Comprehensive environmental perception warning method and system for intelligent rescue helmets
By establishing a virtual three-dimensional coordinate system and twin model in the data processing center, and connecting it with the intelligent rescue helmet, monitoring and updating data in real time, and collaborative early warning analysis combined with the intelligent perception network, it solves the difficulty in providing all-round environmental perception and early warning in the existing technology. The problem has improved the efficiency and safety of rescue operations.
Patent Information
- Application Number
- CN202411050271.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-08-01
AI Technical Summary
Existing rescue helmets are difficult to provide comprehensive environmental perception and early warning support in complex and changing rescue scenarios, making it difficult to ensure the efficiency and safety of rescue operations.
By establishing a virtual three-dimensional coordinate system and twin model in the data processing center, and connecting it with the intelligent rescue helmet, monitoring and updating data in real time, and collaborative early warning analysis combined with the intelligent perception network, comprehensive perception and timely early warning of the rescue site environment can be achieved.
It realizes all-round environmental perception and timely warning of complex rescue scenarios, and improves the efficiency and safety of rescue operations.
Smart Images

Figure CN118570991B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, and in particular to an omnidirectional environment perception and early warning method and system for an intelligent rescue helmet. Background Art
[0002] With the acceleration of urbanization and the frequent occurrence of natural disasters, intelligent rescue technology has become increasingly important in modern rescue operations. However, when dealing with complex and changing rescue scenarios, existing rescue technologies and equipment often have problems such as insufficient real-time perception, large information transmission delays, and incomplete environmental perception, making it difficult to meet the needs of fast, efficient, and safe rescue.
[0003] Traditional rescue helmets mainly rely on the rescuers' experience and limited sensor equipment, making it difficult to obtain and analyze environmental data in real time and unable to warn of potential dangers in a timely manner. In addition, rescuers need to deal with a variety of emergencies in complex rescue environments, such as fires, earthquakes, floods, etc., and existing rescue equipment is difficult to provide full-scale environmental perception and warning support. These problems make it difficult to ensure the efficiency and safety of rescue operations. Summary of the invention
[0004] The present application provides an all-round environmental perception and early warning method and system for smart rescue helmets, which is used to solve the technical problem that the existing technology is difficult to provide all-round environmental perception and early warning support when dealing with complex and changeable rescue scenarios.
[0005] In view of the above problems, the present application provides an all-round environmental perception warning method and system for a smart rescue helmet.
[0006] In a first aspect of the present application, a comprehensive environment perception and early warning method for a smart rescue helmet is provided, the method comprising:
[0007] Acquire regional data of the target area, and establish a virtual three-dimensional coordinate system in the data processing center according to the regional data, wherein the target area is the rescue area for intelligent rescue; take the virtual three-dimensional coordinate system as a benchmark, build a twin model based on the regional data in the data processing center, and establish connection and communication between the data processing center and the intelligent rescue helmet; acquire the real-time coordinates of the intelligent rescue helmet, the real-time coordinates are generated according to the virtual three-dimensional coordinate system, and a time-series update data set is established according to the real-time coordinates and the monitoring direction of the intelligent rescue helmet, and the time-series update data set is fed back to the data processing center, and the twin model is updated, and the updated twin model has a time-series identifier; receive the real-time attention tasks of the linked users, and perform rescue operations through the intelligent rescue helmet according to the real-time attention tasks Perform initial perception recognition, and synchronously call the model data of the updated twin model, fuse the initial perception recognition results and the model data, and establish an initial fusion result; make local recognition decisions for the smart rescue helmet based on the initial fusion result, reconstruct the focus position, and generate collection instructions for the reconstructed focus position; when the linked user moves according to the collection instructions, perform additional data collection of the focus position through the smart rescue helmet, and process the additional data collection results through the intelligent perception network to establish local early warning perception; synchronize the local early warning perception to the data processing center, and perform collaborative early warning analysis based on the real-time updated twin model through the data processing center to establish central early warning perception; establish real-time warning based on local early warning perception and central early warning perception.
[0008] The second aspect of the present application provides an all-round environmental perception warning system for a smart rescue helmet, the system comprising:
[0009] A virtual coordinate construction module, which obtains regional data of the target area, and establishes a virtual three-dimensional coordinate system in the data processing center according to the regional data, wherein the target area is a rescue area for intelligent rescue; a twin model construction module, which takes the virtual three-dimensional coordinate system as a reference, builds a twin model based on the regional data in the data processing center, and establishes a connection and communication between the data processing center and the intelligent rescue helmet; a twin model update module, which obtains the real-time coordinates of the intelligent rescue helmet, which are generated according to the virtual three-dimensional coordinate system, and establishes a time-series update data set according to the real-time coordinates and the monitoring direction of the intelligent rescue helmet, and feeds the time-series update data set back to the data processing center to update the twin model, wherein the updated twin model has a time-series identifier; an initial perception recognition module, which receives the real-time attention tasks of the linked user, and updates the twin model through the intelligent rescue helmet according to the real-time attention tasks. The helmet performs initial perception and recognition, and synchronously calls the model data of the updated twin model, fuses the initial perception and recognition results and the model data, and establishes an initial fusion result; a local recognition decision module, the local recognition decision module performs local recognition decision of the smart rescue helmet according to the initial fusion result, reconstructs the focus position, and generates a collection guide for the reconstructed focus position; an additional data collection module, when the linked user moves according to the collection guide, the additional data collection module performs additional data collection of the focus position through the smart rescue helmet, and processes the additional data collection results through the intelligent perception network to establish local early warning perception; a collaborative early warning analysis module, the collaborative early warning analysis module synchronizes the local early warning perception to the data processing center, and performs collaborative early warning analysis based on the real-time updated twin model through the data processing center to establish a central early warning perception; a real-time early warning module, the real-time early warning module establishes a real-time early warning based on the local early warning perception and the central early warning perception.
