Emergency rescue management method and system based on dual-mode interphone
By optimizing the system configuration of dual-mode walkie-talkies, applying natural language processing and machine learning analysis, and combining multi-source data fusion and dynamic communication management, the problem of low communication efficiency in existing emergency rescue systems has been solved. This has enabled efficient allocation of rescue resources and situational awareness, thereby improving the overall efficiency and effectiveness of emergency rescue.
Patent Information
- Application Number
- CN202410983610.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing dual-mode walkie-talkie emergency rescue systems have failed to fully utilize their advantages, making it difficult to achieve rapid fusion of multi-source heterogeneous data, real-time situational awareness, and dynamic communication strategy adjustments. This results in limited communication coverage, low information transmission efficiency, and an inability to meet the rescue needs of large-scale disaster events.
By acquiring the device status and location data of dual-mode walkie-talkies, configuring system parameters, and combining natural language processing and machine learning to analyze on-site voice data, multi-factor optimization calculations and path planning are performed, dynamic grouping and spectrum management are implemented, a collaborative communication network is constructed, and multi-source data fusion and real-time situation map generation are achieved.
It improved system startup efficiency and communication reliability, enabled comprehensive collection and scientific assessment of key event information, optimized the allocation of rescue resources and route planning, ensured communication quality and rescue efficiency in complex environments, and achieved real-time visualization of the rescue situation and updating of emergency knowledge.
Smart Images

Figure CN118940948A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of dual-mode intercoms, and in particular to an emergency rescue management method and system based on dual-mode intercoms. Background Art
[0002] In modern emergency rescue management, efficient communication and information processing systems are essential for the successful implementation of rescue operations. Traditional emergency communication systems often face problems such as limited communication coverage, low information transmission efficiency, and difficulty in data integration, which are particularly prominent in complex rescue environments. Especially in large-scale disaster events, rescuers need to quickly obtain on-site information, coordinate multiple resources, and grasp the rescue situation in real time, but existing communication systems often cannot meet these needs. As a device with both analog and digital communication capabilities, dual-mode walkie-talkies have the advantages of long communication distance, strong anti-interference ability, and stable data transmission.
[0003] However, how to make full use of the advantages of dual-mode walkie-talkies to build an intelligent and efficient emergency rescue management system is still an urgent problem to be solved. At present, most emergency systems based on dual-mode walkie-talkies are still at the level of simple voice communication and basic data transmission, and have failed to fully realize their potential. In addition, in the actual rescue process, how to achieve rapid fusion of multi-source heterogeneous data, how to conduct real-time situational awareness and decision support, and how to dynamically adjust communication strategies to adapt to the changing rescue environment are all challenges faced by existing emergency rescue management systems. Summary of the invention
[0004] The present application provides an emergency rescue management method and system based on a dual-mode walkie-talkie, which is used to improve the efficiency and accuracy of emergency response, as well as the continuous accumulation and optimization of emergency knowledge.
[0005] In a first aspect, the present application provides an emergency rescue management method based on a dual-mode intercom, and the emergency rescue management method based on a dual-mode intercom includes:
[0006] Obtain the device status data and device location data of the dual-mode intercom and configure the system parameters to build the intercom system;
[0007] Acquire the on-site voice data of the rescued party through the intercom system and perform natural language processing to obtain key information of the incident;
[0008] Perform machine learning analysis on the key information of the incident and the real-time GPS positioning data of the rescued party to obtain an initial incident assessment report;
[0009] Perform multi-factor optimization calculation and rescuer path planning on the initial event assessment report to obtain a target dispatch instruction for the rescuer;
[0010] Based on the target dispatch instruction, the real-time location data of the rescuer is obtained for dynamic grouping, a communication group division scheme is obtained, spectrum management and mode switching are performed, and a collaborative communication network is created;
[0011] The multi-source data of the collaborative communication network are fused and processed to obtain a real-time rescue situation map, and trend prediction and report generation are performed to obtain a target event assessment report and an emergency knowledge map.
[0012] In a second aspect, the present application provides an emergency rescue management system based on a dual-mode intercom, and the emergency rescue management system based on a dual-mode intercom includes:
[0013] The acquisition module is used to obtain the device status data and device location data of the dual-mode intercom and configure the system parameters to build the intercom system;
[0014] A processing module, used to obtain the on-site voice data of the rescued party through the intercom system and perform natural language processing to obtain key information of the event;
[0015] An analysis module, used to perform machine learning analysis on the key information of the incident and the real-time GPS positioning data of the rescued party to obtain an initial incident assessment report;
[0016] A planning module, used to perform multi-factor optimization calculation and rescuer path planning on the initial event assessment report to obtain a target dispatch instruction for the rescuer;
[0017] A grouping module, used to obtain the real-time location data of the rescuer based on the target dispatch instruction, perform dynamic grouping, obtain a communication group division plan, perform spectrum management and mode switching, and create a collaborative communication network;
[0018] A generation module is used to fuse and process the multi-source data of the collaborative communication network to obtain a real-time rescue situation map and perform trend prediction and report generation to obtain a target event assessment report and an emergency knowledge map.
[0019] The third aspect of the present application provides an emergency rescue management device based on a dual-mode walkie-talkie, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the emergency rescue management device based on the dual-mode walkie-talkie executes the above-mentioned emergency rescue management method based on the dual-mode walkie-talkie.
[0020] The fourth aspect of the present application provides a computer-readable storage medium, in which instructions are stored. When the computer-readable storage medium is run on a computer, the computer executes the above-mentioned emergency rescue management method based on the dual-mode walkie-talkie.
[0021] In the technical solution provided by this application, rapid initialization of the walkie-talkie system is achieved, and the system startup efficiency and reliability are improved. Combined with GPS positioning data, comprehensive and accurate collection of key information of the event is achieved, and scientific evaluation and prediction of the severity and development trend of the event are achieved. The multi-objective optimization algorithm and real-time road condition data are used to achieve optimal allocation of rescue resources and dynamic path planning, thereby improving rescue efficiency and resource utilization. Through dynamic grouping, spectrum management and mode switching, a highly adaptable collaborative communication network is constructed to ensure communication quality and reliability in complex environments. By utilizing multi-source data fusion, real-time and comprehensive visualization of the rescue situation is achieved. Through knowledge extraction and semantic association of event evaluation reports, the emergency knowledge map is continuously updated and improved, significantly improving the efficiency and effectiveness of the overall rescue process. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0023] Figure 1 A schematic diagram of an embodiment of an emergency rescue management method based on a dual-mode intercom in an embodiment of the present application;
[0024] Figure 2 This is a schematic diagram of an embodiment of an emergency rescue management system based on a dual-mode walkie-talkie in an embodiment of the present application. DETAILED DESCRIPTION
[0025] The embodiment of the present application provides an emergency rescue management method and system based on a dual-mode walkie-talkie. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0026] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1In the embodiment of the present application, an embodiment of the emergency rescue management method based on the dual-mode walkie-talkie includes:
[0027] Step S101, obtaining device status data and device location data of the dual-mode intercom and configuring system parameters to build an intercom system;
[0028] It is understandable that the execution subject of the present application can be an emergency rescue management system based on a dual-mode intercom, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.
