A Coal Mine Fire Prevention Simulation Training Method Based on Virtual Reality Technology

Abstract

The application discloses a coal mine fire prevention simulation training method based on virtual reality technology, relates to the technical field of coal mine fire prevention simulation training, and is used for solving the problem that the virtual scene and the operation of a trainee are out of synchronization due to sensor data delay or misalignment in the training process, and comprises the following steps: collecting action data of the trainee in real time, including posture change, hand movement and position coordinate information. The application ensures seamless synchronization of the sensor and the virtual scene through a time synchronization algorithm and a dynamic compensation model, improves the safety and effectiveness of the training, and combines a sliding window, a prediction algorithm and a redundancy detection mechanism, so that the system has high fault tolerance, and the training continuity and stability are ensured. A* algorithm is adopted to realize dynamic path planning under the fire spread trend, and the trainee can be quickly adjusted and corrected when deviating from the route, so that the real-time response capability and the immersion of the training are improved, and the trainee can make quick decisions and take actions in an emergency.

Description

Technical Field

[0001] This invention relates to the field of coal mine fire prevention simulation training technology, and specifically to a coal mine fire prevention simulation training method based on virtual reality technology. Background Technology

[0002] Virtual Reality (VR) based coal mine fire prevention simulation training is an immersive training system that simulates various scenarios and sudden fire situations within a coal mine. In this system, trainees intuitively experience the fire's occurrence, smoke spread, and emergency escape routes in a virtual environment, while also learning the use of fire extinguishers and key emergency procedures. This technology enhances trainees' emergency response capabilities, avoids the risks of actual drills, improves training efficiency and effectiveness, and provides an innovative solution for coal mine safety management.

[0003] The existing technology has the following shortcomings:

[0004] In virtual reality-based coal mine fire prevention simulation training, sensor data latency or inaccuracy is a critical issue, potentially leading to desynchronization between the virtual scene and the trainee's actions. VR systems rely on sensors such as head trackers, controllers, and positioning devices to capture trainees' movements and interactions in real time. If network latency, hardware failure, or algorithm processing delays cause a time lag between the fire dynamics (such as fire spread and escape route display) and the trainee's actions, it will severely impact judgment. For example, in escape drills, if the virtual fire spread lags behind the actual situation, trainees may misjudge and choose the wrong escape route, thus facing significant life-threatening danger in a real fire. Therefore, ensuring high-precision synchronization of sensor data and real-time system response is crucial for the safety and effectiveness of training.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a coal mine fire prevention simulation training method based on virtual reality technology. Through time synchronization algorithms (such as PTP or NTP) and dynamic compensation models, it achieves precise matching between sensors and the virtual scene, ensuring seamless synchronization between trainee operations and the screen, avoiding misleading information caused by network fluctuations or equipment malfunctions, and improving training safety. Combining sliding window technology, predictive compensation algorithms, and redundancy detection mechanisms, the system possesses high fault tolerance, maintaining continuous operation in complex network environments and quickly switching even in the event of sensor failure, improving stability and efficiency. Through the A* algorithm and time synchronization scheme, the system can dynamically plan escape routes in real time based on the fire and trainee location, and quickly adjust guidance, enhancing immersion and emergency decision-making capabilities, thereby solving the problems mentioned in the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a coal mine fire prevention simulation training method based on virtual reality technology, used to solve the problem of asynchronous operation between the virtual scene and the trainee due to sensor data delay or inaccuracy during training, comprising the following steps:

[0008] The system collects students' motion data in real time, including posture changes, hand movements, and position coordinates; and performs noise reduction and formatting on the collected raw data through a preprocessing module.

[0009] Monitor the transmission delay of sensor data and detect anomalies in the network transmission process; if the delay exceeds the set threshold, dynamically compensate the data through timestamps and prediction models to generate temporary location information and avoid data loss or scene misalignment caused by short-term network fluctuations.

[0010] The fire spread path is predicted based on the Kalman filter algorithm. The current position and movement trend of the trainees are used as input parameters of the algorithm to dynamically calculate the speed and direction of fire spread; ensuring that the fire scene and the trainees' escape routes are updated synchronously in the virtual environment.

[0011] The system uses the A* path planning algorithm to calculate the optimal escape route in real time and dynamically adjusts the escape route guidance based on the trainee's current location and the fire's spread trend. If the trainee deviates from the planned route during the escape, it will be corrected through voice and visual prompts.

[0012] A time synchronization algorithm is used to accurately match sensor data with simulation data of the virtual scene, ensuring clock consistency between different devices; a synchronization correction algorithm is used to eliminate time deviation between the sensor and the system, achieving seamless synchronization between the scene and the student's operation.

