A method for representing and constructing environmental map data in mining areas and a map management system

Abstract

This invention discloses a method for representing and constructing mining area environmental map data, as well as a map management system. The map management system includes a map client and a map server. The map client is deployed on a vehicle-mounted device, which travels within the work area (section, segment, or sub-section) where the map client needs to collect data. During travel, local map data is collected and constructed, and the constructed local map data is uploaded to the map server. The map server constructs global map data based on the acquired local map data and pushes the right to use the global map data to the map client. This invention enables effective control over the overall or partial operational environment status information, effectively overcoming many problems in traditional mines such as extensive management style, outdated management methods, and low management efficiency, laying the foundation for digital operation and maintenance and unmanned production in mining areas.

Description

Technical Field

[0001] This invention belongs to the field of engineering machinery technology and relates to a method for representing and constructing mining area environmental map data and a map management system. Background Technology

[0002] Traditional manned mines are characterized by extensive management styles, outdated management methods, and low management efficiency. Furthermore, manned mines often present numerous challenges, including high risk, high costs, high labor intensity, and low efficiency. For unmanned mining areas, real-time monitoring of the entire mining environment is necessary, thus requiring the creation of a comprehensive map of the entire mining area. Chinese invention patent CN111443360A discloses an automatic road boundary acquisition device and identification method for an unmanned driving system in a mining area. The acquisition device includes an engineering vehicle, an onboard computer system, an onboard positioning and terrain scanning system, a map data processing platform, and a mining area unmanned driving control platform. The engineering vehicle carries the onboard computer system and the onboard positioning and terrain scanning system. The onboard computer system controls the onboard positioning and terrain scanning system to acquire data. The onboard positioning and terrain scanning system includes a positioning device and a scanning device for acquiring location data and point cloud data. The map data processing platform receives the location data and point cloud data transmitted by the onboard computer system, automatically extracts the road edge vector boundaries, and generates an incremental vector map. The mining area unmanned driving control platform monitors real-time changes in road conditions in the mining area, automatically issues acquisition tasks to the vehicle, and the vehicle acquires and uploads road data within a designated area according to the task assignment. Chinese invention patent CN113156451A discloses an unstructured road boundary detection method, device, storage medium, and electronic device, relating to the field of vehicle environmental perception technology. This invention acquires and preprocesses single-frame scan data from multi-line radar; obtains non-ground points based on the preprocessed multi-line radar single-frame scan data; projects the non-ground points onto a raster map; obtains multiple potential boundary points for different field of view angles based on the raster map and the scanning method; and merges these potential boundary points to obtain a set of boundary points. This invention, based on multi-field-of-view boundary detection, can more accurately extract boundary points of unstructured road depressions and occluded areas under a single field of view, improving the accuracy of road boundary identification. It is evident that existing technologies only address the detection and extraction of road boundaries in mining areas, not providing a complete mining area environmental map data acquisition solution. If the mining area environmental map data is incomplete, collaboration between the control center and operating equipment cannot be achieved from an operational environment perspective. Therefore, it cannot fulfill the fundamental mission of digital operation and maintenance and unmanned production in mining areas. Summary of the Invention

[0003] The purpose of this invention is to provide a method for representing and constructing mining area environmental map data and a map management system. When a vehicle equipped with the map management system is in motion, it can obtain real-time data on the operating environment around the vehicle through the map management system, construct a local map, and push it to the control center. The control center can then maintain and publish the global map data based on the obtained local map data.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of this invention provides a method for representing mining area environmental map data, comprising:

[0006] The global map data of the entire mining area operation environment is represented as a tree-like topology; the global map data of the entire mining area operation environment represents the status information of the entire mining area operation environment, as well as the status information generated by the interaction between vehicles and equipment and the mining area operation environment.

[0007] The tree-like topology includes: a main node, subgraph nodes, and leaf nodes;

[0008] The main node is the root node; the sub-graph nodes are child nodes of the main node; the leaf nodes are used to store local map data of the mining area's operating environment; the main node and sub-graph nodes are used to store the regional attributes to which the mining area belongs.

[0009] The leaf node includes at least a zone / sub-zone node; the zone / sub-zone node is a child node of the sub-graph node; the zone refers to any complete working area in the entire mining area's working environment;

[0010] Any leaf node represents a hierarchical topology, including static and dynamic layers;

[0011] The static layer is used to store the static features of the work area corresponding to the leaf node; the static features refer to environmental state features that do not change within a certain time interval.

[0012] The dynamic layer is used to store the dynamic features of the work area corresponding to the leaf node; the dynamic features refer to the environmental state features that exist only within a specific time interval.

[0013] Furthermore, the leaf node also includes a segment node; the segment node is a child node of the region / subregion node;

[0014] The term "segment" refers to any uncontrolled road segment within a complete work area.

[0015] Furthermore, the leaf node also includes a segment node; the segment node is a child node of the segment node;

[0016] The term "sub-segment" refers to a sub-segment of a non-controlled road segment according to traffic rules.

