基于地心坐标系的传感器极大似然配准算法

An Maximum Likelihood Registration Algorithm for Sensor Alignment Based on Earth-Centered Earth-Fixed Coordinate System

配准是多传感数据无误差坐标转换的必不可少的处理过程。现有的精确极大似然 (EML)配准算法[1]是基于立体几何投影法进行的 ,没有考虑地球地形的影响 ,因此当传感器之间相距很远时将失去实际意义。提出了一种基于地心坐标系 (ECEF)的传感器极大似然配准算法。这种方法考虑到了地球形状对配准的影响。最后给出的仿真结果说明这种配准算法初步解决了远距离的传感器配准。

Registration is the essential process of error-free coordinate conversion of multiple sensor data. The existed EML (exact maximum likelihood) algorithm is based on the stereographic projection, it does not consider the geometry of the global. So it will be failure, while the distance between sensors is far away. An maximum likelihood registration algorithm based on earth-centered earth-fixed (ECEF) coordinate system is presented. This algorithm considers the geometry of the global. Finally, the simulative result based on this registration algorithm is given, which shows it preliminarily solves the registration between the long distance sensors.