Indoor fusion positioning data processing method and system, storage medium and electronic equipment

Abstract

The invention belongs to the technical field of indoor positioning, and provides an indoor fusion positioning data processing method and system, a storage medium and electronic equipment, and the method comprises the steps: obtaining an original pseudo-range observed quantity, a carrier phase, ephemeris data and reference station differential data, carrying out the processing of the original pseudo-range observed quantity according to the reference station differential data and the carrier phase, and carrying out the processing of the original pseudo-range observed quantity; obtaining a target pseudo-range observed quantity, and generating pseudo-range differential positioning data according to the target pseudo-range observed quantity, the reference station differential data and the ephemeris data; generating to-be-measured data of the to-be-measured point according to the triaxial magnetic field intensity, matching the to-be-measured data with feature data in an indoor geomagnetic fingerprint database, and determining geomagnetic matching positioning data; and performing federated filtering calculation on the pseudo-range differential positioning data and the geomagnetic matching positioning data to determine a final positioning result. The robustness, stability and continuity of positioning data processing are improved, a more stable and more accurate indoor positioning effect is provided, and the arrangement cost of indoor positioning related equipment is reduced.

Description

Technical Field

[0001] The present disclosure belongs to the field of indoor positioning technology, and in particular relates to a method, system, storage medium and electronic device for processing indoor fusion positioning data. Background Art

[0002] Currently, the positioning function of mobile terminals, such as smartphones, is mainly realized outdoors by combining GNSS (Global Navigation Satellite System) technology and inertial navigation technology. In ideal open environments, it can achieve positioning accuracy of decimeters to meters. However, in urban canyons and indoors, because satellite signals are blocked by buildings and other environments, other indoor positioning technologies are mostly used.

[0003] Existing indoor distribution systems utilize hardware such as cable antennas or leaky cables installed within buildings to evenly distribute signals across various areas, eliminating signal "blind spots." Indoor distribution systems support transparent transmission of RF signals and distribute, transmit, and receive RF signals along a specified path (transmission medium), enabling users to communicate with their devices anytime, anywhere. Indoor distribution systems are widely applicable, including large indoor public spaces, buildings, tunnels, and underground locations. Passive indoor distribution systems primarily utilize high-power distributed base stations as signal sources. The distribution system consists of components such as POIs (Points of Interest) / combiners, feeders, leaky cables, power splitters, couplers, and antennas. To address the issue of satellite signal shielding indoors and reduce the cost of indoor positioning, technologies have emerged that relay Beidou satellite signals through passive indoor distribution systems and use GNSS fusion multi-source sensors on indoor mobile terminals for positioning. This makes indoor GNSS positioning and navigation possible. However, the current accuracy of indoor GNSS positioning for mobile terminals is generally at the decimeter level, failing to meet the requirements for high-precision positioning and incurring high deployment costs. Summary of the Invention

[0004] To solve the above problems, the present invention provides a method, system, storage medium and electronic device for processing indoor fusion positioning data, which processes the original pseudorange observations according to the base station differential data and carrier phase to obtain the target pseudorange observations, and generates pseudorange differential positioning data according to the target pseudorange observations, the base station differential data and the ephemeris data; pre-constructs an indoor geomagnetic fingerprint library, generates the test data of the test point according to the three-axis magnetic field strength, and uses the weighted K nearest neighbor algorithm to match the test data with the feature data in the indoor geomagnetic fingerprint library to determine the geomagnetic matching positioning data; calculates the pseudorange differential positioning data and the geomagnetic matching positioning data based on the federal filtering algorithm to obtain the global optimal estimation, and determines the global optimal estimation as the final positioning result. It can improve the robustness, stability and continuity of positioning data processing; provide a more stable and accurate indoor positioning effect, and reduce the deployment cost of indoor positioning related equipment.

[0005] The present invention is achieved through the following technical solutions:

[0006] In a first aspect, an embodiment of the present disclosure provides a method for processing indoor fusion positioning data, the method comprising:

[0007] Receiving original observation data, ephemeris data, and reference station differential data, parsing the original observation data to obtain original pseudorange observations and carrier phases, processing the original pseudorange observations according to the reference station differential data and the carrier phases to obtain target pseudorange observations, and generating pseudorange differential positioning data according to the target pseudorange observations, the reference station differential data, and the ephemeris data;

[0008] Acquire the three-axis magnetic field strength, generate the measured data of the measured point according to the three-axis magnetic field strength, and use the measured data to match the feature data in the indoor geomagnetic fingerprint library based on the weighted K-nearest neighbor algorithm to determine the geomagnetic matching positioning data;

[0009] Based on the federated filtering algorithm, the pseudorange differential positioning data and the geomagnetic matching positioning data are calculated to obtain a global optimal estimate, which is determined as a final positioning result.

[0010] Further,

[0011] Setting a plurality of sampling points and a plurality of non-sampling points, and obtaining the three-axis magnetic field strength of the sampling points; wherein the three-axis magnetic field strength includes the magnetic field strength in the x-axis direction, the magnetic field strength in the y-axis direction, and the magnetic field strength in the z-axis direction;

[0012] Generate sampling data according to the three-axis magnetic field strength of the sampling point; wherein the sampling data includes the total magnetic field strength of the sampling point, the magnetic field strength in the horizontal direction relative to the ground, and the magnetic field strength in the vertical direction relative to the ground;

[0013] Determining estimated data of the non-sampling point using an inverse distance interpolation estimation method based on the distance from the non-sampling point to the target sampling point and the sampling data of the target sampling point; wherein the estimated data includes the total magnetic field intensity, the magnetic field intensity in a horizontal direction relative to the ground, and the magnetic field intensity in a vertical direction relative to the ground at the non-sampling point;

[0014] The sampling data of all sampling points and the estimated data of all non-sampling points are determined as feature points and feature data corresponding to the feature points, and the indoor geomagnetic fingerprint library is constructed using all feature points and the feature data corresponding to the feature points.

[0015] Further,

[0016] Based on the magnetic field strength in the x-axis direction, the y-axis direction, and the z-axis direction at the sampling point, the total magnetic field strength, the magnetic field strength in the horizontal direction relative to the ground, and the magnetic field strength in the vertical direction relative to the ground at the sampling point are calculated. The specific formulas include:

[0017]

[0018] in, represents the roll angle, θ represents the pitch angle, M x Indicates the magnetic field strength in the x-axis direction of the sampling point, M y Indicates the magnetic field strength in the y-axis direction of the sampling point, M z Indicates the magnetic field strength in the z-axis direction of the sampling point, M indicates the total magnetic field strength at the sampling point, M h Indicates the magnetic field strength at the sampling point relative to the ground horizontally, M v Indicates the magnetic field strength at the sampling point in the direction perpendicular to the ground.

