An information processing method and system based on open-pit mine 5G communication navigation

Abstract

The application provides an information processing method and system based on open-pit mine 5G communication navigation, and relates to the technical field of artificial intelligence.In the application, a first data angle mine site position data vector and a second data angle mine site position data vector of target mine site position data are determined, and a first data angle mine site position data vector and a second data angle mine site position data vector of each candidate mine site position data in an original mine site position data cluster are determined; based on the first data angle mine site position data vector, a preset number of candidate mine site position data are determined in the original mine site position data cluster, and a corresponding related mine site position data cluster is formed by combination; based on the second data angle mine site position data vector, related mine site position data are determined in the related mine site position data cluster.Based on the above, the accuracy of communication navigation in the open-pit mine can be improved to a certain extent.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and more specifically, to an information processing method and system based on 5G communication navigation in open-pit mines. Background Technology

[0002] 5G communication navigation in open-pit mines is a method that utilizes 5G communication technology to provide navigation and location services for mine workers. It combines high-speed data transmission capabilities with precise positioning technology, effectively guiding the work and movement of mine workers in the complex environment of open-pit mines. Specifically, 5G communication navigation in open-pit mines includes the following main components:

[0003] 5G Communication Network: Utilizing 5G networks to provide high bandwidth and low latency communication support, ensuring real-time transmission and reception of navigation-related data; Positioning Technology: Using precise positioning technologies, such as Global Navigation Satellite System (GPS), Inertial Navigation System (INS), or indoor positioning systems based on Bluetooth, Wi-Fi, etc., to obtain accurate location information of mine workers; Map and Route Planning: Constructing a mine map based on the mine's geographical information and the location information of mine workers, and using route planning algorithms to determine the optimal navigation route to help mine workers reach their target location.

[0004] In other words, during the communication and navigation process, it is necessary to search for and determine the target location. However, in the existing technology, due to the complexity of the area in the open-pit mine, the search based on the area name has the problem of relatively low reliability in determining the target location. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide an information processing method and system based on 5G communication and navigation in open-pit mines, so as to improve the accuracy of communication and navigation in open-pit mines to a certain extent, that is, to solve the problem of how to improve the accuracy of communication and navigation in open-pit mines.

[0006] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions:

[0007] An information processing method based on 5G communication and navigation in open-pit mines includes:

[0008] The target mine location data is determined by first data angle and second data angle, and the target mine location data is also determined by first data angle and second data angle for each candidate mine location data in the original mine location data cluster. The target mine location data is text data obtained by parsing the target navigation request and is used to reflect the attribute characteristics of the navigation target corresponding to the target navigation request.

[0009] The target mine location data vector of the first data angle is compared and analyzed with the mine location data vector of the first data angle of each candidate mine location data in the original mine location data cluster, and the corresponding first correlation parameter is output. Based on the first correlation parameter, a preset number of candidate mine location data are determined in the original mine location data cluster and combined to form a corresponding related mine location data cluster.

[0010] The target mine location data vector at the second data angle is compared and analyzed with the mine location data vector at the second data angle of each candidate mine location data in the relevant mine location data cluster. The corresponding second correlation parameter is output. Based on the second correlation parameter, the relevant mine location data of the target mine location data is determined in the relevant mine location data cluster. The relevant mine location data is used as the location of the navigation target corresponding to the target navigation request.

[0011] In some preferred embodiments, in the above-described information processing method based on 5G communication navigation in open-pit mines, the step of determining the mine location data vector of the first data angle and the mine location data vector of the second data angle of the target mine location data, and determining the mine location data vector of the first data angle and the mine location data vector of the second data angle of each candidate mine location data in the original mine location data cluster, includes:

[0012] The target mine location data is noise processed to form a first noise data and a second noise data of the mine location corresponding to the target mine location data. At least one of the first noise data and the second noise data of the mine location has noise data, and at most one of the first noise data and the second noise data of the mine location does not have noise data.

[0013] The first noise data and the second noise data of the target mine location corresponding to the target mine location data are mined respectively to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the target mine location data.

[0014] Determine the mine location data vectors for the first data angle and the second data angle for each candidate mine location data in the original mine location data cluster.

[0015] In some preferred embodiments, in the above-described information processing method based on 5G communication navigation in open-pit mines, the step of mining the first noise data and the second noise data of the target mine location corresponding to the target mine location data to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the target mine location data includes:

[0016] The first noise data and the second noise data of the target mine location data are respectively subjected to shallow feature mining, and the first shallow vector and the second shallow vector of the target mine location data are output.

[0017] The first shallow vector and the second shallow vector of the target mine location data are transformed by vector dimension, and the first transformed vector and the second transformed vector of the target mine location data are output.

[0018] The first transformation vector and the second transformation vector of the mine location are respectively subjected to feature filtering, and the mine location data vectors of the first data angle and the second data angle corresponding to the target mine location data are output.

[0019] In some preferred embodiments, in the above-mentioned information processing method based on 5G communication navigation in open-pit mines, the step of mining the first noise data and the second noise data of the mine location corresponding to the target mine location data is performed using a feature mining network, which includes a shallow feature mining unit, a vector dimension transformation unit, and a first feature filtering unit.

[0020] The step of mining the first noise data and the second noise data of the target mine location corresponding to the target mine location data to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the target mine location data includes:

[0021] Using the shallow feature mining unit, shallow feature mining is performed on the first noise data of the mine location and the second noise data of the mine location respectively to form the first shallow vector of the mine location and the second shallow vector of the mine location of the target mine location data.

[0022] Using the vector dimension transformation unit, the first shallow vector of the mine location and the second shallow vector of the mine location are transformed by vector dimension to form the first transformation vector and the second transformation vector of the mine location corresponding to the target mine location data.

[0023] Using the first feature filtering unit, the first transformation vector of the mine location is filtered to form a mine location data vector of the first data angle corresponding to the target mine location data;

[0024] The second transformation vector of the mine location is compressed in terms of vector dimension to form a mine location data vector with a second data angle corresponding to the target mine location data.

[0025] In some preferred embodiments, in the above-described information processing method based on 5G communication and navigation in open-pit mines, the feature mining network further includes a second feature filtering unit, and the information processing method based on 5G communication and navigation in open-pit mines further includes:

[0026] The candidate feature mining network and training mine location data were determined.

[0027] Based on the training mine location data, the shallow feature mining units in the candidate feature mining network are updated to form the first feature mining network corresponding to the candidate feature mining network.

[0028] Maintain the network parameters of the shallow feature mining unit in the first feature mining network, and update the second feature filtering unit in the first feature mining network to form a second feature mining network corresponding to the first feature mining network.

