国产POS与SWDC-4A集成检校的研究

发布时间：2018-03-15 12:27

本文选题：国产POS　切入点：数字航空相机　出处：《西南交通大学》2013年硕士论文　论文类型：学位论文

【摘要】：机载POS (Position and Orientation System)技术能够直接获取航摄瞬间影像的外方位元素,以摆脱对地面控制点的依赖,在航空摄影测量中发挥了重要的作用。在POS与航摄相机进行系统集成的过程中,为了将POS系统获取的位置和姿态元素转换为影像的外方位元素,需要对POS与航摄相机集成时的位置和姿态关系进行检校,即偏心角和偏心分量的检校。对多传感器集成的系统误差进行精密检校,是机载POS对地定位的一项关键技术。随着我国高精度POS、数字航空相机和惯性稳定平台的研制,进一步研究国产多传感器集成时系统误差的检校方法,对航测设备的国产化具有重要的现实意义。本文在像空间坐标系到摄影测量坐标系严密转换关系的基础上,通过引入地球曲率和子午线收敛角改正参数建立数学模型,再分别采用最小二乘法和均值法求解偏心角,然后将物方坐标系下的坐标残差转换到像空间坐标系中,进而求出了偏心分量。最后将原始POS数据加上偏心角和偏心分量的误差改正,从而解算出校正后相片的外方位元素。为了验证上述检校模型的可靠性,本文先后利用地面检校场和空对地检校场两种检校平台,根据一系列的模拟飞行和航测实测数据,分别进行了集成检校参数求解、集成传感器定向和直接地理定位数据处理。结果显示： (1)本文所建立的集成检校模型稳定可靠。利用该检校模型求解检校参数时,最小二乘法和均值法求解偏心角的精度相当,残差中误差均处在千分位上,求解偏心分量的残差中误差处在百分位上； (2)地面检校场能够提供高精度的检校参数,可以代替传统的空对地检校场进行作业。利用地面检校场所获取的检校参数中,角元素的中误差不大于0.02°,线元素的中误差不大于0.2m,将该检校参数改正到航测区域中后,可获得检查点平面中误差小于0.3m、高程中误差小于0.4m的精度； (3)国产POS用于集成传感器定向的精度能够满足规范要求。将检校后的POS数据作为初始值带入空三进行区域网平差,基本定向点的平面和高程中误差不大于0.05m,检查点的平而和高程中误差不大于0.2m。
[Abstract]:The airborne POS position and Orientation system can directly obtain the external azimuth elements of the aerial instant image, so as to get rid of the dependence on the ground control points and play an important role in aerial photogrammetry. In order to convert the position and attitude elements acquired by POS system into the external orientation elements of the image, it is necessary to calibrate the position and attitude relationship when the POS is integrated with the aerial camera. It is a key technology for airborne POS to accurately calibrate the system error of multi-sensor integration. With the development of high precision POSs, digital aerial camera and inertial stabilization platform in China, it is of great practical significance to further study the calibration method of system error in multi-sensor integration. In this paper, on the basis of the strict transformation from the image space coordinate system to the photogrammetric coordinate system, the mathematical model is established by introducing the earth curvature and meridian convergence angle correction parameters, and then the least square method and the mean method are used to solve the eccentric angle, respectively. Then the coordinate residuals in the object square coordinate system are transformed into the image space coordinate system, and then the eccentric component is obtained. Finally, the original POS data is added with the error correction of the eccentric angle and the eccentric component, and the external azimuth element of the corrected photograph is solved. In order to verify the reliability of the above calibration model, the integrated calibration parameters are solved based on a series of simulated flight and aerial survey data by using ground calibration field and air-to-ground calibration field, respectively. Integrated sensor orientation and direct geographic positioning data processing-results show:. 1) the integrated calibration model established in this paper is stable and reliable. When the calibration model is used to solve the calibration parameters, the accuracy of the least square method and the mean method to solve the eccentric angle is the same, and the error in the residual error is in thousands of quartiles. The error of the residual error of the eccentric component is on the percentile. 2) the ground calibration yard can provide high precision calibration parameters, which can replace the traditional air-to-ground inspection yard. The median error of angle element is not more than 0.02 掳, and that of line element is not more than 0.2 m. After correcting the calibration parameter into the aerial survey area, the accuracy of error in the checkpoint plane is less than 0.3 m, and the error in elevation is less than 0.4 m. The precision of domestic POS for integrated sensor orientation can meet the requirements of the specification. The calibrated POS data are brought into the space three as initial values for the adjustment of the regional network. The mean error of the plane and elevation of the basic orientation point is not greater than 0.05m, and the error of the check point is not greater than 0.2m.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P231;V245.6

【参考文献】