for object tracking［J］. Computer Vision and Image Understanding, 2011, 115(6): 885900.

［2］牛长峰, 刘玉树. 融合多特征的粒子滤波目标跟踪算法［J］. 华中科技大学学报:自然科学版, 2010, 38(1): 1821.

NIU Changfeng, LIU Yushu. Particle filter targets tracking by fusing multiple features［J］. Journal of Huazhong University of Science & Technology: Natural Science Edition, 2010, 38(1): 1821.

［3］Amir Hooshang Mazinan, Arash AmirLatifi. Applying mean shift, motion information and Kalman filtering approaches to object tracking［J］. ISA Transactions, 2012, 51: 485497.

［4］刘士荣孙凯, 张波涛, 等. 基于改进Camshift算法的移动机器人运动目标跟踪［J］. 华中科技大学学报：自然科学版, 2011, 39(增刊Ⅱ): 223226.

LIU Shirong, SUN Kai, ZHANG Botao，et al. Mobile robot moving object tracking based on modified Camshift algorithm［J］. Journal of Huazhong University of Science & Technology: Natural Science Edition, 2011, 39(Ⅱ): 223226.

［5］Germa T, Lerasle F. Videobased face recognition and tracking from a robot companion［J］. International Journal of Pattern Recognition and Artificial Intelligence, 2009, 23(3): 591616.

［6］戴渊明, 韦巍, 林亦宁. 基于颜色纹理特征的均值漂移目标跟踪算法［J］. 浙江大学学报:工学版, 2012, 46(2): 212217.

DAI Yuanming, WEI Wei, LIN Yining. An improved meanshift tracking algorithm based on color and texture feature［J］. Journal of Zhejiang University: Engineering Science, 2012, 46(2): 212217.

［7］李冠彬, 吴贺丰. 基于颜色纹理直方图的带权分块均值漂移目标跟踪算法［J］. 计算机辅助设计与图形学学报, 2011, 23(12): 26.

LI Guanbin, WU Hefeng. Weighted fragmentsbased meanshift tracking using colortexture histogram［J］. Journal of ComputerAided Design & Computer Graphics, 2011, 23(12): 20592066.

［8］刘婕, 张金艺, 查肇祥, 等. 重叠年轮式无源射频识别定位算法设计［J］. 上海交通大学学报, 2012, 46(11): 18161820.

LIU Jie, ZHANG Jinyi, ZHA Zhaoxiang, et al. Overlapped growth ring passive radio frequency identification positioning algorithm［J］. Journal of Shanghai Jiaotong University, 2012, 46(11): 18161820.

［9］林振华, 王营冠, 夏凌楠, 等. 错误测量下节点选择策略对跟踪精度的影响［J］. 华中科技大学学报：自然科学版, 2012, 40(8): 118122.

LIN Zhenhua, WANG Yingguan, XIA Lingnan, et al. Influence on tracking accuracy of nodes selection strategy with erratic measurement［J］. Journal of Huazhong University of Science & Technology: Natural Science Edition, 2012, 40(8): 118122.

［10］Sanpechuda T, Kovavisaruch L. A review of RFID localization: Applications and techniques［C］∥ Int Conf Electr Eng/Electron, Comput, Telecommun Inf Technol. 5th IEEE IntConf Electrical Engineering\\Electronics, Computer, Telecommunications and Information Technology. 445 Hoes Lane – P O Box 1331, Piscataway, NJ 088551331, United States: Inst of Elec and Elec Eng Computer Society, 2008: 769772.

［11］Parveen Flora, Briens Cedric, Berruti, et al. Effect of particle size, liquid content and location on the stability of agglomerates in a fluidized bed［J］. Powder Technology, 2013, 237: 276385.

［12］王景川, 方毅, 陈卫东. 移动机器人定位的自适应功率调节射频识别系统［J］. 上海交通大学学报, 2012, 46(2): 207212.

WANG Jingchuan, FANG Yi, CHEN Weidong. Mobile robot selflocalization based on RFID system with adaptive power control［J］. Journal of Shanghai Jiaotong University, 2012, 46(2): 207212.

［13］Ouadah N, Cadenat V, Lerasle F, et al. Multisensorbased control strategy for initiating and maintaining interaction between a robot and a human［J］. Advanced Robotics, 2011, 25(910): 12491270.

［14］胡春华, 马旭东, 戴先中. 一种基于行为的移动机器人目标人跟踪控制方法［J］. 控制工程, 2011, 18(2): 318321.

