Map data fusion method for intelligent lookout and collision avoidance of railway train and related device

Abstract

The invention discloses a map data fusion method for intelligent lookout and collision avoidance of a railway train and a related device, and relates to the technical field of map data fusion, the method comprises the following steps: acquiring railway special high-precision map data, train control equipment ground basic data and station yard five-color map grid base map data, and respectively processing the acquired data; the method comprises the steps of obtaining corresponding structured vector data, executing a first operation on the basis of all the structured vector data for a first key node in the high-precision map structured vector data special for the railway to obtain fused structured vector data, and superposing all the fused structured vector data into the high-precision map data special for the railway to obtain the high-precision map data special for the railway. And obtaining the fused high-precision map special for the railway. According to the method, the high-precision map for intelligent lookout collision avoidance of the railway train is generated by fusing the multi-source data information, the false alarm rate of collision avoidance early warning can be reduced by means of the map, and the credibility of intelligent lookout collision avoidance early warning of the railway train is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of map data fusion, in particular to a map data fusion method for intelligent lookout anti-collision of railway trains and related devices. BACKGROUND

[0002] Railway transportation is the artery of the national economy, and its operation safety is crucial. During the operation of railway trains, foreign intrusions into the line, track obstacles, personnel intrusion or abnormal environment in the front line are the main risk sources leading to collisions, derailments and other major accidents.

[0003] At present, the lookout anti-collision warning method used in the operation of railway trains is to analyze and process the two-dimensional images obtained during the operation. This method identifies obstacles on the railway track from the perspective of image recognition. However, when the railway tracks (especially the two rails at the bend) are projected as a straight line in the image, the fixed equipment, buildings and other targets located beside the tracks will "visually coincide" with the tracks on the image due to the shooting angle and the geometric relationship of imaging. From the perspective of image recognition, such legal targets are easily identified as foreign intrusions, which may cause false alarms. SUMMARY

[0004] The purpose of the present application is to provide a map data fusion method for intelligent lookout anti-collision of railway trains and related devices, to generate a high-precision map that fuses multi-source data information, and to rely on this map for intelligent lookout anti-collision warning of railway trains. This can achieve accurate differentiation between legal fixed targets beside the tracks and real intrusion foreign objects, and accurate and reliable anti-collision warning, which can fundamentally reduce the false alarm rate and improve the credibility of intelligent lookout anti-collision warning of railway trains.

[0005] To achieve the above purpose, the present application provides the following solutions: In a first aspect, the present application provides a map data fusion method for intelligent lookout anti-collision of railway trains, comprising: A map data fusion method for intelligent lookout anti-collision of railway trains, characterized in that the map data fusion method for intelligent lookout anti-collision of railway trains comprises: Obtaining railway special high-precision map data, train control equipment ground basic data and station five-color raster base map data.

[0006] The railway-specific high-precision map data, the train control equipment ground base data, and the station yard five-color raster base map data are processed respectively to obtain railway-specific high-precision map structured vector data, train control equipment ground base structured vector data, and station yard five-color map preliminary structured vector data. The railway-specific high-precision map structured vector data includes multiple first key nodes and the latitude and longitude information corresponding to each first key node. The train control equipment ground base structured vector data includes multiple second key nodes and the latitude and longitude information and kilometer marker information corresponding to each second key node. The station yard five-color map preliminary structured vector data includes multiple third key nodes and the latitude and longitude information and track position relationships corresponding to each third key node. The track position relationships are the positional relationships between key nodes and tracks.

[0007] Adjacency relationship encoding is performed on the track position relationships corresponding to all third key nodes in the preliminary structured vector data of the station five-color map to obtain the station five-color map structured vector data; the station five-color map structured vector data includes multiple third key nodes and the structured encoded data of latitude and longitude information and track position relationships corresponding to each third key node.

[0008] For each first key node in the railway-specific high-precision map structured vector data, a first operation is performed based on the railway-specific high-precision map structured vector data, the train control equipment ground foundation structured vector data, and the station five-color map structured vector data to obtain fused structured vector data. The fused structured vector data includes multiple first key nodes and structured encoded data of latitude and longitude information, kilometer marker information, and track position relationships corresponding to the first key nodes.

[0009] The fused structured vector data is overlaid onto the railway-specific high-precision map data to obtain the fused railway-specific high-precision map.

[0010] In a second aspect, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the steps of the map data fusion method for intelligent lookout and collision avoidance of railway trains as described above.

[0011] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the map data fusion method for intelligent lookout and collision avoidance of railway trains as described above.

[0012] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the map data fusion method for intelligent lookout and collision avoidance of railway trains described above.

