A mine safety training detection method and system based on mixed reality

Abstract

The application belongs to the technical field of mine safety mixed reality interactive training, and specifically provides a mine safety training detection method and system based on mixed reality, which mainly comprises the following steps: obtaining a mine basic data set, generating reinforced safety constraint boundary data according to the mine basic data set; generating environment risk features according to the reinforced safety constraint boundary data; generating safety behavior features according to the reinforced safety constraint boundary data and the environment risk features; generating risk evolution features according to the environment risk features and the safety behavior features; generating emergency response features according to the safety behavior features and the risk evolution features; and generating dynamic risk visualization pictures according to the risk evolution features and the emergency response features. The application realizes dynamic training scenes, automatic operation behavior compliance detection, and risk evolution process visualization, and improves the effectiveness of safety training and emergency handling capability.

Description

Technical Field

[0001] This application belongs to the field of mixed reality interactive training technology for mine safety, and specifically relates to a mixed reality-based mine safety training and testing method and system. Background Technology

[0002] Traditional mine safety training mainly focuses on knowledge transmission and static scenario simulation, lacking a mechanism for real-time detection and dynamic evaluation of trainees' operational behavior. For example, theoretical lectures and animated demonstrations are difficult to track trainees' specific actions in virtual or real scenarios. Although simulation exercises involve operational experience, they lack quantitative testing methods for the compliance of actions and the effectiveness of risk response.

[0003] Traditional mine safety training makes it difficult to detect trainees' operational deviations in a timely manner, assess their mastery of safety regulations, or provide personalized feedback based on individual behavioral characteristics. Due to the lack of a closed-loop verification process for training effectiveness, it is difficult to predict whether trainees' behavior will comply with safety regulations when facing real underground risks, thus affecting the efficiency of transforming training into actual safety capabilities. Summary of the Invention

[0004] This application provides a mixed reality-based method and system for detecting mine safety training, which effectively solves the problems of fixed scenarios, lack of real-time behavior detection and closed-loop risk assessment in existing mine safety training processes, resulting in insufficient training effectiveness. It realizes dynamic training scenarios, automated detection of operational behavior compliance, and visualization of risk evolution processes, thereby improving the effectiveness of safety training and emergency response capabilities.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] Firstly, this application provides a mine safety training and testing method based on mixed reality, including:

[0007] Obtain the basic dataset of the mine, perform dynamic scenario construction processing based on the basic dataset of the mine, and generate enhanced safety constraint boundary data.

[0008] Environmental risk characteristics are generated by performing environmental coupling acquisition and processing based on enhanced safety constraint boundary data.

[0009] Based on enhanced safety constraint boundary data and environmental risk characteristics, closed-loop action verification is performed to generate safety behavior characteristics.

[0010] Based on environmental risk characteristics and safety behavior characteristics, collaborative risk extrapolation is conducted to generate risk evolution characteristics.

[0011] Multi-level response processing is performed based on safety behavior characteristics and risk evolution characteristics to generate emergency response characteristics.

[0012] Based on the characteristics of risk evolution and emergency response, holographic decision feedback processing is performed to generate dynamic risk visualization.

[0013] Furthermore, the mine basic dataset includes basic 3D point cloud data of mine roadways and preset standard operating procedure node data.

[0014] Dynamic scenario construction is performed based on the mine's basic dataset to generate risk propagation path data and enhanced safety constraint boundary data, including:

[0015] Based on the three-dimensional point cloud data of mine roadways, the safety clearance of mobile devices is calculated to obtain dynamic equipment safety boundary data; the risk propagation path data is obtained by annotating the node data of the preset standard operation process with risk lines.

[0016] By overlaying risk propagation path data with dynamic device safety boundary data in spatial trajectory, composite safety constraint boundary data is obtained.

[0017] Historical accident scenario reproduction data is acquired, and spatial risk is enhanced by combining historical accident scenario reproduction data with composite safety constraint boundary data to obtain enhanced safety constraint boundary data.

[0018] Furthermore, environmental coupling data collection and processing are performed based on enhanced safety constraint boundary data to generate environmental risk characteristics, including:

[0019] Acquire video stream data of miners' operations and real-time monitoring data of ambient gas concentration.

[0020] Spatial relationship analysis of personnel and equipment is performed on the video stream data of miners' operations to obtain spatial coordinate data of human-computer interaction.

[0021] Dynamic hazard source distribution data is generated based on real-time monitoring data of ambient gas concentrations.

[0022] By integrating dynamic hazard source distribution data with enhanced safety constraint boundary data, collaborative risk field data is constructed.