[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0011] The present application obtains regional data of the target area, and establishes a virtual three-dimensional coordinate system in the data processing center according to the regional data. The target area is the rescue area for intelligent rescue; based on the virtual three-dimensional coordinate system, a twin model is built in the data processing center based on the regional data, and a connection and communication between the data processing center and the intelligent rescue helmet is established; the real-time coordinates of the intelligent rescue helmet are obtained, and the real-time coordinates are generated according to the virtual three-dimensional coordinate system. A time series update data set is established according to the real-time coordinates and the monitoring direction of the intelligent rescue helmet, and the time series update data set is fed back to the data processing center to update the twin model. The updated twin model has a time series identifier; the real-time attention tasks of the linked users are received, and initialization is performed through the intelligent rescue helmet according to the real-time attention tasks. Initial perception and recognition, and synchronously call the model data of the updated twin model, fuse the initial perception and recognition results and the model data, and establish the initial fusion result; make local recognition decisions of the smart rescue helmet according to the initial fusion result, reconstruct the focus position, and generate the collection guide of the reconstructed focus position; when the linked user moves according to the collection guide, the smart rescue helmet performs additional data collection of the focus position, and processes the additional data collection results through the intelligent perception network to establish local early warning perception; synchronize the local early warning perception to the data processing center, and conduct collaborative early warning analysis based on the real-time updated twin model through the data processing center to establish central early warning perception; establish real-time warning based on local early warning perception and central early warning perception. The present invention solves the technical problem that the existing technology is difficult to provide all-round environmental perception and early warning support when dealing with complex and changeable rescue scenes, and realizes the technical effect of comprehensive perception and timely early warning of the rescue site environment through real-time monitoring, data processing, twin model update and collaborative early warning analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0013] Figure 1 A schematic flow chart of a comprehensive environmental perception and early warning method for a smart rescue helmet provided in an embodiment of the present application;
[0014] Figure 2 A schematic diagram of the structure of an all-round environmental perception and warning system for a smart rescue helmet provided in an embodiment of the present application.
[0015] Explanation of the accompanying drawings: virtual coordinate construction module 11, twin model building module 12, twin model updating module 13, initial perception and recognition module 14, local recognition and decision module 15, additional data collection module 16, collaborative early warning analysis module 17, real-time early warning module 18. DETAILED DESCRIPTION
[0016] This application provides a comprehensive environmental perception and warning method and system for smart rescue helmets, aiming to solve the technical problem that existing technologies are difficult to provide comprehensive environmental perception and warning support when dealing with complex and changeable rescue scenarios. Through real-time monitoring, data processing, twin model updates and collaborative warning analysis, the technical effect of comprehensive perception of the rescue site environment and timely warning is achieved.
[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0018] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.
[0019] Embodiment 1, as Figure 1 As shown, the present application provides an all-round environment perception warning method for a smart rescue helmet, the method comprising:
[0020] Step S100: Acquire regional data of a target area, and establish a virtual three-dimensional coordinate system in a data processing center according to the regional data, wherein the target area is a rescue area for performing intelligent rescue.
[0021] In an embodiment of the present application, firstly, regional data of the target area is acquired through data acquisition equipment, such as a GPS device, a laser scanner and a camera built into a smart rescue helmet, wherein the target area is the rescue area for smart rescue. Specifically, the GPS device uses the global positioning system to provide precise geographic coordinates; the laser scanner uses laser ranging technology to scan the terrain and generate high-precision three-dimensional point cloud data; the camera is used to capture and record image data of the target area in real time. Next, the collected data is classified and preprocessed, and the processed data includes two-dimensional data and three-dimensional data. Two-dimensional data is obtained by camera shooting, such as plan views and video images; three-dimensional data consists of three-dimensional point cloud data and models obtained by laser scanning.
[0022] After the data preprocessing is completed, the virtual 3D coordinate system is established in the data processing center. Specifically, a suitable coordinate system such as the geographic coordinate system is selected as the basis of the virtual 3D coordinate system, and the coordinate system definition and parameter setting are performed using the geographic information system software. The coordinate conversion tool is then used to map the preprocessed 2D and 3D data to the selected coordinate system to ensure that all data is processed in the same coordinate system. Through this process, the virtual 3D coordinate system is established.
[0023] Step S200: Taking the virtual three-dimensional coordinate system as a reference, a twin model is built based on regional data in the data processing center, and a connection and communication between the data processing center and the smart rescue helmet is established.
[0024] In the embodiment of the present application, a twin model is first constructed in a data processing center using preprocessed two-dimensional and three-dimensional data based on a virtual three-dimensional coordinate system. In this process, data integration is first performed to import regional data into professional modeling software, such as AutoCAD, and these data are integrated to generate a twin model of the target area.
[0025] At the same time, establish connection and communication between the data processing center and the smart rescue helmet, use network communication protocol to realize the connection between remote devices, and ensure high-speed and stable two-way data transmission between the data processing center and the smart rescue helmet.
[0026] Step S300: Obtain the real-time coordinates of the smart rescue helmet, which are generated according to the virtual three-dimensional coordinate system. Establish a time-series update data set according to the real-time coordinates and the monitoring direction of the smart rescue helmet, and feed the time-series update data set back to the data processing center to update the twin model. The updated twin model has a time-series identifier.
