Abstract: Particle filter is an effective method for non-Gaussian state filter and it has been gained special attention of researchers in various fields. There will be a new mixed particle filter proposed in this paper based on the extended Kalman particle filter. It uses the extended Kalman particle filter and MLP to generate particles for sampling the state at time and MLP is used to re-sample the state the second time. This structure makes use of the latest observation information. It has small error and better stability. The experimental results indicate that the proposed particle filter’s performance outperforms the other particle filters. The result indicates that this algorithm is a useful method for nonlinear filter problems.

Key words: particle filter, re-sampling, multi-modal noise, Multi-Layer Perceptron（MLP）

摘要： 粒子滤波是一种解决非高斯滤波问题的有效方法，受到许多领域的研究人员的重视。在扩展卡尔曼滤波（EKF）的基础上，提出一种基于多层感知器（MLP）的扩展卡尔曼滤波算法。利用扩展卡尔曼粒子滤波器和MLP对当前时刻状态重要性采样，引入MLP对样本进行重采样。该算法能有效利用测量值的最新信息，对状态估计的误差更小。在实验中，对于多模噪声非线性系统，该算法与另外算法进行比较。结果证明，所提算法性能优异于其他算法。

关键词: 粒子滤波, 重采样, 多模噪声, 多层感知器（MLP）