Track maintenance oriented distance positioning method and device

Abstract

This invention discloses a mileage positioning method and apparatus for track maintenance. The method includes: acquiring railway vehicle point cloud data of the railway line and identifying sleeper point cloud data corresponding to each sleeper; merging sleeper point cloud data belonging to the same sleeper to obtain multiple sleeper instances; sequentially numbering the sleeper instances along the mileage direction and establishing a first mapping relationship between the sleeper instance number and its mileage coordinates; acquiring feature location information of at least one key facility in the railway line, and determining the sleeper instance corresponding to the key facility as a key sleeper instance based on the mileage coordinates in the feature location information; converting the sleeper instance number into a sleeper code based on the key sleeper instance to establish a second mapping relationship between the sleeper code, the sleeper instance number, and its mileage coordinates; and determining the mileage position of the target sleeper corresponding to the track maintenance operation based on the sleeper code. This invention can improve the accuracy of mileage positioning.

Description

Technical Field

[0001] This invention relates to the field of railway infrastructure inspection technology, and in particular to a mileage positioning method and device for track maintenance. Background Technology

[0002] In the inspection and maintenance of railway infrastructure, accurate mileage positioning of track defects or faults is fundamental to ensuring work quality and efficiency. Currently, railway inspection processes have begun to utilize inspection trains equipped with advanced sensors, such as lidar, enabling high-precision self-positioning of track mileage through technologies like synchronous positioning and map building. However, in actual maintenance operations, the vast majority of maintenance vehicles and personnel still rely on traditional wheeled odometers for positioning. These devices are susceptible to factors such as wheel wear and wheel spin, meaning that the accurate defect coordinates obtained during inspection cannot be directly and reliably used in maintenance, thus hindering maintenance efficiency.

[0003] To correct for accumulated errors, existing railway mileage calibration technologies are mainly divided into two categories. One category is assisted positioning technology that relies on external infrastructure, such as deploying transponders along the track or utilizing track circuits. The other category is self-localization technology that relies on onboard sensors, mainly based on lidar or visual sensors, which achieves positioning by sensing environmental features and matching them with a pre-built high-precision map.

[0004] However, all of the aforementioned existing technologies have inherent limitations. External assisted positioning technology requires the deployment of a large number of trackside devices along the line, resulting in high construction and long-term maintenance costs, and positioning services will be interrupted once the equipment fails. Although satellite navigation technology is less expensive, it cannot be used in areas with signal obstruction such as tunnels and canyons, and its positioning accuracy is limited. While lidar-based self-positioning technology has high accuracy, it relies on expensive sensors and complex algorithms, and is usually only equipped on advanced inspection trains, making it difficult to popularize it on the numerous and varied ordinary maintenance vehicles. This results in the inspection end having high-precision positioning capabilities, while the positioning methods at the execution end are relatively backward and have large errors.

[0005] In summary, existing technologies are insufficient to fundamentally solve the problem of large mileage positioning errors in railway maintenance operations. Summary of the Invention

[0006] This invention provides a mileage positioning method for track maintenance, aimed at reducing mileage positioning errors and improving the accuracy of mileage positioning during track maintenance. The method includes: Acquire railway vehicle point cloud data of railway lines, and based on the railway vehicle point cloud data, identify the sleeper point cloud data corresponding to each sleeper; Using prior information about the sleepers, point cloud data of sleepers belonging to the same sleeper are merged to obtain multiple sleeper instances; based on the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish the first mapping relationship between the sleeper instance number and its mileage coordinates. Obtain the characteristic location information of at least one key facility in the railway line, the characteristic location information including the type of key facility, the serial number of the key facility and its mileage coordinates; based on the mileage coordinates in the characteristic location information, determine the sleeper instance corresponding to the key facility as the key sleeper instance; Based on key sleeper instances, sleeper instance numbers are converted into sleeper codes to establish a second mapping relationship between sleeper codes and sleeper instance numbers and their mileage coordinates; the sleeper code includes a key facility type identifier, a key facility serial number, and a sleeper serial number; Based on the sleeper code, the mileage location of the target sleeper corresponding to the track maintenance operation is determined.

[0007] This invention also provides a mileage positioning device for track maintenance, used to reduce mileage positioning errors and improve the accuracy of mileage positioning during track maintenance. The device includes: The sleeper point cloud data recognition module is used to acquire railway vehicle point cloud data of railway lines and, based on the railway vehicle point cloud data, identify the sleeper point cloud data corresponding to each sleeper. The sleeper instance numbering module is used to merge sleeper point cloud data belonging to the same sleeper using the prior information of the sleeper to obtain multiple sleeper instances; according to the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish the first mapping relationship between the sleeper instance number and its mileage coordinates. The critical sleeper instance determination module is used to obtain the characteristic location information of at least one critical facility in the railway line. The characteristic location information includes the type of critical facility, the serial number of the critical facility and its mileage coordinates. Based on the mileage coordinates in the characteristic location information, the sleeper instance corresponding to the critical facility is determined as the critical sleeper instance. The sleeper coding conversion module is used to convert sleeper instance numbers into sleeper codes based on key sleeper instances, so as to establish a second mapping relationship between sleeper codes and sleeper instance numbers and their mileage coordinates; the sleeper code includes key facility type identifier, key facility serial number and sleeper serial number; The target sleeper mileage positioning module is used to determine the mileage location of the target sleeper corresponding to track maintenance operations based on the sleeper code.