[0017] Furthermore, the static layer includes a raw data layer and a feature data layer;

[0018] The raw data layer includes static features representing the environmental state of the work area in the form of raw sensor data;

[0019] The feature data layer includes static features representing the environmental state of the work area by processing the raw sensor data; the processing includes thinning, clustering, feature extraction, tracking, and rasterization.

[0020] Furthermore, the dynamic layer includes a real-time data layer and a historical data layer;

[0021] The real-time data layer includes the real-time dynamic characteristics of the work area;

[0022] The historical data layer includes the accumulation of dynamic characteristics of the work area over a certain time interval.

[0023] Furthermore, the dynamic layer also includes a task layer;

[0024] The task layer includes status information representing the task objects contained in the task area.

[0025] A second aspect of the present invention provides a method for constructing mining area environmental map data, comprising:

[0026] Acquire raw state data and preprocess it; the raw state data includes navigation and positioning data and environmental perception data of the corresponding work area in the mining operation environment.

[0027] Feature data extraction and classification are performed on the preprocessed original state data;

[0028] Based on the preprocessed original state data, extracted feature data, and classification results, judgments and filters are performed to form the original map data;

[0029] Based on the original map data, local map data is constructed;

[0030] The intersection of all constructed local map data is calculated based on spatial location coordinates, and each local map data is segmented according to the leaf nodes in the aforementioned mining area environmental map data representation method.

[0031] Based on the local map data segmentation results, global map data is mapped to form global map data represented according to the aforementioned mining area environment map data representation method;

[0032] The mapping relationship between local map data and global map data includes: the mapping relationship between the local map data segmentation results and the leaf nodes in the global map data; and the mapping relationship between each level in the local map data segmentation results and each level in the leaf nodes of the global map data.

[0033] Furthermore, feature data extraction and classification are performed on the preprocessed original state data.

[0034] The feature data includes obstacle pose information, obstacle motion state information, obstacle size information, obstacle shape information, road condition information within the work area, and boundary information of the work area;

[0035] The classification refers to the process of determining the motion state of obstacles, assessing the probability of obstacle movement, and classifying obstacle types based on feature data.

[0036] Furthermore, the determination and filtering based on the preprocessed original state data, extracted feature data, and classification results to form original map data includes:

[0037] Determine the mapping relationship between the preprocessed original state data and the feature data;

[0038] Remove components that do not belong to the local map data and their corresponding original state data.

[0039] Furthermore, the construction of local map data based on the original map data includes:

[0040] In the dynamic layer, dynamic components of local map data are constructed, and historical data of the dynamic components of local map data are saved. In the static layer, static components of local map data are constructed.

[0041] Furthermore, it also includes:

[0042] After constructing the global map data, as data is continuously collected, the steps to repair the global map data based on the segmentation results of each local map data are as follows;

[0043] The repair of global map data refers to repairing the mapping relationship between local map data and global map data based on the segmentation results of each local map data.

[0044] Furthermore, it also includes the step of performing anomaly detection on the global map data.

[0045] Distinguish between normal and abnormal components in the global map data; for normal components, grant usage rights; for abnormal components, generate and publish corresponding local map data collection tasks; normal components refer to components with completely defined attributes in the updated global map data, and the rest are abnormal components.

[0046] Furthermore, it also includes: determining whether the current global map data needs to be updated based on the segmentation results of the local map data; if so, then...

[0047] Based on the mapping relationship between local map data and global map data, as well as the key features of the local map, incremental map data is extracted from the preprocessed original state data on the basis of the global map data.

[0048] The global map data is updated based on incremental map data.

[0049] Furthermore,

[0050] Incremental map data of static graphs are extracted based on the XOR logic of the probability density function of data features.

[0051] The incremental map data of the dynamic graph is extracted based on the weighted average logic of the probability density function of data features.

[0052] A third aspect of the present invention provides a map management system, comprising: a map client and a map server;

[0053] The map client is used to collect and construct local map data using the aforementioned mining area environmental map data construction method, and uploads the constructed local map data to the map server; the map client is deployed on vehicle equipment, and the vehicle equipment travels in the working area of ​​the area, segment, or sub-segment that the map client needs to collect data from;

[0054] The map server is used to construct global map data based on the acquired local map data using the aforementioned mining area environment map data construction method, and to push the right to use the global map data to the map client.

[0055] Furthermore, the map client includes:

[0056] The data acquisition module is used to read and preprocess raw state data in the mining area's working environment; the raw state data includes navigation and positioning data and working environment perception data.

[0057] The detection submodule is used to read the preprocessed raw state data and extract feature data; the feature data includes: obstacle pose information, obstacle motion state information, obstacle size information, obstacle shape information, road condition information within the work area, and boundary information of the work area;

[0058] The classification submodule is used to read the feature data extracted by the detection submodule, and to perform obstacle motion state discrimination, obstacle movement probability determination, and obstacle type classification.

[0059] The analysis submodule is used to simultaneously read the preprocessed raw state data, feature data, and the classification results of the feature data, and to judge and filter the read data to form the raw map data.

[0060] A construction submodule is used to read the original map data from the analysis submodule, construct the local map data, and locate the vehicle or equipment in the local map.