[0029] Specifically, the dual-mode intercom is subjected to hardware self-test to obtain device status data, which includes information such as battery power, antenna status, and storage space. According to the device status data, the dual-mode intercom is switched to realize the dual-mode working state of the default analog mode and the standby digital mode, ensuring the flexible response capability of the device under different communication requirements. The dual-mode intercom in the dual-mode working state is connected to the network to establish a data link with the background management system. The device ID and initial location information are uploaded according to the data link to obtain the basic data packet of the device. According to the basic data packet of the device, parameters are issued to generate system configuration data. The system configuration data includes specific parameters such as frequency configuration and contact scheme. After receiving the system configuration data, the device parses these data, and the result of the parsing includes the setting instructions of the working frequency, scan list, and contact group. After the parsing is completed, the communication parameters of the dual-mode intercom are configured according to the obtained setting instructions to complete the initialization of the communication module. The initialized communication module has the ability to work efficiently in different modes. At the same time, the communication module is GPS activated, and the location information of the device is obtained through GPS activation. This location information is helpful for real-time positioning and dispatching of rescue resources during emergency rescue. The device location data is integrated with the device status data to generate comprehensive status information. The comprehensive status information includes not only the current working status and physical status of the device, but also its location information and other important parameters. Based on the comprehensive status information, a dual-mode intercom system is constructed. The system can monitor the status and location of each intercom in real time, and adjust parameters and switch modes as needed to ensure stable and efficient communication support in various emergency rescue scenarios.
[0030] Step S102: Acquire the on-site voice data of the rescued party through the intercom system and perform natural language processing to obtain key information of the event;
[0031] Specifically, an emergency trigger is performed on the intercom system to obtain a voice collection start signal. According to the voice collection start signal, the audio module of the dual-mode intercom is configured to enter a high-fidelity recording state to ensure that the subsequently collected voice data has high clarity and quality. Data is collected from the audio module in the high-fidelity recording state to obtain the on-site voice data of the rescued party. The collected on-site voice data is segmented according to a preset audio segmentation algorithm to obtain multiple voice segments. Each voice segment represents a part of the on-site voice data, which is helpful for subsequent parallel processing. The voice segments are subjected to parallel voice noise reduction processing to remove background noise and obtain clear voice signals. The clear voice signals are subjected to voice recognition and converted into initial text data. The initial text data is subjected to word embedding to obtain a vector representation of the text. The vector representation converts the text into a numerical form that can be calculated and analyzed. The vector representation is subjected to sequence modeling of the attention mechanism to capture the context-related feature representation in the text. The context-related feature representation is subjected to entity extraction to identify key entity information contained in the text, including event type, location, and time. Relationship extraction is performed on key entity information to identify semantic associations between entities and understand the relationships between entities. Semantic reasoning is performed on key entity information and semantic associations between entities to generate structured event descriptions. Structured event descriptions present the full picture of the event in a systematic way, which facilitates rescue personnel to make quick decisions and respond. In order to improve the effectiveness of event descriptions, criticality scores are performed on structured event descriptions to evaluate the relative importance of each event description and obtain the most critical event information.
[0032] Step S103: Perform machine learning analysis on key event information and the real-time GPS positioning data of the rescued party to obtain an initial event assessment report;
[0033] Specifically, the key information of the event is vectorized. The event information described in the text is converted into numerical form for subsequent calculation and analysis. Through feature vectorization, an event feature matrix is obtained, which contains various descriptive features of the event. In order to simplify the calculation and improve the processing efficiency, the event feature matrix is subjected to dimensionality reduction processing to obtain the event representation vector. At the same time, the real-time GPS positioning data of the rescued party is spatiotemporally serialized, and these data are arranged in the order of time and space to form a spatiotemporal feature sequence. The spatiotemporal feature sequence is dynamically encoded to obtain a position dynamic feature vector. The feature vector can reflect the change of the position of the rescued party during the rescue process. The event representation vector and the position dynamic feature vector are multimodally fused, and the two different types of features are integrated to obtain a comprehensive feature vector. Through feature fusion, the static description information of the event is combined with the dynamic location information of the rescued party to provide a more comprehensive event description. The comprehensive feature vector is nonlinearly mapped to obtain the event severity score, which reflects the urgency and potential harm of the event. The comprehensive feature vector is searched for the nearest neighbor according to the historical case database to find the historical event set most similar to the current event. Statistical analysis of a collection of similar historical events is performed to obtain event development trend prediction results, which can help predict the possible evolution direction and severity of the current event. The event severity score is weighted and fused with the event development trend prediction results to obtain an initial risk assessment index. This index combines the severity of the current event and the historical development trend to provide an overall risk assessment result. The initial risk assessment index is spatially interpolated according to the predefined geographic information system to generate a risk distribution heat map. The heat map can intuitively display the risk distribution in different areas, which helps to quickly identify high-risk areas. Threshold segmentation is performed on the risk distribution heat map to divide areas with different risk levels to obtain multi-level dangerous area divisions, which helps rescuers identify and focus on areas that need emergency response most. The multi-level dangerous area divisions are matched according to the emergency plan knowledge base to generate an initial event assessment report. The assessment report integrates the severity, development trend and spatial risk distribution of the event.
