A working face laser radar monitoring array data acquisition system and acquisition method

Abstract

The application discloses a working face laser radar monitoring array data acquisition system and an acquisition method, numbers laser radar probes, starts the fixed laser radar monitoring array data acquisition system of the coal mining face, and the laser radar probes are self-checked; the underground processing terminal controls the laser radar probes of the laser radar to enter a full scanning mode, that is, all the laser radar probes synchronously scan the coal mining face, and the collected point cloud data is asynchronously sent to an underground explosion-proof switch, the underground explosion-proof switch is sent to the underground processing terminal, the underground processing terminal receives and processes the point cloud data, and an initial model of the working face is obtained; through switching of scanning strategies such as the full scanning mode, the flat scanning mode and the feeding mode, the service life of the laser radar is prolonged, the system deployment and maintenance cost is reduced, the data transmission occupied bandwidth is reduced, and the technical problem of short service life of the underground laser radar monitoring equipment in the prior art is solved.

Description

Technical Field

[0001] This invention belongs to the technical field of coal mines and relates to radar monitoring, specifically a working face lidar monitoring array data acquisition system and acquisition method. Background Technology

[0002] Unmanned mining technology, as the ultimate goal of intelligent coal mining, has received widespread attention from the industry, and the technical requirements for supporting equipment and processes are becoming increasingly urgent. To achieve unmanned mining at the working face, comprehensive monitoring and sensing of the underground working face environment is essential. LiDAR, as an emerging spatial mapping and sensing technology, has made significant progress and is widely used in fields such as ground mapping and environmental monitoring. LiDAR's high detection accuracy, fast scanning and modeling speed, and independence from visible light make it highly suitable for scanning and spatial monitoring of underground coal mine working faces.

[0003] Currently, monitoring of the working face relies on sensors and video surveillance. However, video surveillance data acquisition methods are not suitable for acquiring LiDAR scanning monitoring data. The main problems are:

[0004] (1) At present, there is a lack of a working surface space model acquisition system in the industry. The monitoring of the working surface is limited to camera video observation and sensor data acquisition. Such data can only provide an abstract overview of the working surface and cannot directly provide a dynamic data model for dynamic updating of the working surface model and digital twin of the working surface.

[0005] (2) The lifespan of mainstream lidar hardware on the market is approximately 5,000-20,000 hours. If it were to collect data continuously for 24 hours like the cameras deployed on the hydraulic supports of the working face, its working cycle would be approximately 2.28 years. Considering environmental factors such as high dust, high humidity, and strong vibration underground, the actual equipment lifespan would be even shorter. Moreover, the cost of a single lidar scanning probe is as high as tens of thousands of yuan. If the probe becomes a consumable, the overall cost of the monitoring system will increase significantly. In addition, lidar scanners have high measurement accuracy, and when high-density point cloud data is collected in real time, the instantaneous data volume can reach hundreds of megabytes per second. Due to the limitations of factors such as working face undulations and the scanning range of lidar probes, a dozen or even dozens of probes need to be deployed simultaneously within the working face to achieve full working face coverage. If all probes scan and collect data simultaneously, it will put enormous pressure on the underground ring network, and the central processing unit of the underground lidar monitoring system will also find it difficult to process such a large amount of point cloud data in a timely manner, causing system lag or even system crashes. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a working face lidar monitoring array data acquisition system and method, thereby solving the technical problem that existing downhole lidar monitoring equipment can only provide abstract working face overview data and has a short service life.

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0008] A working face lidar monitoring array data acquisition system includes multiple lidar probes installed on one side of the coal mining face, an underground processing terminal, an underground explosion-proof switch, and an underground explosion-proof power supply.

[0009] Each of the aforementioned lidar probes is connected to an underground explosion-proof switch, which is connected to an underground processing terminal. Both the lidar probes and the underground explosion-proof switch are connected to an underground explosion-proof power supply.

[0010] The aforementioned lidar probe is used to synchronously scan the coal mining face and asynchronously send the collected point cloud data to the underground explosion-proof switch.

[0011] The aforementioned explosion-proof switch is used to send the received point cloud data to the downhole processing terminal;

[0012] The downhole processing terminal is used to receive and process point cloud data, control the scanning mode of the lidar probe, and control the wake-up and sleep modes of the lidar probe.

