Processor and terminal for subway train positioning

Abstract

The invention relates to a processor and a terminal for positioning a subway train, and the processor is in communication connection with an inertial measurement unit of the train, and is used for receiving multi-source data, collected by the inertial measurement unit, of the train; in the processor, an initial correction module performs state estimation on original acceleration and angular velocity data of a train acquired by an inertial measurement unit based on an error Kalman filtering algorithm, and dynamically corrects measurement noise and system errors in combination with a motion model of the system and an actual measurement value of the inertial measurement unit; the curve generation module constructs a time-mileage curve of the train and corrects the time-mileage curve based on the first correction proportion; the turning recognition module calculates the total mileage in the turning section and corrects the total mileage in the turning section based on the second correction proportion of the turning section; the abnormal vibration recognition module recognizes abnormal vibration features based on the high-order change rate of the acceleration and the angular speed of the train and marks the abnormal vibration features in a visual mode.

Description

[0001] The original basis for this divisional application is patent application No. 202411968924.0, filed on December 30, 2024, entitled "A Metro Train Positioning Method and System". Technical Field

[0002] This invention relates to the field of urban rail detection technology, and in particular to a processor and terminal for subway train positioning. Background Technology

[0003] Currently, positioning systems for stability detection in rail transit mainly rely on GPS or BeiDou. However, in underground environments such as subway trains, these satellite positioning systems often malfunction due to signal shielding, resulting in the inability to obtain the train's accurate location. To address this issue, existing systems also use transponders and positioning tags as auxiliary positioning devices. Transponders are installed along the track, transmitting location information via short-range communication as the train passes, while positioning tags provide auxiliary positioning by sensing the train's movement. However, the installation and maintenance costs of these external devices are high, and the subway environment imposes strict limitations on equipment placement, making it difficult to adapt to comprehensive and long-term application needs.

[0004] Therefore, current stability testing instruments typically use pure inertial navigation systems, that is, an inertial navigation unit (INS) is installed on the train to perform inertial navigation calculations, thereby estimating the distance traveled without GPS. However, pure inertial navigation also suffers from error accumulation, making it difficult to maintain positional accuracy after long-term operation.

[0005] Furthermore, in subway rail transit systems, prolonged train operation leads to rail wear and track geometry changes, resulting in abnormal train vibrations. These vibrations not only affect passenger comfort but also pose a potential threat to train safety. Therefore, accurately detecting and locating abnormal vibration areas during train operation has become a crucial technical requirement for subway safety maintenance. However, existing positioning systems face challenges in the subway environment, such as signal limitations and error accumulation, making it difficult to accurately monitor and precisely locate train operational stability.

[0006] Current technologies used for train positioning mainly include GPS and inertial navigation unit (INS) fusion positioning, UWB positioning, or GPS and differential GPS (DGPS) systems and inertial navigation units (INS).

[0007] Currently, most stability testing instruments use a fusion positioning method combining GPS or BeiDou and inertial navigation unit (INS) systems. However, in underground environments such as subway trains, GPS and BeiDou signals are limited, making it impossible to locate the source of abnormal vibrations accurately. Therefore, the effectiveness of this system in subways is limited.

[0008] Another positioning method for underground trains such as subway trains is to perform real-time correction through UWB or differential GPS systems. However, these systems require the installation of base stations, resulting in high installation and maintenance costs. Furthermore, the environment of some subway lines does not allow the installation of external equipment, making them unsuitable for the positioning needs of subway trains in operation.

[0009] Pure inertial navigation does not rely on external equipment and is suitable for signal-shielded environments such as subway trains. However, its main problem is error accumulation. As time increases, the error accumulation effect becomes more pronounced, typically accumulating to approximately 15% of the mileage error within 10 minutes. For high-precision track detection, pure inertial navigation solutions struggle to meet the accuracy requirements of long-term operation.

[0010] For example, CN107976697A discloses a train safety positioning method and system based on BeiDou / GPS combination. The method includes: determining the current train track by reading transponders configured on the track, checking the accuracy of satellite positioning, and calibrating the GNSS module and inertial navigation combination module; obtaining the current train's direction of travel by reading two consecutive sets of transponders passed by the current train, or by traveling a certain distance at a certain opening speed, obtaining satellite positioning information and train inertial navigation information through the GNSS module and inertial navigation combination module, thereby resolving the train's direction of travel; once the current train track and direction of travel are determined... Subsequently, the system continuously acquires the satellite positioning information and inertial navigation information of the current train through the GNSS module and the inertial navigation unit (INS) fusion module, and determines whether there are false or erroneous satellite signals through a predetermined method. If so, it combines the ground differential station system to select the correct positioning method for the GNSS module and the INS fusion module. Finally, by setting the rear safety boundary of the current train to the most recently triggered electronic fence, and adding the train length to the rear safety boundary and extending it forward to the position of the next electronic fence, the train's front safety boundary is obtained. The front and rear safety boundaries together constitute the overall safety footprint of the train, thus forming the train's positioning. This technical solution is a typical GPS and INS fusion positioning scheme. Therefore, in underground environments, the positioning signal is easily unclear or discontinuous, leading to positioning failure and difficulty in accurately identifying the location of abnormal vibrations.

[0011] Therefore, the present invention aims to provide a subway train positioning method and system that can achieve accurate subway positioning using only an inertial navigation system in underground environments where GPS or BeiDou satellite signals are inaccurate.

[0012] Furthermore, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making this invention, but due to space limitations, not all details and contents were listed in detail. However, this does not mean that the present invention does not possess the features of these prior art. On the contrary, the present invention already possesses all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0013] In subway rail transit systems, prolonged train operation leads to rail wear and track geometry changes, resulting in abnormal train vibrations. These vibrations not only affect passenger comfort but also pose a potential threat to train safety. Therefore, accurately detecting and locating abnormal vibration areas during train operation has become a crucial technical requirement for subway safety maintenance. However, existing positioning systems face challenges in the subway environment, such as signal limitations and error accumulation, making it difficult to accurately monitor and precisely locate train running stability.

[0014] Therefore, this invention proposes a subway train positioning system, hoping to solve the following problems: First, there's the positioning problem in environments without GPS. Since subway trains mostly operate underground, signals from satellite positioning systems like GPS and BeiDou are unavailable, causing traditional stability detection equipment to lose its positioning function in underground environments. How to achieve high-precision positioning of train operations without external signal support has become one of the technical challenges.

[0015] Second, there is the problem of error accumulation in pure inertial navigation systems. Inertial navigation units (INS) can measure train motion information, but the errors of the inertial navigation system will gradually accumulate during long-term operation, making it difficult to meet the positioning accuracy requirements.

