Laser scanner-based positioning method and system for drilling holes in steel strips on roof of mine tunnels
By using laser scanner sampling and algorithm processing, high-precision automated positioning of steel strip anchor holes in the roof of mine roadways was achieved, solving the problems of light influence and sample library establishment, and improving the operating efficiency of anchor protection robots.
Patent Information
- Application Number
- CN202210465629.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-29
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-04-29
AI Technical Summary
Existing technologies for locating steel strip anchor holes in mine roadway roofs suffer from several drawbacks. Poor lighting conditions affect imaging results, leading to low positioning accuracy. Furthermore, the need to establish a sample library increases the workflow, and special anchor holes cannot be identified.
A laser scanner-based method for locating boreholes in the roof steel strips of mine roadways is adopted. The method involves sampling cloud point information using a laser scanner, extracting boundary points using the kd-tree method, classifying them using Euclidean distance clustering and segmentation algorithms, extracting anchor hole boundaries using ellipse fitting, calculating the center and axis of the fitted ellipse, and converting the borehole plane coordinates to achieve automated positioning.
It improves anchor hole positioning accuracy under harsh lighting conditions, simplifies the workflow, can identify special anchor holes, realizes automated operation of anchor protection robots, reduces manual support work, and improves work efficiency.
Smart Images

Figure CN114998212B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of anchor robot drilling positioning, and in particular relates to a mine tunnel roof steel belt drilling positioning system and method based on a laser scanner. Background Art
[0002] With the improvement of industrialization and technology, the trend of reducing the number of people in underground mines and unmanned intelligent mining has become a trend. However, as an important guarantee for the stability and reliability of tunnels during their service life, the operation mode of anchor support still requires manual operation, especially in the operation of anchor holes in the steel belts on the roof of the tunnel. Due to the poor working environment underground, limited light, and small diameter of the anchor holes on the roof, workers cannot quickly and accurately locate the anchor holes during operation, which prolongs the working time of anchor support and even affects the stability of the tunnel. In order to realize the automation of anchor support operations, machine vision is also used in this regard. Patent No. CN202110084689.0 provides a system and method, which has the following shortcomings: (1) This method requires the establishment of a sample library in the early stage of the anchor support machine, and the imaging results are compared with the sample library. This will increase the task process on the one hand, and on the other hand, it will not be able to identify the special anchor holes that appear, resulting in empty anchors. (2) This system relies on cameras, and the lighting conditions in the mine seriously affect the imaging effect and weaken the positioning accuracy of the anchor holes.
[0003] With the development of laser scanning technology, laser scanning instruments are more and more widely used. Due to its adaptability and high precision in mining environments, how to use laser scanners to identify drill holes has become an urgent problem to be solved in this field. Summary of the invention
[0004] In order to help realize unmanned and reduced-manpower underground operations and to solve the problem that machine vision technology in anchor protection robots is greatly affected by brightness and is not suitable for mine environments, the present invention provides a mine tunnel roof steel belt drilling positioning system and method based on a laser scanner.
[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for drilling and locating a steel belt on a roof of a mine tunnel based on a laser scanner, comprising the following steps:
[0006] S1. Sampling the steel strip on the top of the roadway by a laser scanner to obtain sampling cloud point information;
[0007] S2, using the kd tree method to extract the boundary of the sampled cloud point information to obtain the boundary cloud point;
[0008] S3, after classifying the boundary cloud points according to the Euclidean distance clustering segmentation algorithm, extracting the anchor hole boundary cloud points by ellipse fitting, removing the steel belt boundary cloud points, and calculating the center coordinates, major axis and minor axis of the fitting ellipse corresponding to the anchor hole boundary cloud points;
[0009] S4, determining the drilling plane by fitting the cloud points near the center of the ellipse, and converting the center of the ellipse to the drilling plane to obtain the three-dimensional coordinates of the drilling center;
[0010] S5. According to the three-dimensional coordinates of the drilling center, control the drill arm to perform anchor support operations.
