A low-cost, all-regional speed measurement and positioning method and device for train control onboard equipment.

Abstract

This invention relates to the field of railway transportation and management, and provides a low-cost, all-area train control onboard equipment speed measurement and positioning method and device. If the satellite signal is in a failed state, an inertial sensor error model is established using a multi-sensor fusion method. The inertial sensor error is estimated by using the difference between the inertial sensor and other sensor measurements and feedback correction is performed. This method can accurately estimate the sensor error value even when the satellite signal is failed. By combining recursion and feedback correction, the final train status information is obtained, which improves the applicability of the method and device and solves the speed measurement and positioning problem in areas where satellite signals are prone to failure, such as mountainous valleys, tunnels, and station roof obstructions.

Description

Technical Field

[0001] This invention relates to the field of railway transportation and management, specifically to a low-cost, all-regional speed measurement and positioning method and device for train control onboard equipment. Background Technology

[0002] my country is implementing the Western Development Strategy in depth and further accelerating the development of railways in the west. However, railways in the west, such as the Qinghai-Tibet Railway, have large altitude variations, harsh environments, and sparse populations, which are not conducive to the operation and maintenance of trackside equipment. This has created a prominent demand for train control systems that require less trackside access and are maintenance-free.

[0003] Currently, there are two main types of speed measurement and positioning methods in existing train control onboard systems. Traditional train control onboard speed measurement and positioning systems mostly use photoelectric / Hall effect speed sensors to achieve speed measurement. However, when the speed sensor detects wheel slippage / skidding, it can only use mathematical models to consider the worst-case scenario for speed compensation, and cannot directly obtain effective measurement data from the sensor. Some systems have added radar systems to assist the speed sensor in achieving speed measurement; however, radar systems are costly and rely on ground-based physical transponders for positioning, which cannot meet the needs of railway construction, operation, and maintenance in western China. New train control systems use virtual transponder technology from the BeiDou satellite system to replace physical transponders for positioning. However, in mountainous terrain, tunnels, and station roof obstructions, satellite signal transmission may fail, preventing the acquisition of absolute positioning information for distance calibration. The accumulated errors in train speed and distance over time cannot meet the control accuracy requirements of the train control onboard system during long-term operation. Satellites measure three-dimensional position information, while speed sensors measure one-dimensional speed and distance information along the train's longitudinal direction. These two measurements need to be converted before they can be used together. However, the new train control system relies on electronic maps to achieve the dimensional conversion between satellite and speed sensor data, making it unable to align and calibrate inertial sensors. Furthermore, the method for using inertial sensor data is still unclear. Therefore, the new system will be unusable if satellites fail for an extended period.

[0004] Therefore, there is an urgent need for a low-cost, all-regional train control onboard equipment speed measurement and positioning method and device, so as to calculate the train's running speed, position and direction in real time. Summary of the Invention

[0005] The purpose of this invention is to provide a low-cost, all-regional speed measurement and positioning method and device for train control onboard equipment, so as to solve the problems mentioned in the background art.

[0006] This invention provides a low-cost, all-regional speed measurement and positioning method for train control onboard equipment, comprising the following steps:

[0007] The satellite signal status is obtained. If the satellite signal status is normal, inertial sensors and satellite receivers are used to obtain sensor error status information. If the satellite signal status is faulty, the first, second, and third correction methods are used to obtain the first, second, and third sensor error status information, respectively. Based on the normal and faulty satellite status, and combined with the information allocation coefficient determined by the confidence interval size corresponding to the sensor error status information, the sensor error status information is fused. The final train status information is obtained by combining recursion and feedback correction.

[0008] The first correction method uses an inertial sensor and a rotation sensor to obtain sensor error status information; the second correction method uses an inertial sensor and an inertial sensor to match location points with an electronic map to obtain second sensor error status information; and the third correction method uses an inertial sensor and a camera to obtain third sensor error status information.

[0009] Preferably, the acquisition of satellite signal status further includes: if the satellite signal status is normal, using an inertial sensor and a satellite receiver to acquire sensor error status information, and using a combination of recursion and feedback correction to obtain the final train status information.

[0010] Preferably, the first correction method includes:

[0011] Initial alignment of the inertial sensor;

[0012] Inertial sensor installation error angle calibration;

[0013] The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system. The attitude angles can be converted into an attitude matrix.