[0061] The interface module is used to determine whether the completed local map data needs to be pushed to the map server. If so, the local map data is pushed to the map server according to certain rules; otherwise, the newly constructed local map data is updated by weighted averaging with historical data.

[0062] Furthermore, the data acquisition module in the map client is a sensor system installed on the vehicle itself.

[0063] Furthermore, the analysis submodule is specifically used for,

[0064] Based on the feature data, remove noise and the corresponding original state data;

[0065] Based on the classification results of the feature data, the mapping relationship between the original state data and the feature data is determined, thus forming the original map data.

[0066] Furthermore, the map server includes:

[0067] The data acquisition module is used to read raw state data and preprocess it; the raw state data includes local map data from various map clients.

[0068] The analysis submodule is used to read the preprocessed raw state data, segment the local map data, and extract key feature data.

[0069] The integration submodule is used to read the segmentation results of each local map data in the analysis submodule and repair the mapping relationship between the segmentation results of the local map data and the global map data;

[0070] The interface module is used to generate corresponding local map data collection tasks for abnormal components in the global map data and push them to the map client; as well as to push the usage rights of normal components in the global map data to the map client.

[0071] Furthermore, the data acquisition module of the map server is a communication system between the map server and each vehicle device equipped with a map client.

[0072] Furthermore, the analysis submodule is specifically used for,

[0073] The intersection of each local map data is calculated based on the spatial coordinates. Each local map data is then segmented and, according to the tree-like topology of the global map data, it is segmented into leaf nodes in the global map data.

[0074] Furthermore, the map server also includes:

[0075] The update submodule is used to determine whether the current global map data needs to be updated based on the processing results of the local map data. If so, it sends a global map update command to the processing submodule; and reads the incremental map data from the processing submodule and updates the global map data accordingly.

[0076] The processing submodule is used to simultaneously read the processing results of the preprocessed original state data and the local map data. Based on the global map update command issued by the update submodule, and the processing results of the local map data, it extracts incremental map data from the preprocessed original state data on the basis of the existing global map data.

[0077] Furthermore, the interface module of the map server also includes:

[0078] The map decision submodule is used to determine the anomalies of the current global map data and distinguish between normal and abnormal components. Normal components refer to components in the updated global map data whose attributes are completely determined, while abnormal components are the opposite.

[0079] Furthermore,

[0080] The map client is deployed on vehicles or equipment in the mining area that do not have a control center.

[0081] The map server is deployed in the control center of the mining area or on a vehicle equipped with a control center.

[0082] The beneficial effects of this invention are as follows:

[0083] This invention enables effective control over overall or partial operational environment status information, effectively overcoming many problems in traditional mines such as extensive management style, outdated management methods, and low management efficiency, laying the foundation for digital operation and maintenance and unmanned production in mining areas. Attached Figure Description

[0084] Figure 1 This is a schematic diagram of the tree-like topology of map data in Embodiment 1 of the present invention;

[0085] Figure 2 This is a schematic diagram of the hierarchical topology of map data in Embodiment 1 of the present invention;

[0086] Figure 3 This is a block diagram illustrating the implementation of the map processor in Embodiment 3 of the present invention;

[0087] Figure 4 This is a block diagram illustrating the implementation of the map management system in Embodiment 3 of the present invention;

[0088] Figure 5 This is a block diagram illustrating the implementation of a map client in one embodiment of the present invention.

[0089] Figure 6 This is an example of the functional architecture and data processing flow of each module of the map client in one embodiment of the present invention;

[0090] Figure 7 This is a block diagram illustrating the implementation of a map server in one embodiment of the present invention.

[0091] Figure 8 This is an example of the functional architecture and data processing flow of each module of the map server in an embodiment of the present invention. Detailed Implementation

[0092] The present invention will now be further described. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0093] Example 1

[0094] This embodiment provides a method for representing environmental map data in a mining area, as detailed below:

[0095] The entire mining area environment is classified into three levels: zones / sub-zones, sections, and sub-sections. The entire mining area environment can be divided into several zones / sub-zones, each of which may contain several sections, and each section may contain several sub-sections.

[0096] It should be noted that a zone refers to a specific, complete work area, such as a loading zone, a safety zone, a driving zone, and an unloading zone; a sub-zone refers to a part of the complete work area that is related to the work task, such as a waiting sub-zone or a control sub-zone.

[0097] It should be noted that a sub-region may or may not be a part of a region. The difference between the two lies not only in the scope but also in whether it is temporary. A region is generally static; once defined, its scope may change at any time but it is generally not arbitrarily revoked. The existence of a sub-region, on the other hand, is dynamic and temporary.

[0098] It should be noted that a "segment" refers to a non-controlled section within a complete work area, while a controlled section in a mining environment may belong to a controlled sub-area and correspond to specific work rules.

[0099] It should be noted that a sub-segment refers to a division of non-controlled road sections due to traffic rule restrictions.

[0100] The map data is represented as follows:

[0101] From a macro perspective, map data can be represented as a tree-like topological structure, called a map tree;

[0102] Map data, or global map data, can represent the status information of the mining area's working environment, as well as the status information generated by the interaction between vehicles and equipment and the working environment.

