CN112817020A - GNSS observation data quality control and positioning method based on SVM model - Google Patents

GNSS observation data quality control and positioning method based on SVM model Download PDF

Info

Publication number
CN112817020A
CN112817020A CN201911146629.6A CN201911146629A CN112817020A CN 112817020 A CN112817020 A CN 112817020A CN 201911146629 A CN201911146629 A CN 201911146629A CN 112817020 A CN112817020 A CN 112817020A
Authority
CN
China
Prior art keywords
svm model
observation data
epoch
svm
gnss observation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911146629.6A
Other languages
Chinese (zh)
Other versions
CN112817020B (en
Inventor
王璇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chihiro Location Network Co Ltd
Original Assignee
Chihiro Location Network Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chihiro Location Network Co Ltd filed Critical Chihiro Location Network Co Ltd
Priority to CN201911146629.6A priority Critical patent/CN112817020B/en
Publication of CN112817020A publication Critical patent/CN112817020A/en
Application granted granted Critical
Publication of CN112817020B publication Critical patent/CN112817020B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
    • G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
    • G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
    • G01S19/42—Determining position
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
    • G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
    • G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
    • G01S19/40—Correcting position, velocity or attitude

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

The invention is suitable for the technical field of positioning, and provides a GNSS observation data quality control and positioning method based on an SVM model, wherein the control method comprises the following steps: collecting GNSS observation data; setting an objective function of the SVM model; constructing feature statistics based on the collected multiple epoch GNSS observation data; setting a corresponding SVM model output result according to the reliability of the pseudo-range observed quantity obtained by each epoch; using the obtained feature statistics and the SVM model output result for training SVM model parameters of an objective function of an SVM model; updating an objective function of the SVM model according to the optimized SVM model parameters to obtain an SVM vector machine; and screening new GNSS observation data in real time according to the obtained SVM vector machine. The method and the device can improve the efficiency of eliminating the gross error of the GNSS observation data.

