一种改进的高动态捷联惯导解算算法  

Improved algorithm for high dynamic strapdown inertial navigation system

作　　者：赵欣[1] 王仕成[1] 杨东方[1,2] 高运广[1] 罗大成[1] 

机构地区：[1]西安第二炮兵工程学院,西安710025 [2]清华大学计算机科学与技术系,北京100084

出　　处：《中国惯性技术学报》2011年第2期163-169,共7页Journal of Chinese Inertial Technology

基　　金：国家863课题资助项目(2009AA7050411);国家自然科学基金(60572061)

摘　　要：针对高动态环境下较大的不可交换误差难以精确补偿的问题,提出了一种改进的等效旋转矢量三子样多回路迭代姿态解算算法。首先分析了高动态环境给捷联惯导系统带来的影响,阐述了捷联解算的基本过程;然后对比分析了几种主要的姿态解算算法,针对一般算法难以较好地补偿不可交换误差问题,在旋转矢量多子样二次迭代优化算法的基础上,从提高解算精度和减小计算复杂度两方面进行考虑,研究了一种改进的等效旋转矢量三子样多回路迭代算法;最后,对改进的算法的性能进行了深入的分析。实验结果表明,在低动态条件下改进的算法和毕卡四元数法姿态解算精度相当,在高动态条件下其精度较毕卡四元数提高约3个数量级。To accurately compensate the large coning error aroused by high dynamic circumstance,an improved three-sample multi-loop equivalent rotation vector iterative algorithm is presented.Firstly,the paper analyses the influence to SINS brought by high dynamic circumstance,and expatiates the basic theory of strapdown solution.Then several main attitude solution algorithms are compared and contrasted.To solve the problem that common algorithms can not compensate the coning error perfectly,this paper,based on the multi-sample dual-loop equivalent rotation vector iterative algorithm,considers the algorithm performance from heightening precision and reducing calculation,and studies an improved three-sample multi-loop equivalent rotation vector iterative algorithm.Finally,the performance of the proposed algorithm is analyzed.The low-dynamic testing experiment and high dynamic simulation experiment results show that the attitude accuracy of the improved algorithm is similar to that of Peano Baker quaternion numbers in low dynamic conditions,but is higher than the attitude accuracy of Peano Baker quaternion numbers by 3 orders of magnitude in high dynamic conditions.

关 键 词：捷联惯导系统 姿态解算 等效旋转矢量 高动态 

分 类 号：U666.1[交通运输工程—船舶及航道工程]