大尺寸复杂形状组合测量系统的全局标定与多视数据融合

为解决大尺寸复杂形状全局测量与局部精度控制的矛盾，提出以大空间测量设备为全局控制手段，集成终端近距离测量设备的组合测量、全局标定与数据融合方法.在多站位下观测测量控制网以获取冗余观测数据，利用测量平差优化技术完成控制网的高精度标定.建立全局测量坐标系与测量控制网的物理关联，实现测量空间基准定义的唯一性.布设扫描仪观测目标并建立基准坐标系，为扫描仪位姿空间定位提供观测目标.建立扫描仪坐标映射模型，基于平差优化技术完成模型的高精度标定.测量过程中通过移动扫描仪获取多视角精密测量数据，利用激光跟踪仪完成局部视角位姿的动态跟踪，结合控制网的坐标观测实现局部视角测量数据的全局标定与数据融合.实验结果表明，所提出的组合测量与标定方法有效地拓展了测量空间并控制了全局测量误差，同时避免了额外标定设备与标定操作的介入对测量工作的干扰.

Global Calibration and Multi-view Data Fusion for Combination Measurement System of Large Complicate Shapes

To eliminate the contradiction between global measurement and local precision control in 3D digitizing of large complex shapes, a combination measurement scheme is proposed, which is characterized by global control means with large space measuring device, and acquiring detail data with terminal close device, as well as global calibration of the system and data fusion of multi-views. In order to calibrate the global measurement control net, observations are performed at a number of sites to acquire redundant data. They are then fed into an optimization algorithm to precisely position the net. In order to uniquely define the measurement foundation during the whole measuring process, a global measurement coordinate system as well as the physical relation to the control network is setup. By means of fixing a number of observing target points on the scanner, a base coordinate system is built for detecting the pose and orientation of the scanner with the laser tracker. Moreover, a coordinate mapping model of the scanner is established, and its high precision calibration is carried out based on measurement adjustment optimization technology. During the measurement, multi-view shape data detail is obtained through moving the scanner, meanwhile the laser tracker performs on-time positioning and posture detecting of the scanner at each site. Combined with observation of the control net, global calibration of data of each local view and data fusion of them are achieved. Experimental results show that the combination measurement and calibration method proposed can effectively expand the measurement space and control the global measurement error. In addition, it avoids interfering with the measurement, due to the fact that additional calibration equipment and operation are no longer needed.