[0013] Preferably, posture changes refer to the rotation and tilting of the trainee's body or head in three-dimensional space; hand movements refer to the changes in the trainee's hand movements during training; and position coordinate information refers to the changes in the trainee's position in three-dimensional virtual space, described using the X, Y, and Z axis coordinate system.

[0014] Preferably, the specific steps for monitoring sensor data transmission delay and performing dynamic compensation are as follows:

[0015] First, the trainee's motion data is collected from multiple sensors, and a timestamp is added to each frame of data to record the exact time point of its generation;

[0016] The latency during transmission is calculated by monitoring the timestamp of each frame of data in real time and the time it arrives at the VR system.

[0017] When sensor data delay or packet loss is detected, the system calls a dynamic compensation algorithm to make predictions based on previously collected motion data.

[0018] After generating temporary data, the temporary data is directly applied to the virtual scene while waiting for lost or delayed data packets to arrive again. Once the original data packets arrive, the system will automatically correct the position information in the scene to keep the virtual environment consistent with the real operation. If the original data fails to arrive within the set time, the temporary predicted data is considered valid and the trainees are notified through the sensor feedback system to ensure that they remain sensitive to scene changes.

[0019] Preferably, the steps for predicting the fire spread path based on the Kalman filter algorithm are as follows:

[0020] In the Kalman filter algorithm, the system's state variables and noise model are first defined. The fire spread path is assumed to be a continuous motion in two-dimensional or three-dimensional space, and the state vector... Describe the flame in the k Location at any moment and velocity components Secondly, it is also necessary to define the student's current location. As system input, it is used to evaluate the relative distance between the flame and the trainee; the noise in the system is modeled as process noise. and measuring noise ,in Q and R This is the noise covariance matrix, used to describe the uncertainties in flame propagation and sensor measurements. The state transition model is as follows: ,in, A It is the state transition matrix. B The input control matrix ensures that the relative positions of the fire and the trainees jointly influence the fire's spread.

[0021] In each iteration, the next state of the flame is calculated using the prediction phase of the Kalman filter algorithm. The prediction equation is as follows: ,in, This predicts the location and velocity of the fire based on the current state, reflecting the fire's spread trend in the absence of additional measurement data. Simultaneously, the system updates the covariance matrix; the update expression is: This ensures that the system estimate includes uncertainties, taking into account the impact of process noise.

[0022] Preferably, the system uses Kalman gain when new measurement data arrives. Adjusting the state prediction to synchronize the virtual scenario with the actual fire dynamics, the Kalman gain is calculated using the following formula: The updated state is as follows: ,in, For the current flame measurement position, H It is a measurement matrix, and the updated covariance is: Through measurement updates, the system will correct the fire spread path in real time and ensure that the direction and speed of the fire in the virtual scene are synchronized with the trainees' actions.

[0023] Preferably, the specific steps for dynamically calculating the optimal escape route using the A* path planning algorithm are as follows:

[0024] First, a gridded map model is constructed in the virtual mine scene, dividing each location in the scene into several nodes. Each node represents a specific location in the mine, and the path connection between each node is defined as an edge. Each edge is assigned a different cost value. During initialization, the system marks the starting point and the target point, and also identifies known obstacle areas.

[0025] Based on the fire simulation's spread trend, the nodes on the map at the edge of the fire source are adjusted in real time. The weight of each node consists of two parts: G value and H value. As the fire spreads, the cost of nodes at the edge of the fire source will increase. The system will mark the nodes at the edge of the fire source as high-risk areas or impassable areas, thereby forcing the A* algorithm to avoid dangerous paths.

[0026] During the path planning process, the search nodes are continuously expanded using the A* algorithm to search for a path from the starting point to the target point.

[0027] During the escape process, the system continuously monitors the deviation between the student's current position and the planned route. If the system detects that the student is not following the guided path, it immediately triggers a correction mechanism. The correction methods include voice prompts and visual feedback to guide the student back to the correct path. If the student continues to deviate, the system will re-invoke the A* algorithm to recalculate a new escape route based on the student's current position and the fire spread status, and update the guidance in real time.

[0028] Preferably, the specific steps for using a time synchronization algorithm to accurately match sensor and virtual scene data are as follows:

[0029] During the initialization phase, a network time protocol or a precision time protocol is used to synchronize all sensor devices with the main system clock;

[0030] When the sensor collects the trainee's operation data, it adds a timestamp to each frame of data to accurately record the time when it was generated;

[0031] A time synchronization algorithm is used to correct the clock errors between each device and the main system;

[0032] After all data has been time-corrected, the sensor data is fused and matched with the simulation data in the virtual scene to ensure that each frame of data can be accurately applied to the virtual environment in chronological order.

