多传感器组网系统空间配准算法

Multi-sensor networking system spatial registration algorithm

传统的偏差配准技术多基于球极投影,当传感器距离较远时,给配准结果引入一定的偏差,使用地心地固(ECEF)坐标系,以一个传感器为融合中心,提出了多传感器动态偏差的估计模型,能较好地解决了远距离偏差估计问题,由于的最小二乘法和广义最小二乘法不能满足实时动态系统配准,提出利用序惯卡尔曼滤波算法实现动态系统偏差估计,通过仿真试验说明该模型和算法能有效地进行传感器动态偏差估计.

Sensor registration is used to system bias estimation and compensation by means of measurements of multi-sensors to spatial common target. Some classical registration algorithms are all based on the stereographic projection, which introduce errors to the registration of the long distance sensors. In this work, we establish multi-sensor dynamic system bias estimation model, which is based on coordinate system of ECEF and selecting a senor as data processing center. This new approach solves the problem of registration between the long distance sensors. Least squares and generalized least squares registration algorithm can＇t do competent real time to dynamical bias registration. We suggest that the sensor bias estimates are obtained dynamically by using sequential process of Kalman filter, the result shows it can solve the registration problem effectively.