[0013] The aforementioned explosion-proof power supply is used to power the lidar probe and the explosion-proof switch in the well.

[0014] A method for acquiring data from a lidar monitoring array on a working surface includes the following steps:

[0015] Step 1: Number the lidar probes;

[0016] The lidar probe near the tail of the scraper conveyor is numbered A, the lidar probe near the head of the scraper conveyor is numbered C, and the lidar probe between lidar probe A and lidar probe C is numbered Bi, i = 1 to n, where n is a positive integer. The front feed area sentry lidar probe and the rear feed area sentry lidar probe are also determined.

[0017] Step 2: Start the data acquisition system of the fixed lidar monitoring array at the coal mining face, and the lidar probe performs a self-test;

[0018] The self-test includes checking the integrity of the lidar probe itself, communication status, interface condition, and visibility between adjacent lidar probes;

[0019] Step 3: The adjacent lidar probes are mutually calibrated, and the reference coordinate system of the working surface laser scanning model is established with the built-in coordinate system of lidar probe A as the reference.

[0020] Step 4: The underground processing terminal controls all lidar probes to enter full-scan mode, that is, all lidar probes scan the coal mining face synchronously and send the collected point cloud data asynchronously to the underground explosion-proof switch. The underground explosion-proof switch sends the data to the underground processing terminal, which receives and processes the point cloud data to obtain the initial model of the working face.

[0021] Step 5: In real time, determine whether the laser radar probe of the front-end cutting area or the rear-end cutting area has detected the coal mining machine. If yes, enter the cutting mode; otherwise, enter the flat sweeping mode.

[0022] The aforementioned cutting mode involves the coal mining machine reciprocating and cutting at the end of the coal body. The lidar probe at the end of the coal body continuously scans. When the corresponding sentry probe detects that the coal mining machine has moved out of the cutting area, the cutting mode is exited and the flat sweeping mode is entered.

[0023] The described flat scanning mode specifically includes the following steps;

[0024] Step 5.1: Establish a sliding sweeping window based on the width of the coal mining machine, wherein the sliding sweeping window covers the lidar probe near the coal mining machine;

[0025] Step 5.2: The lidar probe inside the sliding scanning window scans the coal mining machine and the nearby coal mining face, while all lidar probes outside the sliding scanning window enter a dormant state.

[0026] Step 5.3: When the coal mining machine moves, the sliding scanning window moves accordingly, thereby waking up a lidar probe in front of the coal mining machine and adding it to the sliding scanning window. When the lidar probe that has moved out of the sliding scanning window detects that the hydraulic support it is located has finished moving and there are no moving objects within the scanning range, it enters a sleep state.

[0027] Step 6: Asynchronously send the point cloud data obtained in Step 5 to the underground explosion-proof switch. The underground explosion-proof switch sends the data to the underground processing terminal. The underground processing terminal receives and processes the point cloud data to obtain the local update model.

[0028] Step 7: Fuse the initial working face model obtained in Step 4 and the local update model obtained in Step 6 to obtain the global working face model.

[0029] This invention includes the following technical features:

[0030] Step 3 specifically includes the following steps:

[0031] Step 3.1: Use all lidar probes Bi of the coal mining face to scan the spatial morphology of the face, measure the distance from the lidar probe to the coal wall, and determine the relative positions of adjacent lidar probes; where: i = 1 to n;

[0032] Step 3.2: Using the built-in coordinate system of lidar probe A as the reference, establish the reference coordinate system of the working face laser scanning model. The other lidar probes will use the spatial shape, coal wall distance and relative position with adjacent probes collected in step 3.1 and perform registration transformation calculations to integrate their own coordinate system into the reference coordinate system of the working face laser scanning model.

[0033] In step 4, obtaining the initial model of the working surface includes the following:

[0034] S1, define the lidar probe A at the tail of the scraper conveyor as the origin of the coordinate system;

[0035] S2, determine the 6 degrees of freedom between adjacent lidar probes. The specific steps are as follows:

[0036] The {Δx, Δy, Δz} between adjacent lidar probes is determined by mutual scanning and identification. The distance between adjacent lidar probes is determined by identifying feature points in the overlapping point cloud region.