[0014] By using historically estimated train lateral and longitudinal speeds, a correlation model is established with the geographical characteristics of the current railway line to optimize the non-holonomic constraints of the speed sensor.

[0015] Data from the rotation speed sensor is acquired, and the optimized and constrained data from the rotation speed sensor is converted into three-dimensional data in the northeast-northeast coordinate system based on the attitude matrix of the inertial sensor.

[0016] Substitute the difference between the data from the speed sensor and the inertial sensor in the northeast-central coordinate system into the measurement equation, and estimate the error state information of the first inertial sensor using the error state equation.

[0017] Preferably, the historically estimated train lateral and longitudinal speeds are the train lateral and longitudinal speeds estimated by satellites under effective conditions, and the geographical features of the current railway line include at least the geographical features of curves and gradients.

[0018] Preferably, the method of obtaining the first sensor error state information using the first calibration method further includes:

[0019] Determine whether the wheelset with the speed sensor installed is spinning or slipping based on information from the inertial sensor and the speed sensor.

[0020] If it does not occur, the first correction method is used to estimate the error state information of the first sensor, and the longitudinal running speed of the train is obtained by measuring the rotation of the wheels using the speed sensor;

[0021] If this occurs, during the time of idling / coasting, the error state estimate of the wheelset when no idling / coasting occurs is used as the first sensor error state information.

[0022] Preferably, the second correction method includes:

[0023] Initial alignment of the inertial sensor;

[0024] Inertial sensor installation error angle calibration;

[0025] The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system, and the attitude angles are converted into an attitude matrix.

[0026] Use electronic maps to find the characteristic values ​​of the current railway line's parameters and mark the absolute coordinates of these characteristic values;

[0027] The absolute coordinate position of the railway line is obtained by feature matching between the real-time measurement values ​​of the inertial sensor and the feature values ​​of the current railway line parameters.

[0028] The difference between the inertial sensor measurement and the absolute coordinate position of the line is substituted into the measurement equation, and the error state information of the second sensor is estimated by the error state equation.

[0029] Preferably, the third correction method includes:

[0030] Initial alignment of the inertial sensor;

[0031] Inertial sensor installation error angle calibration;

[0032] The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system, and the attitude angles are converted into an attitude matrix.

[0033] Acquire train image information and use the train image information to calculate train visual positioning information;

[0034] The pose information of the inertial sensor is constrained into the train's visual positioning information. The error state information of the third sensor is estimated by the error state equation. The measurement error of the inertial sensor, the camera pose corresponding to the image, and the spatial position parameters corresponding to the feature points are used as state variables. The difference between the position and attitude of the camera and the inertial sensor is used as the observation to construct the observation model. The MSCKF filtering fusion algorithm of the inertial sensor and the camera is used to estimate the error of the inertial sensor. The error state information of the third inertial sensor is used to correct the navigation parameters in the inertial navigation calculation process to obtain the train's running speed and running position in the northeast-northeast coordinate system, thereby obtaining the error state information of the third sensor.

[0035] Preferably, acquiring train image information and using the train image information to calculate train visual positioning information specifically includes:

[0036] Train image information is obtained using a camera, and absolute positioning information is obtained through feature extraction, matching, and optimization estimation in order to calculate the train's visual positioning information.

[0037] Preferably, acquiring train image information and using the train image information to calculate train visual positioning information specifically includes:

[0038] Multiple images of a moving train are captured using a visual odometry system. The differences between these images are detected to measure the train's travel distance. The train's speed is obtained by comparing the differences between consecutive frames, thus calculating the train's visual positioning information.

[0039] Preferably, the low-cost, all-region train control on-board equipment speed measurement and positioning method further includes:

[0040] A federated Kalman filter is used to perform data fusion processing on the train sensor error state information. Based on the normal and failed satellite states, and considering the information allocation coefficients determined by the confidence interval size corresponding to the sensor error state information, if the state is failed, the first, second, and third sensor error state information obtained from the first, second, and third correction methods are fused to obtain the final sensor error state information. This fused train sensor error state information is then used to correct the corresponding navigation parameters in the calculation process, estimating the train's running speed and position in the northeast-northeast coordinate system to obtain the final train state information.