[0034] Step S104: Perform multi-factor optimization calculation and rescuer path planning on the initial event assessment report to obtain the rescuer's target dispatch instruction;
[0035] Specifically, the initial event assessment report is analyzed to extract a set of key parameters, including important information such as event type, severity, and geographic location. Based on the extracted set of key parameters, matching analysis is performed to generate an initial resource requirement list. The initial resource requirement list specifies the type and quantity of rescue resources required for the current event. In order to ensure the reasonable allocation of rescue resources, the existing rescue resource database is queried in real time to obtain the current available resource list. On this basis, a multi-objective optimization algorithm is applied to calculate the matching degree between the initial resource requirement list and the available resource list, find the best match between resource demand and resource supply, and obtain the optimal resource allocation plan. The geographic location of the rescuer in the optimal resource allocation plan is extracted to generate a set of rescuer starting positions. According to real-time road condition data, the rescuer starting position set is combined with the event geographic location to construct a road network and generate a multi-level road network map. The multi-level road network map not only reflects the path of each rescuer to the event location, but also takes into account the road network information at different levels, which improves the accuracy and practicality of path planning. The shortest path search is performed on the multi-level road network map to preliminarily determine the rescuer's path plan. In order to ensure the rationality and feasibility of the path plan, traffic flow prediction is carried out, and the initial path plan is dynamically adjusted according to the prediction results, and dynamic path adjustment suggestions are proposed. These suggestions are based on real-time traffic conditions to ensure that rescue vehicles can reach the scene of the incident in the shortest time. Multi-agent collaborative optimization is carried out on the dynamic path adjustment suggestions, and a more optimized target path planning scheme is generated through collaboration and coordination between agents. The target path planning scheme not only considers the shortest distance of the path, but also combines the real-time traffic conditions and the dynamic changes of rescue resources. The target path planning scheme is time-serialized to generate a task timeline. The task timeline clarifies the time nodes and execution order of each rescue task, and provides a time arrangement for the rescue operation. According to the communication protocol specification, the task timeline and the optimal resource allocation plan are encoded to generate the target scheduling instructions of the rescue party. The target scheduling instructions stipulate the action path, resource allocation and time arrangement of each rescue unit to ensure the efficient and orderly conduct of the rescue operation.
[0036] Step S105: based on the target dispatch instruction, the real-time location data of the rescuers is obtained for dynamic grouping, a communication group division scheme is obtained, spectrum management and mode switching are performed, and a collaborative communication network is created;
[0037] Specifically, the target dispatch instruction is parsed to obtain a rescuer identification list. According to the identification list, the rescuer is located by walkie-talkie to obtain the rescuer's real-time location data. The obtained real-time location data is clustered and analyzed to preliminarily determine the rescuer's grouping situation and generate the initial grouping result. In order to ensure the rationality and effectiveness of the grouping, the initial grouping result is optimized to obtain the final communication group division scheme. The optimized communication group division scheme can maximize the communication efficiency and resource utilization. The communication group division scheme is evaluated for communication requirements to determine the bandwidth requirements and communication quality requirements of each group. Through the evaluation, the resources and quality assurance required by different groups in the communication process are understood. According to the available spectrum resources, the bandwidth requirements and communication quality requirements are matched to generate an initial spectrum allocation scheme to ensure that each group can obtain sufficient spectrum resources to meet its communication needs. The initial spectrum allocation scheme is analyzed for interference to obtain a potential interference matrix. By analyzing the potential interference matrix, possible interference sources and interference situations are identified. The initial spectrum allocation scheme is reallocated to generate a target spectrum management scheme to minimize interference and ensure the stability and reliability of communication. The communication mode of the target spectrum management scheme is evaluated to determine the optimal communication mode for each group. Different communication modes are suitable for different communication needs and environments. Selecting the optimal communication mode is the key to improving communication efficiency. The optimal communication mode is switched to generate a mode switching instruction set. These instructions guide each group to perform corresponding mode switching to adapt to the current communication needs and environmental changes. The network topology is constructed for the mode switching instruction set to generate the initial communication network. The initial communication network is optimized through the adaptive routing algorithm to generate the final collaborative communication network. The collaborative communication network can be adaptively adjusted according to the actual situation to ensure unimpeded communication between groups, reasonable resource allocation, and stable communication quality.
[0038] Step S106: Fusion processing is performed on multi-source data of the collaborative communication network to obtain a real-time rescue situation map, and trend prediction and report generation are performed to obtain a target event assessment report and an emergency knowledge map.
[0039] Specifically, the multi-source data of the collaborative communication network is cleaned to remove noise and redundant information to obtain a standardized data set. Feature extraction is performed on the standardized data set to extract a fusion feature vector that can reflect the characteristics and development trends of the event. The spatiotemporal pattern of the event evolution is identified by performing spatiotemporal correlation analysis on the fusion feature vector. The spatiotemporal pattern describes the changing law of the event in time and space, and is an important basis for prediction and decision-making. The spatiotemporal pattern is semantically mapped to convert the spatiotemporal pattern into an initial situation description. The initial situation description is rendered by a geographic information system to generate a visual real-time rescue situation map. The situation map can reflect the progress and on-site situation of the current rescue operation in real time. Trend extrapolation is performed on the real-time rescue situation map to predict the development trend of events in the short term. Through trend extrapolation, the short-term situation prediction results are obtained to help rescuers prepare in advance and adjust the rescue strategy in time. Based on the short-term situation prediction results, risk assessment is performed to identify potential risk points and generate a list of potential risk points. In order to formulate an effective response strategy, the list of potential risk points is matched and response strategy suggestions are proposed. The response strategy suggestions are comprehensively analyzed with the real-time rescue situation map to generate a target event evaluation report. The report integrates the dynamic changes of the event, risk assessment and response strategies. After obtaining the target event assessment report, knowledge extraction is performed to extract key concepts and relationship sets. These concepts and relationships describe the main characteristics of the event and the relationship between various elements. The key concepts and relationship sets are semantically associated to generate an emergency knowledge graph. The emergency knowledge graph displays the knowledge structure related to the event in a graphical form, helping rescue personnel to better understand the background and development laws of the event and provide knowledge support for emergency decision-making.
[0040] In the embodiment of the present application, rapid initialization of the intercom system is achieved, and the system startup efficiency and reliability are improved. Combined with GPS positioning data, comprehensive and accurate collection of key information of the event is achieved, and scientific evaluation and prediction of the severity and development trend of the event are achieved. By adopting multi-objective optimization algorithms and real-time road condition data, optimal allocation and dynamic path planning of rescue resources are achieved, and rescue efficiency and resource utilization are improved. Through dynamic grouping, spectrum management and mode switching, a highly adaptable collaborative communication network is constructed to ensure communication quality and reliability in complex environments. By utilizing multi-source data fusion, real-time and comprehensive visualization of the rescue situation is achieved. Through knowledge extraction and semantic association of event evaluation reports, the emergency knowledge map is continuously updated and improved, which significantly improves the efficiency and effectiveness of the overall rescue process.
[0041] In a specific embodiment, the process of executing step S101 may specifically include the following steps:
[0042] (1) Perform hardware self-test on the dual-mode walkie-talkie to obtain device status data, which includes battery power, antenna status, and storage space;
[0043] (2) Switching the dual-mode intercom mode according to the device status data to obtain the dual-mode working states of the default analog mode and the standby digital mode;
[0044] (3) Connect the dual-mode walkie-talkie in dual-mode working state to the network, obtain a data link with the background management system, and upload the device ID and initial location information according to the data link to obtain the basic data packet of the device;
[0045] (4) System parameters are issued according to the basic data packet of the device to obtain system configuration data, which includes frequency configuration and contact plan;
[0046] (5) Parse the system configuration data to obtain the setting instructions for the working frequency, scan list, and contact group;
[0047] (6) According to the setting instructions, the communication parameters of the dual-mode walkie-talkie are configured to obtain an initialized communication module, and the GPS of the communication module is activated to obtain the device location data;
[0048] (7) Integrate the device location data and the device status data to obtain comprehensive status information; and construct a dual-mode intercom system based on the comprehensive status information.
