CN106885577A - Lagrangian aeronautical satellite autonomous orbit determination method - Google Patents

Abstract

It is as follows the invention discloses a kind of Lagrangian aeronautical satellite autonomous orbit determination method, including step：Three groups of H_2O maser information are obtained by including at least four navigation constellations of satellite；Using H_2O maser information updating neural network weight；NONLINEAR PERTURBATION is estimated according to above-mentioned neural network weight；Using NONLINEAR PERTURBATION obtained above constructing neural network state observer, the orbit information of Lagrangian satellite is estimated.The present invention is based on elliptic restricted three body problem, the perturbative force suffered by Lagrangian aeronautical satellite is accurately estimated by neutral net, improve the model accuracy of orbit determination, utilization state observer is accurately estimated the state of Lagrangian aeronautical satellite, any limitation is not done to system noise and observation noise, with preferable versatility.

Description

Autonomous orbit determination method for Lagrange navigation satellite

Technical Field

The invention belongs to the technical field of positioning navigation and control, and particularly relates to an autonomous orbit determination method for a Lagrange navigation satellite based on a neural network state observer.

Background

The deep space exploration is a research hotspot in the field of aerospace at present, and because the deep space exploration is far away from the earth, the requirements of the deep space exploration on navigation instantaneity and high precision are difficult to meet by means of a navigation mode of a ground station. The special dynamic property of the earth-moon system Lagrange point determines that the navigation satellite constellation is arranged at the Lagrange point to provide powerful navigation support for deep space exploration. The Lagrange navigation satellite constellation provides accurate navigation information on the premise that the Lagrange navigation satellite can realize accurate orbit determination.

At present, the research on the autonomous orbit determination technology of the Lagrange navigation satellite is mainly based on a circular restrictive trisomy problem, and the estimation of the Lagrange navigation satellite orbit is realized by combining a filtering algorithm. The round restrictive trisomy problem is an approximate model, and completely ignores the perturbation influence generated by the eccentricity of the moon around the earth orbit and the gravity of large planets such as the sun on Lagrange navigation satellites. The simplification of the dynamic model necessarily affects the autonomous orbit determination precision of the Lagrange navigation satellite. In addition, the filtering algorithm adopted at present assumes the types of system noise and observation noise, and also limits the application range of the filtering algorithm.

Disclosure of Invention

In view of the above problems, the present invention aims to provide an autonomous orbit determination method for a lagrangian navigation satellite, which performs real-time estimation on the state of the lagrangian navigation satellite by using a state observer by improving the accuracy of a dynamic model of the lagrangian navigation satellite, thereby implementing high-accuracy autonomous orbit determination of the lagrangian navigation satellite.

In order to achieve the purpose, the autonomous orbit determination method for the Lagrange navigation satellite comprises the following steps:

obtaining three groups of inter-satellite ranging information through a navigation constellation at least comprising four satellites;

updating the weight of the neural network by using the inter-satellite ranging information;

estimating a nonlinear perturbation item according to the weight of the neural network;

and constructing a neural network state observer by using the obtained nonlinear perturbation term, and estimating the orbit information of the Lagrange satellite.

Preferably, the neural network weight estimation updating law is designed as follows:

in the formula,for a known bounded basis vector to be,in order to observe the residual error,to estimate the state, σ is a correction coefficient.

Preferably, the observer is designed as follows:

where K is a user-defined gain matrix, v (f) is a robust term,for the estimated vector of the non-linear perturbation term, the calculation is as follows:

in the formula,is an estimate of the weights of the neural network,for a known bounded basis vector to be,d and_maxis a positive scalar.

The invention has the beneficial effects that:

according to the method, the autonomous orbit determination of the Lagrange navigation satellite is realized by designing the neural network state observer, all the perturbation forces borne by the Lagrange navigation satellite are approximated by the neural network, the model precision of the autonomous orbit determination is improved, the state of the Lagrange navigation satellite is accurately estimated by the state observer, no limitation is imposed on system noise and observation noise, and the method has good universality. The existing research on the autonomous orbit determination technology of the Lagrange navigation satellite is mainly based on the circular restrictive trisomy problem, and completely ignores the eccentricity of the orbit of the moon around the earth and the perturbation influence of large planets such as the sun on the gravitation of the Lagrange navigation satellite.

According to the method, the state of the Lagrange navigation satellite is directly estimated by the neural network state observer only by using the inter-satellite ranging information, the measuring means is simple, and the orbit determination precision is high.

Drawings

Fig. 1a is a schematic diagram of an orbit determination error curve X-axis of a satellite 1.

FIG. 1b is a Y-axis diagram of an orbit determination error curve of the satellite 1.

FIG. 1c is a Z-axis diagram of an orbit determination error curve of the satellite 1.

Fig. 2a is a schematic diagram of an X-axis orbit determination error curve of the satellite 2.

Fig. 2b is a schematic diagram of the orbit determination error curve Y axis of the satellite 2.

Fig. 2c is a Z-axis diagram of the orbit determination error curve of the satellite 2.

Fig. 3a is a schematic diagram of the estimated X-axis of the perturbation acceleration of the satellite 1.

Fig. 3b is a schematic diagram of the perturbation acceleration estimation Y-axis of the satellite 1.

Fig. 3c is a schematic diagram of the estimated Z-axis of the perturbation acceleration of the satellite 1.

Fig. 4a is a schematic diagram of the satellite 2 perturbation acceleration estimation X-axis.