[0016] Third, the problem of identifying track feature points and mileage calibration. This invention uses features such as turning points, stopping points, and acceleration change points at switches in the subway track as mileage references. Therefore, how to effectively calibrate the inertial navigation system using the natural feature points of the track is also a problem that this invention needs to solve, which involves two difficulties: (1) identifying track feature points from the data of the inertial navigation unit (INS); (2) processing the mileage data of the feature points to obtain the calibrated mileage data.

[0017] To address the shortcomings of existing technologies, this invention provides a subway train positioning system from a first aspect, including a processor configured to: process multi-source train data collected by an inertial navigation unit based on an error Kalman filter algorithm to initially correct acceleration and angular velocity; construct a time-mileage curve of the train and correct the time-mileage curve based on a first correction ratio; identify dynamic change characteristics of turning based on the angular velocity collected by the inertial navigation unit, wherein the total mileage within the turning segment is calculated and the total mileage within the turning segment is corrected based on a second correction ratio of the turning segment; and identify abnormal vibration characteristics based on the higher-order rates of change of the train's acceleration and angular velocity and mark them in a visual manner.

[0018] This invention constructs a track feature-based mileage calibration model by collecting acceleration and angular velocity data of trains during operation and calibrating the processed data, without relying on GPS, thus ensuring accurate train mileage and location information in the absence of GPS.

[0019] According to a preferred embodiment, the processor is further configured to perform data filtering and smoothing on the pre-corrected acceleration and angular velocity before constructing the train's time-mileage curve, in order to remove noise from the data. During data measurement by the inertial measurement unit, the raw acceleration, angular velocity, velocity, and position information typically contain high-frequency noise due to environmental interference, vibration, and sensor noise. This noise can affect subsequent turn detection, mileage calculation, and position estimation. Therefore, this invention employs smoothing and filtering techniques to suppress noise and ensure data stability and accuracy.

[0020] According to a preferred embodiment, the processor initially corrects the acceleration and angular velocity by: predicting the train's state at the next moment based on a preset motion model and the train's acceleration and angular velocity collected by the inertial navigation unit during the initial stage of processor operation; correcting the predicted state information based on Kalman gain and updating the error covariance model related to the prediction, thereby correcting the predicted state of the train. The state estimation is performed on the raw acceleration and angular velocity data collected by the inertial measurement unit based on an error Kalman filter algorithm. By combining the system's motion model and the actual measurement values ​​of the inertial measurement unit, measurement noise and system errors are dynamically corrected. Through real-time estimation of the noise of the inertial measurement unit, the error Kalman filter algorithm can effectively reduce drift and random errors in the data of the inertial measurement unit, thereby providing more accurate basic data for subsequent data processing.

[0021] According to a preferred embodiment, the processor corrects the time-mileage curve by: converting acceleration and angular velocity into position and velocity to construct the time-mileage curve; using the ratio of mileage to a first target mileage as a first correction ratio; and scaling the mileage data at each time point proportionally based on the first correction ratio to obtain the corrected mileage, thereby correcting the time-mileage curve. The time-mileage curve is used to visually display the changes in train mileage.

[0022] According to a preferred embodiment, the processor identifies the dynamic changes in the turning characteristics by: performing nonlinear fitting on the angular velocity curve, calculating the rate of change of angular velocity and angular acceleration, setting a dynamic threshold based on the train's current speed and the track's curvature radius, determining the start time of the turning segment when the angular velocity is greater than the dynamic threshold, and determining the end time of the turning segment based on the negative changes in the rate of change of angular velocity and angular acceleration. Higher-order analysis of the angular velocity is performed, using the rate of change of angular velocity and its second derivative to capture minute changes between the start and end times of the turning segment, and dynamically adjusting the dynamic threshold to adapt to different turning radii and speeds.

[0023] According to a preferred embodiment, the processor corrects the total mileage within a curve segment by: calculating the total mileage within the curve segment based on the train's speed; using the ratio of the total mileage within the curve segment to a second target mileage within the curve segment as a second correction ratio; and scaling the total mileage within the curve segment proportionally based on the second correction ratio. This invention ensures that the corrected mileage within the curve segment matches the input mileage marker and dynamically adjusts the scaling ratio to match the real-time changing speed.

[0024] According to a preferred embodiment, the processor corrects the total mileage within the turning segment by employing a transition function to smoothly calibrate the corrected total mileage within the turning segment, ensuring that the corrected time-mileage curve is continuous and smoothly connected at the calibration point. This avoids discontinuities in the corrected time-mileage curve at the calibration point, ensuring a smooth connection between the time-mileage curve before and after the calibration point and preventing abrupt changes.

[0025] According to a preferred embodiment, the processor identifies abnormal vibration features by: calculating the third derivative functions of the train's acceleration and angular velocity, respectively, with the absolute value of the third derivative function being the vibration intensity; dynamically adjusting the identification threshold of the vibration point based on the train's speed; and identifying the location corresponding to a vibration intensity greater than the identification threshold as an abnormal vibration point.

[0026] This invention determines the size and color of visualized vibration points based on vibration intensity, thereby indicating the severity of the vibration, allowing users to see the specific information of each vibration point at a glance.

[0027] The present invention provides a subway train positioning method from a second aspect. The method includes: processing the acceleration and angular velocity of the train collected by an inertial navigation unit based on an error Kalman filter algorithm to preliminarily correct the acceleration and angular velocity; constructing a time-mileage curve of the train and correcting the time-mileage curve based on a first correction ratio; identifying dynamic change characteristics of turning based on the angular velocity collected by the inertial navigation unit, wherein the total mileage within the turning segment is calculated and the total mileage within the turning segment is corrected based on a second correction ratio of the turning segment; and identifying abnormal vibration characteristics based on the higher-order rates of change of the train's acceleration and angular velocity and marking them in a visual manner.

[0028] The method of this invention has several significant advantages, including improved positioning accuracy and reduced error accumulation. By combining the inertial navigation system with subway track feature points and using an error Kalman filter algorithm to correct the data of the inertial navigation unit (INS), high-precision positioning is achieved in GPS-free environments, with errors controlled within 5%. Furthermore, this invention eliminates dependence on external signals, reducing costs. Compared to existing GPS and UWB solutions, it does not require external signals or additional base station equipment, eliminating the signal limitations of GPS and BeiDou in underground environments, while avoiding the high installation and maintenance costs of base stations, demonstrating excellent economic efficiency and operability. This invention also improves the convenience and effectiveness of subway track inspection. The positioning and stability testing instrument used is highly integrated. Combined with the error Kalman filter algorithm and the calibration strategy of track feature points, it can generate time-mileage curves that more closely resemble real train data. This facilitates accurate detection and maintenance of subway track anomalies in GPS-free environments, significantly improving the accuracy and effectiveness of train stability testing.