Description

GNSS observation data quality control and positioning method based on SVM model
Technical Field
The invention relates to the technical field of Internet of things, in particular to a GNSS observation data quality control method and a GNSS observation data positioning method based on an SVM model.
Background
With the development of the internet of things technology, the positioning technology is applied to multiple industries, the positioning accuracy requirement for the consumer terminal is higher and higher at present, the observation data is the basic data of positioning, and the positioning accuracy is seriously influenced if the error of the observation data is large. In a complex urban scene, GNSS observation data are easily influenced by multipath and building shielding, and if the GNSS observation data containing gross errors cannot be effectively eliminated, a positioning result can generate large offset, and the GNSS positioning accuracy is seriously influenced.
The existing quality control method of GNSS observation data adopts elimination of gross errors, cycle slip detection and the like, and the method has the disadvantages of complex calculation process, large calculated amount, incapability of effectively and accurately eliminating the GNSS observation amount containing the gross errors and incapability of providing accurate navigation experience.
Therefore, a new technical solution is needed to solve the above technical problems.
Disclosure of Invention
In view of this, the embodiment of the invention provides a GNSS observation data quality control method and a positioning method based on an SVM model, which solve the problem of low efficiency of gross error elimination in the prior art.
The first aspect of the embodiments of the present invention provides a GNSS observation data quality control method based on an SVM model, including:
collecting GNSS observation data;
setting an objective function of the SVM model as follows:
Figure BDA0002277880340000021
wherein b and w are both SVM model parameters, xiRepresenting a feature statistic;
constructing feature statistics based on the collected GNSS observation data of a plurality of epochs, wherein the feature statistics comprise an observation signal-to-noise ratio of a single satellite in a single epoch, an observation height angle of the single satellite in the single epoch, a pseudo range observation residual of the single satellite in the single epoch, a Doppler observation residual of the single satellite in the single epoch and a pseudo range Doppler consistency parameter, and the pseudo range Doppler consistency parameter is obtained according to the pseudo range observations and the Doppler observations of the adjacent epochs of the GNSS observation data;
setting a corresponding SVM model output result according to the reliability of pseudo-range observed quantity obtained by each epoch in the GNSS observation data;
using the obtained feature statistics and the SVM model output result for training SVM model parameters of an objective function of the SVM model to obtain optimized SVM model parameters;
updating the objective function of the SVM model according to the optimized SVM model parameters to obtain an SVM vector machine;
and screening new GNSS observation data in real time according to the obtained SVM vector machine.
A second aspect of an embodiment of the present invention provides a positioning method, including:
collecting GNSS observation data;
constructing an objective function of the SVM model;
constructing feature statistics based on the collected GNSS observation data, wherein the feature statistics comprise single epoch statistics of a single satellite;
setting a corresponding SVM model output result according to the reliability of the pseudo-range observed quantity obtained by each epoch;
using the obtained feature statistics and the SVM model output result for training model parameters of a target function of the SVM model to obtain an SVM vector machine;
screening new GNSS observation data in real time according to the obtained SVM vector machine;
and performing Kalman filtering positioning calculation based on the screened GNSS observation data to obtain a corresponding positioning result.
A third aspect of an embodiment of the present invention further provides a positioning apparatus, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the method of the first aspect or the second aspect.
A fourth aspect of embodiments of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of the first or second aspect.
Compared with the prior art, the embodiment of the invention has the following beneficial effects: .
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a schematic flowchart of a GNSS observation data quality control method based on an SVM model according to an embodiment of the present invention;
fig. 2 is a schematic flowchart of a positioning method according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a positioning device according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a positioning device according to a fourth embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
As used in this specification and the appended claims, the term "if" may be interpreted contextually as "when", "upon" or "in response to a determination" or "in response to a detection". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
It should be understood that, the sequence numbers of the steps in this embodiment do not mean the execution sequence, and the execution sequence of each process should be determined by the function and the inherent logic of the process, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
It should be noted that, the descriptions of "first" and "second" in this embodiment are used to distinguish different regions, modules, and the like, and do not represent a sequential order, and the descriptions of "first" and "second" are not limited to be of different types.
In order to illustrate the technical solution of the present invention, the following is illustrated by specific examples.
Example one
Fig. 1 is a flowchart illustrating a GNSS observation data quality control method based on an SVM model according to an embodiment of the present invention, where the method includes the following steps:
step S1, collecting GNSS observation data;
specifically, first, GNSS observation data is collected, the GNSS observation data including GNSS observation data of a plurality of epochs, and the GNSS observation data of each epoch may include: pseudorange observations and doppler observations.
Step S2, setting an objective function of the SVM model;
specifically, an objective function of the SVM model is preset, and the objective function specifically includes:
Figure BDA0002277880340000051
wherein b and w are both SVM model parameters, xiRepresenting the feature statistics. The representation form of the objective function may be various, and the present invention is not limited thereto. The essence of the objective function in this embodiment is to find the support vector machine with maximized interval by the formula.
Step S3, constructing feature statistics based on the collected GNSS observation data of a plurality of epochs;