[0023] Generate environmental risk characteristics that include human-computer interaction spatial coordinate data and collaborative risk field data.

[0024] Furthermore, closed-loop action verification is performed based on enhanced safety constraint boundary data and environmental risk characteristics to generate safety behavior characteristics, including:

[0025] Operational space compliance detection is performed based on human-computer interaction spatial coordinate data and enhanced safety constraint boundary data to obtain equipment operation deviation data; anti-interference and disaster avoidance path data is generated based on human-computer interaction spatial coordinate data and collaborative risk field data.

[0026] Generate safety behavior characteristics that include equipment operation deviation data and anti-interference disaster avoidance path data.

[0027] Furthermore, based on environmental risk characteristics and safety behavior characteristics, collaborative risk extrapolation is conducted to generate risk evolution characteristics, including:

[0028] Reconstruct collaborative risk field data to generate dynamic three-dimensional risk field data; combine equipment operation deviation data to generate risk collaborative evolution prediction data; combine anti-interference disaster avoidance path data to verify escape plans and obtain emergency escape effectiveness index data.

[0029] It generates risk evolution characteristics that include dynamic three-dimensional risk field data, risk co-evolution prediction data, and emergency escape effectiveness index data.

[0030] Furthermore, based on safety behavior characteristics and risk evolution characteristics, multi-level response processing is performed to generate emergency response characteristics, including:

[0031] Acquire location data of the collaborative work group; plan collaborative evacuation routes based on the risk evolution characteristics of the collaborative work group location data, and generate multi-unit collaborative evacuation plan data.

[0032] Based on the risk evolution characteristics, a path safety fusion assessment was conducted on the data of multi-unit coordinated evacuation schemes and anti-interference disaster avoidance routes to obtain the final evacuation safety coefficient data.

[0033] Generate emergency response features that include data on multi-unit coordinated evacuation plans and final evacuation safety factor data.

[0034] Furthermore, based on the risk evolution characteristics, collaborative evacuation route planning is performed on the location data of the collaborative work groups to generate multi-unit collaborative evacuation plan data, including:

[0035] Team risk coupling analysis was performed on the location data and risk co-evolution prediction data of the collaborative work group to obtain the team risk coupling coefficient data.

[0036] Based on the team's risk coupling coefficient data and emergency escape effectiveness index data, evacuation routes were planned, and multi-unit coordinated evacuation plan data were obtained.

[0037] Furthermore, based on the risk evolution characteristics, a path safety fusion assessment was performed on the multi-unit coordinated evacuation plan data and anti-interference disaster avoidance path data to obtain the final evacuation safety coefficient data, including:

[0038] By integrating data from multi-unit coordinated evacuation plans with data from anti-interference disaster avoidance routes, an optimized route is generated, resulting in optimized disaster avoidance route data.

[0039] By collaboratively assessing and optimizing disaster avoidance path data and risk co-evolution prediction data, the path risk is obtained, and the final evacuation safety coefficient data is obtained.

[0040] Furthermore, based on the characteristics of risk evolution and emergency response, holographic decision feedback processing is performed to generate a dynamic risk visualization, including:

[0041] Dynamic three-dimensional risk field data is holographically mapped to generate virtual-real fusion risk field data.

[0042] Based on the virtual-real fusion risk field data and multi-unit collaborative evacuation plan data, collaborative escape guidance animation data is generated; combined with the final evacuation safety coefficient data, mixed reality early warning level signal is generated.

[0043] Drive collaborative escape guidance animation data and early warning signals to generate dynamic risk visualization.

[0044] Secondly, this application provides a mine safety training and detection system based on mixed reality, which includes:

[0045] Dynamic safety constraint modeling module: acquires the basic dataset of the mine, performs dynamic scenario construction processing based on the basic dataset of the mine, and generates enhanced safety constraint boundary data.

[0046] Environmental risk field coupling module: Based on the enhanced safety constraint boundary data, environmental coupling acquisition and processing are performed to generate environmental risk characteristics.

[0047] Safety behavior closed-loop verification module: Based on enhanced safety constraint boundary data and environmental risk characteristics, it performs closed-loop action verification and generates safety behavior characteristics.

[0048] Risk evolution collaborative simulation module: Based on environmental risk characteristics and safety behavior characteristics, collaborative risk simulation is performed to generate risk evolution characteristics.

[0049] Team emergency response module: Performs multi-level response processing based on safety behavior characteristics and risk evolution characteristics to generate emergency response characteristics.

