A method and system for positioning personnel in underground mines
By obtaining underground environment and personnel positioning information of the mine, combining triangulation and drone technology, accurate positioning and safety guarantee of underground personnel is achieved, solving the problems of inaccurate positioning and low response efficiency in the existing technology, and significantly improving the mine Safety management and rescue capabilities.
Patent Information
- Application Number
- CN202411100367.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-12
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-12
AI Technical Summary
The underground environment of the mine is complex, and the existing positioning methods are difficult to achieve accurate positioning. In emergency situations, the response efficiency is low, the warning accuracy is poor, and the safety hazards that cannot be effectively dealt with the comprehensive impact of multiple factors.
By obtaining underground environment information and personnel positioning information of mines, analyzing the failure of positioning information, using triangulation or multilateral measurement technology to accurately track the location of failed personnel, and using environmental information to determine the warning level, using drones to charge and sound-light alarms, improve the Rescue efficiency and targetedness.
It has achieved accurate tracking and safety guarantees of underground personnel, improved the efficiency of mine safety management, reduced personnel safety risks, and improved the timeliness and intelligence of rescue.
Smart Images

Figure CN118646995B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of mine safety, and in particular to a method and system for positioning personnel underground in a mine. Background Art
[0002] As an important resource mining site, the safety of mines has always attracted much attention. In the process of underground mine operations, accurate positioning of personnel is the key to ensuring safety. However, due to the complex and changeable environment of underground mines, traditional positioning methods are often subject to many limitations, such as signal interference and low positioning accuracy. Therefore, how to provide a method for positioning underground personnel in mines that can achieve accurate positioning in complex environments has become a technical problem that needs to be solved urgently.
[0003] Existing methods for locating personnel in underground mines mainly rely on various sensors and positioning systems, but these methods still have many shortcomings in practical applications. For example, some methods may not be able to detect the failure of positioning information in time, resulting in an inability to respond quickly in emergency situations. In addition, when determining the warning level, existing methods often only consider a single environmental parameter and ignore the combined influence of multiple factors, thereby reducing the accuracy and effectiveness of the warning. Summary of the invention
[0004] In order to achieve accurate positioning and safety protection for underground mine personnel and improve emergency response efficiency and personnel safety, the present application provides a method and system for positioning underground mine personnel.
[0005] In the first aspect, the present application provides a method for locating personnel in a mine, which adopts the following technical solution:
[0006] A method for positioning personnel in an underground mine, comprising:
[0007] Obtaining underground mine environment information and personnel location information, where underground mine environment information includes temperature, humidity, and related gas concentrations;
[0008] Analyze whether the personnel location information is invalid;
[0009] If not, the personnel location information is sent back to the command center periodically;
[0010] If yes, then obtain information about the last known location where the positioning information is invalid based on the personnel positioning information analysis, and obtain environmental information about the location area where the invalid last known location is located;
[0011] Analyze and determine the warning level according to the correspondence between the key parameter data range and the warning level of the key parameter data contained in the environmental information of the location area;
[0012] The warning level, the last known location when the positioning information is invalid, and the environmental information are sent back to the command center.
[0013] By adopting the above technical solutions, accurate tracking and safety protection of underground personnel are achieved by comprehensively acquiring environmental information and personnel positioning information. When the positioning information is valid, the personnel location can be sent to the command center regularly to ensure real-time monitoring; once the positioning information fails, the system can quickly analyze the last location before the failure, and combine the environmental information of the area to intelligently determine the warning level, and promptly feedback key information to the command center to provide strong support for rescue work. This method not only improves the efficiency of mine safety management, but also significantly reduces personnel safety risks.
[0014] Optionally, the information of the last known location of the person whose positioning information is invalid may be obtained by analyzing the positioning information of the person, including:
[0015] Analyze whether there are multiple people in the mine, define that all people are equipped with positioning devices and all positioning devices can send location data;
[0016] If no, continue with the next steps;
[0017] If yes, the locations of the remaining people are used as node locations. When the location information of one person fails, triangulation or multilateral measurement is performed using the data of at least three other nodes with known locations that are not on the same straight line.
[0018] The relative distance between the failed device and each reference point is measured, and a triangular or polygonal geometric model is constructed based on the measured relative distance. The positioning information of the failed person is analyzed using the geometric relationship as the information of the last known position of the failed positioning information.
[0019] By adopting the above technical solution, through clever triangulation or multilateral measurement technology, it is possible to accurately track the position of the failed personnel when the positioning information of the personnel in the mine fails. Using the positions of other personnel as reference points, constructing a geometric model, and analyzing the positioning information of the failed personnel, this innovative method not only improves the positioning accuracy, but also enhances the timeliness and pertinence of rescue. The application of this solution will significantly improve the intelligent level of mine safety management and provide strong technical support for protecting the lives of personnel.
[0020] Optionally, analyzing and determining the warning level according to the correspondence between the key parameter data range within which the key parameter data included in the environmental information of the location area falls and the warning level includes:
[0021] Obtain the power of the positioning device before failure;
[0022] The warning level is determined by analyzing the correspondence between the power range that the positioning device's power before failure falls into, the key parameter data range that the key parameter data contained in the environmental information of the location area falls into, and the warning level.
[0023] By adopting the above technical solution, a more accurate early warning level determination is achieved by comprehensively considering the power level of the positioning device before failure and the environmental information of the area where it is located. The joint analysis of the power range and key environmental parameter data not only improves the accuracy and pertinence of the early warning, but also provides timely early warning information before the positioning device fails, buying valuable time to deal with potential risks. The application of this solution will effectively improve the level of safety management, reduce the safety hazards caused by the failure of the positioning device, and ensure the safety of personnel and equipment.
[0024] Optionally, according to the correspondence between the power interval range that the power of the positioning device falls into before failure, the key parameter data range that the key parameter data contained in the environmental information of the location area falls into, and the warning level, analyzing and determining the warning level includes:
[0025] Obtain weather condition information within a future time range, wherein the weather condition information includes rainfall intensity and wind speed;
[0026] The warning level is determined based on the correspondence between the weather condition information within the future time range, the power range within which the positioning device's power before failure falls, the key parameter data range within which the key parameter data contained in the environmental information of the location area falls, and the warning level.