specifically, feature statistics are constructed based on the aforementioned GNSS observation data of the plurality of epochs, the feature statistics comprising: the signal-to-noise ratio (an index of signal strength, which indicates the energy intensity of a GNSS signal captured by a receiver, and the higher the energy, the better the observation quality; the signal-to-noise ratio may also be referred to as "CN 0"), the observation quantity high angle of a single satellite in a single epoch (which is used for distinguishing the quality of observation data, and is the included angle between a connecting line between the satellite and the receiver and a horizontal line, and the higher the height angle, the stronger the signal received by an antenna, and the more reliable the observation quality), the pseudorange observation quantity residual of the single satellite in the single epoch, the Doppler observation quantity residual of the single satellite in the single epoch, and the pseudorange Doppler consistency parameter, wherein the pseudorange Doppler consistency parameter is obtained according to the pseudorange observation quantity and the Doppler observation quantity of adjacent epochs of the GNSS observation data;
in a further preferred aspect of this embodiment, the feature statistic may further include: average signal-to-noise ratio of observations of all satellites for a single epoch and the number of satellites. The correlation data for all satellites of a single epoch may be obtained based on statistics for a single satellite of a single epoch.
In a further preferred aspect of this embodiment, the feature statistic may further include: a median of pseudorange observation residuals and/or a median of doppler observation residuals.
It should be noted that, the observed quantity signal-to-noise ratio of a single satellite in a single epoch, the observed quantity angle of a single satellite in a single epoch, the pseudo range observed quantity residual of a single satellite in a single epoch, the mean signal-to-noise ratio of the observed quantity of all satellites in a single epoch, the number of satellites and the median of the pseudo range observed quantity residual may be used as the feature statistics, or the observed quantity signal-to-noise ratio of a single satellite in a single epoch, the observed quantity residual of a single satellite in a single epoch, the pseudo range observed quantity residual of a single satellite in a single epoch, the mean signal-to-noise ratio of all satellites in a single epoch, the number of satellites and the median of the doppler observed quantity residual may be used as the feature statistics, or the observed quantity signal-to noise ratio of, The observation quantity height angle of a single satellite in a single epoch, the pseudo-range observation quantity residual error of the single satellite in the single epoch, the average signal-to-noise ratio and the satellite number of the observation quantity of all satellites in the single epoch, the median of the pseudo-range observation quantity residual error and the median of the Doppler observation quantity residual error are used as feature statistics and are respectively input into an SVM model for training, and different support vector machines can be obtained. Different support vector machines obtained due to different feature statistics may have different effects when screening new data.
And step S4, setting a corresponding SVM model output result according to the reliability of the pseudo-range observed quantity of each epoch in the GNSS observation data.
Specifically, a corresponding SVM model output result is set according to the reliability of the pseudo-range observed quantity of each epoch in the GNSS observation data.
And step S5, using the obtained feature statistics and the SVM model output result for training SVM model parameters of an objective function of the SVM model to obtain optimized SVM model parameters.
And step S6, updating the objective function of the SVM model according to the optimized SVM model parameters to obtain the SVM vector machine.
Specifically, the SVM vector machine is: and f (x) w x + b, wherein w and b are SVM model parameters x as feature statistics, and f represents the meaning of a function.
And step S7, screening new GNSS observation data in real time according to the obtained SVM vector machine.
Specifically, the GNSS observation data obtained currently is screened according to the current SVM vector machine so as to improve the quality of the GNSS observation data.
In this embodiment, feature statistics are constructed based on GNSS observation data, an SVM vector machine is obtained by training an SVM model according to the feature statistics, and currently acquired GNSS observation data is screened in real time according to the SVM vector machine, so that the efficiency of eliminating gross errors of the GNSS observation data can be improved.
In a preferred embodiment of this embodiment, obtaining the pseudorange doppler consistency parameter according to the pseudorange observed quantity and the doppler observed quantity of the adjacent epoch of the GNSS observation data specifically includes:
according to the formula: pr _ dr _ diff ═ pri-(pri-1+dri) Obtaining the pseudo range Doppler consistency parameter, wherein pr _ dr _ diff represents the pseudo range Doppler consistency parameter pri，pri-1Respectively representing pseudorange observations, dr, of the ith and i-1 th epochsiRepresents the doppler observation for the ith epoch. The pseudorange doppler consistency may be computed in a variety of ways, essentially representing the rate of change of the pseudorange.
In this embodiment, a pseudorange observation residual for a single epoch is computed from a pseudorange observation of GNSS observation data, as follows: pr _ diff ═ prraw-prsTo compute a pseudorange observation residual, where pr _ diff represents the pseudorange observation residual, prrawRepresenting pseudorange observations, prsRepresenting the distance between the satellite and the receiver.
In this embodiment, the doppler observation residuals for a single epoch are calculated from the doppler observations of the GNSS observations, as shown in the following formula: dr _ diff ═ drraw-drsTo calculate a Doppler observation residual, wherein dr _ diff is the Doppler observation residual, drrawRepresenting the Doppler observations, drsWhich represents the relative velocity between the satellite and the receiver (corresponding to the relative velocity of the satellite with respect to the receiver).
In a preferable scheme of this embodiment, the step S4 specifically includes:
acquiring pseudo-range observation according to acquired GNSS observation data;
specifically, pseudo-range observations are obtained according to the obtained GNSS observation data, and it should be noted that each epoch corresponds to one pseudo-range observation;
acquiring a distance reference value;
specifically, a distance reference value is obtained according to the true distance between the satellite and the receiver (the position of the satellite is calculated through ephemeris, and the true distance between the satellite and the receiver is calculated according to the known position of the receiver and the calculated position of the satellite);
and comparing the difference value of the pseudo-range observed quantity and the distance reference value, setting the output result of the corresponding SVM model as a first parameter to represent high reliability when the difference value is greater than a threshold value, and setting the output result of the corresponding SVM model as a second parameter to represent low reliability when the difference value is less than the threshold value, wherein the observation result is likely to deviate from the real distance greatly.