[0050] Mixed Reality Decision Feedback Module: Performs holographic decision feedback processing based on risk evolution characteristics and emergency response characteristics to generate dynamic risk visualization.

[0051] Thirdly, this application provides a mixed reality-based mine safety training and testing device, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the steps of the mixed reality-based mine safety training and testing method as described in the first aspect.

[0052] Fourthly, this application provides a storage medium storing computer program instructions, which are read and executed by a processor to perform the steps of the mixed reality-based mine safety training and testing method described in the first aspect.

[0053] The beneficial effects of this application are:

[0054] This application constructs a dynamic mine scenario based on mixed reality technology, collects environmental risk characteristics and verifies operational actions, and finally deduces risk evolution and generates dynamic visual feedback. It effectively solves the problems of fixed scenarios, lack of real-time behavior detection and closed-loop risk assessment in existing mine safety training processes, which leads to insufficient training effectiveness. It realizes dynamic training scenarios, automated compliance detection of operational behaviors, and visualization of risk evolution process, thereby improving the effectiveness of safety training and emergency response capabilities.

[0055] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures pointed out in the description and the accompanying drawings. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart illustrating a mine safety training and testing method based on mixed reality according to this application is shown.

[0058] Figure 2 A schematic diagram of a module of a mine safety training and testing system based on mixed reality is shown in this application. Detailed Implementation

[0059] To address the problems raised in the background technology, this application constructs a dynamic mine scenario based on mixed reality technology, collects environmental risk characteristics, verifies operational actions, deduces risk evolution, and generates dynamic visual feedback, thereby improving the effectiveness of safety training and emergency response capabilities.

[0060] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0061] In some embodiments, such as Figure 1 As shown, this application provides a mine safety training and testing method based on mixed reality, including:

[0062] S1. Obtain the basic dataset of the mine, perform dynamic scene construction processing based on the basic dataset of the mine, and generate enhanced safety constraint boundary data.

[0063] S2. Based on the enhanced safety constraint boundary data, perform environmental coupling acquisition and processing to generate environmental risk characteristics.

[0064] S3. Based on the enhanced safety constraint boundary data and environmental risk characteristics, perform closed-loop action verification to generate safety behavior characteristics.

[0065] S4. Conduct collaborative risk simulation based on environmental risk characteristics and safety behavior characteristics to generate risk evolution characteristics.

[0066] S5. Perform multi-level response processing based on safety behavior characteristics and risk evolution characteristics to generate emergency response characteristics.

[0067] S6. Perform holographic decision feedback processing based on risk evolution characteristics and emergency response characteristics to generate dynamic risk visualization.

[0068] In some embodiments, the mine basic dataset in S1 includes basic three-dimensional point cloud data of mine roadways and preset standard operation process node data.

[0069] The basic data of the three-dimensional point cloud of mine roadways can be obtained by scanning with lidar and the set of (x,y,z) coordinates represents the spatial outline of the roadway.

[0070] Standard operating procedure (SOP) node data is a structured spatial-logical tag set based on mine operation procedures, describing the key control points and their spatial locations in the mine's standard operating procedures, such as: process number. Location coordinates Equipment type .

[0071] S1 performs dynamic scenario construction processing based on the mine's basic dataset, generating risk propagation path data and enhanced safety constraint boundary data, including:

[0072] S11. Calculate the safety clearance of mobile devices based on the three-dimensional point cloud data of the mine roadway to obtain dynamic equipment safety boundary data; mark the risk movement line of the preset standard operation process node data to obtain risk propagation path data.

[0073] When calculating the safety clearance of mobile devices based on the 3D point cloud data of mine roadways, a bounding box collision detection algorithm is used to obtain dynamic equipment safety boundary data. The specific steps include:

[0074] S111. Determine the outer envelope dimensions of the equipment and establish a minimum circumscribed cuboid model.

[0075] S112. Calculate the dynamic buffer distance , ;in, Representative type of equipment The maximum permissible operating speed, such as 1.5 m / s for a loader, Emergency response time threshold for representatives, such as 0.5 seconds. This represents the safety compensation coefficient of the equipment itself, which can be taken as 0.2m.

[0076] S113. Expand each surface of the outer envelope cuboid. Distance forms a three-dimensional volume space, and the output is dynamic device safety boundary data.

[0077] S12. Spatial trajectory overlay of risk propagation path data and dynamic equipment safety boundary data to obtain composite safety constraint boundary data.

