Novel Particle Filtering Algorithm and Its Application for Underwater Target Tracking

YANG Yubo,WANG Huigang
DOI: https://doi.org/10.3969/j.issn.1002-8684.2011.03.013
2011-01-01
Abstract:In this paper,the fundamental knowledge of particle filtering algorithm and state partition technique with parallel EKF filters are introduced.It is combined with each other logically to obtain a new improved particle tracking method which is called PF-SP-PEKF algorithm,whose computation burden is moderate and enforceable.A distinctive way of obtaining the proposal distribution in the context of sequential importance sampling is provided.Meanwhile,this new method is implemented onto the underwater BTMA area in the passive measurement system,and the case that only one maneuvering target are discussed.At last,comparied with other well-known tracking algorithms,the PF-SP-PEKF algorithm demonstrates the superior performance and robustness.The experiment shows that in BTMA area,PF-SP-PEKF algorithm can yield the preferable tracking results to maneuvering target.
What problem does this paper attempt to address?