论文成果
A Novel Modified Particle Filter Algorithm
- 点击次数：
- 论文类型：会议论文
- 发表时间：2011-04-16
- 收录刊物：EI、CPCI-S、Scopus
- 文献类型：A
- 卷号：204-210
- 页面范围：1895-1899
- 关键字：particle filter; state estimation; probability distribution
- 摘要：The particle filter (PF) algorithm provides an effective solution to the non-linear and non-Gaussian filtering problem. However, when the motion noises or observation noises are strong, the degenerate phenomena will occur, which leads to poor estimation. In this paper, we propose a modified particle filter (MPF) algorithm for improving the estimated precision through a particle optimization method. After calculating the coarse estimation with the traditional PF, we optimize the particles according to their weights and relative positions, then, move the particles toward the optimal probability distribution. The state estimation and target tracking experiments demonstrate the outstanding performance of the proposed algorithm.
