Research On Comprehensive Performance Evaluation Method Of INS/GNSS Integrated Navigation System Based On Combined Weight

Integrated navigation is a common multi-source information fusion technology in navigation and positioning.It has important applications in military and civil fields such as aviation,aerospace,navigation and so on.In order to solve the problems of inaccurate and unreasonable evaluation of integrated navigation system in terms of reliability,availability and data accuracy,focusing on INS/GNSS integrated navigation system,this topic improves many problems in the evaluation of traditional INS/GNSS integrated navigation system,such as incomplete establishment of index system,unscientific index structure,unreasonable establishment of index weight and so on,The hierarchical structure of the index system of INS/GNSS integrated navigation system is optimized,the criterion layer and index layer under the index system are enriched and expanded,a more reasonable weight method is selected for the establishment of index weight under the multi-level index system,and the scientific evaluation of INS/GNSS integrated navigation system is realized by using the establishment method of high reliable comprehensive evaluation value,The evaluation results for the comprehensive performance of the system are obtained.The main research work includes:Firstly,in view of the problems existing in the establishment of the existing INS/GNSS integrated navigation system index system,such as unclear level division,lack of index number,lack of test on the feasibility of single index and the necessity and completeness of overall index,the INS/GNSS integrated navigation system index system is reconstructed and tested,and an improved INS/GNSS integrated navigation system evaluation index system is proposed.Secondly,aiming at the problems of different dimensions of quantitative indicators and difficult quantification of qualitative indicators in the evaluation process of INS/GNSS integrated navigation system,and aiming at the change characteristics and actual meaning of different quantitative indicator data,the nonlinear hybrid dimensionless method is studied to deal with the quantitative indicators dimensionless,reasonably eliminate the dimensions,and normalize or interval the data,This paper solves the problem of incommensurability of index data,establishes the evaluation value of qualitative index by using expert evaluation method,completes the quantitative process of qualitative index,and facilitates the subsequent establishment of index weight.Thirdly,in the process of determining the index weight,when a single weight is adopted,because the subjective weight ignores the objective significance of the data and the objective weight violates the practical importance of the index,resulting in poor authenticity and low reliability of the weight,the combined weight method combining subjective and objective weights is studied to establish the index weight and comprehensively consider the significance of the index including the landlord and objective weight,The fuzzy analytic hierarchy process is adopted to determine the subjective weight,and the entropy weight method is selected to determine the objective weight.In view of the "scaling effect" caused by the simple linear fusion coefficient in the process of subjective and objective weight fusion,a multi expert subjective weight optimization model is constructed,and the particle swarm optimization algorithm is adopted to obtain the optimal combination weight coefficient to improve the reliability of the subjective weight,Then the objective weight is used to modify the subjective weight,and the subjective and objective weights are fused to form a combined weight,so that the subjective and objective combined weight not only reflects the actual application performance represented by the index data,but also represents the practical significance of the index,and improves the rationality of the index weight.Finally,aiming at the problems of fuzzy discrimination and poor reliability in the comprehensive performance evaluation results of INS/GNSS integrated navigation system,the fuzzy comprehensive evaluation method is studied,the fuzzy comprehensive evaluation model is established,the evaluation results of INS/GNSS integrated navigation system are obtained,and then the evaluation method is analyzed and demonstrated,Using the ideal solution based on grey correlation degree,which is widely used in the evaluation of INS/GNSS integrated navigation system and has good evaluation results,and using the INS/GNSS integrated navigation system simulation data used in this subject,the evaluation results are compared with the evaluation results of this subject method to verify the superiority of this subject’s evaluation method.