Train positioning method and system

Abstract

The invention provides a train positioning method and system, and relates to the technical field of train localization, and the method comprises the steps: building a positioning SLAM module through a synchronous map during train detection, carrying out the training through the information collected by a conventional positioning unit at least comprising a speedometer and a beacon reader, and carrying out the recognition of the positioning SLAM module; generating and maintaining information which is marked with an area which is used for marking that the SLAM positioning result in the train operation line has reached the preset reliability; during the train operation period, whether the current position meets the SLAM positioning starting condition or not is judged based on the information; if yes, adopting an SLAM positioning result as a safe position; and if not, adopting a traditional positioning result. According to the method, the SLAM is trained and calibrated through traditional positioning information to form a credible area map, and a positioning source is dynamically selected according to the credible area map, so that progressive and high-reliability application of SLAM positioning in train operation is realized, and a compatible scheme is provided for transformation of an existing system and deployment of a virtual transponder.

Description

Technical Field

[0001] This invention relates to the field of train positioning technology, and in particular to a train positioning method and system. Background Technology

[0002] In the field of train operation control, continuous and precise train positioning is crucial for ensuring operational safety and efficiency. Currently widely used positioning systems typically consist of onboard equipment and trackside equipment working together.

[0003] Currently, Simultaneous Localization and Mapping (SLAM) technology, based on sensors such as LiDAR, cameras, and Inertial Measurement Units (IMUs), is beginning to be applied to rail transit environmental perception. However, relying solely on SLAM for train localization still faces unique challenges: the operating environment contains many repetitive structural features (such as tracks and overhead contact line fixtures), making feature matching difficult; trains travel long distances and follow single trajectories, and the accumulation of IMU errors in unpredictable directions can easily lead to drift; trains operate between multiple tracks, making it difficult to repeatedly scan the same scene to achieve closed-loop correction. These issues make it difficult for pure SLAM solutions to independently meet the requirements of the highest safety integrity level for train operation in terms of reliability and accuracy.

[0004] Therefore, under the premise of ensuring operational safety and continuity, designing a solution that can use an existing high-reliability positioning system as a benchmark to conduct online training and reliability assessment of SLAM positioning performance, and dynamically and safely switch positioning sources accordingly, has become the key to promoting the implementation of new technologies and realizing the low-cost, gradual transformation of existing systems. Summary of the Invention

[0005] This invention provides a train positioning method and system to solve the technical problems of insufficient reliability of pure SLAM positioning in complex train operating environments, and difficulty in compatibility with and gradual replacement of existing trackside beacon systems.

[0006] This invention provides a train positioning method, comprising: during train inspection, constructing a positioning SLAM module using a synchronous map, training it with conventional positioning information from conventional positioning units to generate and maintain SLAM reliable area information; the conventional positioning unit includes at least an odometer and a beacon reader, and the SLAM reliable area information is used to identify areas in the train's operating line where the SLAM positioning results have reached a preset reliability; during train operation, determining whether the train's current position meets the SLAM positioning activation conditions based on the SLAM reliable area information; when the SLAM positioning activation conditions are met, using the SLAM positioning result output by the SLAM module as the train's safe position; when the SLAM positioning activation conditions are not met, using the conventional positioning result output by the conventional positioning unit as the train's safe position.

[0007] According to a train positioning method provided by the present invention, a SLAM module is trained using traditional positioning information to generate and maintain SLAM reliable area information. The method includes: the SLAM module receiving raw measurement information output by a traditional positioning unit, the raw measurement information including speed and distance information output by an odometer and beacon information output by a beacon reader; compensating SLAM positioning data obtained based on an environmental perception sensor using the raw measurement information to obtain a first positioning result; receiving train safety position information determined by the traditional positioning unit; iteratively optimizing the first positioning result using the train safety position information to obtain a second positioning result; and determining a geographical area with acceptable SLAM positioning accuracy based on the degree of agreement between the second positioning result and the train safety position information to generate SLAM reliable area information.

[0008] According to a train positioning method provided by the present invention, during train detection, the SLAM module does not output positioning results for train positioning control to the on-board controller; during train operation, the SLAM module outputs SLAM positioning results to the on-board controller based on SLAM trusted area information and the current position of the train, and when the SLAM positioning activation conditions are met.

[0009] According to the train positioning method provided by the present invention, the conditions for not meeting the SLAM positioning activation include any of the following: SLAM module failure; SLAM trusted area information unavailable; or the current position of the train not covered by SLAM trusted area information.

