Electrical, Information Engineering and Mechatronics 2011 pp 1167-1173 | Cite as
Application and Research on Extended Kalman Prediction Algorithm in Target Tracking System
Target tracking is a critical important part in robot controlling system and it usually cannot get satisfied results because of the complex condition. The Kalman Forecast Algorithm on the camera platform which has two degrees of freedom has been researched in this paper. The motion model has been built based on the particular characteristic of the target tracking system. An extend kalman filter algorithm has been made up on this paper for this nonlinear model to dynamically compensate the error in line procession. The results show that the new algorithm can get fine effect in control of noise, time and punctual. And it meets the requirement of real-time target tracking. Therefore, a great value of application is showing up.
KeywordsExtended kalman prediction Target tracking Camera
This paper is supported by the national natural science fund project of Hubei province (number: 2010CDB02504), and the national natural science fund major project of Hubei province (number: 2010CBB0800).
- 1.Julier SJ (2004) Unscented filtering and nonlinear estimation. Proc IEEE 92(3):401–422 Google Scholar
- 2.Guo W, Gao X (2007) Kalman prediction algorithm based on three-dimensional head tracking. Shanxi University of Technology, 3(25):121–125Google Scholar
- 3.Qu S, Shi Z (2001) Image-based vehicle tracking. Flight Dyn 13(l2):014–020Google Scholar
- 4.Xiao Q, Lei B (2006) Kalman filter-based camera tracking. Xi’an Institute of Science 2(2):056–061Google Scholar
- 5.Haritaolu I, Harwood D et al (1999) Active outdoor surveil-lance. In: Proceedings, international conference on image analysis and processing, pp 1096–1009Google Scholar
- 6.Shen Y, Gong H, Yan X (2007) Application of adaptive Kalman filter in target tracking system. Comput Simul 11(24):143–148Google Scholar