[0033] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0034] This invention achieves precise matching between sensors and virtual scenes by employing time synchronization algorithms (such as PTP or NTP) and dynamic compensation models, ensuring seamless synchronization between each student's operation and the virtual screen. Through high-frequency time calibration and data compensation, even in the event of network fluctuations or brief equipment malfunctions, the system can still guarantee the matching of actions and scenes, avoiding virtual misleading and improving the safety and effectiveness of training.

[0035] This invention, by combining sliding window technology, predictive compensation algorithms, and redundancy detection mechanisms, enables the system to respond quickly and automatically correct errors when faced with packet loss or network latency. This mechanism gives the system high fault tolerance, ensuring continuous training services even in complex network environments. Even if a sensor experiences a temporary malfunction, a backup sensor or predictive model can quickly step in, preventing training interruptions due to equipment problems. This not only improves system stability but also reduces the cost of restarting training after interruptions, enhancing training efficiency and the learning experience for participants.

[0036] This invention, through the A* algorithm combined with a dynamic time synchronization scheme, can replan escape routes in real time based on the fire's spread and the trainees' locations. Due to the guaranteed precision of time synchronization, the system can complete path calculation and feedback within milliseconds and immediately update escape guidance in the virtual scenario. When trainees deviate from the route or the fire situation suddenly changes, the system quickly adjusts and corrects the trainees' actions through visual or voice prompts. This real-time response capability greatly enhances the immersiveness of training, providing trainees with a more realistic drill experience and cultivating their ability to make rapid decisions and act in emergency situations. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0038] Figure 1 This is a flowchart of a coal mine fire prevention simulation training method based on virtual reality technology according to the present invention. Detailed Implementation

[0039] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0040] This invention provides, for example Figure 1 The method for coal mine fire prevention simulation training based on virtual reality technology is shown below. It addresses the problem of asynchronous operation between the virtual scene and the trainee's actions due to sensor data delays or inaccuracies during training. The method includes the following steps:

[0041] Multiple sensors, including head trackers, hand grips, and positioning devices, are used to collect trainees' motion data in real time, including posture changes, hand movements, and position coordinates. The raw data collected by the sensors is denoised and formatted by a preprocessing module to improve the reliability of data transmission.

[0042] Postural changes refer to the rotation and tilt of a trainee's body or head in three-dimensional space, typically represented by Euler angles or quaternions. Specifically, head trackers can capture the trainee's pitch, roll, and yaw angles, reflecting changes in the head's vertical, horizontal, and rotational directions. This information is crucial for ensuring viewpoint synchronization in the virtual environment. For example, when a trainee looks down at the ground or turns their head to observe a fire source during training, the system needs to adjust the VR viewpoint in real time to prevent dizziness and disorientation. Hand movements refer to the changes in the trainee's hand gestures during training, including hand gestures, grasping, releasing, and direction of movement. This information is collected via controllers or hand tracking devices, involving multi-degree-of-freedom movements such as lifting, pressing down, and rotating. Hand movement data is used not only to simulate the grasping and use of a fire extinguisher but also to detect the trainee's interactive reactions in emergency situations (such as pushing aside obstacles or instructing others to escape). Ensuring real-time capture and feedback of hand movements is essential for enhancing the immersion and realism of the training. Position coordinate information refers to the changes in a trainee's position in a three-dimensional virtual space, typically described using an X, Y, and Z coordinate system. Positioning devices track the trainee's displacement in real time, such as moving from one location to another. The system needs to update the fire spread status and escape routes in the scenario based on these position coordinates. This information is particularly crucial for navigation in the simulated environment. For example, when trainees need to avoid flames or find an exit during an escape drill, the system must synchronize the trainee's movement to ensure consistency between the virtual fire scenario and their movement path, avoiding misleading or delayed judgments.

[0043] Monitor the transmission delay of sensor data and detect anomalies in the network transmission process; if the delay exceeds the set threshold, dynamically compensate the data through timestamps and prediction models to generate temporary location information and avoid data loss or scene misalignment caused by short-term network fluctuations.

[0044] The specific steps for monitoring sensor data transmission delay and performing dynamic compensation are as follows:

[0045] The system first collects motion data from multiple sensors (such as head trackers, handles, and positioning devices), and appends a timestamp to each frame of data to record the exact time it was generated. The timestamps are typically recorded with high-precision clocks (e.g., nanoseconds or milliseconds) to determine data transmission latency during subsequent processing. This step ensures the system can accurately track the lifecycle of each data packet and provides a basis for subsequent latency calculations and compensation. For example, if a significant discrepancy is detected between the timestamp of the sensor data and the system's reception time, a potential latency issue can be identified.