[0029] According to a preferred embodiment, the method further includes: performing data filtering and smoothing on the preliminarily corrected acceleration and angular velocity before constructing the train's time-mileage curve, in order to remove noise from the data.

[0030] During data measurement by an inertial measurement unit (IMU), the raw acceleration, angular velocity, velocity, and position information typically contain high-frequency noise due to environmental interference, vibration, and sensor noise. This noise can affect subsequent turn detection, odometer calculation, and position estimation. Therefore, this invention employs smoothing and filtering techniques to suppress noise and ensure data stability and accuracy. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the module connection relationship of the subway train positioning system provided by the present invention; Figure 2 This is a flowchart illustrating the subway train positioning method provided by the present invention; Figure 3This is a subway track diagram provided by the present invention; Figure 4 This is a schematic diagram showing the orientation of the inertial measurement unit (IMU) provided by the present invention placed flat on the train; Figure 5 This is a schematic diagram of the angular velocity change provided by the present invention; Figure 6 This is a schematic diagram showing the operation test data of a subway line provided by the present invention; Figure 7 This is a comparative diagram of the time-mileage curves before and after the correction provided by the present invention; Figure 8 This is a graph showing the absolute velocity changing over time, provided by the present invention. Figure 9 This is a test chart of the preliminary corrected total mileage data in the example data provided by this invention; Figure 10 This is a test chart of the total mileage data calculated from the example data provided by this invention; Figure 11 This is a table of example data obtained through testing of the subway train positioning system provided by this invention.

[0032] List of reference numerals 100: Processor; 110: Initial calibration module; 120: Curve generation module; 130: Turning recognition module; 140: Abnormal vibration recognition module; 200: Stability detector; 300: Memory. Detailed Implementation

[0033] The following is a detailed explanation with reference to the accompanying drawings.

[0034] This invention provides explanations for some terms and nouns.

[0035] The stability monitoring system is a data acquisition hardware system integrating an inertial measurement unit (IMU), a step-down module, a microcontroller, and a power supply. It is specifically designed for measuring and analyzing the stability of vehicles, especially trains. The system captures the train's acceleration and attitude through the IMU, ensures voltage stability through the step-down module, and processes signals from sensors and performs control tasks through the microcontroller. Furthermore, the system communicates with a host computer, which processes the collected data and performs in-depth analysis, including data trend analysis, anomaly detection, and report generation.

[0036] The core of the stability detection system also includes a mileage calibration algorithm based on absolute mileage and speed. This algorithm determines the train's operating speed by calculating the arithmetic mean of the operating speeds at multiple sensor locations and discards any abnormal speed values ​​to improve measurement accuracy. This algorithm helps reduce computational errors, ensures the accuracy of train speed measurements, and provides important data support and analysis tools for train stability and safety.

[0037] IMU: An abbreviation for Inertial Measurement Unit. An IMU is an electronic device used to measure and report the three basic linear motions (acceleration) and three basic angular motions (angular velocity) of an object.

[0038] A 6-axis inertial measurement unit (IMU) is a sensor system that integrates three acceleration axes (X, Y, Z) and three angular velocity axes (X, Y, Z), enabling comprehensive measurement of an object's motion in three-dimensional space. The acceleration axes capture the linear acceleration of the object in three mutually perpendicular directions, while the angular velocity axes measure the object's rotational rate about these axes. This design allows the 6-axis IMU to provide precise data on an object's velocity, position, and attitude.

[0039] Multi-source data: including triaxial acceleration and triaxial angular velocity (angle) collected by inertial measurement unit (IMU), actual mileage between two stations, turning point mileage information, etc.

[0040] Error Kalman Filtering Algorithm: Used to process data acquired by Inertial Measurement Units (IMUs). The error Kalman Filtering algorithm is an application of the Kalman Filtering algorithm when processing measurement data with errors. The Kalman Filtering algorithm is an optimal estimation algorithm for estimating the state of dynamic systems, particularly suitable for situations with noise and uncertainty.

[0041] In practical applications, observational data are often affected by various types of noise, which can be random or systematic. The error Kalman filter algorithm describes the statistical characteristics of these errors using mathematical models, such as their mean and covariance, and then uses this information to optimize state estimation and reduce the impact of errors.

[0042] Example 1 Current technologies used for train positioning mainly include GPS and inertial navigation unit (INS) fusion positioning, UWB positioning, or GPS and differential GPS (DGPS) systems and inertial navigation units (INS).

[0043] Currently, most stability testing instruments use a fusion positioning method combining GPS or BeiDou and inertial navigation units (INS). However, in underground environments such as subway trains, GPS and BeiDou signals are limited, making it impossible to locate the source of abnormal vibrations accurately. Therefore, the effectiveness of this system in subways is limited.

[0044] Another method for positioning subway trains is to perform real-time correction using UWB or differential GPS systems. However, these systems require the installation of base stations, resulting in high installation and maintenance costs. Furthermore, the environment of some subway lines does not allow the installation of external equipment, making them unsuitable for the positioning needs of subway trains during operation.

[0045] Pure inertial navigation does not rely on external equipment and is suitable for signal-shielded environments such as subway trains. However, its main problem is error accumulation. As time increases, the error accumulation effect becomes more pronounced, typically accumulating to approximately 15% of the mileage error within 10 minutes. For high-precision track detection, pure inertial navigation solutions struggle to meet the accuracy requirements of long-term operation.

[0046] This invention provides a subway train positioning method and system, particularly a subway train positioning method and system based on an error Kalman filter algorithm and multi-source data mileage correction. This invention can also provide an electronic device for subway train positioning. This invention can also provide a storage medium storing an encoding program for the subway train positioning method. This invention can also provide a processor 100 for the encoding program of the subway train positioning method. This invention can also provide a terminal that can be connected to the processor 100. This invention can also provide a subway control system, which contains the subway train positioning system of this invention or executes the subway train positioning method of this invention. Preferably, this invention can also provide a subway, on which the subway train positioning system of this invention is installed, or an electronic device executing the subway train positioning method of this invention is installed.