[0008] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described mileage positioning method for track maintenance.

[0009] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described mileage positioning method for track maintenance.

[0010] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described mileage positioning method for track maintenance.

[0011] In this embodiment of the invention, railway vehicle-mounted point cloud data of the railway line is acquired, and sleeper point cloud data corresponding to each sleeper is identified based on the railway vehicle-mounted point cloud data; using the prior information of the sleepers, sleeper point cloud data belonging to the same sleeper are merged to obtain multiple sleeper instances; according to the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish a first mapping relationship between the sleeper instance number and its mileage coordinates; characteristic location information of at least one key facility in the railway line is acquired, the characteristic location information including the key facility type, key facility serial number and its mileage coordinates; based on the mileage coordinates in the characteristic location information, the sleeper instance corresponding to the key facility is determined as the key sleeper instance; using the key sleeper instance as a reference, the sleeper instance number is converted into a sleeper code to establish a second mapping relationship between the sleeper code and the sleeper instance number and its mileage coordinates; the sleeper code includes a key facility type identifier, a key facility serial number and a sleeper serial number; based on the sleeper code, the mileage position of the target sleeper corresponding to the track maintenance operation is determined. In the above process, this embodiment of the invention achieves digital representation of sleepers by establishing a first mapping relationship between sleeper instance numbers and mileage coordinates; and by establishing a second mapping relationship between sleeper codes, sleeper instance numbers, and mileage coordinates, a dual mapping relationship is formed between railway mileage coordinates in point cloud data and sleeper instances in the physical track. Ultimately, this allows maintenance processes to directly locate the mileage position of the target sleeper based on its code, without relying on complex positioning equipment. This effectively eliminates accumulated mileage errors between different stages and significantly improves the accuracy of mileage positioning. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 This is a flowchart of a track maintenance mileage positioning method in an embodiment of the present invention; Figure 2 This is a schematic diagram of the track profile and the ZOY plane coordinate system of the point cloud in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the process of obtaining a sleeper example in an embodiment of the present invention; Figure 4 This is a flowchart of the sleeper coding process in an embodiment of the present invention; Figure 5 This is a schematic diagram of a track maintenance mileage positioning device in an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0014] Figure 1 This is a flowchart of a track maintenance mileage positioning method according to an embodiment of the present invention. The method includes: Step 101: Obtain the railway vehicle point cloud data of the railway line, and based on the railway vehicle point cloud data, identify the sleeper point cloud data corresponding to each sleeper. Step 102: Using the prior information of the sleepers, merge the point cloud data of the sleepers belonging to the same sleeper to obtain multiple sleeper instances; according to the mileage coordinates of each sleeper instance, sequentially number the sleeper instances along the mileage direction and establish the first mapping relationship between the sleeper instance number and its mileage coordinates. Step 103: Obtain the characteristic location information of at least one key facility in the railway line. The characteristic location information includes the type of key facility, the serial number of the key facility and its mileage coordinates. Based on the mileage coordinates in the characteristic location information, determine the sleeper instance corresponding to the key facility as the key sleeper instance. Step 104: Based on the key sleeper instance, convert the sleeper instance number into a sleeper code to establish a second mapping relationship between the sleeper code and the sleeper instance number and its mileage coordinates; the sleeper code includes the key facility type identifier, the key facility serial number and the sleeper serial number. Step 105: Based on the sleeper code, determine the mileage location of the target sleeper corresponding to the track maintenance operation.

[0015] Each step is explained in detail below.

[0016] In step 101, the railway vehicle point cloud data of the railway line is acquired, and the sleeper point cloud data corresponding to each sleeper is identified based on the railway vehicle point cloud data.

[0017] In one embodiment, based on railway vehicle-mounted point cloud data, sleeper point cloud data corresponding to each sleeper is identified, including: The railway vehicle-mounted point cloud data is decoupled into multiple track profiles along the mileage direction; For each track profile, the track profile curve profile of the preset track region is extracted based on the prior structural information of the track. Based on the geometric features of the track profile curve, curve segments that conform to the preset geometric features of the sleepers are identified as candidate sleeper profiles. The proportion of the total width of the candidate sleeper profiles in each profile to the preset track area is calculated as the confidence level of the sleeper. The confidence level of the sleeper is compared with a preset confidence threshold, and candidate sleeper profiles with a confidence level greater than the preset confidence threshold are selected as the final sleeper profiles. In railway vehicle-mounted point cloud data, the point cloud data corresponding to the final sleeper profile is labeled as sleeper point cloud data.

[0018] In a specific embodiment, the complexity of the sleeper segmentation algorithm is significantly reduced by decoupling and dimensionality-reducing the sleeper semantic segmentation task in the vehicle point cloud into several independent sleeper profile segmentation tasks.

[0019] In a specific embodiment, the present invention adopts the railway mileage sleeper coding method (SEL). Based on the characteristics that the sleeper profile is flat and smooth while the ballast profile is rugged, the proportion of the flat profile in the middle of the track bed is calculated as the sleeper existence confidence, and then the sleeper point cloud is identified and labeled.