[0027] By adopting the above technical solutions, a more comprehensive and accurate warning level determination is achieved by integrating future weather conditions, positioning device power and key environmental parameters. The introduction of rainfall intensity and wind speed enables the early warning system to respond to potential risks brought by severe weather in advance, and combines power and environmental data to form a multi-dimensional early warning analysis system. This solution not only improves the accuracy and foresight of early warnings, but also provides more detailed data support for emergency response, thereby effectively reducing safety hazards caused by sudden weather changes and failure of positioning devices.
[0028] Optionally, steps in parallel with analyzing and determining the warning level are also included, as follows:
[0029] Analyze whether the power of the positioning device before failure falls within a preset power range;
[0030] If yes, the drone carrying the charging device for the positioning device is started to fly to the last known location where the positioning information is invalid, and the sound and light alarm is activated to remind the missing person to come over.
[0031] By adopting the above technical solution, the parallel step intelligently analyzes the power of the positioning device before failure, promptly starts the drone carrying the charging device to fly to the last known location of the missing person, and activates the sound and light alarm, thereby achieving rapid positioning and rescue of the missing person. This innovative method not only improves the efficiency of rescue, but also effectively reminds the missing person through sound and light alarm, enhancing the pertinence and effectiveness of the rescue. This solution will significantly improve the rescue capability after the positioning device fails, and provide strong support for ensuring the safety of personnel.
[0032] Optionally, the method further includes steps after initiating an audible and visual alarm to remind the missing person to pass by, as follows:
[0033] Analyze whether the time to activate the sound and light alarm reaches the preset time;
[0034] If yes, the drone is started to inspect within a preset distance range with the current location as the center;
[0035] If a person is detected within the preset distance range, image recognition analysis is performed to determine whether the person is a missing person;
[0036] If yes, the charging device carried will be placed at the location of the missing person for the missing person to charge the positioning device, and the missing person will be guided to a safe location through flight;
[0037] If not, analyze whether the warning level has reached the highest level;
[0038] If not, the warning level is raised by one level;
[0039] If yes, maintain the current warning level.
[0040] By adopting the above technical solutions, efficient search and rescue and safety protection for missing persons are achieved. After the sound and light alarm is activated, the drone is intelligently started to conduct range inspections through the judgment of the preset time, and the missing persons are accurately identified by combining image recognition technology. After confirming the location of the missing person, not only a charging device is provided to restore the positioning function, but also flight guidance is used to ensure the safe transfer of personnel. At the same time, the dynamic adjustment mechanism of the warning level in the solution can flexibly increase or maintain the warning level according to actual conditions, thereby achieving precise control of the rescue operation. This solution will significantly improve the efficiency and safety of rescue and provide all-round protection for the missing persons.
[0041] Optionally, the method further includes the following steps in parallel with starting the drone to inspect within a preset distance with the location as the center of the circle:
[0042] Activate the sound collection device preset on the drone to capture and collect sounds;
[0043] The captured sound is identified as a human voice through voice recognition technology;
[0044] If yes, further extract the directional features of the sound, and input the directional features of the sound into a trained deep learning model for inferring the specific direction and distance of the sound source, and output the inferred specific direction and distance of the sound source;
[0045] Start the drone to fly in the direction and distance of the inferred sound source and identify whether there is a missing person;
[0046] If yes, the charging device carried will be placed at the location of the missing person for the missing person to charge the positioning device, and the missing person will be guided to a safe location through flight;
[0047] If not, analyze whether the warning level has reached the highest level;
[0048] If not, the warning level is raised by one level;
[0049] If yes, maintain the current warning level.
[0050] By adopting the above technical solution, this parallel step realizes the efficient capture and positioning of the voice of the missing person by starting the sound collection and recognition technology on the drone. Combined with the deep learning model, the specific direction and distance of the sound source are accurately inferred, guiding the drone to quickly approach the missing person. After confirming the location of the missing person, a charging device is provided in time and guided to a safe area, ensuring the timeliness and effectiveness of the rescue. At the same time, the dynamic adjustment mechanism of the warning level can respond flexibly according to the actual situation, improving the intelligence and accuracy of the entire rescue process. This solution will significantly improve the efficiency of search and rescue and provide strong safety protection for the missing persons.
[0051] In the second aspect, the present application provides a mine underground personnel positioning system, which adopts the following technical solution:
[0052] A mine underground personnel positioning system comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can implement the mine underground personnel positioning method as described in the first aspect when loaded and executed by the processor. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 It is a flow chart of a method for locating personnel underground in a mine according to an embodiment of the present application.
[0054] Figure 2 It is a flowchart of another embodiment of the present application for obtaining the information of the last known location when the positioning information is invalid based on the personnel positioning information analysis.
[0055] Figure 3This is a flowchart of another embodiment of the present application for analyzing and determining the warning level according to the correspondence between the key parameter data range into which the key parameter data contained in the environmental information of the location area falls and the warning level.
[0056] Figure 4 Another embodiment of the present application is a flowchart for analyzing and determining the warning level based on the correspondence between the power range within which the power of the positioning device before failure falls, the key parameter data range within which the key parameter data contained in the environmental information of the location area falls, and the warning level. DETAILED DESCRIPTION
[0057] The present application is further described in detail below in conjunction with the accompanying drawings.
[0058] Reference Figure 1 , is a method for locating personnel in an underground mine disclosed in the present application, comprising:
[0059] Step S100, obtaining underground mine environment information and personnel location information.
[0060] Among them, the underground environmental information of the mine includes temperature, humidity, and related gas concentrations. The underground environmental information of the mine can be obtained by deploying multiple sensor nodes underground in the mine, monitoring the environmental data in real time and uploading it to the central control system. The personnel positioning information can be obtained through positioning devices such as GPS equipped by underground personnel.