[0078] Generate a spatial path by connecting the coordinates of adjacent nodes according to the process sequence. , When the distance between two paths is less than or equal to a preset distance threshold The point is identified as the intersection point, and then the risk weight is calculated. , ;in, Represents the m-th node or path point, threshold. It can be determined based on the minimum turning radius of the equipment, such as 2m, where f represents the frequency of intersections per day. , These represent the mass of the converging equipment, such as tonnage; the final output includes... The spatial vector line data of the values ​​serves as risk propagation path data.

[0079] When overlaying risk propagation path data with dynamic device safety boundary data into a spatial trajectory, first analyze the path points... Calculate to device boundary volume shortest distance ,when Less than or equal to the preset safety gap threshold The time mark is the constraint area This refers to composite safety constraint boundary data; among which, the safety gap threshold... The safety factor can be 1.2 times the maximum external dimension of the equipment.

[0080] S13. Obtain historical accident scenario reproduction data, and combine historical accident scenario reproduction data with composite safety constraint boundary data to enhance spatial risks and obtain enhanced safety constraint boundary data.

[0081] Establish a three-dimensional Gaussian field at accident point k , ;in, Represents the coordinates of the accident point. Represents the radius of influence, such as in a methane explosion accident. , This represents the Euclidean distance operator.

[0082] Generate reinforced security constraint values ​​at spatial location (x, y, z). Marked as enhanced security constraint boundary data; ;in, , The coefficients are determined through training with a BP neural network, k represents the record number in the accident database, and K represents the number of records in the historical accident scene reproduction data.

[0083] In some embodiments, S2 performs environmental coupling acquisition processing based on enhanced safety constraint boundary data to generate environmental risk characteristics, including:

[0084] S21. Acquire video stream data of miners' operations and real-time monitoring data of ambient gas concentration.

[0085] Infrared cameras deployed on the roof of the tunnel acquire operational video stream data, and sensors collect real-time monitoring data of ambient gas concentrations such as methane and carbon monoxide.

[0086] S22. Perform spatial relationship analysis on the video stream data of miners' operations to obtain human-computer interaction spatial coordinate data.

[0087] When analyzing the spatial relationship between personnel and equipment in video streams of miners' operations, the DeepLabCut deep learning framework is used: a pre-trained ResNet-50 model is used to detect key points of the miner's joints (such as head, hands, and feet) and the contour coordinates of the equipment in the video frames, and the spatial vector relationship between the two is calculated. For example, in a rock drilling rig operation scenario, the position of the miner's right hand is detected. Position of the trolley control handle The spatial distance is used to mark valid operation contact points when it is less than a specified distance (e.g., 0.5m), and the human-computer interaction spatial coordinate data is stored in a three-dimensional coordinate format.

[0088] S23. Generate dynamic hazard source distribution data based on real-time monitoring data of ambient gas concentration.

[0089] When generating dynamic hazard source distribution data based on real-time monitoring data of ambient gas concentration, the Shepard inverse distance weighted interpolation method can be used to obtain the hazard source concentration estimate c(x,y,z) at any point (x,y,z) in the roadway. When c(x,y,z) is greater than the methane explosion threshold, it is marked as a high-risk area, thus obtaining dynamic hazard source distribution data.

[0090] S24. Construct collaborative risk field data by integrating dynamic hazard source distribution data with enhanced safety constraint boundary data.

[0091] The process of constructing collaborative risk field data by integrating dynamic hazard distribution data and enhanced safety constraint boundary data involves two stages: the first stage maps the constraint regions of the enhanced safety constraint boundary data into a Boolean matrix B(x,y,z) (B=1 within the constraint region, 0 otherwise); the second stage overlays the dynamic hazard distribution data to generate risk field intensity values. Data labeled as collaborative risk field data: Among them, the coefficient , , It can be determined through regression analysis of historical accidents.

[0092] S25. Generate environmental risk characteristics including human-computer interaction spatial coordinate data and collaborative risk field data.

[0093] In some embodiments, S3 performs closed-loop action verification based on reinforced safety constraint boundary data and environmental risk characteristics to generate safety behavior characteristics, including:

[0094] S31. Perform operational space compliance detection based on human-computer interaction spatial coordinate data and enhanced safety constraint boundary data to obtain equipment operation deviation data; generate anti-interference and disaster avoidance path data based on human-computer interaction spatial coordinate data and collaborative risk field data.

[0095] Obtain the coordinates of the miner's operating part (e.g., right hand position) from the human-computer interaction spatial coordinate data. Using the 3D constraint zone marker value B(x,y,z) of the enhanced safety constraint boundary data, a spatial topology relationship detection algorithm is employed to calculate the minimum Euclidean distance from the operating part to the nearest constraint boundary. ,when Less than the preset safe distance threshold This was determined to be an operational violation.

