Interacting multiple sensor unscented Kalman filter for accelerating object tracking

Author

Liu, Zhigang ; Wang, Jinkuan ; Qu, Wei

Author_Institution

Dept. of Autom. Eng., Northeastern Univ., Qinhuangdao, China

fYear

2010

fDate

10-12 April 2010

Firstpage

143

Lastpage

146

Abstract

Due to limited sensing range for sensor nodes, moving object tracking has to be realized by relaying from one node to the other in a cluster. By taking object tracking in a fixed cluster as a Markov jump nonlinear system, the interacting multiple sensor unscented Kalman filter(IMSUKF) algorithm is designed to deal with distributed tracking. The proposed method can be divided into two parts: one-step unscented Kalman filter for object tracking and the fusion of the information provided by all the nodes. Finally, simulation results show the effectiveness of the proposed method.

Keywords

Kalman filters; Markov processes; object detection; sensor fusion; target tracking; Markov jump nonlinear system; accelerating object tracking; distributed tracking; fixed cluster; information fusion; interacting multiple sensor; unscented Kalman filter; Acceleration; Algorithm design and analysis; Automation; Clustering algorithms; Collaboration; Filtering; Kalman filters; Nonlinear systems; Sensor systems; State estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

Networking, Sensing and Control (ICNSC), 2010 International Conference on

Conference_Location

Chicago, IL

Print_ISBN

978-1-4244-6450-0

Type

conf

DOI

10.1109/ICNSC.2010.5461518

Filename

5461518