A train intelligent positioning method, system, device, medium and computer program product based on multi-source fusion

Abstract

The application provides a train intelligent positioning method, system, device, medium and computer program product based on multi-source fusion, relates to the technical field of rail transit, and comprises the following steps: in response to a positioning request of a train, first positioning source data generated by a first positioning source and second positioning source data generated by a second positioning source are acquired under a current running environment of the train; the first positioning source data and the second positioning source data are respectively subjected to filtering processing to obtain first corrected data and second corrected data; signal quality parameters of the first positioning source and the second positioning source are acquired; and based on the signal quality parameters, the first corrected data and the second corrected data are subjected to fusion processing to obtain a target positioning result of the train. The application solves the problem that existing railway trains mainly rely on single track circuit positioning, but the track circuit has the disadvantages of high wiring maintenance cost and difficult fault troubleshooting, so that the train is difficult to realize accurate positioning, and has the effect of improving the positioning accuracy of the train.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular to a train intelligent positioning method, system, device, medium, and computer program product based on multi-source fusion. Background Technology

[0002] As train speeds and operating density increase, dispatch centers need to obtain the precise location of trains in real time to ensure operational safety and dispatch efficiency, which places higher demands on railway train positioning systems.

[0003] Currently, railway train positioning mainly relies on a single track circuit. The track circuit determines the section where the train is located by detecting the occupancy status of the rails. When the train wheels enter a certain section, they short-circuit the track circuit of that section. Ground equipment can then determine whether the section is occupied by a train based on the circuit's on / off status.

[0004] However, track circuit positioning has significant limitations: firstly, track circuits require extensive wiring along the railway line, resulting in high construction costs, and their long-term exposure to the outdoor environment makes fault diagnosis and maintenance difficult; secondly, track circuits can only determine the section where the train is located, and cannot provide continuous and precise train position. Therefore, these limitations make it difficult for existing positioning methods to meet the demands for higher-precision train positioning. Summary of the Invention

[0005] In view of the above-mentioned shortcomings in the existing technology, the purpose of this invention is to provide a train intelligent positioning method based on multi-source fusion, which has the feature of improving the accuracy of train positioning.

[0006] The above-mentioned objective of this invention is achieved through the following technical solution:

[0007] A train intelligent positioning method based on multi-source fusion includes:

[0008] In response to the train's positioning request, the system acquires first positioning source data generated by the first positioning source and second positioning source data generated by the second positioning source under the current operating environment of the train.

[0009] The first positioning source data and the second positioning source data are filtered respectively to obtain the first corrected data and the second corrected data.

[0010] The signal quality parameters of the first positioning source and the second positioning source are obtained, and the first corrected data and the second corrected data are fused based on the signal quality parameters to obtain the target positioning result of the train.

[0011] By adopting the above technical solution, data from two independent positioning sources, Beidou and 5G, are integrated. By filtering the data from the two positioning sources and weighting the data based on signal quality parameters, continuous and accurate position information can be output. This effectively overcomes the shortcomings of track circuits, which can only determine section occupancy and have high maintenance costs, and improves the accuracy and reliability of train positioning.

[0012] Preferably, the specific method for obtaining the first positioning source data generated by the first positioning source and the second positioning source data generated by the second positioning source under the current operating environment of the train is as follows:

[0013] The train receives raw observation data broadcast by the BeiDou satellite system via its BeiDou satellite navigation module, which serves as the first positioning source data, wherein the BeiDou satellite system is the first positioning source.

[0014] The train receives positioning reference signals from surrounding base stations via a 5G-R communication module and extracts positioning measurement data from the positioning reference signals as the second positioning source data, wherein the surrounding base stations are the second positioning source.

[0015] By adopting the above technical solution, the BeiDou satellite system is clearly used as the primary positioning source and the 5G communication base station is used as the secondary positioning source to obtain the original observation data and the measurement data in the positioning reference signal, thus eliminating the dependence on a single orbital circuit.

[0016] Preferably, the step of filtering the first positioning source data and the second positioning source data respectively to obtain the first corrected data and the second corrected data includes:

[0017] The first positioning source data is filtered by a preset BeiDou RTK local filter to obtain the first corrected data.

[0018] The second positioning source data is subjected to a second filtering process by a preset 5G assisted positioning local filter to obtain the second corrected data.

[0019] By adopting the above technical solutions, the BeiDou RTK local filter and the 5G assisted positioning local filter are used to perform targeted and independent filtering on the two types of positioning source data, which can effectively filter out noise and errors in their respective data and obtain higher quality corrected data.

[0020] Preferably, the step of obtaining the signal quality parameters of the first positioning source and the second positioning source, and fusing the first corrected data and the second corrected data based on the signal quality parameters to obtain the target positioning result of the train includes:

[0021] Based on the BeiDou satellite navigation module on the train, the number of visible satellites is obtained and used as the first signal quality parameter of the first positioning source.

[0022] Based on the 5G-R communication module on the train, the 5G signal strength is obtained and used as the second signal quality parameter of the second positioning source.

[0023] The first fusion weight corresponding to the first corrected data is determined based on the first signal quality parameter, and the second fusion weight corresponding to the second corrected data is determined based on the second signal quality parameter.

[0024] Calculate the first product of the first corrected data and the first fusion weight, and calculate the second product of the second corrected data and the second fusion weight;

[0025] The first product and the second product are summed to obtain the target positioning result of the train.

[0026] By adopting the above technical solution, the fusion weight of the two positioning data is dynamically determined by using the number of visible satellites and the strength of 5G signal as signal quality parameters. The weighted corrected data is then summed, which can adaptively adjust the fusion weight according to the signal quality in the current environment. This allows the optimal positioning result to be obtained even in complex environments, further improving the accuracy and environmental adaptability of train positioning.