[0096] Equipment operation deviation data is generated based on the cumulative percentage of violation duration: ;in, This represents the degree of deviation in equipment operation. This represents the total duration of the training cycle. This indicates the duration of the violation.

[0097] Dynamic path planning algorithm is used to generate disaster avoidance paths, and the three-dimensional risk distribution of collaborative risk field data is analyzed. Perform the following steps:

[0098] 1. Real-time reading of miner locations from human-computer interaction spatial coordinate data as the starting point of disaster avoidance paths.

[0099] 2. When a high-risk area is detected around the starting point (such as an area with a risk field intensity value R>0.6), a safe detour path is generated with that point as the center and the minimum detour distance (such as 2m).

[0100] 3. Calculate the new direction based on the movement direction vector of the human-computer interaction spatial coordinate data, such as limiting the path turning angle θ < 45°.

[0101] 4. Reload the collaborative risk field data every specified time interval (e.g., 0.5 seconds) to refresh the risk distribution, and replan the path based on the current human-computer interaction space coordinate data position.

[0102] S32. Generate safety behavior features including equipment operation deviation data and anti-interference disaster avoidance path data.

[0103] In some embodiments, S4 performs collaborative risk deduction based on environmental risk characteristics and safety behavior characteristics to generate risk evolution characteristics, including:

[0104] S41. Reconstruct collaborative risk field data to generate dynamic three-dimensional risk field data; combine equipment operation deviation data to generate risk collaborative evolution prediction data; combine anti-interference disaster avoidance path data to verify escape plans and obtain emergency escape effectiveness index data.

[0105] Dynamic reconstruction of the three-dimensional risk value distribution based on collaborative risk field data can be achieved by using Kriging space interpolation algorithms (such as the PyKrige library) to generate dynamic three-dimensional risk field data. For example, the risk values ​​of the original sparse monitoring points with a spacing of 50 meters can be expanded into a three-dimensional grid field with a resolution of 0.5 meters.

[0106] When generating risk co-evolution prediction data by combining equipment operation deviation data, a deviation-risk diffusion correlation model is established: the real-time value of equipment operation deviation data (e.g., rock drilling rig operation deviation). ( ) as the risk diffusion coefficient Input parameters, ;in, This represents the output value of the prediction model, i.e., the actual risk diffusion coefficient. This represents the basic risk diffusion coefficient, which can be determined based on the type of accident; for example, 0.5 is used for gas diffusion. .

[0107] According to partial differential equations Predict the evolution trend of the risk field within a specified time period (e.g., 120 seconds), where, Represents the Laplace operator. For example, when... At that time, the calculation yielded This caused the expansion rate of the core risk area to increase from 0.8 m / s to 1.2 m / s.

[0108] Using the coordinate sequence of anti-interference and disaster avoidance path data as the basic escape route, and superimposing the risk value distribution of dynamic three-dimensional risk field data, the cumulative risk integral of the miner's exposure along the path is calculated. , ;in, Represents dynamic three-dimensional risk field data at path points The risk value, This represents the time it takes for a miner to traverse path segment i.

[0109] When the accumulated risk points Less than the preset first safety threshold (e.g., 15) If the escape is deemed effective, output emergency escape effectiveness index data. : .

[0110] S42. Generate risk evolution characteristics including dynamic three-dimensional risk field data, risk co-evolution prediction data, and emergency escape effectiveness index data.

[0111] In some embodiments, S5 performs multi-level response processing based on safety behavior characteristics and risk evolution characteristics to generate emergency response characteristics, including:

[0112] S51. Obtain the location data of the collaborative work group; based on the risk evolution characteristics, perform collaborative evacuation path planning on the location data of the collaborative work group, and generate multi-unit collaborative evacuation plan data.

[0113] S52. Based on the risk evolution characteristics, conduct a path safety fusion assessment of the multi-unit coordinated evacuation plan data and anti-interference disaster avoidance path data to obtain the final evacuation safety coefficient data.

[0114] S53. Generate emergency response features including multi-unit coordinated evacuation plan data and final evacuation safety factor data.

[0115] In some embodiments, S51 performs coordinated evacuation route planning on the location data of the coordinated work group based on risk evolution characteristics, generating multi-unit coordinated evacuation plan data, including:

[0116] S511. Perform team risk coupling analysis on the location data and risk co-evolution prediction data of the collaborative work group to obtain team risk coupling coefficient data.

[0117] The collaborative work group location data includes a real-time three-dimensional coordinate sequence of all members of the work group.