[0011] In step S2, the specific method for performing boundary extraction is:
[0012] S201, select a cloud point a from the sampled cloud points, set a k-neighborhood, and find the set of k points closest to the cloud point a as the first neighborhood set, denoted by N j (j=0,1,2,…,k-1), for point a and the first neighborhood set N j The local tangent plane of cloud point a is obtained by fitting all point clouds within the cloud using the least square method;
[0013] S202, establish a plane coordinate system on the local tangent plane of cloud point a, with the projection a1 of cloud point a to the tangent plane as the origin of the coordinate system, and the projection N of a1 and N0 on the plane 0,1 The vector composed of is the x-axis, the cross product of the normal vector of the local tangent plane and the x-axis is the y-axis to establish the first plane coordinate system, and the cloud point a and the first neighborhood set N j All points inside are projected into the first plane coordinate system;
[0014] S203, in the first plane coordinate system, take the projection a1 of the cloud point a as the vertex, and move to the first neighborhood set N j The projections of each point in the first plane coordinate system are vectors, and the angle θ between any two adjacent vectors is calculated. j , if maxθ j >θ, it means that cloud point a is a boundary point. If maxθ j <θ, it means that cloud point a is not a boundary, and θ represents the maximum angle threshold;
[0015] S204, select the next point from the sampled cloud points, repeat steps S201 to S203, judge all cloud points, and obtain all boundary cloud point sets a i .
[0016] In step S203, the value range of the maximum angle threshold is:
[0017] In step S3, the specific method for classifying the boundary cloud points is:
[0018] S301, select any point p in the boundary cloud point set and put it into the cluster set Q1, and use the kd tree to find the boundary cloud point set a i The k points closest to the cloud point p are the second k-neighborhood set, denoted as p i (i=0,…,k-1); Calculate the cloud point p to the second k neighborhood set p i The Euclidean distance r of each point in the cluster is set to the point whose Euclidean distance is less than the cluster segmentation threshold r. s The points are put into cluster set Q1;
[0019] S302, select cluster set Q t A cloud point q in , using the kd tree to find the boundary cloud point set a i Among the remaining cloud points, the k points closest to cloud point q are taken as the third k-neighborhood set, and the Euclidean distance from cloud point q to each point in the third k-neighborhood set is calculated. The points whose Euclidean distance is less than the cluster segmentation threshold are put into cluster set Q1;
[0020] S303, select the next cloud point in cluster set Q1, and repeat step S303 until the number of points in cluster set Q1 stops increasing;
[0021] S304, in the boundary cloud point set a i Continue to select any point from the remaining cloud points and put it into cluster set Q2, and repeat steps S301 to S304 until the number of points in cluster set Q2 stops increasing;
[0022] S305, repeat step S304 until all boundary cloud points are put into cluster sets, and finally obtain cluster sets Q1, Q2, ... Q T , T represents the number of cluster sets.
[0023] In step S3, the specific method of extracting the anchor hole boundary by ellipse fitting and removing the steel strip boundary is as follows:
[0024] S306, a first fitting plane is obtained by fitting all cloud points in one cluster set Q1 by the least square method, and a second plane coordinate system is established with any point on the first fitting plane as the origin, an arbitrary vector at the origin as the x-axis of the coordinate system, and the cross product of the plane normal vector and the x-axis as the y-axis. t All cloud points in are projected onto the fitting plane to obtain the coordinate set in the second plane coordinate system
[0025] S307, coordinate set Perform plane ellipse fitting and calculate the fitting error G;
[0026] S308, repeat steps S306 to S307 for other cluster sets;
[0027] S309, determine each cluster set Q1, Q2, ... Q T Whether it is a borehole boundary cloud point set is determined by: Then the cluster set is not a borehole boundary cloud point set and is removed. Then the cluster set is the borehole boundary cloud point set and is retained, where G0 represents the fitting error threshold and N represents the number of point clouds in the cluster set;
[0028] S310 . For the retained cluster set, the ellipse center, the major axis and the minor axis of the fitted ellipse are calculated by using a fitted ellipse expression.
[0029] In step S307, the fitting ellipse expression is:
[0030] A0x 2 +B0xy+C0y 2 +D0x+E0y+F0=0;
[0031] Where A0, B0, C0, D0, F0, F0 are the ellipse parameters, 4A0C0-B0 2 >0, x, y represent the horizontal and vertical coordinates respectively;
[0032] The fitting optimization function expression is:
[0033]
[0034] The calculation formula of the fitting error G is:
[0035]
[0036] in, Represents a coordinate set The coordinates of the nth cloud point in , where N represents the number of cloud points in cluster set Q1.
