Filter for BDS (beidou navigation satellite system) and SINS (strapdown inertial navigation systems) navigation and positioning system and filtering method

Abstract

The invention discloses a filter for a BDS (beidou navigation satellite system) and SINS (strapdown inertial navigation systems) navigation and positioning system and a filtering method, and relates to the technical field of navigation and positioning. The method comprises the steps of building a first Gaussian mixture model formed by a plurality of Gaussian components according to the system state noise distribution condition and the posterior probability density of the navigation output result at the former moment; performing time updating on each Gaussian component in the first Gaussian mixture model according to the unscented kalman filtering, and obtaining the time predicating result corresponding to each Gaussian component; building a second Gaussian mixture model formed by a plurality of Gaussian components according to the measuring noise at the current moment; performing measurement updating according to each time predicating result according to the second Gaussian mixture model to obtain a plurality of measurement updating results; merging the plurality of measurement results to generate the navigation output result of the current moment. After the filter is applied to the BDS / SINS vehicle-mounted combined navigation system, the navigation parameter precision is ensured; meanwhile, the calculation rate is high; the work real-time performance is better.

Description

technical field [0001] The invention relates to the technical field of navigation and positioning, in particular to a filter and a filtering method for BDS and SINS navigation and positioning systems. Background technique [0002] At present, navigation and positioning systems should be widely used in various industries, such as vehicle navigation and positioning systems, ship-borne navigation and positioning systems and UAV navigation and positioning systems. Among them, the BeiDou Navigation Satellite System (BDS) independently developed by China has developed rapidly and can be combined with the Strapdown Inertial Navigation Systems (SINS) to form complementary advantages, low cost, and medium precision. DBS / SINS navigation and positioning system. In the DBS / SINS navigation and positioning system, a filter needs to be set to receive the navigation information sent by DBS and SINS, and carry out information fusion, and finally, output the result of information fusion. ...

AI Technical Summary

Problems solved by technology

Some studies have shown that any non-Gaussian distribution can be approximated by a finite mixed Gaussian model. Therefore, a mixed Gaussian Kalman filter algorithm is proposed. Although this algorithm is more reasonable for estimating the system noise covariance characteristics, there is a problem of model divergence in the estimation of nonlinear systems. disadvantages

[0004] There is an idea to propose an unscented Kalman filter algorithm (GM-UKF) based on a mixed Gaussian model (GuassianMixture, GM), but because the filtering algorithm will cause the number of Gaussian components to grow exponentially during the navigation time, resulting in long-term navigation and positioning The sharp increase in computing costs during work seriously limits the real-time performance of navigation result output and affects its application effect

Images