Research On Method And Quality Control For BDS/GPS Combined Kinematic Relative Positioning

GNSS precise relative positioning is a precise positioning technology developed in 1980s,which determines the position of user carrier in real-time or post processing modes by carrier phase measurements of two or more receivers.It has been widely applied to many areas,such as geomatics,engineering surveying,earth science studies due to its advantages of low-cost,all-climate observation,high-precision,etc.Kinematic relative positioning has wider application prospect than static relative positioning because of its not limited by stations' location and motion states,which plays an important role in mobile mapping,engineering layout,positioning and orientation of moving carrier and the collection of geographic information data.Furthermore,as the fast development of BeiDou satellite navigation system,BDS/GPS combined kinematic positioning has more advantages of visible satellite number,satellite geometrical structure,convergence speed and positioning accuracy than single GPS,so there is a great significance to study BDS/GPS combined kinematic relative positioning.Although kinematic relative positioning is a mature technology of space positioning after more than thirty years of development,there still exist some problems such as satellite number changing frequently and poor data quality because of the complicated and changeable carrier movement and observation environments,which will decrease the fixing rate,positioning accuracy and reliability.Aiming at the above problems,this thesis is intended to design and modify a series of quality control methods from three aspects including data preprocessing,parameters estimation and ambiguity resolution to improve the ambiguity fixing rate and positioning accuracy of kinematic relative positioning.The main work and contributions of this thesis are as follows:(1)The existing challenges of kinematic relative positioning are summarized systematically and the mathematical model and theoretical method of BDS/GPS combined kinematic relative positioning are introduced.The functional model and stochastic model of baselines with different length are given from original observation equations.The standard Kalman filter estimation and kinematic positioning model based on Kalman filter are introduced from the aspect of parameter estimation,and the common methods of ambiguity resolution and validation are given finally.(2)Improving TurboEdit cycle slip detection performance for kinematic data by the construction of threshold model.The method works by adding the threshold model which is adaptive to the change of observation sequence RMS to the MW combination cycle slip detection based on the moving average filter;adding weighted threshold model which is changing with sampling rate and satellite elevation angle to the GF combination cycle slip detection based on the GF difference value between adjacent epochs.A large amount of data is tested and the results show that,the combination method is effective and feasible and not only improves the detection sensitivity to some extent,but also has great improvement in avoiding misjudgments and false negatives in cycle slip detection compared with the original method.(3)Aiming at the defect of undistinguishable between gross errors and small cycle slips of traditional Kalman filter method,we propose a modified robust Kalman filter method,which considers phase measurements in rejected area of traditional Kalman filter method as cycle slips,and judge this case with both residuals and standardized residuals.Meanwhile,reinitialize the ambiguity only when phase measurement of the same satellite locates in the rejected area for at least sequential two epochs to avoid affecting convergence by the mistaken initialization.The real data tests show that the modified robust Kalman filter can avoid the influence to the positioning result caused by the undetected small cycle slips,which may also cause continuous increase of phase residuals in subsequent epochs,improve the positioning accuracy and reliability significantly.(4)For the issues that the number of satellite changes frequently in kinematic positioning which has an impact on filter estimation and ambiguity resolution,we proposed a modified adaptive Kalman filter,which takes the number of satellite continuous tracking epochs as judgment statistics,and scale the ambiguity process noise adaptively in the two aspects of increasing satellite number and decreasing satellite number.Data tests show that the modified adaptive Kalman filter can reduce the impact of emerging satellites on the filter resolution effectively,and improve the positioning performance when the number of satellite increases,in which the epoch fixing rate is increased by 30%,and the accuracy in all directions has improved a lot,especially the height component,which is improved by 35%;and avoid the impact of poor satellite geometrical structure caused by blocking to positioning and keep the stability and continuity of filter when the number of satellite decreases.(5)A partial ambiguity resolution(PAR)strategy combined with rough screening and accurate screening is proposed for the characteristics of kinematic positioning.The modified PAR method considers the influence of the number of continuous tracking epochs,elevation angle and decorrelated ambiguity variance on the ambiguity resolution,and constructs a reasonable ambiguity rejecting index,which is suitable for kinematic positioning situation.A large amount of data tests show that PAR can improve the epoch fixing rate by 15?35%,shorten the average TTFF by 31.7%from 44.1 minutes to 30.1 minutes,and increase the positioning accuracy to some extent compared with full ambiguity resolution(FAR)method.The modified PAR method improves the performance of fixing results further compared with the traditional PAR method,especially for the real kinematic data,which increases the average fixing rate by 4.5%,shortens the TTFF further by 6.3%from 30.1 minutes to 28.2 minutes,and improves the positioning to some degree.