Specifically, the difference value between the pseudo-range observed quantity and the distance reference value is compared, when the difference value is larger than a threshold value, the output result of the corresponding SVM model is set as a first parameter to represent high reliability, and when the difference value is smaller than the threshold value, the output result of the corresponding SVM model is set as a second parameter to represent low reliability. If the difference value of the pseudo-range observed quantity and the distance reference value is compared, according to the formula:
Figure BDA0002277880340000081
to obtain confidence in pseudorange observations, where ε represents a threshold, prrefDenotes a distance reference value, prrawAnd expressing the pseudo-range observed quantity, if the difference value of the pseudo-range observed quantity and the pseudo-range observed quantity is greater than or equal to a threshold value, the output result y of the SVM model is a first parameter 1 and expresses high reliability, and if the difference value is less than the threshold value, the output result y of the SVM model is a second parameter-1 and expresses low reliability.
In a preferable scheme of this embodiment, the step S5 specifically includes:
and taking the feature statistics as a nonlinear separable training set, mapping to a high-dimensional feature space according to a Gaussian kernel function, and constructing a target optimal classification hyperplane.
In particular, the feature statistics are treated as a nonlinear separable training set, such as (x)1，y1)，...(xl，yl)，xi∈Rn，yi＝1，-1，i＝1..l，RnIs a set of real numbers used to represent feature statistics (i.e., a non-linear separable training set).
And mapping to a high-dimensional feature space according to the Gaussian kernel function, and constructing a target optimal classification hyperplane. The feature statistic is used as a nonlinear separable training set, the feature statistic is used as an input vector x, and according to a Gaussian formula:
Figure BDA0002277880340000082
mapping to a high-dimensional feature space, namely transforming to the high-dimensional feature space through nonlinear data to construct a target optimal classification hyperplane, wherein the target optimal classification hyperplane is as follows:
Figure BDA0002277880340000083
in a preferable scheme of this embodiment, the step S7 specifically includes:
acquiring an output result of the SVM vector machine according to the feature statistics of each epoch of the new GNSS observation data; and
and determining whether the GNSS observation data corresponding to the epoch is excluded according to the output result of the SVM vector machine.
Specifically, new GNSS observation data is acquired, the feature statistics of each epoch of the GNSS observation data are input into the SVM vector machine to obtain a corresponding output result, whether the GNSS observation data of the corresponding epoch need to be excluded is determined according to the output result, and if the result is 1,
Figure BDA0002277880340000091
if the output result is-1, the quality of the corresponding GNSS observation data is poor, and the GNSS observation data of the epoch needs to be excludedAnd observing data, thereby realizing the control of the quality of the GNSS observation data.
In this embodiment, feature statistics are constructed based on GNSS observation data, an SVM vector machine is obtained by training an SVM model according to the feature statistics, and currently acquired GNSS observation data is screened in real time according to the SVM vector machine, so that the efficiency of eliminating gross errors of the GNSS observation data can be improved.
And secondly, the parameters of the SVM model are optimized and set, so that the quality of the acquired GNSS observation data can be improved.
Example two
Fig. 2 is a schematic flowchart of a positioning method according to a second embodiment of the present invention, where the method includes the following steps:
step A1, collecting GNSS observation data;
specifically, first, GNSS observation data is collected, the GNSS observation data including GNSS observation data of a plurality of epochs, and the GNSS observation data of each epoch may include: pseudorange observations and doppler observations.
A2, constructing an objective function of the SVM model;
specifically, an objective function of the SVM model is constructed in advance, and the objective function specifically includes:
Figure BDA0002277880340000101
wherein b and w are both SVM model parameters, xiRepresenting the feature statistics.
Step A3, constructing feature statistics based on the collected GNSS observation data;
in particular, feature statistics are constructed based on the aforementioned multi-epoch GNSS observations, the feature statistics including single-satellite single-epoch statistics, which may include: the signal-to-noise ratio of the observed quantity of a single satellite in a single epoch (an index of signal strength, which indicates the energy intensity of a receiver capturing a GNSS signal, and the higher the energy, the better the observation quality), the height angle of the observed quantity of the single satellite in the single epoch (which is used for distinguishing the quality of observation data, and is an included angle between a connecting line between the satellite and the receiver and a horizontal line, and the higher the height angle, the stronger the signal received by an antenna, the more reliable the observation quality), the pseudo-range observed quantity residual error of the single satellite in the single epoch, and the Doppler observed quantity residual error of the single satellite in the single epoch; the feature statistics may further include a pseudorange doppler consistency parameter, wherein the pseudorange doppler consistency parameter is obtained from pseudorange observations and doppler observations of adjacent epochs of the GNSS observation data;
in a further preferred aspect of this embodiment, the feature statistic may further include: average signal-to-noise ratio of observations of all satellites for a single epoch and the number of satellites.
In a further preferred aspect of this embodiment, the feature statistic may further include: a median of pseudorange observation residuals and/or a median of doppler observation residuals.
Step A4, setting a corresponding SVM model output result according to the reliability of pseudo-range observed quantity obtained by each epoch;
specifically, a corresponding SVM model output result is set according to the reliability of the pseudo-range observed quantity obtained by each epoch.
Step A5, using the obtained feature statistics and the SVM model output result for training model parameters of an objective function of an SVM model to obtain an SVM vector machine;
specifically, firstly, the obtained feature statistics and the SVM model output result are used for training SVM model parameters of an objective function of an SVM model to obtain optimized SVM model parameters, and then the objective function of the SVM model is updated according to the optimized SVM model parameters to obtain an SVM vector machine, wherein the SVM vector machine is: and f (x) w x + b, wherein w and b are SVM model parameters x as feature statistics, and f represents the meaning of a function.
Step A6, screening new GNSS observation data in real time according to the obtained SVM vector machine;
specifically, the GNSS observation data obtained currently is screened according to the current SVM vector machine so as to improve the quality of the GNSS observation data.
And A7, performing Kalman filtering positioning calculation based on the screened GNSS observation data to obtain a corresponding positioning result.
Specifically, Kalman filtering positioning calculation is performed according to the screened GNSS observation data to obtain a corresponding calculation result, and the GNSS observation data are screened, so that the positioning calculation precision and reliability can be improved.