[0019] Further,

[0020] Based on the reference station differential data, a pseudorange correction number and a range change rate correction number are obtained, and the original pseudorange observation value is processed using the pseudorange correction number and the range change rate correction number to generate a corrected pseudorange observation value;

[0021] Performing carrier phase smoothing on the corrected pseudorange observations using Hatch filtering to generate smoothed pseudorange observations;

[0022] Performing eigenvalue calculation on the smoothed pseudorange observation to generate the target pseudorange observation.

[0023] Further,

[0024] The pseudorange correction number and the range change rate correction number are used to process the original pseudorange observation value to generate the corrected pseudorange observation value. The specific formula includes:

[0025] PR(t)=PRM(t)+PRC(t0)+RRC×(t-t0); (4)

[0026] Where t0 is the reference epoch, in seconds; t is the current epoch, in seconds; PR(t) is the corrected pseudorange observation at the current epoch, in meters; PRM(t) is the raw pseudorange observation at the current epoch, in meters; PRC(t0) is the pseudorange correction at the reference epoch, in meters; and RRC is the range rate correction, in meters per second.

[0027] Further,

[0028] Hatch filtering is used to perform carrier phase smoothing on the corrected pseudorange observation to generate a smoothed pseudorange observation. The specific formula includes:

[0029]

[0030] in, is the initial condition, the smoothed pseudorange observation at the first epoch is equal to the corrected pseudorange observation; represent the carrier phase of the nth and n-1th epochs respectively; P(n) represents the corrected pseudorange observation of the nth epoch, represents the carrier phase smoothed corrected pseudorange observations at the nth and n-1th epochs, where n is a non-zero natural number.

[0031] Further,

[0032] Calculate the two-dimensional Euclidean distance between the point to be measured and multiple feature points in the indoor geomagnetic fingerprint library. The specific formula includes:

[0033]

[0034] Where d is the two-dimensional Euclidean distance from the point to be measured to the feature point, T represents the total magnetic field intensity of the point to be measured, and T h Indicates the magnetic field strength of the measured point relative to the horizontal direction of the ground, T v Indicates the magnetic field strength of the point to be measured relative to the ground in the vertical direction; F represents the total magnetic field strength of the feature point, F h Indicates the magnetic field strength of the feature point relative to the horizontal direction of the ground, F v Indicates the magnetic field strength of the feature point in the direction perpendicular to the ground;

[0035] According to the two-dimensional Euclidean distance between the test point and the feature point, multiple feature points are sorted, and the k target feature points with the closest two-dimensional Euclidean distance to the test point are selected; where k is a non-zero natural number;

[0036] The feature coordinates of the k target feature points are weighted averaged to determine the coordinates of the target point. The specific formula includes:

[0037]

[0038] Among them, d i is the two-dimensional Euclidean distance from the point to be measured to the i-th target feature point, w i is the weight of the i-th target feature point; W is the total weight; is the coordinate of the point to be measured, P i is the coordinate of the i-th target feature point, is the normalized weight of the i-th target feature point; where i is a non-zero natural number;

[0039] Generate geomagnetic matching positioning data based on the measured coordinates of the measured point.

[0040] Further,

[0041] Based on the Kalman estimation algorithm, a first state vector and a first covariance matrix are generated according to the pseudorange differential positioning data and the ratio of the first optimal estimate, the first estimation error and the first weight factor fed back by the main filtering unit at the previous moment;

[0042] Based on the Kalman estimation algorithm, a second state vector and a second covariance matrix are generated according to the geomagnetic matching positioning data and the ratio of the first optimal estimate value fed back by the main filtering unit at the previous moment, the first estimation error, and the second weight factor;

[0043] Based on the Kalman estimation algorithm, a second optimal estimation value and a second estimation error are generated according to the ratio of the first state vector and the first covariance matrix, the second state vector and the second covariance matrix, and the first optimal estimation value, the first estimation error and the third weight factor; if the preset iteration termination condition is met, or the second estimation error is less than the preset error threshold, the second optimal estimation value is determined as the global optimal estimation value.

[0044] In a second aspect, based on the same inventive concept, an embodiment of the present disclosure further provides an indoor fusion positioning data processing system for executing any of the aforementioned indoor fusion positioning data processing methods, the indoor fusion positioning data processing system comprising: a pseudorange differential positioning module, a geomagnetic matching positioning module, and a data processing module;

[0045] The pseudorange differential positioning module is configured to receive raw observation data, ephemeris data, and reference station differential data, parse the raw observation data, obtain raw pseudorange observations and carrier phase, process the raw pseudorange observations based on the reference station differential data and the carrier phase, obtain target pseudorange observations, and generate pseudorange differential positioning data based on the target pseudorange observations, the reference station differential data, and the ephemeris data;

[0046] The geomagnetic matching and positioning module is used to obtain the three-axis magnetic field strength, generate the measured data of the measured point according to the three-axis magnetic field strength, and use the weighted K nearest neighbor algorithm to match the measured data with the feature data in the indoor geomagnetic fingerprint library to determine the geomagnetic matching positioning data;

[0047] The data processing module is used to calculate the pseudo-range differential positioning data and the geomagnetic matching positioning data based on a federal filtering algorithm to obtain a global optimal estimation, and the global optimal estimation is determined as a final positioning result.

[0048] Further,

[0049] The data processing module includes: a first sub-filtering unit, a second sub-filtering unit and a main filtering unit;

[0050] The first sub-filtering unit is configured to generate a first state vector and a first covariance matrix based on the pseudorange differential positioning data and the ratio of the first optimal estimate, the first estimation error, and the first weight factor fed back by the main filtering unit at the previous moment based on the Kalman estimation algorithm;

[0051] The second sub-filtering unit is configured to generate a second state vector and a second covariance matrix based on the geomagnetic matching positioning data and the ratio of the first optimal estimate, the first estimation error, and the second weight factor fed back by the main filtering unit at the previous moment based on the Kalman estimation algorithm;

[0052] The main filtering unit is used to generate a second optimal estimate and a second estimation error based on the first state vector and the first covariance matrix, the second state vector and the second covariance matrix, and the first optimal estimate, the first estimation error and the ratio of the third weight factor; if the preset iteration termination condition is met, or the second estimation error is less than the preset error threshold, the second optimal estimate is determined as the global optimal estimate.

[0053] On the third aspect, based on the same inventive concept, the embodiment of the present disclosure also provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed, the aforementioned indoor fusion positioning data processing method can be implemented.

[0054] Fourthly, based on the same inventive concept, embodiments of the present disclosure further provide an electronic device comprising a processor, a communication interface, the aforementioned computer-readable storage medium, and a communication bus. The processor, the communication interface, and the computer-readable storage medium communicate with each other via the communication bus. The processor is configured to execute a program stored in the aforementioned computer-readable storage medium.