[0118] Calculate the spatiotemporal intersection duration between each member and the high-risk area within a future time window. When the intersection of the exposure times of any two members exceeds a preset intersection threshold (e.g., 15 seconds), it is marked as a high-risk coupling event.

[0119] Team risk coupling coefficient data for: ;in, Represents the number of risk-coupled events. Represents the total number of members (e.g., in a group of three). =3), Represents exposure duration, coefficient , The empirical coefficient is derived from historical accident data.

[0120] S512. Based on the team risk coupling coefficient data and emergency escape effectiveness index data, plan evacuation routes and obtain multi-unit coordinated evacuation plan data.

[0121] Establish a two-factor decision matrix: when the team risk coupling coefficient data value The data value of the emergency escape effectiveness index is less than or equal to the preset risk threshold (e.g., 0.3). When the value is greater than or equal to the preset second safety threshold (e.g., 80), a centralized evacuation plan is adopted; when Greater than the risk threshold or When the value is below the second safety threshold, the group evacuation plan is activated.

[0122] Evacuation routes are generated using Dijkstra's algorithm combined with roadway topological constraints: the target point is the safety exit, the path turning angle is limited to ≤45°, and highly coupled areas are avoided. (Nodes >0.6) to obtain multi-unit coordinated evacuation plan data.

[0123] In some embodiments, S52 performs a path safety fusion assessment on the multi-unit coordinated evacuation plan data and anti-interference disaster avoidance path data based on risk evolution characteristics to obtain the final evacuation safety coefficient data, including:

[0124] S521. Integrate multi-unit coordinated evacuation plan data with anti-interference disaster avoidance path data to generate an optimized path and obtain optimized disaster avoidance path data.

[0125] Calculate the distance between adjacent nodes in the two types of paths, and merge them into a shared path segment when the node distance is less than or equal to a preset node distance threshold (e.g., 2 meters).

[0126] Specifically, staggered travel is achieved by setting a travel time window, assigning a timestamp t to each path point (e.g., t=0s for point A1, t=12s for point B1), ensuring that the time interval between arrivals of different units at the same location is greater than a specified value (e.g., 3 seconds), and outputting optimized disaster avoidance path data as a collaborative path network with time windows.

[0127] S522. Collaborately assess and optimize the path risks of disaster avoidance path data and risk co-evolution prediction data to obtain the final evacuation safety coefficient data.

[0128] For each path node of the optimized disaster evacuation route data ( Extract the point at the passage time. Risk value R ( , The total risk exposure value is obtained by accumulating the risk exposure values ​​of all nodes. .

[0129] Final evacuation safety factor data for: .

[0130] In some embodiments, S6 performs holographic decision feedback processing based on risk evolution characteristics and emergency response characteristics to generate a dynamic risk visualization, including:

[0131] S61. Perform holographic spatial mapping on dynamic three-dimensional risk field data to generate virtual-real fusion risk field data.

[0132] Based on the actual spatial dimensions of the mine roadway, each risk value point (x, y, z, R) in the dynamic three-dimensional risk field data is mapped to a virtual-real fusion coordinate system. For example, the original risk point coordinates (x, y, z) are mapped to virtual-real fusion coordinates. The risk value R is mapped to holographic transparency, and the output virtual-real fusion risk field data is displayed in real time through a holographic device.

[0133] S62. Generate collaborative escape guidance animation data based on virtual-real fusion risk field data and multi-unit collaborative evacuation plan data; generate mixed reality early warning level signal by combining final evacuation safety coefficient data.

[0134] When generating collaborative escape guidance animation data based on virtual-real fusion risk field data and multi-unit collaborative evacuation plan data, keyframe animation technology is used for processing.

[0135] Specifically, the path sequence in the multi-unit coordinated evacuation plan data is read, and a timestamp animation identifier is added to each path point in the virtual-real fusion coordinate system. Finally, the coordinated escape guidance animation data is output as a time-space-action sequence.

[0136] When generating a mixed reality early warning level signal by combining the final evacuation safety factor data, when the final evacuation safety factor data value... A green safety signal is output when the first threshold (e.g., 80) is reached, and a green safety signal is output when the preset second threshold is less than or equal to the final evacuation safety factor value. A yellow warning signal is output when the threshold is reached, ultimately determining the evacuation safety factor data value. A red danger signal is output when the threshold is less than the second threshold.

[0137] S63. Drive collaborative escape guidance animation data and early warning signals to generate dynamic risk visualization.