[0020] The input data for the SEL is railway vehicle-mounted point cloud data acquired by a lidar system. This point cloud data undergoes point cloud denoising and coordinate transformation processing by the acquisition equipment, and its coordinate system is transformed into the rail surface reference coordinate system. The rail surface reference coordinate system is derived from the top surface of the rail and the center of the track gauge. Its X-axis is parallel to the direction of travel of the vehicle carrying the lidar, its Y-axis is perpendicular to the top surface of the rail, and its Z-axis is parallel to the lidar scanning surface. Furthermore, a key characteristic of railway vehicle-mounted point clouds is that the point cloud data is composed of several profile data obtained from lidar scanning. This allows the vehicle-mounted point cloud to be naturally decoupled into several independent track profile point cloud data. Figure 2 This is a schematic diagram of the track profile and point cloud ZOY plane coordinate system in an embodiment of the present invention. Figure 2The diagram shows a track profile of the onboard point cloud, parallel to the ZOY plane of the rail reference coordinate system, with the origin of the ZOY plane located at the center of the gauge along the horizontal line of the top surfaces of the two rails. Since both the rails and sleepers are located in the central region of the track, the point cloud of the central ballast bed can be extracted from the overall onboard point cloud based on prior information about its dimensions, thus reducing the data complexity of subsequent sleeper segmentation algorithms. Furthermore, because the positions of the left and right rails in the rail reference coordinate system are fixed, the point cloud of the rail region can be directly categorized based on prior information such as the gauge, preventing interference with the sleeper segmentation algorithm.

[0021] Since railway vehicle-mounted point clouds can be decoupled into several track profiles, each track profile can be regarded as two-dimensional discrete profile data. Therefore, if sleeper profile segmentation is performed directly on these independent profile data, the upper limit of the algorithm's time complexity will be lower than that of segmentation algorithms based on two-dimensional images or three-dimensional point clouds. Although this will lose the spatial relative relationship between profiles in the X-axis direction, for sleeper segmentation tasks in railway scenarios, since the laying postures of different sleepers are basically parallel and consistent, the profile profiles of sleepers can be regarded as following independent and identically distributed characteristics along the mileage direction. The loss of information in the X-axis direction will not have a significant impact on the sleeper segmentation algorithm. Through this assumption of independent and identical distribution, the sleeper segmentation problem in the vehicle-mounted point cloud can be decomposed into several less complex sleeper profile segmentation subproblems, reducing the complexity of the sleeper segmentation algorithm. In addition, sleeper profile segmentation does not require point cloud regularization processing such as voxelization or octree generation, which can also reduce the complexity of the overall semantic segmentation algorithm from the preprocessing stage.

[0022] The SEL algorithm, based on the characteristic that the sleeper profile is flat and smooth while the ballast profile is rugged, calculates the proportion of flat profiles in the middle of the track bed as the sleeper presence confidence, and then identifies and labels the sleeper point cloud. The algorithm flow is as follows: \begin{enumerate} \item \textbf{section decoupling}: Decouples the vehicle-mounted point cloud into independent section processing.

[0023] \item \textbf{Curve Decomposition}: Decomposes the profile curve into monotonic sub-curve segments based on inflection points.

[0024] \item \textbf{Smooth / Steep Determination}: Uses height and slope thresholds to determine each sub-curve segment. Those meeting the thresholds (flat and low) are considered smooth curves; otherwise, they are considered steep curves (potentially ballast) and are discarded.

[0025] \item \textbf{Merge and Pseudo-Envelope Removal}: Merges adjacent smooth curves. A sliding window of the merged curves, sized to correspond to a height threshold, is used to scan and detect connections between local extrema. If a connection does not meet a slope threshold, it is identified as a pseudo-envelope (e.g., gravel accumulation), and that smooth curve segment is removed. The remaining curves are true flat curves.

[0026] \item \textbf{Sleeper Confidence Calculation}: \begin{itemize} After removing the fine, flat curves, the remaining curves are suspected to be sleeper profiles (segmented curves).

[0027] The sleeper confidence score is obtained by calculating the proportion of the total width W_S of the segmented curve to the width W_P of the middle section of the track bed. \begin{equation} Score = \frac{W_S}{W_P} \end{equation} The `item` array stores the profile confidence scores, which will then be used as input for the subsequent sleeper instance numbering algorithm.

[0028] \end{itemize}

[0029] \item\textbf{Sleeper Segmentation and Labeling}: Uses a score threshold to filter confidence scores, identifies sleeper profiles, and labels their data points. Completing all profile processing completes the point cloud sleeper segmentation.

[0030] \end{enumerate}

[0031] Using railway sleepers as a natural spatial scale, a discrete positioning framework is constructed, transforming abstract mileage coordinates into intuitive sleeper codes (such as "D0040000"), establishing a two-way mapping relationship between railway mileage coordinates and the physical track. Through feature recognition of key facilities (such as turnout switch rails, bridges, and tunnels), the sleeper codes are anchored to the actual track. SEL can unify the mileage positioning reference system for the inspection and maintenance stages in the railway maintenance chain, thereby overcoming the problem of accumulated positioning errors across multiple stages.

[0032] Experimental results demonstrate that the dimensionality reduction and divide-and-conquer strategy based on vehicle-mounted point cloud profile decoupling significantly improves sleeper segmentation efficiency: in 225 meters of measured data collected at a vehicle speed of 160 km / h, the SEL algorithm achieves a processing speed more than 10 times faster than advanced deep learning semantic segmentation models while maintaining considerable segmentation accuracy. Furthermore, a numbering algorithm designed by integrating prior knowledge of sleeper size and spacing achieves high accuracy in sleeper instance numbering on a 5-kilometer multi-scenario line. Even under strong amplitude random noise interference, the algorithm demonstrates high accuracy in identifying skipped numbers for missing sleepers, showcasing its excellent robustness.