[0055] Compared with the prior art, the present disclosure has the following advantages:

[0056] 1. During the positioning process, pseudo-range differential positioning data and geomagnetic matching positioning data are integrated to provide more stable and accurate positioning results. In addition, it can reduce the equipment required for indoor positioning and reduce the deployment cost of indoor positioning-related equipment.

[0057] 2. When performing pseudo-range differential positioning, the pseudo-range observations are corrected and smoothed, which improves the robustness, stability and continuity of positioning data processing, thereby improving the stability and continuity of positioning.

[0058] Other features and advantages of the present disclosure will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present disclosure. The purposes and other advantages of the present disclosure can be realized and obtained by the structures indicated in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0060] Figure 1 A flow chart of a method for processing indoor fusion positioning data provided by an embodiment of the present disclosure;

[0061] Figure 2 A flow chart of another method for processing indoor fusion positioning data provided by an embodiment of the present disclosure;

[0062] Figure 3 A block diagram of an indoor fusion positioning data processing system provided in an embodiment of the present disclosure;

[0063] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present disclosure without making any creative efforts shall fall within the scope of protection of the present disclosure.

[0065] Firstly, Figure 1 This is a flow chart of a method for processing indoor fusion positioning data provided by an embodiment of the present disclosure. Figure 2 Another flow chart of the indoor fusion positioning data processing method provided by the embodiment of the present disclosure is as follows: Figure 1 、 2 As shown, the embodiment of the present disclosure provides a method for processing indoor fusion positioning data, including:

[0066] S1: Receive original observation data, ephemeris data and reference station differential data, parse the original observation data, obtain original pseudorange observations and carrier phase, process the original pseudorange observations according to the reference station differential data and carrier phase, obtain target pseudorange observations, and generate pseudorange differential positioning data based on the target pseudorange observations, reference station differential data and ephemeris data.

[0067] Specifically, the mobile terminal receives the original observation data through the GNSS data interface, parses the original observation data, obtains the original pseudorange observation quantity and carrier phase, and also obtains Doppler, carrier-to-noise ratio and other data for positioning solution. While obtaining the original observation data, the mobile terminal also obtains ephemeris data and reference station data. The reference station data includes the reference station differential data obtained using the network protocol and related software, and may also include related data such as the reference station position coordinates. The original pseudorange observation quantity is processed according to the reference station differential data and the carrier phase to obtain the target pseudorange observation quantity. The target pseudorange observation quantity is optimized data. Using the target pseudorange observation quantity, the reference station differential data and the ephemeris data to generate pseudorange differential positioning data can improve the precision and accuracy of positioning, and at the same time, it also improves the robustness, continuity and stability of positioning data processing. Mobile terminals include smart phones, tablets, notebooks and other devices.

[0068] S2: Obtain the three-axis magnetic field strength, generate the test data of the test point based on the three-axis magnetic field strength, and use the weighted K-nearest neighbor algorithm to match the test data with the feature data in the indoor geomagnetic fingerprint library to determine the geomagnetic matching positioning data.

[0069] Specifically, the three-axis magnetic field strength includes the magnetic field strength in the x-axis direction, the magnetic field strength in the y-axis direction, and the magnetic field strength in the z-axis direction. Based on the three-axis magnetic field strength, the measured data of the measured point is generated. The measured data includes the total magnetic field strength of the measured point, the magnetic field strength in the horizontal direction relative to the ground, and the magnetic field strength in the vertical direction relative to the ground. The mobile terminal can download and store the indoor geomagnetic fingerprint library from the network, or obtain the indoor geomagnetic fingerprint library through the network during positioning. The mobile terminal includes a magnetometer, which can be used to collect the three-axis magnetic field strength.

[0070] S3: Based on the federated filtering algorithm, the pseudo-range differential positioning data and the geomagnetic matching positioning data are calculated to obtain the global optimal estimate, which is determined as the final positioning result.

[0071] In the disclosed embodiment, during indoor positioning, pseudo-range differential positioning data and geomagnetic matching positioning data are combined to improve positioning stability and accuracy. At the same time, the deployment of some indoor positioning-related equipment can be reduced. For example, Bluetooth devices used for positioning can be saved, or some ground terminal equipment and antennas can be reduced. While reducing the cost of indoor positioning deployment, the cost of subsequent human maintenance is also saved. The existing indoor Beidou fusion positioning solution relies heavily on the signal strength of Bluetooth data as the basis for selecting beacons. Therefore, there is often a positioning delay within the intersection range of multiple Bluetooth beacons, or the positioning point repeatedly jumps significantly due to the wrong beacon selection, which seriously affects the positioning effect. This solution replaces Bluetooth technology with geomagnetic matching, which reduces the impact of Bluetooth during the positioning process. At the same time, geomagnetism is more stable than Bluetooth and is not easily affected by the environment, which greatly improves the stability of indoor positioning.

[0072] In some examples, the indoor fusion positioning data processing method further includes pre-establishing an indoor geomagnetic fingerprint library. The specific process includes:

[0073] S11': Set multiple sampling points and multiple non-sampling points to obtain the three-axis magnetic field strength at the sampling points.

[0074] Specifically, the three-axis magnetic field strength includes the magnetic field strength in the x-axis direction, the magnetic field strength in the y-axis direction, and the magnetic field strength in the z-axis direction.

[0075] S12': Generate sampling data according to the three-axis magnetic field strength at the sampling point.

[0076] Specifically, the total magnetic field intensity, the magnetic field intensity in the horizontal direction relative to the ground, and the magnetic field intensity in the vertical direction relative to the ground at the sampling point are calculated based on the magnetic field intensity in the x-axis direction, the magnetic field intensity in the y-axis direction, and the magnetic field intensity in the z-axis direction at the sampling point. The specific formulas include:

[0077]

[0078] in, represents the roll angle, θ represents the pitch angle, M x Indicates the magnetic field strength in the x-axis direction of the sampling point, M y Indicates the magnetic field strength in the y-axis direction of the sampling point, M z Indicates the magnetic field strength in the z-axis direction of the sampling point, M indicates the total magnetic field strength at the sampling point, M h Indicates the magnetic field strength at the sampling point relative to the ground horizontally, M v Indicates the magnetic field strength at the sampling point relative to the ground in the vertical direction. The total magnetic field strength at the sampling point, the magnetic field strength horizontally relative to the ground, and the magnetic field strength vertically relative to the ground constitute the sampling data.

[0079] S13′: Determine estimated data of the non-sampling point using an inverse distance interpolation estimation method according to the distance from the non-sampling point to the target sampling point and the sampling data of the target sampling point.

[0080] Specifically, for each non-sampling point that needs to be estimated, the sampling points around it are determined as target sampling points. According to M, M h , M v The estimated data of the non-sampling point is calculated by weighted average based on the value and the distance from the non-sampling point to the target sampling point, that is, the total magnetic field intensity of the non-sampling point, the magnetic field intensity in the horizontal direction relative to the ground, and the magnetic field intensity in the vertical direction relative to the ground.

