EPF和PF在水下三维纯方位目标跟踪中的应用

Application of EPF and PF in underwater 3D bearings-only target tracking

针对水下目标跟踪问题,以静止双观测站三维纯方位跟踪系统为研究对象,介绍粒子滤波(Particle Filter,PF)和扩展卡尔曼粒子滤波(Extended Kalman Filter,EPF)的基本思想和算法实现步骤,根据建立的目标运动模型,在目标运动速度不同、粒子数目不同的情况下,将EPF、PF在双观测站三维纯方位目标跟踪系统中进行仿真分析,并结合UKF、EKF算法进行对比,结果表明,EPF算法相较于其他算法有更好的跟踪效果,并且不需要选取过多的粒子数目就可以达到较好的跟踪效果,但跟踪时间长、实时性较差。

Aiming at the problem of underwater target tracking,a three-dimensional bearings-only tracking system with static dual observation stations is taken as the research object,the basic idea and algorithm implementation steps of Particle Filter and Extended Kalman Filter are introduced,according to the established target motion model,when the target motion speed is different and the number of particles is different,simulation analysis was conducted on EPF and PF in a dual observation station 3D bearings-only tracking system,and compare with UKF and EKF algorithms,the results indicate that,EPF algorithm has better tracking performance compared to other algorithms,and it can achieve better tracking effect without selecting too many particles,but the tracking time is long and the real-time performance is poor.