Fig. 4b is a schematic diagram of the perturbation acceleration estimation Y-axis of the satellite 2.

Fig. 4c is a schematic diagram of the satellite 2 perturbation acceleration estimation Z-axis.

Fig. 5 is a flow chart of the orbit determination method.

Detailed Description

In order to facilitate understanding of those skilled in the art, the present invention will be further described with reference to the following examples and drawings, which are not intended to limit the present invention.

Referring to fig. 5, the autonomous orbit determination method for the lagrangian navigation satellite of the present invention includes the following steps:

obtaining three groups of inter-satellite ranging information through a navigation constellation at least comprising four satellites;

updating the weight of the neural network by using the inter-satellite ranging information;

estimating a nonlinear perturbation item according to the weight of the neural network;

and constructing a neural network state observer by using the obtained nonlinear perturbation term, and estimating the orbit information of the Lagrange satellite.

In the embodiment, a dynamic model is established based on an elliptic restrictive trisomy problem and a perturbation term is added; spacecraft in L under elliptical restrictive three-body problem model₁Or L₂The linearized kinetic equation in the center-meeting coordinate system is as follows:

defining a new state vector

Equation (1) can be written as follows:

wherein,

the spacecraft is affected by other perturbations in addition to the gravitational forces from the two main celestial bodies, and when these perturbations are taken into account, equation (1) will become the following form:

wherein,

represents the non-linear perturbation acceleration in the directions of three coordinate axes, and

in this embodiment, the observed quantity is ranging information between satellites, and the relationship between the observed quantity and the state variable is:

wherein [ x y z]^TAnd [ x ]₂y₂z₂]^TSatellite 1 and satellite 2 at L, respectively₁Or L₂The centers meet the coordinates under the coordinate system.

Equation (9) performs taylor series expansion near the estimated state, and ignores the high-order terms, a linear relation between the observed quantity and the state can be obtained:

defining the state estimation error as:

defining the observation residuals as:

then, the following results were obtained:

wherein,

usually only estimates of the position state of the satellite 2 are available, so the matrix C is often calculated by:

in the invention, three groups of ranging information are needed to realize the orbit determination of the Lagrange satellite, namely a navigation constellation needs to comprise four satellites; the C matrix is now re-expressed as:

wherein,andthe estimated positions of satellite 3 and satellite 4, respectively.

In order to estimate the state of the lagrangian satellite using only inter-satellite range observations, the observer is designed in the form:

where K is a user-defined gain matrix, v (f) is a robust term,for the estimated vector of the non-linear perturbation term, the calculation is as follows:

in the formula,is an estimate of the weights of the neural network,for a known bounded basis vector to be,d and_maxis a positive scalar.

In order to ensure the estimation error to be stable, the neural network weight estimation updating law is designed as follows:

in the formula,for a known bounded basis vector to be,in order to observe the residual error,to estimate the state, σ is a correction coefficient.

Fig. 1a to 1c and fig. 2a to 2c are orbit determination position error graphs of the lagrangian satellite 1 and the satellite 2 in the experiment respectively, and it can be known from the graphs that the neural network state observer can well realize that the lagrangian satellite can perform autonomous orbit determination only by using inter-satellite distance measurement.

3a-3c and 4a-4c are perturbation acceleration estimation of the satellite 1 and the satellite 2 in each coordinate axis in the experiment respectively, and it can be seen that the perturbation force can be well estimated by the neural network.

While the invention has been described in terms of its preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims (3)

1. An autonomous orbit determination method for a Lagrange navigation satellite is characterized by comprising the following steps:

obtaining three groups of inter-satellite ranging information through a navigation constellation at least comprising four satellites;

updating the weight of the neural network by using the inter-satellite ranging information;

estimating a nonlinear perturbation item according to the weight of the neural network;

and constructing a neural network state observer by using the obtained nonlinear perturbation term, and estimating the orbit information of the Lagrange satellite.

2. The autonomous orbit determination method of the lagrangian navigation satellite according to claim 1, wherein the neural network weight estimation update law is designed as:

$\stackrel{\·}{\hat{W}}=F\left(\mathrm{\φ}\right(\widehat{\stackrel{\‾}{X}}){\tilde{Y}}^T-\mathrm{\σ}\hat{W})$

in the formula,for a known bounded basis vector to be,in order to observe the residual error,to estimate the state, σ is a correction coefficient.

3. The autonomous orbit determination method for Lagrangian navigation satellites according to claim 1, characterized in that the observer is designed as follows:

$\stackrel{\·}{\widehat{\stackrel{\‾}{X}}}=A\widehat{\stackrel{\‾}{X}}+B\[\hat{g}\left(\widehat{\stackrel{\‾}{X}}\right)-v\left(f\right)\]+K(\stackrel{\‾}{Y}-\widehat{\stackrel{\‾}{Y}})$

where K is a user-defined gain matrix, v (f) is a robust term,for the estimated vector of the non-linear perturbation term, the calculation is as follows:

$\hat{g}\left(\widehat{\stackrel{\‾}{X}}\right)={\hat{W}}^T\mathrm{\φ}\left(\widehat{\stackrel{\‾}{X}}\right)$

$v\left(f\right)=-D\frac{\tilde{Y}}{\left|\right|\tilde{Y}\left|\right|}-{\mathrm{\ϵ}}_{max}\tilde{Y}$

in the formula,is an estimate of the weights of the neural network,for a known bounded basis vector to be,d and_maxis a positive scalar.