A Beidou-based multi-source positioning data fusion processing system

Abstract

The present invention belongs to the technical field of navigation and positioning systems. The present invention discloses a Beidou-based multi-source positioning data fusion processing system, including: a positioning data acquisition module, a preliminary positioning module, a geographic information acquisition module, a spatial feature processing module, an initial differential module, a corrected differential module and a final positioning module; the positioning data of Beidou satellites and differential base stations are simultaneously acquired through the positioning data acquisition module, and a comprehensive data basis is provided in combination with the preliminary positioning and geographic information acquisition modules; the spatial feature processing module analyzes the relative spatial features between the positioning terminal and the differential base station; the initial differential module and the corrected differential module use a pre-trained machine learning model to correct the initial differential to obtain a high-precision corrected differential. The final positioning module uses the corrected differential and terminal positioning data to calculate the corrected distance and obtain the position of the positioning terminal. Through error correction, the accuracy of the final positioning is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of navigation and positioning systems, and more specifically, to a Beidou-based multi-source positioning data fusion processing system. Background Art

[0002] The BeiDou satellite navigation system is a global satellite navigation system independently developed by my country, with all-weather, all-day, high-precision positioning, navigation and timing service capabilities. The BeiDou system consists of three parts: the space segment, the ground segment and the user segment. The space segment includes multiple working satellites, the ground segment includes the master control station, injection station and monitoring station, and the user segment includes various BeiDou user terminals. The BeiDou system sends navigation signals to the ground through satellites, and the user terminals receive and process these signals to achieve positioning, navigation and timing functions.

[0003] In the existing satellite navigation positioning technology, differential positioning is a commonly used method to improve positioning accuracy. This method usually calculates the differential error by setting up a fixed differential base station. Since the differential base station and the positioning terminal are often in the same area, the error caused by atmospheric conditions during signal propagation has a similar impact on the base station and the positioning terminal, and this part of the error can be effectively eliminated by differential. The advantage of this method is that it can significantly improve the positioning accuracy, especially in open areas. However, this traditional differential positioning method also has some limitations. The main problem is that the signal obstruction caused by terrain objects such as buildings and mountains will cause a large error difference between the base station and the positioning terminal. This is because the base station is usually set in a place with a wide field of view, while the positioning terminal may be in various complex environments. This difference makes the differential error calculated by the base station unable to accurately reflect the actual error of the positioning terminal, thereby affecting the accuracy of the final positioning result. Summary of the invention

[0004] In order to overcome the problem that the positioning results of the prior art are not accurate enough, the present invention proposes a Beidou-based multi-source positioning data fusion processing system to solve the above problem.

[0005] The present invention provides the following technical solutions:

[0006] A Beidou-based multi-source positioning data fusion processing system, comprising:

[0007] The positioning data acquisition module is used to establish communication between the positioning terminal and multiple Beidou satellites, record the Beidou satellites that successfully establish communication as communication satellites, and obtain terminal positioning data between the positioning terminal and each communication satellite; it is also used to establish communication between the positioning terminal and any differential base station, obtain the base station position of the differential base station that establishes communication and the base station positioning data between the differential base station and each communication satellite;

[0008] The initial positioning module is used to obtain the current position of the positioning terminal through the Beidou positioning system and record it as the initial position;

[0009] A geographic information acquisition module, which acquires geographic information corresponding to the initial position and geographic information corresponding to the base station position;

[0010] The spatial feature processing module is used to obtain the relative spatial features between the positioning terminal and the differential base station under each communication satellite according to the geographical information corresponding to the initial position, the geographical information corresponding to the base station position, the terminal positioning data between the positioning terminal and each communication satellite, and the base station positioning data between the differential base station and each communication satellite;

[0011] An initial difference module, used to obtain an initial difference between the positioning terminal and each communication satellite according to the terminal positioning data between the positioning terminal and each communication satellite and the base station positioning data between the differential base station and each communication satellite;

[0012] The correction difference module is used to input the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite and the initial difference between the positioning terminal and each communication satellite into a pre-trained correction model to obtain the correction difference between the positioning terminal and each communication satellite;

[0013] The final positioning module is used to obtain the corrected distance between the positioning terminal and each communication satellite based on the corrected difference between the positioning terminal and each communication satellite and the terminal positioning data between the positioning terminal and each communication satellite, and calculate the position of the positioning terminal based on the corrected distance.