[0138] When driving collaborative escape guidance animation data and early warning signals to generate dynamic risk visualization, a spatiotemporal synchronization fusion engine is used to align and update the 3D animation and early warning status every specified time (e.g., 0.1 seconds), and finally outputs dynamic risk visualization.

[0139] In some embodiments, such as Figure 2 As shown, this application provides a mine safety training and detection system based on mixed reality, which includes:

[0140] Dynamic safety constraint modeling module: acquires the basic dataset of the mine, performs dynamic scenario construction processing based on the basic dataset of the mine, and generates enhanced safety constraint boundary data.

[0141] Environmental risk field coupling module: Based on the enhanced safety constraint boundary data, environmental coupling acquisition and processing are performed to generate environmental risk characteristics.

[0142] Safety behavior closed-loop verification module: Based on enhanced safety constraint boundary data and environmental risk characteristics, it performs closed-loop action verification and generates safety behavior characteristics.

[0143] Risk evolution collaborative simulation module: Based on environmental risk characteristics and safety behavior characteristics, collaborative risk simulation is performed to generate risk evolution characteristics.

[0144] Team emergency response module: Performs multi-level response processing based on safety behavior characteristics and risk evolution characteristics to generate emergency response characteristics.

[0145] Mixed Reality Decision Feedback Module: Performs holographic decision feedback processing based on risk evolution characteristics and emergency response characteristics to generate dynamic risk visualization.

[0146] In some embodiments, this application provides a mixed reality-based mine safety training and testing device, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the steps of the mixed reality-based mine safety training and testing method.

[0147] In some embodiments, this application provides a storage medium storing computer program instructions, which are read and executed by a processor to perform the steps of a mixed reality-based mine safety training and testing method.

[0148] Any references to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0149] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0150] Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A mixed reality-based mine safety training detection method, characterized in that, The method comprises the following steps: obtaining a mine foundation data set, performing dynamic scene construction processing according to the mine foundation data set, and generating enhanced safety constraint boundary data; performing environment coupling collection processing according to the enhanced safety constraint boundary data, and generating environment risk characteristics including human-machine interaction space coordinate data and collaborative risk field data; The method comprises the following steps: obtaining miner operation video stream data and environment gas concentration real-time monitoring data; performing personnel equipment space relationship analysis on the miner operation video stream data to obtain human-machine interaction space coordinate data; generating dynamic hazard source distribution data based on the environment gas concentration real-time monitoring data; fusing the dynamic hazard source distribution data and the enhanced safety constraint boundary data to construct the collaborative risk field data; generating environment risk characteristics including human-machine interaction space coordinate data and collaborative risk field data; performing closed-loop action verification according to the enhanced safety constraint boundary data and the environment risk characteristics, and generating safety behavior characteristics including equipment operation deviation data and anti-interference escape path data; performing collaborative risk deduction according to the environment risk characteristics and the safety behavior characteristics, and generating risk evolution characteristics, including: reconstructing the collaborative risk field data to generate dynamic three-dimensional risk field data; combining the equipment operation deviation data to generate risk collaborative evolution prediction data; combining the anti-interference escape path data to verify the escape scheme, and obtaining emergency escape effectiveness index data; performing multi-level response processing according to the safety behavior characteristics and the risk evolution characteristics, and generating emergency response characteristics, including: obtaining collaborative work group position data; performing collaborative evacuation path planning on the collaborative work group position data according to the risk evolution characteristics, and generating multi-machine group collaborative evacuation scheme data; performing path safety fusion evaluation on the multi-machine group collaborative evacuation scheme data and the anti-interference escape path data according to the risk evolution characteristics, and obtaining final evacuation safety coefficient data; performing holographic decision feedback processing according to the risk evolution characteristics and the emergency response characteristics, and generating dynamic risk visualization picture.

2. The mixed reality based mine safety training detection method according to claim 1, characterized in that, The mine foundation data set includes mine roadway three-dimensional point cloud basic data and preset standard operation process node data; performing dynamic scene construction processing according to the mine foundation data set, and generating risk propagation path data and enhanced safety constraint boundary data, including: performing mobile equipment safety gap calculation based on the mine roadway three-dimensional point cloud basic data to obtain dynamic equipment safety boundary data; performing risk dynamic line annotation on the preset standard operation process node data to obtain risk propagation path data; performing space trajectory superposition on the risk propagation path data and the dynamic equipment safety boundary data to obtain composite safety constraint boundary data; obtaining historical accident scene reproduction data, and performing spatial risk enhancement on the historical accident scene reproduction data and the composite safety constraint boundary data to obtain the enhanced safety constraint boundary data.

