量测野值与波束故障条件下SINS/DVL紧组合导航方法

Improved Huber robust filter SINS/DVL compact combination fault processing method based on beam reconstruction

针对传统捷联惯性导航系统(SINS)/声学多普勒测速仪(DVL)组合导航系统导航精度容易受量测信息质量影响的问题,提出了一种基于波束重构的SINS/DVL紧组合导航系统故障处理方案。以现有的紧组合模型为基础,根据四波束DVL结构特点,提出了新的故障波束速度信息重构方案。针对速度测量异常、量测模型误差及波束信息重构误差等综合引起的量测噪声特性变化问题,根据核函数思想,设计了一种改进Huber鲁棒滤波器。试验结果表明,在1 h波束故障情况下,采用所提出故障处理方案最大定位精度相对于纯惯性定位提高了80%以上。

Aiming at the problem that the navigation accuracy of traditional SINS/DVL integrated navigation system is easily affected by the quality of measurement information,a fault handling scheme of SINS/DVL tightly integrated navigation system based on beam reconstruction is proposed.Based on the existing compact combination model and the structural characteristics of four beam DVL,a new reconstruction scheme of fault beam velocity information is proposed.Aiming at the problem of measurement noise characteristic change caused by abnormal measurement speed,measurement model error and beam information reconstruction error,a robust adaptive filter is designed based on kernel function.The test results show that in the case of one hour beam failure,the maximum positioning accuracy of the proposed method is improved by more than 80%compared with the pure inertial positioning.