[0103] From a micro perspective, each leaf node in a map tree can be represented as a hierarchical topology, known as a local map.

[0104] like Figure 1 As shown, in one embodiment of the present invention, the global map data can be represented as a tree-like topology, which may include a global node, sub-map nodes, region / sub-region nodes, segment nodes, and sub-segment nodes, and has the following characteristics:

[0105] (1) The total node is the root node;

[0106] (2) A subgraph node is a child node of the main node;

[0107] (3) A region / subregion node is a child node of a subgraph node;

[0108] (4) Segment nodes are child nodes of region / subregion nodes;

[0109] (5) A segment node is a child node of a segment node;

[0110] It should be noted that area / subarea nodes, segment nodes, and subsegment nodes can all be leaf nodes.

[0111] It should be noted that only leaf nodes store specific map data. Non-leaf nodes do not store map data, but store some regional attributes, such as region number, region boundary information, region status (whether it is available), region level (the node level of the region), region composition (which regions make up the region and what their respective numbers are), etc.

[0112] like Figure 2 As shown, in one embodiment of the present invention, each local map data (i.e., leaf node) can be represented as a hierarchical topology, including static layers and dynamic layers, and has the following characteristics:

[0113] Static layers are used to represent the static characteristics of a work area, that is, environmental features that do not change over a long period of time. Examples include curbs, retaining walls, signal poles, and cable bridges, which are environmental features that do not change over a long period of time.

[0114] Furthermore, the static layer may contain two layers: the raw data layer and the feature data layer. The raw data layer represents the static features of the work area in the form of raw state data, while the feature data layer represents the static features of the work area in the form of feature data derived from the raw state data.

[0115] Furthermore, the raw data layer is a map built based on the raw sensor data; the feature data layer is a map built based on the processing results of the raw sensor data (such as thinning, clustering, feature extraction, tracking, rasterization, etc.).

[0116] Dynamic layers are used to represent the dynamic characteristics of a work area, that is, environmental state characteristics that exist only within a specific time interval. For example, temporarily parked vehicles, obstacles in the road, and other environmental characteristics that exist only within a specific time interval.

[0117] Furthermore, the dynamic layer may contain two layers: a real-time data layer and a historical data layer. The real-time data layer represents the real-time dynamic characteristics of the work area, while the historical data layer represents the accumulation of the dynamic characteristics of the work area over a certain time interval. This data can provide a basis for map management and maintenance.

[0118] It should be noted that some leaf nodes of the map tree may contain task layers in their hierarchical topology, which are used to represent the status information of the task objects contained in the work environment.

[0119] Example 2

[0120] This embodiment provides a method for constructing mining area environmental map data, specifically including:

[0121] S1. Obtain the raw state data and perform preprocessing;

[0122] It should be noted that the raw state data may include navigation and positioning data as well as operational environment perception data.

[0123] S2. Extract and classify feature data from the original state data;

[0124] It should be noted that the feature data may include obstacle pose information, obstacle motion state information, obstacle size information, obstacle shape information, road condition information within the work area, and boundary information of the work area.

[0125] It should be noted that classification refers to determining the motion state of obstacles based on feature data, judging the probability that obstacles may move, and classifying obstacle types.

[0126] S3. Based on the preprocessed original state data and feature data, make judgments and filters to form the original map data.

[0127] S4. Construct local map data based on the original map data;

[0128] It should be noted that constructing local map data refers to constructing dynamic components of local map data in a dynamic layer while saving historical data of the dynamic components of local map data, and constructing static components of local map data in a static layer.

[0129] S5. Calculate the intersection of all constructed local map data based on spatial location coordinates, and segment each local map data according to the leaf nodes in the mining area environment map data representation method of Example 1.

[0130] It should be noted that the intersection of each local map data is obtained, and each local map data is processed into a spatial division of the target area.

[0131] S6. Map the local map data segmentation results to a global map tree to form global map data.

[0132] It should be noted that the mapping relationship between local map data and the global map tree includes: first, the macro-mapping relationship, that is, the mapping relationship between each local map segmentation result and the leaf nodes in the global map tree; second, the micro-mapping relationship based on the macro-mapping relationship, that is, the mapping relationship between each level in the local map segmentation result and each level in the leaf nodes of the map tree.

[0133] In one embodiment, the method further includes, after constructing the global map data, repairing the global map data based on the segmentation results of each local map data as data is continuously collected.

[0134] In one embodiment, the method further includes a step of performing anomaly detection on the repaired global map data, specifically:

[0135] Distinguish between normal and abnormal components in the repaired global map data; for normal components, grant usage rights to the map client; for abnormal components, generate corresponding local map data collection tasks and push them to the map client.

[0136] In one embodiment, the method further includes a step of determining whether the current global map data needs to be updated based on the processing results of the local map data; if so, then...

[0137] Based on the mapping relationship between local map data and global map data, as well as the key features of the local map, incremental map data is extracted from the preprocessed original state data on the basis of the global map data.

[0138] The global map data is updated based on incremental map data.