[0037] in:

[0038] {Δx,Δy,Δz} represents the spatial positional deviation between the lidar probes;

[0039] Indicates rotational deviation;

[0040] S3, point cloud data is stitched together using the point cloud registration method to obtain the initial model of the working surface;

[0041] The point cloud registration method described above involves extracting feature points from two radar point clouds, calculating the error f(R,T) between the two feature points, and finding the registration matrix that minimizes the global error as the optimal registration matrix.

[0042] in:

[0043] This represents the rotational change vector of the point clouds of two adjacent excitation devices;

[0044] T 3×1 =[Δx Δy Δz] T , represents the translational change vector of the point clouds of two adjacent excitation devices;

[0045] S represents the scaling factor of the point clouds of two adjacent excitation devices;

[0046] O 1×3 Represents {0,0,0};

[0047] The formula for calculating the error f(R,T) is as follows:

[0048]

[0049] in:

[0050] R is the rotational transformation vector of the point clouds of two adjacent excitation devices;

[0051] T represents the translational change vector of the point clouds of two adjacent excitation devices;

[0052] P = {p i |p i ∈R 3×3 Let {i = 1, 2, ..., n} represent the source set A;

[0053] Q = {q j |q j ∈R 3×3 The set of target points B1 is represented by the group of points B1, where j = 1, 2, ..., m.

[0054] m and n represent the sizes of the two point sets, respectively.

[0055] In step 7, the initial working face model obtained in step 4 and the local update model obtained in step 6 are fused to obtain the global working face model, which specifically includes the following steps:

[0056] Step 7.1: Determine the position of the local update model in the working surface according to the LiDAR device number of the local update area, and superimpose the local update model on the corresponding position of the initial model on the working surface to obtain the overlapping area of ​​the two positions;

[0057] Step 7.2: Perform point cloud registration operation on the local update model and the initial model of the working face. After registration, remove the expired point cloud data in the overlapping area between the initial model of the working face and the local update model, and retain the point cloud data of the registered local update model to finally obtain the global model of the working face.

[0058] Compared with the prior art, the beneficial technical effects of this invention are:

[0059] (I) In this invention, by deploying a lidar array at the coal mining face, real-time data can be collected across the entire working face space, providing fundamental data for the dynamic updating of the working face's 3D model and the digital twin system of the working face. This solves the problem that existing underground lidar monitoring equipment can only provide abstract overview data of the working face.

[0060] (II) By switching between scanning strategies such as full scan mode, flat scan mode and cutting mode, the service life of the lidar is extended, the system deployment and maintenance costs are reduced, the data transmission bandwidth is reduced, and the technical problem of short service life of downhole lidar monitoring equipment in the prior art is solved. Attached Figure Description

[0061] Figure 1 This is a schematic diagram of the data acquisition system for the working surface lidar monitoring array of the present invention;

[0062] Figure 2 This is a schematic diagram of the laser radar array data deployment in this invention.

[0063] The specific content of the present invention will be further explained in detail below with reference to the embodiments. Detailed Implementation

[0064] It should be noted that, unless otherwise specified, all components in this invention are those known in the art.

[0065] The following are specific embodiments of the present invention. It should be noted that the present invention is not limited to the following specific embodiments. All equivalent modifications made based on the technical solutions of this application fall within the protection scope of the present invention.

[0066] This invention provides a data acquisition system for a working surface lidar monitoring array, such as... Figure 1 As shown, it includes multiple lidar probes and underground processing terminals, underground explosion-proof switches and underground explosion-proof power supplies installed on one side of the coal mining face;

[0067] Each lidar probe is connected to an underground explosion-proof switch, which is connected to an underground processing terminal. Both the lidar probe and the underground explosion-proof switch are connected to an underground explosion-proof power supply.

[0068] The lidar probe is used to synchronously scan the coal mining face and asynchronously send the collected point cloud data to the underground explosion-proof switch.

[0069] The underground explosion-proof switch is used to send the received point cloud data to the underground processing terminal;

[0070] The downhole processing terminal is used to receive and process point cloud data, control the scanning mode of the lidar probe, and control the wake-up and sleep modes of the lidar probe.

[0071] The explosion-proof power supply for underground applications is used to power lidar probes and explosion-proof switches.

