Research On Reliability Of RTK Localization In Complex Environment

RTK(Real-time Kinematic),real-time kinematic carrier phase differential positioning technology,has centimeter-level or even millimeter-level positioning under short baselines,and has a wide range of applications in surveying,monitoring,and navigation.However,due to the special transmission requirements of satellite signals,it causes their susceptibility to interference,when possesses advantages include wide coverage,long transmission distance,and high data quality.In RTK positioning,the satellite signal interfered by the external environment seriously affects the accuracy and reliability.This paper studies the positioning reliability based on the common environmental disturbances in satellite positioning,such as vegetation occlusion,barbed wire and other phenomena,mainly including cycle slip detection,cycle slip recovery,error resistance,integer ambiguity resolution and confirmation algorithms.The main innovations of the paper are as follows:(1)The influence of environmental disturbance on the receiver and RTK positioning model is introduced,and the issues of cycle slip,gross error and outlier are studied.In terms of cycle slip,the principle of two classical cycle slip algorithms is briefly introduced,and an improved high-order difference method is proposed for non-continuous epoch cycle slips.The algorithm uses the loss of lock count to determine the sampling interval,and supplements the data.Experiments show that the algorithm can effectively detect cycle slips in discontinuous epochs.The cycle slip recovery strategy is mainly aimed at the impact of repeated cycle slips on the ambiguity resolution in real-time positioning.The impact of the recovered cycle slip satellite data on the AR(Ambiguity Resolution)ratio is used to decide whether to use the satellite data for ambiguity resolution,and the effectiveness of the method is verified by comparative experiments.(2)In terms of gross errors and outliers,the interference of external environmental disturbances to the KF model is analyzed.For the initial filtering divergence issue,a single-point positioning robust algorithm determined by the least squares residual is proposed.Based on the actual data,the effectiveness of the algorithm is proved.In the relative positioning process,on the basis of Extended Kalman Filter(EKF),the superior gross error processing performance of the robust EKF based on the IGGIII model is verified.The way observations are variance-inflated improves the usability of robust EKF.The above algorithms are integrated into the RTK positioning process to form a reliability strategy,and the excellent anti-interference effect of the reliability strategy in complex environments is verified by grouping.(3)In terms of ambiguity solution,when the search time of LAMBDA algorithm is too long,an adaptive weighted differential evolution algorithm(Adaptive Weighted Differential Evolution Algorithm,AWDE)is proposed to solve the multi-dimensions ambiguity solution,and compared with the six algorithms for the calculation speed and stability.The AWDE algorithm can achieve success rate of 97% in solving 12-dimensional integer ambiguity,the optimal number of iterations is kept within 20 generations,and the average solution speed is 0.16 seconds.In terms of ambiguity confirmation,the shortcomings of the traditional fixed ratio threshold test method are proposed,and the effectiveness of dynamic Ratio threshold adjustment based on the Fixed Failure-rate Ratio test(FFRT)of ambiguity is verified.