In this embodiment, the steps a1 to a4 correspond to the steps S1 to S4 of the first embodiment, the step a5 corresponds to the steps S5 and S6 of the first embodiment, and the step a6 corresponds to the step S7 of the first embodiment, and the specific implementation process can refer to the first embodiment and is not described herein again.
In the embodiment, feature statistics are constructed based on GNSS observation data, an SVM model is trained according to the feature statistics to obtain an SVM vector machine, and the SVM vector machine screens the GNSS observation data obtained currently in real time, so that the efficiency of eliminating gross errors of the GNSS observation data can be improved.
And secondly, GNSS observation data are screened, so that the positioning resolving precision and reliability can be improved.
The second embodiment can be adjusted based on the first embodiment, but the invention is not limited thereto.
EXAMPLE III
Fig. 3 is a schematic structural diagram of a positioning device according to a third embodiment of the present invention. As shown in fig. 3, the positioning device 3 of this embodiment includes: a processor 30, a memory 31 and a computer program 32 stored in said memory 31 and executable on said processor 30. The steps of the first embodiment of the method described above are implemented when the computer program 32 is executed by the processor 30.
The computer program 32 is specifically adapted to cause the processor 30 to perform the following operations:
collecting GNSS observation data;
setting an objective function of the SVM model as follows:
Figure BDA0002277880340000121
wherein b and w are both SVM model parameters, xiRepresenting a feature statistic;
constructing feature statistics based on the collected GNSS observation data of a plurality of epochs, wherein the feature statistics comprise an observation signal-to-noise ratio of a single satellite in a single epoch, an observation height angle of the single satellite in the single epoch, a pseudo range observation residual of the single satellite in the single epoch, a Doppler observation residual of the single satellite in the single epoch and a pseudo range Doppler consistency parameter, and the pseudo range Doppler consistency parameter is obtained according to the pseudo range observations and the Doppler observations of the adjacent epochs of the GNSS observation data;
setting a corresponding SVM model output result according to the reliability of pseudo-range observed quantity obtained by each epoch in the GNSS observation data;
using the obtained feature statistics and the SVM model output result for training SVM model parameters of an objective function of the SVM model to obtain optimized SVM model parameters;
updating the objective function of the SVM model according to the optimized SVM model parameters to obtain an SVM vector machine;
and screening new GNSS observation data in real time according to the obtained SVM vector machine.
In an optional manner, the feature statistics further include:
average signal-to-noise ratio of observations of all satellites for a single epoch and the number of satellites.
In an optional manner, the feature statistics further include:
a median of pseudorange observation residuals and/or a median of doppler observation residuals.
In an alternative approach, the computer program 32 is specifically configured to cause the processor 30 to perform the following operations:
according to pr _ dr _ diff ═ pri-(pri-1+dri) Obtaining the pseudorange Doppler consistency parameter, wherein pr _ dr _ diff represents the pseudorange Doppler consistency parameter, pri，pri-1Respectively representing pseudorange observations, dr, of the ith and i-1 th epochsiRepresents the doppler observation for the ith epoch.
In an alternative approach, the computer program 32 is specifically configured to cause the processor 30 to perform the following operations:
acquiring the pseudo-range observation according to the acquired GNSS observation data;
acquiring a distance reference value;
and comparing the difference value between the pseudo-range observed quantity and the distance reference value, setting the output result of the corresponding SVM model as a first parameter to represent high reliability when the difference value is greater than or equal to a threshold value, and setting the output result of the corresponding SVM model as a second parameter to represent low reliability when the difference value is less than the threshold value.
In an alternative approach, the computer program 32 is specifically configured to cause the processor 30 to perform the following operations:
and taking the feature statistics as a nonlinear separable training set, mapping to a high-dimensional feature space according to a Gaussian kernel function, and constructing a target optimal classification hyperplane.
In an alternative approach, the computer program 32 is specifically configured to cause the processor 30 to perform the following operations:
acquiring an output result of the SVM vector machine according to the feature statistics of each epoch of new GNSS observation data; and
and determining whether the GNSS observation data corresponding to the epoch are excluded according to the output result of the SVM vector machine.
In the embodiment, feature statistics are constructed based on GNSS observation data, an SVM model is trained according to the feature statistics to obtain an SVM vector machine, and the SVM vector machine screens the GNSS observation data obtained currently in real time, so that the efficiency of eliminating gross errors of the GNSS observation data can be improved.
Example four
Fig. 4 is a schematic structural diagram of a positioning device according to a fourth embodiment of the present invention. As shown in fig. 4, the positioning device 4 of this embodiment includes: a processor 40, a memory 41 and a computer program 42 stored in said memory 41 and executable on said processor 40. The steps of the first embodiment of the method described above are implemented when the computer program 42 is executed by the processor 40.
The computer program 42 is specifically adapted to cause the processor 40 to perform the following operations:
collecting GNSS observation data;
constructing an objective function of the SVM model;
constructing feature statistics based on the collected GNSS observation data, wherein the feature statistics comprise single epoch statistics of a single satellite;
setting a corresponding SVM model output result according to the reliability of the pseudo-range observed quantity obtained by each epoch;
using the obtained feature statistics and the SVM model output result for training model parameters of a target function of the SVM model to obtain an SVM vector machine;
screening new GNSS observation data in real time according to the obtained SVM vector machine;
and performing Kalman filtering positioning calculation based on the screened GNSS observation data to obtain a corresponding positioning result.
In an alternative approach, the single satellite single epoch statistic includes an observed quantity signal-to-noise ratio of the single satellite in the single epoch, an observed quantity height angle of the single satellite in the single epoch, a pseudorange observed quantity residual of the single satellite in the single epoch, and a doppler observed quantity residual of the single satellite in the single epoch.