HU Chunhua, MA Xudong, DAI Xianzhong. A robust person following control method for behaviorbased mobile robot［J］. Control Engineering of China, 2011, 18(2): 318321.

［15］Satake J, Iura J. Stereobased multiperson tracking using overlapping silhouette templates［C］// ProceedingsInternational Conference on Pattern Recognition. 20th International Conference on Pattern Recognition，ICPR 2010. Piscataway, NJ 088551331, United States: Institute of Electrical and Electronics Engineers Inc, 2010: 43044307.

［16］Jia S M, Sheng J B, Shang E Z, et al. Robot localization in indoor environments using radio frequency identification technology and stereo vision［J］. Advanced Robotics, 2008, 22: 279297.

［17］Jia S M, Wang S H, Wang L J, et al. Robust human detecting and tracking using varying scale template matching［C］∥ IEEE Int Conf Inf Autom, ICIA. 2012 International Conference on Information and Automation. 2001 L Street N W, Suite 700, Washington, DC 200364928, United States: IEEE Computer Society, 2012: 2530.

［18］郭伟斌, 陈勇. 基于模糊控制的除草机器人自主导航［J］. 机器人, 2010, 32(2): 204209.

GUO Weibing, CHEN Yong. Fuzzy control based autonomous navigation for a weeding robot［J］. Robot, 2010, 32(2): 204209.

[1]	黄鹤, 吴琨, 李昕芮, 王珺, 王会峰, 茹锋. 自适应插值飞蛾扑火优化的多特征粒子滤波车辆跟踪算法[J]. 上海交通大学学报, 2022, 56(2): 143-155.
[2]	何子述, 苏洋, 程婷. 基于共址MIMO雷达组网系统的机会协同目标跟踪方法[J]. 空天防御, 2022, 5(1): 6-11.
[3]	侯煜冠, 韩远鹏, 谢金月, 毛兴鹏. 基于SDE模型的雷达组目标角度跟踪方法[J]. 空天防御, 2021, 4(1): 41-46.
[4]	郭维诚，李蓓智，杨建国，周勤之. 磨削过程信号监测与砂轮磨损预测模型构建[J]. 上海交通大学学报, 2019, 53(12): 1475-1481.
[5]	陈晓,李亚安,李余兴,蔚婧. 基于距离加权的概率数据关联机动目标跟踪算法[J]. 上海交通大学学报（自然版）, 2018, 52(4): 474-479.
[6]	李坦坦，雷明. 基于启发式逻辑的概率假设密度滤波高效航迹起始方法[J]. 上海交通大学学报（自然版）, 2018, 52(1): 63-69.
[7]	戴丽珍. 基于取向性的内发动机[J]. 上海交通大学学报（自然版）, 2017, 51(7): 892-896.
[8]	李家强1a,2，赵春艳1b,2，陈金立1a,1b，赵荣华1b,2. WGS-84坐标系下改进的广义最小二乘配准算法[J]. 上海交通大学学报（自然版）, 2016, 50(05): 771-775.
[9]	张铁，马琼雄. 人机交互中的人体目标跟踪算法[J]. 上海交通大学学报（自然版）, 2015, 49(08): 1213-1219.
[10]	李晓旭1，戴彬1，曹洁1,2. 基于多特征融合的改进UPF目标跟踪算法[J]. 上海交通大学学报（自然版）, 2014, 48(10): 1473-1478.
[11]	刘士荣1,2，吴楚1,3，张波涛1,2，章海兵1,2. 基于极线约束SIFT特征和粒子滤波的目标跟踪算法[J]. 上海交通大学学报（自然版）, 2014, 48(07): 1026-1032.
[12]	陈东岳，陈宗文. 基于特征显著性的均值漂移鲁棒目标跟踪[J]. 上海交通大学学报（自然版）, 2013, 47(11): 1807-1812.
[13]	张铁栋, 万磊, 王博, 曾文静. 基于改进粒子滤波算法的水下目标跟踪 [J]. 上海交通大学学报（自然版）, 2012, 46(06): 943-948.
[14]	王鹏, 李平康, 余俊, 尹长志. 一种多宿主机构架的xPC实时仿真系统[J]. 上海交通大学学报（自然版）, 2011, 45(08): 1181-1185.
[15]	郝燕玲，孟凡彬,王素鑫，孙枫. 一种新的多机动目标跟踪的GMPHD滤波算法[J]. 上海交通大学学报（自然版）, 2010, 44(07): 873-0877.