[0027] Preferably, determining the first fusion weight corresponding to the first corrected data based on the first signal quality parameter, and determining the second fusion weight corresponding to the second corrected data based on the second signal quality parameter, includes:

[0028] The first signal quality parameter and the second signal quality parameter are normalized respectively to obtain a standardized first relative quality index and a standardized second relative quality index.

[0029] Calculate the sum of the first relative quality index and the second relative quality index to obtain the quality assessment benchmark value;

[0030] Divide the first relative quality index by the quality assessment benchmark value to obtain the first fusion weight;

[0031] The second relative quality index is divided by the quality assessment benchmark value to obtain the second fusion weight.

[0032] By adopting the above technical solution, the number of visible satellites and the strength of 5G signals are normalized, and the fusion weight is dynamically allocated based on the sum of the two. This makes the weights of the two positioning data adaptively adjusted with signal quality and the sum always equal to 1, thereby ensuring the rationality of the fusion result, avoiding positioning deviation caused by a sudden drop in the quality of a certain signal, and improving the robustness of fused positioning.

[0033] Preferably, before obtaining the signal quality parameters of the first positioning source and the second positioning source, and fusing the first corrected data and the second corrected data based on the signal quality parameters to obtain the target positioning result of the train, the method further includes:

[0034] Residual analysis was performed on the first corrected data and the second corrected data respectively to obtain the corresponding first residual value and second residual value;

[0035] The first residual value and the second residual value are compared with the corresponding preset residual thresholds to obtain the first comparison result and the second comparison result.

[0036] When the first comparison result is abnormal, the first corrected data is hidden, and the second corrected data is used as the target location result;

[0037] When the second comparison result is abnormal, the second correction data is masked, and the first correction data is used as the target location result.

[0038] When both the first comparison result and the second comparison result are normal, the step of obtaining the signal quality parameters of the first positioning source and the second positioning source, and fusing the first corrected data and the second corrected data based on the signal quality parameters to obtain the target positioning result of the train is executed.

[0039] By adopting the above technical solution, residual analysis and threshold comparison are performed on the two types of corrected data before weighted fusion. This can effectively identify and shield abnormal data sources, ensuring that only normal data participates in the fusion or that a single normal data is used directly as the positioning result. This avoids the contamination of the fusion result by abnormal data and further improves the reliability and safety of train positioning.

[0040] The second objective of this invention is to provide a train intelligent positioning system based on multi-source fusion, which has the characteristic of improving the accuracy of train positioning.

[0041] The second objective of this invention is achieved through the following technical solution:

[0042] A train intelligent positioning system based on multi-source fusion includes: a data acquisition module, used to respond to the positioning request of the train and acquire first positioning source data generated by a first positioning source and second positioning source data generated by a second positioning source in the current operating environment of the train;

[0043] The filtering module is used to filter the first positioning source data and the second positioning source data respectively to obtain the first corrected data and the second corrected data.

[0044] The fusion positioning module is used to acquire the signal quality parameters of the first positioning source and the second positioning source, and based on the signal quality parameters, to perform fusion processing on the first corrected data and the second corrected data to obtain the target positioning result of the train.

[0045] By adopting the above technical solutions, each component works in concert to complete the acquisition of multi-source data, targeted filtering, and adaptive fusion based on signal quality parameters. This enables the output of continuous and accurate train position information, effectively overcoming the inherent drawbacks of track circuit positioning and improving the accuracy and reliability of train positioning.

[0046] The third objective of this invention is to provide an electronic device that improves the accuracy of train positioning.

[0047] The above-mentioned objective three of this invention is achieved through the following technical solution:

[0048] An electronic device includes a memory and a processor, wherein the memory stores a computer program for a train intelligent positioning method based on multi-source fusion that can be loaded by the processor and executed by the processor.

[0049] The fourth objective of this invention is to provide a computer-readable storage medium capable of storing corresponding programs, which facilitates the improvement of train positioning accuracy.

[0050] The fourth objective of this invention is achieved through the following technical solution:

[0051] A computer-readable storage medium storing a computer program for a train intelligent positioning method based on multi-source fusion, which can be loaded by a processor and executed by any of the above-mentioned methods.

[0052] The fifth objective of this invention is to provide a computer program product that facilitates the improvement of train positioning accuracy.

[0053] The fifth objective of this invention is achieved through the following technical solution:

[0054] A computer program product, a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, wherein when the program instructions are executed by a computer, the computer executes the computer program that performs any of the above-mentioned multi-source fusion-based intelligent train positioning methods.

[0055] In summary, the present invention has at least one of the following beneficial technical effects:

[0056] 1. By integrating two independent data sources, BeiDou satellite positioning and 5G base station positioning, and using filtering and weighted fusion processing based on signal quality parameters, continuous and accurate location information can be output. This effectively overcomes the shortcomings of traditional track circuits, which can only determine section occupancy, cannot provide continuous positioning, and have high maintenance costs, thus improving the accuracy and reliability of train positioning.

[0057] 2. By introducing a residual analysis mechanism, anomaly detection and masking are performed on the two types of corrected data before fusion, ensuring that only normal data participates in the fusion, avoiding the contamination of positioning results by abnormal data, and further improving the safety and robustness of train positioning;

[0058] 3. By dynamically allocating fusion weights using satellite visibility and 5G signal strength as signal quality parameters, the fusion strategy can be adaptively adjusted according to the signal quality in the current environment, thus achieving stable positioning results even in complex environments such as tunnels and mountainous areas, enhancing environmental adaptability. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating the steps of a train intelligent positioning method based on multi-source fusion provided in Embodiment 1 of the present invention.