[0037] The step S4 specifically comprises the following steps:
[0038] S401, taking the center of the fitted ellipse as the search center and m times the mean of the major axis and minor axis of the fitted ellipse as the search radius, searching for a cloud point set a in this area using a kd tree structure h (h=1,2,…,H), H represents a h The number of interior points;
[0039] S402, from cloud point set a h (h=1,2,…,H) to obtain the drilling plane. The fitting expression is:
[0040]
[0041] Among them, A1, B1, C1, and D1 represent the fitting plane parameters, and the drilling plane equation obtained by fitting is A1x h +B1y h +C1Z h +D1=0;
[0042] S403, transform the ellipse center coordinates calculated in step 3 from the ellipse fitting plane coordinate system to the drilling plane coordinate system to obtain the drilling center coordinates (x0, y0). The drilling plane coordinate system is based on the cloud point set a h Any point in (h=1,2,…,H) is the origin, any vector is the x-axis, and the cross product of the x-axis and the plane normal vector is the y-axis;
[0043] S404. Calculate the three-dimensional coordinates (x0, y0, z0) of the drilling center using the drilling plane equation.
[0044] In step S403, the calculation formula for coordinate transformation is:
[0045]
[0046] Among them, α is the rotation angle between the two coordinate systems, (u, v) represents the coordinates of the origin of the drilling plane coordinate system on the ellipse fitting plane, and (x0, y0) represents the coordinates of the drilling center in the drilling plane coordinate system after coordinate transformation.
[0047] The step S5 also includes the step of converting the three-dimensional coordinates of the drilling center into the anchor robot coordinate system.
[0048] In addition, the present invention also provides a mine tunnel roof steel belt drilling positioning system based on a laser scanner, comprising: a laser scanner and an airborne data processing terminal; the airborne data processing terminal is provided with:
[0049] Boundary extraction unit: used to extract boundary cloud point sets from the scattered environment cloud points collected by the laser scanner using the kd tree method.
[0050] Cluster segmentation unit: used to select the drilling boundary cloud point set from the boundary cloud point set using the Euclidean distance cluster segmentation algorithm.
[0051] Coordinate calculation unit: used to calculate the three-dimensional coordinates of the center position of the drilling hole through the drilling boundary cloud point set to achieve drilling positioning.
[0052] Compared with the prior art, the present invention has the following beneficial effects:
[0053] (1) The present invention is used for positioning anchor holes on the steel belt of the tunnel roof during the anchor robot support operation. Compared with the existing machine vision using industrial camera technology, the present invention is more suitable for mine tunnel environments with poor lighting conditions.
[0054] (2) The present invention obtains cloud point information of the environment on the steel belt of the tunnel roof through a laser scanner, and does not need to establish a sample library of anchor holes in advance, which not only simplifies the work process, but also can locate special anchor holes on the steel plate.
[0055] (3) The present invention has high positioning accuracy for anchor holes, further realizing the automation of anchor robot operations, reducing the number of support operators and improving work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 A schematic diagram of a process for locating a mine tunnel roof steel strip drilling hole based on a laser scanner provided by an embodiment of the present invention;
[0057] Figure 2 It is a structural schematic diagram of a drilling and positioning device on a steel belt of a tunnel roof in an embodiment of the present invention;
[0058] Figure 3 This is a schematic diagram of cloud points of the roadway roof steel strip obtained by laser scanner;
[0059] Figure 4 It is a schematic diagram of the vector angle in the local tangent plane coordinate system;
[0060] Figure 5 This is a schematic diagram of the boundary point cloud extracted using the kd tree method;
[0061] Figure 6 Schematic diagram of ellipse fitting for borehole boundary cloud points;
[0062] Figure 7 This is a schematic diagram of the structure of the anchor protection robot's mechanical arm;
[0063] Figure 8 for Figure 7 Schematic diagram of the coordinate system of the anchor protection robot's mechanical arm.