[0061] Step S200, analyzing whether the personnel location information is invalid. If not, executing step S300; if yes, executing step S400.
[0062] Among them, whether the personnel positioning information is invalid can be determined by the central control system regularly receiving and verifying the GPS positioning data. If the data is found to be abnormal or lost, the positioning information is determined to be invalid, and the subsequent early warning and rescue mechanism is triggered.
[0063] Step S300: regularly sending personnel location information back to the command center.
[0064] When the positioning information is valid, this step is responsible for regularly transmitting the real-time location data of the underground personnel to the command center so that the management can conduct real-time monitoring and scheduling. For example, a fixed time interval is set, such as every 5 minutes, to send the GPS positioning data of the underground personnel to the command center through the wireless communication network to achieve real-time update and monitoring of the location.
[0065] Step S400, obtaining information on the last known location of the invalid positioning information based on personnel positioning information analysis, and obtaining environmental information on the location area where the invalid last known location is located.
[0066] Specifically, when the GPS positioning data fails, the system automatically traces back to the last valid positioning point, and combines the underground map and environmental sensor data to analyze the specific area where the personnel may be located and its environmental conditions.
[0067] Step S500: Analyze and determine the warning level according to the corresponding relationship between the key parameter data range within which the key parameter data contained in the environmental information of the location area falls and the warning level.
[0068] Specifically, the warning level can be set to three levels: high-level warning, low-level warning, and general reminder. It can also be set to level one, level two, and level three.
[0069] The analysis and determination of the warning level is as follows: the key parameter data contained in the environmental information of the location area is analyzed to determine the key parameter data range within which the key parameter data contained in the environmental information of the location area falls, and the key parameter data range within which the key parameter data contained in the environmental information of the location area falls is used as a query object, and the warning level is obtained by querying from a preset database that stores the correspondence between the key parameter data range within which the key parameter data contained in the environmental information of the location area falls and the warning level.
[0070] For example, if the environmental information in the failure location area shows that the temperature is abnormally high and the concentration of harmful gases exceeds the standard, the system will determine it as a high-level warning based on the preset standards and immediately initiate the corresponding emergency rescue plan.
[0071] Step S600, the warning level, the last known location when the positioning information is invalid, and the environmental information are sent back to the command center.
[0072] For example, once the warning level and failure location are determined, the system immediately sends this critical information to the command center's large-screen display system via a dedicated communication line so that management can quickly make rescue deployments.
[0073] In addition, the last known location of the personnel location information obtained through analysis includes:
[0074] Step S410, start environmental feature matching, compare the current environmental feature with pre-stored environmental feature data and obtain the environmental feature matching result.
[0075] Specifically, step S410 is to start environmental feature matching. The core of this step is to compare the currently collected environmental feature data with the environmental feature data pre-stored in the map database to find the closest match. This process is crucial for determining the last known location of a person when the positioning information fails.
[0076] Environmental feature data may include Wi-Fi signal strength, Bluetooth beacon ID, magnetic field strength, etc. These data can reflect the characteristics of a specific location to a certain extent, and therefore can be used as a basis for location identification.
[0077] During the environmental feature matching process, the system will traverse all pre-stored environmental feature data in the map database and compare each data item with the currently collected environmental feature data. This comparison is performed through a specific algorithm that can comprehensively consider the similarities and weights of various environmental features to obtain a matching score.
[0078] Suppose we have deployed multiple Wi-Fi hotspots and Bluetooth beacons in a mine, and stored the signal strength, ID and other information of these hotspots and beacons in the map database in advance. At a certain moment, a miner's positioning device fails for some reason, and the system automatically starts the environmental feature matching program.
[0079] The program first collects information such as the Wi-Fi signal strength and Bluetooth beacon ID of the current environment. For example, the currently collected data may be as follows: the signal strength of the nearest Wi-Fi hotspot A is -65dBm, and the ID of Bluetooth beacon B is "Beacon001".
[0080] Next, the system compares this set of data with the pre-stored environmental feature data in the map database. In the database, each location is associated with a set of environmental feature data. For example, the database stores such a data item: location C, the corresponding Wi-Fi hotspot A has a signal strength of -67dBm, and the ID of Bluetooth beacon B is "Beacon001".
[0081] Through a specific matching algorithm (such as a comprehensive matching calculation based on signal strength and ID), the system finds that the currently collected environmental feature data is very close to the pre-stored environmental feature data of location C, so it preliminarily determines that the miner's last known location may be near location C. This matching result will be used as the basis for the credibility evaluation in the subsequent step S420.
[0082] The following is an example of a specific matching algorithm that may be used to calculate the combined matching score based on Wi-Fi signal strength and Bluetooth beacon ID:
[0083] Regarding the signal strength matching part, assuming that the difference threshold of Wi-Fi signal strength is ±5dBm, the signal strength of Wi-Fi hotspot A currently collected is -65dBm, which is compared with the signal strength of Wi-Fi hotspot A corresponding to position C in the database, which is -67dBm. The difference value is: |-65-(-67)|=2dBm. Since 2dBm is less than the threshold of 5dBm, the matching degree of Wi-Fi signal strength can be given a higher value, such as 0.8 (the full score is 1).
[0084] Regarding the Bluetooth beacon ID matching part, if the Bluetooth beacon ID is completely matched (that is, the ID of the currently collected Bluetooth beacon B is "Beacon001", which is the same as the ID of the Bluetooth beacon B corresponding to position C in the database), the matching degree of this part can be directly given as 1.
[0085] Regarding the calculation of the comprehensive matching degree, the comprehensive matching degree can be calculated by weighted average. Assume that the weight of the Wi-Fi signal strength is 0.6 and the weight of the Bluetooth beacon ID is 0.4. Then the comprehensive matching degree = 0.6×0.8+0.4×1=0.48+0.4=0.88. Through this comprehensive matching degree of 0.88, it can be judged that the currently collected environmental feature data is close to the pre-stored environmental feature data of location C.
