Rapid Kalman online detection filtering method based on parameter dynamic adjustment

Abstract

The invention discloses a rapid Kalman online detection filtering method based on parameter dynamic adjustment. The method comprises the following steps: placing a to-be-detected object in the same detection environment for pre-sampling to obtain a plurality of groups of noisy data, and calculating a filtering data target value, a filtering data measurement variance and a Kalman gain lower limit threshold value based on the noisy data; calculating an initial estimation value of the filtering data and an initial estimation variance value of the filtering data in the online detection process ofthe detection object; wherein the initial estimation variance is determined according to an interval to which an error between an initial estimation value and a target value belongs; and performing Kalman filtering correction recursion operation based on the initial estimation value and the initial estimation variance value until the filtering data converges to obtain filtering data of online detection of the detection object. According to the method, the problems of high iteration frequency and low convergence speed caused by the fact that an initial estimation value cannot be adjusted in a traditional Kalman filtering method are solved, and the timeliness of filtering detection is improved.

Description

technical field [0001] The invention belongs to the technical field of signal filtering and processing, and in particular relates to a fast Kalman online detection and filtering method based on parameter dynamic adjustment. Background technique [0002] Kalman filtering is a method of optimally estimating the state of the system by observing the input and output data of the system by using the state equation of the linear system. At present, Kalman filter theory is widely used in communication systems, industrial control, radar signal processing, etc., and it is widely used in time series analysis in the field of automatic detection. Nowadays, the computer programming implementation of Kalman filter theory is quite mature, but the traditional implementation methods mostly emphasize the operation logic of correction recursion in Kalman filter theory, ignoring the direct impact of parameter settings on the results. [0003] At present, some adaptive Kalman filter methods (AKF...

AI Technical Summary

Problems solved by technology

Nowadays, the computer programming implementation of Kalman filter theory is quite mature, but the traditional implementation method emphasizes the operation logic of corrective recursion in Kalman filter theory, ignoring the direct impact of parameter settings on the results

[0003] At present, some adaptive Kalman filter methods (AKF) have also made related improvements to parameter settings. Although they can suppress the defect of convergence time uncertainty caused by measurement random errors to a certain extent, there is still room for optimization in terms of timeliness. Difficult to meet the timeliness requirements of online testing

Images