[0064] In the figure: 1-tunnel, 2-roof steel belt, 3-anchor hole, 4-laser scanner, 5-airborne data processing terminal, 6-anchor protection robot, 7-drilling arm, 8-mechanical arm, 9-drilling boundary point cloud, 12-steel belt boundary point cloud; 13-drilling plane, 14-first fitting plane, 15-fitting ellipse; DETAILED DESCRIPTION
[0065] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0066] Embodiment 1
[0067] like Figure 1 As shown, an embodiment of the present invention provides a method for drilling and locating a mine tunnel roof steel strip based on a laser scanner, comprising the following steps:
[0068] S1. The steel strip on the top plate of the tunnel is sampled by a laser scanner to obtain sampling cloud point information.
[0069] like Figure 2 As shown, it is a schematic diagram of the structure of the positioning method of the embodiment of the present invention when the positioning system is used. A supporting steel belt 2 is provided on the top of the tunnel 1, and a row of circular drill holes 3 are provided on the supporting steel belt 2. The positioning system includes a laser scanner 4 and an airborne data processing terminal 5. The laser scanner 4 is fixed on the mechanical arm 8 of the anchor robot to collect environmental cloud point information on the top of the tunnel. The cloud point data collected by the laser scanner 4 is calculated and processed by the airborne data processing terminal to obtain the three-dimensional coordinates of the drilling holes, and then sent to the anchor robot 7. After the anchor robot 7 performs coordinate conversion on the obtained coordinates, it drives the mechanical arm 8 and the drilling arm 7 to perform anchoring.
[0070] When positioning, first turn on the laser scanner power, start the anchor robot, and control the mechanical arm to the position of the supporting steel belt; the laser scanner works to fully sample the steel belt on the top plate of the tunnel, and transmit the sampling cloud point information to the airborne data processing terminal. The schematic diagram of the cloud point obtained by sampling is shown in Figure 3 shown.
[0071] S2. Use the kd tree method to extract the boundaries of the sampled cloud point information to obtain boundary cloud points.
[0072] In step S2, the specific method for performing boundary extraction is:
[0073] S201, select a cloud point a from the sampled cloud points, set a k-neighborhood, and find the set of k points closest to the cloud point a as the first neighborhood set, denoted by N j (j=0,1,2,…,k-1), for point a and the first neighborhood set N j The local tangent plane of cloud point a is obtained by fitting all point clouds within the cloud using the least square method;
[0074] S202, establish a plane coordinate system on the local tangent plane of cloud point a, with the projection a1 of cloud point a to the tangent plane as the origin of the coordinate system, and the projection N of a1 and N0 on the plane 0,1 The vector composed of is the x-axis, the cross product of the normal vector of the local tangent plane and the x-axis is the y-axis to establish the first plane coordinate system, and the cloud point a and the first neighborhood set N j All points inside are projected into the first plane coordinate system;
[0075] S203, in the first plane coordinate system, take the projection a1 of the cloud point a as the vertex, and move to the first neighborhood set N j The projections of each point in the first plane coordinate system are vectors, and the angle θj between any two adjacent vectors is calculated. If maxθ j >θ, it means that cloud point a is a boundary point. If maxθ j <θ, it means that cloud point a is not a boundary, and θ represents the maximum angle threshold; Figure 4 As shown, it is a schematic diagram of each vector, where N 0,1 、N 0,2 , ... N 0,k-1 Represents the first neighborhood set N j The projection of each point in the first plane coordinate system, θ0~θ k-1 Represents the angle between each two adjacent vectors.
[0076] In step S203, the value range of the maximum angle threshold is:
[0077] S204, select the next point from the sampled cloud points, repeat steps S201 to S203, judge all cloud points, and obtain all boundary cloud point sets a i Finally, the schematic diagram of the boundary cloud point set is obtained as follows Figure 5 shown.
[0078] S3. After classifying the boundary cloud points according to the Euclidean distance clustering segmentation algorithm, the anchor hole boundary cloud points are extracted by ellipse fitting, the steel belt boundary cloud points are eliminated, and the center coordinates, major axis and minor axis of the fitting ellipse corresponding to the anchor hole boundary cloud points are calculated.
