一种自适应H_∞滤波的运动学约束惯性导航方法

An adaptive H_∞ filtering approach for inertial navigation with motion constrained

纯捷联惯性导航系统的精度主要受限于传感器精度,为了在固有的传感器精度上进一步提高纯捷联惯性系统导航精度,提出了一种基于自适应H∞滤波的运动学约束辅助惯性导航算法。利用运动约束条件,将载体系中的侧向和上向速度作为量测量,将纯捷联惯性导航系统输出的姿态误差、速度误差及传感器误差作为状态量,采用一种能够自动调节滤波器参数的自适应H∞滤波方法以提高系统性能。实验结果表明,所提出的自适应H∞滤波运动学约束算法与无约束纯惯导算法相比能够提高23.7%~81%的定位精度,与无自适应H∞滤波运动学约束算法相比能够提高16.3%~62.2%的定位精度。

The accuracy of pure strapdown inertial navigation system (SINS) is mainly limited by the accuracy of the sensor. In order to improve the precision of SINS without replacing the sensor, a motion constrained assisted inertial navigation algorithm based on adaptive H∞ filtering is proposed. Based on the motion constraints, the lateral and vertical velocities in the SINS carrier system are chosen as the observation. Taking the attitude error, velocity error and sensor error of SINS as system states, an adaptive H∞ filtering method that can automatically adjust the filter parameters is adopted to improve the system performance. Experimental results show that the proposed adaptive H∞ filtering motion constraint algorithm can improve the positioning accuracy by 23.7%~81% compared with the unconstrained pure inertial navigation algorithm, and by 16.3%~62.2% compared with the normal H∞ filtering motion constraint algorithm.