[0010] According to a train positioning method provided by the present invention, the method further includes: when there are both valid SLAM positioning results and traditional positioning results at the current position of the train, performing consistency verification on the two to obtain a deviation value; if the deviation value exceeds a preset threshold, performing positioning anomaly processing.

[0011] According to a train positioning method provided by the present invention, positioning anomaly handling includes: initiating positioning re-establishment by the on-board controller; and / or, recording abnormal data by the SLAM module and updating SLAM trusted area information using the abnormal data.

[0012] According to a train positioning method provided by the present invention, SLAM trusted area information can be fused and dynamically updated based on training data or abnormal data provided by multiple trains.

[0013] According to a train positioning method provided by the present invention, the area identified by SLAM trusted area information includes areas where ground beacons have been deployed and areas where ground beacons have not been deployed.

[0014] According to a train positioning method provided by the present invention, the method further includes: determining the intersection area of ​​SLAM trusted area information of all individual trains; if the intersection area is at least one discrete geographical location point, then when the train runs to the geographical location point, it is confirmed that it meets the SLAM positioning activation condition, and the SLAM positioning result output by the SLAM module is used as the safe position of the train.

[0015] This invention also provides a train positioning system, comprising: a conventional positioning unit, including at least an odometer and a beacon reader, for providing conventional positioning information; a synchronous map construction positioning SLAM module, configured to train using conventional positioning information from the conventional positioning unit during train inspection to generate and maintain SLAM reliable area information; the SLAM reliable area information is used to identify areas in the train's operating line where the SLAM positioning results have reached a preset reliability; and an onboard controller, communicatively connected to the conventional positioning unit and the SLAM module, configured to determine whether the train's current position meets the SLAM positioning activation conditions based on the SLAM reliable area information during train operation; when the SLAM positioning activation conditions are met, the SLAM positioning result output by the SLAM module is used as the train's safe position; when the SLAM positioning activation conditions are not met, the conventional positioning result output by the conventional positioning unit is used as the train's safe position.

[0016] The present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the train positioning method described in any of the above claims.

[0017] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the train positioning method described in any of the preceding claims.

[0018] The train positioning method and system provided by this invention constructs a reliable electronic map of the SLAM region by training and calibrating the SLAM module using highly reliable traditional positioning information during a specific period. This effectively overcomes the unreliability of SLAM positioning caused by repetitive environmental features and accumulated errors of the IMU. During operation, the positioning source is dynamically determined based on this map. The SLAM positioning result is adopted within the reliable region, and the system reverts to traditional positioning outside the region or under abnormal conditions. This ensures that trains can obtain continuous and safe position information under any track conditions (including multi-track areas). Thus, without interfering with the existing safety logic, this provides a compatible and smooth transition technical solution for eventually replacing physical beacons with virtual transponders, effectively solving the problem of integrating emerging sensing and positioning technologies with mature trackside beacon systems. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of a train positioning system provided in an embodiment of the present invention.

[0021] Figure 2 This is a flowchart illustrating a train positioning method provided in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0024] This invention provides a train positioning system, which is used to achieve the following Figure 2 Any train positioning method in the corresponding method embodiments.

[0025] Figure 1This is a schematic diagram of a train positioning system provided in an embodiment of the present invention. The train positioning system includes a conventional positioning unit 110, a SLAM module 120, and an on-board controller 130.

[0026] The conventional positioning unit 110 includes an odometer 111 and a beacon reader 112. The odometer 111 is a conventional ODO odometer, used to output train speed information and cumulative distance information; the beacon reader 112 is a conventional beacon reader, used to output beacon ID and beacon content.

[0027] In this embodiment of the invention, the vehicle controller 130 is communicatively connected to the conventional positioning unit 110 and the SLAM module 120.

[0028] For example, the vehicle controller 130 can be connected to the odometer 111, beacon reader 112 and SLAM module 120 via a communication bus 140, which is used to transmit input and output information of each device to form a complete information interaction link.

[0029] In this embodiment of the invention, the SLAM module 120 can receive the positioning information from the sensor group 150 and compensate for the positioning information using conventional information collected by the odometer 111 and the beacon reader 112.

[0030] For example, sensor group 150 may include LiDAR, camera, and IMU.

[0031] Specifically, the inputs of the SLAM module 120 include all information on the communication bus 140, specifically covering the speed information and cumulative distance information output by the conventional ODO odometer, the beacon ID and beacon content output by the conventional beacon reader, and the train in-use safe position information output by the on-board controller 130. Its output is the train position and speed determined by SLAM positioning.

[0032] In some embodiments, the SLAM module 120 is configured to: train using conventional positioning information from conventional positioning units during train detection (i.e., the positioning annotation and training phase) to generate and maintain SLAM trusted region information.