[0047] This invention aims to solve the following problems: First, there's the positioning problem in environments without GPS. Since subway trains mostly operate underground, signals from satellite positioning systems like GPS and BeiDou are unavailable, causing traditional stability detection equipment to lose its positioning function in underground environments. How to achieve high-precision positioning of train operations without external signal support has become one of the technical challenges.

[0048] Second, there is the problem of error accumulation in pure inertial navigation systems. Inertial navigation units (INS) can measure train motion information, but the errors of the inertial navigation system will gradually accumulate during long-term operation, making it difficult to meet the positioning accuracy requirements.

[0049] Third, the problem of identifying track feature points and mileage calibration. This invention uses features such as turning points, stopping points, and acceleration change points at switches in the subway track as mileage references. Therefore, how to effectively calibrate the inertial navigation system using the natural feature points of the track is also a problem that this invention needs to solve, which involves two difficulties: (1) identifying track feature points from the data of the inertial navigation unit (INS); (2) processing the mileage data of the feature points to obtain the calibrated mileage data.

[0050] To address the shortcomings of existing technologies, this invention provides a subway train positioning system, such as... Figure 1 As shown, it includes a processor 100. The processor 100 is configured as multiple modules, including an initial correction module 110, a curve generation module 120, a turn recognition module 130, and an abnormal vibration recognition module 140.

[0051] Preferably, when the processor 100 is a standalone processor, the processor 100 can be a CPU, a GPU, or a combination thereof, or a dedicated integrated chip capable of executing the subway train positioning method of the present invention. In this case, the initial correction module 110, the curve generation module 120, the turn recognition module 130, and the abnormal vibration recognition module 140 are different execution task modules within the processor 100.

[0052] Preferably, when the processor 100 is not a standalone processor, it can be composed of multiple dedicated integrated chips and / or processing hardware (CPU and / or GPU, storage module) connected in communication. The initial correction module 110, curve generation module 120, turning recognition module 130, and abnormal vibration recognition module 140 are each standalone processors, or at least two modules are integrated into the same processor 100. Preferably, different standalone processors are connected via communication lines.

[0053] Preferably, the processor 100 can be connected to a terminal to receive user-input data and display the processing process and results of the processor 100.

[0054] Preferably, such as Figure 1 As shown, the processor 100 of the present invention is communicatively connected to the inertial measurement unit (IMU) of the train and is used to receive multi-source data of the train collected by the inertial measurement unit (IMU).

[0055] Preferably, the subway train positioning system of the present invention may include a processor 100 and an inertial measurement unit (IMU) or an inertial navigation unit (INS). More preferably, the subway train positioning system of the present invention may include a processor 100, an inertial measurement unit, and a memory 300.

[0056] The subway train of this invention will be referred to as the train in the following description.

[0057] Preferably, the stability tester 200 on the train is equipped with an inertial measurement unit (IMU), a step-down module, a power supply, and an embedded chip. Alternatively, the stability tester 200 on the train is equipped with an inertial navigation unit (INS, including an inertial measurement unit (IMU)), a step-down module, a power supply, and an embedded chip.

[0058] Inertial Measurement Units (IMUs) or Inertial Navigation Units (INS) acquire high-frequency signals. Embedded chips use an error Kalman filter algorithm to process the acquired data, thereby reducing random errors in the IMUs or INS and improving data accuracy. Characteristic points on the subway track, such as track curvature changes during train turns, and station stopping times and locations, can serve as key reference points for mileage calibration: the location of turns and the start or end times within a turn are obvious and easy to acquire, while stopping times and locations provide a clear mileage calibration benchmark.

[0059] like Figure 2 As shown, the processor 100 of the present invention is configured to perform the following steps.

[0060] S100: The initial correction module 110 processes the multi-source data of the train collected by the inertial measurement unit (IMU) in the inertial navigation unit based on the error Kalman filter algorithm to initially correct the acceleration and angular velocity.

[0061] The error Kalman filter algorithm performs state estimation on the raw acceleration and angular velocity data acquired by the inertial measurement unit (IMU). By combining the system's motion model with the actual measurements from the IMU, it dynamically corrects measurement noise and system errors. Through real-time estimation of IMU noise, the error Kalman filter algorithm effectively reduces drift and random errors in the IMU data, thus providing more accurate foundational data for subsequent data processing.

[0062] S200: Curve generation module 120 constructs the train's time-mileage curve and corrects the time-mileage curve based on a first correction ratio.

[0063] S300: The turn recognition module 130 identifies the dynamic change characteristics of the turn based on the angular velocity collected by the inertial measurement unit (IMU) in the inertial navigation unit. Preferably, the turn recognition module 130 calculates the total mileage within the turn segment and corrects the total mileage within the turn segment based on a second correction ratio of the turn segment.

[0064] S400: Abnormal vibration identification module 140 identifies abnormal vibration characteristics based on the higher-order rate of change of the train's acceleration and angular velocity and marks them in a visual manner.

[0065] This invention constructs a track feature-based mileage calibration model by collecting acceleration and angular velocity data of trains during operation and calibrating the processed data, without relying on GPS, thus ensuring accurate train mileage and location information in the absence of GPS.

[0066] The present invention provides a detailed description of step S100.

[0067] S110: State initialization.

[0068] In the initial stage of system operation according to the present invention, a state vector is set based on the initial acceleration, angular velocity, and position data of the inertial measurement unit (IMU). Initialize the state error covariance matrix based on the characteristics of the inertial measurement unit (IMU). The state error covariance matrix includes several initial state vectors. These initial state vectors can include the train's initial position, velocity, and attitude.

[0069] S120: Predicted state (time update).

[0070] The train's state at the next moment is predicted based on the preset motion model and the train's acceleration and angular velocity collected by the inertial measurement unit (IMU) in the inertial navigation unit.

[0071] The time update formula is: .

[0072] in, Indicates the predicted state. This indicates the previous predicted state. Represents the state transition matrix. This represents the input data (such as acceleration and angular velocity) measured by the inertial measurement unit (IMU). This represents the input matrix.

[0073] S130: Prediction error covariance.

[0074] Error covariance matrix of predicted state for: .

[0075] in, This represents the transpose of the state transition matrix; This represents the process noise covariance matrix, used to describe the uncertainty in the system model.

[0076] S140: Correction prediction state.

[0077] Based on the state information predicted by Kalman gain correction, the error covariance model related to the prediction is updated.

[0078] The predicted state is corrected based on acceleration and angular velocity data collected by the inertial measurement unit (IMU).

[0079] First, calculate the Kalman gain. : .

[0080] in, Represents the measurement matrix. Represents the measurement noise covariance matrix. This represents the state observation matrix.