[0072] This invention also provides a method for acquiring data from a lidar monitoring array on a working surface, comprising the following steps:

[0073] Step 1: Number the lidar probes;

[0074] The lidar probe near the tail of the scraper conveyor is numbered A, the lidar probe near the head of the scraper conveyor is numbered C, and the lidar probe between lidar probe A and lidar probe C is numbered Bi, i = 1 to n, where n is a positive integer. The front feed area sentry lidar probe and the rear feed area sentry lidar probe are also determined.

[0075] n is determined based on the width of the coal mining face;

[0076] Step 2: Start the data acquisition system of the fixed lidar monitoring array at the coal mining face, and the lidar probe performs a self-test;

[0077] The self-test includes checking the integrity of the lidar probe itself, its communication status, interface condition, and the visibility between adjacent lidar probes; if all of the above self-test items are within the normal range, then the self-test is passed.

[0078] Step 3: The adjacent lidar probes are mutually calibrated, and the reference coordinate system of the working surface laser scanning model is established with the built-in coordinate system of lidar probe A as the reference.

[0079] Step 4: The underground processing terminal controls all lidar probes to enter full-scan mode, that is, all lidar probes scan the coal mining face synchronously and send the collected point cloud data asynchronously to the underground explosion-proof switch. The underground explosion-proof switch sends the data to the underground processing terminal, which receives and processes the point cloud data to obtain the initial model of the working face.

[0080] Step 5: In real time, determine whether the laser radar probe of the front-end cutting area or the rear-end cutting area has detected the coal mining machine. If yes, enter the cutting mode; otherwise, enter the flat sweeping mode.

[0081] The cutting mode is when the coal mining machine moves back and forth and cuts at the end of the coal body. The lidar probe at the end of the coal body scans continuously. When the corresponding sentry probe detects that the coal mining machine has moved out of the cutting area, it exits the cutting mode and enters the flat sweep mode.

[0082] The flat scan mode specifically includes the following steps;

[0083] Step 5.1: Establish a sliding scanning window based on the width of the coal mining machine, and cover the lidar probes near the coal mining machine.

[0084] Step 5.2: The lidar probe inside the sliding scanning window scans the coal mining machine and the nearby coal mining face, while all lidar probes outside the sliding scanning window enter a dormant state.

[0085] Step 5.3: When the coal mining machine moves, the sliding scanning window moves accordingly, thereby waking up a lidar probe in front of the coal mining machine and adding it to the sliding scanning window. When the lidar probe that has moved out of the sliding scanning window detects that the hydraulic support it is located has finished moving and there are no moving objects within the scanning range, it enters a sleep state.

[0086] Step 6: Asynchronously send the point cloud data obtained in Step 5 to the underground explosion-proof switch. The underground explosion-proof switch sends the data to the underground processing terminal. The underground processing terminal receives and processes the point cloud data to obtain the local update model.

[0087] Step 7: Fuse the initial working face model obtained in Step 4 and the local update model obtained in Step 6 to obtain the global working face model.

[0088] In the above technical solution, by deploying a lidar array in the coal mining face, real-time data can be collected from the entire working face space, providing basic data for the dynamic updating of the working face's 3D model and the digital twin system of the working face. This solves the problem that existing underground lidar monitoring equipment can only provide abstract overview data of the working face. By switching between scanning strategies such as full scan mode, horizontal scan mode, and cutting mode, the service life of the lidar is extended, the system deployment and maintenance costs are reduced, and the bandwidth occupied by data transmission is reduced, thus solving the technical problem of the short service life of existing underground lidar monitoring equipment.

[0089] Specifically, step 3 includes the following steps:

[0090] Step 3.1: Use all lidar probes Bi of the coal mining face to scan the spatial morphology of the face, measure the distance from the lidar probe to the coal wall, and determine the relative positions of adjacent lidar probes; where: i = 1 to n;

[0091] Step 3.2: Using the built-in coordinate system of lidar probe A as the reference, establish the reference coordinate system of the working face laser scanning model. The other lidar probes will use the spatial shape, coal wall distance and relative position with adjacent probes collected in step 3.1 and perform registration transformation calculations to integrate their own coordinate system into the reference coordinate system of the working face laser scanning model.

