论文部分内容阅读
针对存在粗差或异常数据点时,最小二乘定位方法会产生定位错误的情况,本文提出了基于M-估计的稳健标靶球定位方法。通过M-估计剔除或减弱球体数据中异常点对参数估计的影响,获取稳健的标靶球球心坐标。利用仿真数据与真实数据进行验证,结果表明该法能克服异常值的影响,提高球体定位的准确度,具有稳健性。
In the case of a gross error or anomalous data point, the least square method will lead to positioning errors. In this paper, we propose a robust target ball localization method based on M-estimation. By M-estimation, the influence of the abnormal points in the sphere data on the parameter estimation is eliminated or weakened, and the robust target ball-center coordinates are obtained. The results of simulation and real data show that this method can overcome the influence of outliers and improve the accuracy of ball localization with robustness.