[0081] Correcting the predicted state and error covariance using Kalman gain: .

[0082] in, This represents the actual measurement value of the inertial measurement unit (IMU). This represents the corrected state estimate.

[0083] Update error covariance: .

[0084] S150: Repeat the prediction and update steps.

[0085] Kalman filtering performs state prediction and correction (update) at each time step, continuously reducing the impact of noise and drift on the system state estimation through a recursive process. This filtering process continues until data processing is complete.

[0086] After applying error Kalman filtering to the inertial measurement unit (IMU) data, the system obtains acceleration and angular velocity data with preliminary error correction. Subsequently, this filtered IMU data undergoes further smoothing and integration calculations to generate the train's time-mileage curve. This curve is then combined with track feature points for mileage calibration, ultimately achieving high-precision train trajectory and abnormal vibration detection.

[0087] Preferably, the execution steps of the processor 100 may further include S160. S160 is executed between steps S100 and S200.

[0088] S160: Before constructing the train's time-mileage curve, the preliminarily corrected acceleration and angular velocity are filtered and smoothed to remove noise from the data.

[0089] During data measurement by an inertial measurement unit (IMU), the raw acceleration, angular velocity, velocity, and position information typically contain high-frequency noise due to environmental interference, vibration, and sensor noise. This noise can affect subsequent turn detection, odometer calculation, and position estimation. Therefore, this invention employs smoothing and filtering techniques to suppress noise and ensure data stability and accuracy.

[0090] The present invention will describe step S160 as follows.

[0091] S161: Perform low-pass filtering.

[0092] Acceleration and angular velocity data are often affected by high-frequency noise.

[0093] Acceleration is expressed as: .

[0094] Angular velocity is expressed as: .

[0095] Therefore, the acceleration and angular velocity signals are low-pass filtered to eliminate high-frequency noise.

[0096] Preferably, a Butterworth filter is used for low-pass filtering. Low-pass filtering is characterized by a smooth frequency response and can effectively filter out high-frequency noise without introducing phase shift. Assuming the filter is fourth-order and the cutoff frequency is... Then the transfer function of the filter is: .

[0097] This transfer function describes how the filter responds to signals of different frequencies.

[0098] in, Represents a complex frequency variable. Indicates the damping coefficient. This indicates the cutoff frequency of the filter.

[0099] For example, the ζ value of the Butterworth filter is fixed, and it is usually chosen to be ζ=0.707.

[0100] Low-frequency vibration: Vibrations with lower frequencies caused by uneven tracks, asymmetrical wheels, etc. Typically below 20Hz.

[0101] The aforementioned high-frequency noise may originate from sensor noise, electrical noise, etc., and is typically greater than 50Hz.

[0102] Preferably, when using a Butterworth low-pass filter, the cutoff frequency can be selected. For example, cutoff frequency The frequency is set to 30Hz, and then the frequency response of the Butterworth filter is calculated to determine how to filter out high-frequency noise.

[0103] =2π×30rad / s≈188.4rad / s. This is the cutoff frequency of a low-pass filter at a frequency of 30Hz. The value.

[0104] For example, assuming the cutoff frequency is chosen The cutoff frequency is 30Hz, which determines the frequency at which the filter begins to decay.

[0105] Substitute each cutoff frequency into the filter's transfer function to calculate the corresponding gain. This represents the filter's response at that frequency.

[0106] For signals below the cutoff frequency, the filter attenuates almost nothing.

[0107] For signals above the cutoff frequency, the filter will gradually attenuate the signal and reduce high-frequency noise.

[0108] Specifically, when selecting the cutoff frequency At 30Hz, the frequency components of the train acceleration vibration signal from 1Hz to 30Hz are basically unaffected, while high-frequency noise above 30Hz will be filtered out.

[0109] The transfer function after filtering out high-frequency noise becomes: .

[0110] Set as ,in It is the actual frequency of the signal.

[0111] Complex frequency variable By substituting the transfer function, the effect of the filter on the signal processing can be calculated.

[0112] In this invention, the acceleration and angular velocity data are low-pass filtered by a filter to obtain the filtered signal: .

[0113] .

[0114] in, This represents the convolution operation. and These represent the acceleration and angular velocity signals obtained after filtering, respectively.

[0115] S162: Smooth the data.

[0116] Even after filtering, the acceleration and angular velocity signals may still contain slight noise fluctuations. Therefore, smoothing is employed to further suppress residual noise and enhance data stability. The smoothing process uses a moving average method and a double smoothing technique.

[0117] S1621: Moving average method.

[0118] The moving average is a commonly used smoothing technique that eliminates short-term fluctuations by averaging the data within a window. For position, velocity, and angular velocity data, the smoothing formula for the moving average is: .

[0119] in, Indicates window size. Represents the original signal The value at the Kth point. k is an index representing the value of the original signal at the (t+k)th moment within N time intervals before and after time t. The data measured in this invention are angular velocity and acceleration data; therefore, smoothing is only performed on the angular velocity and acceleration data.

[0120] S1622: Secondary smoothing.

[0121] To further improve the smoothing effect, the system performs a second smoothing of the data. First, a second moving average is applied to the initially smoothed data: .

[0122] This secondary smoothing process further removes high-frequency noise while maintaining signal stability. Particularly in corner detection and odometer calibration, using smoothed data reduces jitter and ensures the accuracy of subsequent calculations.

[0123] S163: Overall formula description of the data processing procedure.

[0124] As shown above, the system processes the raw inertial measurement unit (IMU) data. , Perform a two-step process.

[0125] Low-pass filtering: High-frequency noise is removed using a Butterworth low-pass filter. .

[0126] .

[0127] Secondary smoothing: The filtered signal is further smoothed by two moving averages. .

[0128] .

[0129] It can represent acceleration 'a' or angular velocity. .when When representing acceleration 'a', This represents the filtered acceleration 'a'. This represents the acceleration 'a' after the second smoothing process.

[0130] when Represents angular velocity hour, Represents the filtered angular velocity , Represents the angular velocity after secondary smoothing. .

[0131] Through the aforementioned low-pass filtering and secondary smoothing processes, the system effectively removes high-frequency noise from the inertial measurement unit (IMU) data, ensuring reduced noise interference and providing smoother signals in the calculation of critical data such as acceleration and angular velocity. This processing method is particularly suitable for turn detection and odometer calibration of the inertial measurement unit (IMU) in inertial navigation, improving the system's stability and accuracy.