[0086] Another example: The signal strength of Wi-Fi hotspot A currently collected is -70dBm, which is compared with the signal strength of Wi-Fi hotspot A corresponding to position C in the database, which is -67dBm. The difference is: |-70-(-67)|=3dBm, which is less than the threshold of 5dBm, and the matching degree of this part is given as 0.7. The Bluetooth beacon ID is still completely matched, and the matching degree is 1. When the weight remains unchanged, the comprehensive matching degree = 0.6×0.7+0.4×1=0.42+0.4=0.82, which still indicates a high degree of matching.
[0087] Step S420: Analyze and determine the credibility of the environmental feature matching based on the result of the environmental feature matching and a preset credibility evaluation algorithm.
[0088] The credibility assessment algorithm here is a comprehensive consideration of multiple factors, which usually includes the following aspects of assessment:
[0089] Similarity evaluation: The system compares the similarity between the currently collected environmental feature data and the pre-stored data. For example, in an underground mine environment, if the currently detected Wi-Fi signal strength, Bluetooth beacon ID, and magnetic field strength are highly consistent with the data of a certain location in the map database, the similarity score will be high.
[0090] Stability considerations: Certain environmental characteristics, such as magnetic field strength, may be relatively stable in a specific area. If the matching results are consistent for multiple consecutive times, the credibility is increased.
[0091] Data freshness: For pre-stored environmental feature data, its update time and frequency will be considered. Newer data is more likely to reflect the actual situation of the current environment, thus increasing the credibility of the match.
[0092] For example, assume that there is an area A in the mine, where the pre-stored Wi-Fi signal strength is -65dBm, the Bluetooth beacon ID is XYZ123, and the magnetic field strength is 50μT. When the system detects that the positioning fails and starts environmental feature matching, it obtains data similar to area A: the Wi-Fi signal strength is -63dBm, the Bluetooth beacon ID is also XYZ123, and the magnetic field strength is 48μT.
[0093] The credibility evaluation algorithm may be as follows:
[0094] Similarity score = (1-|current Wi-Fi signal strength-prestored Wi-Fi signal strength| / signal strength range) + (Bluetooth beacon ID match? 1:0) + (1-|current magnetic field strength-prestored magnetic field strength| / magnetic field strength range)
[0095] Considering that the signal may be negative or vary widely, the absolute value is used in the algorithm to calculate the difference, and each feature difference is normalized according to the normal variation range of the feature.
[0096] Assuming that the signal strength range is usually -100dBm to 0dBm, the magnetic field strength range is 0 to 100μT, and a perfect match of the Bluetooth beacon ID is scored as 1, and a mismatch is scored as 0. Similarity score = (1-|-63-(-65)| / 100)+(1)+(1-|48-50| / 100)=0.98+1+0.98=2.96.
[0097] Stability considerations may involve the homogeneity of multiple measurements. Assuming that in the past 5 minutes, the system has measured three times and the matching degree of the features with area A is above 95%, which will further improve the credibility. In terms of data freshness, if the data of area A is updated within the past week, the data is considered to be relatively fresh, which will add additional points to the credibility assessment.
[0098] Step S430: If the assessed credibility exceeds a preset credibility threshold, the last known position is determined based on the result of the environmental feature matching.
[0099] Specifically, when the environmental feature matching program completes the matching and obtains the matching result, the system will calculate the credibility of the matching result based on the preset credibility evaluation algorithm (such as similarity evaluation, stability evaluation, etc.). This credibility score reflects the accuracy and reliability of the matching result.
[0100] If the assessed confidence exceeds the preset confidence threshold, the system will consider the environmental feature match result to be reliable, and will directly determine the person's last known location based on this match result. This approach can provide a relatively accurate location estimate when the positioning information fails.
[0101] Example (based on underground mine information):
[0102] Assume that in an underground mine environment, we have established a detailed environmental feature database and set the credibility threshold to 80%. When the location information of a miner fails, the system automatically starts the environmental feature matching program and obtains a matching result.
[0103] After calculation by the credibility evaluation algorithm, the credibility of the matching result is 85%, which exceeds the preset 80% threshold. Therefore, in step S430, the system will directly adopt this high credibility matching result as the last known location of the miner.
[0104] Specifically, if the matching result shows that the miner’s last location was in a specific area of the mine (such as area A), and the credibility of this result exceeds the threshold, the system will assume that the miner’s last known location before the positioning information expired was area A.
[0105] Step S440, if the assessed credibility is less than or equal to the preset credibility value, switch to the sensor data assisted mode to use the various sensor data built into the positioning device for positioning compensation. In this mode, the inertial navigation algorithm is used to calculate the movement path and posture changes of the positioning device since the last known position, thereby obtaining a relative position estimate as the determined last known position.
[0106] Step S440 is a positioning compensation mechanism adopted when the reliability of the environmental feature matching is insufficient. When the reliability of the environmental feature matching determined in step S420 is lower than the preset threshold, the system will switch to the sensor data auxiliary mode. In this mode, the built-in sensor data of the positioning device, such as accelerometers, gyroscopes, etc., will be used to estimate the position.
[0107] Specifically, in this mode, the system starts the inertial navigation algorithm. This algorithm can analyze the sensor data to infer the movement path and attitude change of the positioning device since the last known position. By integrating and processing data such as acceleration and angular velocity, the displacement and direction change of the device during the failure period can be estimated, thereby giving a relative position estimate.
[0108] The steps of the inertial navigation algorithm are as follows:
[0109] Initial attitude determination: Assume that the initial known position is (X0, Y0, Z0), and the initial attitude is (pitch angle θ0, roll angle φ0, yaw angle ψ0).
[0110] Acceleration integration: The acceleration value (ax, ay, az) measured by the accelerometer is integrated twice to get the displacement change. Assume that the acceleration measurement value over a period of time is as follows (unit: m / s²):
[0111] At t1: ax1=1.5, ay1=0.8, az1=-0.5;
[0112] At t2: ax2=2.0, ay2=-0.3, az2=0.2;
[0113] At t3: ax3=1.2, ay3=0.6, az3=-0.1;
[0114] Calculate the displacement change by integration: First integrate the acceleration to get the velocity: v(t)=v(0)+∫a(t)dt; then integrate the velocity to get the displacement: s(t)=s(0)+∫v(t)dt; but due to errors in acceleration measurement, the errors in the integration results will accumulate over time.