[0041] Preferably, the low-cost, all-region train control on-board equipment speed measurement and positioning method further includes:

[0042] A redundant architecture is used to acquire sensor measurement information. The redundant architecture includes two speed and distance measurement units consisting of two speed sensors and two inertial sensors, and two positioning units consisting of two satellite receivers, two cameras and an electronic map. The results of the two speed and distance measurement units and the two positioning units are compared and then fused by an algorithm model to obtain the speed measurement and positioning result.

[0043] This invention also provides a low-cost, all-region train control onboard equipment speed measurement and positioning device, comprising:

[0044] The fault identification module is used to acquire the satellite signal status and determine whether the satellite signal is in a failed state.

[0045] The speed measurement and positioning module is used to obtain the error status information of the first sensor, the second train sensor, and the third sensor respectively using the first correction method, the second correction method, and the third correction method if the satellite signal status is in a failed state. It uses a combination of recursion and feedback correction to correct the navigation parameters in the solution process using the sensor error status information obtained by the fusion module, that is, to feed it back into the system to obtain the final train status information.

[0046] The fusion module is used to fuse the sensor error state information based on the satellite's normal state and failure state, combined with the information allocation coefficient determined by the confidence interval size corresponding to the sensor error state information, to obtain the final sensor error state information. The failure state includes first sensor error state information, second sensor error state information, and third sensor error state information. Specifically, the first correction method uses an inertial sensor and a rotation speed sensor to obtain the first sensor error state information; the second correction method uses an inertial sensor and an inertial sensor to match location points with an electronic map to obtain the second sensor error state information; and the third correction method uses an inertial sensor and a camera to obtain the third sensor error state information under a third train state.

[0047] Compared to existing technologies, the present invention provides a low-cost, all-terrain speed measurement and positioning method and device for train control onboard equipment. It is compatible with data from multiple sensors, including existing speed measurement and positioning devices on the train, and also incorporates low-cost, universal sensors, including MEMS inertial sensors and cameras. It eliminates the need for computationally complex and costly sensors such as radar and lidar. Even when satellite signals fail, it can utilize existing onboard sensors to obtain train status information, fusing data from multiple sensors to obtain accurate train status information. This allows the train control onboard equipment to monitor the train's status in any location and environment, improving the applicability of the low-cost, all-terrain speed measurement and positioning method. The system adopts a two-out-of-two safety platform architecture design, offering high reliability and low cost, adapting to the development of western plateau railways, and meeting the prominent needs for less trackside access and maintenance-free operation. Attached Figure Description

[0048] Figure 1 Flowchart of a low-cost, all-region train control on-board equipment speed measurement and positioning method provided in this embodiment of the invention;

[0049] Figure 2 A schematic diagram of the federated Kalman filter structure provided in this embodiment of the invention;

[0050] Figure 3 A schematic diagram of the redundant structure provided in an embodiment of the present invention. Detailed Implementation

[0051] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0052] The components of the embodiments of the invention described and shown in the accompanying drawings can typically be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of the invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention.

[0053] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0055] This invention provides a low-cost, all-regional speed measurement and positioning method for train control onboard equipment, such as... Figure 1 As shown, the low-cost, all-region train control on-board equipment speed measurement and positioning method specifically includes:

[0056] The satellite signal status is obtained. If the satellite signal status is normal, inertial sensors and satellite receivers are used to obtain sensor error status information. If the satellite signal status is faulty, the first, second, and third correction methods are used to obtain the first, second, and third sensor error status information, respectively. Based on the normal and faulty satellite status, and combined with the information allocation coefficient determined by the confidence interval size corresponding to the sensor error status information, the sensor error status information is fused. The final train status information is obtained by combining recursion and feedback correction.

[0057] This invention provides a low-cost, all-area train control onboard equipment speed measurement and positioning method. When satellite signal status fails, it employs a federated Kalman filter algorithm, combined with an information allocation coefficient determined by the confidence interval size corresponding to sensor error status information, to fuse sensor error status information. Using a combination of recursive and feedback correction, it obtains accurate train status information, enabling the train control onboard equipment to monitor train status in any location and environment. This improves the applicability of the low-cost, all-area train control onboard equipment speed measurement and positioning method, making it suitable for all regions. Furthermore, this invention can fuse multiple correction methods to obtain train status information. In the case of multiple parallel methods, even if one correction method fails due to sensor malfunction, the other path correction methods can still obtain accurate train status information through result fusion, improving system reliability. Simultaneously, the first, second, and third correction methods use low-cost, universal sensors for data acquisition, eliminating the need to develop dedicated sensors or complex communication methods, thus improving system applicability while reducing system cost.

