AUV视觉与短基线定位自适应融合方法

Adaptive Data Fusion Method of Vision and SBL Position for AUV

为了使AUV(自主水下航行器)在水下自主对接过程中获得更可靠的定位数据,本文研究了基于视觉和短基线定位系统同时工作时的定位数据融合问题;由于卡尔曼滤波算法在进行数据融合时对于过程噪声与测量噪声统计数据具有依赖性的缺点,提出了一种基于模糊逻辑的在线自适应卡尔曼滤波定位数据融合方法;水池试验结果表明,该方法有效地提高了视觉和短基线融合后的定位精度,切实可行。

The vision position and SBL position were applied to underwater docking of AUV,and thus the problem of the data fusion method for vision and SBL was researched for position precision improvements. In order to overcome the shortcomings of Kalman filtering algorithm for its dependence of process noise and measurement noise statistics during the data fusion,an online adaptive Kalman filter positioning data fuzzy method was proposed. The results of underwater docking in pool tests indicate the position precision is im-proved by data fusion of vision and SBL position,which proves that the proposed method is effective.