船用捷联惯导系统运动中对准的UKF设计  

Unscented Kalman filter for in-motion alignment of shipborne SINS

作　　者：顾冬晴[1] 秦永元[1] 

机构地区：[1]西北工业大学自动化学院,陕西西安710072

出　　处：《系统工程与电子技术》2006年第8期1218-1220,共3页Systems Engineering and Electronics

基　　金：总装"十五"预研基金项目资助课题(41309030308)

摘　　要：针对海况恶劣船体大幅晃动,捷联惯导系统无法快速完成自主初始对准的问题,给出了大方位失准角情况下捷联惯导系统的非线性误差方程,采用UKF方法来解决GPS辅助船用捷联惯导的运动中对准问题。仿真结果表明,该方法仅需60 s即可完成初始对准,并且水平对准精度达1′,方位对准精度达6′。与传统的EKF非线性滤波方法相比,采用UKF来实现船用捷联惯导系统运动中对准,不仅无需计算复杂的Jcobian矩阵,而且精度更高。Shipborne strapdown inertial navigation system （SINS） should use an in-motion alignment to obtain initial attitude information when ship rocking is violent in moorage. The in-motion alignment of ship SINS using an unscented Kalman filter （UKF） is described. The attitude, velocity and position error models of SINS under large heading misalignment angle are introduced. UKF for additive noise is developed to complete the inmotion alignment under large heading misalignment angle. Simulation results show that UKF could complete the alignment within 60 s with the estimation error of 1′ and 6′ for horizontal and heading misalignment angle respectively. And then compared with EKF, UKF could be implemented without calculating Jacobian matrices which can be a very difficult and error-prone process. So UKF presented for the in-motion alignment is better than EKF.

关 键 词：捷联惯导系统 对准 失准角 卡尔曼滤波 

分 类 号：U666.1[交通运输工程—船舶及航道工程]