[0029] Maintain the network parameters of the shallow feature mining unit and the network parameters of the second feature filtering unit in the second feature mining network, and update the first feature filtering unit in the second feature mining network to form a third feature mining network corresponding to the second feature mining network. Then, mark the third feature mining network to form an updated feature mining network.

[0030] In some preferred embodiments, in the above-described information processing method based on 5G communication and navigation in open-pit mines, the step of updating the shallow feature mining units in the candidate feature mining network according to the training mine location data to form a first feature mining network corresponding to the candidate feature mining network includes:

[0031] The training mine location data is noise-processed to form first noise data and second noise data of the mine location corresponding to the training mine location data.

[0032] The first noise data of the mine location corresponding to the training mine location data is used as anchor training data, the second noise data group of the mine location corresponding to the training mine location data is used as anchor positive training data, and other training mine location data is used as anchor negative training data.

[0033] Using the shallow feature mining unit in the candidate feature mining network, the anchor training data, the anchor positive training data, and the anchor negative training data are respectively subjected to shallow feature mining to form corresponding anchor training data vectors, anchor positive training data vectors, and anchor negative training data vectors;

[0034] Calculate a first matching parameter between the anchor training data vector and the anchor positive training data vector, and calculate a second matching parameter between the anchor training data vector and the anchor negative training data vector; and calculate a first type of error parameter corresponding to the candidate feature mining network based on the first matching parameter and the second matching parameter.

[0035] Based on the first type of error parameters, the network parameters of the shallow feature mining units in the candidate feature mining network are optimized and adjusted to form the first feature mining network corresponding to the candidate feature mining network.

[0036] In some preferred embodiments, in the above-described information processing method based on 5G communication and navigation in open-pit mines, the step of maintaining the network parameters of the shallow feature mining units in the first feature mining network and updating the second feature filtering units in the first feature mining network to form a second feature mining network corresponding to the first feature mining network includes:

[0037] Using the second feature filtering unit in the first feature mining network, the anchor training data vector, the anchor positive training data vector, and the anchor negative training data vector are respectively subjected to feature filtering to form corresponding anchor training data filtering vector, anchor positive training data filtering vector, and anchor negative training data filtering vector;

[0038] Calculate the third matching parameter between the anchor training data selection vector and the anchor positive training data selection vector, and calculate the fourth matching parameter between the anchor training data selection vector and the anchor negative training data selection vector. Based on the third matching parameter and the fourth matching parameter, calculate the second type of error parameter corresponding to the first feature mining network.

[0039] Maintain the network parameters of the shallow feature mining unit in the first feature mining network, and optimize and adjust the network parameters of the second feature filtering unit in the first feature mining network according to the second type of error parameters to form a second feature mining network corresponding to the first feature mining network.

[0040] In some preferred embodiments, in the above-described information processing method based on 5G communication and navigation in open-pit mines, the steps of maintaining the network parameters of the shallow feature mining units and the network parameters of the second feature filtering units in the second feature mining network, updating the first feature filtering units in the second feature mining network to form a third feature mining network corresponding to the second feature mining network, and marking the third feature mining network to form an updated feature mining network, include:

[0041] A training selection vector cluster is extracted. The training selection vector cluster includes training initial mine data vectors corresponding to multiple training mine location data and training vector supervision data corresponding to each training initial mine data vector. The training vector supervision data is used to characterize the selection vectors for feature selection of the training initial mine data vectors using the first feature selection unit.

[0042] Using the first feature filtering unit in the second feature mining network, feature filtering is performed on each of the initial training mining data vectors, and the estimated filtering vector corresponding to each of the initial training mining data vectors is output.

[0043] Calculate the vector matching parameters between each of the estimated screening vectors and the corresponding training vector supervision data, and perform fusion processing on each of the vector matching parameters to output the third type of error parameters corresponding to the feature mining network;

[0044] Maintain the network parameters of the shallow feature mining unit and the network parameters of the second feature filtering unit in the second feature mining network, and optimize and adjust the network parameters of the first feature filtering unit in the second feature mining network based on the third type of error parameter to form a third feature mining network corresponding to the second feature mining network.

[0045] The third feature mining network is labeled to form an updated feature mining network.

[0046] In some preferred embodiments, in the above-described information processing method based on 5G communication navigation in open-pit mines, the step of determining the mine location data vector of the first data angle and the mine location data vector of the second data angle for each candidate mine location data in the original mine location data cluster includes:

[0047] For each candidate mine location data in the original mine location data cluster:

[0048] The candidate mine location data is noise-processed to form the third noise data of the mine location corresponding to the candidate mine location data.

[0049] The third noise data of the mine location corresponding to the candidate mine location data is subjected to shallow feature mining to form the third shallow vector of the mine location corresponding to the candidate mine location data.

[0050] The third shallow vector of the mine location corresponding to the candidate mine location data is transformed by vector dimension to form the third transformed vector of the mine location corresponding to the candidate mine location data.

[0051] The third transformation vector of the mine location is subjected to two different feature filtering processes to form a mine location data vector with a first data angle and a mine location data vector with a second data angle corresponding to the candidate mine location data.

[0052] This invention also provides an information processing system based on 5G communication and navigation in open-pit mines, including a processor and a memory. The memory is used to store computer programs, and the processor is used to execute the computer programs to implement the above-described information processing method based on 5G communication and navigation in open-pit mines.

[0053] The information processing method and system based on 5G communication navigation in open-pit mines provided in this invention can first determine the mine location data vectors of the first and second data angles of the target mine location data, and determine the mine location data vectors of the first and second data angles of each candidate mine location data in the original mine location data cluster; based on the mine location data vectors of the first data angles, a preset number of candidate mine location data are determined in the original mine location data cluster, and combined to form corresponding related mine location data clusters; based on the mine location data vectors of the second data angles, related mine location data are determined in the related mine location data clusters. Based on the foregoing, since the navigation target is determined by the attribute characteristics of the target navigation request, the basis for determining the navigation target can be enriched to a certain extent. In addition, a two-level comparative analysis is performed based on the attribute characteristics of the navigation target (target mine location data) from the first data angle and the second data angle of the mine location data vector. This allows the location of the navigation target to be narrowed down and matched sequentially, thereby improving the problem of relatively low reliability in determining the target location in the existing technology. In this way, the accuracy of communication and navigation in open-pit mines can be improved to a certain extent.

[0054] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0055] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0056] Figure 1 A structural block diagram of an information processing system based on 5G communication and navigation in an open-pit mine, provided in an embodiment of the present invention;

[0057] Figure 2 A flowchart illustrating the steps of the information processing method based on 5G communication and navigation in an open-pit mine provided in this embodiment of the invention.

[0058] Figure 3 This is a schematic diagram of the modules included in the information processing device based on 5G communication and navigation in an open-pit mine provided in an embodiment of the present invention. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0060] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0061] like Figure 1 As shown, this embodiment of the invention provides an information processing system based on 5G communication and navigation in open-pit mines. The aforementioned information processing system based on 5G communication and navigation in open-pit mines may include a memory and a processor.

[0062] In detail, in one possible implementation, the memory and processor are electrically connected directly or indirectly to enable data transmission or interaction. For example, they can be electrically connected via one or more communication buses or signal lines. The memory may store at least one software functional module (computer program) that exists in the form of software or firmware. The processor can be used to execute the executable computer program stored in the memory, thereby implementing the information processing method based on 5G communication navigation in open-pit mines provided in this embodiment of the invention.

[0063] In detail, in one possible implementation, the aforementioned memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The aforementioned processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), a System on Chip (SoC), etc.; it can also be a Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0064] In detail, in one possible implementation, the aforementioned information processing system based on 5G communication and navigation in open-pit mines can be a server with data processing capabilities.

[0065] Combination Figure 2 This invention also provides an information processing method based on 5G communication and navigation in open-pit mines, which can be applied to the aforementioned information processing system based on 5G communication and navigation in open-pit mines. The method steps defined in the relevant process of the aforementioned information processing method based on 5G communication and navigation in open-pit mines can be implemented by the aforementioned information processing system based on 5G communication and navigation in open-pit mines.

[0066] The following will be about Figure 2 The specific process shown will be explained in detail.

[0067] Step S110: Determine the mine location data vector of the first data angle and the mine location data of the target mine location data, and determine the mine location data vector of the first data angle and the mine location data vector of the second data angle for each candidate mine location data in the original mine location data cluster.

[0068] In this embodiment of the invention, the aforementioned information processing system based on 5G communication navigation in open-pit mines can determine the mine location data vectors for a first data angle and a second data angle of the target mine location data, and determine the mine location data vectors for a first data angle and a second data angle of each candidate mine location data in the original mine location data cluster. The target mine location data is text data obtained by parsing the target navigation request, used to reflect the attribute characteristics of the navigation target corresponding to the target navigation request, such as describing what kind of open-pit mine area the navigation target belongs to. Furthermore, the mine location data vector can be the vectorized result of the target mine location data and the candidate mine location data, i.e., obtained through vectorization processing, facilitating subsequent comparative analysis. The first data angle and the second data angle can refer to reflecting the target mine location data and the candidate mine location data from two different perspectives. Additionally, the candidate mine location data in the original mine location data cluster includes the mine location data of each open-pit mine area in the target open-pit mine; this mine location data is also text data, used to reflect the attribute characteristics of the open-pit mine area. The aforementioned target navigation request can be generated by the corresponding operation of the target terminal device and transmitted to the backend based on 5G communication technology, such as the information processing system based on 5G communication navigation in open-pit mines or the associated server of the aforementioned information processing system based on 5G communication navigation in open-pit mines. 5G networks possess high reliability and stability, capable of coping with harsh environmental conditions and complex terrain. In the challenging working environment of a mine, 5G networks can provide continuous and stable communication services, ensuring the normal operation of navigation work.

[0069] Step S120: Compare and analyze the mine location data vector of the first data angle of the target mine location data with the mine location data vector of the first data angle of each candidate mine location data in the original mine location data cluster, output the corresponding first correlation parameter, and, based on the first correlation parameter, determine a preset number of candidate mine location data in the original mine location data cluster, and combine them to form the corresponding related mine location data cluster.

[0070] In this embodiment of the invention, the information processing system based on 5G communication and navigation in open-pit mines can compare and analyze the mine location data vector of the first data angle of the target mine location data with the mine location data vector of the first data angle of each candidate mine location data in the original mine location data cluster, outputting a corresponding first correlation parameter. Based on the first correlation parameter, a preset number of candidate mine location data are determined in the original mine location data cluster and combined to form a corresponding related mine location data cluster. The specific value of the preset number is not limited and can be configured according to actual needs. Furthermore, the preset number of candidate mine location data with the largest first correlation parameter can be determined and combined to form a related mine location data cluster, thus achieving the first-level candidate mine location data filtering.

[0071] Step S130: Compare and analyze the mine location data vector of the second data angle of the target mine location data with the mine location data vector of the second data angle of each candidate mine location data in the relevant mine location data cluster, output the corresponding second correlation parameter, and determine the relevant mine location data of the target mine location data in the relevant mine location data cluster based on the second correlation parameter.

[0072] In this embodiment of the invention, the information processing system based on 5G communication navigation in open-pit mines can compare and analyze the mine location data vector of the second data angle of the target mine location data with the mine location data vector of the second data angle of each candidate mine location data in the relevant mine location data cluster, outputting a corresponding second correlation parameter. Based on the second correlation parameter, the relevant mine location data of the target mine location data is determined in the relevant mine location data cluster. This relevant mine location data serves as the location of the navigation target corresponding to the target navigation request. Thus, after determining the relevant mine location data, the corresponding open-pit mine area can be determined based on the relevant mine location data. Then, the location coordinates corresponding to the open-pit mine area can be retrieved from the corresponding database, enabling 5G communication navigation based on these location coordinates. Furthermore, for example, the candidate mine location data with the maximum value of the second correlation parameter in the relevant mine location data cluster can be used as the relevant mine location data of the target mine location data. Moreover, the first and second correlation parameters can refer to the cosine similarity between vectors or other parameters representing similarity.

[0073] Because it is based on the attribute characteristics of the navigation target corresponding to the target navigation request, the basis for determining the navigation target can be richer to a certain extent. In addition, a two-level comparative analysis is performed based on the attribute characteristics of the navigation target (target mine location data) from the first data angle and the second data angle of the mine location data vector. This allows the location of the navigation target to be narrowed down and matched sequentially, thereby improving the problem of relatively low reliability in determining the target location in the existing technology. In this way, the accuracy of communication and navigation in open-pit mines can be improved to a certain extent.

[0074] For example, an open-pit mine may include the following open-pit mining areas (in other embodiments, it may also have other divisions, such as more levels of division, etc.):

[0075] (I) Excavation Area:

[0076] Mineralization Pilot Area: Exploration Drilling Points: Geological exploration drilling is conducted in the mineralization pilot area to obtain information on the depth and characteristics of the ore body; Geological Survey Area: Detailed geological surveys, sampling, and testing are carried out in the mineralization pilot area to determine the location and quality of the ore body;

[0077] Mining Zone: Initial Development Zone: The area where ore mining first begins; Production Development Zone: The area where ore mining is currently underway;

[0078] Temporary storage areas: Soil storage area: Soil excavated from the mining area is stored in a designated area; Rock storage area: Rocks removed from the mining area are temporarily stored in a safe location.

[0079] (II) Storage Yard Area:

[0080] Rough ore stockpile: Different ore type stockpile areas: ore is classified and stockpiled according to ore type; Grade zoning: the same type of ore is stockpiled in zones according to grade (content) level;

[0081] Tailings dump; Tailings filling area: A designated area where tailings are used to fill tailings to reduce environmental impact; Tailings storage area: An area for long-term tailings storage and management.

[0082] (III) Ore Body Slope:

[0083] Uphill / Downhill: High slope area: Ore body slope area with a large slope and height; Low slope area: Ore body slope area with a small slope and height;

[0084] Slope monitoring points: Inclinometer measuring points: Inclinometers are installed to monitor slope tilt changes in real time; GPS monitoring points: Global Positioning System (GPS) equipment is used to monitor slope displacement;

[0085] (iv) Water treatment area:

[0086] Rainwater harvesting ponds: rainwater retention areas: areas that collect and temporarily store rainwater; rainwater diversion channels: channels that guide the collected rainwater to appropriate treatment facilities;

[0087] Clarification tank; Sedimentation tank: A large pond or tank used to settle suspended particulate matter; Filtration zone: An area where further clarification and treatment are carried out through filter media;

[0088] (V) Equipment Maintenance Area:

[0089] Repair workshop: Mechanical repair area: a dedicated area for the repair and maintenance of mechanical equipment; Electrical repair area: a dedicated area for handling and repairing electrical equipment malfunctions;

[0090] Parts Warehouse: Spare Parts Storage Area: A specific area for storing spare parts and components, categorized by type and purpose; Parts Distribution Area: An area used to distribute spare parts and components to the maintenance team;

[0091] Lubrication station: Lubricating oil storage tank area: an area for storing various lubricating oils and greases; Gas station: providing supplies for equipment such as excavators and trucks.

[0092] Continuing with the example above, candidate mine location data could be "an area for storing various lubricating oils and greases," "a specific area for storing spare parts and components," "an area for collecting and temporarily storing rainwater," "an area for classifying and stockpiling ore according to type," or "an area for conducting geological exploration drilling to obtain ore body depth and characteristic information," etc. Correspondingly, the target mine location data could be "an area for storing lubricating oil," "an area for classification and stockpiling," etc. Thus, for the target mine location data corresponding to "an area for storing lubricating oil," a comparative analysis can reveal the relevant mine location data "an area for storing various lubricating oils and greases," meaning the navigation target is a lubricating oil storage tank area. For the target mine location data corresponding to "an area for classification and stockpiling," a comparative analysis can reveal the relevant mine location data "an area for classification and stockpiling according to ore type" (wherein, the candidate mine location data included in the analyzed relevant mine location data cluster can be "classified and stockpiled according to ore type" and "storing the same type of ore in zones according to grade (content) level"), meaning the navigation target is a stockpile area for different ore types. It is understood that the above examples are relatively simple. In practical applications, an open-pit mine includes more complex open-pit mining areas, and there may be more similarities in the attribute characteristics between open-pit mining areas. Therefore, the application of the two-level comparative analysis provided by the embodiments of the present invention is more demanding.

[0093] In detail, in one possible implementation, step S110 described above may further include the specific implementation details described below:

[0094] The aforementioned target mine location data is subjected to noise processing to generate first noise data and second noise data corresponding to the target mine location data. At least one of the first noise data and the second noise data contains noisy data, and at most one of the first noise data and the second noise data does not contain noisy data. That is, if only one noise processing is performed (i.e., perturbation is applied), the aforementioned target mine location data can be used as the first noise data, and the noise-processed target mine location data can be used as the second noise data; if two noise processing operations are performed... The noise processing can use data with less noise as the target mine location data as the first noise data, and data with more noise as the target mine location data as the second noise data. That is, the noise application amount is different in the two noise processing steps. In addition, the configuration of the first and second noise data can improve the processing efficiency to a certain extent when there are many comparison analyses in the first level (it is necessary to compare and analyze with every candidate mine location data in the original mine location data cluster).

[0095] The first noise data and the second noise data of the target mine location corresponding to the above target mine location data are mined to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the above target mine location data; the first noise data of the mine location can be mined to obtain the mine location data vector of the first data angle, and the second noise data of the mine location can be mined to obtain the mine location data vector of the second data angle.

[0096] The method for determining the mine location data vectors for the first and second data angles of each candidate mine location data in the original mine location data cluster can be the same as or different from the method for determining the target mine location data.

[0097] Specifically, in one possible implementation, the step of performing noise processing on the target mine location data to form first noise data and second noise data of the mine location corresponding to the target mine location data may further include the specific implementation details described below:

[0098] The order of at least one statement or at least one word in the above target mine location data is transformed (e.g., the third statement is placed in the position of the second statement, or the third word is placed in the position of the second word, etc.) to complete the first noise processing and form the first noise data of the mine location corresponding to the above target mine location data.

[0099] The second noise processing is performed by replacing at least one statement or word in the first noise data of the above-mentioned mine location (such as synonym replacement) to form the second noise data of the mine location corresponding to the above-mentioned target mine location data.

[0100] For example, for the target mine location data corresponding to "the area where lubricating oil is stored", the word order can be changed to obtain the first noise data of the mine location, "the area where lubricating oil is stored". Then, the relevant words in "the area where lubricating oil is stored" can be replaced to obtain the second noise data of the mine location, "the location where lubricating fluid is placed".

[0101] In detail, in one possible implementation, the step of mining the first noise data and the second noise data of the target mine location corresponding to the target mine location data to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the target mine location data may further include the specific implementation details described below:

[0102] The first noise data and the second noise data of the target mine location data are respectively subjected to shallow feature mining, and the first shallow vector and the second shallow vector of the target mine location data are output. The first shallow vector and the second shallow vector of the mine location are used to characterize the surface semantic features of the first noise data and the second noise data of the mine location, respectively. They can usually be captured by convolutional neural networks (CNN) or other shallow models.

[0103] The first shallow vector and the second shallow vector of the target mine location data are transformed by vector dimension transformation to output the first transformed vector and the second transformed vector of the target mine location data. In this way, by transforming the vector dimension, such as compressing or reducing the vector dimension, the complexity of the data can be reduced to a certain extent and the computational efficiency of the model can be improved. That is, the first shallow vector of the mine location can be transformed by vector dimension transformation to obtain the first transformed vector of the mine location, and the second shallow vector of the mine location can be transformed by vector dimension transformation to obtain the second transformed vector of the mine location.

[0104] The first and second transformation vectors of the mine location are subjected to feature filtering, respectively, to output the mine location data vectors for the first and second data angles corresponding to the target mine location data. In other words, the first transformation vector of the mine location can be subjected to feature filtering to obtain the mine location data vector for the first data angle, and the second transformation vector of the mine location can be subjected to feature filtering to obtain the mine location data vector for the second data angle. Feature filtering can also reduce the number of model parameters and computational complexity.

[0105] In detail, in one possible implementation, the aforementioned mining of the first noise data and second noise data of the target mine location corresponding to the target mine location data is performed using a feature mining network. This feature mining network includes a shallow feature mining unit, a vector dimension transformation unit, and a first feature filtering unit. Based on this, the step of mining the first noise data and second noise data of the target mine location corresponding to the target mine location data to form a mine location data vector from a first data angle and a mine location data vector from a second data angle can further include the specific implementation details described below:

[0106] Using the aforementioned shallow feature mining unit, shallow feature mining is performed on the first noise data and the second noise data of the mine location, respectively, to form a first shallow vector and a second shallow vector of the target mine location data. The shallow feature mining unit can be a convolutional network layer, including one or more convolutional kernels, used to perform convolution operations on the first noise data of the mine location to obtain the first shallow vector of the mine location, and to perform convolution operations on the second noise data of the mine location to obtain the second shallow vector of the mine location.

[0107] Using the aforementioned vector dimension transformation unit, the first shallow vector and the second shallow vector of the mine location are transformed into vector dimensions to form the first transformed vector and the second transformed vector of the mine location corresponding to the target mine location data. The vector dimension transformation performed by the aforementioned vector dimension transformation unit can refer to the Pooling operation, which is used to select the maximum value or average value as a representative feature in the window (obtained by sliding window).

[0108] Using the first feature filtering unit, the first transformation vector of the mine location is filtered to form the mine location data vector of the first data angle corresponding to the target mine location data. The first feature filtering unit may include two cascaded fully connected sub-units. The first fully connected sub-unit can map the input data (i.e. the first transformation vector of the mine location) to a high-dimensional feature space and extract richer and more abstract features from it. Then, through the second fully connected sub-unit, the high-dimensional features (the mapping result of the high-dimensional feature space) are mapped back to the low-dimensional feature space to achieve feature compression and dimensionality reduction. Such a structure can help the network learn more discriminative feature representations and reduce the number of model parameters and computational complexity.

[0109] The second transformation vector of the above-mentioned mine location is compressed in terms of vector dimension to form the mine location data vector of the second data angle corresponding to the target mine location data. The method of compressing the vector dimension of the second transformation vector of the above-mentioned mine location can be the same as the method of compressing the vector dimension of the first transformation vector of the above-mentioned mine location, or it can be different.

[0110] Before performing shallow feature mining on the first noise data and the second noise data of the mine location using the shallow feature mining unit, the first noise data and the second noise data of the mine location can be embedded first, which can be achieved using a word embedding model. Then, the shallow feature mining unit can be used to perform shallow feature mining on the obtained embedding vector.

[0111] For example:

[0112] The first noisy data point in the mine location, "the storage area of ​​lubricating oil," has the following embedding vector: [0.12,-0.05,0.18,-0.09,0.23,-0.02,-0.08,0.10,0.15,0.07,-0.14,0.11,-0.06,0.16,0.03,-0.13,0.04,0.20,-0.01,-0.16].

[0113] The second noise data of the mine location, "placement location of lubricating fluid", has the following embedding vector: [-0.08,0.10,-0.15,0.07,-0.05,-0.19,0.22,0.01,-0.04,-0.12,0.26,-0.09,0.13,-0.07,-0.25,0.18,0.06,-0.16,-0.11,0.14].

[0114] In detail, in one possible implementation, the aforementioned feature mining network may further include a second feature filtering unit. This second feature filtering unit may include a fully connected subunit that directly maps the input data to a low-dimensional feature space, achieving feature compression and dimensionality reduction. Based on this, the aforementioned information processing method based on 5G communication and navigation in open-pit mines may further include the specific implementation details described below:

[0115] The candidate feature mining network and training mine location data are determined. The candidate feature mining network can be initially built or built historically and needs further training.

[0116] Based on the aforementioned training mine location data, the aforementioned shallow feature mining units in the aforementioned candidate feature mining network are updated to form the first feature mining network corresponding to the aforementioned candidate feature mining network; that is, the first network update is only for the aforementioned shallow feature mining units.

[0117] The network parameters of the shallow feature mining unit in the first feature mining network are maintained, and the second feature filtering unit in the first feature mining network is updated to form the second feature mining network corresponding to the first feature mining network; that is, the first network update is only for the second feature filtering unit.

[0118] The network parameters of the shallow feature mining unit and the second feature filtering unit in the second feature mining network are maintained, and the first feature filtering unit in the second feature mining network is updated to form a third feature mining network corresponding to the second feature mining network. The third feature mining network is then marked to form an updated feature mining network. In other words, the first network update only targets the first feature filtering unit. Based on this, the shallow feature mining unit, the second feature filtering unit, and the first feature filtering unit can be updated separately, so that the granularity of the update is updated, thereby ensuring the reliability of the update operation.

[0119] In detail, in one possible implementation, the step of updating the shallow feature mining units in the candidate feature mining network based on the training mine location data to form the first feature mining network corresponding to the candidate feature mining network may further include the following specific implementation details:

[0120] The above-mentioned training mine location data is subjected to noise processing to form the first noise data and the second noise data of the mine location corresponding to the above-mentioned training mine location data; as described above, noise processing can be performed once or twice; in addition, the above-mentioned training mine location data is also text data obtained by parsing the training navigation request, which is used to reflect the attribute characteristics of the navigation target corresponding to the above-mentioned training navigation request.

[0121] The first noisy data of the mine location corresponding to the above training mine location data is used as the anchor training data (Anchor, the benchmark sample in the triplet, used to determine the similarity measure between other samples), and the second noisy data group of the mine location corresponding to the above training mine location data is used as the anchor positive training data (Positive, samples similar to or of the same category as the anchor, with similar features or labels to the anchor, indicating instances belonging to the same category or similar to the anchor). Other training mine location data is used as the anchor negative training data (Negative, samples dissimilar to or of a different category than the anchor, with features or labels that are significantly different from the anchor, indicating instances not belonging to the same category or dissimilar to the anchor).

[0122] Using the shallow feature mining unit in the candidate feature mining network, the anchor training data, the positive anchor training data, and the negative anchor training data are respectively subjected to shallow feature mining to form corresponding anchor training data vectors, positive anchor training data vectors, and negative anchor training data vectors. For example, the anchor training data, the positive anchor training data, and the negative anchor training data are respectively subjected to convolution operations to obtain corresponding anchor training data vectors, positive anchor training data vectors, and negative anchor training data vectors.

[0123] Calculate the first matching parameter (such as the cosine similarity between the anchor training data vector and the positive anchor training data vector) between the anchor training data vector and the positive anchor training data vector, and calculate the second matching parameter (such as the cosine similarity between the anchor training data vector and the negative anchor training data vector) between the anchor training data vector and the negative anchor training data vector. Based on the first and second matching parameters, calculate the first type of error parameter corresponding to the candidate feature mining network. For example, the sum of the second matching parameters between the anchor training data vector and each of the negative anchor training data vectors can be calculated first, then the ratio between the first matching parameter and the sum can be calculated, and the logarithm of the ratio can be taken. Finally, the result of the logarithm operation can be inverted to obtain the first type of error parameter corresponding to the candidate feature mining network.

[0124] Based on the first type of error parameter, the network parameters of the shallow feature mining unit in the candidate feature mining network are optimized and adjusted to form the first feature mining network corresponding to the candidate feature mining network. For example, the network parameters of the shallow feature mining unit in the candidate feature mining network can be adjusted along the direction of reducing the first type of error parameter.

[0125] For example, suppose the above anchor training data vector is: [0.12,-0.05,0.18,-0.09,0.23,-0.02,-0.08,0.10,0.15,0.07,-0.14,0.11,-0.06,0.16,0.03,-0.13,0.04,0.20,-0.01,-0.16];

[0126] The above anchor point negative training data vector is: [-0.08, 0.10, -0.15, 0.07, -0.05, -0.19, 0.22, 0.01, -0.04, -0.12, 0.26, -0.09, 0.13, -0.07, -0.25, 0.18, 0.06, -0.16, -0.11, 0.14];

[0127] The cosine similarity between the aforementioned anchor training data vector and the aforementioned positive anchor training data vector can be calculated by dividing the inner product of the two vectors by the product of their magnitudes. Here, the magnitude of the aforementioned anchor training data vector is 0.577, the magnitude of the aforementioned negative anchor training data vector is 0.639, and the inner product between the aforementioned anchor training data vector and the aforementioned positive anchor training data vector is -0.0909. Therefore, the cosine similarity between the aforementioned anchor training data vector and the aforementioned positive anchor training data vector can be -0.2467, indicating opposite directions and no correlation.

[0128] In detail, in one possible implementation, the step of maintaining the network parameters of the shallow feature mining units in the first feature mining network and updating the second feature filtering units in the first feature mining network to form a second feature mining network corresponding to the first feature mining network may further include the specific implementation details described below:

[0129] Using the second feature filtering unit in the first feature mining network, the anchor training data vector, the positive anchor training data vector, and the negative anchor training data vector are respectively subjected to feature filtering to form corresponding anchor training data filtering vector, positive anchor training data filtering vector, and negative anchor training data filtering vector. The feature filtering process is as described above, wherein the anchor training data vector is subjected to feature filtering to obtain the anchor training data filtering vector; the positive anchor training data vector is subjected to feature filtering to obtain the positive anchor training data filtering vector; and the negative anchor training data vector is subjected to feature filtering to obtain the negative anchor training data filtering vector.

[0130] Calculate the third matching parameter (such as the cosine similarity between the anchor training data selection vector and the positive anchor training data selection vector) between the above anchor training data selection vector and the above positive anchor training data selection vector, and calculate the fourth matching parameter (such as the cosine similarity between the anchor training data selection vector and the above negative anchor training data selection vector) between the above anchor training data selection vector and the above negative anchor training data selection vector. Based on the above third matching parameter and the above fourth matching parameter, calculate the second type of error parameter corresponding to the above first feature mining network. The calculation method of the above first type of error parameter can be referred to above.

[0131] Maintaining the network parameters of the shallow feature mining units in the first feature mining network, and optimizing and adjusting the network parameters of the second feature filtering units in the first feature mining network according to the second type of error parameters, a second feature mining network corresponding to the first feature mining network is formed; that is, keeping the network parameters of the shallow feature mining units in the first feature mining network unchanged, and optimizing and adjusting the network parameters of the second feature filtering units in the first feature mining network in the direction of reducing the second type of error parameters.

[0132] In detail, in one possible implementation, the steps of maintaining the network parameters of the shallow feature mining units and the network parameters of the second feature filtering units in the second feature mining network, updating the first feature filtering units in the second feature mining network to form a third feature mining network corresponding to the second feature mining network, and marking the third feature mining network to form an updated feature mining network, may further include the specific implementation details described below:

[0133] A training selection vector cluster is extracted. The training selection vector cluster includes training initial mine data vectors corresponding to multiple training mine location data and training vector supervision data corresponding to each of the training initial mine data vectors. The training vector supervision data is used to characterize the selection vectors for feature selection of the training initial mine data vectors using the first feature selection unit. In addition, the training initial mine data vectors include anchor training data vectors corresponding to the second noise data of the mine location corresponding to the training mine location data, anchor positive training data vectors corresponding to the first noise data of the mine location corresponding to the training mine location data, and anchor negative training data vectors corresponding to other training mine location data.

[0134] Using the first feature filtering unit in the second feature mining network, feature filtering is performed on each of the initial training mining data vectors, and the estimated filtering vectors corresponding to each of the initial training mining data vectors are output.

[0135] Calculate the vector matching parameters (such as cosine similarity) between each of the above estimated screening vectors and the corresponding training vector supervision data, and perform fusion processing on each of the above vector matching parameters, such as mean calculation, to output the third type of error parameter corresponding to the above feature mining network; for example, the difference between the target parameter and each vector matching parameter can be calculated, and then the mean of each difference can be calculated to obtain the third round error parameter, which can be equal to 1;

[0136] Maintaining the network parameters of the shallow feature mining unit and the second feature filtering unit in the second feature mining network, and based on the third type of error parameter, optimizing and adjusting the network parameters of the first feature filtering unit in the second feature mining network to form a third feature mining network corresponding to the second feature mining network; that is, keeping the network parameters of the shallow feature mining unit and the second feature filtering unit in the second feature mining network unchanged, optimizing and adjusting the network parameters of the first feature filtering unit in the second feature mining network in the direction of reducing the third type of error parameter;

[0137] The third feature mining network described above is labeled to form an updated feature mining network.

[0138] In detail, in one possible implementation, the steps described above for determining the mine location data vector of the first data angle and the mine location data vector of the second data angle for each candidate mine location data in the original mine location data cluster may further include the specific implementation details described below:

[0139] For each candidate mine location data in the above original mine location data cluster:

[0140] The above candidate mine location data is processed to generate third noise data of the mine location corresponding to the above candidate mine location data. In other words, noise processing is performed only once.

[0141] Shallow feature mining is performed on the third noise data of the mine location corresponding to the above candidate mine location data to form the third shallow vector of the mine location corresponding to the above candidate mine location data.

[0142] Transform the vector dimension of the third shallow vector of the mine location corresponding to the above candidate mine location data to form the third transformed vector of the mine location corresponding to the above candidate mine location data.

[0143] The above-mentioned third transformation vector of the mine location is subjected to two different feature screenings to form the mine location data vector of the first data angle and the mine location data vector of the second data angle corresponding to the above-mentioned candidate mine location data. That is, the mine location data vector of a data angle belongs to the result of the above-mentioned third transformation vector of the mine location being subjected to one feature screening.

[0144] Specifically, in one possible implementation, the step of performing two different feature filtering operations on the third transformation vector of the mine location to form the mine location data vector at the first data angle and the mine location data vector at the second data angle corresponding to the candidate mine location data can further include the specific implementation details described below:

[0145] First and second dimension reduction parameters are determined for dimension reduction processing of the third transformation vector of the above-mentioned mine location. The first and second dimension reduction parameters can both be parameter matrices, used to perform different dimension reductions on the third transformation vector of the above-mentioned mine location. Furthermore, the first and second dimension reduction parameters can be network parameters in the corresponding neural network, and can be updated and formed during the network training process of the neural network.

[0146] The first dimensionality reduction parameter and the third transformation vector of the mine location are fused (e.g., multiplied) to form a corresponding first fused vector; and the second dimensionality reduction parameter and the third transformation vector of the mine location are fused (e.g., multiplied) to form a corresponding second fused vector.

[0147] Activating the first fusion vector outputs a mine location data vector with a first data angle corresponding to the candidate mine location data; and activating the second fusion vector outputs a mine location data vector with a second data angle corresponding to the candidate mine location data; wherein the activation function can be:

[0148] (I) Sigmoid function (S-shaped function):

[0149] f(x) = 1 / (1 + exp(-x));

[0150] The Sigmoid function compresses the input value into a continuous output between 0 and 1, making it suitable for binary classification problems or tasks that require output probabilities.

[0151] (ii) ReLU function (corrected linear unit):

[0152] f(x) = max(0,x);

[0153] The ReLU function returns the input value if the input is greater than zero, otherwise it returns zero.

[0154] Combination Figure 3 This invention also provides an information processing device based on 5G communication and navigation in open-pit mines, which can be applied to the aforementioned information processing system based on 5G communication and navigation in open-pit mines. The aforementioned information processing device based on 5G communication and navigation in open-pit mines may include:

[0155] The location data vector determination module is used to determine the mine location data vector of the first data angle and the mine location data of the target mine location data, and to determine the mine location data vector of the first data angle and the mine location data vector of the second data angle for each candidate mine location data in the original mine location data cluster. The aforementioned target mine location data is text data obtained by parsing the target navigation request, and is used to reflect the attribute characteristics of the navigation target corresponding to the aforementioned target navigation request.

[0156] The mine location data filtering module is used to compare and analyze the mine location data vector of the first data angle of the target mine location data with the mine location data vector of the first data angle of each candidate mine location data in the original mine location data cluster, output the corresponding first correlation parameter, and, based on the first correlation parameter, determine a preset number of candidate mine location data in the original mine location data cluster and combine them to form the corresponding related mine location data cluster.

[0157] The mine location data determination module is used to compare and analyze the mine location data vector of the second data angle of the target mine location data with the mine location data vector of the second data angle of each candidate mine location data in the relevant mine location data cluster, output the corresponding second correlation parameter, and, based on the second correlation parameter, determine the relevant mine location data of the target mine location data in the relevant mine location data cluster, and the relevant mine location data is used as the location of the navigation target corresponding to the target navigation request.

[0158] In summary, the information processing method and system based on 5G communication navigation in open-pit mines provided by this invention can first determine the mine location data vectors of the first and second data angles of the target mine location data, and determine the mine location data vectors of the first and second data angles of each candidate mine location data in the original mine location data cluster; based on the mine location data vectors of the first data angles, a preset number of candidate mine location data are determined in the original mine location data cluster, and combined to form corresponding related mine location data clusters; based on the mine location data vectors of the second data angles, related mine location data are determined in the related mine location data clusters. Based on the foregoing, since the navigation target is determined by the attribute characteristics of the target navigation request, the basis for determining the navigation target can be enriched to a certain extent. In addition, a two-level comparative analysis is performed based on the attribute characteristics of the navigation target (target mine location data) from the first data angle and the second data angle of the mine location data vector. This allows the location of the navigation target to be narrowed down and matched sequentially, thereby improving the problem of relatively low reliability in determining the target location in the existing technology. In this way, the accuracy of communication and navigation in open-pit mines can be improved to a certain extent.

[0159] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An information processing method based on 5G communication and navigation in open-pit mines, characterized in that, include: The target mine location data is determined by first data angle and second data angle, and the target mine location data is also determined by first data angle and second data angle for each candidate mine location data in the original mine location data cluster. The target mine location data is text data obtained by parsing the target navigation request and is used to reflect the attribute characteristics of the navigation target corresponding to the target navigation request. The target mine location data vector of the first data angle is compared and analyzed with the mine location data vector of the first data angle of each candidate mine location data in the original mine location data cluster, and the corresponding first correlation parameter is output. Based on the first correlation parameter, a preset number of candidate mine location data are determined in the original mine location data cluster and combined to form a corresponding related mine location data cluster. The target mine location data vector at the second data angle is compared and analyzed with the mine location data vector at the second data angle of each candidate mine location data in the relevant mine location data cluster. The corresponding second correlation parameter is output. Based on the second correlation parameter, the relevant mine location data of the target mine location data is determined in the relevant mine location data cluster. The relevant mine location data is used as the location of the navigation target corresponding to the target navigation request. The second noise data is greater than the first noise data. The steps of determining the mine location data vectors for the first and second data angles of the target mine location data, and determining the mine location data vectors for the first and second data angles of each candidate mine location data in the original mine location data cluster, include: The target mine location data is noise processed to form a first noise data and a second noise data of the mine location corresponding to the target mine location data, and at least one of the first noise data and the second noise data of the mine location has noise data; The first noise data and the second noise data of the target mine location corresponding to the target mine location data are mined respectively to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the target mine location data. Determine the mine location data vector of the first data angle and the mine location data vector of the second data angle for each candidate mine location data in the original mine location data cluster; The step of mining the first noise data and the second noise data of the target mine location corresponding to the target mine location data to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the target mine location data includes: The first noise data and the second noise data of the target mine location data are respectively subjected to shallow feature mining, and the first shallow vector and the second shallow vector of the target mine location data are output. The first shallow vector and the second shallow vector of the target mine location data are transformed by vector dimension, and the first transformed vector and the second transformed vector of the target mine location data are output. The first transformation vector and the second transformation vector of the mine location are respectively subjected to feature filtering, and the mine location data vectors of the first data angle and the second data angle corresponding to the target mine location data are output. The step of mining the first noise data and the second noise data of the target mine location corresponding to the target mine location data is performed using a feature mining network, which includes a shallow feature mining unit, a vector dimension transformation unit, and a first feature filtering unit. The step of mining the first noise data and the second noise data of the target mine location corresponding to the target mine location data to form a mine location data vector of the first data angle and a mine location data vector of the second data angle of the target mine location data includes: Using the shallow feature mining unit, shallow feature mining is performed on the first noise data of the mine location and the second noise data of the mine location respectively to form the first shallow vector of the mine location and the second shallow vector of the mine location of the target mine location data. Using the vector dimension transformation unit, the first shallow vector of the mine location and the second shallow vector of the mine location are transformed by vector dimension to form the first transformation vector and the second transformation vector of the mine location corresponding to the target mine location data. Using the first feature filtering unit, the first transformation vector of the mine location is filtered to form a mine location data vector of the first data angle corresponding to the target mine location data; The second transformation vector of the mine location is compressed in terms of vector dimension to form a mine location data vector with a second data angle corresponding to the target mine location data; The step of determining the mine location data vector of the first data angle and the mine location data vector of the second data angle for each candidate mine location data in the original mine location data cluster includes: For each candidate mine location data in the original mine location data cluster: The candidate mine location data is noise-processed to form the third noise data of the mine location corresponding to the candidate mine location data. The third noise data of the mine location corresponding to the candidate mine location data is subjected to shallow feature mining to form the third shallow vector of the mine location corresponding to the candidate mine location data. The third shallow vector of the mine location corresponding to the candidate mine location data is transformed by vector dimension to form the third transformed vector of the mine location corresponding to the candidate mine location data. The third transformation vector of the mine location is subjected to two different feature filtering processes to form a mine location data vector with a first data angle and a mine location data vector with a second data angle corresponding to the candidate mine location data.

2. The information processing method based on 5G communication and navigation in open-pit mines as described in claim 1, characterized in that, The feature mining network further includes a second feature filtering unit, and the information processing method based on 5G communication and navigation in open-pit mines further includes: The candidate feature mining network and training mine location data were determined. Based on the training mine location data, the shallow feature mining units in the candidate feature mining network are updated to form the first feature mining network corresponding to the candidate feature mining network. Maintain the network parameters of the shallow feature mining unit in the first feature mining network, and update the second feature filtering unit in the first feature mining network to form a second feature mining network corresponding to the first feature mining network. Maintain the network parameters of the shallow feature mining unit and the network parameters of the second feature filtering unit in the second feature mining network, and update the first feature filtering unit in the second feature mining network to form a third feature mining network corresponding to the second feature mining network. Then, mark the third feature mining network to form an updated feature mining network.

3. The information processing method based on 5G communication and navigation in open-pit mines as described in claim 2, characterized in that, The step of updating the shallow feature mining units in the candidate feature mining network based on the training mine location data to form the first feature mining network corresponding to the candidate feature mining network includes: The training mine location data is noise-processed to form first noise data and second noise data of the mine location corresponding to the training mine location data. The first noise data of the mine location corresponding to the training mine location data is used as anchor training data, the second noise data group of the mine location corresponding to the training mine location data is used as anchor positive training data, and other training mine location data is used as anchor negative training data. Using the shallow feature mining unit in the candidate feature mining network, the anchor training data, the anchor positive training data, and the anchor negative training data are respectively subjected to shallow feature mining to form corresponding anchor training data vectors, anchor positive training data vectors, and anchor negative training data vectors; Calculate a first matching parameter between the anchor training data vector and the anchor positive training data vector, and calculate a second matching parameter between the anchor training data vector and the anchor negative training data vector; and calculate a first type of error parameter corresponding to the candidate feature mining network based on the first matching parameter and the second matching parameter. Based on the first type of error parameters, the network parameters of the shallow feature mining units in the candidate feature mining network are optimized and adjusted to form the first feature mining network corresponding to the candidate feature mining network.

4. The information processing method based on 5G communication and navigation in open-pit mines as described in claim 3, characterized in that, The step of maintaining the network parameters of the shallow feature mining units in the first feature mining network and updating the second feature filtering units in the first feature mining network to form a second feature mining network corresponding to the first feature mining network includes: Using the second feature filtering unit in the first feature mining network, the anchor training data vector, the anchor positive training data vector, and the anchor negative training data vector are respectively subjected to feature filtering to form corresponding anchor training data filtering vector, anchor positive training data filtering vector, and anchor negative training data filtering vector; Calculate the third matching parameter between the anchor training data selection vector and the anchor positive training data selection vector, and calculate the fourth matching parameter between the anchor training data selection vector and the anchor negative training data selection vector. Based on the third matching parameter and the fourth matching parameter, calculate the second type of error parameter corresponding to the first feature mining network. Maintain the network parameters of the shallow feature mining unit in the first feature mining network, and optimize and adjust the network parameters of the second feature filtering unit in the first feature mining network according to the second type of error parameters to form a second feature mining network corresponding to the first feature mining network.

5. The information processing method based on 5G communication and navigation in open-pit mines as described in claim 4, characterized in that, The steps of maintaining the network parameters of the shallow feature mining units and the network parameters of the second feature filtering units in the second feature mining network, updating the first feature filtering units in the second feature mining network to form a third feature mining network corresponding to the second feature mining network, and marking the third feature mining network to form an updated feature mining network include: A training selection vector cluster is extracted. The training selection vector cluster includes training initial mine data vectors corresponding to multiple training mine location data and training vector supervision data corresponding to each training initial mine data vector. The training vector supervision data is used to characterize the selection vectors for feature selection of the training initial mine data vectors using the first feature selection unit. Using the first feature filtering unit in the second feature mining network, feature filtering is performed on each of the initial training mining data vectors, and the estimated filtering vector corresponding to each of the initial training mining data vectors is output. Calculate the vector matching parameters between each of the estimated screening vectors and the corresponding training vector supervision data, and perform fusion processing on each of the vector matching parameters to output the third type of error parameters corresponding to the feature mining network; Maintain the network parameters of the shallow feature mining unit and the network parameters of the second feature filtering unit in the second feature mining network, and optimize and adjust the network parameters of the first feature filtering unit in the second feature mining network based on the third type of error parameter to form a third feature mining network corresponding to the second feature mining network. The third feature mining network is labeled to form an updated feature mining network.

6. An information processing system based on 5G communication and navigation in open-pit mines, characterized in that, It includes a processor and a memory, the memory being used to store a computer program, and the processor being used to execute the computer program to implement the information processing method based on 5G communication and navigation in open-pit mines as described in any one of claims 1-5.

Images