Chinese Journal of Computers   Full Text
Title	A New Particle Filter for Nonlinear Filtering Problems
Authors	WANG Fa-Sheng ZHAO Qing-Jie
Address	(Beijing Key Laboratory of Intelligent Information, School of Computer Science & Technology, Beijing Institute of Technology, Beijing 100081)
Year	2008
Issue	No.2(346—352)
Abstract &   Background	Abstract Particle filters have gained special attention of researchers in various fields. The key idea of this technique is to represent the posterior density by sets of weighed samples. This paper proposes a new particle filter which is based on the extended Kalman filter and the Unscented Kalman filter. It first uses the former to generate an estimate of the state at time k, and then uses the latter to repeat the process and to gain the final estimate of the state and corresponding covariance at time k. In the experiments, the authors test five different particle filters on two different nonlinear systems. The experimental results indicate that the proposed particle filter has much better performance than the other four particle filters do. keywords nonlinear filtering; extended Kalman filter; Unscented Kalman filter; mixed Kalman particle filter background Nonlinear filtering problems generally exist in many fields such as robotics, artificial intelligence, biomedicine, computer vision, statistics, industrial control and signal processing. Studying on nonlinear filtering technique is beneficial to the development of related subjects. The achievements will possibly bring active influence to industrial control, finance, geological exploration, aeronautics and astronautics. Recently particle filtering technique has attracted more and more researchers’ attention due to its capacity to handle nonlinear filtering problems, and it has been used successfully in some fields. The group has been concentrating on "high-efficient particle filtering technique and corresponding applications" under the support of National Natural Science Foundation of China under grant No.60772063 and the foundational fund of Beijing Institute of Technology under grant No.200501F4210. In this project, the authors are aiming to find more efficient nonlinear filtering algorithms in order to supply effective tools to solve practical problems. The group has acquired considerable achievements and there are more than seven papers being published. The results presented in this work are associated with new nonlinear particle filtering algorithm adopting different strategies, which can achieve better results than others. Corresponding applications in machine vision and some other fields are being done.