[0081] S14′: Determine the sampled data of all sampling points and the estimated data of all non-sampling points as feature points and feature data corresponding to the feature points, and use all feature points and the feature data corresponding to the feature points to build an indoor geomagnetic fingerprint library.

[0082] Specifically, after determining the sampled data for all sampling points and obtaining the estimated data for all non-sampling points, the sampled data for the sampling points and the estimated data for the non-sampling points are determined as feature points and corresponding feature data. All feature points and corresponding feature data are used to construct an indoor geomagnetic fingerprint library for subsequent geomagnetic matching and positioning. The indoor geomagnetic fingerprint library can be updated in real time according to actual needs. For example, when the indoor environment or indoor structure changes, the indoor geomagnetic fingerprint library is updated to prevent positioning errors.

[0083] In some examples, processing the raw pseudorange observations based on the reference station differential data and the carrier phase to obtain the target pseudorange observations includes:

[0084] S11: Based on the differential data of the reference station, a pseudorange correction number and a range change rate correction number are obtained, and the original pseudorange observation value is processed using the pseudorange correction number and the range change rate correction number to generate a corrected pseudorange observation value.

[0085] Specifically, the pseudorange correction number and the range change rate correction number are used to process the original pseudorange observation to generate the corrected pseudorange observation. The specific formula includes:

[0086] PR(t)=PRM(t)+PRC(t0)+RRC×(t-t0); (4)

[0087] Where t0 is the reference epoch, in seconds; t is the current epoch, in seconds; PR(t) is the corrected pseudorange observation at the current epoch, in meters; PRM(t) is the raw pseudorange observation at the current epoch, in meters; PRC(t0) is the pseudorange correction at the reference epoch, in meters; and RRC is the range rate correction, in meters per second.

[0088] S12: Hatch filtering is used to perform carrier phase smoothing on the corrected pseudorange observations to generate smoothed pseudorange observations.

[0089] Specifically, Hatch filtering is used to perform carrier phase smoothing on the corrected pseudorange observations to generate smoothed pseudorange observations. The specific formula includes:

[0090]

[0091] in, is the initial condition, the smoothed pseudorange observation of the first epoch is equal to the corrected pseudorange observation; φ(n) and φ(n-1) represent the carrier phase of the nth and n-1th epochs respectively; P(n) represents the corrected pseudorange observation of the nth epoch, represents the carrier phase smoothed corrected pseudorange observations at the nth and n-1th epochs, where n is a non-zero natural number.

[0092] S13: Calculate the eigenvalue of the smoothed pseudorange observation to generate the target pseudorange observation.

[0093] Specifically, eigenvalue calculation is performed on the smoothed pseudorange observations, including calculating the residual and root mean square error of the observations based on the smoothed pseudorange observations. Alternatively, a covariance matrix may be constructed using multiple smoothed pseudorange observations, and then eigenvalue calculation may be performed on the smoothed pseudorange observations based on the constructed covariance matrix. The aforementioned method for calculating eigenvalues of the smoothed pseudorange observations is merely an example and is not further specifically limited. Persons skilled in the relevant art may adjust the method for calculating eigenvalues based on actual deployment conditions and actual indoor environments.

[0094] In some examples, based on a weighted K-nearest neighbor algorithm, the measured data is matched with feature data in an indoor geomagnetic fingerprint library to determine geomagnetic matching positioning data, including:

[0095] S21: Calculate the two-dimensional Euclidean distance between the point to be measured and multiple feature points in the indoor geomagnetic fingerprint library. The specific formula includes:

[0096]

[0097] Where d is the two-dimensional Euclidean distance from the point to be measured to the feature point, T represents the total magnetic field intensity of the point to be measured, and T h Indicates the magnetic field strength of the measured point relative to the horizontal direction of the ground, T v Indicates the magnetic field strength of the point to be measured relative to the ground in the vertical direction; F represents the total magnetic field strength of the feature point, F h Indicates the magnetic field strength of the feature point relative to the horizontal direction of the ground, F vIndicates the magnetic field strength of the feature point in the direction perpendicular to the ground;

[0098] S22: sorting the multiple feature points according to the two-dimensional Euclidean distance between the test point and the feature point, and selecting k target feature points whose two-dimensional Euclidean distance is closest to the test point; wherein k is a non-zero natural number;

[0099] S23: Perform weighted averaging on the feature coordinates of the k target feature points to determine the coordinates of the target point. The specific formula includes:

[0100]

[0101] Among them, d i is the two-dimensional Euclidean distance from the point to be measured to the i-th target feature point, w i is the weight of the i-th target feature point; W is the total weight; is the coordinate of the point to be measured, P i is the coordinate of the i-th target feature point, is the normalized weight of the i-th target feature point; i is a nonzero natural number. Determining the coordinates of the target point using a weighted average of the feature coordinates of k target feature points can prevent the influence of factors such as people and electronic devices near the target point on geomagnetic matching positioning.

[0102] S24: Generate geomagnetic matching positioning data according to the coordinates of the points to be measured.

[0103] In some examples, a federated filtering algorithm is used to calculate pseudorange differential positioning data and geomagnetic matching positioning data to obtain a global optimal estimate, including:

[0104] S31: Based on the Kalman estimation algorithm, a first state vector and a first covariance matrix are generated according to the pseudorange differential positioning data and the ratio of the first optimal estimate, the first estimation error and the first weight factor fed back by the main filtering unit at the previous moment.

[0105] S32: Based on the Kalman estimation algorithm, a second state vector and a second covariance matrix are generated according to the geomagnetic matching positioning data and the ratio of the first optimal estimation value, the first estimation error and the second weight factor fed back by the main filtering unit at the previous moment.

[0106] S33: Based on the Kalman valuation algorithm, a second optimal estimate and a second estimation error are generated according to the ratio of the first state vector and the first covariance matrix, the second state vector and the second covariance matrix, the first optimal estimate, the first estimation error and the third weight factor; if the preset iteration termination condition is met, or the second estimation error is less than the preset error threshold, the second optimal estimate is determined as the global optimal estimate.

[0107] Specifically, after generating the second optimal estimate and the second estimation error, if the preset iteration termination condition is met, the iteration is stopped, and the second optimal estimate is determined as the global optimal estimate; the preset iteration termination condition includes a preset number of iterations, and if the current actual number of iterations does not reach the preset number of iterations, the iteration continues until the iteration termination condition is met, the iteration is stopped, and the global optimal estimate is determined; or, if the second estimation error is less than the preset error threshold, the second optimal estimate is determined as the global optimal estimate.

[0108] In the embodiment of the present disclosure, after the final positioning result is determined, it can be displayed through the display module of the mobile terminal to inform the user of the final positioning result.

