A self-adaptive Kalman filtering algorithm

Abstract

The invention discloses a self-adaptive Kalman filtering algorithm, which is used for removing noise. The method provided by the invention comprises the following steps: establishing a state vector model and a measurement vector model of a filtering system; After the system is initialized, taking a Kalman filtering prediction value as a sample regression fitting value, and calculating a fitting goodness determination coefficient and a correction coefficient; Correcting a process noise covariance matrix according to the correction coefficient; And according to the transmission matrix, the measurement matrix and the corrected process noise covariance matrix in the state vector model and the measurement vector model, calculating Kalman gain, and calculating a state estimation value and an estimation error covariance. Through the technical scheme provided by the invention, real-time and efficient filtering can be realized, and the filtering precision is ensured.

Description

technical field [0001] The invention relates to the field of signal processing, in particular to an adaptive Kalman filter algorithm. Background technique [0002] Kalman filtering is a set of recursive estimation algorithms that use the state space model of signal and noise to describe the system, and the estimation principle is based on the minimum mean square error. It has been widely used because it can remove noise in real time and restore real data. [0003] However, the currently commonly used Kalman filter algorithm can only rely on the accuracy of the mathematical model of the system and the integrity of the statistical characteristics of the noise to ensure accuracy. Influenced by factors, the mathematical models and noise statistical characteristics of many systems are unknown or approximately known. Using inaccurate mathematical models and noise statistical characteristics usually leads to a decrease in the accuracy of traditional Kalman filtering, or even diverg...

AI Technical Summary

Problems solved by technology

[0003] However, the currently commonly used Kalman filter algorithm can only rely on the accuracy of the mathematical model of the system and the integrity of the statistical characteristics of the noise to ensure accuracy. Influenced by factors, the mathematical models and noise statistical characteristics of many systems are unknown or approximately known. The use of inaccurate mathematical models and noise statistical characteristics usually leads to a decrease in the accuracy of traditional Kalman filtering, or even divergence, making it difficult to guarantee filtering accuracy.

Images