[0060] Figure 2 This is a structural block diagram of a train intelligent positioning system based on multi-source fusion provided in Embodiment 2 of the present invention. Detailed Implementation

[0061] This invention provides a train intelligent positioning method, system, device, medium, and computer program product based on multi-source fusion. It addresses the technical problem that existing railway train positioning primarily relies on a single track circuit, which suffers from high wiring and maintenance costs and difficulties in fault diagnosis, hindering accurate train positioning. The invention improves the accuracy of train positioning.

[0062] It should be noted that the embodiments of the present invention are applied to a railway intelligent module platform, which includes a Beidou satellite navigation module, a 5G-R communication module, a data fusion module, a core control module, a power supply module, and an interface module.

[0063] The BeiDou satellite navigation module includes: a BeiDou dual-mode positioning chip, supporting dual-mode operation of BeiDou-2 and BeiDou-3; an RTK (Real-Time Kinematic) ground-based augmentation interface, used to receive differential correction signals sent by ground-based augmentation stations; an anti-multipath interference antenna, used to reduce reflection interference from buildings, overhead contact lines, etc. along railway lines to satellite signals; and a short message communication unit, used to achieve emergency data transmission when 5G-R communication is interrupted.

[0064] The 5G-R communication module includes: a railway-specific 5G-R (railway fifth-generation mobile communication) module chip that supports network slicing and edge computing adaptation; a power amplifier to enhance uplink signal transmission power; an anti-interference filter to suppress out-of-band interference in the strong electromagnetic environment of railways; and a data encryption unit for encrypting and transmitting dispatching instructions and positioning data.

[0065] The data fusion module includes: a BeiDou RTK local filter, used for error correction and filtering of the raw BeiDou observation data; a 5G-assisted positioning local filter, used for filtering of 5G-R positioning measurement data; and a main filter, which adopts an improved federated Kalman filter algorithm, adaptively allocates fusion weights based on signal quality parameters, and outputs the fused target positioning result.

[0066] The core control module includes: a high-performance microcontroller as the main control chip; a large-capacity flash memory and random access memory for storing positioning data, communication logs, and device status data, with a storage time of no less than 30 days; a real-time clock for providing accurate timestamps for all data; and a fault diagnosis unit for real-time monitoring of the working status of each module and triggering alarms or redundancy switching.

[0067] The power supply module and interface module include: a wide voltage input power converter adapted to the DC power supply system of railway trains; overvoltage, overcurrent and short circuit protection circuits; a lithium battery backup power supply that automatically switches when power is lost, with a battery life of no less than 2 hours; a low power management unit used to control each module to enter sleep mode when there is no task; and an interface module including an RS485 interface, an Ethernet interface and a USB interface. Each interface has electromagnetic interference resistance and is used to connect to train on-board equipment, track monitoring equipment and dispatch center terminal.

[0068] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0069] It should be noted that in this embodiment of the invention, all content involving object data must be obtained with the object's authorization and consent, and must comply with current laws and standards. If the embodiment involves personal information, it must ensure that the individual's consent has been obtained; if it involves sensitive information, the separate consent of the information subject must be obtained. The implementation of the entire embodiment should also be based on the object's authorization and consent.

[0070] It should be noted that the terms "first," "second," etc., used in this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this disclosure.

[0071] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0072] Example 1:

[0073] Please see Figure 1 The present invention provides a train intelligent positioning method based on multi-source fusion, comprising:

[0074] Step 101: In response to the train's positioning request, obtain the first positioning source data generated by the first positioning source and the second positioning source data generated by the second positioning source under the current operating environment of the train.

[0075] In response to a train's positioning request, a positioning data acquisition and processing process is initiated in response to the train's power-on, positioning instructions issued by the dispatch center, or preset periodic positioning trigger events.

[0076] The current operating environment of the train refers to the real-time external scene in which the train is located during operation, including but not limited to different geographical and signal propagation conditions such as open areas, tunnels, mountainous areas, urban canyons, multipath interference areas, and areas with strong electromagnetic interference.

[0077] The primary positioning source refers to the BeiDou Navigation Satellite System (BDS) as the main source of satellite positioning information, used to provide raw observation data for the train.

[0078] The second positioning source refers to the 5G-R communication base station as an auxiliary positioning information source. It receives the positioning reference signal (PRS) sent by the base station and extracts measurement data such as the time difference between the base station and the signal angle of arrival and signal strength for auxiliary positioning.

[0079] The first positioning source data refers to the raw observation data broadcast by the BeiDou satellite system and received by the BeiDou satellite navigation module on the train. This includes, but is not limited to, pseudorange observations, carrier phase observations, Doppler shift, satellite ephemeris, satellite number, signal strength, and timestamp information, which are used to calculate the train's initial position, speed, and attitude.

[0080] The second positioning source data refers to the positioning reference signals received by the 5G-R communication module on the train from surrounding base stations, and the positioning measurement data extracted from these signals, including but not limited to parameters such as reference signal time difference, angle of arrival, and signal strength.

[0081] Preferably, the specific method for obtaining the first positioning source data generated by the first positioning source and the second positioning source data generated by the second positioning source under the current operating environment of the train is as follows:

[0082] S11. The train receives raw observation data broadcast by the BeiDou satellite system via the BeiDou satellite navigation module, which serves as the primary positioning source data, with the BeiDou satellite system as the primary positioning source.

[0083] Raw observation data refers to the low-level measurement information directly output by the BeiDou satellite navigation module after down-converting, sampling, and demodulating the received broadcast signals from the BeiDou satellite system.

[0084] In this embodiment of the invention, the Beidou satellite navigation module installed on the top of the train receives L-band navigation signals broadcast by the Beidou satellite system within the visible range in real time. The module supports dual-mode operation of Beidou-2 and Beidou-3 and integrates an anti-multipath interference antenna, which can effectively reduce the reflection interference of the contact network, bridges and mountains along the railway to the satellite signals. After down-converting, sampling and demodulating the received radio frequency signals, the module outputs the original observation data.

