CDKF在GPS/SINS组合导航系统非线性模型中的应用

摘要

The nonlinearity of GPS/SINS navigation system model affects the estimation accuracy of extend Kalman filter (EKF). Central difference Kalman filter (CDKF) is a new nonlinear filtering method, which adopts the interpolation formula for the state estimation approximation of nonlinear systems and reduces the influence of the linearization on system accuracy.According to the characteristics of GPS/SINS navigation system, the nonlinear error model of an integrated navigation system was established. EKF and CDKF are respectively applied to the model of the navigation system toconduct the simulation comparative study. The simulation results show that CDKF method is easy to implement, meets the navigation requirement of the system in a nonlinear model, and has higher precision and convergence.%GPS/SINS组合导航系统模型的非线性会导致扩展卡尔曼滤波(EKF)的估计精度降低.而中心差分卡尔曼滤波(CDKF)的新型非线性滤波方法,则利用插值公式对非线性系统的状态估计进行逼近,从而减小线性化误差对系统精度的影响.针对GPS/SINS导航系统的特点,建立了一种非线性误差模型,并将EKF与CDKF分别应用于组合导航系统模型中进行仿真比较.仿真结果表明,该算法简单易实现,且能满足系统在非线性模型下的导航要求,并具有较高的精度和收敛性.