[0013] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a map data fusion method and related apparatus for intelligent collision avoidance of railway trains. By employing this method, the latitude and longitude information corresponding to key nodes in high-precision railway-specific map data, the kilometer marker information corresponding to key nodes in the ground-based data of train control equipment, and the track position relationships corresponding to key nodes in the five-color raster base map data of the station yard are obtained. A fused structured vector data is then established, and this fused structured vector data is superimposed onto the high-precision railway-specific map data to obtain a fused high-precision railway-specific map. This application generates a high-precision railway-specific map by fusing multi-source data, and uses this map for intelligent collision avoidance early warning of railway trains. This enables accurate differentiation between legally fixed targets and actual intruding foreign objects beside the tracks, and provides accurate and reliable collision avoidance early warning, fundamentally and significantly reducing the alarm rate of false collision avoidance warnings and improving the reliability of intelligent collision avoidance early warning for railway trains. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is an application environment diagram of a map data fusion method for intelligent lookout and collision avoidance of railway trains according to an embodiment of this application; Figure 2 A schematic flowchart illustrating a map data fusion method for intelligent lookout and collision avoidance of railway trains, provided as an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] The map data fusion method for intelligent lookout and collision avoidance of railway trains provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send the railway-specific high-precision map data, train control equipment ground base data, and station five-color raster base map data to be processed to server 104. After receiving the railway-specific high-precision map data, train control equipment ground base data, and station five-color raster base map data, server 104 processes the railway-specific high-precision map data, the train control equipment ground base data, and the station five-color raster base map data respectively to obtain railway-specific high-precision map structured vector data, train control equipment ground base structured vector data, and station five-color map preliminary structured vector data. The railway-specific high-precision map structured vector data includes multiple first key nodes and the latitude and longitude information corresponding to each first key node. The train control equipment ground base structured vector data includes multiple second key nodes and the latitude and longitude information and kilometer marker information corresponding to each second key node. The station five-color map preliminary structured vector data includes multiple third key nodes and the latitude and longitude information and track position relationship corresponding to each third key node. The track position relationship is the positional relationship between the key nodes and the tracks. The adjacency relationship encoding is performed on the track position relationships corresponding to all third key nodes in the preliminary structured vector data of the station five-color map to obtain the station five-color map structured vector data. The station five-color map structured vector data includes multiple third key nodes and structured encoded data of latitude and longitude information and track position relationships corresponding to each third key node. For each first key node in the railway-specific high-precision map structured vector data, a first operation is performed based on the railway-specific high-precision map structured vector data, the train control equipment ground foundation structured vector data, and the station five-color map structured vector data to obtain fused structured vector data. The fused structured vector data includes multiple first key nodes and structured encoded data of latitude and longitude information, kilometer marker information, and track position relationships corresponding to the first key nodes. The fused structured vector data is superimposed on the railway-specific high-precision map data to obtain the fused railway-specific high-precision map. The server 104 can feed back the obtained fused railway-specific high-precision map to the terminal 102.In addition, in some embodiments, the map data fusion method for intelligent observation and collision avoidance of railway trains can also be implemented separately by the server 104 or the terminal 102. For example, the terminal 102 can directly perform data fusion processing on the railway-specific high-precision map data, train control equipment ground base data and station five-color raster base map data to be processed. Alternatively, the server 104 can obtain the railway-specific high-precision map data, train control equipment ground base data and station five-color raster base map data to be processed from the data storage system and perform data fusion processing on the obtained data.

[0019] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, and IoT devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0020] In one exemplary embodiment, such as Figure 2 As shown, a map data fusion method for intelligent lookout and collision avoidance of railway trains is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 205. Wherein: Step 201: Obtain high-precision railway-specific map data, ground foundation data for train control equipment, and station five-color raster base map data.

[0021] Step 202: Process the railway-specific high-precision map data, the train control equipment ground base data, and the station yard five-color raster base map data respectively to obtain railway-specific high-precision map structured vector data, train control equipment ground base structured vector data, and station yard five-color map preliminary structured vector data. The railway-specific high-precision map structured vector data includes multiple first key nodes and the latitude and longitude information corresponding to each first key node. The train control equipment ground base structured vector data includes multiple second key nodes and the latitude and longitude information and kilometer marker information corresponding to each second key node. The station yard five-color map preliminary structured vector data includes multiple third key nodes and the latitude and longitude information and track position relationship corresponding to each third key node. The track position relationship is the positional relationship between the key nodes and the tracks.

[0022] Step 203: Encode the adjacency relationship of the track positions corresponding to all third key nodes in the preliminary structured vector data of the station five-color map to obtain the structured vector data of the station five-color map; the structured vector data of the station five-color map includes multiple third key nodes and the structured encoded data of the latitude and longitude information and track positions corresponding to each third key node.

