Research On Optimized Satellite Selection Method Of GNSS Baseline Network

In the Global Navigation Satellite System(GNSS)combined positioning strategys,the technology of setting up a GNSS baseline network is widely used in many fields of deformation detection.At present,the main factors affecting the accuracy of the GNSS baseline network are:the layout of the network type,the accuracy of the baseline composition,and the combination strategy of the satellite.The network layout technology is now becoming mature,and how to improve the latters need further research.In this paper,in order to solve the GNSS baseline network solution's questions that tthe network adjustment solution is inefficient and inaccurate because of satellite data redundancy and constellation structure being unreasonable,.This paper introduces Nondominated Sorting Genetic Algorithm ?(NSGA-?)to determine reasonable satellite combinations in the baseline network and make improvements.At the same time,to improve the accuracy of the random model in the baseline,Helmert variance component estimation is used to determine the weight ratio between the systems to provide more reliable and accurate positioning values for the star selection strategy.The main research contents and results of this article are as follows:(1)This paper analyzes the three core steps of unifying the spatiotemporal baseline,the baseline solution model,the free network adjustment model,and the accuracy evaluation in the GNSS baseline network solution.And to solve the inaccuracy of the random model in the GNSS baseline solution,Helmert's variance component estimation is used to determine the weight ratio between the systems to improve the baseline accuracy and the GNSS network adjustment results.This method belongs to posterior difference,and uses the residual correction number after pre-adjustment to estimate the weight ratio between the carrier phase observations of each system in real time to optimize the stochastic model.The accuracy of the algorithm is verified by comparison experiments using Helmert fixed weight models and equal weight models between the systems.The experimental results show that,compared with the equal weight model,the accuracy of the solution under the Helmert fixed weight model is significantly improved in the X,Y,Z direction and the position error,which is more reliable.(2)Aiming at the problems of irrational constellation structure and reduced positioning accuracy caused by redundant satellite data in the GNSS baseline network,the NSGA-? multi-target genetic algorithm is proposed to provide a reasonable satellite subset solution to meet the overall accuracy of each monitoring point in the network.This algorithm belongs to the intelligent optimization algorithm.By imitating the natural evolution rule of "survival of the fittest",the excellent"individuals" are retained in the population along the evolutionary direction,and the optimal solution in the optimization problem is obtained by multiple iterations.The evolution direction is the GNSS network adjustment points accuracy,and each group of satellites is regarded as "individuals".NSGA-? multi-target genetic algorithm is used to provide the optimal satellite selection strategy for the GNSS network.The experimental results show that the satellite selection combination obtained by the NSGA-? multi-objective genetic algorithm can comprehensively consider the impact of multiple errors on the positioning result,and break through the traditional star selection's dependence on the Geometric Dilution of Precision(GDOP).In theory,compared with the positioning result of the optimal satellite combination with GDOP value,the positioning accuracy is higher and more reliable.(4)Based on the GNSS baseline network adjustment principle and the research content of this paper,a baseline network solution software based on the GNSS multi-system combination is established-The software has the following three functions:?Realize the functions of multi-system single-point positioning,relative positioning,and free network adjustment calculation;?In the multi-system solution module,give the Weight ratio;?In the GNSS baseline network solution module,the NSGA-II algorithm is used to give a reasonable satellite selection scheme.Figures[21]Tables[4]References[90]...