CN-116429100-B - Low-rail navigation signal positioning method based on factor graph method

Abstract

The invention designs a low-orbit navigation signal positioning method based on a factor graph method, and belongs to the field of low-orbit satellites for satellite navigation. Various measurement information is obtained through the low-orbit satellite and the inertial navigation system, then a factor graph model is built based on the low-orbit satellite and the inertial navigation system, finally constraint and factor nodes are set, a cost function is confirmed, and when the cost function takes the minimum value, partial derivatives are obtained on the state quantity to obtain the estimation of the state quantity. According to the method, an inertial navigation system is introduced, and a factor graph method is used for replacing a least square method to finish the positioning of the low-rail navigation signal, so that the complexity of the original positioning method is reduced. The method has important research significance and application value for the field of low-orbit satellites.

Inventors

• HAO SHUO
• ZHAO JINGBO
• WEI BAOGUO
• YI QINGWU
• LIU LIANG
• YANG JIANLEI
• Hui Shenying

Assignees

• 中国电子科技集团公司第五十四研究所

Dates

Publication Date

20260505

Application Date

20230412

Claims (2)

1. 1. The low-rail navigation signal positioning method based on the factor graph method is characterized by comprising the following steps of: (101) Various measurement information is obtained through a low-orbit satellite and an inertial navigation system; (102) Establishing a factor graph model based on a low-orbit satellite and an inertial navigation system, and defining a state vector of the navigation system as a variable node of the factor graph, wherein various measurement information obtained by the low-orbit satellite and the inertial navigation system is the factor node of the factor graph; (103) Setting constraint and factor nodes, confirming a cost function, and solving partial derivatives of state quantity when the cost function takes the minimum value to obtain state quantity estimation, wherein the specific process comprises the following steps: The constraint is set to: Wherein, the To be positioned at On the joint distribution function, a factor node is written as Representing the difference value between the predicted measurement information and the actual measurement information obtained by the factor node, constructing a corresponding index function to obtain an estimate of the state variable, As a result of the actual measurement value, As a state quantity of the system, , Is an error vector, a factor node Representing a local function in factorization; N represents a collection of the number of all factor nodes; h is a transformation relation matrix from the estimated quantity to the measured data, and the measured value of the sensor is predicted according to given state estimation in a navigation framework; At the cost function When taking the minimum value, the system state quantity Obtaining the partial derivative to obtain the state quantity of the system Is determined by the estimation of (a); At the cost function Taking the minimum value: wherein W is a priori weighting matrix, and the W pairs are used according to different user requirements Weighting each state quantity, and taking the partial derivative of the state quantity to be 0: Thereby obtaining a current state Is estimated as: ; And (5) completing the low-rail navigation signal positioning based on the factor graph method.
2. 2. The method for positioning a low-rail navigation signal based on the factor graph method of claim 1, wherein the bipartite graph model of the factor graph is: G=(F,X,E) Wherein, variable node Representing variables in global multiple functions, factor nodes Representing local functions, edges in factorization Representing if and only if a state variable node in a factor graph And factor node corresponding thereto When the two are related, a connecting edge exists between the two; Wherein, the ; For the purpose of the attitude angle error, In order to be a speed error, In the event of a position error, For the constant value drift of the gyro, Is offset for angular velocity meter constant.

Description

Low-rail navigation signal positioning method based on factor graph method Technical Field The invention relates to a low-orbit navigation signal positioning method based on a factor graph method, and belongs to the field of satellite navigation low-orbit satellites. Background GNSS navigation currently faces some problems. The navigation satellites in the GNSS are middle earth orbit satellites with orbit heights of 2-3 ten thousands of kilometers. The satellite transmitting signals reach the ground through space loss and have relatively large attenuation. And the high orbital heights result in slow changes in satellite geometry, which results in GNSS failing to meet some of the requirements for fast positioning. Meanwhile, the transmission rate of the existing GNSS navigation message is low, and the speed and instantaneity of the positioning navigation service are directly affected. The low orbit satellite orbit has low height, generally about hundreds of kilometers to two thousands of kilometers, small space loss of the transmitted signal, rapid geometric change of the satellite and quick positioning. And the low orbit satellite information transmission rate is high. The defects of GNSS can be overcome due to the two characteristics of low orbit satellite orbit height and high communication rate. Currently, the Doppler positioning principle and the least square method solution are commonly used for low-orbit satellite positioning. The least square method has larger resolving redundancy and complex computation. The factor graph is a graph model tool, can reduce the complexity of calculation, and has wide application in a plurality of fields such as statistics, artificial intelligence and the like. But less research is done with factor graph calculation using low orbit satellite positioning. Therefore, the low-orbit navigation signal positioning method based on the factor graph method has important research significance and application value. Disclosure of Invention The invention aims to overcome the complexity of least square method positioning and resolving, and provides a low-rail navigation signal positioning method based on a factor graph method. The invention is realized by the following technical scheme: A low-orbit navigation signal positioning method based on a factor graph method comprises the following steps: (101) Various measurement information is obtained through a low-orbit satellite and an inertial navigation system; (102) Establishing a factor graph model based on a low-orbit satellite and an inertial navigation system, and defining a state vector of the navigation system as a variable node of the factor graph, wherein various measurement information obtained by the low-orbit satellite and the inertial navigation system is the factor node of the factor graph; (103) Setting constraint and factor nodes, confirming a cost function, and solving partial derivatives of the state quantity when the cost function takes the minimum value to obtain the estimation of the state quantity; And (5) completing the low-rail navigation signal positioning based on the factor graph method. The bipartite graph model of the factor graph is as follows: G=(F,X,E) Wherein, the variable node X j epsilon X represents the variable in the global multi-element function, the factor node F i epsilon F represents the local function in the factorization, the edge E i,j epsilon E represents a connecting edge between the state variable node X j in the factor graph and the corresponding factor node F i if and only if the state variable node X j in the factor graph is related to the factor node F i; Wherein, the Wherein, the Is attitude angle error, δv= [ δv e δVn δVu]T is velocity error, δp= [ δlδλδh ] T is position error, b g＝[bgx bgy bgz]T is gyro constant drift, and b a＝[bax bay baz]T is angular speedometer constant bias. Compared with the prior art, the invention has the following advantages: Compared with the traditional least square method low-orbit satellite positioning solution, the low-orbit satellite positioning solution based on the factor graph only performs calculation when the measurement quantity changes, so that the redundancy of the solution and the complexity of calculation can be greatly reduced. Drawings FIG. 1 is a flow chart of an embodiment of the present invention. Detailed Description For a better description of the objects and advantages of the present invention, the technical solution of the present invention will be further described with reference to fig. 1 and the embodiment. A low-orbit navigation signal positioning method based on a factor graph method comprises the following steps: Step one, various measurement information is obtained through a low-orbit satellite and an inertial navigation system. And step two, establishing a factor graph model based on the low-orbit satellite and the inertial navigation system. The state vector of the navigation system is defined as a variable node of the fact