[0027] In an embodiment of the present application, the real-time geographic coordinates of the smart rescue helmet are obtained through the built-in GPS device of the smart rescue helmet, and the coordinate data is converted into a format matching the virtual three-dimensional coordinate system to obtain the real-time coordinates of the smart rescue helmet.
[0028] Next, a time-series update data set is established based on the real-time coordinates and the monitoring direction of the smart rescue helmet. The posture and direction of the helmet are obtained in real time through the built-in sensors of the smart rescue helmet, such as accelerometers and gyroscopes. Through the sensor fusion algorithm, these sensor data are fused and processed to determine the monitoring direction of the helmet. Then the real-time coordinates and monitoring direction data are combined to generate an update data set containing a time series. The time-series update data set records the location information and monitoring direction of the helmet at different time points. Its structure includes key data fields such as timestamp, coordinates, and monitoring direction. After that, the time-series update data set is transmitted from the smart rescue helmet to the data processing center in real time through wireless network transmission.
[0029] After receiving the time series update data set, the data processing center updates the twin model. By using data processing algorithms, such as ETL processes, the received data is processed and integrated into the existing twin model. The twin model is dynamically updated according to the latest time series data to ensure that the model can reflect the position and monitoring direction of the helmet in real time. Finally, in order to enhance the practicality and traceability of the twin model, the timestamp information is integrated into the twin model through the time series database, so that the twin model has a time series mark.
[0030] Step S400: Receive the real-time attention task of the linked user, perform initial perception and recognition through the smart rescue helmet according to the real-time attention task, and synchronously call the model data of the updated twin model, fuse the initial perception and recognition result and the model data, and establish an initial fusion result.
[0031] In an embodiment of the present application, firstly, a real-time attention task of a linked user is received through the user interface of the smart rescue helmet, wherein the linked user is a rescuer using the smart rescue helmet, and the real-time attention task includes specific attention targets and area information.
[0032] Next, the intelligent rescue helmet performs initial perception and recognition based on the real-time attention task. The intelligent rescue helmet is equipped with a camera and a variety of sensors such as temperature and humidity sensors to capture images and videos of the target area in real time and collect environmental data. Image processing algorithms such as edge detection and target recognition are used to process the captured images and videos to preliminarily identify the target of interest. The sensor data is combined with image data through data fusion algorithms such as weighted average and Kalman filtering to improve the accuracy of the recognition results. While performing initial perception and recognition, the intelligent rescue helmet sends a data request to the data processing center through the wireless network to call the updated twin model data. Finally, the fused data is further processed and analyzed to generate the initial fusion result. Specifically, the data processing module on the intelligent rescue helmet analyzes the fused data, extracts key features and information, and generates an initial fusion result that includes the current location of the target, environmental data, and related features.
[0033] Step S500: Perform local recognition decision of the smart rescue helmet according to the initial fusion result, reconstruct the focus position, and generate collection guidance of the reconstructed focus position.
[0034] In an embodiment of the present application, the local identification decision of the smart rescue helmet is made based on the initial fusion result by analyzing through preset logical rules, and the preset logical rules include temperature threshold judgment, gas concentration detection and structural change monitoring. The temperature threshold judgment is that when the temperature of a certain area exceeds the set safety threshold, the area is marked as a point of concern. Gas concentration detection is based on the concentration of harmful gases detected by the sensor. If the concentration exceeds the safety standard, the area is marked as a dangerous area. Structural change monitoring is to determine whether there is a potential structural danger by analyzing the deformation and displacement data of the structure.
[0035] Next, the smart rescue helmet reconstructs the location of interest based on the results of the recognition decision. The purpose of reconstructing the location of interest is to redefine the areas and objects that rescuers need to pay special attention to based on the latest recognition decision. Specifically, the data of the location of interest that needs to be reconstructed is extracted from the initial fusion results. The extracted data is spatially analyzed using a geometric algorithm to determine the new area of interest. For example, the precise location of the structural deformation is calculated using triangulation. Finally, based on the results of the geometric calculation, the coordinates and related information of the location of interest are updated.
[0036] Then, the smart rescue helmet generates collection instructions for the reconstructed location of interest. The collection instructions are to guide rescuers on how to further collect data and perform subsequent operations. Specifically, based on the reconstructed location of interest, the path planning algorithm is used to generate the optimal path to guide rescuers on how to reach the location of interest and avoid possible obstacles and dangerous areas. At the same time, detailed operation steps are provided to rescuers, including the angle of image shooting, the method of collecting sensor data, etc. For example, rescuers are guided to take high-resolution images at specific locations and record temperature and gas concentration data at the same time. According to the environmental characteristics of the reconstructed location of interest, rescuers are reminded of safety matters that need to be paid attention to during the collection process. For example, protective equipment needs to be worn when operating in high-temperature areas, or a respirator needs to be used in areas with toxic gases.
[0037] Through the above steps, the intelligent rescue helmet makes local recognition decisions based on the initial fusion results, reconstructs the focus position, and generates collection instructions for the reconstructed focus position.
[0038] Step S600: After the linked user moves according to the collection instructions, additional data collection of the concerned location is performed through the smart rescue helmet, and the additional data collection results are processed through the intelligent perception network to establish local early warning perception.
[0039] In the embodiment of the present application, when the linked user moves to the reconstructed location of interest according to the collection instructions, the smart rescue helmet activates its built-in camera and sensor. The camera takes high-resolution images or videos, and the sensor records environmental parameters in real time. These data collection devices work together to generate a complete additional data set.