[0033] The SLAM trusted area information is used to identify areas along the train's operating line where the SLAM positioning results have reached a preset reliability. Preset reliability means meeting the positioning accuracy requirements of relevant standards for train operation safety and dispatch management.

[0034] In some embodiments, the on-board controller 130 is configured to: during train operation (i.e., the formal operation phase), determine whether the current position of the train meets the SLAM positioning activation conditions based on SLAM trusted area information; when the SLAM positioning activation conditions are met, use the SLAM positioning result output by the SLAM module 120 as the safe position of the train; when the SLAM positioning activation conditions are not met, use the conventional positioning result output by the conventional positioning unit 110 as the safe position of the train. Optionally, the train positioning system also includes independent BeiDou devices at the front and rear and Safety Integrity Level 4 (SIL4) safety devices.

[0035] In this embodiment of the invention, the independent BeiDou devices at the front and rear ends are respectively installed at the front and rear ends of the train, and have the functions of data storage during hibernation and location reporting during wake-up; the SIL4 safety device is a safety-related device that meets the SIL4 requirements, used to store the synchronization information of the independent BeiDou devices at the front and rear ends, and to verify whether the position offset of the front and rear ends meets the set requirements.

[0036] For example, the independent Beidou devices at the beginning and end and the SIL4 safety devices establish communication with the vehicle controller 130 and SLAM module 120 through the communication bus 140 to ensure information synchronization and interaction.

[0037] In some embodiments, the vehicle controller 130 is also configured to perform security verification-related logic, including receiving reporting information from independent BeiDou devices at both ends and triggering security warnings or control commands after cold movement detection.

[0038] It should be noted that a detailed description of the steps performed by the SLAM module 120 and the vehicle controller 130 can be found in the following... Figure 2 The relevant descriptions in the corresponding method embodiments.

[0039] The following is combined Figure 2 The train positioning method of the present invention is described. The entity performing this method may be as described above. Figure 1 The train positioning system shown can also be a device or module within a train positioning system.

[0040] Figure 2 This is a flowchart illustrating the train positioning method provided by the present invention, as shown below. Figure 2 As shown, the method includes the following: S201. During train inspection, a localization SLAM module is constructed using a synchronous map and trained using traditional localization information from traditional localization units to generate and maintain SLAM trusted area information.

[0041] In this embodiment of the invention, SLAM trusted area information is used to identify areas in the train running line where the SLAM positioning results have reached a preset reliability.

[0042] For example, the areas identified by the SLAM trusted area information include areas with deployed ground beacons and areas without deployed ground beacons. Through training and annotation, the trusted area can completely cover the entire train operating area, or it can only cover a portion of the area that meets the accuracy requirements.

[0043] Specifically, after the train stops at any point and starts, each module starts working. This detection period is the positioning, labeling, and training phase. During this phase, the SLAM module does not output positioning results for train positioning control to the onboard controller, but focuses only on optimizing its own positioning results and building a reliable region.

[0044] Thus, by completing the verification of SLAM positioning accuracy and area calibration in an independent training phase, this invention avoids interference with the normal positioning control of the train during the training process, ensuring operational safety while laying the foundation for the reliable application of SLAM positioning.

[0045] In some embodiments, the SLAM module may receive the raw measurement information output by the conventional positioning unit and use the raw measurement information to compensate the SLAM positioning data obtained based on the environmental perception sensor to obtain a first positioning result.

[0046] In this embodiment of the invention, the raw measurement information includes speed information and distance information output by the odometer, as well as beacon information output by the beacon reader.

[0047] For example, the conventional ODO odometer in the conventional positioning unit outputs the train's running speed and cumulative travel distance in real time, and the conventional beacon reader outputs the unique identifier of the beacon, i.e., the beacon ID, as well as the beacon content such as the line parameters stored in the beacon when passing a ground beacon; the SLAM module generates initial SLAM positioning data based on environmental feature information (such as trackside structures, catenary suspension points, etc.) collected by lidar, cameras, and IMU.

[0048] Specifically, due to the repetitive features in the train operating environment (such as the repetitive rails on elevated sections and the suspension clamps of the power grid), and the deviation that the IMU will produce after long-term operation, the initial SLAM positioning data will have errors. The SLAM module uses the received speed, cumulative distance, beacon ID and beacon content as compensation basis to correct the initial SLAM positioning data, reduce the positioning drift caused by the repetition of environmental features and IMU deviation, and thus obtain a first positioning result with better accuracy.

[0049] Furthermore, the system receives train safety location information determined by a traditional positioning unit and uses this information to iteratively optimize the first positioning result to obtain a second positioning result.