[0132] Figure 3 The image shows a subway track route map. Figure 3 In this context, XK0+139 represents the 0th kilometer and the 139th meter. XK1 represents 1 kilometer. Figure 3 XK1+355.944 indicates that the turn begins at 1355.944m. In other words, there is a turning point between the two stations: XK1+355.944.

[0133] like Figure 4 As shown, when the inertial measurement unit (IMU) is placed flat on the train, the X or Y direction of the IMU represents the direction of train travel. The curved arrow indicates the angular velocity around the Z-axis.

[0134] When the train passes through a curve, the inertial measurement unit (IMU) detects and records the change in angular velocity, such as... Figure 5 As shown. In Figure 5 In the diagram, the horizontal axis represents time, and the vertical axis represents the magnitude of the angular velocity about the z-axis. The angular velocity changes significantly during the turning segment.

[0135] The present invention describes step S200 as follows.

[0136] S210: Converts acceleration and angular velocity into position and velocity to construct time-mileage curves.

[0137] Numerical integration of acceleration and angular velocity data transforms the acceleration and angular velocity data of the train at different time points into position and velocity information, generating the train's time-mileage curve.

[0138] S211: Acceleration data is used to calculate velocity through integration.

[0139] The integral formula is: .

[0140] in, Indicates time speed, Represents acceleration data. This indicates the initial velocity.

[0141] S212: Perform a second integration on the speed data to obtain the train's mileage.

[0142] .

[0143] in, Indicates the time of the train Mileage, Indicates the initial mileage.

[0144] S220: Use the ratio of the mileage to the first target mileage as the first correction ratio.

[0145] like Figure 6 As shown, the test data for the operation of Metro Line 9 of XX is presented. Figure 6 The information shown includes total mileage settings, query time, query results, running mileage, running time, correction file options, algorithm selection box, etc.

[0146] S221: Enter the first target mileage.

[0147] Enter the actual mileage S in "Total Mileage Settings". target S target Used to correct the total mileage calculated by integration from the inertial measurement unit (IMU).

[0148] S222: Total mileage correction.

[0149] The mileage is calculated by integrating the inertial measurement unit (IMU). The first target mileage S entered by the user target The ratio of the two values ​​yields the first correction ratio. .

[0150] .

[0151] when When the value is greater than 1, it indicates that the calculated mileage is less than the first target mileage, and stretching is required; when... When the value is less than 1, it indicates that the calculated mileage is greater than the first target mileage and compression is required.

[0152] S230: The mileage data at each time point is scaled proportionally based on the first correction ratio to obtain the corrected mileage, thus correcting the time-mileage curve. The time-mileage curve is used to visually display the changes in train mileage.

[0153] S231: Use the first correction ratio The mileage data at each time point is scaled proportionally.

[0154] .

[0155] in, This indicates the corrected mileage.

[0156] S232: Plot the time-mileage curve.

[0157] like Figure 7 As shown, the corrected time-mileage curve is generated and displayed to visually show the changes in the train's mileage.

[0158] Figure 7 The table on the right shows time on the horizontal axis and mileage on the vertical axis. It displays curves plotted from the original data and curves plotted from the corrected data.

[0159] Compare the time-mileage curves before and after the correction. The corrected time-mileage curve more accurately shows the train's mileage changes.

[0160] The present invention describes step S300 as follows.

[0161] To more accurately identify the moments when the train enters and exits a turn, the Z-axis angular velocity is... Advanced analysis is performed, using the rate of change of angular velocity and the second derivative to capture minute changes at the start and end points of a turn, and dynamic threshold adjustments are made to adapt to different turning radii and speeds.

[0162] S310: Perform nonlinear fitting on the angular velocity curve to calculate the rate of change of angular velocity and angular acceleration.

[0163] The change of Z-axis angular velocity over time can be modeled using a third-order Bézier curve: .

[0164] in, , , , These represent the coefficients obtained through least squares fitting. Here, nonlinear fitting of the angular velocity curve is used to capture the dynamic changes during turning.

[0165] S320: Based on the current speed of the train and the curvature radius of the track, a dynamic threshold is set. When the angular velocity is greater than the dynamic threshold, the start time of the train's turning segment is determined. The end time of the train's turning segment is determined based on the rate of change of angular velocity and the negative change of angular acceleration.

[0166] S321: Determining the start time within the turning segment.

[0167] The start time within the turning segment is determined by the first and second derivatives of the angular velocity.

[0168] Calculate the rate of change of Z-axis angular velocity and acceleration Its definition is as follows: .

[0169] .

[0170] Set dynamic threshold This dynamic threshold is determined based on the train's current speed and the track's curvature radius, and is calculated as follows: .

[0171] in, Represents the proportionality coefficient. Indicates the current speed. This represents the radius of curvature of the orbit.

[0172] Preferably, α is used to adjust the sensitivity of dynamic detection during turns. It controls the value of the threshold, thereby determining how changes in the train's angular velocity are detected during turns. The dynamic threshold is adjusted as the train's speed and the radius of curvature of the track change. The value range is 0.8 to 1.5.

[0173] When the rate of change of angular velocity reaches the threshold, the train begins to turn. .

[0174] The start time of a train's turn depends not only on the change in angular velocity, but also on the positive change in angular acceleration, in order to ensure the dynamic characteristics of the turn.

[0175] S322: Determination of the end time within the turning segment.

[0176] The end time within the turning segment is determined by the rate of change of angular velocity and the negative change of acceleration.

[0177] The specific formula is as follows: .

[0178] At this point, the angular velocity decreases and the angular acceleration becomes negative, indicating that the train is leaving the curve section.

[0179] A high-order analysis of angular velocity is performed, using the rate of change of angular velocity and the second derivative to capture the minute changes at the start and end of the turning segment, and dynamically adjusted through a dynamic threshold to adapt to different turning radii and turning speeds.

[0180] S330: Corrects the total mileage within the turning section.

[0181] S331: Calculate the total mileage within the turning section based on the train's speed.

[0182] To calculate the total distance the train travels within each curve segment, the train's speed is integrated. Speed The acceleration data from the inertial measurement unit (IMU) is obtained by integrating it twice: .

[0183] in, Indicates acceleration. This indicates the initial velocity.

[0184] Integrating the speed further, we obtain the total distance traveled within the turning segment: .

[0185] The calculation of the total mileage within the turning section not only takes into account speed, but also corrects for nonlinear changes in acceleration by integrating the acceleration.

[0186] S332: The ratio of the total mileage within the turning segment to the second target mileage within the turning segment is used as the second correction ratio.

