| Studying the uncertainty of aircraft trajectory is of great significance for exploring aircraft conflicts and solving aircraft flow problems.A trajectory uncertainty model was constructed,which is introduces from the perspective of trajectory point uncertainty.And then,based on the above model,a trajectory data evaluation model was constructed,and a comparative study of different trajectory data was carried out.First,Three different model was introduced,which is by studying the aircraft motion of cruise phase,namely cruise dynamics model,SIMPLE motion model,and uniform motion model.And then,the performance of the three models was evaluated by using traditional analysis methods,which is to find the mean error.The results show that the predictive performance of the cruise dynamics model is superior to others.Secondly,in view of the problems in using the mean error to analysis trajectory error,such as one-sidedness,limitations and non-intuition,a new calculation method of trajectory uncertainty is proposed.Introduce the uncertainty principle in measurement,and add the calculation of trajectory uncertainty.By considering the uncertainty of each trajectory point and discusses the relation between the trajectory and trajectory point error,scientifically,and then builds a trajectory uncertainty model.Next,this model was validated with the different aircraft motion models.The results show that compared with the traditional error method,the model can evaluate the performance of aircraft model more scientifically,digitally and intuitively.Finally,based on the establishment of the trajectory uncertainty model,a trajectory data evaluation model was constructed,which is combined the Kalman filter algorithm with the Monte Carlo’s ideas.In order to verify the model,some trajectory data are used,which come from different trajectory numbers of the same radar.The results show that the track data evaluation model is correct and effective.Moreover,for the same radar,when enough data is acquired,the stability of the radar equipment can be evaluated more accurately.Next,the model is used to analyze the ADS-B data,which under different degrees of WAAS coverage.The results show that,the higher the coverage of WAAS,the better the quality of ADS-B data. |