Research On Experimental Data Fusion And Processing Of Hypersonic Vehicle

With the rapid development of hypersonic vehicles,various research institutes actively carry out flight tests for such aircraft.In this paper,based on the flight test of hypersonic vehicle,the fusion and processing of measurement data are studied with the combination of external system source data,wireless data,wired data and other test data.These measurement data may be incomplete or redundant and inconsistent,it is necessary to fuse and process the data for more accurate observation results.so that we can describe the flight test relatively accurately.In this paper,Around the real-time tracking data fusion,flight test data fusion and error detection and calibration in the fusion process,the selection and comparison of relevant algorithms are studied,and relevant conclusions are drawn.(1)Research on the design of ground plan and system modeling of flight test.The ground plan of flight test is explained from the two aspects of the overall plan and data processing flow.A three-degree-of-freedom dynamic model is established combining with the motion characteristics of hypersonic vehicle in glide phase.Taking the radar observation system as an example,the process of coordinate transformation is analyzed,and the observation system model is constructed,which lays a foundation for data fusion and processing.(2)Aiming at the real-time tracking requirement of target aircraft in flight test,combining with the practical application of Multi-observation system,the real-time tracking capability of Kalman filter and its extended form is studied and analyzed.On this basis,the centralized fusion and distributed fusion are studied.The advantages and disadvantages of each algorithm and its applicable scenarios are analyzed through simulation,which provides ideas for the research of real-time data fusion and processing in flight test.(3)Aiming at the needs of the post-flight evaluation and analysis,considering the lack of prior information in the flight test and the unknown or inaccurate motion equation in the new aircraft test,the post-flight fusion is studied.Combining the saving parameter modeling technology of flight path,the precision analysis and fusion algorithm are studied based on the application of B-spline function,and the effectiveness of the algorithm is verified by simulation.(4)Combined with the possible systematic errors and gross errors in the process of observation,the adverse effects of these errors on data fusion are analyzed,and the robust approaches are studied.The best estimation method of ballistic error model is used to calibrate the system error,and the filtering method is improved for gross error.The effectiveness of robust method is verified by simulation analysis.In a word,this paper makes a preliminary study on data fusion and processing in hypersonic vehicle flight test,and gives different fusion methods according to different requirements of real-time tracking and post-evaluation,which provides theoretical support for the application of data fusion and processing in the software platform of flight test.