[0040] Next, the smart rescue helmet processes the collected additional data through the embedded intelligent perception network. Specifically, the collected data is first cleaned and preprocessed. The data from different sensors are fused using a multi-sensor data fusion algorithm. For example, the data from the temperature sensor and the gas sensor are combined with the image data captured by the camera to generate comprehensive environmental perception information. The fused data is analyzed through built-in pattern recognition algorithms (such as edge detection and target recognition) and anomaly detection algorithms (such as threshold judgment) to identify potential dangers and abnormal situations. For example, when a sharp rise in temperature or an excessive concentration of toxic gas is detected, an early warning signal is immediately generated. Then, the intelligent perception network generates early warning information based on the preset early warning rules.
[0041] Through the above process, local early warning perception is established, that is, early warning information is generated to provide data support for subsequent processing.
[0042] Step S700: Synchronize the local warning perception to the data processing center, and establish central warning perception by performing collaborative warning analysis based on the twin model updated in real time by the data processing center.
[0043] In an embodiment of the present application, the local warning perception data generated by the smart rescue helmet is packaged and synchronized to the data processing center. In the data processing center, collaborative warning analysis is performed through a twin model based on real-time updates. Specifically, first, the received local warning perception data is integrated with the twin model through data fusion technologies such as data merging and association analysis. Use collaborative analysis algorithms, such as multidimensional data analysis and association rule mining, to conduct in-depth analysis of the integrated data. The analysis content includes identifying potential associated risks, predicting future risk development trends, etc. Based on the analysis results, the twin model is updated in real time to ensure that the model always reflects the latest environmental status and warning information.
[0044] Finally, a central warning perception is established, that is, based on the results of collaborative analysis, more comprehensive warning information is generated. The central warning information includes detailed risk assessment, spatial distribution of potential threats, and global response measures.
[0045] Step S800: Establish a real-time warning based on local warning perception and central warning perception.
[0046] In an embodiment of the present application, information of local warning perception and central warning perception is collected from a smart rescue helmet and a data processing center. A Bayesian network or a fuzzy logic algorithm is applied to fuse the local and central warning data to generate comprehensive warning information. Based on the fused comprehensive warning information, real-time warning information is generated. The real-time warning information includes specific warning content and can be presented to rescuers in a timely manner on the display screen of the smart rescue helmet. Real-time warning is completed through the above process.
[0047] Furthermore, in the method provided in the embodiment of the application, the collaborative early warning analysis based on the twin model updated in real time by the data processing center further includes:
[0048] The collaborative identification network of the data processing center is called, and the associated position analysis of local early warning perception is performed through the collaborative identification network to establish a collaborative position identification. The collaborative identification network is constructed based on regional data, and the collaborative identification network performs synchronous network updates based on the twin model updated in real time. The twin model is called back for data based on the collaborative position identification, and after data screening based on the result of the back for data call, the data is input into the collaborative prediction model, and the location development prediction corresponding to the collaborative position identification is predicted according to the collaborative prediction model, and the collaborative early warning analysis is completed according to the development prediction result.
[0049] In an embodiment of the present application, the collaborative identification network of the data processing center is first called. The collaborative identification network is built based on regional data and can comprehensively analyze data from different sources, and the collaborative identification network is based on a twin model that is updated in real time for synchronous network updates. Next, an associated position analysis of local early warning perception is performed in the collaborative identification network to identify key locations related to the early warning information. Specifically, data fusion algorithms, such as weighted average and Kalman filtering, are used to fuse local early warning perception data with regional data, and then association rule mining algorithms, such as the Apriori algorithm, are applied to analyze the relationship between the data and identify potential associated locations. Finally, based on the results of the association analysis, a collaborative location identifier is established to mark locations that require special attention.
[0050] Next, based on the collaborative location identifier, the twin model's backtracking data call is performed. Specifically, by sending a data request to the twin model database, the historical data related to the collaborative location identifier is called. Then, a backtracking algorithm, such as time series analysis, is used to extract the historical data of the corresponding location from the twin model to ensure the timeliness and relevance of the data.
[0051] After obtaining the retrospective data, the data is screened and then input into the collaborative forecasting model for analysis and forecasting. Specifically, data cleaning and screening techniques, such as filtering algorithms and outlier detection, are applied to remove irrelevant or noisy data, and the screened data is input into the collaborative forecasting model, and forecasting algorithms, such as the ARIMA model, are used to forecast location development. Based on the forecast results, the future development trend of the location corresponding to the collaborative location identifier is analyzed to identify potential risks and abnormal situations.
[0052] Finally, based on the prediction analysis results, the collaborative early warning analysis is completed. According to the prediction analysis results, specific early warning information is generated, including risk assessment, potential threat types and response measures.
[0053] Furthermore, in the method provided in the embodiment of the application, after the data is screened based on the result of the backtracking data call, the data is input into the collaborative prediction model, and the location development prediction corresponding to the collaborative location identifier is performed according to the collaborative prediction model, which also includes:
[0054] Obtain the backtracking node identifier of the backtracking data call result and establish a data richness evaluation; create an adaptive retention window according to the data richness evaluation result, and perform retention screening of the backtracking data call result based on the adaptive retention window; input the retention screening result into the collaborative prediction model, and execute the position development prediction corresponding to the collaborative position identifier.