[0049] Specifically, the device is comprehensively checked to ensure that all its components are working properly. Hardware self-check includes checking the battery power, antenna status and storage space. Battery power detection is completed by a voltage sensor, which can monitor the voltage level of the battery in real time. Antenna status detection is achieved through an antenna matching circuit. When the antenna matches the transmitting circuit well, the antenna status is considered normal. Storage space detection is completed by a built-in storage controller, which checks the available space of the storage chip to ensure that it is sufficient to store the data to be generated. After completing the hardware self-check, a set of device status data is obtained, which includes information such as battery power, antenna status and storage space. According to the device status data, the dual-mode walkie-talkie is switched to a mode, thereby selecting between analog mode and digital mode to ensure that the device can work stably in different environments. In the dual-mode working state, the dual-mode walkie-talkie is connected to the network to establish a data link with the background management system. The establishment of the data link is achieved through the built-in network module of the walkie-talkie, which can automatically search and connect to an available network. After the network connection is established, the walkie-talkie uploads the device ID and initial location information to the background management system according to the data link to form a basic data packet of the device. The basic device data package includes the unique identification (ID) of the device and the current geographic location information (obtained through GPS). The system parameters are issued according to the basic device data package. The system parameters include specific configuration data such as frequency configuration and contact scheme. Frequency configuration refers to the radio frequency used when the walkie-talkie works, and the contact scheme defines the specific way to communicate with other devices. These system configuration data are sent to the walkie-talkie through the data link and parsed by the walkie-talkie. The process of parsing system configuration data is to convert the data into specific operation instructions. According to the setting instructions obtained by parsing, the communication parameters of the dual-mode walkie-talkie are configured. The communication parameter configuration includes operations such as setting the working frequency, updating the scan list, and configuring the contact group. Through these configurations, the communication module of the walkie-talkie can be initialized to ensure that it can work normally in different modes. At the same time, the communication module is activated by GPS to obtain the location information of the device. After GPS activation, the walkie-talkie can obtain and update its geographic location information in real time to ensure the accuracy of the location information. After completing the above operations, the device location data and device status data are integrated to form comprehensive status information. The comprehensive status information includes the current geographic location of the device, as well as detailed information such as the working status and communication status of the device. Based on the comprehensive status information, a complete dual-mode intercom system is built. The system can monitor the status and location of each device in real time, and adjust parameters and switch modes as needed to ensure that the device can provide stable and efficient communication support in various emergency situations. For example, in an emergency rescue scenario, when the rescue team arrives at the scene, the system will automatically switch to a communication mode suitable for the current environment to ensure smooth communication between the rescue teams.At the same time, the system can dynamically adjust the frequency configuration and contact plan based on the real-time geographic location information to ensure the reliability and stability of the communication network.
[0050] In a specific embodiment, the process of executing step S102 may specifically include the following steps:
[0051] (1) Emergency triggering of the intercom system to obtain a voice collection start signal, and configuring the audio module of the dual-mode intercom according to the voice collection start signal to obtain a high-fidelity recording state;
[0052] (2) collecting data from the audio module in a high-fidelity recording state to obtain on-site voice data of the rescued party, and segmenting the on-site voice data according to a preset audio segmentation algorithm to obtain multiple voice segments;
[0053] (3) performing speech noise reduction on multiple speech segments in parallel to obtain clear speech signals, and performing speech recognition on the clear speech signals to obtain initial text data;
[0054] (4) Perform word embedding on the initial text data to obtain a vector representation of the text, and perform sequence modeling of the attention mechanism on the vector representation of the text to obtain a context-related feature representation;
[0055] (5) Entity extraction is performed on the context-related feature representation to obtain key entity information including event type, location, and time, and relationship extraction is performed on the key entity information to obtain semantic associations between entities;
[0056] (6) Perform semantic reasoning on key entity information and semantic associations between entities to obtain a structured event description, and perform criticality scoring on the structured event description to obtain key event information.
[0057] Specifically, the intercom system is triggered in an emergency. It is started by an emergency command or a trigger mechanism. When the system receives the emergency trigger signal, a voice collection start signal is generated. Through the start signal, the audio module of the intercom will be configured to the best state to ensure that it can record with high fidelity quality. After the audio module of the dual-mode intercom enters the high-fidelity recording state, data collection is performed to obtain the on-site voice data of the rescued party. The on-site voice data is segmented according to the preset audio segmentation algorithm, and the long voice recording is divided into multiple shorter voice segments. Multiple voice segments are processed for voice noise reduction in parallel. The on-site environment is usually noisy, and the voice signal often contains a lot of noise. Through noise reduction processing, a clearer voice signal is obtained. The clear voice signal is recognized to generate initial text data. The speech recognition system uses a machine learning algorithm to accurately convert speech into text. The initial text data is processed by word embedding. Word embedding is a method of converting words in a text into a vector representation. In this method, the text is converted into a numerical form that can be calculated and processed. The vector representation of the text is represented by a sequence modeling of the attention mechanism. The attention mechanism is a deep learning method that can capture the relationship between contexts in a text and generate context-dependent feature representations. These feature representations can better reflect the actual meaning of the text. Entity extraction is performed through context-dependent feature representations to identify key entity information contained in the text. This information includes important content such as event type, location, and time. Relationship extraction is performed on key entity information to identify semantic associations between entities. Through relationship extraction, the relationship between different entities is understood, so as to better understand the overall situation of the event. Semantic reasoning is performed on key entity information and semantic associations between entities. Semantic reasoning is to structure the extracted information through logical reasoning and rule matching to generate structured event descriptions. Criticality scores are performed on structured event descriptions to evaluate the relative importance of each description and generate event key information. This key information can help rescuers quickly understand the core situation of the event and make timely decisions.