[0115] Angular velocity integration: The angular velocity values (ωx, ωy, ωz) measured by the gyroscope are integrated to update the attitude information. Assume that the angular velocity measurements over a period of time are as follows (unit: rad / s):
[0116] At time t1: ωx1=0.15, ωy1=-0.05, ωz1=0.08;
[0117] At t2: ωx2=0.20, ωy2=0.03, ωz2=-0.06;
[0118] Time t3: ωx3=0.12, ωy3=0.04, ωz3=0.02;
[0119] Sensor data fusion: Use methods such as Kalman filtering to integrate data from accelerometers, gyroscopes, and magnetometers to improve positioning accuracy and stability.
[0120] For example: In a mine, the preset credibility threshold is 0.8, while the credibility assessment result of environmental feature matching is 0.6.
[0121] The initial position is (100, 200, 50) meters, and the initial attitude is (15°, 20°, 30°).
[0122] After the above algorithm is used to calculate, the relative displacement after a period of time is (20, -15, 10) meters, and the attitude change is (pitch angle increases by 5°, roll angle decreases by 3°, yaw angle increases by 2°). The final estimated relative position is (120, 185, 60) meters.
[0123] exist Figure 1 In step S400, further consideration may be given to how to more accurately obtain the information of the last known location where the positioning information fails. Figure 2 The illustrated embodiment is described in detail.
[0124] Reference Figure 2 , according to the personnel location information analysis, the information of the last known location after the location information is invalid is obtained, including:
[0125] Step SA00, analyze whether there are multiple people underground in the mine. If not, execute step SB00; if yes, execute step SC00.
[0126] It is defined that all persons are equipped with a positioning device and all positioning devices can send location data.
[0127] Step SB00, continue to subsequent steps.
[0128] Step SC00, taking the locations of the remaining persons as node locations, when the positioning information of one person fails, triangulation or multilateral measurement is performed using the node data of at least three other known locations that are not on the same straight line.
[0129] In a multi-person scenario, when the positioning information of a certain person fails, this step will use the positions of at least three other known miners that are not on the same straight line as reference points to perform triangulation or multilateral measurement to achieve an accurate estimation of the position of the failed person.
[0130] For example, among three miners underground, suppose the positioning device of one of the miners suddenly fails. At this time, the system will automatically select the positions of the remaining two miners as reference points and perform triangulation to estimate the approximate position of the failed miner.
[0131] Step SD00, measure the relative distance between the failed device and each reference point, and construct a triangular or polygonal geometric model based on the measured relative distance, and use the geometric relationship to analyze the positioning information of the failed person as the information of the last known position of the failed positioning information.
[0132] First, accurate relative distance measurements are performed using the data of at least three nodes with known locations that are not in the same straight line determined in the previous step. These node data come from the positioning devices of other personnel underground in the mine, and their positions are used as reference points at this time. Through advanced ranging techniques, such as measurement methods based on signal propagation time or signal attenuation, we can accurately obtain the relative distance between the failed device and each reference point.
[0133] Next, based on the relative distances obtained from these measurements, we construct a triangular or polygonal geometric model. This model can intuitively reflect the spatial relationship between the invalid person and each reference point. By applying basic principles in geometry, such as the relationship between the side length and angle of a triangle, or the relationship between the side length and internal angle of a polygon, we can further analyze the specific location of the invalid person.
[0134] Let's take an example as follows: suppose there are four people underground in a mine, and the positioning device of one of them suddenly fails. In step SC00, we have determined the positions of the other three people as reference points. Now, in step SD00, we measure the relative distances between the failed device and the three reference points, which are d1, d2, and d3. Next, we construct a triangular geometric model on the plane based on these three relative distances. By applying the relationship between the side length and angle of the triangle, such as the law of cosines or the law of sines, we can calculate the specific position of the failed person in this triangle. This position is the last known position after the positioning information we obtained failed.
[0135] Reference Figure 3 According to the corresponding relationship between the key parameter data range and the warning level, the warning level is analyzed and determined to include:
[0136] Step S5a0, obtaining the power of the positioning device before failure.
[0137] The main task of the system is to obtain the remaining power information of the positioning device before it fails. This usually involves communicating with the positioning device to read its internal power monitoring data. The power data obtained will serve as an important basis for subsequent analysis of the warning level. For example: Suppose a positioning device records that its remaining power is 20% through the built-in power detection module before it fails. This data is then successfully obtained by the system for use in subsequent steps.
[0138] Step S5b0, analyzing and determining the warning level according to the correspondence between the power interval range that the power of the positioning device falls into before failure, the key parameter data range that the key parameter data contained in the environmental information of the location area falls into, and the warning level.
[0139] First, the system will consider the power of the positioning device before it fails and compare it with the preset power range. Secondly, the system will also examine the environmental information of the location area, especially those key parameter data, such as temperature, humidity, air pressure, etc., and determine which key parameter data range these data fall into. Finally, combined with the preset correspondence between the power range and the key parameter data range and the warning level, the system can analyze and determine the current warning level that should be issued.
[0140] For example: Continuing with the above example, the remaining power of the positioning device before failure was 20%, which falls within the power range of the "low power warning". At the same time, the system detects that the ambient temperature in the area where the positioning device is located exceeds the preset "high temperature warning" threshold. Based on the correspondence between these two factors and the warning level, the system may determine to issue a "comprehensive warning", which reflects the dual risks of low power and high temperature environment, thereby reminding relevant personnel to take timely countermeasures.
[0141] Reference Figure 4 According to the corresponding relationship between the power interval range of the power of the positioning device before failure, the key parameter data range of the key parameter data contained in the environmental information of the location area, and the warning level, the warning level is analyzed and determined to include:
[0142] Step S5b1, obtaining weather condition information within a future time range.