[0092] In the above technical solution, the lidar probe is installed on the hydraulic support around the underground coal mining face. When the coal mining machine is mining normally, that is, after entering the flat sweep mode, after the coal mining machine moves forward, the lidar probe behind it will move out of the hydraulic support where the flat sweep window is located and move towards the coal wall. After the movement is completed, it will enter the dormant state.

[0093] Specifically, in step 4, the initial model of the working surface includes the following:

[0094] S1, define the lidar probe A at the tail of the scraper conveyor as the origin of the coordinate system;

[0095] S2, determine the 6 degrees of freedom between adjacent lidar probes. The specific steps are as follows:

[0096] The {Δx, Δy, Δz} between adjacent lidar probes is determined by mutual scanning and identification. The distance between adjacent lidar probes is determined by identifying feature points in the overlapping point cloud region.

[0097] in:

[0098] {Δx,Δy,Δz} represents the spatial positional deviation between the lidar probes;

[0099] Indicates rotational deviation;

[0100] S3, point cloud data is stitched together using the point cloud registration method to obtain the initial model of the working surface;

[0101] The point cloud registration method involves extracting feature points from two radar point clouds, calculating the error f(R,T) between the two feature points, and finding the registration matrix that minimizes the global error as the optimal registration matrix.

[0102] in:

[0103] This represents the rotational change vector of the point clouds of two adjacent excitation devices;

[0104] T 3×1 =[Δx Δy Δz] T , represents the translational change vector of the point clouds of two adjacent excitation devices;

[0105] S represents the scaling factor of the point clouds of two adjacent excitation devices;

[0106] O 1×3 Represents {0,0,0};

[0107] The formula for calculating the error f(R,T) is as follows:

[0108]

[0109] in:

[0110] R is the rotational transformation vector of the point clouds of two adjacent excitation devices;

[0111] T represents the translational change vector of the point clouds of two adjacent excitation devices;

[0112] P = {p i|p i ∈R 3×3 Let {i = 1, 2, ..., n} represent the source set A;

[0113] Q = {q j |q j ∈R 3×3 The set of target points B1 is represented by the group of points B1, where j = 1, 2, ..., m.

[0114] m and n represent the sizes of the two point sets, respectively.

[0115] Specifically, in step 7, the initial working face model obtained in step 4 and the local update model obtained in step 6 are fused to obtain the global working face model, which includes the following steps:

[0116] Step 7.1: Determine the position of the local update model in the working surface according to the LiDAR device number of the local update area, and superimpose the local update model on the corresponding position of the initial model on the working surface to obtain the overlapping area of ​​the two positions;

[0117] Step 7.2: Perform point cloud registration operation on the local update model and the initial model of the working face. After registration, remove the expired point cloud data in the overlapping area between the initial model of the working face and the local update model, and retain the point cloud data of the registered local update model to finally obtain the global model of the working face.

[0118] Example:

[0119] This embodiment presents a data acquisition method using a working face lidar monitoring array data acquisition system. The basic conditions of the working face are as follows: the working face is 200 meters long, and a total of 131 hydraulic supports are installed within the working face; the head direction of the scraper conveyor is at the left end of the working face, and the tail direction is at the right end of the working face; based on the site conditions, one lidar probe is installed for every 10 hydraulic supports, for a total of 14 lidar probes are installed; the coal mining machine itself is 10 meters long, and before the system is started, the coal mining machine is located in the middle of the working face, cutting coal from left to right.

[0120] The specific method includes the following steps:

[0121] Each device in the system is numbered and defined. The probe at the head of the machine is numbered A, the probe at the tail of the machine is numbered C, and the 12 probes between A and C are numbered B1, B2, ..., B 11 B 12 The planar scan window is defined as including three consecutive radars with a spacing of 20 meters between adjacent radars. The area between probes A-B3 is defined as the front-end infeed area, and probe B... 10 The area between -C is the rear-end feed area, probe B3, probe B 10Separately, the front-end feed area sentry lidar probe and the rear-end feed area sentry lidar probe;

[0122] When the system starts up, it first performs a self-test on all probes, checking their integrity, communication status, interface condition, and visibility between adjacent probes. After the self-test is complete, adjacent probes perform mutual position calibration and establish a unified spatial coordinate system based on the position of probe A. Finally, all probes enter the working state, synchronously scan the working surface space, and asynchronously send the collected point cloud data to the central processing unit inside the equipment train. After the central processing unit completes the data reception and processing, it establishes an initial model for the working surface.

