用于多传感器最优融合的优化节点样条数据平滑方法

对多个精度不同的传感器观测数据采用了加权平均法，给出了观测数据的最优融合方法；在此基础上，利用优化理论和最小二乘准则，对高机动目标的运动轨迹实现了B样条函数逼近，使逼近过程适应目标的运动规律，仿真计算结果表明其明显地优于微分平滑和等距节点平滑。

AN OPTIMAL KNOTS SPLINE MEASUREMENT SMOOTHING METHOD USED IN THE OPTIMAL FUSION OF MULTI-SENSOR DATA

When fusing multi sensor measurements of dissimilar precisions, the weighted mean method was often used. An optimal fusion method was proposed for multi sensor measurements. Based on the optimal fusion of multi sensors, applied to optimization theory and the least square principle, the movement of objective is described by the B spline with optimal knots. Thus the approximation process is adaptively fit for the track of moving target with high velocity. Results of calculation show that the optimal knots spline method evidently excels the differential coefficient smoothing method or the isometry knots smoothing method.