Vision-based object tracking method of mobile robot

被引：0
|
作者：
Yu D. [1 ,2 ]
Wang Y. [1 ,2 ]
Mao J. [1 ,2 ]
Zheng H. [1 ,2 ]
Zhou X. [1 ,2 ]
机构：
[1] College of Electrical and Information Engineering, Hunan University, Changsha
[2] National Engineering Laboratory for Robot Vision Perception and Control Technology, Changsha
来源：
Yi Qi Yi Biao Xue Bao/Chinese Journal of Scientific Instrument | 2019年 / 40卷 / 01期
关键词：
Fast discriminative scale space tracking(fDSST) algorithm; Kalman filter; Machine vision; Mobile robot; Target tracking;
D O I：
10.19650/j.cnki.cjsi.J1804340
中图分类号：
学科分类号：
摘要：
To achieve fast and stable tracking of pedestrian targets and simplify the robot system, one kind of tracking method combining fast discriminative scale space tracking (fDSST) correlation filtering and Kalman filtering is proposed. The problem of loss of target coordinate information caused by occlusion during tracking is solved. According to the oscillating severity of the relevant filter response graph, the occlusion judgment criterion is set. The occlusion judgment criterion is utilized to realize the switching between the fDSST tracking algorithm. The position coordination information of the target is continuously output. In this way, the robustness of the algorithm is improved. Based on the image coordinations provided by the visual tracking algorithm, the mobile robot uses the image-based servo control strategy to complete the follow-up task to the target. The structure of the mobile robot system is simplified. Finally, the method is evaluated on the OTB2013 test set and mobile robot. Experimental results show that the proposed method has strong robustness and accuracy for target occlusion and scale change. Meanwhile, the real-time requirements are satisfied. © 2019, Science Press. All right reserved.
引用
收藏
页码：227 / 235
页数：8
相关论文
共 25 条
  • [1] Gritti A.P., Tarabini O., Guzzi J., Et al., Kinect-based people detection and tracking from smallfootprint ground robots, IEEE/RSJ Internatio-nal Conference on Intelligent Robots and Systems, pp. 4096-4103, (2014)
  • [2] Wang J.E., Xiao X.Q., Mobile robot integrated navigation method based on QR code vision positioning, Chinese Journal of Scientific Instrument, 39, 8, pp. 230-238, (2018)
  • [3] Li P., Zhang J.Y., Han G.C., Et al., Local search tracking algorithm for LiDAR mobile robot, Electronic Measurement Technology, 41, 11, pp. 6-9, (2018)
  • [4] Vincze M., Schlemmer M., Gemeiner P., Et al., Vision for robotics: A tool for model-based object tracking, IEEE Robotics & Automation Magazine, 12, 4, pp. 53-64, (2005)
  • [5] Su D., Miro J.V., An ultrasonic/RF GP-based sensor model robotic solution for indoors/outdoors person tracking, 13th International Conference on Control Automation Robotics & Vision (ICARCV), pp. 1662-1667, (2014)
  • [6] Ullah I., Ullah Q., Ullah F., Et al., Integrated collision avoidance and tracking system for mobile robot, International Conference of Robotics and Artificial Intelligence, pp. 68-74, (2012)
  • [7] Bellotto N., Hu H.S., Multisensor-based human detection and tracking for mobile service robots, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 39, 1, pp. 167-181, (2009)
  • [8] Cui J.S., Zha H.B., Zhao H.J., Et al., Multi-modal tracking of people using laser scanners and video camera, Image and Vision Computing, 26, 2, pp. 240-252, (2008)
  • [9] Yuan J., Li Y., Dong X.L., Et al., Target tracking with a mobile robot based on on-line classification for motion patterns, Chinese Journal of Scientific Instrument, 38, 3, pp. 568-577, (2017)
  • [10] Jean J.H., Wang J.L., Development of an indoor patrol robot based on ultrasonic and vision data fusion, IEEE International Conference on Mechatronics and Automation, pp. 1234-1238, (2013)
123