基于支持度和自适应加权的MEMS陀螺信息融合算法

针对MEMS陀螺阵列进行信息融合可以大幅提高其测量精度和可靠性，而传统信息融合算法大部分依赖于观测信息的先验知识而受到应用上的限制。在不依赖于先验知识的基于支持度的融合算法的基础上，提出一种新的加权系数构造方法，利用自适应加权法的思想改变各个传感器观测值方差对权系数的影响程度，既保证了阵列输出结果的可靠性，又能使融合后得到目标参数的总方差最小；而且自适应加权算法中，对各个传感器方差的计算考虑到环境因素等带来的噪声，与各传感器方差的真实值更加接近。实验结果表明，本文提出的融合算法优于传统的平均值估计融合算法和支持度融合算法。

Information Fusion Algorithm of MEMS-gyro Based on Support Degree and Random Weighting

The information fusion of MEMS-gyro can improve its measurement accuracy and reliability a lot. Howev?er,most traditional information fusion algorithms depend on prior knowledge of measuring information. This paper put forward a new construction method of weighting coefficient,which is on the foundation of the fusion algorithm based on support degree. The new method utilizes idea of adaptive weighting which can change the effects of the observation covariance matrix on weighting coefficients. Thus,the total variance of target parameter will be minimum and the reli?ability of gyroscope array can be guaranteed. Moreover,the new method also considers the environment noise,so the observation covariance approaches the true value closer. The result shows that the new method has a better effect.