[0143] Weather condition information includes rainfall intensity and wind speed, which can be obtained in real time or periodically by interfacing with a weather service provider or using built-in weather sensors (if available).
[0144] For example, suppose the system is set to obtain weather information for the next 24 hours. At a certain moment, the system obtains hourly rainfall intensity and wind speed data for the next 24 hours by querying the API of the weather service provider. This data will be used for subsequent warning level analysis.
[0145] Step S5b2, analyze and determine the warning level based on the correspondence between the weather condition information within the future time range, the power range within which the power of the positioning device falls before failure, the key parameter data range within which the key parameter data contained in the environmental information of the location area falls, and the warning level.
[0146] First, the system will analyze weather conditions in the future, especially rainfall intensity and wind speed, to assess their possible impact on the environment where the positioning device is located. Secondly, the system will combine the power range of the positioning device before failure, and the key parameter data (such as temperature, humidity, etc.) contained in the environmental information of the location area. Finally, based on the comprehensive consideration of these factors and the corresponding relationship between the preset warning levels, the system can analyze and determine the current warning level that should be issued.
[0147] For example: Continuing with the above example, the system has obtained weather information for the next 24 hours. At the same time, the positioning device recorded that its remaining power was 30% before it failed, and this power value fell into the power interval range of "medium power". In addition, the system also detected that the ambient temperature in the area where the positioning device is located is in the "normal temperature" range, but the humidity is slightly high, falling into the "high humidity" environmental parameter data range. Combining this information, as well as the preset warning level correspondence (such as high wind speed + medium power + high humidity = yellow warning), the system finally determines to issue a yellow warning to remind relevant personnel to pay attention to potential risks and take corresponding countermeasures.
[0148] A method for locating personnel in an underground mine also includes steps in parallel with analyzing and determining the warning level, which are as follows:
[0149] Step Sa00, analyzing whether the power level of the positioning device before failure falls within a preset power range.
[0150] The system will first analyze the power of the positioning device before it fails to determine whether it falls within the preset power range. This preset power range can be set according to actual needs, such as the range from low power threshold to full power. Through this step, the system can promptly detect the low power of the positioning device and provide accurate information for subsequent emergency response.
[0151] For example: Assuming that the preset power range is 10%-100%, when the power of a positioning device drops to 15%, the system determines that it falls within the preset power range and then triggers the subsequent emergency response process.
[0152] Step Sb00, if yes, start the drone carrying the charging device of the positioning device to fly to the last known location where the positioning information is invalid, and start the sound and light alarm to remind the missing person to come over.
[0153] If the battery level of the positioning device falls within the preset battery level range, step Sb00 will be executed. In this step, the system will start the drone and let it carry the charging device of the positioning device to fly to the last known location where the positioning information is invalid. At the same time, the drone will activate the sound and light alarm to remind the missing person to pay attention and go to the location. In this way, charging support can be quickly provided for the positioning device with low battery and help find the missing person.
[0154] For example: A drone carrying a charging device flew to the last known location, emitting flashing lights and loud alarms, successfully attracting the attention of the missing person.
[0155] Step Sc00, analyzing whether the time for starting the sound and light alarm reaches the preset time.
[0156] In step Sc00, the system will continue to monitor the start time of the sound and light alarm to determine whether it has reached the preset time. The preset time can be set according to the actual situation to ensure that the missing person has enough time to respond. If the preset time is reached, the system will execute the subsequent steps; otherwise, it will continue to wait.
[0157] For example: Set the preset time to 5 minutes. When the sound and light alarm continues for 5 minutes, the system automatically enters the next step.
[0158] Step Sd00, if yes, start the drone to inspect within a preset distance range with the current location as the center.
[0159] If the sound and light alarm time reaches the preset time, step Sd00 will be triggered. In this step, the drone will patrol the preset distance range with the location as the center of the circle. This patrol method can expand the search range and increase the possibility of finding the missing person.
[0160] For example: The drone cruises within a radius of 100 meters and searches for the missing person through the onboard cameras and sensors.
[0161] Step Se00: If a person is detected within the preset distance range, image recognition analysis is performed to determine whether the person is a missing person.
[0162] If the drone detects a person within the preset distance, the system will immediately perform image recognition analysis to confirm whether it is the missing person. Through image recognition technology, the target person can be quickly and accurately identified, providing key information for subsequent rescue operations.
[0163] For example: A drone discovered a miner during an inspection and confirmed that he was a missing person after comparing him using image recognition technology.
[0164] Step Sf00, if yes, place the carried charging device at the location of the missing person for the missing person to charge the positioning device, and guide the missing person to a safe location area by flying.
[0165] Once the missing person is confirmed to be found, step Sf00 will be executed. In this step, the drone will place the charging device it carries at the location of the missing person so that it can charge the positioning device. At the same time, the drone will also guide the missing person to a safe location through flight to ensure their safe escape.
[0166] For example: A drone places a charging device next to a missing person and sends a guidance signal to guide him to a nearby safe exit.
[0167] Step Sg00, if no, then analyze whether the warning level has reached the highest level. If no, then execute step Sh00; if yes, then execute step Si00.
[0168] If the drone does not find the missing person within the preset distance range, step Sg00 is executed. In this step, the system analyzes whether the current warning level has reached the highest level. If it has not reached the highest level, the subsequent steps are executed to increase the warning level; if it has reached the highest level, the current warning level is maintained unchanged. In this way, the system can flexibly adjust the warning level according to the actual situation to better respond to emergencies.
[0169] For example: The current warning level is level 2, and the system determines that it has not reached the highest level (level 3), so it decides to perform subsequent steps to increase the warning level.
[0170] Step Sh00, raising the warning level by one level.
[0171] If the warning level does not reach the highest level, the system will raise the warning level by one level. This helps to improve the response speed and attention to emergency situations, ensuring that relevant personnel can take quick measures to deal with potential risks.
[0172] For example: The system raises the warning level from level 2 to level 3, triggering a higher level of emergency response process.
[0173] Step Si00, maintain the existing warning level.