[0055] In an embodiment of the present application, the result of the backtracking data call is first obtained, the data is parsed and structured, and key information such as timestamps and event descriptions are extracted. Time series analysis technology is used to identify key nodes in historical data and generate backtracking node identifiers. These identifiers mark the time points that are of great significance in historical data. Next, a data richness evaluation is established. Specifically, data quality analysis techniques, such as data integrity checks and data consistency checks, are applied to evaluate the quality of the backtracking data and determine the integrity and accuracy of the data. Information entropy or other information measurement methods are used to calculate the data richness of each backtracking node. The higher the amount of information, the more valuable information the data node contains. Based on the data quality and information calculation results, each backtracking node is scored for richness, and a data richness evaluation index is established.
[0056] Next, an adaptive retention window is created based on the data richness evaluation results. The adaptive retention window is used to filter out the most valuable historical data. Specifically, first, a richness threshold is set based on the data richness evaluation results. Only nodes with scores higher than the threshold will be retained. Adaptive algorithms such as sliding window algorithms and dynamic adjustment algorithms are used to adjust the size and range of the retention window according to real-time data conditions to ensure that the retained data has the highest amount of information and quality.
[0057] After that, based on the adaptive retention window, the back-tracking data call results are retained and filtered, and the filtering algorithm is applied to filter out qualified back-tracking data nodes according to the range of the adaptive retention window. Data with scores below the threshold are eliminated, and data with high scores are retained. The retained and filtered data are sorted and formatted to ensure that the data structure is consistent, which is convenient for subsequent input into the prediction model.
[0058] Finally, the retained screening results are input into the collaborative prediction model to perform the location development prediction corresponding to the collaborative location identifier. The collaborative prediction model is a pre-built model. The specific construction process is to train the time series prediction model using historical data. The historical data includes sensor data, image and video data, geographic location information, events and response records. The collaborative prediction model is obtained through training. The collaborative prediction model uses a collaborative prediction algorithm such as a time series prediction algorithm to predict the future development trend of the location corresponding to the collaborative location identifier, obtain the prediction analysis results, and complete the location development prediction corresponding to the collaborative location identifier.
[0059] Furthermore, the method provided in the application embodiment also includes:
[0060] The strength of the feedback signal of the smart rescue helmet is verified, and a weak signal set is configured according to the strength verification result; the real-time coordinates of the weak signal set are obtained, and the helmets of the smart rescue helmets are clustered according to the real-time focus tasks and the real-time coordinates to generate an adaptive clustering result; an interconnected local area network is established according to the adaptive clustering result, collaborative early warning analysis is performed through the interconnected local area network, local early warning perception is established, and real-time early warning is established according to the local early warning perception and local early warning perception.
[0061] In an embodiment of the present application, the strength of the feedback signal of the smart rescue helmet is first verified, and the feedback signal strength of each helmet is measured using the built-in signal strength detection module of the smart rescue helmet. The signal strength data of all helmets are collected, the signal strength values and their corresponding geographical locations are recorded, and a signal strength analysis algorithm, such as RSSI analysis, is used to evaluate the signal strength of each helmet, and signals below a preset threshold are marked as weak signals. Helmets with signal strengths below a preset threshold are marked as weak signal helmets, and all weak signal helmets and their corresponding real-time coordinate information are added to the weak signal set to form a complete weak signal data set.
[0062] Next, the real-time coordinates of each helmet are extracted from the weak signal set, and clustering algorithms, such as K-means clustering and DBSCAN, are applied to perform cluster analysis on the smart rescue helmets according to the real-time focus tasks and real-time coordinates to generate adaptive clustering results. The clustering parameters are adjusted according to different environments and task requirements to adapt to the dynamically changing rescue environment. Adaptive algorithms, such as the sliding window algorithm, are used to dynamically adjust the clustering results.
[0063] According to the results of adaptive clustering, an interconnected LAN is established, and collaborative early warning analysis is performed through the interconnected LAN. Specifically, wireless communication technology is used to use the wireless communication module between devices to establish an interconnected LAN to ensure efficient data transmission and communication between rescue helmets. Data synchronization and sharing are performed within the interconnected LAN, and the network time protocol is used to ensure time synchronization, and a distributed database is used to ensure data consistency. Through collaborative analysis algorithms, such as multi-sensor data fusion and joint analysis, comprehensive analysis is performed using MATLAB or Python to generate local early warning perception.
[0064] Finally, information fusion technologies, such as Bayesian networks and fuzzy logic, are used to integrate regional warning perception with local warning perception, comprehensively consider risk factors at all levels, and publish warning information in real time through the display screen of the smart rescue helmet to complete real-time warning.
[0065] Furthermore, the method provided in the application embodiment also includes:
[0066] A strong signal set is established based on the strength verification results; a load forecast analysis is performed on the strong signal set to establish a load forecast evaluation result; a media intelligent rescue helmet is determined based on the load forecast evaluation result, and a media mapping is established based on the distance between the media intelligent rescue helmet and the interconnected local area network; data interaction is performed between the interconnected local area network and the data processing center based on the media mapping.
[0067] In an embodiment of the present application, helmets with signal strength higher than a preset threshold are marked as strong signal helmets, and these helmets and their corresponding real-time coordinate information are added to a strong signal set to form a complete strong signal data set.
[0068] When conducting load forecasting analysis, historical load data of strong signal helmets, such as data transmission volume and communication frequency, are collected, and data preprocessing techniques, such as data cleaning and normalization, are used to ensure data consistency. Statistical analysis methods, such as moving average and linear regression, are used to analyze the possible load conditions of each helmet in the future, and the load capacity and potential load pressure of each helmet are evaluated. Based on the forecast results, load forecasting evaluation results are established to determine the load capacity and potential load pressure of each strong signal helmet.