[0085] S12. Receive positioning reference signals from surrounding base stations via the 5G-R communication module on the train, and extract positioning measurement data from the positioning reference signals as the second positioning source data, wherein the surrounding base stations are the second positioning source.

[0086] The positioning reference signal refers to a dedicated downlink signal that is periodically broadcast by a 5G-R base station according to a preset time-frequency resource grid, and is used by terminal equipment for positioning measurement.

[0087] In this embodiment of the invention, the positioning reference signals broadcast by the surrounding 5G-R base stations are monitored in real time by the 5G-R communication module on the train. The module adopts the railway-dedicated 5G-R frequency band, supports network slicing technology, and can keep synchronized with multiple adjacent base stations at the same time. When the module receives the positioning reference signals sent by each base station according to the preset time and frequency resources, it first performs synchronization acquisition, channel estimation and interference suppression processing on the signals, and then extracts the positioning measurement data from them.

[0088] Step 102: Filter the first positioning source data and the second positioning source data respectively to obtain the first corrected data and the second corrected data.

[0089] Preferably, step 102 may include the following sub-steps:

[0090] S21. The first positioning source data is subjected to a first filtering process through a preset BeiDou RTK local filter to obtain the first corrected data.

[0091] The preset BeiDou RTK (Real-Time Kinematic) local filter refers to the data processing unit that is pre-configured inside the data fusion module and uses an improved federated Kalman filter algorithm.

[0092] The first correction data refers to the train positioning result output by the Beidou RTK local filter after error correction and filtering smoothing, specifically including the three-dimensional position coordinates of the train in the station center coordinate system, the three-dimensional velocity components, and the corresponding timestamp information.

[0093] In this embodiment of the invention, a BeiDou RTK local filter pre-configured within the data fusion module receives first positioning source data from the BeiDou satellite navigation module. The filter first extracts broadcast ephemeris from the first positioning source data and calculates the real-time three-dimensional spatial position of each visible satellite. Simultaneously, it extracts pseudorange observations, carrier phase observations, and Doppler frequency shift. The filter uses the train's three-dimensional position, three-dimensional velocity, and receiver clock error in the station-centric coordinate system as the local state vector, and uses the pseudorange observations and carrier phase observations as observation inputs to construct pseudorange observation equations and carrier phase observation equations. This serves as a local filter of the improved federated Kalman filter. This filter independently runs the prediction and update steps of the extended Kalman filter algorithm: the prediction step predicts the prior estimate of the local state and the covariance matrix at the current moment based on the local state estimate and the train motion model at the previous moment; the update step calculates the Kalman gain using the observations at the current moment, performs weighted correction on the prior estimate of the local state, and obtains the posterior estimate of the local state. The filter outputs the calculated three-dimensional position coordinates, three-dimensional velocity components, and corresponding timestamps as the first corrected data.

[0094] S22. The second positioning source data is subjected to a second filtering process through a preset 5G assisted positioning local filter to obtain the second corrected data.

[0095] The preset 5G assisted positioning local filter refers to the data processing unit that is pre-configured inside the data fusion module and uses an improved federated Kalman filtering algorithm.

[0096] The second correction data refers to the train assisted positioning result output by the 5G assisted positioning local filter after filtering and smoothing, specifically including the three-dimensional position coordinates of the train in the station center coordinate system, the three-dimensional velocity components, and the corresponding timestamp information.

[0097] In this embodiment of the invention, a 5G-assisted positioning local filter pre-configured within the data fusion module receives second positioning source data from the 5G-R communication module. The filter first extracts the unique identifiers and coordinate information of each surrounding base station from the second positioning source data, and simultaneously extracts positioning measurement parameters such as reference signal time difference, angle of arrival, and signal strength. The filter uses the train's three-dimensional position, three-dimensional velocity, and receiver clock error in the station center coordinate system as the local state vector, and the reference signal time difference and angle of arrival as the observation input to construct an observation equation based on the time difference of arrival and angle of arrival. This serves as a local filter of the improved federated Kalman filter. The filter independently runs the prediction and update steps of the extended Kalman filter algorithm: the prediction step predicts the prior estimate of the local state and the covariance matrix at the current moment based on the local state estimate and train motion model at the previous moment; the update step calculates the Kalman gain using the observation value at the current moment, performs weighted correction on the prior estimate of the local state, and obtains the posterior estimate of the local state. The filter uses the calculated three-dimensional position coordinates, three-dimensional velocity components, and corresponding timestamps as the second correction data.

[0098] Step 103: Obtain the signal quality parameters of the first positioning source and the second positioning source, and based on the signal quality parameters, fuse the first correction data and the second correction data to obtain the target positioning result of the train.

[0099] Preferably, step 103 may include the following sub-steps:

[0100] S31. Based on the BeiDou satellite navigation module on the train, the number of visible satellites is obtained as the first signal quality parameter of the first positioning source.

[0101] The number of visible satellites refers to the total number of satellites from which the BeiDou satellite navigation module can successfully receive and demodulate valid navigation signals under the current operating environment of the train.

[0102] The first signal quality parameter refers to the characteristic value that characterizes the current signal state and positioning capability of the first positioning source.

[0103] In this embodiment of the invention, the navigation signals broadcast by each BeiDou satellite are captured and tracked through the BeiDou satellite navigation module receiving channel. Based on the demodulated ephemeris data and the preset carrier-to-noise ratio threshold (e.g., set to 35 dB-Hz), satellites that are unusable or have low signal quality are eliminated, thereby selecting the number of valid satellites that meet the positioning calculation conditions. This value is used as the first signal quality parameter.