[0174] A method for locating personnel in a mine, characterized in that it also includes the steps of starting a drone to patrol a preset distance range with the location as the center of the circle, which are as follows:
[0175] Step Sj00, start the sound collection device preset on the drone to capture and collect sound.
[0176] In step Sj00, the system will start the sound collection device pre-installed on the drone. This device has a highly sensitive sound capture capability and can accurately capture and collect various sound signals in a noisy mining environment. This step provides basic data for subsequent sound recognition and processing.
[0177] For example: When a drone flies to a certain area underground, the sound collection device successfully captures a faint cry for help, providing key clues for subsequent rescue.
[0178] In step Sk00, the captured sound is identified as a human voice through sound recognition technology.
[0179] Step Sk00 is the process of identifying the captured sound. The system will use advanced sound recognition technology to conduct in-depth analysis of the collected sound to determine whether it is a human voice. This step can effectively filter out non-human noise interference and improve rescue efficiency.
[0180] For example: After being processed by voice recognition technology, the system accurately determines that the captured sound is a human voice and determines the distress information contained in the sound.
[0181] Step Sl00, if yes, further extract the directional characteristics of the sound, and input the directional characteristics of the sound into the trained deep learning model for inferring the specific direction and distance of the sound source, and output the inferred specific direction and distance of the sound source.
[0182] Among them, the acquisition process of the deep learning model is as follows: 1. Data collection: First, a large amount of sound sample data needs to be collected. These data should include human voices and non-human voices in various environments. In particular, for a specific environment such as underground mines, corresponding sound data needs to be collected to train the model to better adapt to this environment. 2. Data annotation: The collected sound samples need to be annotated in detail, including the type of sound (such as human voice, mechanical sound, etc.), the direction of the sound, and the distance. These annotated data will be used as supervisory signals during model training. 3. Model selection: According to the task requirements, select a suitable deep learning model architecture, such as convolutional neural network (CNN). 4. Model training: Use the annotated data set to train the selected deep learning model. Through continuous iterative optimization, the model can accurately extract directional features from the sound signal and infer the specific direction and distance of the sound source. 5. Model evaluation and tuning: After the training is completed, the model needs to be evaluated to verify whether its performance meets expectations. If the performance is poor, the model needs to be tuned, including adjusting model parameters and optimizing training strategies. 6. Model deployment: The deep learning model after evaluation and tuning will be deployed to the actual underground mine personnel positioning system to assist drones in extracting directional features of sound and locating sound sources.
[0183] Extraction of sound directional features: With the help of deep learning models, directional features can be extracted from captured sound signals. Specifically, when the sound signal is input into the trained model, the model will analyze and process it and extract feature information related to the direction of the sound. This feature information may include the arrival time difference of the sound, the sound intensity distribution, etc., which together constitute the directional characteristics of the sound. Through these features, we can accurately infer the specific direction and distance of the sound source, thereby providing strong support for subsequent rescue operations.
[0184] For example: The deep learning model successfully inferred the approximate location and distance of the missing person based on the directional characteristics of the sound, providing clear guidance for the flight direction of the drone.
[0185] Step Sm00, start the drone to fly towards the specific direction and distance of the inferred sound source, and identify whether there is a missing person. If yes, execute step Sn00. If no, execute step So00.
[0186] Step Sm00 is a key step in the rescue operation. After determining the specific direction and distance of the sound source, the system will start the drone to fly in that direction and continuously identify whether there are missing persons during the flight. If the missing person is found, the next step will be executed immediately for rescue; if the missing person is not found, the rescue strategy will be adjusted according to the judgment of the warning level.
[0187] For example: The drone flew in the direction of the inferred sound source and successfully found the missing person on the way, winning valuable time for subsequent rescue work.
[0188] Step Sn00, placing the carried charging device at the location of the missing person so that the missing person can charge the positioning device, and guiding the missing person to a safe location area through flight.
[0189] In step Sn00, once the drone finds the missing person, the system will immediately place the charging device it carries at the location of the missing person so that it can charge the positioning device. At the same time, the drone will also guide the missing person to a safe location through flight to ensure their safe escape. This step fully reflects the humanity and efficiency of the rescue operation.
[0190] For example: The drone accurately placed the charging device next to the missing person and guided him to the safe exit through flight, successfully completing the rescue mission.
[0191] Step So00, analyzing whether the warning level has reached the highest level. If not, executing step Sp00; if yes, executing step Sq00.
[0192] If no missing person is found in step Sm00, the system will execute step So00 to analyze the current warning level. This step is to flexibly adjust the warning level according to the actual situation to better deal with possible risks. The system will determine whether the warning level has reached the highest level, and then decide whether to increase the warning level or maintain the current level.
[0193] For example: The current warning level is level 2. After system analysis, the system believes that it has not reached the highest level, so it decides to raise the warning level to strengthen rescue efforts.
[0194] Step Sp00, raising the warning level by one level.
[0195] In step Sp00, if the warning level has not reached the highest level, the system will raise it by one level. This helps to increase the urgency and importance of the rescue operation, ensuring that relevant personnel can take quick measures to deal with potential risks. At the same time, raising the warning level can also trigger a higher level of emergency response process, providing more resources and support for the rescue operation.
[0196] For example: The system raised the warning level from level 2 to level 3, and then initiated higher-level emergency response measures, including dispatching additional rescue personnel and deploying more rescue equipment.
[0197] Step Sq00, maintain the existing warning level.
[0198] If the warning level has reached the highest level, the system will execute step Sq00, that is, maintain the existing warning level unchanged. In this step, the system will continue to monitor and evaluate the current situation so that necessary response measures can be taken in a timely manner. At the same time, maintaining the highest level of warning will also help ensure that relevant personnel continue to pay attention to and attach great importance to the emergency, providing stable support for rescue operations.
[0199] For example: Since the warning level has reached the highest level (level three), the system decides to maintain the current warning level unchanged and concentrate all efforts on rescue work to ensure the safety of every missing person.
