TY - GEN
T1 - A novel adaptive filtering algorithm for maneuvering target tracking
AU - Shi, Haoqin
AU - Zhou, Deyun
AU - Chen, Chuxin
PY - 2012
Y1 - 2012
N2 - For maneuvering target tracking, and based on the traditional "current" statistical model, the value of maneuvering frequency and maximum acceleration are set up in advance not in adaptive manner which is irrational for the reality. In the present work, a new adaptive filtering algorithm based on the "current" statistical model is proposed. In the proposed algorithm, the maneuvering frequency is adjusted based on fuzzy inference according to the value of the innovation and its change. At the same time, the maximum acceleration is adjusted according to the innovation and the estimated value. Compared to the traditional "current" statistical algorithm, the proposed algorithm improves not only the tracking accuracy and the robustness, but also the adaptation and the rapid response capabilities, for tracking non-maneuvering or weak maneuvering targets. The performance of the proposed algorithm is verified through the Monte Carlo simulations which validate the superiority of the proposed algorithm over the traditional "current" statistical algorithm from the point of view of rationality and validity.
AB - For maneuvering target tracking, and based on the traditional "current" statistical model, the value of maneuvering frequency and maximum acceleration are set up in advance not in adaptive manner which is irrational for the reality. In the present work, a new adaptive filtering algorithm based on the "current" statistical model is proposed. In the proposed algorithm, the maneuvering frequency is adjusted based on fuzzy inference according to the value of the innovation and its change. At the same time, the maximum acceleration is adjusted according to the innovation and the estimated value. Compared to the traditional "current" statistical algorithm, the proposed algorithm improves not only the tracking accuracy and the robustness, but also the adaptation and the rapid response capabilities, for tracking non-maneuvering or weak maneuvering targets. The performance of the proposed algorithm is verified through the Monte Carlo simulations which validate the superiority of the proposed algorithm over the traditional "current" statistical algorithm from the point of view of rationality and validity.
KW - "current" statistical model
KW - adaptive filter
KW - maneuvering frequency
KW - maximum acceleration
KW - target tracking
UR - http://www.scopus.com/inward/record.url?scp=84869216298&partnerID=8YFLogxK
U2 - 10.1109/MSNA.2012.6324551
DO - 10.1109/MSNA.2012.6324551
M3 - 会议稿件
AN - SCOPUS:84869216298
SN - 9781467324670
T3 - Proceedings - 2012 International Symposium on Instrumentation and Measurement, Sensor Network and Automation, IMSNA 2012
SP - 209
EP - 213
BT - Proceedings - 2012 International Symposium on Instrumentation and Measurement, Sensor Network and Automation, IMSNA 2012
T2 - 2012 International Symposium on Instrumentation and Measurement, Sensor Network and Automation, IMSNA 2012
Y2 - 25 August 2012 through 28 August 2012
ER -