[0187] The total mileage calculated by the inertial measurement unit (IMU) within the turning segment may have a significant error compared to the actual total mileage within the turning segment. A nonlinear correction model is used to calibrate the total mileage, assuming the second target mileage input by the terminal is... (representing the mileage starting at the first turn and the mileage ending at the first turn, respectively), second correction ratio. It is a dynamic function that varies with time to adapt to changes in train speed and turning conditions: .

[0188] The denominator is the total mileage actually integrated within the turning segment.

[0189] S333: The total mileage within the turning segment is scaled proportionally based on the second correction ratio.

[0190] Specifically, the total mileage within each turning segment is non-linearly scaled using a second correction ratio: .

[0191] This represents the total mileage within the original turning segment. This indicates the total mileage within the corrected turning segment. This indicates the start time within the first turning segment. This indicates the end time within the first turning segment.

[0192] This formula ensures that within each turning segment, the mileage within the corrected turning segment matches the input mileage marker, and dynamically adjusts the scaling ratio to match the real-time changing speed.

[0193] S334: A transition function is used to smooth the total mileage within the corrected turning section so that the corrected time-mileage curve is continuous and smoothly connected at the calibration point.

[0194] The transition function is: .

[0195] in, This represents the parameter controlling the transition. The corrected total mileage can be expressed as: .

[0196] This represents the total mileage within the turning section after smoothing calibration. This avoids discontinuities in the calibrated time-mileage curve at the calibration point, ensuring a smooth transition between the curves before and after the calibration point and preventing abrupt changes.

[0197] The present invention describes step S400 as follows.

[0198] The identification of abnormal vibrations is based on the analysis of higher-order rates of change of acceleration and angular velocity. This is achieved by analyzing acceleration... and angular velocity The third derivative is used to capture dramatic vibrational changes.

[0199] S410: Calculate the third derivative functions of the train's acceleration and angular velocity respectively. The absolute value of the third derivative function is the vibration intensity.

[0200] To capture vibrations more accurately, the third derivative of acceleration is calculated. and the third derivative of angular velocity : .

[0201] .

[0202] These third derivatives represent the rates of change of acceleration and angular velocity, respectively. In other words, these third derivatives are used to capture rapid changes in acceleration and angular velocity. Based on the magnitude of the rate of change, vibration points with large rates of change can be quickly identified, that is, the vibration location and duration with large rates of change in angular velocity and acceleration can be quickly determined, thus effectively capturing vibration points that change drastically within a short period of time.

[0203] S420: Dynamically adjusts the vibration point identification threshold based on train speed. The location corresponding to the vibration intensity exceeding the identification threshold is the abnormal vibration point.

[0204] The threshold for identifying vibration points varies depending on the train's speed and environmental conditions. and It is also dynamically adjusted: .

[0205] .

[0206] in, and It is a dynamic adjustment coefficient. ≥1, ≤5, to ensure the accuracy of vibration point identification under different speed conditions. Indicates the current speed. This represents the radius of curvature of the orbit.

[0207] γ and These are dynamic adjustment coefficients used to adjust the threshold for identifying vibration points. The function of these two coefficients is based on the train's speed. and the radius of curvature of the orbit The recognition thresholds for vibrations based on acceleration and angular velocity are dynamically adjusted to avoid misjudging large vibrations caused by turning or acceleration as abnormal vibrations.

[0208] When the train speed is low or the track is relatively straight, the detection threshold is low so that small vibration changes can be detected.

[0209] Preferably, points in the time-mileage curves of the rate of change of angular velocity and the rate of change of acceleration that are greater than the identification threshold are identified as abnormal vibration points.

[0210] S430: Mark vibration points in a visual manner.

[0211] At each detected abnormal vibration point, not only is its time and distance marked, but the severity of the vibration is also indicated by the intensity of the color. This invention determines the size and color of the visualized vibration point based on the vibration intensity, thereby representing the severity of the vibration, allowing users to see the specific information of each vibration point at a glance.

[0212] Figure 8 The graph in the middle shows the absolute velocity as a function of time, and also serves as a visual representation of the abnormal vibration point. The horizontal axis of the curve represents time (seconds), and the vertical axis represents absolute velocity (meters per second). Figure 8 As shown by the curve, the circular dots on the curve represent abnormal vibration points of the train, and the triangular dots represent vibration changes caused by the train turning (which are normal phenomena and can be ignored). Through the visualization of abnormal vibration points, users can intuitively see the number and frequency of abnormal vibrations of the train, and can also intuitively determine the location of abnormal vibration points.

[0213] By employing nonlinear correction models and higher-order derivative analysis, the accuracy of identifying turning moments was improved, and the accumulation of mileage errors was reduced. Furthermore, by analyzing acceleration and angular velocity using third-order derivatives, abnormal vibration points can be identified more sensitively, providing more accurate track maintenance information.

[0214] Preferably, by performing mileage correction at turning points, the present invention ensures that at least three mileage corrections are performed at each station. Compared with the prior art, which only performs mileage correction at the end, the time-mileage curve obtained by the present invention is closer to the actual time-mileage curve of train operation.

[0215] As mentioned above, due to the existence of the turning point, the Z-axis angular velocity data measured by the inertial measurement unit (IMU) when the train passes through this point will be significantly different from that when the train is moving in a straight line. Based on this principle, the moment when the train enters the turning point can be found, and the mileage information calculated after data processing can be corrected to the information at the starting point of the turn.

[0216] This invention also provides example data obtained through testing of the subway train positioning system, such as... Figure 11 As shown. In Figure 11 In the test chart of the preliminary corrected total mileage data, as shown below. Figure 9 As shown ( Figure 9 (The triangle markers in the diagram can be ignored); the test graph of the calculated total mileage data is shown below. Figure 10 As shown.

[0217] The method and system of the present invention can achieve the following objectives: First, improve positioning accuracy and reduce error accumulation.

[0218] This invention combines an inertial navigation system with subway track feature points (such as turning points, stopping points, and acceleration abrupt change points at switches) and uses an error Kalman filter algorithm to correct the data of the inertial navigation unit (INS). By utilizing track feature points as natural calibration references, high-precision positioning is achieved in GPS-free environments, keeping the error within a reasonable range (5%).

[0219] Second, reduce reliance on external signals and lower costs.

[0220] Compared with existing GPS and UWB solutions, this invention does not require external signals or additional base station equipment, completely eliminating the signal limitations of the underground environment of the subway on GPS and BeiDou, while avoiding the high installation and maintenance costs of base stations, thus possessing good economic efficiency and operability.

[0221] Third, improve the convenience and effectiveness of train track inspection.