[0109] Example 1:

[0110] Build an indoor geomagnetic fingerprint library:

[0111] A sampling area is set up indoors, and multiple sampling points and non-sampling points are set up in the sampling area. The magnetic field information acquisition module (e.g., magnetometer) of a mobile terminal or a separate magnetic field information acquisition device is used to collect the three-axis magnetic field intensity at the sampling point. The three-axis magnetic field intensity includes the magnetic field intensity in the x-axis direction, the magnetic field intensity in the y-axis direction, and the magnetic field intensity in the z-axis direction. The three-axis magnetic field intensity is substituted into formulas (1) to (3) for calculation to determine the magnetic field intensity in the horizontal and vertical directions of the sampling point relative to the ground, as well as the total magnetic field intensity at the sampling point. The specific formula is:

[0112]

[0113] in, represents the roll angle, θ represents the pitch angle, M x Indicates the magnetic field strength in the x-axis direction of the sampling point, M y Indicates the magnetic field strength in the y-axis direction of the sampling point, M z Indicates the magnetic field strength in the z-axis direction of the sampling point, M indicates the total magnetic field strength at the sampling point, M h Indicates the magnetic field strength at the sampling point relative to the ground horizontally, M v Indicates the magnetic field strength at the sampling point in the direction perpendicular to the ground.

[0114] When building an indoor geomagnetic fingerprint library, the network interval must be determined based on the scenario and positioning requirements. The network interval is the distance between geomagnetic data collected in an indoor space. In indoor spaces with complex layouts and high positioning requirements, the network interval can be set relatively close, for example, 1 or 2 meters. In indoor spaces with simple layouts and low positioning requirements, the network interval can be set relatively far, for example, 5 or 10 meters. The network interval directly affects the density of sampling and non-sampling points, which in turn affects the amount of data in the indoor geomagnetic fingerprint library and the accuracy of geomagnetic matching positioning.

[0115] Collect the three-axis magnetic field strength of all sampling points and use formulas (1) to (3) to calculate the total magnetic field strength M of the sampling point and the magnetic field strength M relative to the ground horizontal direction h , the magnetic field strength M in the direction perpendicular to the ground v The sampling data of the sampling point includes the total magnetic field intensity of the sampling point, the magnetic field intensity in the horizontal direction relative to the ground, and the magnetic field intensity in the vertical direction relative to the ground. The sampling data can be expressed as (M, M h , M v ).

[0116] Due to the limited number of sampling points, it is necessary to use the inverse distance interpolation estimation method based on the sampling data of known sampling points to determine the estimated data of non-sampling points. The estimated data include the total magnetic field intensity of non-sampling points, the magnetic field intensity in the horizontal direction relative to the ground, and the magnetic field intensity in the vertical direction relative to the ground.

[0117] Furthermore, the distance from the non-sampling point to the target sampling point and the sampling data of the target sampling point are used to determine the estimated data of the non-sampling point using the inverse distance interpolation estimation method. In this process, for each non-sampling point to be estimated, the sampling points around it are determined as the target sampling points. According to the M and M of the target sampling point, h , M v The estimated data of the non-sampling point is calculated by weighted average based on the value and the distance from the non-sampling point to the target sampling point, that is, the total magnetic field intensity of the non-sampling point, the magnetic field intensity in the horizontal direction relative to the ground, and the magnetic field intensity in the vertical direction relative to the ground.

[0118] It is understandable that the target sampling point is not a single sampling point, but multiple sampling points around the non-sampling point. The closer the sampling point is to the non-sampling point, the greater its weight is, and the greater its contribution to the estimated data of the non-sampling point is.

[0119] After determining the sampled data for all sampling points and obtaining the estimated data for all non-sampling points, the sampled data for the sampling points and the estimated data for the non-sampling points are identified as feature points and their corresponding feature data. This data is then used to construct an indoor geomagnetic fingerprint library for subsequent geomagnetic matching and positioning. The indoor geomagnetic fingerprint library can be updated in real time based on actual needs. For example, if the indoor environment or indoor structure changes, the indoor geomagnetic fingerprint library can be updated to prevent positioning errors.

[0120] Pseudorange differential positioning:

[0121] The mobile terminal receives raw observation data through the GNSS data interface, analyzes it, and obtains raw pseudorange observations and carrier phase. It also obtains Doppler, carrier-to-noise ratio, and other data for positioning calculations. Along with the raw observation data, the mobile terminal also obtains ephemeris data and reference station data. Reference station data includes reference station differential data obtained using network protocols and related software, and may also include related data such as the reference station's position coordinates.

[0122] The raw pseudorange observations are processed based on the base station differential data and carrier phase to obtain the target pseudorange observations. The target pseudorange observations are optimized data. Using the target pseudorange observations, base station differential data, and ephemeris data to generate pseudorange differential positioning data improves positioning precision and accuracy, while also enhancing the robustness, continuity, and stability of positioning data processing.

[0123] The original pseudorange observations are processed according to the differential data of the reference station and the carrier phase to obtain the target pseudorange observations, including correcting the original pseudorange observations based on the differential data of the reference station to generate corrected pseudorange observations; and also including using Hatch filtering to perform carrier phase smoothing on the corrected pseudorange observations. Through data processing, the pseudorange observations are optimized to improve the measurement accuracy of pseudorange differential positioning and reduce errors.

[0124] Performing pseudorange differential correction on the original pseudorange observations to generate corrected pseudorange observations can improve the precision and accuracy of positioning. Based on the differential data of the reference station, the pseudorange correction number and the range change rate correction number are obtained. The pseudorange correction number and the range change rate correction number are used to process the original pseudorange observations to generate corrected pseudorange observations. The specific formula includes:

[0125] PR(t)=PRM(t)+PRC(t0)+RRC×(t-t0); (4)

[0126] Where t0 is the reference epoch, in seconds; t is the current epoch, in seconds; PR(t) is the corrected pseudorange observation at the current epoch, in meters; PRM(t) is the raw pseudorange observation at the current epoch, in meters; PRC(t0) is the pseudorange correction at the reference epoch, in meters; RRC is the range rate correction, in meters per second. (t-t0) is in seconds, so the product RRC×(t-t0) is in meters.

[0127] Hatch filtering is used to perform carrier phase smoothing on the corrected pseudorange observations to generate smoothed pseudorange observations. The specific formula includes:

[0128]

[0129] in, is the initial condition, the smoothed pseudorange observation of the first epoch is equal to the corrected pseudorange observation; φ(n) and φ(n-1) represent the carrier phase of the nth and n-1th epochs respectively; P(n) represents the corrected pseudorange observation of the nth epoch, Represents the carrier phase-smoothed corrected pseudorange observations for the nth and n-1th epochs, where n is a nonzero natural number. The carrier phase-smoothed corrected pseudorange observations can be used as smoothed pseudorange observations. Hatch filtering is a method for smoothing pseudorange observations in satellite navigation systems, leveraging the high precision of the carrier phase to improve the accuracy and stability of pseudorange observations.