[0139] It should be noted that the extraction of incremental map data for static maps can be accomplished based on the XOR logic of the data feature probability density function; the extraction of incremental map data for dynamic maps can be accomplished based on the weighted average logic of the data feature probability density function.

[0140] Example 3

[0141] This embodiment provides a map management system. When a vehicle equipped with this map management system is in motion, it can obtain real-time data on the operating environment around the equipment, construct a local map, and push it to the control center. The control center then maintains and publishes the global map data based on the obtained local map data, laying the foundation for digital operation and maintenance and unmanned production in the mining area.

[0142] like Figure 4 As shown, the map management system disclosed in this embodiment mainly consists of two map processors: a map client and a map server.

[0143] The map client is used to collect and construct local map data, and then upload the constructed local map data to the map server, as shown in the process represented by number 1 in the figure.

[0144] The map server is used to construct and update global map data based on the acquired local map data, and to push the right to use the global map data to the map client, as shown in the process represented by number 2 in the figure.

[0145] In some embodiments of the present invention, the map server is also used to send local map acquisition tasks to the map client.

[0146] In one embodiment, the map client can be deployed on vehicles or equipment in the mining area that do not have control center functions.

[0147] In one embodiment, the map server can be deployed in the control center of the mining area and on vehicles equipped with control center functions.

[0148] like Figure 3 As shown, a map processor includes at least a data acquisition module, a data fusion module, a map module, and an interface module.

[0149] The acquisition module is used to read the raw state data and perform preprocessing, as shown in the process represented by number 1 in the figure;

[0150] The fusion module is used to read the preprocessed raw state data from the acquisition module, perform feature extraction and classification, and push the results of feature extraction and classification to the interface module, as shown in the process represented by numbers 2 and 3 in the figure.

[0151] The map module is used to read the preprocessed raw state data from the acquisition module and the feature extraction and classification results from the fusion module, to construct or update the map data, and to push the map data construction or update results to the interface module, as shown in the processes represented by numbers 4, 5 and 6 in the figure.

[0152] The interface module is used to push the received data to other modules for their use, as shown in the process represented by number 7 in the figure.

[0153] like Figure 5 As shown, in one embodiment of the present invention, for the map client: the fusion module can be composed of two parts: a detection submodule and a classification submodule; the map module can be composed of two parts: an analysis submodule and a construction submodule.

[0154] like Figure 6 As shown, in one embodiment of the present invention,

[0155] The map client's data collection module is specifically used to read raw state data and perform preprocessing, which manifests as follows: Figure 5 The data stream transmission process represented by sequence number 1 in the middle;

[0156] The detection submodule is specifically used to read the preprocessed original state data and extract feature data, which manifests as follows: Figure 5 The data stream transmission process represented by serial number 2 in the middle;

[0157] The classification submodule is specifically used to read feature data from the detection submodule and classify the feature data, which is manifested as follows: Figure 5 The data stream transmission process represented by serial number 3 in the middle;

[0158] The classification submodule is also used to send the feature data extraction and classification results to the interface module, which is manifested as follows: Figure 5 The data stream transmission process represented by number 4 in the middle;

[0159] The analysis submodule is specifically used to simultaneously read the preprocessed raw state data and the feature data and classification results from the fusion module, and to judge and filter the read data to form the raw map data; this is manifested as... Figure 5 The data stream transmission process represented by numbers 5 and 6 in the middle;

[0160] The construction submodule is specifically used to read raw map data from the analysis submodule, construct local map data, and locate the vehicle / equipment within the local map; this is manifested as follows: Figure 5 The data stream transmission process represented by serial number 7 in the middle;

[0161] The construction submodule is also used to send the completed local map data to the interface module, which is manifested as follows: Figure 5 The data stream transmission process represented by number 8 in the middle;

[0162] The interface module is specifically used to determine whether the constructed local map data needs to be pushed to the map server. If so, it pushes the local map data to the map server according to certain rules, which is manifested as follows: Figure 5 The data flow transmission process represented by serial number 9; otherwise, the newly constructed local map data is weighted and averaged with historical data to update the local map data.

[0163] In one embodiment, the acquisition module comprises three parts: a data acquisition submodule, a data processing submodule, and a data verification submodule.

[0164] The data acquisition submodule is specifically used to collect raw state data;

[0165] The data processing submodule is specifically used to preprocess the raw state data;

[0166] The data validation submodule is specifically used to determine the validity of the preprocessed original state data.

[0167] In one embodiment, the raw state data read by the acquisition module in the map client may include navigation and positioning data and operational environment perception data.

[0168] In one embodiment, the data acquisition module in the map client may refer to the sensor system installed on the vehicle itself.

[0169] In one embodiment, the feature data extracted by the detection submodule may include obstacle pose information, obstacle motion state information, obstacle size information, obstacle shape information, road condition information within the work area, and boundary information of the work area.

[0170] In one embodiment, the classification submodule is specifically used to determine the obstacle's motion state, the probability of the obstacle moving, and the type of obstacle based on the obstacle features from the detection submodule, thereby obtaining classification information for the feature data. The obstacle type classification is related to the specific scenario. For a mining scenario, there may be the following obstacle types: vehicles, people, falling rocks, cable bridges, curbs (retaining walls), and piles of earth.

