粒子滤波理论及其在目标跟踪中的应用

非线性估计领域的经典算法是扩展Kalman滤波（EKF），它采用了Taylor展开的线性变换来近似非线性模型，因而存在计算量大、实时性差、估计精度低等缺点。而粒子滤波采用一些带有权值的随机样本（粒子）来表示所需要的后验概率密度，而不是采用传统的线性变换，从而得到基于物理模型的近似最优数值解，具有精度高、收敛速度快等特点。对经典的纯方位跟踪问题进行了仿真。仿真结果表明，粒子滤波器的跟踪性能要远优于EKF的性能。

Particle filtering theory and its application in target tracking

The Extended Kalman Filter (EKF) is the most popular approach to recursive nonlinear estimation.Because it is a linearization technique based on a first order Taylor series expansion of the nonlinear system and measurement functions about the current estimate of the state,it often provides an insufficiently accurate representation in many cases.The particle filtering method has become an important alternative to the EKF.It does not involve linearizations around current estimates but rather represent the desired distributions by discrete random measures,which are composed of weighted particles.It has a high accuracy and a rapid convergence.A simulation example of the bearings-only tracking problem is presented,and the result proves that the performance of the particle filter is greatly superior to that of the EKF.