[0058] In a specific embodiment, the process of executing step S103 may specifically include the following steps:
[0059] (1) Feature vectorization is performed on the key information of the event to obtain an event feature matrix, and the event feature matrix is reduced in dimension to obtain an event representation vector;
[0060] (2) The real-time GPS positioning data of the rescued party is serialized in time and space to obtain a time and space feature sequence, and the time and space feature sequence is dynamically encoded to obtain a position dynamic feature vector;
[0061] (3) Perform multimodal feature fusion on the event representation vector and the position dynamic feature vector to obtain a comprehensive feature vector, and perform nonlinear mapping on the comprehensive feature vector to obtain the event severity score;
[0062] (4) Performing a nearest neighbor search on the comprehensive feature vector based on the historical case database to obtain a set of similar historical events, and performing statistical analysis on the set of similar historical events to obtain a prediction result of the event development trend;
[0063] (5) Perform weighted fusion on the event severity score and the event development trend prediction results to obtain an initial risk assessment index, and perform spatial interpolation on the initial risk assessment index according to a predefined geographic information system to obtain a risk distribution heat map;
[0064] (6) Threshold segmentation is performed on the risk distribution heat map to obtain multi-level dangerous area divisions, and the multi-level dangerous area divisions are matched according to the emergency plan knowledge base to obtain an initial event assessment report.
[0065] Specifically, the key information of the event is vectorized and converted into numerical form for subsequent calculation and processing. The key information of the event may include the type, location, time, participants, etc. of the event. Through natural language processing technology, the text information is converted into a feature vector to form an event feature matrix, which contains various descriptive features of the event. The event feature matrix is subjected to dimensionality reduction processing to reduce the dimension of the feature while retaining the main information to obtain the event representation vector. At the same time, the real-time GPS positioning data of the rescued party is spatiotemporally serialized, and the real-time GPS positioning data is arranged in the order of time and space to form a spatiotemporal feature sequence. The spatiotemporal feature sequence can reflect the dynamic changes in the position of the rescued party during the rescue process. The spatiotemporal feature sequence is dynamically encoded to obtain a position dynamic feature vector. Multimodal feature fusion is performed on the event representation vector and the position dynamic feature vector, and the static description information of the event is combined with the dynamic position information of the rescued party to generate a comprehensive feature vector. The comprehensive feature vector can comprehensively reflect the various characteristics of the event and provide a basis for the event severity score. The comprehensive feature vector is nonlinearly mapped to obtain the event severity score. Through nonlinear functions, the comprehensive feature vector is mapped into a scoring range to reflect the urgency and potential harm of the event. In order to predict the development trend of the event, the comprehensive feature vector is searched for the nearest neighbor according to the historical case database to find the historical event set most similar to the current event. Through statistical analysis of the similar historical event set, the prediction results of the event development trend are obtained. These results can help predict the possible evolution direction and severity of the current event. The event severity score is weighted and fused with the event development trend prediction results to generate an initial risk assessment index. The initial risk assessment index combines the severity of the current event and the historical development trend to provide an overall risk assessment result. The initial risk assessment index is spatially interpolated according to the predefined geographic information to generate a risk distribution heat map. The heat map can intuitively show the risk distribution of different areas and help to quickly identify high-risk areas. The risk distribution heat map is threshold segmented to divide the areas with different risk levels to obtain multi-level dangerous area divisions. These divisions can help rescuers identify and focus on the areas that need emergency response most. The multi-level dangerous area divisions are matched according to the emergency plan knowledge base to generate an initial event assessment report. The emergency plan knowledge base contains plans and response measures for different risk levels, and through matching, provides specific response strategies and suggestions for each hazardous area.
[0066] In a specific embodiment, the process of executing step S104 may specifically include the following steps:
[0067] (1) Performing data analysis on the initial incident assessment report to obtain a set of key parameters, including: incident type, severity, and geographic location;
[0068] (2) Match the key parameter set to obtain an initial resource requirement list;
[0069] (3) Conduct real-time query on the existing rescue resource database to obtain a list of available resources, and calculate the matching degree between the initial resource requirement list and the available resource list based on a multi-objective optimization algorithm to obtain the optimal resource allocation plan;
[0070] (4) Extracting the geographic location of the rescuers in the optimal resource allocation scheme to obtain the rescuers’ starting location set, and constructing a road network for the rescuers’ starting location set and geographic location based on real-time traffic data to obtain a multi-level road network map;
[0071] (5) Search for the shortest path on the multi-level road network graph to obtain the rescue party's initial path plan, and predict the traffic flow of the initial path plan to obtain dynamic path adjustment suggestions;
[0072] (6) Perform multi-agent collaborative optimization on the dynamic path adjustment suggestions to obtain the target path planning solution, and then perform time serialization on the target path planning solution to obtain the task timeline;
[0073] (7) Encode the task timeline and optimal resource allocation plan according to the communication protocol specifications to obtain the rescue party’s target scheduling instructions.
[0074] Specifically, extract important information from the initial incident assessment report. This information usually includes the type of incident, severity, and geographic location. By analyzing and processing the report content, a set of key parameters containing this information is formed. According to the key parameter set, match the resource requirements of the incident. Determine the type and quantity of resources required for the type of incident, as well as the urgency of the resources required by the severity of the incident. Generate an initial resource requirement list based on these parameters. The initial resource requirement list clarifies the various types of resources required for the current incident. Perform real-time queries on the existing rescue resource database to obtain the current list of available resources. Match the initial resource requirement list with the list of available resources. Through a multi-objective optimization algorithm, calculate the degree of match between the two and obtain the optimal resource allocation plan. Multi-objective optimization algorithms usually take into account multiple factors, such as the type, quantity, location, and urgency of resources, and comprehensively consider these factors to ensure the rationality and effectiveness of resource allocation. The calculation of the optimal resource allocation plan can be expressed by the following formula:
[0075]
[0076] Among them, wi is the weight of the ith factor, f i is the matching function of the i-th factor, D is the initial resource demand list, and R is the list of available resources. The geographic location of the rescuer in the optimal resource allocation plan is extracted. Assuming that the rescuer in the optimal resource allocation plan is A, the rescuer's starting location set P can be obtained by extracting its geographic location. According to the real-time traffic data, the rescuer's starting location set is combined with the geographic location of the event to construct a road network. By analyzing the real-time traffic data, a multi-level road network graph N is generated, which reflects the traffic conditions at different road levels. After the multi-level road network graph is generated, a shortest path search is performed to determine the best path for the rescuer from the starting position to the event location. After the initial path plan is determined, traffic flow prediction is performed. By analyzing the real-time traffic data, the traffic conditions in the future are predicted, and dynamic path adjustment suggestions are proposed. The calculation of dynamic path adjustment suggestions can be expressed by the following formula:
[0077]
[0078] Among them, t j is the estimated travel time of the jth path, and P′ is the adjusted path set. Multi-agent collaborative optimization is performed on the dynamic path adjustment suggestions. Multi-agent collaborative optimization can find a more optimized target path planning scheme through collaboration and coordination between agents. The target path planning scheme not only considers the shortest distance of the path, but also combines the real-time traffic conditions and the dynamic changes of rescue resources. The target path planning scheme is time-serialized to generate a task timeline. The task timeline clarifies the time nodes and execution order of each rescue task, and provides a time schedule for the rescue operation. According to the communication protocol specification, the task timeline and the optimal resource allocation plan are encoded to generate the target scheduling instructions of the rescue party. The target scheduling instructions specify the action path, resource allocation and time schedule of each rescue unit to ensure the efficient and orderly conduct of the rescue operation.