[0171] In one embodiment, the analysis submodule is specifically used to: reconstruct the mapping relationship between the raw state data from the acquisition module and the feature data from the fusion module, that is, to determine the raw state data corresponding to the obstacle features, so that the two correspond one-to-one; based on the feature data from the fusion module, remove components that do not belong to the local map data (such as dynamic obstacles or noise) and their corresponding raw state data; based on the classification results of the feature data from the fusion module, distinguish between dynamic components (such as obstacles with the possibility of movement) and static components belonging to the local map data and their corresponding raw state data. The remaining data components after removing the feature data that do not belong to the local map data and their corresponding raw state data form the raw map data.

[0172] In one embodiment, the construction submodule is specifically used to construct dynamic components of local map data in a dynamic layer while saving historical data of the dynamic components of local map data, and to construct static components of local map data in a static layer.

[0173] In one embodiment, the interface module is specifically used for,

[0174] If a local map construction task is received, a request is sent to the map server before submitting the local map data construction results. If so, the local map data is pushed.

[0175] It should be noted that the map client initialization includes the following two methods:

[0176] In one embodiment, the initialization of the map client is completed along with the startup of the hardware system, and then the constructed local map data can be uploaded to the map server in real time or in time periods.

[0177] In another embodiment, the initialization of the map client can occur after receiving a local map collection task from the map server. After initialization, the map client completes the collection of the local map based on the task data and uploads the collection results to the map server.

[0178] like Figure 7 As shown, in one embodiment of the present invention, for the map server: the fusion module can be composed of two parts: an analysis submodule and a synthesis submodule; the map module can be composed of two parts: a processing submodule and an update submodule; the interface module can contain a map decision submodule.

[0179] like Figure 8 As shown, in one embodiment of the present invention,

[0180] The map server's data collection module is specifically used to read raw state data and perform preprocessing, which manifests as follows: Figure 7 The data stream transmission process represented by sequence number 1 in the middle;

[0181] The analysis submodule is specifically used to read the preprocessed raw state data, segment the local map data, and extract key feature data; this manifests as follows: Figure 7 The data stream transmission process represented by serial number 2 in the middle;

[0182] The integration submodule is specifically used to read the segmentation results of various local map data from the analysis submodule and repair the mapping relationship between the segmentation results of the local map data and the global map tree; this manifests as... Figure 7 The data stream transmission process represented by serial number 3 in the middle;

[0183] The integration submodule is also used to send the data processing results to the interface module; this is manifested as... Figure 7 The data stream transmission process represented by number 4 in the middle;

[0184] The processing submodule is specifically used to simultaneously read the preprocessed raw state data and the processing results of the fusion module on the local map data; this manifests as... Figure 7 The data flow transmission process represented by numbers 5 and 6 in the middle; and based on the processing results of local map data, extracting incremental map data from the preprocessed original state data on the basis of the existing global map data, according to the global map update command issued by the update submodule;

[0185] The update submodule is specifically used to determine whether the existing global map data needs to be updated based on the processing results of the fusion module on the local map data. If so, it sends a global map update command to the processing submodule; this is manifested as... Figure 7The data flow transmission process represented by number 7 in the middle; determining whether an update is needed means comparing the newly obtained local map data with its corresponding map leaf node data. If the data change is large, it is "yes"; otherwise, it is "no".

[0186] The update submodule is also used to read incremental map data from the processing submodule, perform a weighted summation of the incremental map data and the existing data in the leaf nodes of the map tree, and update the global map data; this is manifested as... Figure 7 The data stream transmission process represented by number 8 in the middle;

[0187] The update submodule is also used to send the updated global map data to the interface module; this is manifested as... Figure 7 The data stream transmission process represented by number 9 in the middle;

[0188] The map decision submodule is specifically used to determine the anomalies in the current global map data, distinguishing between normal and abnormal components. Normal components refer to components in the updated map data whose attributes are completely determined, while abnormal components need to be re-collected.

[0189] The interface module is specifically used to generate corresponding local map data collection tasks for abnormal components in the global map data and push them to the map client; this manifests as... Figure 7 The data flow transmission process represented by serial number 11, and the right to use normal components in the publicly available global map data; manifested as Figure 7 The data stream transmission process represented by serial number 10.

[0190] In one embodiment, the raw state data processed by the map server's acquisition module may include local map data from various map clients. The map server's acquisition module performs filtering preprocessing on the raw state data.

[0191] In one embodiment, the data acquisition module of the map server can refer to the communication system installed in the control center and the vehicle equipment installed with map clients.

[0192] In one embodiment, the analysis submodule is used to calculate the intersection of each local map data based on the spatial location coordinates, and to complete the segmentation of each local map data and extract its key features.

[0193] In one embodiment, the synthesis submodule is used to repair the mapping relationship between the local map segmentation results from the analysis submodule and the global map tree, wherein...

