ISBN: 978-981-11-3671-9 DOI: 10.18178/wcse.2017.06.175
Self-tuning Filter for High-Mobility Target Tracking System
Abstract— Since the conventional self-tuning kalman filter can not deal with the filtering problem in the high-mobility target tracking system, we present a kind of modified self-tuning kalman filter. This filter is based on the jerk model and modern time series analysis method. Firstly, we use system identification method to estimate the noise statistic information, and then jerk model is used to generate matrix information which could be substituted into the kalman filter to adapt to the high-mobility environment. The parameters of the jerk model could be set to fit the kalman filter. The excellent convergence is proved by mathematical way, and a simulation example is given to illustrate the effectiveness of this filter.
Index Terms— kalman filter, target tracking, self-tuning, jerk model
Li Heng1, LI Hong, Han Bo, Wang Hao, LI Huai-min, LI A-min
School of Computer and Information Engineering, Fuyang Normal University, CHINA
Cite: Li Heng, LI Hong, Han Bo, Wang Hao, LI Huai-min, LI A-min, "Self-tuning Filter for High-Mobility Target Tracking System," Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering, pp. 1010-1014, Beijing, 25-27 June, 2017.