[0023] Step 204: For each first key node in the railway-specific high-precision map structured vector data, based on the railway-specific high-precision map structured vector data, the train control equipment ground foundation structured vector data, and the station five-color map structured vector data, perform a first operation to obtain fused structured vector data; the fused structured vector data includes multiple first key nodes and structured encoded data of latitude and longitude information, kilometer marker information, and track position relationships corresponding to the first key nodes.

[0024] Step 205: The fused structured vector data is overlaid onto the railway-specific high-precision map data to obtain the fused railway-specific high-precision map.

[0025] Implementing steps 201 to 205 above provides a map data fusion method and related device for intelligent lookout and collision avoidance of railway trains. By adopting the method of this application, a fused high-precision railway-specific map is output, which can be applied to intelligent lookout and collision avoidance early warning of railway trains. The fused high-precision railway-specific map integrates multi-source data such as high-precision railway-specific map data, ground basic data of train control equipment, and five-color map data of station yards. It can perform comprehensive matching between the recognition results of images obtained during railway train operation and the fused high-precision railway-specific map data, accurately identifying legal fixed targets (such as signal lights and catenary supports) and real intruding foreign objects. This solves the problem in the prior art that relies solely on images obtained during operation for lookout and collision avoidance early warning, where overlapping images due to line-of-sight angles make it impossible to distinguish between fixed targets and foreign objects, leading to false collision warnings by identifying legal fixed targets as foreign objects (e.g., in curved sections of the track, the legal fixed targets on the trackside overlap with the rail projection in the acquired image, leading to misjudgment as illegal foreign object targets).

[0026] In another exemplary embodiment of this application, the detailed description of the railway-specific high-precision map data, the ground foundation data of the train control equipment, and the station yard five-color raster base map data in step 201 is as follows: High-precision map data for railways: Pre-stores the geographic coordinates (latitude and longitude), mileage, spatial dimensions (length, width, height) and safety clearance (minimum safe distance from the track centerline) of all fixed equipment (such as signal lights, catenary supports), buildings (such as bridges, tunnel entrances) and legal obstacles (such as guardrails) along the line, as well as the spatial dimensions (length, width, height) and safety clearance (minimum safe distance from the track centerline) of these fixed equipment, buildings, and legal obstacles. Supports quick retrieval of the location information of fixed targets by mileage segment or geographic area.

[0027] Station yard five-color raster base map data: First, a high-precision scanner is used to scan the railway dedicated station yard five-color map, obtaining a raster image base map containing ground control point elements, tracks, etc. The railway dedicated station yard five-color map includes the positional relationship between ground control points (signals, earth retainers, derailers, catenary endpoints, etc.) and tracks. The purpose of high-precision scanning is to convert paper / non-digital raw data into a raster base map. Second, multiple ground control points (such as signals with known latitude and longitude, track intersections) are selected on the raster image base map, and their coordinates are matched with the precise latitude and longitude coordinates of corresponding points in the high-precision geographic map (i.e., railway dedicated high-precision map data). A geometric transformation algorithm is then used to correct the deformation error of the raster base map, obtaining the station yard five-color raster base map data. The purpose of geometric correction is to eliminate paper deformation; the geometric correction algorithm is a polynomial transformation algorithm.

[0028] Ground-based data for train control equipment: Ground-based data for LKJ or GYK train operation monitoring equipment, including kilometer marker information for key nodes (such as signals, tunnel start and end points, and bridge start and end points).

[0029] In another exemplary embodiment of this application, the railway-specific high-precision map data, train control equipment ground base data, and station five-color raster base map data are all preprocessed data. The purpose of preprocessing is to eliminate errors in the original data, thus making the conversion of unstructured data (i.e., railway-specific high-precision map data, train control equipment ground base data, and station five-color raster base map data) into structured vector data more accurate. The preprocessing of the train control equipment ground base data also includes matching the latitude and longitude coordinates of key nodes in the train control equipment ground base data with the corresponding nodes in the railway-specific high-precision map data to obtain the latitude and longitude coordinates of the key nodes in the train control equipment ground base data.

[0030] By adopting the method of this application, a fused high-precision railway-specific map is output, which can be applied to intelligent lookout and collision avoidance warning for railway trains. The fused high-precision railway-specific map integrates multi-source data, including high-precision railway-specific map data, ground base data of train control equipment, and five-color map data of the station yard. It includes all legal fixed targets along the track (first key node, fixed equipment, buildings, and legal obstacles, etc.) as well as the latitude and longitude information, spatial dimensions, kilometer marker information, and structured coded data of track position relationships corresponding to the legal fixed targets. It can perform comprehensive matching between the recognition results of images obtained during railway train operation and the fused high-precision railway-specific map data, accurately identifying legal fixed targets along the track (such as signal lights and catenary supports) and real intruding foreign objects. This solves the problem in the existing technology that relies solely on images obtained during operation for lookout and collision avoidance warning, where overlapping images due to line-of-sight angles make it impossible to distinguish between fixed targets and foreign objects, leading to false collision warnings (such as in curved sections of the track, where the image of a legal fixed target along the track overlaps with the rail projection, leading to misjudgment as an illegal foreign object).