[0014] Preferably, the terminal positioning data includes a first pseudorange and a satellite position, and the base station positioning data includes a second pseudorange and a satellite position; wherein the first pseudorange is a pseudorange between the positioning terminal and the satellite, and the second pseudorange is a pseudorange between a differential base station and the satellite, and the geographic information includes a geographic type and a digital elevation model.

[0015] Preferably, the step of obtaining the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite includes:

[0016] S11, obtaining any communication satellite as a target satellite;

[0017] S12, obtaining the satellite position, initial position, geographic information corresponding to the initial position, base station position, and geographic information corresponding to the base station position of the target satellite;

[0018] S13, obtaining terminal positioning data between the target satellite and the positioning terminal, and obtaining base station positioning data between the target satellite and the differential base station;

[0019] S14, extracting a feature array using a spatial feature extraction method according to the initial position, the geographic information corresponding to the initial position, and the satellite position in the terminal positioning data, and using the extracted feature array as the terminal spatial feature;

[0020] S15, extracting a feature array using a spatial feature extraction method according to the base station position, the geographic information corresponding to the base station position, and the satellite position in the base station positioning data, and using the extracted feature array as the base station spatial feature;

[0021] S16, using the feature array corresponding to the terminal spatial feature, subtracting the feature array corresponding to the base station spatial feature to obtain a relative array;

[0022] S17, using the relative array as the relative spatial feature between the target satellite, the positioning terminal and the differential base station;

[0023] S18. Repeat steps S11 to S17 to obtain the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite.

[0024] Preferably, the steps of the spatial feature extraction method include:

[0025] The satellite position is obtained and recorded as the first position, the initial position or the base station position is obtained and recorded as the second position, and the geographic information corresponding to the second position is obtained and recorded as the target geographic information;

[0026] Get the geographic type in the target geographic information, perform one-hot encoding according to the geographic type, and save the encoding in array form as a one-hot array;

[0027] Calculate the propagation vector, the calculation formula of the propagation vector is as follows:

[0028] ;

[0029] In the formula, represents the propagation vector, represents the coordinates of the first position, represents the coordinates of the second position;

[0030] Preset number of sampling points and sampling interval , and make The value is less than the propagation vector Mould length;

[0031] According to the number of sampling points Sampling interval Calculate the coordinates of each sampling point. The coordinate calculation formula of the sampling point is as follows:

[0032] ;

[0033] In the formula, Represents an index, Indicates The coordinates of the sampling points, Representation vector Mould length.

[0034] Traverse each sampling point and obtain the total number of blocked sampling points based on the coordinates of the sampling point and the digital elevation model in the target geographic information ;

[0035] use Get the occlusion distance, insert the value of the occlusion distance into the predetermined position in the one-hot array, and use the one-hot array after the inserted value as the feature array.

[0036] Preferably, the process of traversing each sampling point and obtaining the total number of blocked sampling points according to the coordinates of the sampling point and the digital elevation model in the target geographic information is as follows: The steps include:

[0037] Get the coordinates of the sampling points as ,in Indicates longitude, Indicates latitude, Indicates elevation;

[0038] The longitude and latitude obtained from the digital elevation model are The maximum terrain height at ;

[0039] like , then the sampling point is recorded as an occluded sampling point;

[0040] Traverse all sampling points and count the total number of occluded sampling points .

[0041] Preferably, the step of obtaining the initial difference between the positioning terminal and each communication satellite comprises:

[0042] S21, obtaining any communication satellite as a reference satellite;

[0043] S22, obtaining the first pseudo-range between the reference satellite and the positioning terminal , obtain the second pseudo-range between the reference satellite and the differential base station ;

[0044] S23, obtaining the base station position of the differential base station and the satellite position in the base station positioning data between the reference satellite and the differential base station, and calculating the real distance based on the obtained satellite position and base station position ;

[0045] S24. Calculate the initial difference between the positioning terminal and the reference satellite using the following formula: :

[0046] ;

[0047] S25. Repeat steps S21 to S24 to obtain the initial difference between the positioning terminal and each communication satellite.

[0048] Preferably, the training process of the correction model includes:

[0049] Obtain historical data, process the historical data to obtain a learning data set, and divide the learning data set into a training set and a test set according to a preset ratio;

[0050] Build a machine learning framework and define the input layer and output layer, set the loss function and optimizer, and build a machine learning model;

[0051] Set the batch size and number of training rounds to train the constructed machine learning model using the training set;

[0052] Use the test set to test the trained machine learning model, and output the machine learning model that meets the preset accuracy as the revised model.

