基于鱼群优化粒子滤波的捷联惯导系统初始对准方法

Research on SINS Initial Alignment Method Based on Fish Swarm Optimization Particle Filter

摘要:

针对传统捷联惯导系统静基座初始对准模型的维数较高,导致滤波算法的解算实时性较差的问题,设计出一种基于鱼群优化粒子滤波的两位置初始对准方法。首先,建立了捷联惯导系统的两位置初始对准模型。由于该模型中不存在惯性器件的随机常值影响,因此,在确保初始对准精度的前提下,有效降低了初始对准模型维数；然后,利用鱼群优化算法改善了粒子滤波算法中粒子样本的分布,提高了粒子滤波算法的收敛速度和预测精度。仿真结果验证,采用该初始对准方法,可以有效提高初始对准的精度,且满足系统对实时性的要求。

Abstract:

Since the high dimension model of the conventional stationary base initial alignment for strapdown in. ertial navigation system(SINS) cannot ensure the real-time performance of filter algorithm, a two-position ini. tial alignment method using fish swarm optimization particle filter(FSOPF) is put forward. Firstly, a two-posi. tion initial alignment model is constructed. Because the new model removes the stochastic constant errors of iner. tial instruments, the dimension of the proposed model will be greatly decreased while its alignment precision can still be guaranteed. Secondly, by using fish swarm optimization, particles of particle filter algorithm are moved to the regions where they have larger values of posterior density function. Accordingly, the convergence speed and estimation accuracy of FSOPF are remarkably improved. Simulation results show that the speed of ini. tial alignment can be effectively improved and the real-time performance of filter algorithm can also be satisfied by using the proposed method.