[0031] In another exemplary embodiment of this application, step 202 processes the railway-specific high-precision map data, the train control equipment ground foundation data, and the station yard five-color raster base map data respectively to obtain railway-specific high-precision map structured vector data, train control equipment ground foundation structured vector data, and station yard five-color map preliminary structured vector data, specifically including: Extract all map features and the boundaries of each map feature from the railway-specific high-precision map data, and determine the feature type of each map feature based on the boundaries of each map feature.

[0032] Based on the feature type of each map element, the railway-specific high-precision map data is divided into vector layers to obtain structured vector data of the railway-specific high-precision map.

[0033] Extract all train control equipment elements and the boundary of each train control equipment element from the ground basic data of the train control equipment, and determine the element type of each train control equipment element based on the boundary of each train control equipment element.

[0034] Based on the element type of each train control equipment element, the ground foundation data of the train control equipment is divided into vector layers to obtain the structured vector data of the ground foundation of the train control equipment.

[0035] Extract all station features and the boundaries of each station feature from the station five-color raster base map data, and determine the feature type of each station feature based on the boundaries of each station feature.

[0036] Based on the element type of each station element, the raster base map data of the station five-color map is divided into vector layers to obtain the preliminary structured vector data of the station five-color map.

[0037] The feature types include point features, line features, and polygon features.

[0038] In another exemplary embodiment of this application, the adjacency relationship encoding in step 203 is a digital storage method of topological relationships. This transforms the unstructured positional relationships between key nodes and tracks into structured codes, ensuring the logical consistency of the fused data, recording spatial relationships of map elements (such as adjacency and connectivity), and establishing spatial data associations. Specifically, the encoding method is topological structure encoding. This method expresses adjacency relationships by recording the topological relationships (such as adjacency, association, and inclusion) of geographic element entities. It is a core method of GIS spatial data encoding and can accurately reflect the spatial logical relationships between element entities. For example, encoding G001-J005 indicates that track G001 is associated with node J005, G001-L-G002 indicates that track G001 is adjacent to track G002, and J005→J006 indicates that node J005 is the preceding adjacent node of node J006. Data encoded with adjacency relationships establishes a topological structure, endowing the fused data with analytical capabilities. This provides reliable data support for path planning in network analysis support processes (such as road network topology in navigation systems), significantly improving the efficiency of spatial relationship queries and reducing data redundancy.

[0039] In another exemplary embodiment of this application, step 205, which involves overlaying the fused structured vector data onto the railway-specific high-precision map data to obtain the fused railway-specific high-precision map, specifically includes: High-precision railway-specific map data provides a realistic geographic spatial background and latitude and longitude coordinate reference. The structured vector data, after being fused in vector form, is overlaid on the high-precision railway-specific map data. Through geometric registration, the latitude and longitude information, kilometer marker information, and positional relationship between the key node and the track in each first key node in the fused structured vector data are bound to the corresponding first key node on the high-precision railway-specific map data. The combination of the two can simultaneously display the geographic spatial location and image details.

[0040] In another exemplary embodiment of this application, after overlaying the fused structured vector data onto the railway-specific high-precision map data to obtain the fused railway-specific high-precision map, the method further includes: Based on the feature types of all map elements in the railway-specific high-precision map data, the fused railway-specific high-precision map is subjected to feature layer coloring to obtain a colored railway-specific high-precision map.

[0041] This embodiment divides the fused high-precision railway map into layers according to element types (point elements, line elements, and polygon elements). For example, the first layer is the point element layer, the second layer is the line element layer, and the third layer is the polygon element layer. Different layers and types of elements are assigned unique colors, with color differences reflecting data indicators; for example, signal nodes are red, and derailment nodes are blue. Through overlay and coloring, the relationships between the three types of attributes (latitude and longitude information, kilometer marker information, and track position relationships) of each key node can be presented intuitively. The resulting colored high-precision railway map combines accurate attribute information with detailed geographical background. It can clearly display the vector elements of track alignment and node distribution, and can also overlay a high-precision railway map raster base to present the real geographical environment. Layered coloring allows for intuitive differentiation of different types of nodes, different tracks, and different areas, facilitating manual verification and review.

[0042] In another exemplary embodiment of this application, after overlaying the fused structured vector data onto the railway-specific high-precision map data to obtain the fused railway-specific high-precision map, the method further includes: The fused high-precision railway map is abstracted as an undirected graph; the undirected graph includes a vertex set and an edge set; the vertex set includes multiple vertices and information for each vertex, each vertex including a first key node, and the information of each vertex is structured encoded data of latitude and longitude information, kilometer marker information, and track position relationship corresponding to the first key node; the edge set includes multiple edges, and each edge represents the spatial connection relationship between a first key node and other first key nodes.