[0053] Preferably, the steps of acquiring historical data and processing the historical data to obtain a learning data set include:

[0054] Respectively obtain the base station positioning data between two base stations and the same Beidou satellite at any time in history, and record the two base stations as the first base station and the second base station respectively, and record the two base station positioning data as the first base station positioning data and the second base station positioning data respectively;

[0055] Acquire base station locations of the first base station and the second base station and geographic information corresponding to the base station locations;

[0056] Extracting a feature array using a spatial feature extraction method according to the base station position of the first base station, geographic information corresponding to the base station position, and satellite positions in the first base station positioning data, and using the extracted feature array as a first feature;

[0057] extracting a feature array using a spatial feature extraction method according to the base station position of the second base station, the geographic information corresponding to the base station position, and the satellite position in the second base station positioning data, and using the extracted feature array as a second feature;

[0058] Use the feature array corresponding to the first feature and subtract the feature array corresponding to the second feature to obtain the array as the relative spatial feature;

[0059] The first difference is obtained by subtracting the true distance between the Beidou satellite and the first base station from the pseudo distance between the Beidou satellite and the first base station;

[0060] The second difference is obtained by subtracting the true distance between the Beidou satellite and the second base station from the pseudo distance between the Beidou satellite and the second base station;

[0061] The relative spatial feature, the first difference and the second difference are used to form a piece of learning data, and multiple pieces of learning data are obtained to form a learning data set.

[0062] Preferably, the step of acquiring the corrected distance includes: acquiring a first pseudorange and a corrected difference between the positioning terminal and the communication satellite; and obtaining the corrected distance by subtracting the corrected difference from the first pseudorange.

[0063] Preferably, the step of calculating the position of the positioning terminal according to the corrected distance includes:

[0064] The spherical equation is constructed using the satellite position of each communication satellite as the center of the circle and the corrected distance between the positioning terminal and each communication satellite as the radius;

[0065] The optimization algorithm is used to solve the intersection points of all spherical equations, and the obtained intersection points are used as the position of the positioning terminal.

[0066] The present invention provides a Beidou-based multi-source positioning data fusion processing system, which has the following beneficial effects:

[0067] The positioning data acquisition module simultaneously acquires the positioning data of Beidou satellites and differential base stations, and combines the preliminary positioning and geographic information acquisition modules to provide a comprehensive data basis for subsequent processing. The spatial feature processing module analyzes the relative spatial characteristics between the positioning terminal and the differential base station, providing a data basis for analyzing the impact of complex terrain environments on positioning accuracy. Through the initial differential module and the corrected differential module, the pre-trained machine learning model is used as the correction model to correct the initial differential, obtain a more accurate corrected differential, and then obtain a more accurate positioning result. In addition, the modular design and the application of machine learning methods make the system have good scalability and adaptability, and can continuously optimize positioning performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a module schematic diagram of a Beidou-based multi-source positioning data fusion processing system of the present invention;

[0069] Figure 2 The present invention is a schematic diagram of the relative spatial feature acquisition process of a Beidou-based multi-source positioning data fusion processing system. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] Example 1, please refer to Figure 1 In this embodiment, a Beidou-based multi-source positioning data fusion processing system includes:

[0072] The positioning data acquisition module is used to establish communication between the positioning terminal and multiple Beidou satellites, record the Beidou satellites that successfully establish communication as communication satellites, and obtain terminal positioning data between the positioning terminal and each communication satellite; it is also used to establish communication between the positioning terminal and any differential base station, obtain the base station position of the differential base station that establishes communication and the base station positioning data between the differential base station and each communication satellite;

[0073] The initial positioning module is used to obtain the current position of the positioning terminal through the Beidou positioning system and record it as the initial position;

[0074] A geographic information acquisition module, which acquires geographic information corresponding to the initial position and geographic information corresponding to the base station position;

[0075] The terminal positioning data includes a first pseudorange and a satellite position, and the base station positioning data includes a second pseudorange and a satellite position; wherein the first pseudorange is a pseudorange between the positioning terminal and the satellite, and the second pseudorange is a pseudorange between a differential base station and the satellite, and the geographic information includes a geographic type and a digital elevation model.