[0079] In a specific embodiment, the process of executing step S105 may specifically include the following steps:
[0080] (1) Parse the target dispatch instruction to obtain a rescuer identification list, and locate the intercom according to the rescuer identification list to obtain the rescuer's real-time location data;
[0081] (2) Perform cluster analysis on the real-time location data to obtain an initial grouping result, and optimize the initial grouping result to obtain a communication group division scheme;
[0082] (3) Evaluate the communication requirements of the communication group division scheme to obtain the bandwidth requirements and communication quality requirements of each group, and match the bandwidth requirements and communication quality requirements according to the available spectrum resources to obtain an initial spectrum allocation scheme;
[0083] (4) performing interference analysis on the initial spectrum allocation scheme to obtain a potential interference matrix, and performing spectrum reallocation on the potential interference matrix to obtain a target spectrum management scheme;
[0084] (5) evaluating the communication mode of the target spectrum management scheme to obtain the optimal communication mode of each group, and performing mode switching on the optimal communication mode to obtain a mode switching instruction set;
[0085] (6) Constructing a network topology for the mode switching instruction set to obtain an initial communication network, and optimizing the initial communication network according to an adaptive routing algorithm to obtain a collaborative communication network.
[0086] Specifically, the target dispatch instruction is parsed to extract the walkie-talkie identifiers of all rescuers, which represent specific rescuers or equipment. The extracted rescuer identifier list is used to obtain the real-time location data of each rescuer through walkie-talkie positioning technology. These location data can be GPS coordinates or other positioning information, and the current status of the rescuer is reflected by updating these data in real time. The acquired real-time location data is clustered and the rescuers are preliminarily grouped according to their geographical location and rescue mission. Cluster analysis can be implemented by K-means algorithm. After cluster analysis, the initial grouping scheme is obtained, and optimization is performed according to the scheme. The optimization process considers the specific rescue mission of each group, the coordination of members and the communication requirements, and obtains the communication group division scheme. The generation of the communication group division scheme involves complex multi-factor optimization, which needs to ensure that the members in each group can communicate efficiently and the interference between different groups is minimized. After obtaining the communication group division scheme, the communication requirements of each group are evaluated, including calculating the bandwidth requirements and communication quality requirements of each group. The bandwidth requirements and communication quality requirements are matched according to the available spectrum resources. The matching process can be solved by linear programming to obtain the initial spectrum allocation scheme, which aims to maximize the use of available spectrum resources and meet the communication needs of each group. The initial spectrum allocation scheme is subjected to interference analysis to obtain a potential interference matrix, which describes the interference relationship between different groups. According to the interference matrix, spectrum reallocation is performed to minimize interference and obtain the target spectrum management scheme. Based on the target spectrum management scheme, the communication mode of each group is evaluated to determine the most suitable communication mode for each group, ensure the stability and efficiency of communication, and obtain the optimal communication mode set. According to the evaluation results, a mode switching instruction set is generated, which is used to guide each group to switch the communication mode. After the mode switching instruction set is generated, the network topology is constructed according to these instructions to obtain the initial communication network. The construction of the network topology takes into account the geographical location and communication needs of each group to ensure the coverage and connection quality of the communication network. In order to optimize the communication network, an adaptive routing algorithm is applied to dynamically adjust the initial communication network. The adaptive routing algorithm adjusts the network path according to the real-time network status and communication needs to improve the communication efficiency and stability. The optimized communication network is the cooperative communication network.
[0087] In a specific embodiment, the process of executing step S106 may specifically include the following steps:
[0088] (1) Perform data cleaning on the multi-source data of the collaborative communication network to obtain a standardized data set, and perform feature extraction on the standardized data set to obtain a fused feature vector;
[0089] (2) Performing spatiotemporal correlation analysis on the fused feature vector to obtain the spatiotemporal pattern of event evolution, and semantically mapping the spatiotemporal pattern to obtain the initial situation description;
[0090] (3) Rendering the initial situation description using a geographic information system to obtain a visualized real-time rescue situation map, and then extrapolating the trend of the real-time rescue situation map to obtain short-term situation prediction results;
[0091] (4) Conduct risk assessment on the short-term situation forecast results to obtain a list of potential risk points, and match the list of potential risk points to obtain response strategy recommendations;
[0092] (5) Comprehensively analyze the response strategy recommendations and real-time rescue situation map to obtain a target event assessment report;
[0093] (6) Perform knowledge extraction on the target event assessment report to obtain key concepts and relationship sets, and perform semantic association on the key concepts and relationship sets to obtain an emergency knowledge graph.
[0094] Specifically, data cleaning is performed on multi-source data to remove noise, correct errors and fill missing values to ensure data quality and consistency, and obtain a standardized data set. Feature extraction is performed on the standardized data set to extract key attributes that can represent data characteristics from the data set. Statistical methods and machine learning algorithms are used to extract representative features. Spatiotemporal correlation analysis is performed on the fused feature vector to identify potential spatiotemporal patterns in the data. By building a spatiotemporal correlation model, the evolution laws of events in time and space are identified. These laws are called spatiotemporal patterns of event evolution. Semantic mapping is performed on spatiotemporal patterns to convert abstract spatiotemporal patterns into easy-to-understand descriptions. Natural language processing technology is used to convert the pattern into an initial situation description. The initial situation description can include the cause of the event, the current state and possible evolution trends. The initial situation description is rendered by a geographic information system to generate a visual real-time rescue situation map. The situation map shows the geographical location and evolution of the event in the form of a map, so that rescuers can intuitively understand the current rescue situation. Trend extrapolation is performed on the real-time rescue situation map to predict the development trend of the event in the short term. By analyzing the current situation and historical data, the time series prediction model is used for extrapolation. Assuming that f(t) represents the time series prediction model and the current time is t0, the short-term situation prediction result can be expressed as f(t0+Δt), where Δt is the prediction time step. After obtaining the short-term situation prediction result, perform risk assessment on it, identify potential risk points, and evaluate their possible impact. By analyzing the prediction results, a list of potential risk points is generated. Assuming that R iRepresents the i-th risk point, and the risk assessment model can be expressed as R = g(f(t0+Δt)), where g is the risk assessment function. Match the list of potential risk points and generate response strategy recommendations. By consulting the emergency plan knowledge base, find the corresponding response strategy for each risk point. The response strategy recommendations include specific action plans and resource allocation plans to ensure that measures can be taken quickly when risk points appear. After completing the response strategy recommendations, they are comprehensively analyzed with the real-time rescue situation map. Through comprehensive analysis, a target event assessment report is generated. The target event assessment report comprehensively summarizes the current situation, prediction results, risk assessment and response strategies of the event, providing a comprehensive reference for decision makers. Knowledge extraction is performed on the target event assessment report, and key concepts and relationship sets are extracted from the report. Semantic association is performed on key concepts and relationship sets to generate an emergency knowledge graph. The emergency knowledge graph graphically displays the relationship between the knowledge structure and elements related to the event, helping rescue personnel to better understand the event background and evolution law.