[0130] After determining the smoothed pseudorange observations, eigenvalue calculations are performed on the smoothed pseudorange observations to generate target pseudorange observations. The eigenvalue calculations for the smoothed pseudorange observations may include calculating residuals and root mean square errors of the observations based on the smoothed pseudorange observations. Alternatively, the eigenvalue calculations may be performed on the smoothed pseudorange observations by constructing a covariance matrix using multiple smoothed pseudorange observations and then performing eigenvalue calculations based on the constructed covariance matrix.

[0131] Pseudorange differential positioning data is generated based on target pseudorange observations, reference station differential data, and ephemeris data. In addition to the original pseudorange observations and carrier phase, the original observation data can also include Doppler and carrier-to-noise ratio data. Pseudorange differential positioning is performed based on target pseudorange observations, reference station differential data, and ephemeris data, combined with Doppler data and carrier-to-noise ratio data. This can further improve positioning accuracy by smoothing the original pseudorange measurements.

[0132] Geomagnetic matching positioning:

[0133] Obtain the three-axis magnetic field strength of the test point. According to the three-axis magnetic field strength, use formulas (1) to (3) to calculate the test data of the test point. The test data includes the total magnetic field strength T of the test point, the magnetic field strength T relative to the horizontal direction of the ground, and the h and the magnetic field strength T in the direction perpendicular to the ground v Based on the weighted K-nearest neighbor algorithm, the measured data of the measured point is matched with the indoor geomagnetic fingerprint library to obtain geomagnetic matching positioning data. The specific process includes:

[0134] Calculate the two-dimensional Euclidean distance between the point to be measured and multiple feature points in the indoor geomagnetic fingerprint library. The specific formula includes:

[0135]

[0136] Where d is the two-dimensional Euclidean distance from the point to be measured to the feature point, T represents the total magnetic field intensity of the point to be measured, and T h Indicates the magnetic field strength of the measured point relative to the horizontal direction of the ground, T v Indicates the magnetic field strength of the point to be measured relative to the ground in the vertical direction; F represents the total magnetic field strength of the feature point, F h Indicates the magnetic field strength of the feature point relative to the horizontal direction of the ground, F v Indicates the magnetic field strength of the feature point in the direction perpendicular to the ground.

[0137] According to the two-dimensional Euclidean distance between the test point and the feature point, multiple feature points are sorted, and the k target feature points with the closest two-dimensional Euclidean distance to the test point are selected; where k is a non-zero natural number;

[0138] The feature coordinates of the k target feature points are weighted averaged to determine the coordinates of the target point. The specific formula includes:

[0139]

[0140] Among them, d i is the two-dimensional Euclidean distance from the point to be measured to the i-th target feature point, w i is the weight of the i-th target feature point; W is the total weight, and the weights w of k target feature points are i Add them together to get the total weight W; is the coordinate of the point to be measured, is the normalized weight of the i-th target feature point, and the coordinates P of the k target feature points are i Perform weighted averaging to obtain the coordinates of the points to be measured Wherein, i is a non-zero natural number.

[0141] Generate geomagnetic matching positioning data based on the measured coordinates of the measured point.

[0142] Fusion positioning:

[0143] Based on the federated filtering algorithm, multiple frames of pseudo-range differential positioning data and multiple frames of geomagnetic matching positioning data are calculated to obtain the global optimal estimate, which is determined as the final positioning result. The specific steps include:

[0144] (a) receiving multiple frames of pseudo-range differential positioning data, and generating a first state vector and a first covariance matrix based on the multiple frames of pseudo-range differential positioning data based on a Kalman estimation algorithm;

[0145] (b) receiving multiple frames of geomagnetic matching positioning data, and generating a second state vector and a second covariance matrix based on the multiple frames of geomagnetic matching positioning data based on a Kalman estimation algorithm;

[0146] (c) generating a first optimal estimate and a first estimation error based on the first state vector and the first covariance matrix, the second state vector and the second covariance matrix based on a Kalman estimation algorithm;

[0147] (d) Feedback the ratio of the first optimal estimate value, the first optimal estimate error and the first weight factor to the first sub-filtering unit, feed back the ratio of the first optimal estimate value, the first optimal estimate error and the second weight factor to the second sub-filtering unit, and cache the first optimal estimate value and the first optimal estimate error; the first weight factor corresponds to the first sub-filtering unit, that is, corresponds to pseudorange differential positioning; the second weight factor corresponds to the second sub-filtering unit, that is, corresponds to geomagnetic matching positioning.

[0148] (e) receiving one or more frames of pseudorange differential positioning data again, and generating a third state vector and a third covariance matrix based on the pseudorange differential positioning data, the first optimal estimate, and the ratio of the first optimal estimate error to the first weight factor based on the Kalman estimation algorithm;

[0149] (f) receiving one or more frames of geomagnetic matching positioning data again, and generating a fourth state vector and a fourth covariance matrix based on the geomagnetic matching positioning data, the first optimal estimate, and the ratio of the first optimal estimate error to the second weight factor based on the Kalman estimation algorithm;

[0150] (g) Based on the Kalman valuation algorithm, a second optimal estimate and a second estimate error are generated according to the third state vector and the third covariance matrix, the fourth state vector and the fourth covariance matrix, the first optimal estimate, the ratio of the first optimal estimate error and the third weight factor; the third weight factor corresponds to the optimal estimate and the estimate error at the previous moment. When the main filtering unit performs Kalman valuation on the two sets of state vectors and covariance matrices at the current moment, the optimal estimate and the estimate error at the previous moment also need to be combined.

[0151] It should be noted that the first weight factor, the second weight factor and the third weight factor are set according to actual conditions and can be adjusted according to positioning accuracy requirements and current environmental conditions, and no further specific limitations are given here.

[0152] If the second estimation error meets the preset conditions, for example, the second estimation error is less than the preset error threshold, the iteration is stopped, and the second optimal estimation value is determined as the global optimal valuation; if the second estimation error is not less than the preset error threshold or the preset iteration termination condition is the preset number of iterations, and the current actual number of iterations does not reach the preset number of iterations, the iteration continues until the iteration termination condition is met, the iteration is stopped, and the global optimal valuation is determined.

[0153] It should be understood that, during the iteration process, the first optimal estimate and the second optimal estimate do not refer to two fixed values. The first optimal estimate represents the optimal estimate at the previous moment in the iteration process, and the second optimal estimate represents the optimal estimate at the current moment in the iteration process. Similarly, the first estimation error and the second estimation error, the first state vector and the third state vector, the first covariance matrix and the third covariance matrix, the second state vector and the fourth state vector, the second covariance matrix and the fourth covariance matrix, etc., all represent different moments in the iteration process.