[0076] In this embodiment, it should be noted that in order to complete the satellite positioning work, the positioning terminal usually needs to establish communication with at least four satellites. In this implementation, the establishment of communication with four Beidou satellites is taken as an example, and these four satellites are recorded as communication satellites. The terminal positioning data between the positioning terminal and each communication satellite is obtained. Then a differential base station is selected to establish communication between the positioning terminal and it. Usually, this communication is established based on a 4G or 5G network. When selecting a differential base station, it can usually be based on the minimum network delay. The base station positioning data between the differential base station and each communication satellite is obtained.

[0077] Among them, the terminal positioning data includes the first pseudorange and the satellite position, and the base station positioning data includes the second pseudorange and the satellite position. The first pseudorange is the pseudorange between the positioning terminal and the satellite, and the second pseudorange is the pseudorange between the differential base station and the satellite. It should be noted that the satellite position is the position of the satellite during communication. The pseudorange is an important parameter used for positioning calculation in the satellite navigation system. It represents the distance obtained by multiplying the time taken for the satellite signal to be transmitted from the satellite to the receiver by the signal speed. However, since the signal speed is affected by many factors during the propagation process and is not a constant value, this distance is not the true geometric distance between the satellite and the receiver, so it is called a pseudorange.

[0078] In this embodiment, the current position of the positioning terminal can be obtained by using communication base station positioning, pseudo-range direct calculation positioning, etc. The position obtained in this way usually has lower accuracy, but the acquisition speed is usually faster, so it can provide a reference for subsequent geographic information acquisition.

[0079] In this embodiment, the location of the differential base station is usually known and fixed, and the geographic information corresponding to the initial location and the geographic information corresponding to the base station location are obtained; the coordinates of the initial location and the base station location can be obtained in the GIS, i.e., geographic information system. For example, using the GIS system, with the initial location or the base station location as the center, a capture range such as a radius of 1000 meters is set, and within the capture range, the geographic type and digital elevation model are accurately captured. Ensuring the comprehensiveness and pertinence of the data can avoid unnecessary data redundancy and improve the efficiency of subsequent work.

[0080] The spatial feature processing module is used to obtain the relative spatial features between the positioning terminal and the differential base station under each communication satellite according to the geographical information corresponding to the initial position, the geographical information corresponding to the base station position, the terminal positioning data between the positioning terminal and each communication satellite, and the base station positioning data between the differential base station and each communication satellite;

[0081] See also Figure 2 The step of obtaining the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite includes:

[0082] S11, obtaining any communication satellite as a target satellite;

[0083] S12, obtaining the satellite position, initial position, geographic information corresponding to the initial position, base station position, and geographic information corresponding to the base station position of the target satellite;

[0084] S13, obtaining terminal positioning data between the target satellite and the positioning terminal, and obtaining base station positioning data between the target satellite and the differential base station;

[0085] S14, extracting a feature array using a spatial feature extraction method according to the initial position, the geographic information corresponding to the initial position, and the satellite position in the terminal positioning data, and using the extracted feature array as the terminal spatial feature;

[0086] S15, extracting a feature array using a spatial feature extraction method according to the base station position, the geographic information corresponding to the base station position, and the satellite position in the base station positioning data, and using the extracted feature array as the base station spatial feature;

[0087] S16, using the feature array corresponding to the terminal spatial feature, subtracting the feature array corresponding to the base station spatial feature to obtain a relative array;

[0088] S17, using the relative array as the relative spatial feature between the target satellite, the positioning terminal and the differential base station;

[0089] S18. Repeat steps S11 to S17 to obtain the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite.

[0090] The steps of the spatial feature extraction method include:

[0091] The satellite position is obtained and recorded as the first position, the initial position or the base station position is obtained and recorded as the second position, and the geographic information corresponding to the second position is obtained and recorded as the target geographic information;

[0092] Get the geographic type in the target geographic information, perform one-hot encoding according to the geographic type, and save the encoding in array form as a one-hot array;

[0093] Calculate the propagation vector, the calculation formula of the propagation vector is as follows:

[0094] ;

[0095] In the formula, represents the propagation vector, represents the coordinates of the first position, indicating coordinates of a second position;

[0096] Preset number of sampling points and sampling interval , and make The value is less than the propagation vector Mould length;

[0097] According to the number of sampling points Sampling interval Calculate the coordinates of each sampling point. The coordinate calculation formula of the sampling point is as follows:

[0098] ;

[0099] In the formula, Represents an index, Indicates The coordinates of the sampling points, Representation vector Mould length.

[0100] Traverse each sampling point and obtain the total number of blocked sampling points based on the coordinates of the sampling point and the digital elevation model in the target geographic information ;

[0101] use Get the occlusion distance, insert the value of the occlusion distance into the predetermined position in the one-hot array, and use the one-hot array after the inserted value as the feature array.