[0050] For example, after the on-board controller establishes its position using traditional methods, it outputs the train's safe location information. This traditional method of establishing position means that the on-board controller relies solely on the speed and cumulative distance information output by the traditional ODO odometer and the beacon ID and beacon content output by the traditional beacon reader to collaboratively calculate the train's position and complete the positioning. The positioning results have been verified through long-term operation and meet the SIL4 safety integrity level requirements.

[0051] Specifically, after receiving the train's safe location information, the SLAM module uses it as a baseline and employs at least one of clustering algorithms and deep learning algorithms to iteratively train and correct the first positioning result.

[0052] For example, a positioning error prediction model can be constructed using deep learning algorithms. The deviation between the first positioning result and the safe location information can be used as training samples to continuously optimize the model parameters, so that the SLAM positioning result can continuously approach the safe location information and finally obtain the second positioning result. For example, clustering algorithms can be used to classify and aggregate multiple sets of location data, and outlier data points can be removed.

[0053] Furthermore, based on the degree of agreement between the second positioning result and the train's safe location information, the geographical area with acceptable SLAM positioning accuracy is determined to generate SLAM reliable area information.

[0054] For example, a positioning deviation threshold is set (this positioning deviation threshold conforms to the train operation safety standards, such as the positioning deviation threshold being within the range of ±0.5 meters to ±2 meters, and the specific value is determined according to the line operation level).

[0055] Specifically, when the deviation between the second positioning result and the train's safe location information is lower than the positioning deviation threshold multiple times consecutively, the geographical area is determined to be a qualified SLAM positioning area; if the deviation is higher than the positioning deviation threshold, the area is temporarily not included in the trusted area.

[0056] Thus, by using a quantified consistency judgment standard, this invention clarifies the reliable application scope of SLAM positioning, providing a clear and executable basis for selecting positioning methods in subsequent operational phases.

[0057] In this embodiment of the invention, during train inspection, the SLAM module does not output positioning results for train positioning control to the on-board controller.

[0058] For example, the first few runs after the train starts can be set as a detection period. During this period, the on-board controller only relies on the traditional positioning unit to complete the positioning. The output of the SLAM module is only used for its own training and trusted region construction, and does not participate in the decision of the train's safe position.

[0059] Specifically, the core tasks of the SLAM module during detection are to collect traditional positioning information, optimize its own positioning algorithm, and calibrate reliable areas to avoid affecting train operation safety due to the incomplete maturity of SLAM positioning.

[0060] Thus, by separating the training phase and the positioning control phase, this invention achieves a gradual improvement in SLAM positioning capabilities while ensuring operational safety, thus balancing safety and technological innovation.

[0061] S202. During train operation, determine whether the current position of the train meets the conditions for SLAM positioning activation based on SLAM trusted area information.

[0062] In some embodiments, the on-board controller can perform region matching and judgment by combining a pre-stored SLAM trusted area information map with the train's current location information obtained by itself through traditional positioning or other methods.

[0063] For example, the SLAM trusted area information map is stored in the form of an electronic map, which includes the geographical coordinate range of each trusted area. The on-board controller obtains the current latitude and longitude, line mileage and other location information of the train in real time and compares it with the trusted area range in the electronic map.

[0064] Specifically, if the coordinates of the train's current location fall within the geographical coordinates of a trusted area, then the SLAM positioning activation condition is met; if it falls outside the trusted area or no corresponding trusted area information is found, then the activation condition is not met.

[0065] In this embodiment of the invention, during train operation, the SLAM module outputs the SLAM positioning result to the on-board controller based on the SLAM trusted area information and the current position of the train, and when the SLAM positioning activation conditions are met.

[0066] For example, the SLAM module receives the train's current position information sent by the on-board controller via the communication bus, and combines it with its stored SLAM trusted area information to determine whether the current position belongs to a trusted area. If it does, the positioning result output function is activated, and the SLAM-positioned train position and speed information is sent to the on-board controller; if it does not, the output is stopped, or an invalid flag is output.

[0067] Specifically, the SLAM module outputs positioning results only within trusted areas during operation, avoiding the output of erroneous positioning information in unreliable areas and ensuring the security and availability of positioning data.

[0068] Thus, this invention further enhances the security of the positioning system through the conditional output mechanism of the SLAM module, providing reliable support for the dynamic selection of positioning methods.

[0069] S203. When the SLAM positioning activation conditions are met, the SLAM positioning result output by the SLAM module will be used as the safe position of the train.

[0070] In some embodiments, the intersection region of SLAM trusted region information for all individual trains can be determined.