[0104] S32. Based on the 5G-R communication module on the train, obtain the 5G signal strength as the second signal quality parameter of the second positioning source.

[0105] 5G signal strength refers to an indicator that characterizes the current downlink signal power level and coverage quality of the second positioning source.

[0106] The second signal quality parameter refers to the characteristic value that characterizes the current signal state and positioning capability of the second positioning source.

[0107] In this embodiment of the invention, the 5G-R communication module on the train receives synchronization signals and reference signals broadcast by surrounding 5G-R base stations. By analyzing the received power of the reference signal, the signal coverage quality of the current wireless link is comprehensively evaluated. The module samples and averages the signal strength of all base stations that can be successfully demodulated, and uses the processed average received power value of the reference signal as the second signal quality parameter. The received power of the reference signal refers to the average power contribution value of the resource particles carrying the cell-specific reference signal within the set measurement frequency band.

[0108] S33. Determine the first fusion weight corresponding to the first corrected data based on the first signal quality parameter, and determine the second fusion weight corresponding to the second corrected data based on the second signal quality parameter.

[0109] Preferably, S33 may include the following sub-steps:

[0110] S1. Normalize the first signal quality parameter and the second signal quality parameter respectively to obtain the standardized first relative quality index and the second relative quality index.

[0111] The first relative quality index is a dimensionless value obtained after normalizing the first signal quality parameter. Its value range is [0, 1], and it is used to characterize the relative reliability of the Beidou positioning source in the current environment.

[0112] The second relative quality index is a dimensionless value obtained after normalizing the second signal quality parameter. Its value range is [0, 1], and it is used to characterize the relative reliability of the 5G assisted positioning source in the current environment.

[0113] For the first signal quality parameter, its maximum reference value is preset. (For example, set to 12, corresponding to the theoretical upper limit of the number of BeiDou system satellites visible in open environments) and minimum reference value For example, setting it to 4 satellites (corresponding to the minimum number of satellites required for independent positioning calculation), will determine the currently measured number of visible satellites. The first relative quality index is obtained by performing linear normalization according to the following formula. :

[0114] Understandably, when the number of visible satellites reaches or exceeds 12, the first relative quality index is set to 1; when the number of visible satellites is less than or equal to 4, the first relative quality index is set to 0; when it is in between, it is calculated linearly proportionally, with a value between 0 and 1.

[0115] For the second signal quality parameter (5G signal strength, i.e., RSRP value), its maximum reference value is preset. (For example, set to -70 dBm, corresponding to an ideal coverage scenario with extremely strong signal) and minimum reference value (For example, set to -110 dBm, corresponding to a weak signal threshold that can maintain basic measurements). Since RSRP is a negative value and the larger the value (the smaller the absolute value), the stronger the signal, the current measured RSRP value will be used. The second relative quality index is obtained by performing linear normalization according to the following formula. :

[0116] Understandably, when RSRP reaches or exceeds -70 dBm, the second relative quality index is set to 1; when RSRP is below or equal to -110 dBm, the second relative quality index is set to 0; and when it is in between, it is calculated linearly proportionally, with a value between 0 and 1.

[0117] In this embodiment of the invention, the first signal quality parameter and the second signal quality parameter are normalized so that both the first relative quality index and the second relative quality index are dimensionless values ​​between 0 and 1.

[0118] S2. Calculate the sum of the first relative quality index and the second relative quality index to obtain the quality assessment benchmark value.

[0119] The quality assessment benchmark value refers to the sum of the first relative quality index and the second relative quality index.

[0120] In this embodiment of the invention, after obtaining the standardized first relative quality index and the second relative quality index respectively, the two are summed, and the sum is used as the quality assessment benchmark value. It should be noted that the quality assessment benchmark value reflects the comprehensive signal quality level of the Beidou positioning source and the 5G positioning source under the current operating environment. When both positioning sources are in an ideal state (i.e., both the first signal quality parameter and the second signal quality parameter are 1), the quality assessment benchmark value reaches the maximum value of 2; when both positioning sources are in an unavailable state (i.e., both the first signal quality parameter and the second signal quality parameter are 0), the quality assessment benchmark value reaches the minimum value of 0.

[0121] S3. Divide the first relative quality index by the quality assessment benchmark value to obtain the first fusion weight.

[0122] The first fusion weight refers to the weighting coefficient assigned to the first corrected data. This coefficient is determined by the ratio of the first relative quality index to the quality assessment benchmark value, and its value ranges from [0, 1], and satisfies the following condition: The higher the weight, the higher the credibility of the BeiDou positioning source in the current environment, and the greater its contribution to the final fusion positioning result.

[0123] It is understandable that after calculating the quality assessment benchmark value... Then, the first relative quality index is divided by the quality assessment benchmark value, and the resulting quotient is the first fusion weight, i.e. .

[0124] It should be noted that when the quality assessment benchmark value is zero, the division operation has no mathematical meaning. At this time, it is determined that the current environment cannot obtain a valid result through fusion positioning, which will trigger the preset positioning failure handling mechanism. For example, the valid positioning result of the previous moment will be output, auxiliary positioning sensors (such as odometers) will be enabled for recursive positioning, or a positioning anomaly alarm will be reported to the dispatch center. This step and the subsequent weighted summation steps will not be executed.

[0125] In this embodiment of the invention, the first relative quality index is divided by the quality assessment benchmark value to obtain the first fusion weight.

[0126] S4. Divide the second relative quality index by the quality assessment benchmark value to obtain the second fusion weight.

