Title :
A yaw-aided interacting multiple model tracking algorithm
Author :
Bing Ji ; Gan-lin Shan ; Hai Chen
Author_Institution :
Dept. of Opt. & Electron. Eng., Mech. Eng. Coll., Shijiazhuang, China
Abstract :
Based on the analysis of correlation among three pose angles, the paper proposed a yaw-aided interacting multiple model tracking algorithm (YAIMM). Firstly, instant angular velocity of the target was estimated from yaw information. Then, applicable models were chosen from preparative model set based on fuzzy association method, and their model probabilities were also computed. At last, weighted sum of state estimations from all selected models was received as final state estimation. Comparing with IMM, YAIMM improves the adaptability of model set and reduces the computation of model probability, which shows that YAIMM has better precision and real-time performance.
Keywords :
fuzzy set theory; object tracking; probability; state estimation; YAIMM; fuzzy association method; model set adaptability improvement; model set probability computation reduction; pose angle correlation analysis; preparative model set; state estimation weighted sum; target angular velocity estimation; yaw information; yaw-aided interacting multiple model tracking algorithm; Acceleration; Adaptation models; Angular velocity; Computational modeling; Correlation; Estimation; Target tracking; IMM; angular velocity; fuzzy association; model probability; yaw;
Conference_Titel :
Image and Signal Processing (CISP), 2012 5th International Congress on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4673-0965-3
DOI :
10.1109/CISP.2012.6469912