[0102] The method traverses each sampling point and obtains the total number of blocked sampling points according to the coordinates of the sampling point and the digital elevation model in the target geographic information. The steps include:

[0103] Get the coordinates of the sampling points as ,in Indicates longitude, Indicates latitude, Indicates elevation;

[0104] The longitude and latitude obtained from the digital elevation model are The maximum terrain height at ;

[0105] like , then the sampling point is recorded as an occluded sampling point;

[0106] Traverse all sampling points and count the total number of occluded sampling points .

[0107] In this embodiment, the position information of the target satellite, the initial position of the positioning terminal, the position of the differential base station and the corresponding geographic information are first obtained. Then, through the spatial feature extraction method, feature arrays are generated for the positioning terminal and the differential base station respectively. These feature arrays contain two key information: the unique hot coding of the geographical type and the blocking distance on the signal propagation path. The unique hot coding of the geographical type reflects the type of the main obstruction on the signal propagation path, such as mountains, high-rise buildings, etc. This encoding method can intuitively represent the potential impact of different types of obstructions on signal propagation. By presetting the sampling points and spacing, such as 100 points with a spacing of 20 meters, sampling is performed along the signal propagation path, and it is determined whether each sampling point is blocked, and finally the blocking distance is obtained. The blocking distance is inserted into a predetermined position in the unique hot coding array, such as the first position, to form a comprehensive feature array. In this way, it contains both the type information of the obstruction and the quantitative data of the blocking distance, which can fully reflect the blocking situation on the signal propagation path.

[0108] By calculating the spatial features of the positioning terminal and the differential base station, and then taking the difference, the relative spatial features are obtained. This relative spatial feature can effectively reflect the difference in the occlusion of the two under the same satellite. In addition, the relative spatial features obtained by this method can provide a basis for subsequent positioning algorithms. These features can be used to establish a machine learning model to learn the influence of obstructions on satellite signal propagation, so as to make more accurate corrections to the data during the actual positioning process.

[0109] An initial difference module, used to obtain an initial difference between the positioning terminal and each communication satellite according to the terminal positioning data between the positioning terminal and each communication satellite and the base station positioning data between the differential base station and each communication satellite;

[0110] The step of obtaining the initial difference between the positioning terminal and each communication satellite comprises:

[0111] S21, obtaining any communication satellite as a reference satellite;

[0112] S22, obtaining the first pseudo-range between the reference satellite and the positioning terminal , obtain the second pseudo-range between the reference satellite and the differential base station ;

[0113] S23, obtaining the base station position of the differential base station and the satellite position in the base station positioning data between the reference satellite and the differential base station, and calculating the real distance based on the obtained satellite position and base station position ;

[0114] S24. Calculate the initial difference between the positioning terminal and the reference satellite using the following formula: :

[0115] ;

[0116] S25. Repeat steps S21 to S24 to obtain the initial difference between the positioning terminal and each communication satellite.

[0117] In this embodiment, a communication satellite is first selected as a reference satellite. For the reference satellite, pseudoranges between it and the positioning terminal and the differential base station, namely, the first pseudorange and the second pseudorange, are obtained respectively. These pseudorange values ​​contain various error factors in the signal propagation process.

[0118] Subsequently, the precise position of the differential base station and the position information of the reference satellite are obtained. Using these two known precise positions, the true geometric distance between the differential base station and the reference satellite is calculated. Based on these data, the initial differential is calculated using a specific formula.

[0119] This initial difference reflects the comprehensive error in the propagation process of the reference satellite signal. However, although the differential base station and the positioning terminal are usually located in relatively close areas and share similar atmospheric conditions, resulting in similar climate-related errors, there may be significant differences in the signal propagation paths between the two and the reference satellite. These differences mainly come from different terrain features, such as physical obstacles such as mountain occlusion or building obstruction. Therefore, although the initial difference provides a valuable error estimate, it fails to fully consider the different parts of the signal propagation path between the positioning terminal and the reference satellite, that is, the path occlusion, so using it as the positioning terminal error may result in the final positioning being inaccurate.

