摘要
卡尔曼滤波器对线性高斯滤波问题能提供最优解,而对目标运动模型、观测方程等要求的非线性就不再适合,提出了一种机动目标自适应非线性粒子滤波算法-“粒子滤波器”(Particle Filters PF)法,这种方法不受线性化误差和高斯噪声假定的限制,适用于任何状态转换或测量模型,分析比较了粒子滤波(PF)与扩展卡尔曼滤波算法(EKF)的滤波精度、运算量等方面指标。给出了基于典型非线性模型的算法仿真,仿真结果表明粒子滤波新方法优于EKF对机动目标跟踪。
The Kalman Filter is optimal solution to the filter problem for linear Gaussian model, but it is not suitable for nonlinear problem to maneuvering model and observation equation. In this paper, a novel nonlinear Particle Filter （PF） algorithm is proposed for maneuvering target adaptive tracking. The approach suits any model of state trans ition and observation. And it is not restricted by error of linearization and assumption of Gaussian to noise. The Particle Filters is compared with the Extended Kalman Filter （EKF） to filter precision and calculating complexity. Some simulation are performed to typical nonlinear model, the results have shown that the performance of PF is better than that of KEF for Maneuvering Target Tracking.
出处
《火力与指挥控制》
CSCD
北大核心
2007年第6期15-17,24,共4页
Fire Control & Command Control
基金
国家自然科学基金资助项目(60372081)
关键词
机动目标跟踪
粒子滤波
序列蒙特卡洛
贝叶斯估计
Maneuvering Target Tracking, Particle Filter, sequence Monte-Carlo, Bayesian estimation