[0154] In the disclosed embodiment, a federated filtering algorithm is used based on pseudo-range differential positioning data and geomagnetic matching positioning data to achieve real-time positioning of a mobile terminal in an indoor environment. In this process, a frame of pseudo-range differential positioning data and a frame of geomagnetic matching positioning data can be used to determine the final positioning result; in order to improve the accuracy of positioning, multiple frames of pseudo-range differential positioning data and multiple frames of geomagnetic matching positioning data are used to determine the global optimal estimate, and then the final positioning result is determined; at the same time, during the positioning process, the current position information can be determined in combination with the positioning information of the previous moment. For example, when using the Kalman estimation algorithm, the data of the previous moment and the data of the current moment are used for estimation to determine the final positioning result; further, multiple frames of continuous final positioning results can be used to implement navigation and path planning.

[0155] Secondly, Figure 3 This is a block diagram of an indoor fusion positioning data processing system provided by an embodiment of the present disclosure, such as Figure 3 As shown, based on the same inventive concept, an embodiment of the present disclosure further provides an indoor fusion positioning data processing system for executing any of the aforementioned indoor fusion positioning data processing methods, the indoor fusion positioning data processing system comprising: a pseudorange differential positioning module, a geomagnetic matching positioning module, and a data processing module;

[0156] The pseudorange differential positioning module is used to receive the original observation data, ephemeris data and reference station differential data, parse the original observation data, obtain the original pseudorange observation amount and carrier phase, process the original pseudorange observation amount according to the reference station differential data and carrier phase, obtain the target pseudorange observation amount, and generate pseudorange differential positioning data according to the target pseudorange observation amount, the reference station differential data and the ephemeris data;

[0157] The geomagnetic matching and positioning module is used to obtain the three-axis magnetic field strength, generate the test data of the test point based on the three-axis magnetic field strength, and use the weighted K nearest neighbor algorithm to match the test data with the feature data in the indoor geomagnetic fingerprint library to determine the geomagnetic matching positioning data;

[0158] The data processing module is used to calculate the pseudo-range differential positioning data and the geomagnetic matching positioning data based on the federal filtering algorithm to obtain the global optimal estimation, which is determined as the final positioning result.

[0159] In some examples, the data processing module includes: a first sub-filtering unit, a second sub-filtering unit, and a main filtering unit;

[0160] The first sub-filtering unit is configured to generate a first state vector and a first covariance matrix based on the pseudorange differential positioning data and the ratio of the first optimal estimate, the first estimation error, and the first weight factor fed back by the main filtering unit at the previous moment based on the Kalman estimation algorithm;

[0161] The second sub-filtering unit is configured to generate a second state vector and a second covariance matrix based on the geomagnetic matching positioning data and the ratio of the first optimal estimate, the first estimation error, and the second weight factor fed back by the main filtering unit at the previous moment based on the Kalman estimation algorithm;

[0162] The main filtering unit is used to generate a second optimal estimate and a second estimation error based on the first state vector and the first covariance matrix, the second state vector and the second covariance matrix, the first optimal estimate, the first estimation error and the ratio of the third weight factor; if the preset iteration termination condition is met, or the second estimation error is less than the preset error threshold, the second optimal estimate is determined as the global optimal estimate.

[0163] On the third aspect, based on the same inventive concept, the embodiment of the present disclosure also provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed, the aforementioned indoor fusion positioning data processing method can be implemented.

[0164] Fourthly, Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure is shown in FIG. Figure 4As shown, based on the same inventive concept, an embodiment of the present disclosure further provides an electronic device, including a processor, a communication interface, the aforementioned computer-readable storage medium, and a communication bus. The processor, the communication interface, and the computer-readable storage medium communicate with each other via the communication bus. The processor is configured to execute a program stored in the aforementioned computer-readable storage medium.

[0165] It should be noted that the electrical connection between the above-mentioned units does not necessarily mean the connection between lines. An indirect connection method can be applied to the embodiments of the present disclosure as long as the purpose of the present disclosure is achieved.

[0166] Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A method for processing indoor fusion positioning data, characterized in that: The method comprises: Receiving original observation data, ephemeris data, and reference station differential data, parsing the original observation data to obtain original pseudorange observations and carrier phases, processing the original pseudorange observations according to the reference station differential data and the carrier phases to obtain target pseudorange observations, and generating pseudorange differential positioning data according to the target pseudorange observations, the reference station differential data, and the ephemeris data; Acquire the three-axis magnetic field strength, generate the measured data of the measured point according to the three-axis magnetic field strength, and use the measured data to match the feature data in the indoor geomagnetic fingerprint library based on the weighted K-nearest neighbor algorithm to determine the geomagnetic matching positioning data; Based on the federated filtering algorithm, the pseudorange differential positioning data and the geomagnetic matching positioning data are calculated to obtain a global optimal estimate, which is determined as a final positioning result.

2. The method according to claim 1, characterized in that The indoor fusion positioning data processing method further includes: Setting a plurality of sampling points and a plurality of non-sampling points, and obtaining the three-axis magnetic field strength of the sampling points; wherein the three-axis magnetic field strength includes the magnetic field strength in the x-axis direction, the magnetic field strength in the y-axis direction, and the magnetic field strength in the z-axis direction; Generate sampling data according to the three-axis magnetic field strength of the sampling point; wherein the sampling data includes the total magnetic field strength of the sampling point, the magnetic field strength in the horizontal direction relative to the ground, and the magnetic field strength in the vertical direction relative to the ground; Determining estimated data of the non-sampling point using an inverse distance interpolation estimation method based on the distance from the non-sampling point to the target sampling point and the sampling data of the target sampling point; wherein the estimated data includes the total magnetic field intensity, the magnetic field intensity in a horizontal direction relative to the ground, and the magnetic field intensity in a vertical direction relative to the ground at the non-sampling point; The sampling data of all sampling points and the estimated data of all non-sampling points are determined as feature points and feature data corresponding to the feature points, and the indoor geomagnetic fingerprint library is constructed using all feature points and the feature data corresponding to the feature points.

3. The method according to claim 2, characterized in that Generating sampling data according to the three-axis magnetic field strength of the sampling point includes: Based on the magnetic field strength in the x-axis direction, the y-axis direction, and the z-axis direction at the sampling point, the total magnetic field strength, the magnetic field strength in the horizontal direction relative to the ground, and the magnetic field strength in the vertical direction relative to the ground at the sampling point are calculated. The specific formulas include: in, represents the roll angle, θ represents the pitch angle, M x Indicates the magnetic field strength in the x-axis direction of the sampling point, M y Indicates the magnetic field strength in the y-axis direction of the sampling point, M z Indicates the magnetic field strength in the z-axis direction of the sampling point, M indicates the total magnetic field strength at the sampling point, M h Indicates the magnetic field strength at the sampling point relative to the ground horizontally, M v Indicates the magnetic field strength at the sampling point in the direction perpendicular to the ground.