[0120] The correction difference module is used to input the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite and the initial difference between the positioning terminal and each communication satellite into a pre-trained correction model to obtain the correction difference between the positioning terminal and each communication satellite;

[0121] The training process of the correction model includes:

[0122] Obtain historical data, process the historical data to obtain a learning data set, and divide the learning data set into a training set and a test set according to a preset ratio;

[0123] Build a machine learning framework and define the input layer and output layer, set the loss function and optimizer, and build a machine learning model;

[0124] Set the batch size and number of training rounds to train the constructed machine learning model using the training set;

[0125] Use the test set to test the trained machine learning model, and output the machine learning model that meets the preset accuracy as the revised model.

[0126] The steps of obtaining historical data and processing the historical data to obtain a learning data set include:

[0127] Respectively obtain the base station positioning data between two base stations and the same Beidou satellite at any time in history, and record the two base stations as the first base station and the second base station respectively, and record the two base station positioning data as the first base station positioning data and the second base station positioning data respectively;

[0128] Acquire base station locations of the first base station and the second base station and geographic information corresponding to the base station locations;

[0129] Extracting a feature array using a spatial feature extraction method according to the base station position of the first base station, geographic information corresponding to the base station position, and satellite positions in the first base station positioning data, and using the extracted feature array as a first feature;

[0130] extracting a feature array using a spatial feature extraction method according to the base station position of the second base station, the geographic information corresponding to the base station position, and the satellite position in the second base station positioning data, and using the extracted feature array as a second feature;

[0131] Use the feature array corresponding to the first feature and subtract the feature array corresponding to the second feature to obtain the array as the relative spatial feature;

[0132] The first difference is obtained by subtracting the true distance between the Beidou satellite and the first base station from the pseudo distance between the Beidou satellite and the first base station;

[0133] The second difference is obtained by subtracting the true distance between the Beidou satellite and the second base station from the pseudo distance between the Beidou satellite and the second base station;

[0134] The relative spatial feature, the first difference and the second difference are used to form a piece of learning data, and multiple pieces of learning data are obtained to form a learning data set.

[0135] In this embodiment, when building a machine learning framework, the TensorFlow framework can be used to build a neural network model. The model structure includes an input layer, two hidden layers, and an output layer. The input layer receives relative spatial features and the first difference; the output layer generates 1 value corresponding to the second difference. The hidden layer uses the ReLU activation function to enhance the nonlinear expression ability of the model. For the loss function, the mean square error (MSE) is selected to minimize the square difference between the predicted value and the actual value. The optimizer selects the Adam algorithm, which can adaptively adjust the learning rate, which is conducive to the rapid convergence of the model. When compiling the model, you can also set the mean absolute error (MAE) as an additional evaluation indicator.

[0136] During the training process, the batch size can be set to 32 and the number of training rounds can be set to 100. These hyperparameters can be adjusted according to the actual situation to achieve the best training effect. During training, 80% of the training set can be used for training, and the remaining 20% ​​can be used for validation to monitor whether the model is overfitting. After the training is completed, the previously divided test set can be used to evaluate the model performance. If the preset accuracy requirement is that the mean absolute error (MAE) is less than 0.05, if the test result meets this condition, the model will be used as the final correction model. The model can calculate the corrected difference based on the relative spatial features of the input and the initial difference. The corrected difference fully considers the different parts of the signal propagation path between the positioning terminal and the reference satellite, that is, the path obstruction, based on the initial difference. Therefore, using the corrected difference as the positioning terminal error can obtain more accurate positioning data.

[0137] The final positioning module is used to obtain the corrected distance between the positioning terminal and each communication satellite based on the corrected difference between the positioning terminal and each communication satellite and the terminal positioning data between the positioning terminal and each communication satellite, and calculate the position of the positioning terminal based on the corrected distance.

[0138] The step of acquiring the corrected distance includes: acquiring a first pseudorange and a corrected difference between the positioning terminal and the communication satellite; and obtaining the corrected distance by subtracting the corrected difference from the first pseudorange.

[0139] In this embodiment, the corrected distance obtained by the above method fully considers various error conditions, so that the error between the corrected distance and the true distance is smaller, thereby making the positioning information calculated based on the corrected distance more accurate.

[0140] Embodiment 2, based on embodiment 1, the step of calculating the position of the positioning terminal according to the corrected distance includes:

[0141] The spherical equation is constructed using the satellite position of each communication satellite as the center of the circle and the corrected distance between the positioning terminal and each communication satellite as the radius;

[0142] The optimization algorithm is used to solve the intersection points of all spherical equations, and the obtained intersection points are used as the position of the positioning terminal.

