Research On UWB Ranging Error Compensation And Positioning Method

With the development of economy,location-based service plays a more and more important role in people's life.In the indoor positioning technology,UWB has the advantages of high positioning accuracy,strong anti multipath ability,fast and safe information transmission,and is one of the best methods to realize high-precision indoor positioning.In this paper,the UWB positioning method is deeply studied,mainly including the following contents:1.Aiming at the problem that the number of UWB base stations is limited,a threedimensional UWB positioning algorithm is proposed.The measured experiments show that the positioning accuracy of this method on X and Y reaches decimeter level,which can meet the positioning requirements under emergency conditions.2.Aiming at the systematic ranging error of UWB due to system hardware,clock drift and other problems,the polynomial fitting method based on least square and exponential function fitting method are used to fit the absolute error and relative error of ranging.The experimental results show that the correction of the established absolute error model reduces the ranging error by about 70.98%,The ranging error corrected by the relative error model is reduced by about 73.37%,which effectively improves the ranging accuracy.Aiming at the non line of sight error,a method of identifying the non line of sight observation value by using BP neural network to classify the ranging environment based on the channel eigenvalue is proposed.The experimental results show that the recognition accuracy of the method for the non line of sight observation value is as high as 0.911,and the recognition rate for the unobstructed ranging value is as high as 0.906.3.A Chan REKF algorithm is proposed,that is,the robust Kalman filter algorithm is used as the core algorithm of UWB position estimation,and the Chan algorithm is used to provide the initial value for the robust Kalman filter algorithm to accelerate its convergence speed.The measured experiments show that when the ranging value is disturbed,the positioning accuracy of Chan REKF is improved by 29.3% in the X direction,63.7% in the Y direction and 76.7% in the Z direction compared with the extended Kalman filter algorithm,Compared with the extended Kalman filter algorithm,the trajectory obtained by Chan REKF is smoother and less affected by gross errors.4.A GNSS / UWB pseudo range combined positioning method based on robust Kalman filter is proposed to enhance the GNSS positioning results.Compared with only using GNSS system for solution,the combined system can effectively improve the positioning accuracy and the reliability of positioning results,and weaken the influence of various errors on the positioning results.For indoor and outdoor transition areas with complex environment,GNSS / UWB positioning can better complete the positioning task than using GNSS only.