船用捷联惯性系统的IMU测量模型优化研究

Research on optimization model of marine strapdown inertial measurement unit

为了提高捷联系统惯性测量组件(IMU)的测量精度,在建立完整的惯组测量数学模型的基础上,采用逐步回归法对测量模型的自变量选择进行最优化选择,寻优时用复相关系数R2作衡量测量方程优劣的指标。试验结果表明,在陀螺测量模型中,角速度二次交叉耦合项系数和与比力的二次项系数非常小,加速度计的测量模型中的加速度二次耦合项系数也很小,模型加入该组变量后,R2变化并不大,均可以忽略不计。采用逐步回归分析法从众多的影响因素中,挑选对响应变量贡献大的因素,从而建立了惯组最优化测量模型。

To improve measurement precision of strapdown system IMU,stepwise regression method was employed to select optimized independent variables for measurement model based on integrated inertial mathematic model. The measurement equation adopted multiple correlation coefficient R^2 as superior-inferior index to find best solution. Experiments result indicate that the second cross coupling coefficient of angular velocity ,the quadratic coefficient of specific force, acceleration second coupling coefficient in accelerator measurement model and the varieties of R^2 are too small to attach importance to. Therefore, stepwise regression analytical method contributes to the establishment of inertial optimization measurement model by picking out large contribution factors to responding variables from multitudinous factors.