[0033] The core value of the SEL method lies in bridging the gap between inspection data and maintenance execution. By directly associating the location of the target defect with a specific sleeper code, it eliminates the need for secondary on-site verification, significantly shortening the inspection and maintenance response chain.

[0034] In step 102, using the prior information of the sleepers, the point cloud data of sleepers belonging to the same sleeper are merged to obtain multiple sleeper instances; according to the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish the first mapping relationship between the sleeper instance number and its mileage coordinates.

[0035] Figure 3 This is a flowchart illustrating the process of obtaining sleeper instances in an embodiment of the present invention. In one embodiment, prior information about the sleepers is used to merge point cloud data of sleepers belonging to the same sleeper to obtain multiple sleeper instances, including: Step 301: Sort the sleeper point cloud data according to the mileage coordinates of each sleeper point cloud data in the railway vehicle point cloud data; Step 302: Calculate the mileage distance between two adjacent sleeper point cloud data in order of sorting; Step 303: If the mileage distance is less than or equal to the preset maximum width threshold of the sleeper in the prior information, it is determined that the two sleeper point cloud data belong to the same sleeper, and they are merged into one sleeper instance.

[0036] In a specific embodiment, the sleeper is a key component of the track bed, undertaking important functions such as fixing the rails and distributing loads. To ensure its effectiveness, the laying spacing, model and size of the sleepers, and the number per kilometer must strictly follow the specifications and be recorded and archived by the railway bureau in the sleeper ledger. Specifically, the width of the sleeper along the train's direction of travel (including the width of the upper and lower bottom surfaces) is between 150-280mm, while the spacing between adjacent sleepers is between 440-720mm. Using this prior information, sleeper profiles belonging to the same sleeper can be merged into a single sleeper instance, distinguishing different sleeper profiles.

[0037] In one embodiment, the sleeper instances are sequentially numbered along the mileage direction based on their mileage coordinates, and the method further includes: During the numbering process, based on the mileage coordinates of each sleeper instance, it is determined whether the mileage distance between adjacent sleeper instances exceeds the preset mileage range threshold in the prior information; the preset mileage range threshold includes: the minimum allowable distance between adjacent sleepers and the maximum allowable distance between adjacent sleepers; If the preset mileage range threshold is exceeded, and it is determined that there are missed sleepers, the following skip number operation will be performed: Based on the mileage distance between adjacent sleeper instances, calculate the number of skipped sleeper instances, and determine the number corresponding to the next sleeper instance to be numbered based on the number of skipped sleeper instances.

[0038] In one embodiment, the number of skipped sleeper instances is calculated based on the mileage distance between adjacent sleeper instances, and the number corresponding to the next sleeper instance to be numbered is determined based on the number of skipped sleeper instances, including: Based on the round function, the number of skipped sleeper instances is calculated using the following formula: ; in, skip This represents the number of skipped sleeper instances. p a and p c These represent adjacent sleeper examples. d min and d max These represent the minimum allowable spacing between adjacent sleepers and the maximum allowable spacing between adjacent sleepers, respectively.

[0039] In a specific embodiment, during the sleeper instance numbering process, the prior knowledge of sleeper dimensions and laying spacing is concretized into four key parameter thresholds: minimum sleeper width, maximum sleeper width, minimum sleeper spacing, and maximum sleeper spacing. The minimum sleeper width is used to filter out invalid sleeper instances that are too narrow due to erroneous merging. The maximum sleeper width serves as a criterion; if the mileage distance between two identified sleeper profiles is not greater than this threshold, they are determined to belong to the same physical sleeper and should be merged into one sleeper instance. The minimum sleeper spacing is used to eliminate false identifications; if the mileage distance between the end of the current sleeper instance and the beginning of the next sleeper profile to be judged is less than this threshold, the profile will be considered a false identification and discarded. The maximum sleeper spacing is used to detect missed detections; if the spacing between two consecutive sleeper instances exceeds this threshold, it is determined that there may be unidentified sleepers within that interval. When a missed detection is detected, the system will calculate and estimate the number of missed sleepers based on the relationship between the mileage difference between the previous and subsequent instances and the average sleeper spacing. Then, it will perform a skip numbering operation in the subsequent numbering sequence to ensure the continuity of the numbering and the consistency with the actual physical order.

[0040] In step 103, characteristic location information of at least one key facility in the railway line is obtained. The characteristic location information includes the type of key facility, the serial number of the key facility and its mileage coordinates. Based on the mileage coordinates in the characteristic location information, a sleeper instance corresponding to the key facility is determined as a key sleeper instance.

[0041] In one embodiment, determining a sleeper instance corresponding to a critical facility as a critical sleeper instance based on the mileage coordinates in the feature location information includes: Compare the mileage coordinates of key facilities with the first mapping relationship; From all sleeper instances in the first mapping relationship, find the sleeper instance whose mileage coordinates are closest to the mileage coordinates of the critical facility; The closest sleeper instance is identified as the critical sleeper instance corresponding to the critical facility.