[0143] In this embodiment, first, the precise position information of each communication satellite is obtained, and the previously calculated corrected distance is used as the radius. A series of spherical equations are constructed with each satellite position as the center and the corrected distance as the radius. To solve this set of equations, a nonlinear least squares optimization algorithm, such as the Levenberg-Marquardt algorithm, can be used. Through iterative optimization, a point is found that minimizes the sum of squares of all spherical equation errors. The position of the positioning terminal is finally output, providing a high-precision three-dimensional positioning result. By combining the information of multiple satellites and the corrected distance data, the positioning accuracy can be effectively improved.

[0144] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0145] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0146] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A Beidou-based multi-source positioning data fusion processing system, characterized in that: include: The positioning data acquisition module is used to establish communication between the positioning terminal and multiple Beidou satellites, record the Beidou satellites that successfully establish communication as communication satellites, and obtain terminal positioning data between the positioning terminal and each communication satellite; it is also used to establish communication between the positioning terminal and any differential base station, obtain the base station position of the differential base station that establishes communication and the base station positioning data between the differential base station and each communication satellite; The initial positioning module is used to obtain the current position of the positioning terminal through the Beidou positioning system and record it as the initial position; A geographic information acquisition module, which acquires geographic information corresponding to the initial position and geographic information corresponding to the base station position; The spatial feature processing module is used to obtain the relative spatial features between the positioning terminal and the differential base station under each communication satellite according to the geographical information corresponding to the initial position, the geographical information corresponding to the base station position, the terminal positioning data between the positioning terminal and each communication satellite, and the base station positioning data between the differential base station and each communication satellite; An initial difference module, used to obtain an initial difference between the positioning terminal and each communication satellite according to the terminal positioning data between the positioning terminal and each communication satellite and the base station positioning data between the differential base station and each communication satellite; The correction difference module is used to input the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite and the initial difference between the positioning terminal and each communication satellite into a pre-trained correction model to obtain the correction difference between the positioning terminal and each communication satellite; The final positioning module is used to obtain the corrected distance between the positioning terminal and each communication satellite based on the corrected difference between the positioning terminal and each communication satellite and the terminal positioning data between the positioning terminal and each communication satellite, and calculate the position of the positioning terminal based on the corrected distance.

2. The Beidou-based multi-source positioning data fusion processing system according to claim 1 is characterized in that: The terminal positioning data includes a first pseudorange and a satellite position, and the base station positioning data includes a second pseudorange and a satellite position; wherein the first pseudorange is a pseudorange between the positioning terminal and the satellite, and the second pseudorange is a pseudorange between a differential base station and the satellite, and the geographic information includes a geographic type and a digital elevation model.

3. The Beidou-based multi-source positioning data fusion processing system according to claim 2 is characterized in that: The step of obtaining the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite comprises: S11, obtaining any communication satellite as a target satellite; S12, obtaining the satellite position, initial position, geographic information corresponding to the initial position, base station position, and geographic information corresponding to the base station position of the target satellite; S13, obtaining terminal positioning data between the target satellite and the positioning terminal, and obtaining base station positioning data between the target satellite and the differential base station; S14, extracting a feature array using a spatial feature extraction method according to the initial position, the geographic information corresponding to the initial position, and the satellite position in the terminal positioning data, and using the extracted feature array as the terminal spatial feature; S15, extracting a feature array using a spatial feature extraction method according to the base station position, the geographic information corresponding to the base station position, and the satellite position in the base station positioning data, and using the extracted feature array as the base station spatial feature; S16, using the feature array corresponding to the terminal spatial feature, subtracting the feature array corresponding to the base station spatial feature to obtain a relative array; S17, using the relative array as the relative spatial feature between the target satellite, the positioning terminal and the differential base station; S18. Repeat steps S11 to S17 to obtain the relative spatial characteristics between the positioning terminal and the differential base station under each communication satellite.

4. The Beidou-based multi-source positioning data fusion processing system according to claim 3 is characterized in that: The steps of the spatial feature extraction method include: The satellite position is obtained and recorded as the first position, the initial position or the base station position is obtained and recorded as the second position, and the geographic information corresponding to the second position is obtained and recorded as the target geographic information; Get the geographic type in the target geographic information, perform one-hot encoding according to the geographic type, and save the encoding in array form as a one-hot array; Calculate the propagation vector, the calculation formula of the propagation vector is as follows: ； In the formula, represents the propagation vector, represents the coordinates of the first position, represents the coordinates of the second position; Preset number of sampling points and sampling interval , and make The value is less than the propagation vector Mould length; According to the number of sampling points Sampling interval Calculate the coordinates of each sampling point. The coordinate calculation formula of the sampling point is as follows: ； In the formula, Represents an index, Indicates The coordinates of the sampling points, Representation vector Mould length; Traverse each sampling point and obtain the total number of blocked sampling points based on the coordinates of the sampling point and the digital elevation model in the target geographic information ; use Get the occlusion distance, insert the value of the occlusion distance into the predetermined position in the one-hot array, and use the one-hot array after the inserted value as the feature array.

