论文部分内容阅读
工业测量中,对于大型工件如大型飞行器、船舶分段搭建等的三维坐标测量,通常在一个固定测站点是无法覆盖全所有待测区域,这时常用通过公共点来衔接并移站测量的方法来解决。然而传统的移站测量法会随着移站次数的增加,点位测量误差不断累积增加,导致最终的测量精度不可控。提出了基于原始观测值和基于坐标观测值的两种整体平差数据融合方法,两种方法的区别在于,前者是指利用原始的测距数据和测角数据来估计所有未知测站点坐标以及公共点坐标,而后者则是先基于原始测距测角数据得到三位坐标值,再将这些坐标值作为观测数据来估计参数。与传统移站测量法相比,这两种整体平差解算的点位精度较高且稳定,实验表明,即使融合的函数模型和随机模型等价,采用不同的融合策略,得到的融合结果也并不完全等价。具体体现为,不同的融合估计策略在抗粗差稳定性、参数估计精度、随机模型偏差稳定性等方面都存在不同的表现。在实际应用中需根据具体情况选择不同的融合方法,从而得到解算精度可靠的待估计参数结果。
Industrial measurement, for large-scale workpieces such as large aircraft, three-dimensional coordinates of the shipbuilding structures, usually at a fixed station is unable to cover all the test area, then commonly used by the public point to connect and move the station measurement method To solve. However, with the traditional method of station measurement, as the number of station moves increases, the error of point measurement continuously accumulates, resulting in the final measurement accuracy being uncontrollable. The two methods based on the original observation data and the coordinate observation data are proposed. The difference between the two methods is that the former method is to estimate the coordinates of all the unknown stations by using the original distance measurement data and the angle measurement data, Point coordinates, while the latter is the first based on the original distance measurement angle data obtained three coordinates, then these coordinates as observations to estimate the parameters. Compared with the traditional station measurement method, the accuracy of the two point solutions is high and stable. Experiments show that even if the fusion function model and the stochastic model are equivalent, different fusion strategies are adopted, and the fusion result Not exactly equivalent. The concrete embodiment is that different fusion estimation strategies have different performance in terms of resistance to gross error, accuracy of parameter estimation, stability of stochastic model deviations, and the like. In practical application, different fusion methods should be selected according to the specific conditions, so as to obtain the result of the parameter to be estimated which has reliable calculation accuracy.