[0046] The system calculates transmission latency by monitoring the timestamp of each frame of data in real time and its arrival time at the VR system. If the latency exceeds a set threshold (e.g., 50 milliseconds), the system marks it as abnormal latency. Simultaneously, the system assesses network transmission stability through packet loss rate, jitter, and network bandwidth usage. Upon detecting abnormal fluctuations, such as packet loss or a sudden increase in latency, the system records this information and activates response mechanisms. This process ensures that data transmission problems are detected promptly, preventing latency accumulation or network instability from interfering with the training process.

[0047] When sensor data delay or packet loss is detected, the system invokes a dynamic compensation algorithm to predict the current position based on previously collected motion data. Typical compensation models include linear interpolation or Kalman filtering algorithms, utilizing the learner's historical position and movement trends to calculate the current possible position. For example, if the controller's data packets are delayed, the system predicts the current hand position based on the learner's previously received data to generate temporary position information. In fire scenarios, this predictive function can prevent asynchronous display of fire spread due to short delays, thus ensuring that the learner's judgment is not interfered with.

[0048] After generating temporary data, the system applies it directly to the virtual scene while waiting for lost or delayed data packets to re-arrive. Once the original data packets arrive, the system automatically corrects the positional information in the scene to ensure consistency between the virtual environment and real-world operations. If the original data fails to arrive within a set time, the system treats the temporary prediction data as valid and notifies the trainee via sensor feedback (such as vibration or voice prompts) to ensure their sensitivity to scene changes. This entire process ensures that virtual training can proceed smoothly even under network fluctuations, without causing scene misalignment or incorrect feedback due to data latency.

[0049] The fire spread path is predicted based on the Kalman filter algorithm. The current position and movement trend of the trainees are used as input parameters of the algorithm to dynamically calculate the speed and direction of fire spread; ensuring that the fire scene and the trainees' escape routes are updated synchronously in the virtual environment.

[0050] The steps for predicting fire spread paths based on the Kalman filter algorithm are as follows:

[0051] In the Kalman Filter algorithm, the system's state variables and noise model are first defined. It is assumed that the fire spread path is a continuous motion in two-dimensional or three-dimensional space, and the state vector... Describe the flame in the k Location at any moment and velocity components The student's current location also needs to be defined. This serves as a system input for evaluating the relative distance between the flame and the trainee. Simultaneously, noise in the system is modeled as process noise. and measuring noise ,in Q and R This is the noise covariance matrix, used to describe the uncertainties in flame propagation and sensor measurements. The state transition model is as follows: ,in, A It is the state transition matrix. B The input control matrix ensures that the relative positions of the fire and the trainees jointly influence the fire's spread.

[0052] In each iteration, the system uses the prediction phase of the Kalman filter algorithm to calculate the next state of the flame. The prediction equation is: ,in, This predicts the location and velocity of the fire based on the current state, reflecting the fire's spread trend in the absence of additional measurement data. Simultaneously, the system updates the covariance matrix; the update expression is: This process ensures that the system estimate includes uncertainty, taking into account the effects of process noise. For example, when trainees move quickly or escape routes change, the covariance matrix expands, indicating increased uncertainty in the prediction.

[0053] When new measurement data (such as the current location of the flame spread) arrives, the system uses Kalman gain. Adjust the state prediction to synchronize the virtual scenario with the actual fire dynamics. The Kalman gain is calculated using the following formula: The updated state is as follows: ,in, For the current flame measurement position, H This is the measurement matrix. The updated covariance is: Through measurement and updates, the system corrects the fire's spread path in real time and ensures that the direction and speed of the fire in the virtual scene are synchronized with the trainees' movements, reducing misalignment caused by latency. Simultaneously, this step guarantees the real-time dynamic interaction between fire spread and escape routes, providing trainees with precise escape guidance.

[0054] The system uses the A* (A-star) path planning algorithm to calculate the optimal escape route in real time and dynamically adjusts the escape route guidance based on the trainee's current location and the fire spread trend. If the trainee deviates from the planned route during the escape, the system will correct the deviation through voice and visual prompts.

