递推阻尼最小二乘及其在INS/双星组合中应用

由于双星系统的误差模型未知,且稳定性较差,采用Kalman滤波实现惯导(INS)/双星组合导航时,其滤波特性较差;而此时最小二乘法是实际工程应用中常用的方法,但易产生参数爆发现象.因此,提出了一种基于多输入-多输出(MI-MO)系统的递推阻尼最小二乘法,并将其应用到INS/双星组合导航系统中.仿真结果表明,利用递推阻尼最小二乘法能达到和递推最小二乘法同样的精度,且能明显抑制参数爆发现象.

Recursive damped least square and its application in INS/DS integrated system

The filtering performance is unsatisfactory while INS/DS (Double-star) integrated navigation system is implemented using Kalman filtering technology, because the model error of double-star system is unknown and the stability is n o t good, either. As an alternative, the least square(LS)scheme is widely used in practical engineering application. But LS scheme is easy to bring about the p roblem of parameter burst-off. Therefore, a modified LS scheme based on multi- in put-multi-output system, i. e. recursive damped least square (RDLS) scheme, wa s d eveloped, and be applied to INS/DS integrated navigation system. Simulation resu lts show that the RDLS scheme has the same estimation precision as the recursive least square (RLS) scheme, and furthermore, the parameter burst-off is obvious ly restricted.