[0069] Next, a strong signal helmet with better load forecasting evaluation results is selected as the media intelligent rescue helmet. The geographic information system technology is used to calculate the distance between the media intelligent rescue helmet and other nodes in the interconnected local area network. According to the distance calculation results, a mapping algorithm, such as the shortest path algorithm, is used to establish a media mapping between the media intelligent rescue helmet and other nodes, optimize the data transmission path, and ensure the efficiency of data transmission.
[0070] Finally, based on media mapping, standard data interaction protocols such as HTTP are used to ensure efficient transmission of data between the interconnected LAN and the data processing center, completing the data interaction between the interconnected LAN and the data processing center.
[0071] Furthermore, the method provided in the application embodiment also includes:
[0072] Establishing an early warning trigger threshold; determining whether the real-time early warning meets the early warning trigger threshold; if the real-time early warning meets the early warning trigger threshold, generating a mission hazard mark, and displaying the mission hazard mark through an intelligent rescue helmet.
[0073] In the embodiment of the present application, the early warning trigger threshold is first established. The early warning trigger threshold is a set of standards set according to historical data and environmental parameters, which is used to determine whether the real-time early warning meets the trigger condition and is pre-set by technical experts. The early warning trigger threshold is based on a variety of factors, including temperature, gas concentration, sound intensity, etc. For example, the temperature exceeding 50 degrees Celsius and the concentration of a certain harmful gas exceeding 100ppm are set as the trigger threshold.
[0074] Next, during real-time monitoring, it is determined whether the real-time warning meets the warning trigger threshold. Specifically, the built-in sensors of the smart rescue helmet, such as temperature sensors and gas sensors, are used to collect environmental data in real time. The real-time collected data is compared with the preset warning trigger threshold to determine whether the collected data exceeds the set threshold. When it is detected that the real-time data exceeds the warning trigger threshold, a task hazard indicator is generated, including information such as the hazard type and location.
[0075] Finally, the mission hazard mark is transmitted from the data processing module to the display module of the intelligent rescue helmet through the wireless communication module, and the mission hazard mark is displayed in real time on the display screen of the intelligent rescue helmet.
[0076] Furthermore, the method provided in the application embodiment also includes:
[0077] A signal anomaly identification of the smart rescue helmet is established according to the time series update data set; an emergency response instruction is generated through the signal anomaly identification, and a regional warning is reported based on the emergency response instruction; and abnormal management of the smart rescue helmet is performed through the regional warning.
[0078] In an embodiment of the present application, operating data is first collected in real time from the smart rescue helmet, including signal strength, sensor readings, etc. Using the data acquisition module, these data are recorded in chronological order and added to the time series update data set. Next, a time series analysis method, such as moving average and anomaly detection algorithm, is used to analyze the changing trend of the time series update data set to identify the moment when the signal is abnormal. Moving average is used to smooth the data, and anomaly detection algorithms are used to detect abnormal fluctuations in the data. When an abnormality is detected in the signal strength or other key indicators, a signal anomaly flag is generated to record the specific time, location, and type of the anomaly.
[0079] Next, emergency response instructions are automatically generated based on the signal anomaly identification. Emergency response instructions contain the anomaly type, location, and time to ensure that rescuers can respond in a timely manner. Emergency response instructions are sent to relevant rescuers and command centers through wireless communication modules. Broadcast communication technology is used to ensure that all relevant personnel can receive warning information in a timely manner and complete regional warnings.
[0080] Finally, the abnormal management of the smart rescue helmet is carried out through regional early warning. Specifically, the operation status of the smart rescue helmet is continuously monitored, and new abnormal situations are detected in real time through sensors and data analysis. According to the regional early warning information, corresponding response measures are taken, such as adjusting the rescue route.
[0081] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:
[0082] The present application obtains regional data of the target area, and establishes a virtual three-dimensional coordinate system in the data processing center according to the regional data. The target area is the rescue area for intelligent rescue; based on the virtual three-dimensional coordinate system, a twin model is built in the data processing center based on the regional data, and a connection and communication between the data processing center and the intelligent rescue helmet is established; the real-time coordinates of the intelligent rescue helmet are obtained, and the real-time coordinates are generated according to the virtual three-dimensional coordinate system. A time series update data set is established according to the real-time coordinates and the monitoring direction of the intelligent rescue helmet, and the time series update data set is fed back to the data processing center to update the twin model. The updated twin model has a time series identifier; the real-time attention tasks of the linked users are received, and initialization is performed through the intelligent rescue helmet according to the real-time attention tasks. Initial perception and recognition, and synchronously call the model data of the updated twin model, fuse the initial perception and recognition results and the model data, and establish the initial fusion result; make local recognition decisions of the smart rescue helmet according to the initial fusion result, reconstruct the focus position, and generate the collection guide of the reconstructed focus position; when the linked user moves according to the collection guide, the smart rescue helmet performs additional data collection of the focus position, and processes the additional data collection results through the intelligent perception network to establish local early warning perception; synchronize the local early warning perception to the data processing center, and conduct collaborative early warning analysis based on the real-time updated twin model through the data processing center to establish central early warning perception; establish real-time warning based on local early warning perception and central early warning perception. The present invention solves the technical problem that the existing technology is difficult to provide all-round environmental perception and early warning support when dealing with complex and changeable rescue scenes, and realizes the technical effect of comprehensive perception and timely early warning of the rescue site environment through real-time monitoring, data processing, twin model update and collaborative early warning analysis.