[0127] The second fusion weight refers to the weighting coefficient assigned to the second corrected data. This coefficient is determined by the ratio of the second relative quality index to the quality assessment benchmark value, and its value ranges from [0, 1]. It also satisfies that the sum of the coefficient and the first fusion weight is 1. A larger weight indicates higher reliability of the 5G-assisted positioning source in the current environment, and a greater contribution to the final fusion positioning result.

[0128] After calculating the quality assessment benchmark value, the second relative quality index is divided by the quality assessment benchmark value, and the resulting quotient is the second fusion weight. .

[0129] In this embodiment of the invention, the second relative quality index is divided by the quality assessment benchmark value to obtain the second fusion weight.

[0130] S34. Calculate the first product of the first corrected data and the first fusion weight, and calculate the second product of the second corrected data and the second fusion weight.

[0131] The first product refers to the weighted result vector obtained by multiplying the first fusion weight (scalar) by the first corrected data (vector). This product is one of the addends in the weighted summation operation and reflects the contribution of the BeiDou positioning source to the final fusion positioning result.

[0132] The second product refers to the weighted result vector obtained by multiplying the second fusion weight (scalar) by the second corrected data (vector). This product is the second addend in the weighted summation operation and reflects the contribution of the 5G assisted positioning source to the final fusion positioning result.

[0133] The specific calculation method for the first product is as follows: The first correction data is the train positioning result output after processing by the Beidou RTK local filter, represented in vector form, specifically including the three-dimensional position coordinate components of the train in the station center coordinate system (eastward coordinates). North coordinates 1. Celestial coordinates ) and the corresponding three-dimensional velocity components (eastward velocity) Northbound speed Horizontal speed ), first fusion weight As a dimensionless scalar between 0 and 1, during multiplication, this scalar weight is multiplied by each component of the first correction data to obtain the weighted first product vector, i.e. .

[0134] The calculation method for the second product is similar: it includes the three-dimensional position coordinate components in the stationary coordinate system. and the corresponding three-dimensional velocity components , with the second fusion weight Performing the multiplication operation yields the weighted second product vector, i.e. .

[0135] In this embodiment of the invention, the first product of the first corrected data and the first fusion weight and the second product of the second corrected data and the second fusion weight are calculated respectively.

[0136] S35. Summing the first product and the second product yields the target positioning result of the train.

[0137] The target positioning result refers to the final train positioning information output after processing by the data fusion module. It is represented in the form of a six-dimensional vector, which includes the three-dimensional position coordinates (eastward, northward, and upward) of the train in the station center coordinate system and the three-dimensional velocity components (eastward velocity, northward velocity, and upward velocity).

[0138] Understandably, for the position component, the eastward coordinate component in the first product is added to the eastward coordinate component in the second product to obtain the fused eastward coordinate. Similarly, northward coordinates celestial coordinates .

[0139] For the velocity components, the same summation method is used to obtain the fused eastward velocity. Northbound speed And the speed of the sky .

[0140] The above six fused components together constitute the target positioning result of the train, which is represented in the form of a six-dimensional vector. .

[0141] In this embodiment of the invention, by summing the first product and the second product, it is ensured that the fusion result is within the linear space spanned by the BeiDou positioning result and the 5G assisted positioning result, and that each component of the fusion result is between the corresponding components of the two positioning sources. This avoids extrapolation or distortion of the positioning result due to improper weight allocation. The system outputs the target positioning result to the core control module, which uploads it to the dispatch center via the 5G-R communication module for train operation monitoring, and can also be used for train self-driving, precise stopping and other control decisions.

[0142] Based on the above-described feasible embodiments, in order to improve the reliability, fault tolerance, and anti-abnormal interference capability of the fused positioning results, and to avoid contamination of the final fusion results due to a momentary anomaly or data jump in a certain positioning source, step 103 may further include the following sub-steps:

[0143] A1. Perform residual analysis on the first and second corrected data respectively to obtain the corresponding first and second residual values.

[0144] Residual analysis refers to the process of statistically analyzing and evaluating the difference (i.e., residual) between the actual measured value and the model predicted value of the positioning source. It is used to determine whether the current measurement data is abnormal or consistent with historical movement trends.

[0145] The first residual value refers to the difference between the actual observation value of the BeiDou positioning source at the current moment and the predicted observation value by the Kalman filter.

[0146] The second residual value refers to the difference between the actual measurement value of the 5G assisted positioning source at the current moment and the predicted observation value by the Kalman filter.

[0147] The specific method for calculating the first residual value is as follows:

[0148] The standard extended Kalman filter algorithm running inside the BeiDou RTK local filter is used as the carrier for residual analysis. This filter uses the train's three-dimensional position, three-dimensional velocity, and receiver clock error in the station-centric coordinate system as the state vector. (Based on the prediction of the current moment from the previous moment), and simultaneously using pseudorange observations from the original BeiDou observation data as the observation vector. (The actual measured value at the current moment), according to the filter's observation equation (Theoretical calculations of mapping the predicted state vector to the pseudorange observation space), calculate the innovation vector at the current time. The innovation vector reflects the difference between the pseudorange measurement and the predicted value for each visible satellite at the current moment. Then, the innovation vector is calculated. The weighted norm is used to obtain the first residual value. ,in, Let T be the new information covariance matrix, and T be the matrix transpose operation, which is calculated and output in real time by the filter.

[0149] Similarly, the calculation method for the second residual value is as follows:

[0150] The filter uses the train's three-dimensional position, three-dimensional speed, and receiver clock error in the station-centric coordinate system as the state vector, and the reference signal time difference (RSTD) extracted from the 5G-R base station positioning reference signal as the observation vector. (The actual measured value at the current moment), the filter predicts the state vector based on the current state vector. Given the base station's coordinates, the predicted reference signal time difference is calculated using a hyperbolic positioning model. Then calculate the innovation vector. And further calculate the second residual value. ,in Let T be the information covariance matrix of the 5G filter, and T be the matrix transpose operation.