[0123] After the initial model of the working face is established, the probe scanning mode is entered. The coal mining machine is located near probe B8, therefore the area between probes B6, B7, and B8 is the scanning window area, maintaining continuous scanning, while the other probes enter sleep mode. When the coal mining machine moves out of the scanning range of probe B6, the scanning window slides to the right. Probe B6 enters sleep mode, and probe B9 is activated and added to the scanning window. At this time, probes B7, B8, and B9 in the scanning window maintain continuous scanning, while the other probes enter sleep mode. The scanning window always scans areas near the coal mining machine where the coal body morphology changes, reducing the data acquisition and transmission load and improving model update efficiency. The start-sleep working mode also effectively extends the average service life of the probes.

[0124] When the coal mining machine moves near the cutting area, the laser radar probe B at the rear cutting area sentry post... 10 When the coal mining machine is detected, it enters the coal mining machine feed mode. During feed, the coal mining machine will move back and forth at the end of the coal body and feed. This requires probe B near the tail of the machine. 10 B 11 B 12 C continues scanning until the coal mining machine finishes its feed. 10 After the probe detects that the coal mining machine has moved out of the cutting area and started normal coal cutting, it returns to the flat sweeping mode.

[0125] The real-time point cloud data collected by the lidar probes within the horizontal scanning window area or the cutting area is sent to the underground explosion-proof processing terminal via an underground explosion-proof switch for local point cloud stitching. Taking the current coal mining machine position B9 as an example, the sliding horizontal scanning window covers three lidars: B7, B8, and B9. The system controls the three lidars (B7, B8, and B9) to continuously collect point cloud data of the working face at the current position, and sends it to the underground processing terminal via the underground explosion-proof switch. After receiving the data, the system in the underground processing terminal stitches this data into a local update model M7-9.

[0126] The downhole processing terminal reads the positions of three lidar sensors (B7, B8, and B9) in the working face, obtains the matching position of the local update model M7-9 in the initial working face model M, and performs local updates on the initial model using point cloud stitching and fusion algorithms to generate a global working face model.

Claims

1. A method for acquiring data from a lidar monitoring array on a working face, characterized in that, A working face lidar monitoring array data acquisition system is adopted. The working face lidar monitoring array data acquisition system includes multiple lidar probes installed on one side of the coal mining face, an underground processing terminal, an underground explosion-proof switch, and an underground explosion-proof power supply. Each of the aforementioned lidar probes is connected to an underground explosion-proof switch, which is connected to an underground processing terminal. Both the lidar probes and the underground explosion-proof switch are connected to an underground explosion-proof power supply. The aforementioned lidar probe is used to synchronously scan the coal mining face and asynchronously send the collected point cloud data to the underground explosion-proof switch. The aforementioned explosion-proof switch is used to send the received point cloud data to the downhole processing terminal; The downhole processing terminal is used to receive and process point cloud data, control the scanning mode of the lidar probe, and control the wake-up and sleep modes of the lidar probe. The aforementioned explosion-proof power supply is used to power the lidar probe and the explosion-proof switch in the well. Includes the following steps: Step 1: Number the lidar probes; The lidar probe near the tail of the scraper conveyor is numbered A, the lidar probe near the head of the scraper conveyor is numbered C, and the lidar probe between lidar probe A and lidar probe C is numbered Bi, i=1~n, where n is a positive integer. The front feed area sentry lidar probe and the rear feed area sentry lidar probe are also determined. Step 2: Start the data acquisition system of the working face lidar monitoring array, and the lidar probe performs a self-test; The self-test includes checking the integrity of the lidar probe itself, communication status, interface condition, and visibility between adjacent lidar probes; Step 3: The adjacent lidar probes are mutually calibrated, and the reference coordinate system of the working surface laser scanning model is established with the built-in coordinate system of lidar probe A as the reference. Step 4: The underground processing terminal controls all lidar probes to enter full-scan mode, that is, all lidar probes scan the coal mining face synchronously and send the collected point cloud data asynchronously to the underground explosion-proof switch. The underground explosion-proof switch sends the data to the underground processing terminal, which receives and processes the point cloud data to obtain the initial model of the working face. Step 5: In real time, determine whether the laser radar probe of the front-end cutting area or the rear-end cutting area has detected the coal mining machine. If yes, enter the cutting mode; otherwise, enter the flat sweeping mode. The aforementioned cutting mode involves the coal mining machine reciprocating and cutting at the end of the coal body. The lidar probe at the end of the coal body continuously scans to obtain point cloud data. When the corresponding sentry probe detects that the coal mining machine has moved out of the cutting area, it exits the cutting mode and enters the flat sweeping mode. The described flat scanning mode specifically includes the following steps; Step 5.1: Establish a sliding sweeping window based on the width of the coal mining machine, wherein the sliding sweeping window covers the lidar probe near the coal mining machine; Step 5.2: The lidar probes inside the sliding scanning window scan the coal mining machine and the nearby coal mining face to obtain point cloud data, and all lidar probes outside the sliding scanning window enter a dormant state. Step 5.3: When the coal mining machine moves, the sliding scanning window moves accordingly, thereby waking up a lidar probe in front of the coal mining machine and adding it to the sliding scanning window. When the lidar probe that has moved out of the sliding scanning window detects that the hydraulic support it is located has finished moving and there are no moving objects within the scanning range, it enters a sleep state. Step 6: Asynchronously send the point cloud data obtained in Step 5 to the underground explosion-proof switch. The underground explosion-proof switch sends the data to the underground processing terminal. The underground processing terminal receives and processes the point cloud data to obtain the local update model. Step 7: Fuse the initial working face model obtained in Step 4 and the local update model obtained in Step 6 to obtain the global working face model.

