基于递推最小二乘的在线罗差校正方法

现有罗差校正方法不同程度地存在着需要精确控制采样点、校正结果受噪声影响大、不能实现现场环境下的在线校正等问题。针对这一问题,设计了一种基于递推最小二乘的在线罗差校正方法。介绍了罗差校正的基本原理,推导了基于递推最小二乘罗差校正的实现过程,设计了转台试验及车载试验进行验证分析。两次转台试验中,罗差校正前后系统航向角误差标准差分别由30.9418°与3.2407°降低至3.8861°与1.2964°;车载试验中,车辆仅需原地转圈即可实现罗差校正,15 min跑车结果显示,校正后航向角误差标准差由17.2037°降低至2.8818°。试验结果表明,该罗差校正方法简单易用,可应用于现场在线校正,相比传统方法高效、稳定。

Online magnetic deviation calibration method based on recursive least square algorithm

In current magnetic deviation calibration methods,there are varying degrees of problems such as the sample points need to be precisely controlled,noises could cause huge impact,and online calibration is unrealizable in field condition.This paper design an online method based on recursive least square algorithm to resolve these problems.The theory and realization are presented,and table tests and vehicle experiments are made.In the two table tests,the standard deviations of magnetic heading errors were reduced from 30.9418° and 3.2407° to 3.8861° and 1.2964° before and after the calibration,respectively.In the vehicle experiments,the calibration could be realized only by making the car to turn around.The standard deviations of magnetic heading errors same after calibration were reduced from 17.2037° to 2.8818° through 15 minutes’ moving.Experiment results show that this method is simple and easy to use,and it can be applied in field condition.This method is more effective and steady compared with traditional ones.