5. The Beidou-based multi-source positioning data fusion processing system according to claim 4 is characterized in that: The method traverses each sampling point and obtains the total number of blocked sampling points according to the coordinates of the sampling point and the digital elevation model in the target geographic information. The steps include: Get the coordinates of the sampling points as ,in Indicates longitude, Indicates latitude, Indicates elevation; The longitude and latitude obtained from the digital elevation model are The maximum terrain height at ; like , then the sampling point is recorded as an occluded sampling point; Traverse all sampling points and count the total number of occluded sampling points .

6. The Beidou-based multi-source positioning data fusion processing system according to claim 5, characterized in that: The step of obtaining the initial difference between the positioning terminal and each communication satellite comprises: S21, obtaining any communication satellite as a reference satellite; S22, obtaining the first pseudo-range between the reference satellite and the positioning terminal , obtain the second pseudo-range between the reference satellite and the differential base station ; S23, obtaining the base station position of the differential base station and the satellite position in the base station positioning data between the reference satellite and the differential base station, and calculating the real distance based on the obtained satellite position and base station position ; S24. Calculate the initial difference between the positioning terminal and the reference satellite using the following formula: : ； S25. Repeat steps S21 to S24 to obtain the initial difference between the positioning terminal and each communication satellite.

7. The Beidou-based multi-source positioning data fusion processing system according to claim 6, characterized in that: The training process of the correction model includes: Obtain historical data, process the historical data to obtain a learning data set, and divide the learning data set into a training set and a test set according to a preset ratio; Build a machine learning framework and define the input layer and output layer, set the loss function and optimizer, and build a machine learning model; Set the batch size and number of training rounds to train the constructed machine learning model using the training set; Use the test set to test the trained machine learning model, and output the machine learning model that meets the preset accuracy as the revised model.

8. The Beidou-based multi-source positioning data fusion processing system according to claim 7, characterized in that: The steps of obtaining historical data and processing the historical data to obtain a learning data set include: Respectively obtain the base station positioning data between two base stations and the same Beidou satellite at any time in history, and record the two base stations as the first base station and the second base station respectively, and record the two base station positioning data as the first base station positioning data and the second base station positioning data respectively; Acquire base station locations of the first base station and the second base station and geographic information corresponding to the base station locations; Extracting a feature array using a spatial feature extraction method according to the base station position of the first base station, geographic information corresponding to the base station position, and satellite positions in the first base station positioning data, and using the extracted feature array as a first feature; extracting a feature array using a spatial feature extraction method according to the base station position of the second base station, the geographic information corresponding to the base station position, and the satellite position in the second base station positioning data, and using the extracted feature array as a second feature; Use the feature array corresponding to the first feature and subtract the feature array corresponding to the second feature to obtain the array as the relative spatial feature; The first difference is obtained by subtracting the true distance between the Beidou satellite and the first base station from the pseudo distance between the Beidou satellite and the first base station; The second difference is obtained by subtracting the true distance between the Beidou satellite and the second base station from the pseudo distance between the Beidou satellite and the second base station; The relative spatial feature, the first difference and the second difference are used to form a piece of learning data, and multiple pieces of learning data are obtained to form a learning data set.

9. The Beidou-based multi-source positioning data fusion processing system according to claim 8, characterized in that: The step of acquiring the corrected distance includes: acquiring a first pseudorange and a corrected difference between the positioning terminal and the communication satellite; and obtaining the corrected distance by subtracting the corrected difference from the first pseudorange.

10. The Beidou-based multi-source positioning data fusion processing system according to claim 9, characterized in that: The step of calculating the position of the positioning terminal according to the corrected distance comprises: The spherical equation is constructed using the satellite position of each communication satellite as the center of the circle and the corrected distance between the positioning terminal and each communication satellite as the radius; The optimization algorithm is used to solve the intersection points of all spherical equations, and the obtained intersection points are used as the position of the positioning terminal.

Images