[0042] Find the sleeper instance whose mileage coordinates are closest to those of the critical facility. This includes: calculating the absolute value of the difference between the mileage coordinates of the critical facility and the mileage coordinates of each sleeper instance; and identifying the sleeper instance with the smallest absolute value as the closest sleeper instance.

[0043] In step 104, based on the key sleeper instance, the sleeper instance number is converted into a sleeper code to establish a second mapping relationship between the sleeper code and the sleeper instance number and its mileage coordinates; the sleeper code includes the key facility type identifier, the key facility serial number and the sleeper serial number.

[0044] In one embodiment, based on a key sleeper instance, the sleeper instance number is converted into a sleeper code, including: Using key sleeper instances as baseline sleepers, the key facility type and key facility serial number in the feature location information of key facilities are combined into a coding prefix; Starting from the mileage coordinates of the key sleeper instance, the sleeper number of the key sleeper instance is set as the starting sleeper number. Based on the order of the sleeper instance numbers, each sleeper instance is assigned a continuously increasing sleeper number. The coding prefix is ​​combined with the sleeper serial number assigned to each sleeper instance to generate the sleeper code corresponding to each sleeper instance.

[0045] In a specific embodiment, through the aforementioned sleeper segmentation and sleeper instance numbering process, a second mapping relationship has been established between sleeper codes, sleeper instance numbers, and their mileage coordinates. However, at this point, the sleeper instance number is only an internal number of the point cloud data, and it still needs to be anchored to the key sleepers of the actual physical track in order to be sequentially converted into sleeper codes that can ultimately be used for precise construction positioning.

[0046] Because railways contain numerous unique and critical features, the anchoring of critical sleepers can be achieved through the identification of these features. Taking turnouts as an example, the railway integrated inspection vehicle, the onboard point cloud data acquisition vehicle of this invention, is also equipped with a precision track detection system. This system can detect the mileage coordinates, type, and sequence number of the turnout switch tip. Through the scheduling of the inspection vehicle's spatiotemporal synchronization system, the algorithm of this invention can obtain the identification information of critical features and, based on this, identify the critical sleeper corresponding to that feature, converting the critical sleeper number into a regularized sleeper code. Subsequently, through reasoning from several critical sleepers and their sequential order, the conversion from sleeper instance number to sleeper code can be achieved.

[0047] The railway sleeper coding and naming rule designed in this invention is: "Key Facility Type + Key Facility Number + Sleeper Number". Wherein: \begin{itemize} \item \textbf{Key Facility Type}: Represented by a single letter, such as turnout (D), bridge (Q), tunnel (S), platform (Z), etc.; \item \textbf{Key Facility Number}: A three-digit number representing the sequential number of each key facility from the starting point to the end point of the railway line. The relevant information is recorded in the line ledger. \item \textbf{sleeper number}: A four-digit number, starting from 0000 for the critical sleeper and increasing with mileage.

[0048] \end{itemize}

[0049] Figure 4 This is a flowchart of the sleeper coding process in an embodiment of the present invention. Figure 4 Taking the sleeper code corresponding to turnout No. 004 on a certain line as an example, based on the critical facility mileage at the tip of the turnout's switch rail, the sleeper closest to that mileage coordinate is found. This sleeper is the critical sleeper instance, and its corresponding code is named "D0040000". Then, starting from the critical sleeper instance, all subsequent sleeper codes begin with "D004", with the sleeper number gradually increasing until the next critical facility is encountered.

[0050] In step 105, the mileage location of the target sleeper corresponding to the track maintenance operation is determined based on the sleeper code.

[0051] In one embodiment, determining the mileage location of the target sleeper corresponding to track maintenance work based on sleeper coding includes: The sleeper code of the target sleeper is parsed to obtain the parsed sleeper code; Based on the parsed sleeper code, a query is performed in the second mapping relationship to find the mileage coordinates that match the parsed sleeper code, which are then used as the mileage position of the target sleeper.

[0052] By directly querying the mapping relationship between sleeper codes and mileage coordinates, the precise location of the target sleeper can be quickly determined, thereby achieving precise positioning at the maintenance site without complex calculations or equipment dependence. This eliminates errors in the transmission and use of mileage information and realizes accurate mileage positioning.

[0053] This invention also provides a track maintenance mileage positioning device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to the track maintenance mileage positioning method, the implementation of this device can refer to the implementation of the track maintenance mileage positioning method; repeated details will not be elaborated further.

[0054] Figure 5 This is a schematic diagram of a track maintenance mileage positioning device according to an embodiment of the present invention. The device includes: The sleeper point cloud data recognition module 501 is used to acquire railway vehicle point cloud data of railway lines and, based on the railway vehicle point cloud data, identify the sleeper point cloud data corresponding to each sleeper. The sleeper instance numbering module 502 is used to merge sleeper point cloud data belonging to the same sleeper using the prior information of the sleeper to obtain multiple sleeper instances; according to the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish the first mapping relationship between the sleeper instance number and its mileage coordinates. The critical sleeper instance determination module 503 is used to obtain the characteristic location information of at least one critical facility in the railway line. The characteristic location information includes the type of critical facility, the serial number of the critical facility and its mileage coordinates. Based on the mileage coordinates in the characteristic location information, the sleeper instance corresponding to the critical facility is determined as the critical sleeper instance. The sleeper coding conversion module 504 is used to convert the sleeper instance number into a sleeper code based on a key sleeper instance, so as to establish a second mapping relationship between the sleeper code and the sleeper instance number and its mileage coordinates; the sleeper code includes a key facility type identifier, a key facility serial number and a sleeper serial number; The target sleeper mileage positioning module 505 is used to determine the mileage position of the target sleeper corresponding to the track maintenance operation based on the sleeper code.