[0058] In some optional embodiments, the first correction method includes:

[0059] Initial alignment of the inertial sensor;

[0060] Inertial sensor installation error angle calibration;

[0061] The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system. The attitude angles can be converted into an attitude matrix.

[0062] By using historically estimated train lateral and longitudinal speeds, a correlation model is established with the geographical characteristics of the current railway line to optimize the non-holonomic constraints of the speed sensor.

[0063] Data from the rotation speed sensor is acquired, and the optimized and constrained data from the rotation speed sensor is converted into three-dimensional data in the northeast-northeast coordinate system based on the attitude matrix of the inertial sensor.

[0064] The difference between the data from the speed sensor and the inertial sensor in the northeast-central coordinate system is substituted into the measurement equation. The error state equation is used to estimate the error state information of the first sensor, and this error state information is then used to correct the navigation parameters in the inertial navigation calculation process, i.e., fed back into the system. The historically estimated lateral and longitudinal speeds of the train are those estimated by the satellite under effective conditions. The current geographical features of the railway line include at least curves and gradients. The first correction method provided by this invention can continuously bring the error state information closer to the true error value through feedback correction.

[0065] The first calibration method is used to obtain the error state information of the first sensor, which specifically includes:

[0066] Based on the information from the inertial sensor and the speed sensor, it is determined whether the wheelset with the speed sensor is engaged in freewheeling / slipping. If it is not engaged, the first correction method is used to estimate the error state information of the first sensor, and the longitudinal running speed of the train is obtained by measuring the rotation of the wheels using the speed sensor. If it is engaged, the error state estimate of the wheelset with the speed sensor engaged when it is not engaged in freewheeling / slipping is used as the error state information of the first sensor during the time when freewheeling / slipping occurs.

[0067] When the speed sensor is working properly, compared with the non-integrity constraints provided for speed sensor data in the prior art, the present invention optimizes the non-integrity constraints in a targeted manner. Specifically, it applies integrity constraints to the train field. In the non-integrity constraints, the speed in the horizontal and vertical directions of the vehicle coordinate system is zero. However, when going uphill or downhill, turning, or at high speeds, the non-integrity constraints are no longer accurate and will cause large errors. The present invention improves the accuracy of data conversion and calculation.

[0068] Existing technologies, when the speed sensor detects wheel slippage / skidding, can only use mathematical models to consider the worst-case scenario for speed compensation, failing to obtain effective measurement data. However, the method provided by this invention can still acquire short-term train status data using inertial sensors even when the speed sensor detects wheel slippage / skidding, ensuring stable operation of train monitoring.

[0069] The second correction method includes:

[0070] Initial alignment of the inertial sensor;

[0071] Inertial sensor installation error angle calibration;

[0072] The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system, and the attitude angles are converted into an attitude matrix.

[0073] Use electronic maps to find the characteristic values ​​of the current railway line's line parameters. The characteristic values ​​of the current railway line's line parameters include at least the curve curvature, the reciprocal of the curvature, and the line gradient, and mark the absolute coordinate position of the characteristic value.

[0074] The absolute coordinate position of the railway line is obtained by matching the real-time measurement values ​​of the inertial sensor with the feature values ​​of the line parameters of the current railway line. Furthermore, a pattern recognition method is used for feature matching.

[0075] The difference between the inertial sensor measurement and the absolute coordinate position of the line is substituted into the measurement equation, and the error state information of the second sensor is estimated from the error state equation. The second correction method provided by this invention can calibrate the inertial sensor using an electronic map, and through feedback correction, make the sensor error state information continuously approach the true error.

[0076] The method provided by this invention addresses the issue of inertial sensors failing to operate autonomously for extended periods when satellite signals are unavailable. To address this, an electronic map is used to estimate the inertial sensor error. By matching real-time inertial navigation system measurements with feature parameters in the electronic map, the train's absolute positioning information can be obtained and used as feedback correction data. This approach effectively estimates sensor errors when satellite signals fail, resulting in more accurate speed and position estimates and improving the accuracy of train status information.