[0200] Based on the same inventive concept, an embodiment of the present invention provides a mine underground personnel positioning system, including a memory and a processor, wherein the memory stores a program that can be executed on the processor to implement the following Figures 1 to 4 Procedure for either method.
[0201] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.
Claims
1. A method for locating personnel in a mine, characterized in that: include: Obtaining underground mine environment information and personnel location information, where the underground mine environment information includes but is not limited to temperature, humidity, and related gas concentrations; Analyze whether the personnel location information is invalid; If not, the personnel location information is sent back to the command center periodically; If yes, then obtain information about the last known location where the positioning information is invalid based on the personnel positioning information analysis, and obtain environmental information about the location area where the invalid last known location is located; According to the correspondence between the key parameter data range within which the key parameter data contained in the environmental information of the location area falls and the warning level, the warning level is determined by analysis, the key parameters include temperature, humidity, and relevant gas concentrations, and the key parameter is determined by analyzing the key parameter data contained in the environmental information of the location area to determine the key parameter data range within which the key parameter data contained in the environmental information of the location area falls, and taking the key parameter data range within which the key parameter data contained in the environmental information of the location area falls as the query object, querying and obtaining the warning level from a preset database storing the correspondence between the key parameter data range within which the key parameter data contained in the environmental information of the location area falls and the warning level; Send the warning level, the last known location when the positioning information fails, and the environmental information back to the command center; The information about the last known location obtained by analyzing the personnel location information includes: Start environmental feature matching, compare the current environmental features with the pre-stored environmental feature data and obtain the environmental feature matching result, where the environmental feature data includes Wi-Fi signal strength, Bluetooth beacon ID and magnetic field strength; According to the result of the environmental feature matching and the preset credibility evaluation algorithm, the credibility of the environmental feature matching is analyzed and determined, wherein the credibility evaluation algorithm is as follows: Similarity score = (1-|current Wi-Fi signal strength-prestored Wi-Fi signal strength| / signal strength range) + (Bluetooth beacon ID match? 1:0) + (1-|current magnetic field strength-prestored magnetic field strength| / magnetic field strength range); If the assessed credibility exceeds a preset credibility threshold, the last known location is determined based on the result of the environmental feature matching; If the assessed credibility is less than or equal to the preset credibility value, the system switches to sensor data assisted mode and uses various sensor data built into the positioning device for positioning compensation. In this mode, the inertial navigation algorithm is used to calculate the movement path and posture changes of the positioning device since the last known position, thereby obtaining a relative position estimate as the determined last known position. It is defined that all people are equipped with positioning devices and all positioning devices can send position data.
2. A method for locating personnel in an underground mine according to claim 1, characterized in that: The information about the last known location obtained by analyzing the personnel location information includes: Analyze whether there are multiple people underground in the mine; If no, continue with the next steps; If yes, the locations of the remaining people are used as node locations. When the location information of one person fails, triangulation or multilateral measurement is performed using the data of at least three other nodes that are known and not on the same straight line. The relative distance between the failed device and each reference point is measured, and a triangular or polygonal geometric model is constructed based on the measured relative distance. The positioning information of the failed person is analyzed using the geometric relationship as the information of the last known position of the failed positioning information.
3. A method for locating personnel in an underground mine according to claim 2, characterized in that: According to the corresponding relationship between the key parameter data range and the warning level, the warning level is analyzed and determined to include: Obtain the power of the positioning device before failure; The warning level is determined by analyzing the correspondence between the power range that the positioning device's power before failure falls into, the key parameter data range that the key parameter data contained in the environmental information of the location area falls into, and the warning level.
4. A method for locating personnel in an underground mine according to claim 3, characterized in that: According to the corresponding relationship between the power range of the positioning device before failure, the key parameter data range of the key parameter data contained in the environmental information of the location area, and the warning level, the warning level is analyzed and determined to include: Obtain weather condition information within a future time range, wherein the weather condition information includes rainfall intensity and wind speed; The warning level is determined based on the correspondence between the weather condition information within the future time range, the power range within which the positioning device's power before failure falls, the key parameter data range within which the key parameter data contained in the environmental information of the location area falls, and the warning level.
5. A method for locating personnel in an underground mine according to claim 3 or 4, characterized in that: It also includes steps that run parallel to the analysis and determination of the warning level, as follows: Analyze whether the power of the positioning device before failure falls within a preset power range; If yes, the drone carrying the charging device for the positioning device is started to fly to the last known location where the positioning information is invalid, and the sound and light alarm is activated to remind the missing person to come over.
6. A method for locating personnel in an underground mine according to claim 5, characterized in that: The steps after the sound and light alarm is activated to remind the missing person to pass by are as follows: Analyze whether the time to activate the sound and light alarm reaches the preset time; If yes, the drone is started to inspect within a preset distance range with the current location as the center; If a person is detected within the preset distance range, image recognition analysis is performed to determine whether the person is a missing person; If yes, the charging device carried will be placed at the location of the missing person for the missing person to charge the positioning device, and the missing person will be guided to a safe location through flight; If not, analyze whether the warning level has reached the highest level; If not, the warning level is raised by one level; If yes, maintain the current warning level.
7. A method for locating personnel in an underground mine according to claim 6, characterized in that: The steps of starting the drone to inspect the preset distance range with the location as the center are also included, as follows: Activate the sound collection device preset on the drone to capture and collect sounds; The captured sounds are then used to identify whether they are human voices through voice recognition technology; If yes, further extract the directional features of the sound, and input the directional features of the sound into a trained deep learning model for inferring the specific direction and distance of the sound source, and output the inferred specific direction and distance of the sound source; Start the drone to fly in the direction and distance of the inferred sound source and identify whether there is a missing person; If yes, the charging device carried will be placed at the location of the missing person for the missing person to charge the positioning device, and the missing person will be guided to a safe location through flight; If not, analyze whether the warning level has reached the highest level; If not, the warning level is raised by one level; If yes, maintain the current warning level.
8. A mine underground personnel positioning system, characterized in that: The invention comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a method for locating personnel in an underground mine as claimed in any one of claims 1 to 7.