[0043] This embodiment can generate an undirected graph, which reflects spatial relationships and clearly shows that all element nodes are not isolated, but rather logically connected to surrounding nodes through the "vertex-edge" structure of the undirected graph. For example, "a certain signal belongs to track 3 and is connected to the adjacent signals before and after it." The undirected graph contains a structured vector data topology. Based on the undirected graph model, it can also directly support practical applications such as railway train route planning, node spatial relationship analysis, and facility operation and maintenance management.

[0044] In another exemplary embodiment of this application, for each first key node in the railway-specific high-precision map structured vector data, based on the railway-specific high-precision map structured vector data, the train control equipment ground foundation structured vector data, and the station yard five-color map structured vector data, a first operation is performed to obtain fused structured vector data, specifically including: For each first key node in the structured vector data of the high-precision map dedicated to the railway, a second key node that coincides with the latitude and longitude information of the first key node is found in the structured vector data of the ground foundation of the train control equipment, and the kilometer marker information corresponding to the second key node is read.

[0045] For each first key node in the railway-specific high-precision map structured vector data, a third key node that overlaps with the latitude and longitude information of the first key node is found in the station five-color map structured vector data, and the structured coded data of the track position relationship corresponding to the third key node is read.

[0046] The structured coded data of the read kilometer marker information and track position relationship is bound to the first key node, so that each first key node integrates latitude and longitude information, kilometer marker information, and structured coded data of track position relationship, resulting in fused structured vector data.

[0047] In another exemplary embodiment of this application, the method further includes: encoding the fused structured vector data using a quadtree encoding method to obtain a spatially compressed set; the spatially compressed set includes a quadtree index table, a block geometry table, and a block attribute table.

[0048] The attribute information in the block attribute table is encoded using run-length encoding to obtain a block attribute compression table.

[0049] In the above embodiments, quadtree encoding is used to achieve recursive hierarchical compression encoding of spatial geographic range, subdividing the space and encapsulating the data for easy retrieval. The fused structured vector data is recursively layered according to accuracy requirements, such as top layer (dividing the geographic range of the entire railway line), middle layer (subdividing into areas such as stations and sections), and bottom layer (subdividing into smaller areas such as signals and derailment devices), achieving hierarchical storage and fast retrieval.

[0050] The run-length encoding method in the above embodiments is a lossless encoding method for compressing consecutive units with the same attribute (compression ratio 5-10 times). The structured vector data fused in this application contains a large number of elements with consecutive repetitive attributes, like consecutive signals on a track. Run-length encoding compresses these node or line data, replacing the original repetitive data stored element by element with the form of "attribute value + number of consecutive repetitions". For example, instead of storing "track 1 - signal A, track 1 - signal B, track 1 - signal C" individually, it can be simplified to "track 1 + 3 consecutive signals" after encoding.

[0051] The above embodiments utilize quadtree coding to optimize hierarchical management of the "spatial dimension," while run-length encoding is responsible for compressing repetitive data in the "attribute dimension." The combination of these two methods results in fused structured vector data (including structured coded data of latitude and longitude information, kilometer marker information, and track position relationships) that is small in size, easy to store, and capable of rapid retrieval and flexible application. Through quadtree coding and run-length encoding, the fused structured vector data is processed to obtain a spatially compressed set, namely, multiple key node association datasets with complete attributes. Each key node data integrates the core attributes of three types of original data, forming a standardized data record. Each key node data can be summarized as: a unique identifier for the key node, latitude and longitude information (latitude and longitude coordinates) from railway-specific high-precision map data, kilometer marker information from ground base data of train control equipment, and track position relationships from station five-color raster base map data. These track position relationships refer to the positional relationship between the key node and the track.

[0052] In another exemplary embodiment of this application, after obtaining the fused high-precision railway-specific map, the method further includes: A topology check operation is performed on the fused high-precision railway map to check the topological relationships and connectivity between map features and to verify the attribute information corresponding to the features.

[0053] Topology checks are primarily implemented through three methods: topology rule validation, spatial relationship analysis, and attribute consistency verification. Specifically, these include: Using mainstream GIS software such as ArcGIS and GGIS, topology rules can be preset to detect topology errors between features in batches.

[0054] Verify the consistency between the attribute information of the elements and the spatial topological relationships, and check the integrity of the attribute fields (whether there are null values ​​or missing values ​​and data format errors).

[0055] Using libraries such as ArcPy and GeoPandas, scripts are written to extract spatial attributes (such as area, length, and adjacent features) of features in batches, automatically compare them with records in the attribute table, and generate error reports for verification.