[0055] In one embodiment, the sleeper point cloud data recognition module 501 is specifically used for: The railway vehicle-mounted point cloud data is decoupled into multiple track profiles along the mileage direction; For each track profile, the track profile curve profile of the preset track region is extracted based on the prior structural information of the track. Based on the geometric features of the track profile curve, curve segments that conform to the preset geometric features of the sleepers are identified as candidate sleeper profiles. The proportion of the total width of the candidate sleeper profiles in each profile to the preset track area is calculated as the confidence level of the sleeper. The confidence level of the sleeper is compared with a preset confidence threshold, and candidate sleeper profiles with a confidence level greater than the preset confidence threshold are selected as the final sleeper profiles. In railway vehicle-mounted point cloud data, the point cloud data corresponding to the final sleeper profile is labeled as sleeper point cloud data.

[0056] In one embodiment, the sleeper instance numbering module 502 is specifically used for: The sleeper point cloud data are sorted sequentially according to their mileage coordinates in the railway vehicle point cloud data. Calculate the mileage distance between two adjacent sleeper point cloud data in the sorted order; If the distance between the two sleepers is less than or equal to the preset maximum width threshold of the sleeper in the prior information, it is determined that the two sleeper point cloud data belong to the same sleeper and they are merged into one sleeper instance.

[0057] In one embodiment, the sleeper instance numbering module 502 is specifically used for: During the numbering process, based on the mileage coordinates of each sleeper instance, it is determined whether the mileage distance between adjacent sleeper instances exceeds the preset mileage range threshold in the prior information; the preset mileage range threshold includes: the minimum allowable distance between adjacent sleepers and the maximum allowable distance between adjacent sleepers; If the preset mileage range threshold is exceeded, and it is determined that there are missed sleepers, the following skip number operation will be performed: Based on the mileage distance between adjacent sleeper instances, calculate the number of skipped sleeper instances, and determine the number corresponding to the next sleeper instance to be numbered based on the number of skipped sleeper instances.

[0058] In one embodiment, the sleeper instance numbering module 502 is specifically used for: Based on the round function, the number of skipped sleeper instances is calculated using the following formula: ; in, skip This represents the number of skipped sleeper instances. p a and p c These represent adjacent sleeper examples. d min and dmax These represent the minimum allowable spacing between adjacent sleepers and the maximum allowable spacing between adjacent sleepers, respectively.

[0059] In one embodiment, the critical sleeper instance determination module 503 is specifically used for: Compare the mileage coordinates of key facilities with the first mapping relationship; From all sleeper instances in the first mapping relationship, find the sleeper instance whose mileage coordinates are closest to the mileage coordinates of the critical facility; The closest sleeper instance is identified as the critical sleeper instance corresponding to the critical facility.

[0060] In one embodiment, the sleeper encoding conversion module 504 is specifically used for: Using key sleeper instances as baseline sleepers, the key facility type and key facility serial number in the feature location information of key facilities are combined into a coding prefix; Starting from the mileage coordinates of the key sleeper instance, the sleeper number of the key sleeper instance is set as the starting sleeper number. Based on the order of the sleeper instance numbers, each sleeper instance is assigned a continuously increasing sleeper number. The coding prefix is ​​combined with the sleeper serial number assigned to each sleeper instance to generate the sleeper code corresponding to each sleeper instance.

[0061] In one embodiment, the target sleeper mileage positioning module 505 is specifically used for: The sleeper code of the target sleeper is parsed to obtain the parsed sleeper code; Based on the parsed sleeper code, a query is performed in the second mapping relationship to find the mileage coordinates that match the parsed sleeper code, which are then used as the mileage position of the target sleeper.

[0062] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described mileage positioning method for track maintenance.

[0063] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described mileage positioning method for track maintenance.

[0064] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described mileage positioning method for track maintenance.

[0065] In this embodiment of the invention, railway vehicle-mounted point cloud data of the railway line is acquired, and sleeper point cloud data corresponding to each sleeper is identified based on the railway vehicle-mounted point cloud data; using the prior information of the sleepers, sleeper point cloud data belonging to the same sleeper are merged to obtain multiple sleeper instances; according to the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish a first mapping relationship between the sleeper instance number and its mileage coordinates; characteristic location information of at least one key facility in the railway line is acquired, the characteristic location information including the key facility type, key facility serial number and its mileage coordinates; based on the mileage coordinates in the characteristic location information, the sleeper instance corresponding to the key facility is determined as the key sleeper instance; using the key sleeper instance as a reference, the sleeper instance number is converted into a sleeper code to establish a second mapping relationship between the sleeper code and the sleeper instance number and its mileage coordinates; the sleeper code includes a key facility type identifier, a key facility serial number and a sleeper serial number; based on the sleeper code, the mileage position of the target sleeper corresponding to the track maintenance operation is determined. In the above process, this embodiment of the invention achieves digital representation of sleepers by establishing a first mapping relationship between sleeper instance numbers and mileage coordinates; and by establishing a second mapping relationship between sleeper codes, sleeper instance numbers, and mileage coordinates, a dual mapping relationship is formed between railway mileage coordinates in point cloud data and sleeper instances in the physical track. Ultimately, this allows maintenance processes to directly locate the mileage position of the target sleeper based on its code, without relying on complex positioning equipment. This effectively eliminates accumulated mileage errors between different stages and significantly improves the accuracy of mileage positioning.