[0077] The third correction method includes:

[0078] Initial alignment of the inertial sensor;

[0079] Inertial sensor installation error angle calibration;

[0080] The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system, and the attitude angles are converted into an attitude matrix.

[0081] Train image information is acquired, and train visual positioning information is calculated using the train image information. One optional embodiment involves using a camera to acquire train image information, obtaining absolute positioning information through feature extraction, matching, and optimization estimation, and then calculating the train visual positioning information. Another optional embodiment involves using a visual odometer to capture multiple images of a moving train, detecting the differences between the multiple images to measure the train's travel distance, and comparing the differences between consecutive frames to obtain the train's speed, thus calculating the train visual positioning information.

[0082] The pose information of the inertial sensor is constrained into the train's visual positioning information. The error state information of the third sensor is estimated by the error state equation. The measurement error of the inertial sensor, the camera pose corresponding to the image, and the spatial position parameters corresponding to the feature points are used as state variables. The difference between the position and attitude of the camera and the inertial sensor is used as the observation to construct the observation model. The MSCKF filtering fusion algorithm of the inertial sensor and the camera is used to estimate the error of the inertial sensor. The error state information of the third sensor is used to correct the navigation parameters in the inertial navigation solution process, that is, to feed back to the inertial navigation system so that the error state information continuously approaches the true error value. The train's running speed and running position in the northeast-northeast coordinate system are obtained to acquire the error state information of the third sensor.

[0083] As an optional embodiment, multiple cameras can be used to acquire images, specifically two cameras. In this embodiment of the invention, when satellite signals fail, the method is not limited to traditional image recognition and positioning methods based on track and trackside features. Camera technology is applied to measure the distance by comparing the images from two cameras, and the train's speed is obtained by comparing the differences between consecutive frames. The predictions from the inertial sensor are then corrected and updated, improving the accuracy of the status information. When satellite signals fail, camera technology is applied to the train speed measurement and positioning device. By analyzing relevant image sequences, the train's position and orientation are determined, and the inertial sensor error is effectively estimated, thereby obtaining more accurate train speed and position information.

[0084] When satellite signals fail in areas such as tunnels, speed and distance are recursively calculated using rotation speed sensors and inertial sensors, and kinematic constraints are added to optimize the calculation results. Furthermore, to improve the accuracy of train status information obtained in the above manner, key positioning point information is determined by matching inertial sensor measurements with feature parameters in electronic maps. Cameras can also be used for information fusion processing. Regardless of whether the satellite fails over a long distance or a short distance, there are backup methods for acquiring train status information, thus improving the applicability of the system.

[0085] Specifically, the train status information includes the train's real-time speed, position, and attitude. The train status information is obtained by fusing satellite normal and failure status information. The failure status information includes first sensor error status information, second sensor error status information, and third sensor error status information. This process involves: sub-filters estimating sensor errors using observation equations and providing these estimates to the main filter; determining information allocation coefficients based on the confidence intervals of the sensor error status information; using the main filter's estimate as the global estimate; and feedback resetting to improve the accuracy of each sub-filter. Finally, the real-time speed, position, and attitude of the train in the northeast-northeast coordinate system are obtained. This can be converted into one-dimensional speed and distance measurement information along the train's longitudinal direction using the attitude matrix, i.e., the train's running speed, position, and direction.

[0086] A federated Kalman filter is used to perform data fusion processing on the train sensor error state information. Based on the normal and failed satellite states, and considering the information allocation coefficients determined by the confidence interval size corresponding to the sensor error state information, if the state is failed, the first, second, and third sensor error state information obtained from the first, second, and third correction methods are fused to obtain the final sensor error state information. This fused train sensor error state information is then used to correct the corresponding navigation parameters in the calculation process, estimating the train's running speed and position in the northeast-northeast coordinate system to obtain the final train state information.

[0087] Specifically, such as Figure 2 As shown, a federated Kalman filter is used to fuse the sensor error state information obtained from the first, second, and third correction methods. The error state estimates using inertial and rotational speed sensors as observations are set as the first sub-filter; the estimates using inertial sensors and location points matched with the electronic map as observations are set as the second sub-filter; the estimates using inertial sensors and satellite receivers as observations are set as the third sub-filter; and the estimates using inertial sensors and cameras as observations are set as the fourth sub-filter. The federated Kalman information allocation coefficients are configured based on the confidence interval of the sensor error state information. A smaller confidence interval results in a higher information allocation coefficient. Accurate sensor error state information is obtained using inertial sensors, rotational speed sensors, electronic maps, satellite receivers, and cameras. Through feedback correction, the error state information continuously approaches the true error value, thereby obtaining the train's running speed and position in the northeast-southeast coordinate system, and thus acquiring train state information.