[0056] Topology checks can correct dangling nodes and overlapping polygons, ensuring the consistency of fused data. During vectorization, issues such as dangling nodes (e.g., track endpoints not connected to signal points) and overlapping polygons (e.g., duplicate administrative division boundaries) may occur. By checking the topological relationships and connectivity between features, verifying feature attribute information, and correcting errors, problems in the vectorization process are eliminated, resulting in a high-precision vector data source that provides a data foundation for subsequent operations.

[0057] In another exemplary embodiment of this application, the process of obtaining the fused structured vector data can be defined as follows: through a data preparation management process, following the serial flow of "high-precision scanning of the station five-color map → station five-color map raster base map → geometric correction → layered vectorization of railway-specific high-precision map data, train control equipment ground base data and station five-color map → fused structured vector data → topology check", the three types of data (railway-specific high-precision map data, train control equipment ground base data and station five-color map) are processed to unify the data expression form, and the three originally scattered original data are merged into a unified vector element attribute to form structured vector data of "spatial location + attribute information", thereby obtaining the fused structured vector data.

[0058] In this process, a unified data representation format is established through a basic data structure model. Key nodes in railway-specific high-precision map data, train control equipment ground base data, and station five-color raster base map data are defined as point elements; railways / rivers are defined as line elements; and station areas and administrative divisions are defined as area elements. Core attribute fields are clarified, and a unified geographic entity representation framework is established, providing a fusionable "carrier" for the three types of data. By expanding the data structure to record spatial relationships, key node kilometer marker information, track location relationships, and key node latitude and longitude information are processed and transformed into vector data. The attributes and spatial locations of nodes are presented in the form of point, line, and area elements, thus obtaining fused structured vector data.

[0059] The map data fusion method for intelligent observation and collision avoidance of railway trains proposed in this application can ultimately output a unified, structured core dataset containing the core attribute information of three types of raw data: railway-specific high-precision map data, train control equipment ground base data, and station five-color raster base map data. This dataset can be directly used for spatial analysis, as well as an undirected graph and a colored railway-specific high-precision map.

[0060] Specifically, the results can be broken down into three levels: (1) Core dataset results: The fused structured vector data is compressed and encoded using quadtree coding and run-length coding to obtain a spatial compressed set. The spatial compressed set includes multiple key node association datasets with complete attributes. Each key node data integrates the core attributes of three types of original data: railway-specific high-precision map data, train control equipment ground base data, and station five-color raster base map data, forming a standardized data record. The format can be summarized as: {unique node identifier; kilometer marker information (from train control equipment ground base data); structured coded data of track position relationships (from station five-color raster base map data); latitude and longitude information (from railway-specific high-precision map data); spatial association relationship (topological structure reflected by adjacency relationship coding)}.

[0061] (2) Undirected graph: An undirected graph shows that all key nodes do not exist in isolation, but are logically connected with surrounding nodes through the "vertex-edge" structure of the undirected graph, such as "a certain signal belongs to track 3 and is connected to the adjacent signals in front and behind". Undirected graphs contain structured vector data topology. Based on the undirected graph model, it can also directly support practical applications such as railway train route planning, node spatial relationship analysis, and facility operation and maintenance management.

[0062] (3) Visualization and Application Results: The elements in the fused high-precision railway map are layered and colored according to point, line, and area elements to obtain a colored high-precision railway map. The colored high-precision railway map is based on the fused structured vector data, combined with vector-raster integration technology and feature layering coloring method to generate a railway map that combines accurate attribute information and geographic background details. This map can clearly display the vector elements of track alignment and node distribution, and can also overlay the geographic raster base map of the high-precision railway map to present the real geographic environment. Through layering coloring, different types of nodes, different tracks, and different areas can be intuitively distinguished, which is convenient for manual verification and viewing.

[0063] In summary, the map data fusion method for intelligent observation and collision avoidance of railway trains in this application does not ultimately result in three fragmented types of raw data, but rather a three-in-one railway spatial data asset consisting of "attributes + location + relationships". This asset can be directly used for map visualization and can also support subsequent business analysis such as railway operation and planning.

[0064] The integrated high-precision railway-specific map can be applied to intelligent train lookout and collision avoidance early warning systems, and has the following advantages: (1) When the railway train is running, when there is a positioning signal, the railway train realizes navigation and sighting and collision avoidance of fixed facilities based on satellite positioning information. When the positioning is interrupted in scenarios such as tunnels and mountainous areas, i.e. when the positioning signal is lost, the railway train combines the offline matching function of the railway-specific high-precision map and realizes navigation and sighting and collision avoidance early warning based on the kilometer marker information, so as to ensure continuous monitoring in the entire line scenario and improve the robustness of the system.

