Hybrid processing method for DVL (Doppler velocity log) failures in integrated navigation

Abstract

The invention discloses a hybrid processing method for DVL (Doppler velocity log) failures in integrated navigation. The hybrid processing method comprises the following steps: when DVL is valid, collecting calculation information of an SINS (strapdown inertial navigation system) and measurement information of DVL to form a data table, using partial least squares regression to establish a linear prediction model, then subtracting the measurement information of DVL and a result obtained by prediction of the least squares regression model to obtain a residual part, and taking the residual part as a training target and using support vector regression training to obtain a corresponding prediction model; when the DVL fails, using the established partial least square regression model and the support vector regression model to respectively predict a measuring linear part of DVL and the residual part, and taking the sum of the measuring linear part of DVL and the residual part as predicted measurement information of DVL to ensure the reliability of SINS / DVL integrated navigation results under intermittent failures of DVL. The hybrid processing method for the DVL failures in the integrated navigation uses the partial least squares regression and support vector regression to establish the models, and uses double-model hybrid prediction to effectively improve the accuracy of prediction results.

Description

technical field [0001] The invention relates to the field of integrated navigation, in particular to a method suitable for dealing with intermittent failure of a Doppler Velocity Log (DVL) in integrated navigation. Background technique [0002] Strapdown Inertial Navigation System (SINS) is widely used for its advantages of high reliability and good concealment. However, due to the inherent errors of gyroscopes and accelerometers, the error of SINS solution results will accumulate over time, so it is necessary to introduce an auxiliary navigation system to form an integrated navigation. For aircraft such as Unmanned Surface Vehicle (USV) and Autonomous Underwater Vehicle (AUV), the introduction of Doppler Velocity Log (DVL) to form SINS / DVL integrated navigation is a A common way of navigation. [0003] DVL uses the Doppler frequency shift of reflected echo waves to measure the absolute velocity of a surface or underwater vehicle relative to the bottom or relative to the c...

AI Technical Summary

Problems solved by technology

[0004] At present, the existing integrated navigation system sensor failure processing methods mostly use a single prediction model to estimate the measurement information that the sensor should have at the time of failure or directly estimate the navigation error.

Due to the existence of model errors, the accuracy of the estimation results of the single prediction model tends to decrease with the increase of sensor failure time

Images