[0095] The above describes the emergency rescue management method based on the dual-mode intercom in the embodiment of the present application. The following describes the emergency rescue management system based on the dual-mode intercom in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of an emergency rescue management system based on a dual-mode intercom includes:
[0096] The acquisition module 201 is used to acquire the device status data and device location data of the dual-mode intercom and configure the system parameters to build the intercom system;
[0097] The processing module 202 is used to obtain the on-site voice data of the rescued party through the intercom system and perform natural language processing to obtain key information of the incident;
[0098] Analysis module 203, used to perform machine learning analysis on key event information and real-time GPS positioning data of the rescued party to obtain an initial event assessment report;
[0099] Planning module 204, used to perform multi-factor optimization calculation and rescuer path planning on the initial event assessment report to obtain the rescuer's target dispatch instruction;
[0100] The grouping module 205 is used to obtain the real-time location data of the rescuers based on the target dispatch instruction, perform dynamic grouping, obtain the communication group division plan, perform spectrum management and mode switching, and create a cooperative communication network;
[0101] The generation module 206 is used to perform fusion processing on the multi-source data of the collaborative communication network, obtain a real-time rescue situation map and perform trend prediction and report generation, and obtain a target event assessment report and an emergency knowledge map.
[0102] Through the collaborative cooperation of the above components, the walkie-talkie system is quickly initialized, and the system startup efficiency and reliability are improved. Combined with GPS positioning data, comprehensive and accurate collection of key information of the incident is achieved, and scientific evaluation and prediction of the severity and development trend of the incident are achieved. The multi-objective optimization algorithm and real-time road condition data are used to achieve the optimal allocation of rescue resources and dynamic path planning, thereby improving rescue efficiency and resource utilization. Through dynamic grouping, spectrum management and mode switching, a highly adaptable collaborative communication network is constructed to ensure the communication quality and reliability in complex environments. By using multi-source data fusion, real-time and comprehensive visualization of the rescue situation is achieved. Through knowledge extraction and semantic association of event evaluation reports, the emergency knowledge map is continuously updated and improved, which significantly improves the efficiency and effectiveness of the overall rescue process.
[0103] The present application also provides an emergency rescue management device based on a dual-mode walkie-talkie, and the emergency rescue management device based on the dual-mode walkie-talkie includes a memory and a processor, and the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the steps of the emergency rescue management method based on the dual-mode walkie-talkie in the above-mentioned embodiments.
[0104] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the emergency rescue management method based on a dual-mode walkie-talkie.
[0105] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0106] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.
[0107] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application 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 embodiments of the present application.
Claims
1. An emergency rescue management method based on a dual-mode intercom, characterized in that: The method comprises: Obtain the device status data and device location data of the dual-mode intercom and configure the system parameters to build the intercom system; Acquire the on-site voice data of the rescued party through the intercom system and perform natural language processing to obtain key information of the incident; Perform machine learning analysis on the key information of the incident and the real-time GPS positioning data of the rescued party to obtain an initial incident assessment report; Perform multi-factor optimization calculation and rescuer path planning on the initial event assessment report to obtain a target dispatch instruction for the rescuer; Based on the target dispatch instruction, the real-time location data of the rescuer is obtained for dynamic grouping, a communication group division scheme is obtained, spectrum management and mode switching are performed, and a collaborative communication network is created; The multi-source data of the collaborative communication network are fused and processed to obtain a real-time rescue situation map, and trend prediction and report generation are performed to obtain a target event assessment report and an emergency knowledge map.
2. The emergency rescue management method based on dual-mode intercom according to claim 1 is characterized in that: The method of obtaining device status data and device location data of the dual-mode intercom and configuring system parameters to build an intercom system includes: Perform hardware self-check on the dual-mode walkie-talkie to obtain device status data, which includes battery power, antenna status and storage space; Switching the dual-mode intercom mode according to the device status data to obtain a dual-mode working state of a default analog mode and a standby digital mode; Connecting the dual-mode intercom in the dual-mode working state to the network to obtain a data link with the background management system, and uploading the device ID and initial location information according to the data link to obtain a basic data packet of the device; Sending system parameters according to the device basic data packet to obtain system configuration data, wherein the system configuration data includes frequency configuration and contact plan; Parsing the system configuration data to obtain setting instructions for working frequency, scanning list and contact group; According to the setting instruction, communication parameters of the dual-mode intercom are configured to obtain an initialized communication module, and GPS is activated on the communication module to obtain device location data; The device location data and the device status data are integrated to obtain comprehensive status information; and an intercom system of the dual-mode intercom is constructed according to the comprehensive status information.
3. The emergency rescue management method based on dual-mode intercom according to claim 2 is characterized in that: The on-site voice data of the rescued party is obtained through the intercom system and natural language processing is performed to obtain key information of the event, including: Performing emergency triggering on the intercom system to obtain a voice collection start signal, and configuring the audio module of the dual-mode intercom according to the voice collection start signal to obtain a high-fidelity recording state; Collecting data from the audio module in the high-fidelity recording state to obtain on-site voice data of the rescued party, and segmenting the on-site voice data according to a preset audio segmentation algorithm to obtain multiple voice segments; Performing speech noise reduction on the multiple speech segments in parallel to obtain clear speech signals, and performing speech recognition on the clear speech signals to obtain initial text data; Performing word embedding on the initial text data to obtain a vector representation of the text, and performing sequence modeling of an attention mechanism on the vector representation of the text to obtain a context-related feature representation; Performing entity extraction on the context-related feature representation to obtain key entity information including event type, location, and time, and performing relationship extraction on the key entity information to obtain semantic associations between entities; Semantic reasoning is performed on the key entity information and the semantic associations between the entities to obtain a structured event description, and a criticality score is performed on the structured event description to obtain event key information.