[0065] (2) The fused railway-specific high-precision map includes all legal fixed targets along the track (first key node, fixed equipment, buildings and legal obstacles, etc.) as well as the latitude and longitude information, spatial dimensions, kilometer marker information and structured coded data of track position relationships corresponding to the legal fixed targets. It can match the recognition results of images obtained during the operation of railway trains with the fused railway-specific high-precision map data in all aspects, accurately identify legal fixed targets along the track (such as signal lights and catenary support pillars) and real intruding foreign objects. It solves the problem in the existing technology that relies solely on images obtained during operation for lookout and collision avoidance warning, where the images overlap due to the angle of view, making it impossible to distinguish between fixed targets and foreign objects, and misidentifying legal fixed targets along the track as foreign objects, resulting in false collision avoidance warnings (such as in track curve sections, where the legal fixed targets along the track overlap with the rail projection in the obtained image, and are mistakenly judged as illegal foreign objects). It avoids the problem of misidentifying legal fixed targets along the track as foreign objects, resulting in false alarms.

[0066] (3) For dynamic targets (such as pedestrians and vehicles that suddenly break in) and illegal intruders (such as falling rocks and fallen trees) not included in the fused high-precision railway map, multiple information verification can be performed with the multi-source data in the fused high-precision railway map to achieve accurate detection of the targets.

[0067] The integrated high-precision railway-specific map provides highly reliable, timely, and comprehensive multi-dimensional information for intelligent railway collision avoidance, facilitating vehicle navigation and fixed facility lookout and collision avoidance early warning, and significantly reducing the false alarm rate.

[0068] This application also provides an application scenario in which the above-mentioned map data fusion method for intelligent observation and collision avoidance of railway trains is applied. Specifically, the map data fusion method for intelligent observation and collision avoidance of railway trains provided in this embodiment can be applied in a map data fusion scenario. The map data fusion scenario includes a data acquisition stage, a data processing stage, and a result display stage; data enters the data processing stage from the content acquisition stage, is processed to obtain the corresponding results, and then enters the downstream result display stage. The map data fusion method for intelligent observation and collision avoidance of railway trains provided in this embodiment belongs to the data processing stage. Specifically, in the data processing process of the map data fusion method for intelligent observation and collision avoidance of railway trains, high-precision map data for railway use, ground base data of train control equipment, and five-color raster base map data of the station yard are acquired. The railway-specific high-precision map data, the train control equipment ground base data, and the station yard five-color raster base map data are processed respectively to obtain railway-specific high-precision map structured vector data, train control equipment ground base structured vector data, and station yard five-color map preliminary structured vector data. The railway-specific high-precision map structured vector data includes multiple first key nodes and the latitude and longitude information corresponding to each first key node. The train control equipment ground base structured vector data includes multiple second key nodes and the latitude and longitude information and kilometer marker information corresponding to each second key node. The station yard five-color map preliminary structured vector data includes multiple third key nodes and the latitude and longitude information and track position relationships corresponding to each third key node. Adjacency relationship encoding is performed on the track position relationships corresponding to all third key nodes in the station yard five-color map preliminary structured vector data to obtain station yard five-color map structured vector data. The station yard five-color map structured vector data includes multiple third key nodes and the latitude and longitude information and track position relationships corresponding to each third key node, as well as structured encoded data. For each first key node in the railway-specific high-precision map structured vector data, a first operation is performed based on the railway-specific high-precision map structured vector data, the train control equipment ground foundation structured vector data, and the station yard five-color map structured vector data to obtain fused structured vector data. The fused structured vector data includes multiple first key nodes and structured coded data of the latitude and longitude information, kilometer marker information, and track position relationships corresponding to the first key nodes. The fused structured vector data is then overlaid onto the railway-specific high-precision map data to obtain the fused railway-specific high-precision map.

[0069] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 3As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores high-precision railway-specific map data, ground-based data for train control equipment, and five-color raster base map data for stations. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a map data fusion method for intelligent collision avoidance of railway trains.

[0070] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0071] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0072] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0073] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use and processing of the relevant data are carried out in compliance with the relevant data protection laws and policies of the country where the location is located, and with the authorization granted by the owner of the corresponding device.