[0194] The mapping relationship includes two aspects: first, the macro-mapping relationship, which is the mapping relationship between each local map segmentation result and the leaf nodes in the global map tree; second, the micro-mapping relationship based on the macro-mapping relationship, which is the mapping relationship between each level in the local map segmentation result and each level in the leaf nodes of the map tree.

[0195] In one embodiment, the processing submodule is used to extract incremental map data based on the global map data and using the raw state data from the acquisition module, according to the mapping relationship between the local map data and the global map data from the fusion module and the key features of the local map.

[0196] In one embodiment, the extraction of incremental map data for a static graph can be accomplished based on the XOR logic of the data feature probability density function.

[0197] In one embodiment, the extraction of incremental map data for dynamic graphs can be accomplished based on the weighted average logic of the data feature probability density function.

[0198] It should be noted that the initialization of the map server is completed along with the startup of the hardware system.

[0199] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for representing environmental map data in a mining area, characterized in that, include: The global map data of the entire mining area's operating environment is represented as a tree-like topology; The global map data of the entire mining area operation environment represents the status information of the entire mining area operation environment, as well as the status information generated by the interaction between vehicles and equipment and the mining area operation environment. The tree-like topology includes: a main node, subgraph nodes, and leaf nodes; The main node is the root node; the sub-graph nodes are child nodes of the main node; the leaf nodes are used to store local map data of the mining area's operating environment; the main node and sub-graph nodes are used to store the regional attributes to which the mining area belongs. The leaf node includes at least a zone / sub-zone node; the zone / sub-zone node is a child node of the sub-graph node; the zone refers to any complete working area in the entire mining area's working environment; Any leaf node represents a hierarchical topology, including static and dynamic layers; The static layer is used to store the static features of the work area corresponding to the leaf node; the static features refer to environmental state features that do not change within a set time interval. The dynamic layer is used to store the dynamic features of the work area corresponding to the leaf node; the dynamic features refer to the environmental state features that exist only within a specified time interval.

2. The method for representing mining area environmental map data according to claim 1, characterized in that, The leaf node also includes a segment node; the segment node is a child node of the region / subregion node; The term "segment" refers to any uncontrolled road segment within a complete work area.

3. The method for representing mining area environmental map data according to claim 2, characterized in that, The leaf node further includes a segment node; the segment node is a child node of the segment node; The term "sub-segment" refers to a sub-segment of a non-controlled road segment according to traffic rules.

4. The method for representing mining area environmental map data according to claim 1, characterized in that, The static layer includes a raw data layer and a feature data layer; The raw data layer includes static features representing the environmental state of the work area in the form of raw sensor data; The feature data layer includes static features representing the environmental state of the work area by processing the raw sensor data; the processing includes thinning, clustering, feature extraction, tracking, and rasterization.

5. The method for representing mining area environmental map data according to claim 1, characterized in that, The dynamic layer includes a real-time data layer and a historical data layer; The real-time data layer includes the real-time dynamic characteristics of the work area; The historical data layer includes the accumulation of dynamic characteristics of the work area over a set time interval.

6. The method for representing mining area environmental map data according to claim 5, characterized in that, The dynamic layer also includes a task layer; The task layer includes status information representing the task objects contained in the task area.

7. A method for constructing environmental map data in a mining area, characterized in that, include: Acquire raw state data and perform preprocessing; The original state data includes navigation and positioning data and environmental perception data of the corresponding work area in the mining operation environment. Feature data extraction and classification are performed on the preprocessed original state data; Based on the preprocessed original state data, extracted feature data, and classification results, judgments and filters are performed to form the original map data; Based on the original map data, local map data is constructed; The intersection of all constructed local map data is calculated based on spatial location coordinates, and each local map data is segmented according to the leaf nodes in the mining area environmental map data representation method described in any one of claims 1 to 6. Based on the local map data segmentation results mapped to global map data, global map data represented by the mining area environment map data representation method according to any one of claims 1 to 6 is formed; The mapping relationship between local map data and global map data includes: the mapping relationship between the segmentation results of each local map data and the leaf nodes in the global map data; In addition, the mapping relationship between each level in the local map data segmentation results and each level in the leaf nodes of the global map data.

8. The method for constructing mining area environmental map data according to claim 7, characterized in that, Feature data extraction and classification are performed on the preprocessed original state data. The feature data includes obstacle pose information, obstacle motion state information, obstacle size information, obstacle shape information, road condition information within the work area, and boundary information of the work area; The classification refers to the process of determining the motion state of obstacles, assessing the probability of obstacle movement, and classifying obstacle types based on feature data.

9. A method for constructing mining area environmental map data according to claim 7, characterized in that, The process of judging and filtering based on the preprocessed original state data, extracted feature data, and classification results forms the original map data, including: Determine the mapping relationship between the preprocessed original state data and the feature data; Remove components that do not belong to the local map data and their corresponding original state data.

10. A method for constructing mining area environmental map data according to claim 7, characterized in that, The construction of local map data based on the original map data includes: In the dynamic layer, dynamic components of local map data are constructed, and historical data of the dynamic components of local map data are saved. In the static layer, static components of local map data are constructed.