[0222] The positioning and stability testing instrument 200 used in this invention has a high degree of integration. Combined with the error Kalman filter algorithm and the calibration strategy of track feature points, it can generate a time-mileage curve that is closer to the real train. This facilitates the accurate detection and maintenance of abnormal points on subway lines in the absence of GPS, and greatly improves the accuracy and effectiveness of train stability testing.

[0223] By achieving the above objectives, this invention provides a low-cost, high-precision solution for accurate positioning and track detection in underground environments, effectively addressing the shortcomings of existing positioning technologies in situations where GPS is unavailable or base stations cannot be installed.

[0224] It should be noted that the specific embodiments described above are exemplary. Those skilled in the art can devise various solutions inspired by the disclosure of this invention, and these solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and not intended to limit the scope of the claims. The scope of protection of this invention is defined by the claims and their equivalents. This specification contains multiple inventive concepts; phrases such as "preferredly" or "according to a preferred embodiment" indicate that the corresponding paragraph discloses an independent concept. The applicant reserves the right to file divisional applications based on each inventive concept.

Claims

1. A processor for subway train positioning, the processor (100) being communicatively connected to the train's inertial measurement unit (IMU) for receiving multi-source data of the train collected by the IMU; characterized in that, The processor (100) includes: The initial correction module (110) performs state estimation on the original acceleration and angular velocity data of the train collected by the inertial measurement unit based on the error Kalman filter algorithm, and dynamically corrects the measurement noise and system error by combining the system's motion model and the actual measurement value of the inertial measurement unit. The curve generation module (120) constructs the train's time-mileage curve and corrects the time-mileage curve based on the first correction ratio; The turning recognition module (130) calculates the total mileage within the turning segment and corrects the total mileage within the turning segment based on the second correction ratio of the turning segment; The abnormal vibration identification module (140) identifies abnormal vibration features based on the higher-order rate of change of the train's acceleration and angular velocity and marks them in a visual manner.

2. The processor according to claim 1, characterized in that, The initial correction module (110) is configured as follows: Perform state initialization; The train's state at the next moment is predicted based on the acceleration and angular velocity of the train collected by the preset motion model and inertial measurement unit. Calculate the error covariance matrix of the predicted state; The state information predicted by Kalman gain correction is used to update the error covariance matrix related to the prediction. The predicted state is corrected based on acceleration and angular velocity data.

3. The processor according to claim 1 or 2, characterized in that, The initial correction module (110) is also configured to: Before constructing the train's time-mileage curve, the preliminarily corrected acceleration and angular velocity data are filtered and smoothed to remove noise from the data.

4. The processor according to any one of claims 1 to 3, characterized in that, The curve generation module (120) is configured as follows: Acceleration and angular velocity are converted into position and velocity to construct a time-mileage curve; wherein, acceleration data is integrated to obtain velocity data; the velocity data is then integrated twice to obtain the train's mileage; The ratio of the stated mileage to the first target mileage is used as the first correction ratio; Based on the first correction ratio, the mileage data at each time point is scaled proportionally to obtain the corrected mileage, and the time-mileage curve is corrected to intuitively display the changes in the train's mileage.

5. The processor according to any one of claims 1 to 4, characterized in that, The steps of the turning recognition module (130) in calculating the total mileage within the turning segment include: Nonlinear fitting is performed on the angular velocity curve; Calculate the rate of change of angular velocity and angular acceleration; A dynamic threshold is set based on the train's current speed and the curve radius of the track; When the angular velocity is greater than the dynamic threshold, the start time within the train's turning segment is determined; The end time within the train's turning segment is determined based on the rate of change of angular velocity and the negative change of angular acceleration. The total mileage within the curve section is calculated based on the train's speed.

6. The processor according to any one of claims 1 to 5, characterized in that, The step of the turning recognition module (130) correcting the total mileage within the turning segment based on the second correction ratio of the turning segment includes: Calculate the total mileage within the curve section based on the train's speed; The ratio of the total mileage within the turning segment to the second target mileage within the turning segment is used as the second correction ratio; The total mileage within the turning segment is scaled proportionally based on the second correction ratio. A transition function is used to smooth the total mileage within the corrected turning segment so that the corrected time-mileage curve is continuous and smoothly connected at the calibration point.

7. The processor according to any one of claims 1 to 6, characterized in that, The abnormal vibration identification module (140) is configured to: Calculate the third derivative functions of the train's acceleration and angular velocity, respectively, to represent the rate of change of the acceleration and angular velocity. The absolute value of the third derivative function is the vibration intensity. Quickly identify vibration points with large rates of change based on the magnitude of the rate of change; The threshold for identifying vibration points is dynamically adjusted based on the train's speed. Points in the time-mileage curves of the rate of change of angular velocity and the rate of change of acceleration that exceed the identification threshold are identified as abnormal vibration points. The size and color of the visualized vibration points are determined based on the vibration intensity, thus marking the vibration points in a visual manner.

8. The processor according to any one of claims 1 to 7, characterized in that, When the processor (100) is connected to the inertial measurement unit, the initial calibration module (110) performs the following state initialization steps: Based on the initial acceleration, angular velocity, and position data of the inertial measurement unit, the state vector is set; The state error covariance matrix is ​​initialized based on the characteristics of the inertial measurement unit; the state error covariance matrix includes several initial state vectors; The initial state vector includes the train's initial position, speed, and attitude.

9. A terminal for positioning subway trains, characterized in that, The terminal is connected to a processor (100) for subway train positioning to display the processing process and processing results of the processor (100); The terminal visually marks vibration points; for each detected abnormal vibration point, it not only marks its time and distance, but also uses color intensity to indicate the severity of the vibration. The processor (100) includes: The initial correction module (110) performs state estimation on the original acceleration and angular velocity data of the train collected by the inertial measurement unit based on the error Kalman filter algorithm, and dynamically corrects the measurement noise and system error by combining the system's motion model and the actual measurement value of the inertial measurement unit. The curve generation module (120) constructs the train's time-mileage curve and corrects the time-mileage curve based on the first correction ratio; The turning recognition module (130) calculates the total mileage within the turning segment and corrects the total mileage within the turning segment based on the second correction ratio of the turning segment; The abnormal vibration identification module (140) identifies abnormal vibration features based on the higher-order rate of change of the train's acceleration and angular velocity and marks them in a visual manner.

10. The terminal according to claim 9, characterized in that, The terminal displays the corrected time-mileage curve, which is used to intuitively show the changes in the train's mileage. The terminal displays curves drawn from the original data and curves drawn from the corrected data.

Images