[0074] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0075] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0076] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0077] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A map data fusion method for intelligent lookout and collision avoidance of railway trains, characterized in that, The map data fusion method for intelligent observation and collision avoidance of railway trains includes: Acquire high-precision railway-specific map data, ground-based data for train control equipment, and five-color raster base map data for stations; The railway-specific high-precision map data, the train control equipment ground base data, and the station yard five-color raster base map data are processed respectively to obtain railway-specific high-precision map structured vector data, train control equipment ground base structured vector data, and station yard five-color map preliminary structured vector data. The railway-specific high-precision map structured vector data includes multiple first key nodes and the latitude and longitude information corresponding to each first key node. The train control equipment ground base structured vector data includes multiple second key nodes and the latitude and longitude information and kilometer marker information corresponding to each second key node. The station yard five-color map preliminary structured vector data includes multiple third key nodes and the latitude and longitude information and track position relationships corresponding to each third key node. The track position relationships are the positional relationships between key nodes and tracks. Adjacency relationship encoding is performed on the track position relationships corresponding to all third key nodes in the preliminary structured vector data of the station five-color map to obtain the station five-color map structured vector data; the station five-color map structured vector data includes multiple third key nodes and the structured encoded data of the latitude and longitude information and track position relationships corresponding to each third key node; For each first key node in the railway-specific high-precision map structured vector data, based on the railway-specific high-precision map structured vector data, the train control equipment ground foundation structured vector data, and the station five-color map structured vector data, a first operation is performed to obtain fused structured vector data; the fused structured vector data includes multiple first key nodes and structured encoded data of latitude and longitude information, kilometer marker information, and track position relationships corresponding to the first key nodes. The fused structured vector data is overlaid onto the railway-specific high-precision map data to obtain the fused railway-specific high-precision map.

2. The map data fusion method for intelligent observation and collision avoidance of railway trains according to claim 1, characterized in that, The railway-specific high-precision map data, the train control equipment ground foundation data, and the station yard five-color raster base map data are processed respectively to obtain structured vector data of the railway-specific high-precision map, structured vector data of the train control equipment ground foundation, and preliminary structured vector data of the station yard five-color map, specifically including: Extract all map features and the boundaries of each map feature from the railway-specific high-precision map data, and determine the feature type of each map feature based on the boundaries of each map feature. Based on the feature type of each map element, the railway-specific high-precision map data is divided into vector layers to obtain structured vector data of the railway-specific high-precision map. Extract all train control equipment elements and the boundary of each train control equipment element from the ground basic data of the train control equipment, and determine the element type of each train control equipment element based on the boundary of each train control equipment element. Based on the element type of each train control equipment element, the ground foundation data of the train control equipment is divided into vector layers to obtain the structured vector data of the ground foundation of the train control equipment. Extract all station features and the boundary of each station feature from the station five-color raster base map data, and determine the feature type of each station feature based on the boundary of each station feature. Based on the element type of each station element, the raster base map data of the station five-color map is divided into vector layers to obtain the preliminary structured vector data of the station five-color map; The feature types include point features, line features, and polygon features.

3. The map data fusion method for intelligent observation and collision avoidance of railway trains according to claim 2, characterized in that, Also includes: Based on the feature types of all map elements in the railway-specific high-precision map data, the fused railway-specific high-precision map is subjected to feature layer coloring to obtain a colored railway-specific high-precision map.

4. The map data fusion method for intelligent observation and collision avoidance of railway trains according to claim 1, characterized in that, Also includes: The fused high-precision railway-specific map is abstracted into an undirected graph. The undirected graph includes a set of vertices and a set of edges; The vertex set includes multiple vertices and information for each vertex. Each vertex includes a first key node. The information of each vertex is structured encoded data of latitude and longitude information, kilometer marker information, and track position relationship corresponding to the first key node. The edge set includes multiple edges, and each edge represents the spatial connection relationship between a first key node and other first key nodes.

5. The map data fusion method for intelligent observation and collision avoidance of railway trains according to claim 1, characterized in that, For each first key node in the railway-specific high-precision map structured vector data, based on the railway-specific high-precision map structured vector data, the train control equipment ground foundation structured vector data, and the station yard five-color map structured vector data, a first operation is performed to obtain fused structured vector data, specifically including: For each first key node in the railway-specific high-precision map structured vector data, a second key node that coincides with the latitude and longitude information of the first key node is found in the ground foundation structured vector data of the train control equipment, and the kilometer marker information corresponding to the second key node is read. For each first key node in the railway-specific high-precision map structured vector data, a third key node that overlaps with the latitude and longitude information of the first key node is found in the station five-color map structured vector data, and the structured coded data of the track position relationship corresponding to the third key node is read. The structured coded data of the read kilometer marker information and track position relationship is bound to the first key node, so that each first key node integrates latitude and longitude information, kilometer marker information, and structured coded data of track position relationship, resulting in fused structured vector data.

6. The map data fusion method for intelligent observation and collision avoidance of railway trains according to claim 1, characterized in that, Also includes: The fused structured vector data is encoded using a quadtree encoding method to obtain a spatially compressed set; The spatial compression set includes a quadtree index table, a block geometry table, and a block attribute table.

7. The map data fusion method for intelligent observation and collision avoidance of railway trains according to claim 6, characterized in that, Also includes: The attribute information in the block attribute table is encoded using run-length encoding to obtain a block attribute compression table.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the map data fusion method for intelligent lookout and collision avoidance of railway trains according to any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the map data fusion method for intelligent lookout and collision avoidance of railway trains as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the map data fusion method for intelligent lookout and collision avoidance of railway trains as described in any one of claims 1-7.

Images