[0066] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0067] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0068] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0069] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0070] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for mileage positioning in track maintenance, characterized in that, include: Acquire railway vehicle point cloud data of railway lines, and based on the railway vehicle point cloud data, identify the sleeper point cloud data corresponding to each sleeper; Using prior information about the sleepers, point cloud data of sleepers belonging to the same sleeper are merged to obtain multiple sleeper instances; based on the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish the first mapping relationship between the sleeper instance number and its mileage coordinates. Obtain the characteristic location information of at least one key facility in a railway line, the characteristic location information including the type of key facility, the serial number of the key facility and its mileage coordinates; Based on the mileage coordinates in the feature location information, the sleeper instances corresponding to the key facilities are identified as key sleeper instances. Based on key sleeper instances, sleeper instance numbers are converted into sleeper codes to establish a second mapping relationship between sleeper codes and sleeper instance numbers and their mileage coordinates; the sleeper code includes a key facility type identifier, a key facility serial number, and a sleeper serial number; Based on the sleeper code, the mileage location of the target sleeper corresponding to the track maintenance operation is determined.

2. The method as described in claim 1, characterized in that, Based on railway vehicle-mounted point cloud data, the point cloud data corresponding to each sleeper is identified, including: The railway vehicle-mounted point cloud data is decoupled into multiple track profiles along the mileage direction; For each track profile, the track profile curve profile of the preset track region is extracted based on the prior structural information of the track. Based on the geometric features of the track profile curve, curve segments that conform to the preset geometric features of the sleepers are identified as candidate sleeper profiles. The proportion of the total width of the candidate sleeper profiles in each profile to the preset track area is calculated as the confidence level of the sleeper. The confidence level of the sleeper is compared with a preset confidence threshold, and candidate sleeper profiles with a confidence level greater than the preset confidence threshold are selected as the final sleeper profiles. In railway vehicle-mounted point cloud data, the point cloud data corresponding to the final sleeper profile is labeled as sleeper point cloud data.

3. The method as described in claim 1, characterized in that, Using prior information about the sleepers, point cloud data of sleepers belonging to the same sleeper are merged to obtain multiple sleeper instances, including: The sleeper point cloud data are sorted sequentially according to their mileage coordinates in the railway vehicle point cloud data. Calculate the mileage distance between two adjacent sleeper point cloud data in the sorted order; If the distance between the two sleepers is less than or equal to the preset maximum width threshold of the sleeper in the prior information, it is determined that the two sleeper point cloud data belong to the same sleeper and they are merged into one sleeper instance.

4. The method as described in claim 1, characterized in that, Based on the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction, which also includes: During the numbering process, based on the mileage coordinates of each sleeper instance, it is determined whether the mileage distance between adjacent sleeper instances exceeds the preset mileage range threshold in the prior information; the preset mileage range threshold includes: the minimum allowable distance between adjacent sleepers and the maximum allowable distance between adjacent sleepers; If the preset mileage range threshold is exceeded, and it is determined that there are missed sleepers, the following skip number operation will be performed: Based on the mileage distance between adjacent sleeper instances, calculate the number of skipped sleeper instances, and determine the number corresponding to the next sleeper instance to be numbered based on the number of skipped sleeper instances.

5. The method as described in claim 4, characterized in that, Based on the mileage distance between adjacent sleeper instances, calculate the number of skipped sleeper instances. Then, based on the number of skipped sleeper instances, determine the number corresponding to the next sleeper instance to be numbered, including: Based on the round function, the number of skipped sleeper instances is calculated using the following formula: ； in, skip This represents the number of skipped sleeper instances. p a and p c These represent adjacent sleeper examples. d min and d max These represent the minimum allowable spacing between adjacent sleepers and the maximum allowable spacing between adjacent sleepers, respectively.

6. The method as described in claim 1, characterized in that, Based on the mileage coordinates in the feature location information, sleeper instances corresponding to critical facilities are identified as critical sleeper instances, including: Compare the mileage coordinates of key facilities with the first mapping relationship; From all sleeper instances in the first mapping relationship, find the sleeper instance whose mileage coordinates are closest to the mileage coordinates of the critical facility; The closest sleeper instance is identified as the critical sleeper instance corresponding to the critical facility.

7. The method as described in claim 1, characterized in that, Based on key sleeper instances, the sleeper instance numbers are converted into sleeper codes, including: Using key sleeper instances as baseline sleepers, the key facility type and key facility serial number in the feature location information of key facilities are combined into a coding prefix; Starting from the mileage coordinates of the key sleeper instance, the sleeper number of the key sleeper instance is set as the starting sleeper number. Based on the order of the sleeper instance numbers, each sleeper instance is assigned a continuously increasing sleeper number. The coding prefix is ​​combined with the sleeper serial number assigned to each sleeper instance to generate the sleeper code corresponding to each sleeper instance.

