Research And Implementation Of High Dynamic Beidou Positioning Algorithm Based On Interactive Multi-model

With the gradual improvement of the Beidou III satellite system,the application scope of Beidou navigation satellite is becoming more and more extensive,and its role in military and civil fields is also becoming increasingly prominent.High dynamic Beidou receiver has also attracted much attention due to its military applications.For the receiver,navigation and positioning solution is the last step to realize positioning and velocity measurement,and also the most critical step.The Beidou satellite positioning in the high dynamic scenario has put forward more stringent requirements on the performance of the receiver equipment,as well as more severe challenges to the positioning and velocity measurement algorithms.Under the premise of high dynamic maneuver,the carrier will cause large positioning and velocity measurement error,especially the increase of velocity measurement error.However,in the actual movement,any carrier will not always maintain a high maneuvering state.It is of great significance to improve the positioning performance of non-maneuvering state while improving the positioning accuracy of high dynamic carrier and velocity measurement.This paper mainly studies the Beidou satellite positioning algorithm with high mobile motion state aircraft,and mainly improves and optimizes the satellite positioning algorithm from two aspects: model and filtering algorithm.The specific work of the paper is as follows:First of all,analyses the working principle and working process of the receiver,through the analysis of the principle of the pseudorange positioning algorithm,combed the navigation calculating process,and the paper analyzed the Beidou observation and navigation file,mainly according to the principle of pseudorange positioning and doppler velocity principle the positioning equations and velocity equations is established in this paper,and least squares algorithm,and through the experimental analysis and positioning speed error correction methods.By analyzing the problems existing in the positioning solution method under the high dynamic environment,the overall scheme design of the paper is completed,which provides the theoretical basis for the implementation of the subsequent scheme.Secondly,according to the design of three single model positioning solutions in the overall scheme of the paper,several models of maneuvering target tracking theory are introduced.The Beidou positioning experiment design of three schemes was completed by using Beidou observation and navigation documents combined with model parameter setting.In the three single model schemes,the maximum positioning error can be reduced from5.5667 m calculated by CV model to 1.1701 m calculated by CS model,and the corresponding RMS value can be reduced from 5.2203 to 0.1641.The maximum error of velocity measurement can be reduced from 3.6899m/s calculated by CV model to 2.6102m/s calculated by CS model,and the corresponding RMS value decreases from 2.4748 to 0.0104.According to the motion characteristics of high dynamic aircraft,such as hovering,uniform cruising and high dynamic maneuvering,the positioning and velocity measurement performance of three single model positioning schemes under three motion forms were analyzed through experiments,and the motion model more suitable for high dynamic and low dynamic scenes was selected respectively.Provide theoretical and data support for the implementation of the follow-up program.Thirdly,according to the design of nonlinear filtering scheme in the overall scheme,the nonlinear extended Kalman filtering and traceless Kalman filtering are briefly introduced and analyzed,and then a neural network optimized traceless Kalman filtering algorithm is proposed combining with the characteristics of high maneuvering motion state.The performance of three nonlinear filtering schemes is analyzed by experiments.A more suitable high dynamic filter is selected for the implementation of the following scheme.Finally,according to the design of multi-model scheme in the overall scheme,the interactive multi-model algorithm is briefly introduced and analyzed,and the neural network optimization of interactive multi-model satellite positioning navigation algorithm is completed by combining the model selected by single model scheme and the filtering algorithm selected by nonlinear filtering scheme.According to the motion characteristics of high dynamic aircraft,the simulation experiment data is designed with the aid of satellite signal simulator.The feasibility of the proposed algorithm is demonstrated through the experimental analysis of multi-model positioning scheme.