[0055] The specific steps for dynamically calculating the optimal escape route using the A* (A-star) path planning algorithm are as follows:

[0056] The system first constructs a gridded map model within the virtual mine scenario, dividing each location in the scenario into several nodes. Each node represents a specific location in the mine, such as a corridor, exit, or obstacle. The path connections between each node are defined as edges, and each edge is assigned a different cost, such as distance or the degree of fire risk along the path. During initialization, the system marks the starting point (the student's current location) and the target point (the safety exit), while also identifying known obstacle areas (such as fire sources or blocked passages). This step ensures that the map model provides the foundation for the A* algorithm, enabling it to reasonably calculate feasible paths.

[0057] The system adjusts the weights of nodes close to the fire source on the map in real time based on the fire's spread trend in the simulation. Each node's weight consists of two parts: a G-value (the actual cost from the starting point to the node) and an H-value (the estimated cost from the node to the target point, typically calculated using Euclidean or Manhattan distance). As the fire spreads, the cost of nodes near the fire source increases significantly, and the system marks these nodes as high-risk or impassable areas, forcing the A* algorithm to avoid dangerous paths. This dynamic update mechanism ensures that escape routes adapt to changes in the fire scenario in real time.

[0058] During path planning, the system continuously expands the search nodes using the A* algorithm, searching for a path from the starting point to the target point. The A* algorithm prioritizes expanding the node with the lowest total cost (G + H) and checks if adjacent nodes offer even lower-cost paths. If a better path is found, the system updates the path information and adds the node to the Open List to continue expanding until the target point is reached. Once an optimal path is calculated, the system immediately displays it in the virtual scene and provides visual guidance (such as arrows or path markers) to instruct the participant to escape along that route. This process ensures that the escape path is always the best choice under the current circumstances.

[0059] During the escape, the system continuously monitors the deviation between the participant's current location and the planned route. If the system detects that a participant is not following the indicated path, it immediately triggers a correction mechanism. Correction methods include voice prompts (such as "Please return to the safe route") and visual feedback (such as flashing arrows or red markers) to guide the participant back to the correct path. If the participant continues to deviate, the system will re-invoke the A* algorithm, recalculate a new escape route based on the participant's current location and the fire's spread, and update the guidance in real time. This feedback mechanism ensures that participants do not get lost in complex or changing environments, minimizing the risks caused by misjudgments.

[0060] A time synchronization algorithm is used to accurately match sensor data with simulation data of the virtual scene to ensure clock consistency between different devices; a synchronization correction algorithm is used to eliminate time deviation between the sensor and the system to achieve seamless synchronization between the scene and the student's operation.

[0061] The specific steps for using a time synchronization algorithm to accurately match sensor and virtual scene data are as follows:

[0062] During the initialization phase, the system employs either Network Time Protocol (NTP) or Precision Time Protocol (PTP) to synchronize all sensor devices (such as head trackers, handheld devices, and positioners) with the main system clock. NTP is suitable for millisecond-level accuracy requirements, while PTP provides microsecond-level time synchronization, meeting the needs of high-precision scenarios. This step ensures that the internal clock of each device is consistent with the standard time of the central server, avoiding timestamp deviations in data packets caused by clock drift. Furthermore, to prevent errors caused by network jitter, the system performs periodic calibrations to ensure that all devices remain time-synchronized at all times.

[0063] When sensors collect student action data (such as position, posture, and hand movements), they add a timestamp to each frame of data, precisely recording the time of its generation. The system pre-aligns the data packets with the main system clock based on the received timestamps. If a discrepancy is detected between the timestamp of data transmitted by the device and the system's current clock, the system marks the data as a "delayed packet" and performs compensation processing. This process ensures that even slight delays in data transmission will not affect the synchronization between the virtual scene and the student's actions.

[0064] The system employs time synchronization algorithms (such as the Berkeley algorithm or the Cristian algorithm) to correct clock errors between each device and the main system. Specifically, when the system detects that a device's clock deviation exceeds a set threshold (e.g., ±2 milliseconds), it sends a synchronization request to that device and adjusts its clock to match the system time. Furthermore, to avoid frequent corrections impacting system performance, the system uses a sliding window technique, dynamically adjusting only devices with persistent deviations. This synchronization correction process not only resolves data packet timing errors but also ensures smooth data flow during interactions between multiple devices.

[0065] After all data has been time-corrected, the system fuses and matches sensor data with simulation data in the virtual scene, ensuring that each frame of data is accurately applied to the virtual environment in chronological order. For example, when a trainee turns their head, the system immediately refreshes the corresponding scene view, keeping the virtual image consistent with the real action. If a device's data packet is briefly lost, the system will predict and compensate based on the timestamps and motion trends of the previous frames, ensuring that the image does not stutter or misalign. The entire process guarantees that every action of the trainee is seamlessly integrated with the virtual scene, enhancing the immersion and realism of the training.