8. The method as described in claim 1, characterized in that, Based on sleeper codes, the mileage location of the target sleeper corresponding to track maintenance work is determined, including: The sleeper code of the target sleeper is parsed to obtain the parsed sleeper code; Based on the parsed sleeper code, a query is performed in the second mapping relationship to find the mileage coordinates that match the parsed sleeper code, which are then used as the mileage position of the target sleeper.

9. A mileage positioning device for track maintenance, characterized in that, include: The sleeper point cloud data recognition module is used to acquire railway vehicle point cloud data of railway lines and, based on the railway vehicle point cloud data, identify the sleeper point cloud data corresponding to each sleeper. The sleeper instance numbering module is used to merge sleeper point cloud data belonging to the same sleeper using the prior information of the sleeper to obtain multiple sleeper instances; according to the mileage coordinates of each sleeper instance, the sleeper instances are sequentially numbered along the mileage direction to establish the first mapping relationship between the sleeper instance number and its mileage coordinates. The critical sleeper instance determination module is used to obtain the characteristic location information of at least one critical facility in a railway line. The characteristic location information includes the type of critical facility, the serial number of the critical facility and its mileage coordinates. Based on the mileage coordinates in the feature location information, the sleeper instances corresponding to the key facilities are identified as key sleeper instances. The sleeper coding conversion module is used to convert sleeper instance numbers into sleeper codes based on key sleeper instances, so as to establish a second mapping relationship between sleeper codes and sleeper instance numbers and their mileage coordinates; the sleeper code includes key facility type identifier, key facility serial number and sleeper serial number; The target sleeper mileage positioning module is used to determine the mileage location of the target sleeper corresponding to track maintenance operations based on the sleeper code.

10. The apparatus as claimed in claim 9, characterized in that, The sleeper point cloud data recognition module is specifically used for: The railway vehicle-mounted point cloud data is decoupled into multiple track profiles along the mileage direction; For each track profile, the track profile curve profile of the preset track region is extracted based on the prior structural information of the track. Based on the geometric features of the track profile curve, curve segments that conform to the preset geometric features of the sleepers are identified as candidate sleeper profiles. The proportion of the total width of the candidate sleeper profiles in each profile to the preset track area is calculated as the confidence level of the sleeper. The confidence level of the sleeper is compared with a preset confidence threshold, and candidate sleeper profiles with a confidence level greater than the preset confidence threshold are selected as the final sleeper profiles. In railway vehicle-mounted point cloud data, the point cloud data corresponding to the final sleeper profile is labeled as sleeper point cloud data.

11. The apparatus as claimed in claim 9, characterized in that, The sleeper instance numbering module is specifically used for: The sleeper point cloud data are sorted sequentially according to their mileage coordinates in the railway vehicle point cloud data. Calculate the mileage distance between two adjacent sleeper point cloud data in the sorted order; If the distance between the two sleepers is less than or equal to the preset maximum width threshold of the sleeper in the prior information, it is determined that the two sleeper point cloud data belong to the same sleeper and they are merged into one sleeper instance.

12. The apparatus as claimed in claim 9, characterized in that, The sleeper instance numbering module is specifically used for: During the numbering process, based on the mileage coordinates of each sleeper instance, it is determined whether the mileage distance between adjacent sleeper instances exceeds the preset mileage range threshold in the prior information; The preset mileage range thresholds include: the minimum allowable spacing between adjacent sleepers and the maximum allowable spacing between adjacent sleepers; If the preset mileage range threshold is exceeded, and it is determined that there are missed sleepers, the following skip number operation will be performed: Based on the mileage distance between adjacent sleeper instances, calculate the number of skipped sleeper instances, and determine the number corresponding to the next sleeper instance to be numbered based on the number of skipped sleeper instances.

13. The apparatus as claimed in claim 12, characterized in that, The sleeper instance numbering module is specifically used for: Based on the round function, the number of skipped sleeper instances is calculated using the following formula: ； in, skip This represents the number of skipped sleeper instances. p a and p c These represent adjacent sleeper examples. d min and d max These represent the minimum allowable spacing between adjacent sleepers and the maximum allowable spacing between adjacent sleepers, respectively.

14. The apparatus as claimed in claim 9, characterized in that, The critical sleeper instance determination module is specifically used for: Compare the mileage coordinates of key facilities with the first mapping relationship; From all sleeper instances in the first mapping relationship, find the sleeper instance whose mileage coordinates are closest to the mileage coordinates of the critical facility; The closest sleeper instance is identified as the critical sleeper instance corresponding to the critical facility.

15. The apparatus as claimed in claim 9, characterized in that, The sleeper encoding conversion module is specifically used for: Using key sleeper instances as baseline sleepers, the key facility type and key facility serial number in the feature location information of key facilities are combined into a coding prefix; Starting from the mileage coordinates of the key sleeper instance, the sleeper number of the key sleeper instance is set as the starting sleeper number. Based on the order of the sleeper instance numbers, each sleeper instance is assigned a continuously increasing sleeper number. The coding prefix is ​​combined with the sleeper serial number assigned to each sleeper instance to generate the sleeper code corresponding to each sleeper instance.

16. The apparatus as claimed in claim 9, characterized in that, The target sleeper mileage positioning module is specifically used for: The sleeper code of the target sleeper is parsed to obtain the parsed sleeper code; Based on the parsed sleeper code, a query is performed in the second mapping relationship to find the mileage coordinates that match the parsed sleeper code, which are then used as the mileage position of the target sleeper.

17. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 8.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 8.

19. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 8.

Images