[0151] In this embodiment of the invention, before the first and second corrected data are fed into the fusion weighting, residual analysis is performed on the two sets of data to detect whether there are abnormal measurement values ​​for each positioning source.

[0152] A2. Compare the first residual value and the second residual value with the corresponding preset residual thresholds to obtain the first comparison result and the second comparison result.

[0153] The preset residual threshold refers to the pre-set critical value used to determine whether the residual value is normal. These are the first residual threshold and the second residual threshold.

[0154] For the BeiDou positioning source, a first residual threshold is pre-set. The threshold is determined based on the chi-square distribution characteristics of the Kalman filter innovation vector: Under the premise that the filter is working normally and the measurement values ​​are not abnormal, the first residual value follows a chi-square distribution with m degrees of freedom (m is the dimension of the current BeiDou observations, i.e., the number of pseudorange observations of visible satellites). The corresponding chi-square critical value is obtained by looking up a table according to a preset significance level (e.g., 0.01, corresponding to a confidence level of 99%), and this critical value is used as the first residual threshold. Subsequently, the currently calculated first residual value is compared with this threshold: if... If the first comparison result is "normal", it indicates that the data is reliable; if If the first comparison result is determined to be "abnormal", it indicates that the measurement value of the Beidou positioning source at the current moment has an abnormal jump or quality deterioration.

[0155] Similarly, for 5G-assisted positioning sources, a second residual threshold is first set. Based on the degrees of freedom of the 5G observations (i.e., the number of effective reference signal time difference observations) and the same significance level (e.g., 0.01), the currently calculated second residual value is compared with the second residual threshold: if If the result is "normal", then the second comparison result is determined to be "normal"; if... If so, the second comparison result is determined to be "abnormal".

[0156] In this embodiment of the invention, the first residual value and the first residual threshold are compared to obtain a first comparison result, and the second residual value and the second residual threshold are compared to obtain a second comparison result.

[0157] A3. When the first comparison result is abnormal, the first corrected data is hidden, and the second corrected data is used as the target location result.

[0158] In this embodiment of the invention, when the first comparison result is determined to be abnormal, it indicates that the measurement data of the Beidou positioning source at the current moment has abnormal jumps or quality degradation, and its output first corrected data is unreliable. At this time, a data masking operation is performed: the main filter in the data fusion module marks the first corrected data as invalid, does not participate in the subsequent weighted summation operation, and does not use it as an output candidate. Instead, the second corrected data is directly output as the target positioning result at the current moment. The output result includes the three-dimensional position coordinates and three-dimensional velocity components of the train in the station center coordinate system. Its data format is completely consistent with the target positioning result of the fusion output, ensuring the uniformity of the output interface and the compatibility of subsequent modules. At the same time, the abnormal event is recorded in the fault log of the core control module, and a Beidou positioning source abnormal alarm is reported to the dispatch center through the 5G-R communication module, prompting the operation and maintenance personnel to pay attention or investigate.

[0159] A4. When the second comparison result is abnormal, the second correction data is masked and the first correction data is used as the target location result.

[0160] In this embodiment of the invention, when the second comparison result is determined to be abnormal, it indicates that the measurement data of the 5G assisted positioning source at the current moment is abnormal and the second corrected data output by it is unreliable. At this time, a data masking operation is performed: the main filter in the data fusion module marks the second corrected data as invalid and does not participate in the subsequent weighted summation operation. Instead, the first corrected data is directly output as the target positioning result at the current moment. Similarly, the abnormal event is recorded in the fault log and a 5G positioning source abnormal alarm is reported to the dispatch center.

[0161] It is worth mentioning that the two anomaly handling mechanisms mentioned above are independent of each other, ensuring that no matter which positioning source is abnormal, the abnormal data can be automatically blocked and the system can switch to a single reliable positioning source to ensure uninterrupted output of train positioning. When both comparison results are abnormal at the same time, the aforementioned positioning failure handling mechanism is triggered, the valid positioning result of the previous moment is output and a dual-source failure alarm is reported.

[0162] A5. When both the first comparison result and the second comparison result are normal, execute the steps of obtaining the signal quality parameters of the first positioning source and the second positioning source, and fusing the first correction data and the second correction data based on the signal quality parameters to obtain the target positioning result of the train.

[0163] In this embodiment of the invention, when the first comparison result and the second comparison result are both normal, it indicates that the measurement data of the Beidou positioning source and the 5G assisted positioning source at the current moment are both within the normal range, without significant abnormalities or jumps. Therefore, the specific steps of step 103 are executed.

[0164] Example 2:

[0165] Please see Figure 2 The present invention provides a train intelligent positioning system based on multi-source fusion, comprising:

[0166] The data acquisition module 201 is used to respond to the train's positioning request and acquire first positioning source data generated by the first positioning source and second positioning source data generated by the second positioning source under the current operating environment of the train.

[0167] The filtering processing module 202 is used to filter the first positioning source data and the second positioning source data respectively to obtain the first corrected data and the second corrected data;

[0168] The fusion positioning module 203 is used to acquire the signal quality parameters of the first positioning source and the second positioning source, and to fuse the first correction data and the second correction data based on the signal quality parameters to obtain the target positioning result of the train.

[0169] Since the above is a system corresponding to a train intelligent positioning method based on multi-source fusion, its implementation principle is the same as that of a train intelligent positioning method based on multi-source fusion. For the sake of convenience and brevity, those skilled in the art can clearly understand that the specific working process of the system and modules described above can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0170] Example 3:

[0171] An electronic device according to an embodiment of the present invention includes: a memory and a processor, wherein the memory stores a computer program; when the computer program is executed by the processor, the processor performs a train intelligent positioning method based on multi-source fusion as described in any of the above embodiments.