[0083] Embodiment 2 is based on the same inventive concept as the omnidirectional environment perception warning method for the smart rescue helmet in the previous embodiment. Figure 2 As shown, the present application provides an all-round environmental perception warning system for a smart rescue helmet, and the system and method embodiments in the present application are based on the same inventive concept. The system includes:
[0084] A virtual coordinate construction module 11, the virtual coordinate construction module 11 obtains the regional data of the target area, and establishes a virtual three-dimensional coordinate system in the data processing center according to the regional data, wherein the target area is the rescue area for intelligent rescue; a twin model construction module 12, the twin model construction module 12 takes the virtual three-dimensional coordinate system as a reference, builds a twin model based on the regional data in the data processing center, and establishes a connection and communication between the data processing center and the intelligent rescue helmet; a twin model update module 13, the twin model update module 13 obtains the real-time coordinates of the intelligent rescue helmet, the real-time coordinates are generated according to the virtual three-dimensional coordinate system, a time-series update data set is established according to the real-time coordinates and the monitoring direction of the intelligent rescue helmet, the time-series update data set is fed back to the data processing center, the twin model is updated, and the updated twin model has a time-series identifier; an initial perception recognition module 14, the initial perception recognition module 14 receives the real-time attention task of the linked user, and updates the twin model through the intelligent rescue helmet according to the real-time attention task The helmet performs initial perception and recognition, and synchronously calls the model data of the updated twin model, fuses the initial perception and recognition results and the model data, and establishes an initial fusion result; a local recognition decision module 15, the local recognition decision module 15 performs local recognition decision of the smart rescue helmet according to the initial fusion result, reconstructs the focus position, and generates a collection guide for the reconstructed focus position; an additional data collection module 16, when the linked user moves according to the collection guide, the additional data collection module 16 performs additional data collection of the focus position through the smart rescue helmet, and processes the additional data collection results through the intelligent perception network to establish local early warning perception; a collaborative early warning analysis module 17, the collaborative early warning analysis module 17 synchronizes the local early warning perception to the data processing center, and performs collaborative early warning analysis based on the real-time updated twin model through the data processing center to establish a central early warning perception; a real-time early warning module 18, the real-time early warning module 18 establishes a real-time early warning based on the local early warning perception and the central early warning perception.
[0085] Furthermore, the system is also used to implement the following functions:
[0086] The collaborative identification network of the data processing center is called, and the associated position analysis of local early warning perception is performed through the collaborative identification network to establish a collaborative position identification. The collaborative identification network is constructed based on regional data, and the collaborative identification network performs synchronous network updates based on the twin model updated in real time. The twin model is called back for data based on the collaborative position identification, and after data screening based on the result of the back for data call, the data is input into the collaborative prediction model, and the location development prediction corresponding to the collaborative position identification is predicted according to the collaborative prediction model, and the collaborative early warning analysis is completed according to the development prediction result.
[0087] Furthermore, the system is also used to implement the following functions:
[0088] Obtain the backtracking node identifier of the backtracking data call result and establish a data richness evaluation; create an adaptive retention window according to the data richness evaluation result, and perform retention screening of the backtracking data call result based on the adaptive retention window; input the retention screening result into the collaborative prediction model, and execute the position development prediction corresponding to the collaborative position identifier.
[0089] Furthermore, the system is also used to implement the following functions:
[0090] The strength of the feedback signal of the smart rescue helmet is verified, and a weak signal set is configured according to the strength verification result; the real-time coordinates of the weak signal set are obtained, and the helmets of the smart rescue helmets are clustered according to the real-time focus tasks and the real-time coordinates to generate an adaptive clustering result; an interconnected local area network is established according to the adaptive clustering result, collaborative early warning analysis is performed through the interconnected local area network, local early warning perception is established, and real-time early warning is established according to the local early warning perception and local early warning perception.
[0091] Furthermore, the system is also used to implement the following functions:
[0092] A strong signal set is established based on the strength verification results; a load forecast analysis is performed on the strong signal set to establish a load forecast evaluation result; a media intelligent rescue helmet is determined based on the load forecast evaluation result, and a media mapping is established based on the distance between the media intelligent rescue helmet and the interconnected local area network; data interaction is performed between the interconnected local area network and the data processing center based on the media mapping.
[0093] Furthermore, the system is also used to implement the following functions:
[0094] Establishing an early warning trigger threshold; determining whether the real-time early warning meets the early warning trigger threshold; if the real-time early warning meets the early warning trigger threshold, generating a mission hazard mark, and displaying the mission hazard mark through an intelligent rescue helmet.
[0095] Furthermore, the system is also used to implement the following functions:
[0096] A signal anomaly identification of the smart rescue helmet is established according to the time series update data set; an emergency response instruction is generated through the signal anomaly identification, and a regional warning is reported based on the emergency response instruction; and abnormal management of the smart rescue helmet is performed through the regional warning.