[0088] This invention employs a first, second, and third correction method to estimate sensor error state information. A federated Kalman filter is used to determine the information allocation coefficient based on the confidence interval corresponding to the sensor error state information; a smaller confidence interval results in a higher information allocation coefficient. Feedback correction continuously brings the error state information closer to the true error value. Error compensation for gyroscope and accelerometer drift is performed on the real-time measurement data of the inertial sensor. The train's running speed, position, and attitude information are obtained through inertial navigation system (INS) calculation. The estimated inertial sensor speed error, positioning error, and INS attitude misalignment angle are compensated into the calculation results to obtain the train's running speed and position in the northeast-northeast coordinate system.

[0089] Sensor malfunctions are inevitable during train operation. For example, speed sensors may slip or coast due to poor road conditions; satellite signals may be blocked in complex conditions such as tunnels and mountainous areas, posing a risk of failure; and camera output may contain errors due to factors such as camera parameters and measurement distance range. The method provided in this invention, based on fault diagnosis and prediction results, combines the confidence interval size corresponding to sensor error state information to configure federated Kalman information fusion coefficients, ultimately achieving a low-cost, all-terrain speed measurement and positioning method for train control onboard equipment. Furthermore, in existing technologies, the use of virtual transponder technology, which calibrates the cumulative speed and distance measurement error value at the transponder positioning point by correcting the measured distance value to the link distance from the previous transponder in the transponder message, can cause train position reversal. The method provided in this invention, during error correction, periodically compares the absolute positioning point of the BeiDou receiver with the relative measurement value to estimate the systematic and random errors of the speed sensor and inertial sensor. A partial feedback mechanism optimizes the estimated value, making the fusion model estimation result closer to the true value, thereby avoiding train position reversal caused by absolute position calibration.

[0090] Furthermore, such as Figure 3 As shown, a redundant architecture is used to acquire sensor measurement information. The redundant architecture includes two speed and distance measurement units consisting of two speed sensors and two inertial sensors, and two positioning units consisting of two satellite receivers, two cameras, and an electronic map. The results of the two speed and distance measurement units and the two positioning units are compared and then fused by an algorithm model to obtain the speed and positioning results. This invention provides a redundant architecture that outputs the train's running speed and position through dual-machine comparison, further improving the reliability of the system.

[0091] This invention also provides a low-cost, all-terrain train control onboard equipment speed measurement and positioning device, comprising:

[0092] The fault identification module is used to acquire the satellite signal status and determine whether the satellite signal is in a failed state.

[0093] The speed measurement and positioning module is used to obtain the error status information of the first sensor, the second train sensor, and the third sensor respectively using the first correction method, the second correction method, and the third correction method if the satellite signal status is in a failed state. It uses a combination of recursion and feedback correction to correct the navigation parameters in the solution process using the sensor error status information obtained by the fusion module, that is, to feed it back into the system to obtain the final train status information.

[0094] The fusion module is used to fuse the sensor error state information based on the satellite's normal state and failure state, combined with the information allocation coefficient determined by the confidence interval size corresponding to the sensor error state information, to obtain the final sensor error state information. The failure state includes first sensor error state information, second sensor error state information, and third sensor error state information. Specifically, the first correction method uses an inertial sensor and a rotational speed sensor to acquire the first sensor error state information under a first train state; the second correction method uses an inertial sensor and an inertial sensor to match location points with an electronic map to acquire the second sensor error state information under a second train state; and the third correction method uses an inertial sensor and a camera to acquire the third sensor error state information under a third train state.

[0095] Furthermore, the fault identification module is also used to acquire data from the speed sensor and inertial sensor to determine whether the sensor mounting shaft is spinning / slipping, acquire camera data, and determine whether the camera data meets the preset reliability requirements.

[0096] Furthermore, the present invention also includes a comparison module: used to compare the measurement data and output results based on the comparison function in the two-out-of-two safe computer platform architecture, and finally output train control vehicle onboard control data that meets the requirements of reliability and safety.