[0066] Implementation method 1: A high-precision solution for microsecond-level time synchronization using PTP;

[0067] In this implementation, the Precision Time Protocol (PTP) is employed to achieve microsecond-level clock synchronization, ensuring complete time consistency among all sensor devices and meeting the requirements for high-precision real-time interaction. The core of the PTP protocol is the continuous exchange of time signals between a dedicated master clock device (time server) and the sensor slave devices, adjusting the local clocks of each device to maintain a unified time reference. Unlike NTP's network-level time synchronization, PTP is particularly suitable for local area network (LAN) environments requiring low latency and high precision, reducing time synchronization errors to the microsecond level. This is crucial for coal mine fire simulation training, as every action and fire spread change in the virtual scenario requires a millisecond-level response, preventing trainees from perceiving any delay.

[0068] In practical applications, all sensor devices (such as head trackers, handles, and positioning devices) are connected to a time server. The server broadcasts time synchronization packets to the devices and collects feedback information from them. PTP calculates network transmission latency through the round-trip transmission of timestamps, thereby eliminating time deviations caused by network fluctuations. Whenever a sensor's clock drifts, the system immediately triggers synchronization adjustments to correct its deviation to the standard time. To further improve accuracy, dedicated network cables and switches can be used between the sensors and the server to reduce jitter and fluctuations during data transmission.

[0069] Furthermore, this solution boasts rapid fault response capabilities. When a sensor experiences connection instability or time drift exceeding permissible limits, the system will switch to a backup sensor or a pre-set predictive model in real time, preventing data synchronization lag or misalignment in the virtual scenario. Because PTP's synchronization error is at the microsecond level, the system can update the fire spread status and escape routes in an extremely short time, guiding trainees to make timely and correct judgments, ensuring the realism and immersion of the training.

[0070] Implementation method 2: Employing a time synchronization strategy of dynamic correction and data compensation;

[0071] This implementation utilizes the Berkeley algorithm and sliding window technique to dynamically adjust the time synchronization between sensor data and the virtual scene, ensuring stable system operation even under complex network conditions. The Berkeley algorithm is an average time synchronization method that collects the local clock times of all sensor devices, calculates the average, and corrects the clocks of each device to that average time. This method is suitable for environments with significant network fluctuations, such as wireless networks inside mines. The sliding window technique is used for historical data management, ensuring the integrity and order of each frame of data during the correction process.

[0072] In practice, the system periodically (e.g., every 5 seconds) triggers global time synchronization, collecting timestamps from all sensors, calculating the current average time, and pushing the result to each device. This effectively reduces time drift between sensors and avoids inconsistencies in data packets across multiple devices. If a sensor loses data packets due to network latency, the system uses a sliding window technique to store the data from the previous few frames and predict the current position based on trends. For example, when controller data packets are delayed, the system calculates the student's possible actions using historical records to ensure the virtual scene remains uninterrupted.

[0073] This data compensation strategy is particularly suitable for responding to emergencies during fire drills. Even in situations of network instability, the trainees' movements can still be accurately captured by the system and fed back into the virtual scenario. During the escape, if trainees experience temporary location inaccuracies due to network fluctuations, the system can rely on predictive models to maintain scenario synchronization, avoiding erroneous escape guidance. The use of sliding windows also ensures smooth data processing, allowing the system to immediately return to normal after each synchronization correction.

[0074] Implementation method 3: Fault-tolerant scheme for clock synchronization and redundancy detection based on NTP;

[0075] This implementation uses Network Time Protocol (NTP) for clock synchronization, combined with redundancy detection and data retransmission mechanisms to ensure high system stability and fault tolerance. NTP is suitable for time synchronization scenarios requiring medium precision, controlling clock errors between different devices to within milliseconds. Compared to PTP, NTP has lower implementation costs and is suitable for complex network environments, making it particularly suitable for scenarios requiring long-term stable system operation, such as in coal mine simulation training.

[0076] In practice, the system establishes a time synchronization connection with the sensor devices through an NTP server. Each sensor periodically requests time updates from the NTP server and adjusts its own clock based on the received time signal. To prevent network jitter from affecting synchronization accuracy, NTP uses multiple time queries and calculates an average value to offset some of the errors caused by network fluctuations. Furthermore, to ensure data reliability, the system appends a timestamp and sequence number to each sensor data packet to check the order and integrity of the packets.