[0079] In step S3, the specific method for classifying the boundary cloud points is:
[0080] S301, select any point p in the boundary cloud point set and put it into the cluster set Q1, and use the kd tree to find the boundary cloud point set a i The k points closest to the cloud point p are the second k-neighborhood set, denoted as p i (i=0,…,k-1); Calculate the cloud point p to the second k neighborhood set p i The Euclidean distance r of each point in the cluster is set to the point whose Euclidean distance is less than the cluster segmentation threshold r. s The points are put into cluster set Q1;
[0081] S302, select cluster set Q t A cloud point q in , using the kd tree to find the boundary cloud point set a iAmong the remaining cloud points, the k1 points closest to cloud point q are taken as the third k-neighborhood set, the Euclidean distance from cloud point q to each point in the third k-neighborhood set is calculated, and the points whose Euclidean distance is less than the cluster segmentation threshold are put into cluster set Q1; wherein k1 can be less than k.
[0082] S303, select the next cloud point in cluster set Q1, and repeat step S303 until the number of points in cluster set Q1 stops increasing;
[0083] S304, in the boundary cloud point set a i Continue to select any point from the remaining cloud points and put it into cluster set Q2, and repeat steps S301 to S304 until the number of points in cluster set Q2 stops increasing;
[0084] S305, in the boundary cloud point set a i Continue to select any point from the remaining cloud points and put it into cluster set Q3, repeat steps S301 to S304 until the number of points in cluster set Q3 stops increasing, and repeat the operation until all boundary cloud points are put into cluster sets, and finally get cluster sets Q1, Q2, ... Q T , T represents the number of cluster sets.
[0085] In step S3, the specific method of extracting the anchor hole boundary by ellipse fitting and removing the steel strip boundary is as follows:
[0086] S306, a first fitting plane is obtained by fitting all cloud points in one cluster set Q1 by the least square method, and a second plane coordinate system is established with any point on the first fitting plane as the origin, an arbitrary vector at the origin as the x-axis of the coordinate system, and the cross product of the plane normal vector and the x-axis as the y-axis. t All cloud points in are projected onto the fitting plane to obtain the coordinate set in the second plane coordinate system
[0087] S307, coordinate set Perform plane ellipse fitting and calculate the fitting error G;
[0088] In step S307, the fitting ellipse expression is:
[0089] A0x 2 +B0xy+C0y 2 +D0x+E0y+F0=0; (1)
[0090] In the formula, A0, B0, C0, D0, E0, F0 are the ellipse parameters, x, y represent the horizontal and vertical coordinates respectively;
[0091] The fitting optimization function expression is:
[0092]
[0093] The calculation formula of the fitting error G is:
[0094]
[0095] in, Represents a coordinate set The coordinates of the nth cloud point in , where N represents the number of cloud points in cluster set Q1.
[0096] S308, repeat steps S306 to S307 for other cluster sets;
[0097] S309, determine each cluster set Q1, Q2, ... Q T Whether it is a borehole boundary cloud point set is determined by: Then the cluster set is not a borehole boundary cloud point set and is removed. Then the cluster set is the borehole boundary cloud point set and is retained, where G0 represents the fitting error threshold and N represents the number of point clouds in the cluster set;
[0098] S310, for the retained cluster set, calculate the ellipse center (x, y), major axis and minor axis of the fitted ellipse by using the fitted ellipse expression. Figure 6 As shown, the fitting ellipse is 15 and the first fitting plane is 14.
[0099] S4. Obtain the drilling plane 13 by fitting the ellipse, and convert the center of the ellipse to the drilling plane to obtain the three-dimensional coordinates of the drilling center.
[0100] The step S4 specifically comprises the following steps:
[0101] S401, using the center (x, y) of the fitted ellipse as the search center and m times the mean of the major axis and minor axis of the fitted ellipse as the search radius, using the kd tree structure to search for the cloud point set a in this area h (h=1,2,…,H), H represents a h The number of interior points.
[0102] S402, from cloud point set a h (h=1,2,…,H) plane fitting is performed to obtain the drilling plane (i.e. the plane where the steel strip is located), and the fitting expression is:
[0103]
[0104] Among them, A1, B1, C1, D1 represent the fitting plane parameters, and the fitting drilling plane equation is A1x h +B1y h +C1Z h +D1=0. Figure 6As shown, the plane where the fitting ellipse 15 obtained by fitting the borehole boundary cloud point set is located is the first fitting plane 14. By finding the kd number to search the cloud point set a h , the drilling plane 13 can be determined, and the drilling plane 13 and the first fitting plane 14 obtained in step S306 may be on different planes.