11. A method for constructing mining area environmental map data according to claim 7, characterized in that, Also includes: After constructing the global map data, as data is continuously collected, the steps to repair the global map data based on the segmentation results of each local map data are as follows; The repair of global map data refers to repairing the mapping relationship between local map data and global map data based on the segmentation results of each local map data.

12. The method for constructing mining area environmental map data according to claim 7, characterized in that, Also includes: The steps for anomaly detection in global map data. Distinguish between normal and abnormal components in global map data; For normal components, the right to use the normal components is granted; for abnormal components, a corresponding local map data collection task is generated and published; the normal components refer to components with completely defined attributes in the updated global map data, and the rest are abnormal components.

13. The method for constructing mining area environmental map data according to claim 7, characterized in that, Also includes: Based on the segmentation results of the local map data, determine whether the current global map data needs to be updated. If so, then... Based on the mapping relationship between local map data and global map data, as well as the key features of the local map, incremental map data is extracted from the preprocessed original state data on the basis of the global map data. The global map data is updated based on incremental map data.

14. A method for constructing mining area environmental map data according to claim 13, characterized in that, Incremental map data of static graphs are extracted based on the XOR logic of the probability density function of data features. The incremental map data of the dynamic graph is extracted based on the weighted average logic of the probability density function of data features.

15. A map management system, characterized in that, include: Map client and map server; The map client is used to collect and construct local map data using the mining area environmental map data construction method according to any one of claims 8 to 14, and upload the constructed local map data to the map server; the map client is deployed on a vehicle, and the vehicle travels in the working area of ​​the area, segment, or sub-segment that the map client needs to collect data from; The map server is used to construct global map data based on the acquired local map data using the mining area environment map data construction method described in any one of claims 8 to 14, and to push the right to use the global map data to the map client.

16. A map management system according to claim 15, characterized in that, The map client includes: The data acquisition module is used to read and preprocess raw state data in the mining area's working environment; the raw state data includes navigation and positioning data and working environment perception data. The detection submodule is used to read the preprocessed raw state data and extract feature data; the feature data includes: obstacle pose information, obstacle motion state information, obstacle size information, obstacle shape information, road condition information within the work area, and boundary information of the work area; The classification submodule is used to read the feature data extracted by the detection submodule, and to perform obstacle motion state discrimination, obstacle movement probability determination, and obstacle type classification. The analysis submodule is used to simultaneously read the preprocessed raw state data, feature data, and the classification results of the feature data, and to judge and filter the read data to form the raw map data. A construction submodule is used to read the original map data from the analysis submodule, construct the local map data, and locate the vehicle or equipment in the local map. The interface module is used to determine whether the completed local map data needs to be pushed to the map server. If so, the local map data is pushed to the map server according to the set rules; otherwise, the newly constructed local map data is updated by weighted averaging with historical data.

17. A map management system according to claim 16, characterized in that, The data acquisition module in the map client is a sensor system installed on the vehicle itself.

18. A map management system according to claim 16, characterized in that, The analysis submodule is specifically used for, Based on the feature data, remove noise and the corresponding original state data; Based on the classification results of the feature data, the mapping relationship between the original state data and the feature data is determined, thus forming the original map data.

19. A map management system according to claim 15, characterized in that, The map server includes: The data acquisition module is used to read raw state data and preprocess it; the raw state data includes local map data from various map clients. The analysis submodule is used to read the preprocessed raw state data, segment the local map data, and extract key feature data. The integration submodule is used to read the segmentation results of each local map data in the analysis submodule and repair the mapping relationship between the segmentation results of the local map data and the global map data; The interface module is used to generate corresponding local map data collection tasks for abnormal components in the global map data and push them to the map client; as well as to push the usage rights of normal components in the global map data to the map client.

20. A map management system according to claim 19, characterized in that, The data acquisition module of the map server is a communication system between the map server and each vehicle device equipped with a map client.

21. A map management system according to claim 19, characterized in that, The analysis submodule is specifically used for, The intersection of each local map data is calculated based on the spatial coordinates. Each local map data is then segmented and, according to the tree-like topology of the global map data, it is segmented into leaf nodes in the global map data.

22. A map management system according to claim 19, characterized in that, The map server also includes: The update submodule is used to determine whether the current global map data needs to be updated based on the processing results of the local map data. If so, it sends a global map update command to the processing submodule; and reads the incremental map data from the processing submodule and updates the global map data accordingly. The processing submodule is used to simultaneously read the processing results of the preprocessed original state data and the local map data. Based on the global map update command issued by the update submodule, and the processing results of the local map data, it extracts incremental map data from the preprocessed original state data on the basis of the existing global map data.

23. A map management system according to claim 19, characterized in that, The interface module of the map server also includes: The map decision submodule is used to determine the anomalies of the current global map data and distinguish between normal and abnormal components. Normal components refer to components in the updated global map data whose attributes are completely determined, while abnormal components are the opposite.

24. A map management system according to any one of claims 15 to 23, characterized in that, The map client is deployed on vehicles or equipment in the mining area that do not have a control center. The map server is deployed in the control center of the mining area or on a vehicle equipped with a control center.

Images