4. The method according to claim 1, wherein The processing of the original pseudorange observation quantity according to the reference station differential data and the carrier phase to obtain the target pseudorange observation quantity includes: Based on the reference station differential data, a pseudorange correction number and a range change rate correction number are obtained, and the original pseudorange observation value is processed using the pseudorange correction number and the range change rate correction number to generate a corrected pseudorange observation value; Performing carrier phase smoothing on the corrected pseudorange observations using Hatch filtering to generate smoothed pseudorange observations; Performing eigenvalue calculation on the smoothed pseudorange observation to generate the target pseudorange observation.

5. The method according to claim 4, characterized in that The pseudorange correction number and the range change rate correction number are used to process the original pseudorange observation value to generate the corrected pseudorange observation value. The specific formula includes: PR(t)=PRM(t)+PRC(t0)+RRC×(t-t0); (4) Where t0 is the reference epoch, in seconds; t is the current epoch, in seconds; PR(t) is the corrected pseudorange observation at the current epoch, in meters; PRM(t) is the raw pseudorange observation at the current epoch, in meters; PRC(t0) is the pseudorange correction at the reference epoch, in meters; and RRC is the range rate correction, in meters per second.

6. The method according to claim 4, wherein the carrier phase of the corrected pseudorange observation is smoothed by using Hatch filtering to generate a smoothed pseudorange observation. The specific formula includes: in, is the initial condition, the smoothed pseudorange observation at the first epoch is equal to the corrected pseudorange observation; represent the carrier phase of the nth and n-1th epochs respectively; P(n) represents the corrected pseudorange observation of the nth epoch, represents the carrier phase smoothed corrected pseudorange observations at the nth and n-1th epochs, where n is a non-zero natural number.

7. The method according to claim 1, characterized in that The method uses the weighted K-nearest neighbor algorithm to match the data to be measured with the feature data in the indoor geomagnetic fingerprint library to determine the geomagnetic matching positioning data, including: Calculate the two-dimensional Euclidean distance between the point to be measured and multiple feature points in the indoor geomagnetic fingerprint library. The specific formula includes: Where d is the two-dimensional Euclidean distance from the point to be measured to the feature point, T represents the total magnetic field intensity of the point to be measured, and T h Indicates the magnetic field strength of the measured point relative to the horizontal direction of the ground, T v Indicates the magnetic field strength of the point to be measured relative to the ground in the vertical direction; F represents the total magnetic field strength of the feature point, F h Indicates the magnetic field strength of the feature point relative to the horizontal direction of the ground, F v Indicates the magnetic field strength of the feature point in the direction perpendicular to the ground; According to the two-dimensional Euclidean distance between the test point and the feature point, multiple feature points are sorted, and the k target feature points with the closest two-dimensional Euclidean distance to the test point are selected; where k is a non-zero natural number; The feature coordinates of the k target feature points are weighted averaged to determine the coordinates of the target point. The specific formula includes: Among them, d i is the two-dimensional Euclidean distance from the point to be measured to the i-th target feature point, w i is the weight of the i-th target feature point; W is the total weight; is the coordinate of the point to be measured, P i is the coordinate of the i-th target feature point, is the normalized weight of the i-th target feature point; where i is a non-zero natural number; Generate geomagnetic matching positioning data based on the measured coordinates of the measured point.

8. The method according to claim 1, characterized in that The method of calculating the pseudo-range differential positioning data and the geomagnetic matching positioning data based on the federated filtering algorithm to obtain the global optimal estimate includes: Based on the Kalman estimation algorithm, a first state vector and a first covariance matrix are generated according to the pseudorange differential positioning data and the ratio of the first optimal estimate, the first estimation error and the first weight factor fed back by the main filtering unit at the previous moment; Based on the Kalman estimation algorithm, a second state vector and a second covariance matrix are generated according to the geomagnetic matching positioning data and the ratio of the first optimal estimate value fed back by the main filtering unit at the previous moment, the first estimation error, and the second weight factor; Based on the Kalman estimation algorithm, a second optimal estimation value and a second estimation error are generated according to the ratio of the first state vector and the first covariance matrix, the second state vector and the second covariance matrix, and the first optimal estimation value, the first estimation error and the third weight factor; if the preset iteration termination condition is met, or the second estimation error is less than the preset error threshold, the second optimal estimation value is determined as the global optimal estimation value.

9. An indoor fusion positioning data processing system, used to execute the indoor fusion positioning data processing method according to any one of claims 1 to 8, characterized in that: The system includes: a pseudo-range differential positioning module, a geomagnetic matching positioning module and a data processing module; The pseudorange differential positioning module is configured to receive raw observation data, ephemeris data, and reference station differential data, parse the raw observation data, obtain raw pseudorange observations and carrier phase, process the raw pseudorange observations based on the reference station differential data and carrier phase, obtain target pseudorange observations, and generate pseudorange differential positioning data based on the target pseudorange observations, reference station differential data, and ephemeris data; The geomagnetic matching and positioning module is used to obtain the three-axis magnetic field strength, generate the measured data of the measured point according to the three-axis magnetic field strength, and use the weighted K nearest neighbor algorithm to match the measured data with the feature data in the indoor geomagnetic fingerprint library to determine the geomagnetic matching positioning data; The data processing module is used to calculate the pseudo-range differential positioning data and the geomagnetic matching positioning data based on a federal filtering algorithm to obtain a global optimal estimation, and the global optimal estimation is determined as a final positioning result.

10. The system according to claim 9, characterized in that The data processing module includes: a first sub-filtering unit, a second sub-filtering unit and a main filtering unit; The first sub-filtering unit is configured to generate a first state vector and a first covariance matrix based on the pseudorange differential positioning data and the ratio of the first optimal estimate, the first estimation error, and the first weight factor fed back by the main filtering unit at the previous moment based on the Kalman estimation algorithm; The second sub-filtering unit is configured to generate a second state vector and a second covariance matrix based on the geomagnetic matching positioning data and the ratio of the first optimal estimate, the first estimation error, and the second weight factor fed back by the main filtering unit at the previous moment based on the Kalman estimation algorithm; The main filtering unit is used to generate a second optimal estimate and a second estimation error based on the first state vector and the first covariance matrix, the second state vector and the second covariance matrix, and the first optimal estimate, the first estimation error and the ratio of the third weight factor; if the preset iteration termination condition is met, or the second estimation error is less than the preset error threshold, the second optimal estimate is determined as the global optimal estimate.

11. A computer-readable storage medium storing one or more programs, characterized in that: When the one or more programs are executed, the indoor fusion positioning data processing method described in any one of claims 1 to 8 can be implemented.

12. An electronic device comprising a processor, a communication interface, the computer-readable storage medium according to claim 11, and a communication bus; wherein: The processor, the communication interface, and the computer-readable storage medium communicate with each other via a communication bus; It is characterized by: The processor is configured to execute a program stored in a computer-readable storage medium.

Images