[0071] For example, for multiple trains on the same operating line, each train will generate its own SLAM trusted area information after its respective detection period. The trusted area information of all trains will be collected through the ground control center or vehicle-to-vehicle communication, and the intersection of their geographical ranges will be calculated.

[0072] Specifically, the intersection area is the area where all trains are determined to have qualified SLAM positioning accuracy. The reliability of SLAM positioning in this area has been verified by multiple trains and has higher credibility.

[0073] Furthermore, if the intersection area is at least one discrete geographic location, when the train runs to the geographic location, it is confirmed that it meets the SLAM positioning activation conditions, and the SLAM positioning result output by the SLAM module is used as the safe position of the train.

[0074] For example, if the intersection of the SLAM trusted regions of all trains is a number of discrete points on the line (such as specific mileage markers, the midpoint of an unobstructed straight section, etc.), these discrete points form the application scenario of the virtual transponder.

[0075] Among them, a virtual transponder refers to a virtual positioning identifier that does not require a physical trackside beacon unit and achieves the equivalent transponder positioning function through SLAM positioning.

[0076] Specifically, when the train reaches these discrete points, the onboard controller determines that the SLAM positioning activation conditions are met and directly uses the positioning result output by the SLAM module as the safe position, without relying on physical beacons, thus realizing the positioning function of the virtual transponder.

[0077] Thus, this invention forms a virtual transponder through discrete intersection points, reducing the deployment requirements of physical beacons, lowering the cost of line construction and maintenance, and simultaneously being compatible with the application scenarios of virtual transponders.

[0078] In one alternative implementation, when the SLAM trusted area information of a single train covers the entire line, the existing onboard traditional beacon readers and traditional ODO odometers of the train can be gradually removed.

[0079] For example, after long-term testing and training, a train on a certain line has a SLAM trusted area that fully covers the entire line and the positioning accuracy continuously meets the operational requirements. At this point, the traditional beacon reader and ODO odometer on the train can be gradually removed, and positioning can be completed solely by the SLAM module and the on-board controller.

[0080] In another alternative implementation, when the SLAM trusted area information of all trains along the entire line covers the entire line, the physical beacon units arranged along the trackside of the entire line can be further removed.

[0081] For example, when the SLAM trusted area of ​​all trains on a certain line completely covers the entire line and the positioning accuracy continuously meets the operational requirements, the physical beacon units along the line can be removed to achieve full-line SLAM positioning.

[0082] Specifically, during the dismantling process, the core safety logic of the on-board controller remains unchanged. Only the positioning input source is switched to achieve a seamless replacement or modification of the safety positioning function without adjusting the original train driving operation and dispatch management process.

[0083] Thus, this invention achieves a smooth transition from traditional positioning to SLAM positioning through a gradual equipment removal scheme, taking into account both the utilization efficiency of existing equipment and the application of new technologies, thereby reducing the cost and risk of modification.

[0084] S204. When the SLAM positioning activation conditions are not met, the traditional positioning result output by the traditional positioning unit shall be used as the safe position of the train.

[0085] Optionally, the conditions for enabling SLAM positioning are not met, including any of the following: SLAM module failure, SLAM trusted area information unavailable, or the train's current position not covered by SLAM trusted area information.

[0086] In one example, SLAM module failure means that the SLAM module is unable to receive input information from the sensor group or traditional positioning unit, or the output positioning result exceeds the reasonable error range.

[0087] For example, a malfunction in the lidar may prevent the collection of environmental features, or the camera may be obstructed, rendering the image information invalid.

[0088] Specifically, the vehicle controller monitors the working status of the SLAM module in real time through the communication bus. If it fails to receive a valid output from the SLAM module multiple times in a row, or receives a fault alarm signal from the SLAM module, it determines that the SLAM module has failed and immediately switches to the traditional positioning method.

[0089] In another example, the unavailability of SLAM trusted area information refers to situations such as the SLAM trusted area information not being generated, information loss due to damage to the storage medium, or the failure of the original trusted area information due to significant changes in the line environment (such as the addition of new structures or track modifications).

[0090] Specifically, after the vehicle controller starts up, it will first verify the integrity and validity of the SLAM trusted area information. If the information is missing, has an incorrect format, or has expired, it will be determined to be unusable. At this time, the vehicle controller will continue to use the traditional positioning method and trigger the SLAM module to re-enter the detection period to regenerate the trusted area information.

[0091] In another example, the train's current position not being covered by SLAM trusted area information refers to the train running to a newly added section of the line, a section that has not been trained by SLAM, or a section in the original trusted area where the SLAM positioning accuracy has decreased due to environmental changes.