[0097] It is not difficult to see that this embodiment is a device embodiment corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and will not be repeated here to reduce repetition. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the first embodiment.

[0098] 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. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0099] Finally, it should be noted that this invention is not limited to the optional embodiments described above, and anyone can derive other various forms of products under the guidance of this invention. The specific embodiments described above should not be construed as limiting the scope of protection of this invention, which should be determined by the claims, and the specification can be used to interpret the claims.

Claims

1. A low-cost, all-regional speed measurement and positioning method for train control onboard equipment, characterized in that, Includes the following steps: The satellite signal status is obtained. If the satellite signal status is normal, inertial sensors and satellite receivers are used to obtain sensor error status information. If the satellite signal status is faulty, the first, second, and third correction methods are used to obtain the first, second, and third sensor error status information, respectively. Based on the normal and faulty satellite status, and combined with the information allocation coefficient determined by the confidence interval size corresponding to the sensor error status information, the sensor error status information is fused. The final train status information is obtained by combining recursion and feedback correction. The first correction method uses an inertial sensor and a rotation sensor to obtain sensor error status information; the second correction method uses an inertial sensor and an inertial sensor to match the location points with the electronic map to obtain second sensor error status information; and the third correction method uses an inertial sensor and a camera to obtain third sensor error status information. The first correction method includes: Initial alignment of the inertial sensor; Inertial sensor installation error angle calibration; The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system. The attitude angles can be converted into an attitude matrix. By using historically estimated train lateral and longitudinal speeds, a correlation model is established with the geographical characteristics of the current railway line to optimize the non-holonomic constraints of the speed sensor. Data from the rotation speed sensor is acquired, and the optimized and constrained data from the rotation speed sensor is converted into three-dimensional data in the northeast-northeast coordinate system based on the attitude matrix of the inertial sensor. Substitute the difference between the data from the speed sensor and the inertial sensor in the northeast-central coordinate system into the measurement equation, and estimate the error state information of the first inertial sensor using the error state equation.

2. The low-cost, all-region train control on-board equipment speed measurement and positioning method as described in claim 1, characterized in that, The acquisition of satellite signal status also includes: if the satellite signal status is normal, using an inertial sensor and a satellite receiver to acquire sensor error status information, and using a combination of recursion and feedback correction to obtain the final train status information.

3. The low-cost, all-region train control on-board equipment speed measurement and positioning method as described in claim 1, characterized in that, The historical estimates of train lateral and longitudinal speeds are the train lateral and longitudinal speeds estimated by satellites under effective conditions. The geographical features of the current railway line include at least the geographical features of curves and gradients.

4. The low-cost, all-regional speed measurement and positioning method for train control onboard equipment as described in claim 1, characterized in that, The method of obtaining the error state information of the first sensor using the first calibration method further includes: Determine whether the wheelset with the speed sensor installed is spinning or slipping based on information from the inertial sensor and the speed sensor. If it does not occur, the first correction method is used to estimate the error state information of the first sensor, and the longitudinal running speed of the train is obtained by measuring the rotation of the wheels using the speed sensor; If this occurs, during the time of idling / coasting, the error state estimate of the wheelset when no idling / coasting occurs is used as the first sensor error state information.

5. The low-cost, all-region train control on-board equipment speed measurement and positioning method as described in claim 1, characterized in that, The second correction method includes: Initial alignment of the inertial sensor; Inertial sensor installation error angle calibration; The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system, and the attitude angles are converted into an attitude matrix. Use electronic maps to find the characteristic values ​​of the current railway line's parameters and mark the absolute coordinates of these characteristic values; The absolute coordinate position of the railway line is obtained by feature matching between the real-time measurement values ​​of the inertial sensor and the feature values ​​of the current railway line parameters. The difference between the inertial sensor measurement and the absolute coordinate position of the line is substituted into the measurement equation, and the error state information of the second sensor is estimated by the error state equation.