[0077] Redundancy detection mechanisms play a crucial role in data synchronization. If the system detects lost or out-of-order data packets (such as misaligned timestamps), it immediately triggers a retransmission mechanism to request the missing packets from the sensors. If a device becomes unresponsive for an extended period, the system will activate the data stream from a backup device to ensure the continuity of the virtual scenario. For example, in fire simulation training, when trainees' escape routes need to be updated in real time, if a sensor's time is out of sync or its data is lost, the system will use data from other devices to replace it, ensuring that the spread of fire and route guidance are not interrupted.

[0078] This approach ensures time synchronization while offering high fault tolerance. Even in complex network environments, the application of data retransmission and redundancy mechanisms guarantees the continuity of the training process, preventing trainees from being misled by system errors. Furthermore, NTP's low cost and ease of deployment make it ideal for resource-constrained environments such as mines.

[0079] This invention achieves precise matching between sensors and the virtual scene by employing time synchronization algorithms (such as PTP or NTP) and dynamic compensation models, ensuring seamless synchronization between each trainee's action and the virtual screen. This high-precision synchronization is particularly crucial for escape training, as trainees rely on real-time feedback from the system for path selection and hazard avoidance. If sensor data is out of sync with the scene, trainees may choose dangerous routes due to incorrect judgment. However, through high-frequency time calibration and data compensation, even in the event of network fluctuations or brief equipment malfunctions, the system can still guarantee the matching of actions with the scene, avoiding virtual misleading and improving the safety and effectiveness of training.

[0080] This invention, by combining sliding window technology, predictive compensation algorithms, and redundancy detection mechanisms, enables the system to respond quickly and automatically correct errors when faced with packet loss or network latency. This mechanism gives the system high fault tolerance, ensuring continuous training services even in complex network environments. Even if a sensor experiences a temporary malfunction, a backup sensor or predictive model can quickly step in, preventing training interruptions due to equipment problems. This feature not only improves system stability but also reduces the cost of restarting training after interruptions, enhancing training efficiency and the learning experience for participants.

[0081] This invention utilizes the A* algorithm combined with a dynamic time synchronization scheme, enabling the system to replan escape routes in real time based on the fire's spread and the trainees' locations. Due to the guaranteed precision of time synchronization, the system can complete path calculation and feedback within milliseconds and immediately update escape guidance in the virtual scenario. When trainees deviate from the route or the fire situation suddenly changes, the system quickly adjusts and corrects the trainees' actions through visual or voice prompts. This real-time response capability greatly enhances the immersiveness of training, providing trainees with a more realistic drill experience and cultivating their ability to make rapid decisions and act in emergency situations.

[0082] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A coal mine fire prevention simulation training method based on virtual reality technology, used to solve the problem that the virtual scene is out of sync with the operation of the trainee due to sensor data delay or misalignment during the training process, characterized in that, The method comprises the following steps: Real-time acquisition of student motion data, including posture changes, hand movements, and position coordinate information; and denoising and formatting of the collected raw data through a preprocessing module; Monitoring the transmission delay of sensor data and detecting abnormal conditions during network transmission; if the delay is found to exceed the set threshold, dynamically compensating the data through time stamping and a prediction model to generate temporary position information, thereby avoiding data loss or scene misplacement caused by short-term network fluctuations; Based on the Kalman filter algorithm, the fire spread path is predicted, the current position and movement trend of the student are taken as input parameters of the algorithm, and the speed and direction of the fire spread are dynamically calculated; ensuring that the fire scene and the escape route of the student are updated synchronously in the virtual environment; An A* path planning algorithm is used to calculate the optimal escape route in real time, and the escape route is dynamically adjusted according to the current position of the student and the trend of the fire spread; If the student deviates from the planned route during the escape process, the student is corrected through voice and visual prompts; A time synchronization algorithm is used to accurately match the sensor data with the simulation data of the virtual scene, ensuring the clock consistency between different devices; a synchronization correction algorithm is used to eliminate the time deviation between the sensor and the system, achieving seamless synchronization between the scene and the student's operation; The specific steps of dynamically calculating the optimal escape route using the A* path planning algorithm are as follows: First, a grid map model is constructed in the virtual mine scene, each position in the scene is divided into nodes, each node represents a specific position in the mine, the connection between each node is defined as an edge, and each edge is assigned a different generation value. At the initialization stage, the system marks the current position and target point of the student, and identifies the known obstacle areas; According to the trend of fire simulation, the nodes on the edge of the fire source in the map are adjusted in real time, and the weight of each node is composed of two parts: G value and H value. G value is the actual cost from the starting point to the node, and H value is the estimated cost from the node to the target point. As the fire spreads, the cost of the nodes on the edge of the fire source will increase, and the system will mark the nodes on the edge of the fire source as high-risk areas or impassable areas, forcing the A* algorithm to avoid dangerous paths; During path planning, the A* algorithm continuously expands search nodes from the student's current position to the target point to search for a path; The A* algorithm selects the node with the smallest total cost G value + H value for expansion each time, and checks whether the adjacent nodes can provide a lower cost path. If a better path is found, the system updates the path information and adds the node to the open list for further expansion until the target point is reached. Once an optimal path is calculated, the system will immediately display it in the virtual scene and guide the student to escape along the route.

