Algorithm Study With GPS/BDS And PL High Precise RTK Positioning

GPS conventional RTK positioning technology has been widely used in many fields,although the basic positioning algorithm is relatively mature,on some challenged observation environments,such as the downtown area,the serious signal obstruction by trees,and so on,the positioning performance is limited due to the insufficient visible satellites and the weak geometric structure.With the gradual improvement of BDS.system in China,the combination systems can significantly increase the number of visible satellites and improve the distribution of spatial geometry,compared with the single system,the combination system can improve the positioning performance to some extent,thus this paper firstly studies the GPS/BDS conventional RTK positioning algorithm and the corresponding software system;aiming at the performance limitation of the GPS/BDS conventional RTK due to the larger distance between rover station and base station,the medium long RTK is further researched in this paper,in order to broaden the conventional RTK base-rover station distance,and provides a foundation for the application of high precision RTK measurement in a relatively lager areas(such as the sea or desert areas),as well as can provide a reference for network RTK error modeling in the city area;aiming at the problem that GPS/BDS will lose the positioning capability in the condition of complete signal occlusion,this paper finally studied indoor pseudolite(PL)high precision RTK positioning algorithm and the corresponding software system,to make up for the defects of GPS/BDS could not positioning in the indoor environment,apply the feasibility of indoor high precision measurement such as for the industrial applications.The purpose of this paper is to deeply study the medium long RTK GPS/BDS and conventional RTK,and indoor pseudolite RTK algorithm,mainly related to the basic positioning mathematical model,error sources and well-known processing method,ambiguity resolution(AR)for carrier phase measurements,quality control and other fields,based on the existing algorithm,according to some certain limitations,deeply analyze the causes and make the corresponding algorithm improvement and innovation,to meet the high precision RTK positioning adapt to more environment,as well as to provide a feasible basis for broadening the application fields of RTK.The main work and contributions of this paper are as follows:(1)The basic theory of RTK positioning is summarized systematically,mainly including the basic positioning mathematical model(function model and stochastic model),parameter estimation method,common error sources and processing methods,some well-known methods of cycle slip detection and repair,ambiguity resolution,quality control.(2)The basic positioning function of GPS/BDS conventional RTK is realized,and the static and kinematic observation data are used to analyze the positioning performance,the results shows that,compared with the single system,the GPS/BDS dual system significantly increases the number of satellites and improves the spatial distribution of the satellites,on the challenged observation environment,such as the downtown area,the combined system can significantly improve the positioning performance;(3)Aiming at the problem that the gross error is easy to be eliminated too much or too little due to the unreasonable threshold value when conduct the residual test,one quality control method of "two steps" is proposed,namely according to the residual sum of squares,to judge whether exist gross errors in observations firstly,if exist,then the gross errors can be located based on the method of trials and errors,the effectiveness of the proposed method is verified by real vehicle data;(4)Due to the Doppler measurement has not been fully applied in the GPS kinematic positioning,an single-epoch ambiguity resolution with velocity information is proposed,the method for coordinated initialization used the velocity information gained by the Doppler measurement is introduced firstly,then a strategy for ambiguity resolution based on velocity information constraint is given.The experiment result shows that,the method improve the ambiguity fixed rate and the final positioning accuracy contrast with the conventional algorithm on GPS single-epoch kinematic positioning;(5)An improved particle swarm algorithm is proposed and apply it to the ambiguity solution(namely IPSO-AR),through the GPS/BDS data with various dimensions and accuracy of float resolution,the performance of the proposed method is deeply analyzed,the experimental results show that the efficiency of this method mainly depends on the dimensions,the success rate mainly depends on the precision of float solution,the proposed IPSO-AR method has been proved its practical application value,especially under the conditions of fewer visible satellites or constraint baseline length.(6)The main error sources of medium range RTK positioning in GPS/BDS are analyzed firstly,several commonly used methods to deal with the tropospheric and ionospheric delay errors are given,compared with the measured data,the method based on parameter estimation is proved to be the best method,and the disadvantages of other methods are also analyzed.The performance of RTK based on the random walk process is verified by the method of parameter estimation using the data of 100km length baseline and the large height difference.The experimental results show that the positioning accuracy can reach cm or even mm level under the condition of static observation;(7)With the global or regional VTEC interpolation model,it is difficult to obtain the high accuracy ionospheric delay due to the interpolation,proposed a method for accurately solving the regional receiver hardware delay(DCB)by using the single difference(SD)between stations,furthermore,the non-differential ionospheric delay of base station and rover station can be obtained by using the dual frequency pseudorange and the satellite DCB published by IGS.The SD method to solve the receiver DCB avoids the problem of the complex ionosphere model and the satellite DCB benchmark,especially suitable for the current BDS system;(8)Aiming at the problem that the conventional RTK and the medium-long RTK are in fuzzy critical state,the unified mathematical model of the two models is discussed,i.e.,the ionosphere weighted model and the inverse model of ionospheric delay based on dual frequency ambiguity,then analyzes its advantages and disadvantages as well as the problems that need further breakthrough.(9)The main error sources of indoor pseudolite(PL)positioning and the common processing methods are analyzed firstly,focusing on the problem of pseudolite system,such as the severe influence of multipath for pseudo-range observation,invariant constellation geometry structure and large linearization error,a method for indoor pseudolite positioning based on fixed point static initialization(KPI)is introduced,the LAMBDA method is used in ambiguity resolution,and the iterative extended Kalman filter(EKF)is adopted for the realization of indoor pseudolite real-time kinematic positioning.The experiment results show that,the positioning accuracy of the indoor pseudolite positioning system can reaches cm-level,and the system has certain anti-interference ability;(10)Aiming at the problem that the KPI based on LAMBDA method is very dependent on the accuracy of initial coordinates,a new method for solving the single-epoch and nonlinear AR with the indoor PL positioning is proposed,the method is based on the ambiguity function method(AFM),and an improved particle swarm optimization algorithm is adopted to improve the searching efficiency,and has more elaborate search ability as compared with the conventional grid searching method for the multi-peak characteristics of AFM,the corresponding strategies to improve its reliability are given,the experimental results show that when the initial coordinates are better than 0.2m,this method can provide high computational efficiency(about 20ms)and the reliability of indoor cm level vector epoch-by-epoch kinematic positioning.(11)Based on the above research contents and VS2010 platform,developing the corresponding data process:ing software system,which can achieve GPS/BDS conventional RTK,medium-long RTK and indoor pseudolite(PL)high precision RTK positioning function.