[0105] S403, convert the ellipse center coordinates (x, y) calculated in step 3 from the ellipse fitting plane coordinate system to the drilling plane coordinate system to obtain the drilling center coordinates (x0, y0). The drilling plane coordinate system is based on the cloud point set a h Any point in (h=1,2,…,H) is the origin, any vector is the x-axis, and the cross product of the x-axis and the plane normal vector is the y-axis.
[0106] In step S403, the calculation formula for coordinate transformation is:
[0107]
[0108] Among them, α is the rotation angle between the two coordinate systems, (u, v) represents the coordinates of the origin of the drilling plane coordinate system on the ellipse fitting plane, and (x0, y0) represents the coordinates of the drilling center in the drilling plane coordinate system after coordinate transformation.
[0109] S404, substituting the coordinates of the drilling center (x0, y0) into the drilling plane equation, and calculating the three-dimensional coordinates of the drilling center (x0, y0, z0).
[0110] S5. According to the three-dimensional coordinates of the drilling center, control the drill arm to perform anchor support operations.
[0111] The step S5 also includes the step of converting the three-dimensional coordinates of the drilling center into the anchor robot coordinate system.
[0112] The anchor robot control system transforms the three-dimensional coordinates of the drilling center to obtain the coordinates of the drilling center relative to the anchor robot. The three-dimensional coordinates of the drilling center obtained in step S4 are the coordinates with the anchor robot as the reference object, recorded as (x r ,y r ,z r ),use Figure 7 and Figure 8 To illustrate the coordinate transformation process.
[0113] The robot arm has 6 degrees of freedom, including rotation θ1 around the z1 axis, rotation θ2 around the z2 axis, extension d3 along the z3 axis, rotation θ4 around the z4 axis, rotation θ5 around the z5 axis, and rotation θ6 around the z6 axis. O0 and O1 are the upper surfaces of the base, corresponding to coordinate systems S0 and S1 respectively; O2 is the center of the outer sleeve swivel pin, corresponding to coordinate system S2; O3 and O4 are the center of the leveling seat swivel pin, corresponding to coordinate systems S3 and S4; O5 is the upper surface of the leveling seat, corresponding to coordinate system S5; O6 is the rotary cylinder, corresponding to coordinate system S6; O7 is the jaws of the drill rig, corresponding to coordinate system S7.
[0114] The laser scanner is placed in the S2 coordinate system, the coordinates of O2 in S2 are (x2, y2, 0), and the coordinates of O2 in S1 are (x1, 0, z1). The above parameters x1, z1, x2, y2 can be obtained by measurement.
[0115] From this, we can get the homogeneous transformation matrix between the coordinate systems of the robot:
[0116]
[0117]
[0118]
[0119]
[0120] in, represents the transformation matrix from coordinate system S1 to coordinate system S0, represents the transformation matrix from coordinate system S2 to coordinate system S1, represents the transformation matrix from coordinate system S3 to coordinate system S2, Represents the transformation matrix from coordinate system S3 to coordinate system S0.
[0121] The coordinates of the drilling center relative to the anchor robot are:
[0122]
[0123] In the formula, (x0 y0 z0 1) is obtained by making (x0, y0, z0) homogeneous. r ,y r ,z r )=(d 14 , d 24 , d 34 ). The hole center position coordinates (x r ,y r ,z r ) is transmitted to the anchor robot control system; then the anchor robot control system calculates the coordinates of the drilling center position (xr ,y r ,z r ) to adjust the position of the drill arm so that it is directly below the borehole. The position of the drill arm in the anchor robot coordinate system is determined by the design and is denoted by (x w ,y w ,z w ), when the drill arm is directly below the borehole,
[0124]
[0125] In the formula, z a It is a fixed value, which is determined by the working condition and design of the drill arm. It can be any value as long as the working conditions are met.