2. The coal mine fire prevention simulation training method based on virtual reality technology according to claim 1, characterized in that, Posture changes refer to the rotation and inclination of the student's body or head in three-dimensional space; hand movements refer to the changes in the student's hands during training; position coordinate information refers to the student's position changes in a three-dimensional virtual space, described using X, Y, and Z axis coordinate systems.

3. The coal mine fire prevention simulation training method based on virtual reality technology according to claim 1, characterized in that, The specific steps of monitoring the transmission delay of sensor data and dynamically compensating are as follows: First, the action data of the trainee is collected from multiple sensors, and a timestamp is attached to each frame of data to record the exact time point of its generation; By monitoring the timestamp of each frame of data and its reception time when it arrives at the VR system in real time, the delay in the transmission process is calculated; When sensor data delay or packet loss is detected, the system calls the dynamic compensation algorithm to make predictions based on previously collected action data; After generating temporary data, the temporary data is directly applied to the virtual scene, while waiting for the missing or delayed data packets to arrive again. Once the original data packet arrives, the system will automatically correct the position information in the scene, making the virtual environment consistent with the real operation. If the original data fails to arrive within the set time, the temporary prediction data will be considered valid, and the trainee will be notified through the sensor feedback system to ensure their sensitivity to changes in the scene.

4. The coal mine fire prevention simulation training method based on virtual reality technology according to claim 1, characterized in that, The steps for predicting the fire spread path based on Kalman filtering algorithm are as follows: In the Kalman filter algorithm, first define the state variable and noise model of the system, set the fire spread path as continuous motion in two or three-dimensional space, the state vector Describe the position k and velocity component of the flame at the first time Secondly, the current position of the student is also needed as the system input to evaluate the relative distance between the flame and the student; The noise in the system is modeled as process noise and measurement noise , where Q and R are noise covariance matrices to describe the uncertainty of the flame spread and sensor measurement, and the state transition model is: , where A is the state transition matrix, B is the input control matrix to ensure that the relative position of the fire and the student jointly affects the spread trend of the fire; In each iteration, the next state of the fire is calculated using the prediction phase of the Kalman filter algorithm, the prediction equation is: where, is the predicted position and velocity based on the current state, reflecting the spreading trend of the fire without additional measurement data, at the same time, the system updates the covariance matrix, the expression of the updated is: , ensuring that the system estimate contains uncertainty, taking into account the influence of process noise.

5. The coal mine fire prevention simulation training method based on virtual reality technology according to claim 4, characterized in that, When new measurement data arrives, the system uses the Kalman gain Adjust the state prediction to synchronize the virtual scene with the actual fire dynamics. The formula for calculating the Kalman gain is: The updated state after measurement is: Where, is the current flame measurement position, H is the measurement matrix, and the updated covariance is: Through measurement updating, the system will real-time correct the fire spread path and ensure that the direction and speed of the fire in the virtual scene are synchronized with the actions of the trainees.

6. The coal mine fire prevention simulation training method based on virtual reality technology according to claim 1, characterized in that, During the trainee's escape, the system continuously monitors the deviation of their current position from the planned route. If it detects that the trainee is not following the guided path, the system will immediately trigger a correction mechanism, including voice prompts and visual feedback, to guide the trainee back to the correct path. If the trainee continues to deviate, the A* algorithm will be called again to calculate a new escape route based on the trainee's current position and the state of fire spread, and the guidance will be updated in real time.

7. The coal mine fire prevention simulation training method based on virtual reality technology according to claim 1, characterized in that, The specific steps for using time synchronization algorithm to accurately match sensor and virtual scene data are as follows: In the initialization phase, use Network Time Protocol or Precision Time Protocol to synchronize all sensor devices with the main system clock; When collecting the trainee's operation data, the sensor attaches a timestamp to each frame of data to accurately record the time point of its generation; Use the time synchronization algorithm to correct the clock error of each device and the main system; After all data is time-corrected, the sensor data is fused and matched with the simulation data in the virtual scene to ensure that each frame of data can be accurately applied to the virtual environment in chronological order.

Images