光电探测系统参数与非参数模型运动学标定

Parametric and non-parametric model kinematic calibration of photoelectric detecting system

为了提高光电探测系统指向精度,提出并对比基于参数模型和非参数模型的运动学标定算法。首先,根据系统组成,全面分析光电探测系统指向误差来源。接着,针对参数模型,运用多体系统理论建立系统指向误差模型,并应用最小二乘法对模型进行标定。然后,针对非参数模型,应用双线性插值算法进行指向误差模型标定。最后,搭建实验平台,获得用于标定和验证的两组实验数据。实验结果表明:经过参数模型标定,指向精度从141.7″提高到22.2″;经过非参数模型标定,指向精度从141.7″提高到27.9″。两种方法均能提高光电探测系统指向精度,参数模型标定指向精度略高于非参数模型标定,但是非参数模型运动学标定具有过程简单、计算量小的优势。

In order to improve the pointing precision of photoelectric detecting system, parametric and non-parametric model kinematic calibration algorithms were proposed. Firstly, based on the component of system, the pointing error sources of photoelectric detecting system were analyzed. Subsequently, in times of the parametric model, pointing error model was established by multi-body system theory, and the least square method was used to calibrate the model. Then, aiming at the non-parametric model, bilinear interpolation algorithm was used to calibrate the pointing error. Finally, structure of the experimental platform, and two groups of data were measured to do the kinematic calibration and verify the calibration model. The experimental results demonstrate that the pointing precision has been improved from 141.7 ″ to22.2″ by parametric model calibration, and from 141.7″ to 27.9″ by non-parametric model calibration.These two algorithms can both improve the pointing precision. As for the system mentioned in this paper,the pointing precision of parametric model calibration is slightly higher, but non-parametric model calibration has simple modeling process and small computational complexity.