[0097] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0098] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0099] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A comprehensive environmental perception and early warning method for a smart rescue helmet, characterized in that: The method comprises: Acquire regional data of a target area, and establish a virtual three-dimensional coordinate system in a data processing center according to the regional data, wherein the target area is a rescue area for performing intelligent rescue; Taking the virtual three-dimensional coordinate system as a reference, a twin model is built based on regional data in the data processing center, and a connection and communication between the data processing center and the intelligent rescue helmet is established; Acquire the real-time coordinates of the smart rescue helmet, the real-time coordinates are generated according to the virtual three-dimensional coordinate system, establish a time series update data set according to the real-time coordinates and the monitoring direction of the smart rescue helmet, feed back the time series update data set to the data processing center, update the twin model, and the updated twin model has a time series identifier; Receive the real-time attention task of the linked user, perform initial perception and recognition through the smart rescue helmet according to the real-time attention task, and synchronously call the model data of the updated twin model, fuse the initial perception and recognition result and the model data, and establish an initial fusion result; Make local recognition decisions of the smart rescue helmet based on the initial fusion results, reconstruct the focus position, and generate collection guidance for the reconstructed focus position; When the linked user moves according to the collection instructions, the smart rescue helmet performs additional data collection of the concerned location, and the additional data collection results are processed through the intelligent perception network to establish local early warning perception; Synchronize the local warning perception to the data processing center, and establish central warning perception by performing collaborative warning analysis based on the twin model updated in real time through the data processing center; Establish real-time warning based on local warning perception and central warning perception; The method further comprises: Verify the strength of the feedback signal of the smart rescue helmet and configure the weak signal set according to the strength verification result; Acquire the real-time coordinates of the weak signal set, and perform helmet clustering of the smart rescue helmets according to the real-time focus tasks and the real-time coordinates to generate an adaptive clustering result; Establishing an interconnected local area network according to the adaptive clustering result, performing collaborative early warning analysis through the interconnected local area network, establishing local early warning perception, and establishing real-time early warning according to the local early warning perception and local early warning perception; The method further comprises: Establish a strong signal set based on the strength verification results; Performing load forecasting analysis on the strong signal set to establish a load forecasting evaluation result; Determine the media intelligent rescue helmet according to the load forecast evaluation result, and establish media mapping according to the distance between the media intelligent rescue helmet and the interconnected local area network; Data interaction between the interconnected local area network and the data processing center is performed according to the media mapping.
2. The omnidirectional environment perception and early warning method for a smart rescue helmet according to claim 1, characterized in that: The collaborative early warning analysis based on the twin model updated in real time by the data processing center also includes: Calling the collaborative identification network of the data processing center, performing the associated location analysis of the local early warning perception through the collaborative identification network, and establishing a collaborative location identification, wherein the collaborative identification network is constructed based on regional data, and the collaborative identification network performs synchronous network updates based on the twin model updated in real time; Based on the collaborative location identifier, the twin model’s back-tracing data is called. After data screening based on the back-tracing data call results, the data is input into the collaborative prediction model. The location development prediction corresponding to the collaborative location identifier is made according to the collaborative prediction model, and the collaborative early warning analysis is completed based on the development prediction results.
3. The omnidirectional environment perception and early warning method for a smart rescue helmet according to claim 2, characterized in that: After the data is screened based on the result of the back-tracing data call, the result is input into the collaborative prediction model, and the location development prediction corresponding to the collaborative location identifier is performed according to the collaborative prediction model, which also includes: Obtain the backtracking node identifier of the backtracking data call result and establish data richness evaluation; Create an adaptive retention window based on the data richness evaluation results, and perform retention screening of the back-tracking data call results based on the adaptive retention window; The retained screening results are input into the collaborative prediction model to perform the location development prediction corresponding to the collaborative location identifier.
4. The omnidirectional environment perception and early warning method for a smart rescue helmet according to claim 1, characterized in that: The method further comprises: Establish warning trigger thresholds; Determining whether the real-time warning meets the warning triggering threshold; If the real-time warning meets the warning triggering threshold, a mission hazard mark is generated, and the mission hazard mark is displayed through the smart rescue helmet.
5. The omnidirectional environment perception and early warning method for a smart rescue helmet according to claim 1, characterized in that: The method further comprises: Establishing a signal anomaly identification of the smart rescue helmet according to the time series update data set; Generate an emergency response instruction through the signal abnormality indicator, and issue a regional warning based on the emergency response instruction; Abnormal management of the smart rescue helmet is carried out through the regional early warning.
6. An all-round environmental perception warning system for smart rescue helmets, characterized in that: The system is used to execute the method according to any one of claims 1 to 5, and the system comprises: A virtual coordinate construction module, wherein the virtual coordinate construction module obtains regional data of a target area and establishes a virtual three-dimensional coordinate system in a data processing center according to the regional data, wherein the target area is a rescue area for performing intelligent rescue; A twin model building module, which uses the virtual three-dimensional coordinate system as a reference, builds a twin model based on regional data in a data processing center, and establishes a connection and communication between the data processing center and the intelligent rescue helmet; A twin model update module, wherein the twin model update module obtains the real-time coordinates of the smart rescue helmet, wherein the real-time coordinates are generated according to the virtual three-dimensional coordinate system, establishes a time series update data set according to the real-time coordinates and the monitoring direction of the smart rescue helmet, feeds the time series update data set back to the data processing center, updates the twin model, and the updated twin model has a time series identifier; An initial perception recognition module, which receives the real-time attention task of the linked user, performs initial perception recognition through the smart rescue helmet according to the real-time attention task, and synchronously calls the model data of the updated twin model, fuses the initial perception recognition result and the model data, and establishes an initial fusion result; A local recognition and decision module, wherein the local recognition and decision module performs local recognition and decision of the intelligent rescue helmet according to the initial fusion result, reconstructs the focus position, and generates a collection guide for the reconstructed focus position; An additional data collection module, when the linked user moves according to the collection guide, the additional data collection module performs additional data collection of the concerned location through the intelligent rescue helmet, and processes the additional data collection results through the intelligent perception network to establish local early warning perception; A collaborative early warning analysis module, which synchronizes the local early warning perception to the data processing center, performs collaborative early warning analysis based on the twin model updated in real time by the data processing center, and establishes central early warning perception; A real-time warning module is used to establish a real-time warning based on local warning perception and central warning perception.