4. The emergency rescue management method based on dual-mode intercom according to claim 1 is characterized in that: The machine learning analysis of the key information of the incident and the real-time GPS positioning data of the rescued party is performed to obtain an initial incident assessment report, including: Performing feature vectorization on the key information of the event to obtain an event feature matrix, and performing dimension reduction on the event feature matrix to obtain an event representation vector; Performing spatiotemporal serialization on the real-time GPS positioning data of the rescued party to obtain a spatiotemporal feature sequence, and dynamically encoding the spatiotemporal feature sequence to obtain a position dynamic feature vector; Performing multimodal feature fusion on the event representation vector and the position dynamic feature vector to obtain a comprehensive feature vector, and performing nonlinear mapping on the comprehensive feature vector to obtain an event severity score; Performing a nearest neighbor search on the comprehensive feature vector according to the historical case database to obtain a set of similar historical events, and performing statistical analysis on the set of similar historical events to obtain a prediction result of the event development trend; Performing weighted fusion on the event severity score and the event development trend prediction result to obtain an initial risk assessment index, and performing spatial interpolation on the initial risk assessment index according to a predefined geographic information system to obtain a risk distribution heat map; Threshold segmentation is performed on the risk distribution heat map to obtain multi-level dangerous area divisions, and the multi-level dangerous area divisions are matched according to the emergency plan knowledge base to obtain an initial event assessment report.
5. The emergency rescue management method based on dual-mode intercom according to claim 4 is characterized in that: The performing of multi-factor optimization calculation and rescuer path planning on the initial event assessment report to obtain a target dispatch instruction for the rescuer includes: Performing data analysis on the initial event assessment report to obtain a key parameter set, wherein the key parameter set includes: event type, severity, and geographic location; Matching the key parameter set to obtain an initial resource requirement list; Conducting real-time query on the existing rescue resource database to obtain a list of available resources, and calculating the matching degree between the initial resource requirement list and the available resource list according to a multi-objective optimization algorithm to obtain an optimal resource allocation plan; Extracting the geographic locations of the rescuers in the optimal resource allocation scheme to obtain a set of rescuer starting locations, and constructing a road network for the rescuer starting location set and the geographic locations based on real-time traffic data to obtain a multi-level road network map; Performing a shortest path search on the multi-level road network diagram to obtain an initial path plan for the rescue party, and performing traffic flow prediction on the initial path plan to obtain dynamic path adjustment suggestions; Perform multi-agent collaborative optimization on the dynamic path adjustment suggestion to obtain a target path planning solution, and perform time serialization on the target path planning solution to obtain a task timeline; The task timeline and the optimal resource allocation scheme are encoded according to the communication protocol specification to obtain the target scheduling instruction of the rescuer.
6. The emergency rescue management method based on dual-mode intercom according to claim 1 is characterized in that: The real-time location data of the rescuer is obtained based on the target dispatch instruction for dynamic grouping, a communication group division scheme is obtained, spectrum management and mode switching are performed, and a cooperative communication network is created, including: Parsing the target dispatch instruction to obtain a rescuer identification list, and performing intercom positioning according to the rescuer identification list to obtain real-time location data of the rescuer; Performing cluster analysis on the real-time location data to obtain an initial grouping result, and optimizing the initial grouping result to obtain a communication group division scheme; Performing communication demand evaluation on the communication group division scheme to obtain bandwidth demand and communication quality requirement of each group, and matching the bandwidth demand and the communication quality requirement according to available spectrum resources to obtain an initial spectrum allocation scheme; Performing interference analysis on the initial spectrum allocation scheme to obtain a potential interference matrix, and performing spectrum reallocation on the potential interference matrix to obtain a target spectrum management scheme; Performing communication mode evaluation on the target spectrum management scheme to obtain an optimal communication mode for each group, and performing mode switching on the optimal communication mode to obtain a mode switching instruction set; A network topology is constructed for the mode switching instruction set to obtain an initial communication network, and the initial communication network is optimized according to an adaptive routing algorithm to obtain a cooperative communication network.
7. The emergency rescue management method based on dual-mode intercom according to claim 1 is characterized in that: The multi-source data of the collaborative communication network is fused and processed to obtain a real-time rescue situation map, and trend prediction and report generation are performed to obtain a target event assessment report and an emergency knowledge map, including: Performing data cleaning on multi-source data of the collaborative communication network to obtain a standardized data set, and performing feature extraction on the standardized data set to obtain a fused feature vector; Performing spatiotemporal correlation analysis on the fused feature vector to obtain the spatiotemporal pattern of event evolution, and performing semantic mapping on the spatiotemporal pattern to obtain an initial situation description; Performing geographic information system rendering on the initial situation description to obtain a visualized real-time rescue situation map, and performing trend extrapolation on the real-time rescue situation map to obtain a short-term situation prediction result; Performing risk assessment on the short-term situation forecast results to obtain a list of potential risk points, and matching the list of potential risk points to obtain response strategy recommendations; Comprehensively analyzing the response strategy suggestions and the real-time rescue situation map to obtain a target event assessment report; Knowledge extraction is performed on the target event assessment report to obtain key concepts and relationship sets, and semantic association is performed on the key concepts and the relationship sets to obtain an emergency knowledge graph.
8. An emergency rescue management system based on dual-mode intercom, characterized in that: Used to execute the emergency rescue management method based on a dual-mode intercom according to any one of claims 1 to 7, the system comprises: The acquisition module is used to obtain the device status data and device location data of the dual-mode intercom and configure the system parameters to build the intercom system; A processing module, used to obtain the on-site voice data of the rescued party through the intercom system and perform natural language processing to obtain key information of the event; An analysis module, used to perform machine learning analysis on the key information of the incident and the real-time GPS positioning data of the rescued party to obtain an initial incident assessment report; A planning module, used to perform multi-factor optimization calculation and rescuer path planning on the initial event assessment report to obtain a target dispatch instruction for the rescuer; A grouping module, used to obtain the real-time location data of the rescuer based on the target dispatch instruction, perform dynamic grouping, obtain a communication group division plan, perform spectrum management and mode switching, and create a collaborative communication network; A generation module is used to fuse and process the multi-source data of the collaborative communication network to obtain a real-time rescue situation map and perform trend prediction and report generation to obtain a target event assessment report and an emergency knowledge map.
9. An emergency rescue management device based on a dual-mode intercom, characterized in that: The emergency rescue management device based on the dual-mode intercom includes: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instruction in the memory so that the emergency rescue management device based on the dual-mode walkie-talkie executes the emergency rescue management method based on the dual-mode walkie-talkie as described in any one of claims 1-7.
10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by the processor, an emergency rescue management method based on a dual-mode walkie-talkie as described in any one of claims 1-7 is implemented.
Citation Information
Cited By
Elevator emergency rescue data intelligent analysis and early warning method and system
CN119612305A
Emergency path planning method and system based on big data analysis
CN120027808A
An emergency path planning method and system based on big data analysis
CN120027808B
Human-computer interaction method and system in emergency scene
CN120029448A
A human-computer interaction method and system for emergency rescue scenarios
CN120029448B