6. The low-cost, all-regional speed measurement and positioning method for train control onboard equipment as described in claim 1, characterized in that, The third correction method includes: Initial alignment of the inertial sensor; Inertial sensor installation error angle calibration; The velocity, position, and attitude angles in the northeast-central coordinate system are calculated from the measured values ​​of triaxial acceleration and angular acceleration in the inertial coordinate system, and the attitude angles are converted into an attitude matrix. Acquire train image information and use the train image information to calculate train visual positioning information; The pose information of the inertial sensor is constrained into the train's visual positioning information. The error state information of the third sensor is estimated by the error state equation. The measurement error of the inertial sensor, the camera pose corresponding to the image, and the spatial position parameters corresponding to the feature points are used as state variables. The difference between the position and attitude of the camera and the inertial sensor is used as the observation to construct the observation model. The MSCKF filtering fusion algorithm of the inertial sensor and the camera is used to estimate the error of the inertial sensor. The error state information of the third inertial sensor is used to correct the navigation parameters in the inertial navigation calculation process to obtain the train's running speed and running position in the northeast-northeast coordinate system, thereby obtaining the error state information of the third sensor.

7. The low-cost, all-region train control on-board equipment speed measurement and positioning method as described in claim 6, characterized in that, Acquiring train image information and using it to calculate train visual positioning information specifically includes: Train image information is obtained using a camera, and absolute positioning information is obtained through feature extraction, matching, and optimization estimation in order to calculate the train's visual positioning information.

8. The low-cost, all-region train control on-board equipment speed measurement and positioning method as described in claim 6, characterized in that, Acquiring train image information and using it to calculate train visual positioning information specifically includes: Multiple images of a moving train are captured using a visual odometry system. The differences between these images are detected to measure the train's travel distance. The train's speed is obtained by comparing the differences between consecutive frames, thus calculating the train's visual positioning information.

9. The low-cost, all-regional speed measurement and positioning method for train control onboard equipment as described in claim 1, characterized in that, The aforementioned low-cost, all-region train control on-board equipment speed measurement and positioning method also includes: A federated Kalman filter is used to perform data fusion processing on the train sensor error state information. Based on the normal and failed states of the satellite, and combined with the information allocation coefficient determined by the size of the confidence interval corresponding to the sensor error state information, if it is a failed state, the first, second, and third sensor error state information obtained by the first, second, and third correction methods are fused to obtain the final sensor error state information. The fused train sensor error state information is then used to correct the corresponding navigation parameters in the calculation process, and the train's running speed and position in the northeast-northeast coordinate system are calculated to obtain the final train state information.

10. The low-cost, all-region train control on-board equipment speed measurement and positioning method as described in claim 1, characterized in that, The aforementioned low-cost, all-region train control on-board equipment speed measurement and positioning method further includes: A redundant architecture is used to acquire sensor measurement information. The redundant architecture includes two speed and distance measurement units consisting of two speed sensors and two inertial sensors, and two positioning units consisting of two satellite receivers, two cameras and an electronic map. The results of the two speed and distance measurement units and the two positioning units are compared and then fused by an algorithm model to obtain the speed measurement and positioning result.

11. A low-cost, all-terrain train control on-board speed measurement and positioning device, used to implement the positioning method as described in any one of claims 1-10, comprising: The fault identification module is used to acquire the satellite signal status and determine whether the satellite signal is in a failed state. The speed measurement and positioning module is used to obtain the error status information of the first sensor, the second train sensor, and the third sensor respectively using the first correction method, the second correction method, and the third correction method if the satellite signal status is in a failed state. It uses a combination of recursion and feedback correction to correct the navigation parameters in the solution process using the sensor error status information obtained by the fusion module, that is, to feed it back into the system to obtain the final train status information. The fusion module is used to fuse the sensor error state information based on the satellite's normal state and failure state, combined with the information allocation coefficient determined by the confidence interval size corresponding to the sensor error state information, to obtain the final sensor error state information. The failure state includes first sensor error state information, second sensor error state information, and third sensor error state information. The first correction method uses an inertial sensor and a rotation speed sensor to obtain the first sensor error state information; the second correction method uses an inertial sensor and an inertial sensor to match location points with an electronic map to obtain the second sensor error state information; and the third correction method uses an inertial sensor and a camera to obtain the third sensor error state information under a third train state. The fault identification module is also used to acquire data from the speed sensor and inertial sensor, determine whether the sensor mounting shaft is spinning / slipping, acquire camera data, and determine whether the camera data meets the preset reliability requirements. It also includes a comparison module: used to compare measurement data and output results based on the comparison function in the two-out-of-two safe computer platform architecture, and finally output train control vehicle onboard control data that meets the requirements of reliability and safety.

Images