Multi-stage observation data processing method based on nonlinear Gauss-Helmert model

Abstract

The invention discloses a multi-stage observation data processing method based on nonlinear Gauss-Helmert model, which comprises the following steps: step 1, establishing a non-linear Gauss-Helmert model; 2, substituting the first-stage observation data into the nonlinear Gauss-Helmert model to obtain parametric solution and precision information of the first-stage observation data; step 3, takingthe parametric solution and the precision information in the step 1 as the prior precision information of the parameters needing to be solved of the next period of observation data to perform parameter solution; 4, repeating the step 3 until the last stage of observation data, and obtaining the optimal solution of the multi-stage observation data; according to the method, the existing result is effectively utilized, compared with a single adjustment result, the parameter estimation precision is improved on the basis that existing information is fully mined, a globally optimal unique solutionis obtained under the criterion of least sum of square error, and the problem of multi-solution condition selection is solved.

Description

technical field [0001] The invention relates to the field of measurement, in particular to a multi-period observation data processing method based on a nonlinear Gauss-Helmert model. Background technique [0002] Sequential adjustment is also called sequential correlation indirect adjustment. It divides the observations into two or more groups, and performs correlation indirect adjustment according to the order of the groups. It does not need to consider the observations of the previous stage, but the previous group can be used The result of the adjustment can achieve the purpose of parameter solution combination. Because the parameters do not change with time, it is sometimes called static Kalman filter (Survey Adjustment Discipline Group, School of Surveying and Mapping, Wuhan University, 2003). Sequential adjustment effectively solves the problem of multi-period observation data processing in geodesy, and theoretically conducts rigorous formula derivation to combine param...

AI Technical Summary

Problems solved by technology

[0005] In the frame conversion problem, when multiple (two or more) independent observations are made on the common point, whether it is estimated by classical least squares theory or estimated by overall least squares correlation algorithm, there will be multiple sets of parameters In the case of the solution, because these parameters are not related to time, it is not possible to use Kalman filtering (Kalman fitting, KF) for parameter estimation

Images