[0126] Embodiment 2
[0127] Embodiment 2 of the present invention provides a mine tunnel roof steel strip drilling positioning system based on a laser scanner, including a laser scanner and an airborne data processing terminal, wherein the airborne data processing terminal is provided with a boundary extraction unit, a cluster segmentation unit and a coordinate calculation unit.
[0128] The boundary extraction unit is used to extract the boundary cloud point set from the scattered environment cloud points collected by the laser scanner using the kd tree method.
[0129] The cluster segmentation unit is used to select a borehole boundary cloud point set from a boundary cloud point set by using a Euclidean distance cluster segmentation algorithm.
[0130] The coordinate calculation unit is used to calculate the three-dimensional coordinates of the center position of the drilling hole through the drilling boundary cloud point set to achieve drilling positioning.
[0131] The airborne data processing terminal transmits the coordinates of the drilling center to the anchor protection robot control system, and the anchor protection robot control system converts the obtained coordinates to complete the subsequent anchor protection operations.
[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for drilling and locating holes in a steel belt on a mine tunnel roof based on a laser scanner, characterized in that: The following steps are involved: S1. Sampling the steel strip on the top of the roadway by laser scanner to obtain sampling point cloud information; S2, using the kd tree method to extract the boundary of the sampling point cloud information to obtain the boundary point cloud; S3, after classifying the boundary point cloud according to the Euclidean distance clustering segmentation algorithm, extracting the anchor hole boundary point cloud by ellipse fitting, removing the steel belt boundary point cloud, and calculating the center coordinates, major axis and minor axis of the fitting ellipse corresponding to the anchor hole boundary point cloud; S4, determining the drilling plane by fitting the point cloud near the center of the ellipse, and converting the center of the ellipse to the drilling plane to obtain the three-dimensional coordinates of the drilling center; S5. Control the drill arm to perform bolt support operations according to the three-dimensional coordinates of the drilling center; In step S2, the specific method for performing boundary extraction is: S201, select a point cloud a from the sampled point cloud, set a k-neighborhood, and find the set of k points closest to the point cloud a as the first neighborhood set, denoted as , for point a and the first neighbor set The local tangent plane of point cloud a is obtained by fitting all point clouds within the least square method; S202, establish a plane coordinate system on the local tangent plane of point cloud a, project point cloud a onto the tangent plane is the coordinate origin, and Projection on a plane The vector composed of is the x-axis, the cross product of the normal vector of the local tangent plane and the x-axis is the y-axis to establish the first plane coordinate system, and the point cloud a and the first neighborhood set All points inside are projected into the first plane coordinate system; S203, in the first plane coordinate system, take the projection a1 of the point cloud a as the vertex, and move to the first neighborhood set The projections of each point in the first plane coordinate system are vectors, and the angle between any two adjacent vectors is calculated. ,like , then it means that point cloud a is a boundary point. If , then it means that point cloud a is not a boundary, Indicates the maximum angle threshold; S204: Select the next point from the sampled point cloud, repeat steps S201 to S203, and make judgments on all point clouds to obtain all boundary point cloud sets. ; In step S3, the specific method for classifying the boundary point cloud is: S301, select any point p in the boundary point cloud set and put it into the cluster set , use kd tree to find the boundary point cloud set The k points closest to the point cloud p are taken as the second k-neighborhood set, denoted as ; Calculate the point cloud p to the second k neighborhood set The Euclidean distance r of each point in the cluster is less than the cluster segmentation threshold. The points are put into the cluster set ; S302, select cluster set A point cloud q in , using kd tree to find the boundary point cloud set In the remaining point cloud, the k points closest to point cloud q are used as the third k-neighborhood set, and the Euclidean distance from point cloud q to each point in the third k-neighborhood set is calculated. Points whose Euclidean distance is less than the cluster segmentation threshold are put into the cluster set. ; S303, select cluster set Repeat step S303 for the next point cloud in the cluster set The number of points in the no longer increases; S304, at the boundary point cloud collection Continue to select any point from the remaining point cloud and put it into cluster set Q2, repeat steps S301 to S304 until cluster set Q2 is reached. The number of points in the no longer increases; S305, repeat step S304 until all boundary point clouds are put into the cluster set, and finally obtain the cluster set , T represents the number of cluster sets; Step S4 specifically includes the following steps: S401, taking the center of the fitted ellipse as the search center and m times the mean of the major axis and minor axis of the fitted ellipse as the search radius, using the kd tree structure to search for the point cloud set in this area (h=1,2,…,H), H represents The number of interior points; S402, Point Cloud Collection (h=1,2,…,H) to obtain the drilling plane. The fitting expression is: ; in, represents the fitting plane parameters, and the drilling plane equation obtained by fitting is ; S403, transforming the ellipse center coordinates calculated in step S3 from the ellipse fitting plane coordinate system to the drilling plane coordinate system to obtain the drilling center coordinates (x0, y0). The drilling plane coordinate system is a point cloud collection. Any point in (h=1,2,…,H) is the origin, any vector is the x-axis, and the cross product of the x-axis and the plane normal vector is the y-axis; S404, calculate the three-dimensional coordinates of the drilling center through the drilling plane equation ( ).