In an alternative, the feature statistics further include statistics for all satellites of a single epoch, including an average signal-to-noise ratio of observations for all satellites of a single epoch and a number of satellites.
In an alternative, the statistics for all satellites of the single epoch further include a median of pseudorange observation residuals and/or a median of doppler observation residuals.
In the embodiment, feature statistics are constructed based on GNSS observation data, an SVM model is trained according to the feature statistics to obtain an SVM vector machine, and the SVM vector machine screens the GNSS observation data obtained currently in real time, so that the efficiency of eliminating gross errors of the GNSS observation data can be improved.
EXAMPLE five
A fifth embodiment of the present invention provides a computer-readable storage medium including a computer program stored on a computer storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to execute the SVM model-based GNSS observation data quality control method in the first embodiment described above.
The executable instructions may be specifically configured to cause the processor to:
collecting GNSS observation data;
setting an objective function of the SVM model as follows:
Figure BDA0002277880340000151
wherein b and w are both SVM model parameters, xiRepresenting a feature statistic;
constructing feature statistics based on the collected GNSS observation data of a plurality of epochs, wherein the feature statistics comprise an observation signal-to-noise ratio of a single satellite in a single epoch, an observation height angle of the single satellite in the single epoch, a pseudo range observation residual of the single satellite in the single epoch, a Doppler observation residual of the single satellite in the single epoch and a pseudo range Doppler consistency parameter, and the pseudo range Doppler consistency parameter is obtained according to the pseudo range observations and the Doppler observations of the adjacent epochs of the GNSS observation data;
setting a corresponding SVM model output result according to the reliability of pseudo-range observed quantity obtained by each epoch in the GNSS observation data;
using the obtained feature statistics and the SVM model output result for training SVM model parameters of an objective function of the SVM model to obtain optimized SVM model parameters;
updating the objective function of the SVM model according to the optimized SVM model parameters to obtain an SVM vector machine;
and screening new GNSS observation data in real time according to the obtained SVM vector machine.
In an optional manner, the feature statistics further include:
average signal-to-noise ratio of observations of all satellites for a single epoch and the number of satellites.
In an optional manner, the feature statistics further include:
a median of pseudorange observation residuals and/or a median of doppler observation residuals.
In an alternative, the executable instructions may be specifically configured to cause the processor to:
according to pr _ dr _ diff ═ pri-(pri-1+dri) Obtaining the pseudorange Doppler consistency parameter, wherein pr _ dr _ diff represents the pseudorange Doppler consistency parameter, pri，pri-1Respectively representing pseudorange observations, dr, of the ith and i-1 th epochsiRepresents the doppler observation for the ith epoch.
In an alternative, the executable instructions may be specifically configured to cause the processor to:
acquiring the pseudo-range observation according to the acquired GNSS observation data;
acquiring a distance reference value;
and comparing the difference value between the pseudo-range observed quantity and the distance reference value, setting the output result of the corresponding SVM model as a first parameter to represent high reliability when the difference value is greater than or equal to a threshold value, and setting the output result of the corresponding SVM model as a second parameter to represent low reliability when the difference value is less than the threshold value.
In an alternative, the executable instructions may be specifically configured to cause the processor to:
and taking the feature statistics as a nonlinear separable training set, mapping to a high-dimensional feature space according to a Gaussian kernel function, and constructing a target optimal classification hyperplane.
In an alternative, the executable instructions may be specifically configured to cause the processor to:
acquiring an output result of the SVM vector machine according to the feature statistics of each epoch of new GNSS observation data; and
and determining whether the GNSS observation data corresponding to the epoch are excluded according to the output result of the SVM vector machine.
In the embodiment, feature statistics are constructed based on GNSS observation data, an SVM model is trained according to the feature statistics to obtain an SVM vector machine, and the SVM vector machine screens the GNSS observation data obtained currently in real time, so that the efficiency of eliminating gross errors of the GNSS observation data can be improved.
EXAMPLE six
A sixth embodiment of the present invention provides a computer-readable storage medium including a computer program stored on a computer storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to perform the positioning method in the second embodiment described above.
The executable instructions may be specifically configured to cause the processor to:
collecting GNSS observation data;
constructing an objective function of the SVM model;
constructing feature statistics based on the collected GNSS observation data, wherein the feature statistics comprise single epoch statistics of a single satellite;
setting a corresponding SVM model output result according to the reliability of the pseudo-range observed quantity obtained by each epoch;
using the obtained feature statistics and the SVM model output result for training model parameters of a target function of the SVM model to obtain an SVM vector machine;
screening new GNSS observation data in real time according to the obtained SVM vector machine;
and performing Kalman filtering positioning calculation based on the screened GNSS observation data to obtain a corresponding positioning result.
In an alternative approach, the single satellite single epoch statistic includes an observed quantity signal-to-noise ratio of the single satellite in the single epoch, an observed quantity height angle of the single satellite in the single epoch, a pseudorange observed quantity residual of the single satellite in the single epoch, and a doppler observed quantity residual of the single satellite in the single epoch.
In an alternative, the feature statistics further include statistics for all satellites of a single epoch, including an average signal-to-noise ratio of observations for all satellites of a single epoch and a number of satellites.
In an alternative, the statistics for all satellites of the single epoch further include a median of pseudorange observation residuals and/or a median of doppler observation residuals.
In the invention, feature statistics are constructed based on GNSS observation data, an SVM model is trained according to the feature statistics to obtain an SVM vector machine, and the currently obtained GNSS observation data is screened in real time by the SVM vector machine, so that the efficiency of eliminating gross errors of the GNSS observation data can be improved.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that has been appropriately increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media may not include electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (13)

