| This thesis takes the GNSS baseline solution under the condition of medium and long baseline as the research topic,and mainly analyzes,discusses and systematically studies the factors affecting the baseline solution results,such as random model,ambiguity solution,ambiguity fixation,ionospheric and tropospheric errors.Through theoretical research and derivation,combined with rtklib open source software platform,the baseline calculation function is realized by C programming language,and the experimental results of the algorithm program are analyzed combined with the example of medium and long baseline.In summary,the main research contents of this thesis are as follows:(1)This thesis systematically expounds the mathematical models of relative positioning,such as non difference,single difference,double difference and so on.The extended Kalman filter model that needs to be applied in data solution is deduced,which lays a theoretical foundation for this thesis.Through the comparison of multiple random models,the most suitable random model is selected,and the random model is tested by fixing the ambiguity,accuracy and probability.(2)Through the comparison of multiple random models,the most suitable random model is selected,and the random model is tested by fixing the ambiguity,accuracy and probability.Through the solution of four different medium and long baselines,it is concluded that Helmert posterior variance model is better than height angle model with the increase of baseline.(3)The processing results of different ionospheric and tropospheric models under medium and long baseline are compared.Through the calculation of actual data,it is obtained that the deionization random model is better than other models with the increase of baseline length.(4)By comparing the fixed rate and degree of fuzziness under different random models,the fuzziness under the optimal solution of ionospheric model and tropospheric model is solved,and the final positioning residual is compared.(5)Based on the rtklib software platform,the program is modified according to different random models.Taking the measured data of Hong Kong CORS network,the Countermeasures for the selection of random model and ionospheric tropospheric delay under the condition of long baseline are obtained.Finally,through the analysis of Hong Kong CORS network data,we get a good ambiguity fixing rate and the final three-dimensional coordinate offset.The medium and long baseline solution of high-precision relative positioning is realized,and the correctness and feasibility of random model,ionospheric model,tropospheric model and ambiguity fixing method are verified. |