2. The method for drilling and locating holes in the roof steel strip of a mine tunnel based on a laser scanner according to claim 1, characterized in that: In step S203, the value range of the maximum angle threshold is: [ ].
3. The method for drilling and locating holes in a mine tunnel roof steel strip based on a laser scanner according to claim 1, characterized in that: In step S3, the specific method of extracting the anchor hole boundary by ellipse fitting and removing the steel strip boundary is as follows: S306: For one of the cluster sets All point clouds in are fitted by the least squares method to get the first fitting plane. Any point on the first fitting plane is taken as the origin, and any vector at the origin is taken as the x-axis of the coordinate system. The plane normal vector and the x-axis are cross-producted as the y-axis to establish the second plane coordinate system. All point clouds in are projected onto the fitting plane to obtain the coordinate set in the second plane coordinate system ; S307, coordinate set Perform plane ellipse fitting and calculate the fitting error G; S308, repeat steps S306-S307 for other cluster sets; S309, judging each cluster set Whether it is a drilling boundary point cloud is determined by: , then the cluster set is not a drilling boundary point cloud set and is removed. , then the cluster set is the drilling boundary point cloud set, which is retained, among which, represents the fitting error threshold, and N represents the number of point clouds in the cluster set; S310 . For the retained cluster set, the ellipse center, the major axis and the minor axis of the fitted ellipse are calculated by using a fitted ellipse expression.
4. The method for drilling and locating holes in a mine tunnel roof steel strip based on a laser scanner according to claim 1, characterized in that: In step S307, the fitting ellipse expression is: ; In the formula, , , , , , are the ellipse parameters, , x , y Respectively represent the horizontal and vertical coordinates; The fitting optimization function expression is: ; The calculation formula of the fitting error G is: ; in,( ) represents the coordinate set The coordinates of the nth point cloud in , N Represents cluster set The number of point clouds in .
5. The method for drilling and locating holes in a mine tunnel roof steel strip based on a laser scanner according to claim 1, characterized in that: In step S403, the calculation formula for coordinate transformation is: ; in, is the rotation angle between the two coordinate systems, ( ) represents the coordinates of the origin of the drilling plane coordinate system on the ellipse fitting plane, and (x0, y0) represents the coordinates of the drilling center in the drilling plane coordinate system after the coordinate transformation.
6. The method for drilling and locating holes in a mine tunnel roof steel strip based on a laser scanner according to claim 1, characterized in that: Step S5 also includes the step of converting the three-dimensional coordinates of the drilling center into the coordinate system of the anchor robot.
7. A mine tunnel roof steel belt drilling positioning system based on laser scanner, characterized in that: The method for locating a hole in a mine tunnel roof steel strip based on a laser scanner according to claim 1 comprises: a laser scanner and an airborne data processing terminal; the airborne data processing terminal is provided with: Boundary extraction unit: used to extract boundary point cloud sets from the scattered environment point clouds collected by the laser scanner using the kd tree method; Cluster segmentation unit: used to select the drilling boundary point cloud set from the boundary point cloud set by using the Euclidean distance cluster segmentation algorithm; Coordinate calculation unit: used to calculate the three-dimensional coordinates of the center position of the drilling hole through the cloud of drilling boundary points to achieve drilling positioning.