[0092] Specifically, after comparing the train's current position with the SLAM trusted area information map, the on-board controller finds that there is no corresponding trusted area record for the current position. At this time, it relies solely on the traditional ODO odometer and beacon reader to collaboratively calculate the train's position to ensure that the positioning accuracy meets operational requirements.

[0093] Thus, by using traditional positioning as a fallback solution, this invention achieves continuous positioning of the entire train line, ensuring operational safety regardless of whether the SLAM trusted area is covered.

[0094] In the train positioning method provided by this invention, a reliable electronic map of the SLAM region is constructed by training and calibrating the SLAM module using highly reliable traditional positioning information during a specific period. During operation, the positioning source is dynamically determined based on this map, adopting the SLAM positioning result within the reliable region and reverting to traditional positioning outside the region or in abnormal situations. This method uses the traditional system as a benchmark and "trainer," realizing the progressive, reliable verification and expansion of SLAM positioning capabilities across the entire train line. It provides a compatible and smooth transition technical solution for the seamless transformation of existing lines and the deployment of virtual transponders, effectively solving the problem of integrating emerging sensing and positioning technologies with mature trackside beacon systems.

[0095] Optionally, when both valid SLAM positioning results and traditional positioning results exist for the current position of the train, a consistency check is performed on the valid SLAM positioning results and the traditional positioning results to obtain the deviation value.

[0096] For example, the vehicle controller simultaneously receives the SLAM positioning results (including position coordinates and velocity) output by the SLAM module and the traditional positioning results output by the traditional positioning unit. It calculates the position deviation value by calculating the difference between the position coordinates of the two at the same time and the velocity deviation value by calculating the velocity difference.

[0097] Specifically, the position deviation can be calculated using the difference in latitude and longitude or the difference in route mileage; the speed deviation is the absolute difference between the output speeds of the two. This verification process is performed in real time to ensure the consistency of the positioning results.

[0098] Furthermore, if the deviation value exceeds a preset threshold, a positioning anomaly is detected.

[0099] It should be noted that the preset threshold here is not the same as the positioning deviation threshold used in the training phase to determine whether the SLAM positioning accuracy is qualified. The two have different application scenarios, core functions and value criterion.

[0100] For example, the preset threshold is set according to the positioning consistency verification requirements of the train operation phase (e.g., it can be set to a range of ±1 meter to ±3 meters, and the specific value is determined based on factors such as line scheduling accuracy and train operating speed). In one alternative implementation, the onboard controller may initiate a re-establishment of the positioning.

[0101] For example, if the position deviation value or speed deviation value exceeds a preset threshold for multiple consecutive sampling periods, the vehicle controller determines that the positioning is lost, immediately terminates the current positioning method, re-executes the positioning establishment process, and prioritizes the rapid recovery of positioning through traditional ODO odometers and traditional beacon readers.

[0102] Specifically, during the re-establishment of positioning, the on-board controller stops outputting the safe position. Once the traditional positioning result is stable and meets the accuracy requirements, the safe position output is resumed, and a positioning anomaly alarm message is sent to the train driver's cab.

[0103] In another alternative implementation, the SLAM module can record abnormal data and use the abnormal data to update the SLAM trusted region information.

[0104] For example, the SLAM module automatically records the data and time of inconsistencies in positioning.

[0105] The inconsistent positioning data includes traditional positioning results, SLAM positioning results, deviation values, and corresponding train operating environment information. Train operating environment information includes the terrain features of the current area (such as inside tunnels or elevated sections), lighting conditions (such as strong light or darkness), and the distribution of obstacles (such as construction barriers or tree obstructions).

[0106] Specifically, the SLAM module analyzes abnormal data. If it finds that the positioning deviation frequently exceeds the preset threshold in a certain trusted area, it will narrow the range of the trusted area or adjust the positioning accuracy judgment standard of the area. If the abnormal data is concentrated in an untrusted area, it will include the area in the key training range and trigger the SLAM module to re-enter the detection period to train in the area.

[0107] Thus, by recording abnormal data and dynamically updating trusted areas, this invention achieves continuous optimization of SLAM positioning capabilities, thereby improving the adaptability and robustness of the positioning system.

[0108] Optionally, the aforementioned SLAM trusted region information can be fused and dynamically updated based on training data or abnormal data provided by multiple trains.

[0109] For example, the ground control center collects SLAM training data (including compensated positioning results, iterative optimization parameters, and reliable region annotations) and abnormal data from all trains on the same line, and uses a data fusion algorithm to integrate these data, remove abnormal data from individual trains, and retain valid data with common characteristics.