[0172] The memory can be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. The memory has storage space for program code used to perform any of the method steps described above. For example, the storage space for program code may include individual program codes for implementing the various steps in the methods described above. This program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact discs (CDs), memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When run by a computing processing device, this code causes the computing processing device to perform the various steps in the methods described above.

[0173] Example 4:

[0174] This invention provides a computer-readable storage medium storing a computer program thereon, which, when executed, implements the train intelligent positioning method based on multi-source fusion of any of the above embodiments.

[0175] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0176] Example 5:

[0177] This invention provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer performs the train intelligent positioning method based on multi-source fusion as described in any of the above embodiments.

[0178] In the embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0179] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0180] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0181] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0182] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions 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 invention.

Claims

1. A train intelligent positioning method based on multi-source fusion, characterized in that, include: In response to the train's positioning request, the system acquires first positioning source data generated by the first positioning source and second positioning source data generated by the second positioning source under the current operating environment of the train. The first positioning source data and the second positioning source data are filtered respectively to obtain the first corrected data and the second corrected data. The signal quality parameters of the first positioning source and the second positioning source are obtained, and the first corrected data and the second corrected data are fused based on the signal quality parameters to obtain the target positioning result of the train.

2. The train intelligent positioning method based on multi-source fusion according to claim 1, characterized in that, The specific method for obtaining the first positioning source data generated by the first positioning source and the second positioning source data generated by the second positioning source under the current operating environment of the train is as follows: The train receives raw observation data broadcast by the BeiDou satellite system via its BeiDou satellite navigation module, which serves as the first positioning source data, wherein the BeiDou satellite system is the first positioning source. The train receives positioning reference signals from surrounding base stations via a 5G-R communication module and extracts positioning measurement data from the positioning reference signals as the second positioning source data, wherein the surrounding base stations are the second positioning source.

3. The train intelligent positioning method based on multi-source fusion according to claim 1, characterized in that, The step of filtering the first location source data and the second location source data respectively to obtain the first corrected data and the second corrected data includes: The first positioning source data is filtered by a preset BeiDou RTK local filter to obtain the first corrected data. The second positioning source data is subjected to a second filtering process by a preset 5G assisted positioning local filter to obtain the second corrected data.

4. The train intelligent positioning method based on multi-source fusion according to claim 1, characterized in that, The step of obtaining the signal quality parameters of the first positioning source and the second positioning source, and fusing the first corrected data and the second corrected data based on the signal quality parameters to obtain the target positioning result of the train includes: Based on the BeiDou satellite navigation module on the train, the number of visible satellites is obtained and used as the first signal quality parameter of the first positioning source. Based on the 5G-R communication module on the train, the 5G signal strength is obtained and used as the second signal quality parameter of the second positioning source. The first fusion weight corresponding to the first corrected data is determined based on the first signal quality parameter, and the second fusion weight corresponding to the second corrected data is determined based on the second signal quality parameter. Calculate the first product of the first corrected data and the first fusion weight, and calculate the second product of the second corrected data and the second fusion weight; The first product and the second product are summed to obtain the target positioning result of the train.

5. The train intelligent positioning method based on multi-source fusion according to claim 4, characterized in that, The step of determining the first fusion weight corresponding to the first corrected data based on the first signal quality parameter, and determining the second fusion weight corresponding to the second corrected data based on the second signal quality parameter, includes: The first signal quality parameter and the second signal quality parameter are normalized respectively to obtain a standardized first relative quality index and a standardized second relative quality index. Calculate the sum of the first relative quality index and the second relative quality index to obtain the quality assessment benchmark value; Divide the first relative quality index by the quality assessment benchmark value to obtain the first fusion weight; The second relative quality index is divided by the quality assessment benchmark value to obtain the second fusion weight.

6. The train intelligent positioning method based on multi-source fusion according to claim 1, characterized in that, Before obtaining the signal quality parameters of the first positioning source and the second positioning source, and fusing the first corrected data and the second corrected data based on the signal quality parameters to obtain the target positioning result of the train, the method further includes: Residual analysis was performed on the first corrected data and the second corrected data respectively to obtain the corresponding first residual value and second residual value; The first residual value and the second residual value are compared with the corresponding preset residual thresholds to obtain the first comparison result and the second comparison result. When the first comparison result is abnormal, the first corrected data is hidden, and the second corrected data is used as the target location result; When the second comparison result is abnormal, the second correction data is masked, and the first correction data is used as the target location result. When both the first comparison result and the second comparison result are normal, the step of obtaining the signal quality parameters of the first positioning source and the second positioning source, and fusing the first corrected data and the second corrected data based on the signal quality parameters to obtain the target positioning result of the train is executed.

7. A train intelligent positioning system based on multi-source fusion, characterized in that, include: The data acquisition module is used to respond to the train's positioning request and acquire first positioning source data generated by the first positioning source and second positioning source data generated by the second positioning source under the current operating environment of the train; The filtering module is used to filter the first positioning source data and the second positioning source data respectively to obtain the first corrected data and the second corrected data. The fusion positioning module is used to acquire the signal quality parameters of the first positioning source and the second positioning source, and based on the signal quality parameters, to perform fusion processing on the first corrected data and the second corrected data to obtain the target positioning result of the train.

8. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 6, which is a train intelligent positioning method based on multi-source fusion.

9. A computer-readable storage medium, characterized in that, The system stores a computer program capable of being loaded by a processor and executing the train intelligent positioning method based on multi-source fusion as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, wherein when the program instructions are executed by a computer, the computer performs the multi-source fusion intelligent train positioning method as described in any one of claims 1 to 6.

Images