1. A GNSS observation data quality control method based on SVM model is characterized by comprising the following steps:
collecting GNSS observation data;
setting an objective function of the SVM model as follows:
Figure FDA0002277880330000011
wherein b and w are both SVM model parameters, xiRepresenting a feature statistic;
constructing feature statistics based on the collected GNSS observation data of a plurality of epochs, wherein the feature statistics comprise an observation signal-to-noise ratio of a single satellite in a single epoch, an observation height angle of the single satellite in the single epoch, a pseudo range observation residual of the single satellite in the single epoch, a Doppler observation residual of the single satellite in the single epoch and a pseudo range Doppler consistency parameter, and the pseudo range Doppler consistency parameter is obtained according to the pseudo range observations and the Doppler observations of the adjacent epochs of the GNSS observation data;
setting a corresponding SVM model output result according to the credibility of the pseudo-range observed quantity of each epoch in the GNSS observation data;
using the obtained feature statistics and the SVM model output result for training SVM model parameters of an objective function of the SVM model to obtain optimized SVM model parameters;
updating the objective function of the SVM model according to the optimized SVM model parameters to obtain an SVM vector machine;
and screening new GNSS observation data in real time according to the obtained SVM vector machine.
2. The method of claim 1, wherein the feature statistics further comprise:
average signal-to-noise ratio of observations of all satellites for a single epoch and the number of satellites.
3. The method of claim 2, wherein the feature statistics further comprise:
a median of pseudorange observation residuals and/or a median of doppler observation residuals.
4. The method of claim 1, wherein obtaining the pseudorange doppler consistency parameter from pseudorange observations and doppler observations of adjacent epochs of the GNSS observation data comprises:
according to pr _ dr _ diff ═ pri-(pri-1+dri) Obtaining the pseudorange Doppler consistency parameter, wherein pr _ dr _ diff represents the pseudorange Doppler consistency parameter, prii，pri-1Respectively representing pseudorange observations, dr, of the ith and i-1 th epochsiRepresents the doppler observation for the ith epoch.
5. The method of claim 1, wherein setting the corresponding SVM model output based on the confidence level of the pseudorange observations obtained for each epoch comprises:
acquiring the pseudo-range observation according to the acquired GNSS observation data;
acquiring a distance reference value;
and comparing the difference value between the pseudo-range observed quantity and the distance reference value, setting the output result of the corresponding SVM model as a first parameter to represent high reliability when the difference value is greater than or equal to a threshold value, and setting the output result of the corresponding SVM model as a second parameter to represent low reliability when the difference value is less than the threshold value.
6. The method of claim 1, wherein using the obtained feature statistics and the SVM model output results for training of SVM model parameters of an objective function of the SVM model comprises:
and taking the feature statistics as a nonlinear separable training set, mapping to a high-dimensional feature space according to a Gaussian kernel function, and constructing a target optimal classification hyperplane.
7. The method of claim 1, wherein the real-time filtering of new GNSS observations from the obtained SVM vector machine comprises:
acquiring an output result of the SVM vector machine according to the feature statistics of each epoch of new GNSS observation data; and
and determining whether the GNSS observation data corresponding to the epoch are excluded according to the output result of the SVM vector machine.
8. A method of positioning, comprising:
collecting GNSS observation data;
constructing an objective function of the SVM model;
constructing feature statistics based on the collected GNSS observation data, wherein the feature statistics comprise single epoch statistics of a single satellite;
setting a corresponding SVM model output result according to the reliability of the pseudo-range observed quantity obtained by each epoch;
using the obtained feature statistics and the SVM model output result for training model parameters of a target function of the SVM model to obtain an SVM vector machine;
screening new GNSS observation data in real time according to the obtained SVM vector machine;
and performing Kalman filtering positioning calculation based on the screened GNSS observation data to obtain a corresponding positioning result.
9. The method of claim 8, wherein the single satellite single epoch statistics comprise an observed volume signal-to-noise ratio of the single satellite in the single epoch, an observed volume height angle of the single satellite in the single epoch, a pseudorange observed volume residual of the single satellite in the single epoch, and a doppler observed volume residual of the single satellite in the single epoch.
10. The method of claim 8, wherein the feature statistics further comprise statistics for all satellites of a single epoch, including average signal-to-noise ratio of observations for all satellites of a single epoch and number of satellites.
11. The method of claim 10, wherein the statistics of all satellites for the single epoch further include a median of pseudorange observation residuals and/or a median of doppler observation residuals.
12. A positioning device for carrying out the method of any one of claims 1 to 7 or 8 to 11.
13. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7 or 8 to 11.
CN201911146629.6A 2019-11-18 2019-11-18 GNSS observation data quality control and positioning method based on SVM model Active CN112817020B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911146629.6A CN112817020B (en) 2019-11-18 2019-11-18 GNSS observation data quality control and positioning method based on SVM model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911146629.6A CN112817020B (en) 2019-11-18 2019-11-18 GNSS observation data quality control and positioning method based on SVM model