2. The data acquisition method for the working face lidar monitoring array as described in claim 1, characterized in that, Step 3 specifically includes the following steps: Step 3.1: Use all lidar probes Bi of the coal mining face to scan the spatial morphology of the face, measure the distance from the lidar probe to the coal wall, and determine the relative positions of adjacent lidar probes; where: i=1~n; Step 3.2: Using the built-in coordinate system of lidar probe A as the reference, establish the reference coordinate system of the working face laser scanning model. The other lidar probes will use the spatial shape, coal wall distance and relative position with adjacent probes collected in step 3.1 and perform registration transformation calculations to integrate their own coordinate system into the reference coordinate system of the working face laser scanning model.

3. The data acquisition method for the working face lidar monitoring array as described in claim 1, characterized in that, In step 4, obtaining the initial model of the working surface includes the following: S1, define the lidar probe A at the tail of the scraper conveyor as the origin of the coordinate system; S2, determine the 6 degrees of freedom between adjacent lidar probes. The specific steps are as follows: The relationship between adjacent lidar probes is determined by mutual scanning and identification. By identifying feature points in the overlapping point cloud, the distance between adjacent lidar probes can be determined. ; in: Indicates the spatial positional deviation between lidar probes; Indicates rotational deviation; S3, point cloud data is stitched together using the point cloud registration method to obtain the initial model of the working surface; The point cloud registration method described above involves extracting feature points from two radar point clouds respectively, and then calculating the error between the two feature points. Find the registration matrix that minimizes the global error as the optimal registration matrix. ; in: , represents the rotational change vector of the point clouds of two adjacent excitation devices; , represents the translational change vector of the point clouds of two adjacent excitation devices; This represents the scaling factor of the point clouds of two adjacent excitation devices; express ; The error The calculation formula is as follows: in: The rotational change vector of the point clouds of two adjacent excitation devices; This represents the translational change vector of the point clouds of two adjacent excitation devices; Represent the source set A; Represents the target point set B1; and These represent the sizes of the two point sets, respectively.

4. The data acquisition method for the working face lidar monitoring array as described in claim 1, characterized in that, In step 7, the initial working face model obtained in step 4 and the local update model obtained in step 6 are fused to obtain the global working face model, which specifically includes the following steps: Step 7.1: Determine the position of the local update model in the working surface according to the LiDAR device number of the local update area, and superimpose the local update model on the corresponding position of the initial model on the working surface to obtain the overlapping area of ​​the two positions; Step 7.2: Perform point cloud registration operation on the local update model and the initial model of the working face. After registration, remove the expired point cloud data in the overlapping area between the initial model of the working face and the local update model, and retain the point cloud data of the registered local update model to finally obtain the global model of the working face.

Images