[0110] Specifically, based on the fused data, the line-level SLAM trusted area information is updated and transmitted to the onboard controllers and SLAM modules of each train via wireless communication, so that all trains can share the optimized trusted area information and improve the consistency and reliability of SLAM positioning across the entire line.

[0111] Thus, this invention achieves global optimization of SLAM trusted area information through a multi-train data fusion and update mechanism, accelerating the promotion and application of SLAM positioning technology across the entire railway line.

[0112] Optionally, the train positioning method in this embodiment of the invention further includes a safety verification stage, which is executed in parallel with the positioning marking and training stage and the formal operation stage to ensure positioning safety during the train's hibernation and wake-up process.

[0113] In some embodiments, when the train is in hibernation, the independent BeiDou devices at the front and rear simultaneously save the train location information, working mode information, and safety level information they have collected to the SIL4 safety device.

[0114] For example, the operating mode information includes the location operating mode and the sleep trigger mode.

[0115] In other embodiments, when the train is woken up, the first and last independent Beidou devices report the train's position information to the SIL4 safety device independently, and the dual-end SIL4 safety devices verify whether the position offset reported by the first and last devices exceeds the set requirements.

[0116] For example, the dual-end SIL4 safety devices calculate the difference between the first and last position information they receive, and determine whether the difference meets the train operation safety requirements; through the first and last position offset verification results, the cold movement detection function of the European Train Control System (ETCS) standard is directly realized.

[0117] Specifically, the cold movement detection function of the ETCS standard refers to the function of detecting whether the train has moved unexpectedly when it is not in operation. When the verification result of the head and tail position offset exceeds the cold movement detection setting requirements, it is determined that the train has moved coldly, and the on-board controller triggers the corresponding safety warning or control command (such as emergency braking or prohibition of starting).

[0118] Optionally, the aforementioned security verification phase may also include a non-dependent sleep control step.

[0119] In this embodiment of the invention, non-dependent hibernation control refers to a hibernation control method that does not rely on real-time confirmation from ground equipment and supports multiple triggering methods.

[0120] For example, the sleep triggering method includes at least one of manual power-off triggering, local press triggering, and remote command triggering. After receiving the sleep triggering signal, the device does not need to wait for the confirmation command from the ground equipment, but directly performs the sleep operation and saves the relevant working status information.

[0121] Specifically, manual power-off triggering is achieved by manually operating the train's onboard power supply switch; local press triggering is achieved by pressing the sleep control button on the train; and remote command triggering is achieved by sending a remote sleep command from the ground control center.

[0122] Thus, by integrating end-to-end redundant positioning verification, independent sleep control, and cold movement detection functions, this invention improves the safety redundancy of the train positioning system without adding extra dedicated equipment, meets the SIL4 safety integrity level requirements, and further ensures the operational safety of the train throughout its entire life cycle.

[0123] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other through the communications bus 340.

[0124] The processor 310 can call logic instructions in the memory 330 to execute a train positioning method, which includes: during train detection, constructing a positioning SLAM module from a synchronized map, training it using conventional positioning information from conventional positioning units to generate and maintain SLAM reliable area information; the conventional positioning unit includes at least an odometer and a beacon reader, and the SLAM reliable area information is used to identify areas in the train's operating line where the SLAM positioning results have reached a preset reliability; during train operation, determining whether the train's current position meets the SLAM positioning activation conditions based on the SLAM reliable area information; when the SLAM positioning activation conditions are met, using the SLAM positioning result output by the SLAM module as the train's safe position; when the SLAM positioning activation conditions are not met, using the conventional positioning result output by the conventional positioning unit as the train's safe position.

[0125] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0126] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the train positioning method provided by the above methods. The method includes: during train inspection, constructing a positioning SLAM module from a synchronous map, training it using conventional positioning information from conventional positioning units to generate and maintain SLAM reliable area information; the conventional positioning unit includes at least an odometer and a beacon reader, and the SLAM reliable area information is used to identify areas in the train running line where the SLAM positioning result has reached a preset reliability; during train operation, determining whether the current position of the train meets the SLAM positioning activation conditions based on the SLAM reliable area information; when the SLAM positioning activation conditions are met, using the SLAM positioning result output by the SLAM module as the safe position of the train; when the SLAM positioning activation conditions are not met, using the conventional positioning result output by the conventional positioning unit as the safe position of the train.