Publications (2)

Publication Number Publication Date
CN112817020A true CN112817020A (en) 2021-05-18
CN112817020B CN112817020B (en) 2023-01-10

Family

ID=75852915

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911146629.6A Active CN112817020B (en) 2019-11-18 2019-11-18 GNSS observation data quality control and positioning method based on SVM model

Country Status (1)

Country Link
CN (1) CN112817020B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113281796A (en) * 2021-07-23 2021-08-20 腾讯科技（深圳）有限公司 Position determining method, speed determining method, device, equipment and storage medium
CN113642252A (en) * 2021-08-31 2021-11-12 中国人民解放军国防科技大学 Target positioning method and device based on single satellite, electronic equipment and medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170131408A1 (en) * 2014-03-28 2017-05-11 Mitsubishi Electric Corporation Positioning device
CN109993191A (en) * 2018-01-02 2019-07-09 ***通信有限公司研究院 Information processing method and device, electronic equipment and storage medium

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170131408A1 (en) * 2014-03-28 2017-05-11 Mitsubishi Electric Corporation Positioning device
CN109993191A (en) * 2018-01-02 2019-07-09 ***通信有限公司研究院 Information processing method and device, electronic equipment and storage medium

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
TAO TINGYE ET AL.: "Gross error detection and correction based on wavelet transform and support vector machine", 《SPIE》 *
张闯 等: "LS-SVM 辅助的小幅值及缓变故障检测与容错方法", 《中国惯性技术学报》 *
李飞等: "初探支持向量机算法在海洋站观测数据质量控制中的应用", 《海洋预报》 *
杨志飞等: "基于短时LS-SVM的GNSS铁路路基沉降监测数据分析研究", 《测控技术》 *
邹荣等: "姿态角高精度数据预测方法", 《遥感信息》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113281796A (en) * 2021-07-23 2021-08-20 腾讯科技（深圳）有限公司 Position determining method, speed determining method, device, equipment and storage medium
CN113281796B (en) * 2021-07-23 2021-10-15 腾讯科技（深圳）有限公司 Position determining method, speed determining method, device, equipment and storage medium
CN113642252A (en) * 2021-08-31 2021-11-12 中国人民解放军国防科技大学 Target positioning method and device based on single satellite, electronic equipment and medium
CN113642252B (en) * 2021-08-31 2022-05-17 中国人民解放军国防科技大学 Target positioning method and device based on single satellite, electronic equipment and medium

Also Published As

Publication number Publication date
CN112817020B (en) 2023-01-10

Similar Documents

Publication Publication Date Title
US20180165548A1 (en) Systems and methods for object tracking
CN112817020B (en) GNSS observation data quality control and positioning method based on SVM model
CN111160210B (en) Video-based water flow rate detection method and system
US20140347219A1 (en) Receivers and methods for multi-mode navigation
CN110058222B (en) Double-layer particle filter tracking-before-detection method based on sensor selection
CN113138402B (en) RTK-based ambiguity fixing method and device and storage medium
CN111965632A (en) Radar target detection method based on Riemann manifold dimensionality reduction
CN113393497B (en) Ship target tracking method, device and equipment of sequence remote sensing image under condition of broken clouds
CN109143224A (en) A kind of multiple target correlating method and device
CN111077550A (en) Gross error detection method and system applied to RTD positioning of intelligent terminal
CN111220960A (en) Target feature extraction method and device under multi-station radar
CN114972339B (en) Data enhancement system for bulldozer structural member production abnormity detection
CN104298650B (en) Multi-method fusion based Kalman filtering quantization method
CN108596032B (en) Detection method, device, equipment and medium for fighting behavior in video
CN116012364A (en) SAR image change detection method and device
CN111366936A (en) Multi-beam sounding data processing method and device
CN105184829B (en) A kind of tight quarters target detection and high-precision method for positioning mass center
Schroth et al. Failure detection and exclusion via range consensus
CN114002722A (en) GNSS positioning method and device
CN109001782B (en) Method and device for fixing residual part fuzzy after inspection
CN114488227B (en) Multipath error correction method based on spatial correlation
CN113608190B (en) Sea surface target detection method and system based on three characteristics of singular space
Trygonis et al. An operational system for automatic school identification on multibeam sonar echoes
Lu et al. Automatic outlier detection in multibeam bathymetric data using robust LTS estimation
CN111830488B (en) Echo signal data processing method and system based on GM-APD laser ranging system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant