Automatic situation plotting method and device and electronic equipment

Abstract

The invention provides an automatic situation plotting method and device and electronic equipment, and the method comprises the steps: carrying out the key information extraction of an initial combat plan text based on a large language model, obtaining the key combat information in the initial combat plan text, and carrying out the retrieval in a military mark library based on the key combat information, obtaining an initial military mark corresponding to the key battle information, and plotting the initial military mark in a military map to obtain an initial situation map of a battlefield; and performing key information extraction on the received battlefield real-time data, and updating the initial situation map based on the extracted real-time key information to obtain a real-time situation map of the battlefield. Based on the combination of the large language model and the military standard library, the whole process automation from the battle plan to the real-time situation icon drawing update is realized, the battlefield situation awareness efficiency and accuracy are obviously improved, and powerful support is provided for command decision making.

Description

Technical Field

[0001] The present invention relates to the technical field of military situation mapping, and in particular to an automatic situation mapping method, device and electronic equipment. Background Art

[0002] With the advent of the information age, battlefield information visualization technology has ushered in new opportunities for rapid development. Against this backdrop, situational mapping has evolved. Using military maps as a foundational basemap, it uses military symbols to digitally represent detailed information such as enemy and friendly forces' combat intentions, troop deployments, and military operations.

[0003] The existing military situation mapping system mainly relies on manual operation, and has problems such as low efficiency and insufficient ability to process complex data. Summary of the Invention

[0004] The present invention provides a method, device and electronic equipment for automatic situation mapping, which are used to solve the defects of the existing military situation mapping system, which mainly relies on manual operation, has low efficiency and insufficient ability to process complex data. It realizes the intelligence and automation of the entire process of military situation mapping and improves the efficiency of situation mapping.

[0005] The present invention provides a situation automatic plotting method, comprising the following steps: Extract key information from the initial combat plan text based on the large language model to obtain key combat information in the initial combat plan text, wherein the key combat information includes place name information and military action information; Based on the key combat information, searching in a military standard database to obtain an initial military standard corresponding to the key combat information, and plotting the initial military standard on a military map to obtain an initial battlefield situation map; Based on the large language model, key information is extracted from the received real-time battlefield data, and based on the extracted real-time key information, the initial situation map is updated to obtain a real-time situation map of the battlefield.

[0006] According to an automatic situation plotting method provided by the present invention, the initial military symbol is plotted on a military map to obtain an initial situation map of the battlefield, comprising: Mapping the place name information to longitude and latitude based on the large language model to obtain longitude and latitude information corresponding to the place name information; The initial military symbol is plotted on the military map according to the latitude and longitude information to obtain an initial situation map of the battlefield.

[0007] According to the automatic situation mapping method provided by the present invention, after obtaining the latitude and longitude information corresponding to the place name information, the method further includes: Correcting the latitude and longitude information based on a position information correction formula to obtain corrected latitude and longitude information; The position information correction formula is: ; in, For the moment The state vector of the corrected latitude and longitude information, For the moment The optimal estimate of the state vector, is the state transition matrix, is the control input matrix, is the control vector, is the Kalman gain, is the observation value of satellite positioning data, is the observation matrix.

[0008] According to an automatic situation plotting method provided by the present invention, the key information of the initial combat plan text is extracted based on a large language model to obtain the key combat information in the initial combat plan text, including: Based on the prompt word template, the large language model is guided to determine the lexical weight of each word in the initial combat plan text based on the attention mechanism. Based on the lexical weight of each word, all words in the initial combat plan text are screened to obtain key words, and the large language model is guided to identify the key words to determine the place name information and military action information in the initial combat plan text.

[0009] According to a situation automatic plotting method provided by the present invention, the vocabulary weight of each vocabulary is determined based on a vocabulary weight determination formula, and the vocabulary weight determination formula is: ; in, It is the first The attention weight of each word, It is the first words, is the vocabulary importance scoring function, is the total number of words in the initial battle plan text.

[0010] According to a situation automatic plotting method provided by the present invention, based on the key combat information, searching in a military standard database to obtain an initial military standard corresponding to the key combat information includes: A word embedding method based on the BERT model is used to extract word vectors of the key combat information; Based on the similarity between the word vector of the key combat information and the word vector of each military standard in the military standard library, the military standard library is searched to obtain the initial military standard corresponding to the key combat information.

[0011] According to an automatic situation plotting method provided by the present invention, key information is extracted from received real-time battlefield data based on the large language model, and the initial situation map is updated based on the extracted real-time key information to obtain a real-time situation map of the battlefield, including: receiving real-time battlefield data of the battlefield, wherein the real-time battlefield data includes multi-source sensor data, reconnaissance data, force deployment data, and action trajectory data; Based on the large language model, key information is extracted from the battlefield real-time data to obtain real-time combat information in the battlefield real-time data; Based on the real-time combat information, searching in a military standard database to obtain a real-time military standard corresponding to the real-time combat information; Based on the real-time military symbol, the military symbol in the initial situation map is updated to obtain the real-time situation map of the battlefield.

[0012] According to an automatic situation plotting method provided by the present invention, after updating the initial situation map to obtain the real-time situation map of the battlefield, the method further includes: Based on the large language model, the real-time situation map is analyzed to determine the command decision information of the battlefield.

[0013] The present invention also provides a situation automatic plotting device, comprising the following modules: A key information extraction module is used to extract key information from the initial combat plan text based on a large language model to obtain key combat information in the initial combat plan text, wherein the key combat information includes place name information and military action information; An initial situation map determining module is used to search a military standard database based on the key combat information to obtain an initial military standard corresponding to the key combat information, and to plot the initial military standard on a military map to obtain an initial situation map of the battlefield; The situation map updating module is used to extract key information from the received real-time battlefield data based on the large language model, and to update the initial situation map based on the extracted real-time key information to obtain a real-time situation map of the battlefield.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the program, any of the above-mentioned automatic situation plotting methods is implemented.

[0015] The automatic situation mapping method, device, and electronic device provided by this invention use a large language model to extract key operational information from the initial battle plan text, search the information in a military standard library, and automatically plot the information on a military map based on the retrieved initial military standard to obtain an initial situation map. This initial situation map is then updated based on received real-time battlefield data. The combination of the large language model and the military standard library automates the entire process, from battle planning to real-time situation map mapping and updating, significantly improving the efficiency and accuracy of battlefield situation awareness and providing strong support for command decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 It is a flow chart of the automatic situation plotting method provided by the present invention.

[0018] Figure 2 This is a schematic diagram of the situation plotting process provided by the present invention.

[0019] Figure 3 It is a structural diagram of the automatic situation plotting system provided by the present invention.

[0020] Figure 4 This is a schematic diagram of the situation map update process provided by the present invention.

[0021] Figure 5 It is a structural schematic diagram of the automatic situation plotting device provided by the present invention.

[0022] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0023] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0024] Figure 1 It is a flow chart of the automatic situation plotting method provided by the present invention, such as Figure 1 As shown, the method includes the following: Step 110: extract key information from the initial combat plan text based on the large language model to obtain key combat information in the initial combat plan text, wherein the key combat information includes place name information and military action information; Step 120: Based on the key combat information, search the military standard database to obtain an initial military standard corresponding to the key combat information, and plot the initial military standard on a military map to obtain an initial battlefield situation map; Step 130 , based on the large language model, extract key information from the received real-time battlefield data, and based on the extracted real-time key information, update the initial situation map to obtain a real-time situation map of the battlefield.

[0025] The automatic situation mapping method provided by the present invention can be executed by an electronic device, a component of an electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), while the non-mobile electronic device can be a server, a network attached storage (NAS), or a personal computer (PC), etc., although the present invention does not impose any specific limitations thereon.

[0026] The following takes the automatic situation plotting method provided by the present invention executed by a computer as an example to describe the technical solution of the present invention in detail.

[0027] In step 110, the initial combat plan text is obtained, and key information is extracted from the obtained initial combat plan text.

[0028] It should be noted that the initial operational plan is a written document used to clarify core elements of military operations, such as operational objectives, division of tasks, action steps, and resource allocation. It is typically compiled by military command or operational planning departments to guide troops in executing combat missions.

[0029] After obtaining the initial battle plan text, key information is extracted from the text based on the large language model. Prompt words can be set to guide the large language model to extract key information from the initial battle plan text.

[0030] Specifically, the set prompt words can guide the large language model to calculate the weight of each word in the initial combat plan text based on the attention mechanism. Based on the calculated weight of each word, all words are filtered to obtain key information. After obtaining the key information, the large language model is further guided to identify key information, including place names and military action information.

[0031] The specific implementation formula for key information extraction is: ; in, It is the first The attention weight of each word, It is the first words, is the vocabulary importance scoring function, is the total number of words in the initial battle plan text, It is the key information to be determined.

[0032] It's important to note that a large language model is an AI model based on deep learning. By learning from massive amounts of text data, it mastered the grammar, semantics, logic, and knowledge of a language, enabling it to perform natural language processing tasks. Specifically, a large language model can be implemented based on the Transformer architecture, including an encoder-decoder structure.

[0033] The large language model in the embodiment of the present invention is a trained large language model, and specifically the training process can be achieved by optimizing model parameters by gradient descent (such as minimizing the cross entropy loss between predicted words and real words) based on massive text data.

[0034] In step 120 , based on the place name information and military action information obtained in step 110 , a search is performed in the military standard database to determine the corresponding initial military standard.

[0035] A pre-built military symbol library is a database that stores military symbols (military symbols). It can include terrain symbols (such as mountains and rivers), troop symbols (such as infantry and armor), and action symbols (such as attack arrows and fortifications). Each military symbol can be represented by specific semantics and visual representations.

[0036] Based on the extracted place name information and military action information, text similarity calculation can be performed with the description information of military symbols in the military symbol database, and military symbols with similarity greater than a preset similarity threshold are determined as military symbols that match the place name information or military action information.

[0037] Optionally, a matching military standard can be determined from the military standard library based on the pre-similarity. The cosine similarity formula is: ; in, is the text feature vector of key combat information, The first sentence in the library k The characteristic vector of a military standard.

[0038] After obtaining the initial military symbol, plot it on a military map to create an initial battlefield situation map. Specifically, the matching initial military symbol can be overlaid on a military map to generate the initial situation map. The map must support geographic information system functionality to ensure that the location of the military symbol aligns with the actual terrain.

[0039] In step 130, after obtaining the initial situation map, real-time battlefield data is received. This real-time battlefield data reflects data related to the current battlefield situation acquired in real-time. This data may include multi-source sensor data, reconnaissance data, troop deployment data, and movement trajectory data.

[0040] After acquiring real-time battlefield data, the system extracts key information from the data based on a large language model, generating real-time critical battlefield information. Based on this real-time critical information, the initial situation map is updated to create a real-time battlefield situation map. This real-time battlefield situation map allows for real-time understanding of battlefield dynamics and rapid decision-making.

[0041] The specific implementation process can be as follows Figure 2 The situation mapping process provided by the present invention is shown in the schematic diagram. The specific process includes: Receive user input: The user enters the initial battle plan text in natural language through a conversational interface.

[0042] Parsing and Understanding: Leveraging a large language model, we extract key information from the initial operational plan text, such as place names and military operations. Deep learning algorithms are then used to further filter this key information to accurately understand user needs and intent, place name information, and military operations.

[0043] Icon Recommendation and Geolocation: Based on extracted key information, it automatically recommends relevant military symbols and determines the corresponding geographical location of place names. Based on a pre-set military symbol library, it combines key information for matching and positioning, providing users with appropriate military symbol icons and accurate geographical coordinates.

[0044] Plotting: Automatically draw and mark military symbols on the map to generate an initial situation map. Based on the recommended military symbols and the determined locations, the system automatically draws and marks them on the electronic map to form a preliminary situation map for the user.

[0045] Dynamic update and optimization: Continuously receive real-time battlefield data, and dynamically adjust and update the situation map through the analysis and prediction functions of the large model to enable it to reflect the latest situation on the battlefield in real time.

[0046] The automatic situation mapping method provided by this invention uses a large language model to extract key operational information from the initial battle plan text, retrieves it from a military standards library, and automatically plots it on a military map based on the retrieved initial military standards, generating an initial situation map. This initial situation map is then updated based on received real-time battlefield data. The combination of the large language model and the military standards library automates the entire process, from battle planning to real-time situation map mapping and updating, significantly improving the efficiency and accuracy of battlefield situation awareness and providing strong support for command decision-making.

[0047] In one embodiment, the initial military symbol is plotted on a military map to obtain an initial situation map of the battlefield, including: mapping the place name information into longitude and latitude based on the large language model to obtain longitude and latitude information corresponding to the place name information; and plotting the initial military symbol on the military map according to the longitude and latitude information to obtain an initial situation map of the battlefield.

[0048] In the process of constructing the initial battlefield situation map, we first use a large language model to deeply analyze and semantically understand the place name information extracted from the initial combat plan text. With the help of the model's built-in geographic knowledge base and place name-coordinate mapping algorithm, we can accurately convert unstructured place name text into structured latitude and longitude coordinates.

[0049] Using military maps as the carrier and the converted longitude and latitude information as the spatial reference, the coordinate positioning and symbol rendering functions of the geographic information system (GIS) are used to accurately superimpose the initial military symbols associated with place names (such as icons representing troop deployment and symbols representing fortifications) on the corresponding positions on the map to generate an initial battlefield situation map.

[0050] Optionally, the large language model can be guided to implement the process of mapping place name information to longitude and latitude based on the longitude and latitude mapping formula.

[0051] The latitude and longitude mapping formula is: ; in, is the place name embedding vector, is the probability distribution of the position predicted by the multilayer perceptron, are all possible geographic location candidates.

[0052] In one embodiment, after obtaining the latitude and longitude information corresponding to the place name information, the method further includes: Correcting the latitude and longitude information based on a position information correction formula to obtain corrected latitude and longitude information; The position information correction formula is: ; in, For the moment The state vector of the corrected latitude and longitude information; For the moment The optimal estimate of the state vector of is the state transition matrix, which describes the system from time At the time The state change relationship; is the control input matrix, which controls the vector Mapping into state space; is the control vector, representing the external control quantity of the test drive at time k; is the Kalman gain, which is used to balance the weights of prediction information and observation information; The observation value of satellite positioning data can be the longitude and latitude information obtained by the GPS sensor; is the observation matrix, which maps the state space to the observation space and is used to convert the state estimate into the same form as the observation value.

[0053] Kalman filtering is a recursive state estimation algorithm suitable for processing noisy data in dynamic systems.

[0054] By combining system models with observational data, the Kalman filter enables high-precision, low-latency positioning optimization in noisy environments. Its core principle is to provide reliable state estimation in uncertain environments, making it particularly suitable for military scenarios that require both real-time performance and accuracy.

[0055] In one embodiment, based on a large language model, key information is extracted from an initial combat plan text to obtain key combat information in the initial combat plan text, including: based on a prompt word template, guiding the large language model to determine the vocabulary weight of each word in the initial combat plan text based on an attention mechanism; based on the vocabulary weight of each word, screening all words in the initial combat plan text to obtain key words; and guiding the large language model to identify the key words to determine place name information and military action information in the initial combat plan text.

[0056] During the processing of the initial battle plan text, a pre-designed structured prompt word template is used to guide the large language model to dynamically calculate the weight distribution of each word in the text based on its built-in attention mechanism. The attention mechanism determines the importance of each word by capturing the semantic associations and contextual dependencies between words.

[0057] Key words are filtered out based on preset weight thresholds or sorting rules, and the large language model is further guided to classify and identify the filtered words, extracting place name information and military action information.

[0058] In one embodiment, the vocabulary weight of each vocabulary word is determined based on a vocabulary weight determination formula, which is: ; in, It is the first The attention weight of each word, It is the first words, is the vocabulary importance scoring function, is the total number of words in the initial battle plan text.

[0059] Based on the attention mechanism, the semantic association and context dependency between words are captured to determine the lexical weight of each word, which provides a basis for the subsequent lexical screening process.

[0060] The vocabulary screening process can be determined based on the screening formula, which is: ; in, For key words.

[0061] In one embodiment, based on the key combat information, a search is performed in a military standard library to obtain an initial military standard corresponding to the key combat information, including: extracting a word vector of the key combat information based on a word embedding method based on a BERT model; and searching in the military standard library based on the similarity between the word vector of the key combat information and the word vectors of each military standard in the military standard library to obtain the initial military standard corresponding to the key combat information.

[0062] In the military standard retrieval process based on the BERT (Bidirectional Encoder Representations from Transformers) model, the BERT model is used to perform deep semantic encoding on key combat information extracted from real-time battlefield data to generate word vectors.

[0063] The specific word vector extraction formula is: ; in, It is the first real-time battlefield data i words, through the sliding window n Capture contextual semantics.

[0064] Through its bidirectional Transformer architecture, the BERT model can capture contextual semantic information in text and convert key operational information into vector representations.

[0065] The generated word vector is similar to the word vector of the description text of each military standard in the pre-built military standard library. The degree of semantic matching between the two can be quantified by using measurement methods such as cosine similarity or Euclidean distance.

[0066] According to the similarity sorting results, the military standard with the closest semantics to the key combat information is selected as the retrieval output.

[0067] In one embodiment, based on the large language model, key information is extracted from the received real-time battlefield data, and based on the extracted real-time key information, the initial situation map is updated to obtain a real-time situation map of the battlefield, including: receiving real-time battlefield data of the battlefield, the real-time battlefield data including multi-source sensor data, reconnaissance data, force deployment data and action trajectory data; based on the large language model, key information is extracted from the real-time battlefield data to obtain real-time combat information in the real-time battlefield data; based on the real-time combat information, a military standard library is searched to obtain a real-time military standard corresponding to the real-time combat information; based on the real-time military standard, the military standard in the initial situation map is updated to obtain a real-time situation map of the battlefield.

[0068] Since battlefield combat information will change dramatically over time and with the execution of combat missions, after the initial situation map is generated, it is necessary to update the initial situation map based on the received real-time battlefield data to meet the subsequent requirements for battlefield situation analysis and command decision-making.

[0069] Specifically, it first receives multi-source real-time data from the battlefield. These data include multi-source sensor data (such as environmental and target information collected by radar, infrared detectors and other equipment), reconnaissance data (such as drone reconnaissance and enemy intelligence obtained by special forces infiltration), force deployment data (such as the current location of each unit, organizational size and deployment form) and action trajectory data (such as the movement path of troops or equipment).

[0070] Then, with the help of the powerful natural language processing and semantic understanding capabilities of the large language model, we conduct in-depth analysis and key information extraction of the massive and complex real-time battlefield data, and accurately extract key combat information from the real-time battlefield data, including but not limited to the dynamics of enemy and friendly forces, tactical actions, target status, etc.

[0071] Then, based on the extracted real-time combat information, a quick search is performed in the pre-built military standard library to find the real-time military standards that precisely match it. These military standards intuitively represent various combat elements in the form of standardized symbols.

[0072] Finally, the retrieved real-time military symbols are dynamically superimposed on the initial situation map, and the existing military symbols are updated in position, changed in status, or newly labeled, thereby generating a situation map that reflects the latest battlefield situation in real time, providing timely, accurate, and comprehensive battlefield information support, and helping them make scientific and efficient combat decisions.

[0073] In one embodiment, after updating the initial situation map to obtain the real-time situation map of the battlefield, the method further includes: analyzing the real-time situation map based on the large language model to determine the command decision information of the battlefield.

[0074] During battlefield situation analysis, the real-time situation map is parsed through a large language model. Specifically, the geographical elements, military symbols, and dynamic annotations in the situation map can be converted into structured text descriptions. Subsequently, the contextual understanding capability of the large language model is utilized, combined with the military knowledge base and tactical rules, to conduct in-depth reasoning on the situation information, ultimately forming battlefield command and decision-making information.

[0075] Optionally, after obtaining the real-time situation map, predictions can also be performed on the map. Combining real-time battlefield data with the predictive capabilities of the large language model, the large model is used to analyze and predict the received real-time battlefield data, generating a dynamic situation map reflecting the current battlefield situation. The map's content is then updated promptly based on changes in the battlefield situation.

[0076] Situation prediction can use the Transforme model to predict battlefield changes: ; in, , , are query, key, and value matrices respectively, is the dimension scaling factor.

[0077] The fusion of multi-source data in real-time battlefield data can be achieved based on weighted fusion. The fusion implementation formula is: ; in, is the sensor confidence weight, For the i Situational score of each sensor.

[0078] The dynamic update rules of the situation map are: ; in is the learning rate, The model is in time The predicted value of It's time The true observed value of .

[0079] The present invention also provides a situation automatic plotting system, such as Figure 3 As shown in the structural diagram of the situation automatic plotting system provided by the present invention, the system specifically includes a display layer, an interface layer, a service layer and a data storage layer.

[0080] The system is based on a B / S architecture. The front-end user interface is built using HTML5, CSS3, and JavaScript. The back-end server logic is written in Java, and the database uses PostgreSQL to store plotting data. A large language model is integrated into the back-end server, providing intelligent support for various functional modules.

[0081] The display layer specifically includes the user interface display and map display, which can be implemented based on Vue and WebGIS; the interface layer includes multiple API interfaces; the service layer specifically includes user interaction optimization module, automatic icon recognition and recommendation module, intelligent geographic positioning module and dynamic situation generation module.

[0082] The user interaction optimization module is used to allow users to input the initial operation plan text in natural language through a conversational interface and automatically perform plotting operations.

[0083] The automatic icon recognition and recommendation module is used to automatically parse the text information in the initial combat plan text based on the natural language processing capabilities of the large language model, extract key place names and military operations, and recommend relevant military symbols.

[0084] The intelligent geolocation module, leveraging the geographic information processing capabilities of a large language model, automatically identifies the geographic locations corresponding to place names, reducing manual search time. Combining geocoding technology with the semantic understanding capabilities of a large model, this module quickly and accurately converts place names in text into specific coordinates on a map, improving the efficiency and accuracy of geolocation.

[0085] The dynamic situation generation module is used to automatically generate and update the situation map by combining real-time battlefield data and the predictive capabilities of large language models.

[0086] The specific update process of the situation map update by the above situation automatic plotting system can be as follows: Figure 4 As shown in the schematic diagram of the situation map update process provided by the present invention, the specific process includes: Automated Icon Recognition and Recommendation: After a user enters a combat plan, the system leverages the natural language processing capabilities of a large language model to perform semantic analysis of the initial plan, extracting key place names and military action information. The system then matches and recommends relevant military icons based on a pre-defined military standard library, reducing the user's manual search time.

[0087] Intelligent geolocation: For extracted place names, the system uses a large language model combined with geocoding technology to quickly and accurately determine their specific coordinates on the map. This process eliminates the need for users to manually search and locate on the map, greatly improving geolocation efficiency.

[0088] Dynamic Situation Generation: The system receives real-time battlefield data, including information on enemy and friendly troop deployments and movement trajectories. Leveraging the predictive capabilities of large language models, it analyzes and predicts battlefield trends, automatically generating a situation map reflecting the current battlefield situation. The map is dynamically updated based on changes in real-time data, ensuring its timeliness and accuracy.

[0089] Improved user interaction: Users can enter commands in natural language through a conversational interface, such as "Mark our tank unit at a certain location." The system automatically understands the command and performs the corresponding marking operation, eliminating the need for users to navigate complex menus. This improves the convenience and efficiency of human-computer interaction.

[0090] The user interaction optimization process includes intent recognition, which can be implemented based on a multi-task learning instruction classification model: ; For user input text, The plotting operation category (such as "mark", "delete", "update").

[0091] Through the above system, the automation and intelligence of military situation mapping are realized, the mapping efficiency is improved, the time and energy consumption of manual operation are reduced, and more timely and accurate situation information support is provided for military command decision-making.

[0092] The automatic situation plotting device provided by the present invention is described below. The automatic situation plotting device described below and the automatic situation plotting method described above can be referenced to each other.

[0093] like Figure 5 As shown, the device includes: A key information extraction module 510 is configured to extract key information from the initial combat plan text based on a large language model to obtain key combat information from the initial combat plan text, wherein the key combat information includes place name information and military action information; The initial situation map determining module 520 is configured to search a military standard database based on the key combat information to obtain an initial military standard corresponding to the key combat information, and plot the initial military standard on a military map to obtain an initial situation map of the battlefield; The situation map updating module 530 is used to extract key information from the received battlefield real-time data based on the large language model, and update the initial situation map based on the extracted real-time key information to obtain a real-time situation map of the battlefield.

[0094] The automatic situation mapping device provided by this invention uses a large language model to extract key operational information from the initial battle plan text, searches it against a military standards library, and automatically plots the retrieved initial military standards on a military map to produce an initial situation map. This initial situation map is then updated based on received real-time battlefield data. The combination of the large language model and the military standards library automates the entire process, from battle planning to real-time situation map mapping and updating. This significantly improves the efficiency and accuracy of battlefield situation awareness and provides strong support for command decision-making.

[0095] In one embodiment, the initial situation map determination module 520 is specifically configured to: The initial military symbol is plotted on a military map to obtain an initial battlefield situation map, including: Mapping the place name information to longitude and latitude based on the large language model to obtain longitude and latitude information corresponding to the place name information; The initial military symbol is plotted on the military map according to the latitude and longitude information to obtain an initial situation map of the battlefield.

[0096] In one embodiment, the initial situation map determination module 520 is further specifically configured to: After obtaining the latitude and longitude information corresponding to the place name information, the method further includes: Correcting the latitude and longitude information based on a position information correction formula to obtain corrected latitude and longitude information; The position information correction formula is: ; in, is the state vector of longitude and latitude information, is the optimal estimate of the state vector at the previous moment, is the state transition matrix, is the control input matrix, is the control vector, is the Kalman gain, is the observation value of satellite positioning data, is the observation matrix.

[0097] In one embodiment, the key information extraction module 510 is specifically configured to: Based on the large language model, key information is extracted from the initial combat plan text to obtain key combat information in the initial combat plan text, including: Based on the prompt word template, the large language model is guided to determine the lexical weight of each word in the initial combat plan text based on the attention mechanism. Based on the lexical weight of each word, all words in the initial combat plan text are screened to obtain key words, and the large language model is guided to identify the key words to determine the place name information and military action information in the initial combat plan text.

[0098] In one embodiment, the key information extraction module 510 is further specifically configured to: The vocabulary weight of each word is determined based on a vocabulary weight determination formula, which is: ; in, The first The attention weight of each word, The first words, is the vocabulary importance scoring function, is the total number of words in the initial battle plan text.

[0099] In one embodiment, the key information extraction module 510 is further specifically configured to: Based on the key combat information, the military standard database is searched to obtain the initial military standard corresponding to the key combat information, including: A word embedding method based on the BERT model is used to extract word vectors of the key combat information; Based on the similarity between the word vector of the key combat information and the word vector of each military standard in the military standard library, the military standard library is searched to obtain the initial military standard corresponding to the key combat information.

[0100] In one embodiment, the situation map updating module 530 is specifically configured to: Extracting key information from the received real-time battlefield data based on the large language model, and updating the initial situation map based on the extracted real-time key information to obtain a real-time situation map of the battlefield, including: receiving real-time battlefield data of the battlefield, wherein the real-time battlefield data includes multi-source sensor data, reconnaissance data, force deployment data, and action trajectory data; Based on the large language model, key information is extracted from the battlefield real-time data to obtain real-time combat information in the battlefield real-time data; Based on the real-time combat information, searching in a military standard database to obtain a real-time military standard corresponding to the real-time combat information; Based on the real-time military symbol, the military symbol in the initial situation map is updated to obtain the real-time situation map of the battlefield.

[0101] In one embodiment, the situation map updating module 530 is further configured to: After updating the initial situation map to obtain the real-time situation map of the battlefield, the method further includes: Based on the large language model, the real-time situation map is analyzed to determine the command decision information of the battlefield.

[0102] Figure 6 An example of a physical structure diagram of an electronic device is shown below. Figure 6 As shown, the electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communications bus 640, wherein the processor 610, the communications interface 620, and the memory 630 communicate with each other via the communications bus 640. The processor 610 may call logic instructions in the memory 630 to execute a method for automatic situation mapping, which includes: extracting key information from an initial combat plan text based on a large language model to obtain key combat information from the initial combat plan text, wherein the key combat information includes place name information and military action information; Based on the key combat information, searching in a military standard database to obtain an initial military standard corresponding to the key combat information, and plotting the initial military standard on a military map to obtain an initial battlefield situation map; Based on the large language model, key information is extracted from the received real-time battlefield data, and based on the extracted real-time key information, the initial situation map is updated to obtain a real-time situation map of the battlefield.

[0103] Furthermore, the logic instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0104] On the other hand, the present invention further provides a computer program product, comprising a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the automatic situation mapping method provided by the above methods, the method comprising: extracting key information from an initial combat plan text based on a large language model to obtain key combat information in the initial combat plan text, wherein the key combat information includes place name information and military action information; Based on the key combat information, searching in a military standard database to obtain an initial military standard corresponding to the key combat information, and plotting the initial military standard on a military map to obtain an initial battlefield situation map; Based on the large language model, key information is extracted from the received real-time battlefield data, and based on the extracted real-time key information, the initial situation map is updated to obtain a real-time situation map of the battlefield.

[0105] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for automatically plotting a situation provided by the above methods is implemented, the method comprising: extracting key information from an initial combat plan text based on a large language model to obtain key combat information in the initial combat plan text, wherein the key combat information includes place name information and military action information; Based on the key combat information, searching in a military standard database to obtain an initial military standard corresponding to the key combat information, and plotting the initial military standard on a military map to obtain an initial battlefield situation map; Based on the large language model, key information is extracted from the received real-time battlefield data, and based on the extracted real-time key information, the initial situation map is updated to obtain a real-time situation map of the battlefield.

[0106] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0107] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A situation automatic plotting method, characterized in that: include: Extract key information from the initial combat plan text based on the large language model to obtain key combat information in the initial combat plan text, wherein the key combat information includes place name information and military action information; Based on the key combat information, searching in a military standard database to obtain an initial military standard corresponding to the key combat information, and plotting the initial military standard on a military map to obtain an initial battlefield situation map; Based on the large language model, key information is extracted from the received real-time battlefield data, and based on the extracted real-time key information, the initial situation map is updated to obtain a real-time situation map of the battlefield.

2. The automatic situation plotting method according to claim 1, characterized in that: The process of plotting the initial military symbol on a military map to obtain an initial battlefield situation map includes: Mapping the place name information to longitude and latitude based on the large language model to obtain longitude and latitude information corresponding to the place name information; The initial military symbol is plotted on the military map according to the latitude and longitude information to obtain an initial situation map of the battlefield.

3. The automatic situation plotting method according to claim 2, characterized in that: After obtaining the latitude and longitude information corresponding to the place name information, the method further includes: Correcting the latitude and longitude information based on a position information correction formula to obtain corrected latitude and longitude information; The position information correction formula is: ； in, For the moment The state vector of the corrected latitude and longitude information, For the moment The optimal estimate of the state vector, is the state transition matrix, is the control input matrix, is the control vector, is the Kalman gain, is the observation value of satellite positioning data, is the observation matrix.

4. The automatic situation plotting method according to claim 1, characterized in that: The key information of the initial combat plan text is extracted based on the large language model to obtain the key combat information in the initial combat plan text, including: Based on the prompt word template, the large language model is guided to determine the lexical weight of each word in the initial combat plan text based on the attention mechanism. Based on the lexical weight of each word, all words in the initial combat plan text are screened to obtain key words, and the large language model is guided to identify the key words to determine the place name information and military action information in the initial combat plan text.

5. The automatic situation plotting method according to claim 4, characterized in that: The vocabulary weight of each word is determined based on a vocabulary weight determination formula, which is: ； in, It is the first The attention weight of each word, It is the first words, is the vocabulary importance scoring function, is the total number of words in the initial battle plan text.

6. The automatic situation plotting method according to claim 1, characterized in that: The searching in the military standard database based on the key combat information to obtain the initial military standard corresponding to the key combat information includes: A word embedding method based on the BERT model is used to extract word vectors of the key combat information; Based on the similarity between the word vector of the key combat information and the word vector of each military standard in the military standard library, the military standard library is searched to obtain the initial military standard corresponding to the key combat information.

7. The automatic situation plotting method according to claim 1, characterized in that: The method of extracting key information from the received real-time battlefield data based on the large language model and updating the initial situation map based on the extracted real-time key information to obtain the real-time situation map of the battlefield includes: receiving real-time battlefield data of the battlefield, wherein the real-time battlefield data includes multi-source sensor data, reconnaissance data, force deployment data, and action trajectory data; Based on the large language model, key information is extracted from the battlefield real-time data to obtain real-time combat information in the battlefield real-time data; Based on the real-time combat information, searching in a military standard database to obtain a real-time military standard corresponding to the real-time combat information; Based on the real-time military symbol, the military symbol in the initial situation map is updated to obtain the real-time situation map of the battlefield.

8. The automatic situation plotting method according to claim 1, characterized in that: After the initial situation map is updated to obtain the real-time situation map of the battlefield, the method further includes: Based on the large language model, the real-time situation map is analyzed to determine the command decision information of the battlefield.

9. A situation automatic plotting device, characterized in that: include: A key information extraction module is used to extract key information from the initial combat plan text based on a large language model to obtain key combat information in the initial combat plan text, wherein the key combat information includes place name information and military action information; An initial situation map determining module is configured to search a military standard database based on the key combat information to obtain an initial military standard corresponding to the key combat information, and to plot the initial military standard on a military map to obtain an initial situation map of the battlefield; The situation map updating module is used to extract key information from the received real-time battlefield data based on the large language model, and to update the initial situation map based on the extracted real-time key information to obtain a real-time situation map of the battlefield.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the automatic situation plotting method according to any one of claims 1 to 8 is implemented.

Images