3. The mixed reality based mine safety training detection method according to claim 1, characterized in that, performing closed-loop action verification according to the enhanced safety constraint boundary data and the environment risk characteristics, and generating safety behavior characteristics, including: performing operation space compliance detection according to the human-machine interaction space coordinate data and the enhanced safety constraint boundary data to obtain equipment operation deviation data; generating anti-interference escape path data based on the human-machine interaction space coordinate data and the collaborative risk field data; Generate safety behavior features including device operation deviation data and anti-interference disaster avoidance path data.

4. The mixed reality based mine safety training detection method of claim 1, wherein, According to the risk evolution characteristics, the collaborative evacuation path of the collaborative operation group position data is planned, and the multi-machine group collaborative evacuation scheme data is generated, including: Perform team risk coupling analysis on collaborative operation group position data and risk collaborative evolution prediction data to obtain team risk coupling coefficient data; Plan the evacuation channel according to the team risk coupling coefficient data and the emergency escape effectiveness index data to obtain the multi-machine group collaborative evacuation scheme data.

5. The mixed reality based mine safety training detection method of claim 1, wherein, According to the risk evolution characteristics, the path safety fusion evaluation of the multi-machine group collaborative evacuation scheme data and the anti-interference disaster avoidance path data is performed, and the final evacuation safety coefficient data is obtained, including: Fuse the multi-machine group collaborative evacuation scheme data and the anti-interference disaster avoidance path data to generate an optimized path, and obtain optimized disaster avoidance path data; Collaboratively evaluate the path risk of the optimized disaster avoidance path data and the risk collaborative evolution prediction data to obtain the final evacuation safety coefficient data.

6. The mixed reality based mine safety training detection method of claim 1, wherein, According to the risk evolution characteristics and the emergency response characteristics, perform holographic decision feedback processing to generate dynamic risk visualization pictures, including: Perform holographic space mapping on dynamic three-dimensional risk field data to generate virtual-real fusion risk field data; Generate collaborative escape guidance animation data based on virtual-real fusion risk field data and multi-machine group collaborative evacuation scheme data; generate mixed reality warning level signal combined with final evacuation safety coefficient data; Drive the collaborative escape guidance animation data and the warning signal to generate dynamic risk visualization pictures.

7. A mixed reality based mine safety training detection system characterized in that, It includes: Dynamic safety constraint modeling module: obtain mine basic data set, perform dynamic scene construction processing according to the mine basic data set, and generate enhanced safety constraint boundary data; Environment risk field coupling module: perform environment coupling collection processing according to the enhanced safety constraint boundary data to generate environment risk characteristics including human-machine interaction space coordinate data and collaborative risk field data; It includes: Obtain miner operation video stream data and environment gas concentration real-time monitoring data; Perform personnel equipment space relationship analysis on the miner operation video stream data to obtain human-machine interaction space coordinate data; Generate dynamic hazard source distribution data based on environment gas concentration real-time monitoring data; Fuse dynamic hazard source distribution data and enhanced safety constraint boundary data to construct collaborative risk field data; Generate environment risk characteristics including human-machine interaction space coordinate data and collaborative risk field data; Safety behavior closed-loop verification module: perform closed-loop action verification according to the enhanced safety constraint boundary data and the environment risk characteristics to generate safety behavior features including device operation deviation data and anti-interference disaster avoidance path data; Risk evolution collaborative deduction module: perform collaborative risk deduction according to the environment risk characteristics and the safety behavior characteristics to generate risk evolution characteristics, including: reconstructing the collaborative risk field data to generate dynamic three-dimensional risk field data; combining the device operation deviation data to generate risk collaborative evolution prediction data; verifying the escape scheme combined with the anti-interference disaster avoidance path data to obtain emergency escape effectiveness index data; Team emergency collaborative response module: perform multi-level response processing according to the safety behavior characteristics and the risk evolution characteristics to generate emergency response characteristics; Mixed Reality Decision Feedback Module: Performs multi-level response processing based on safety behavior characteristics and risk evolution characteristics to generate emergency response characteristics, including: Acquire location data of the collaborative work group; plan collaborative evacuation routes based on the location data of the collaborative work group according to the risk evolution characteristics, and generate multi-unit collaborative evacuation plan data; Based on the risk evolution characteristics, a path safety fusion assessment was conducted on the multi-unit coordinated evacuation plan data and anti-interference disaster avoidance path data to obtain the final evacuation safety coefficient data; Based on the characteristics of risk evolution and emergency response, holographic decision feedback processing is performed to generate dynamic risk visualization.

Images