[0127] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the train positioning method provided by the above methods. This method includes: during train inspection, constructing a positioning SLAM module from a synchronized map, training it using conventional positioning information from conventional positioning units to generate and maintain SLAM reliable area information; the conventional positioning unit includes at least an odometer and a beacon reader, and the SLAM reliable area information is used to identify areas in the train's operating line where the SLAM positioning result has reached a preset reliability; during train operation, determining whether the train's current position meets the SLAM positioning activation conditions based on the SLAM reliable area information; when the SLAM positioning activation conditions are met, using the SLAM positioning result output by the SLAM module as the train's safe position; when the SLAM positioning activation conditions are not met, using the conventional positioning result output by the conventional positioning unit as the train's safe position.

[0128] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0129] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A train positioning method, characterized in that, include: During train inspection, a localization SLAM module is constructed using a synchronous map and trained using traditional localization information from traditional localization units to generate and maintain SLAM trusted area information. The traditional localization units include at least an odometer and a beacon reader. The SLAM trusted area information is used to identify areas in the train running line where the SLAM localization results have reached a preset reliability. During train operation, the current position of the train is determined based on the SLAM trusted area information to determine whether the SLAM positioning activation conditions are met. When the SLAM positioning activation conditions are met, the SLAM positioning result output by the SLAM module is used as the safe position of the train. When the SLAM positioning activation conditions are not met, the traditional positioning result output by the traditional positioning unit is taken as the safe position of the train.

2. The method according to claim 1, characterized in that, The localization SLAM module constructed from the synchronized map is trained using traditional localization information from traditional localization units to generate and maintain SLAM trusted area information, including: The SLAM module receives raw measurement information output by the conventional positioning unit, including speed and distance information output by the odometer and beacon information output by the beacon reader; The original measurement information is used to compensate the SLAM positioning data obtained based on the environmental perception sensor to obtain a first positioning result; Receive train safety location information determined by the conventional positioning unit; The first positioning result is iteratively optimized using the train safety location information to obtain a second positioning result; Based on the degree of agreement between the second positioning result and the train safety location information, a geographical area with acceptable SLAM positioning accuracy is determined to generate the SLAM reliable area information.

3. The method according to claim 1, characterized in that, During the train inspection, the SLAM module does not output positioning results for train positioning control to the on-board controller; During train operation, the SLAM module outputs the SLAM positioning result to the on-board controller based on the SLAM trusted area information and the train's current position, when the SLAM positioning activation conditions are met.

4. The method according to claim 1, characterized in that, The failure to meet the SLAM positioning activation condition includes any of the following: The SLAM module failed. The SLAM trusted region information is unavailable; The train's current location is not covered by the SLAM trusted region information.

5. The method according to claim 1, characterized in that, The method further includes: When both valid SLAM positioning results and traditional positioning results exist at the current position of the train, a consistency check is performed on the valid SLAM positioning results and the traditional positioning results to obtain the deviation value. If the deviation value exceeds a preset threshold, a positioning anomaly is handled.

6. The method according to claim 5, characterized in that, The location anomaly handling includes: Location re-establishment is initiated by the vehicle controller; and / or, The SLAM module records abnormal data and uses the abnormal data to update the SLAM trusted region information.

7. The method according to claim 6, characterized in that, The SLAM trusted region information can be fused and dynamically updated based on training data or abnormal data provided by multiple trains.

8. The method according to claim 1, characterized in that, The areas identified by the SLAM trusted area information include areas where ground beacons have been deployed and areas where ground beacons have not been deployed.

9. The method according to claim 1, characterized in that, The method further includes: Determine the intersection region of SLAM trusted region information for all individual trains; If the intersection area is at least one discrete geographical location, then when the train runs to the geographical location, it is confirmed that it meets the SLAM positioning activation condition, and the SLAM positioning result output by the SLAM module is used as the safe position of the train.

10. A train positioning system, characterized in that, The system for implementing the method as described in any one of claims 1 to 9 comprises: Traditional positioning units include at least an odometer and a beacon reader, used to provide traditional positioning information; The synchronous map building and localization SLAM module is configured as follows: During train inspection, traditional positioning information from the traditional positioning unit is used for training to generate and maintain SLAM trusted region information; the SLAM trusted region information is used to identify areas in the train running line where the SLAM positioning results have reached a preset reliability. The vehicle-mounted controller is communicatively connected to the conventional positioning unit and the SLAM module, and is configured as follows: During train operation, the current position of the train is determined based on the SLAM trusted area information to determine whether the SLAM positioning activation conditions are met. When the SLAM positioning activation conditions are met, the SLAM positioning result output by the SLAM module is used as the safe position of the train. When the SLAM positioning activation conditions are not met, the traditional positioning result output by the traditional positioning unit is taken as the safe position of the train.

Images