Aggregate inventory balance detection method for concrete mixing station based on point cloud

Abstract

The present invention discloses a method for detecting the inventory balance of aggregates in a concrete mixing station based on point cloud, which belongs to the field of concrete production. It solves the problem that the aggregate bin is not of conventional shape and is built with an inclination angle due to the situation by preprocessing the data, excludes the inclined ground plane factor from the calculation of the inventory balance, and further improves the accuracy of the inventory balance measured in this application scenario by polygon cutting. It mainly includes the following steps: installing a two-dimensional radar at the geometric center of two connected aggregate bins, the two-dimensional radar performs rotation scanning driven by a motor, and the radar controller receives and temporarily stores the data sent by the radar, including distance and angle information; the computer obtains the temporarily stored distance and angle information read by the radar controller, which is the point cloud data collected in the clearing state, and includes the following steps: calculating the plane model of the ground by random sampling, and iterating by self-learning. The present invention is mainly used for detecting the inventory balance of aggregates in a concrete mixing station.

Description

Technical Field

[0001] The present invention relates to the field of concrete production, and in particular to a method for detecting aggregate inventory surplus of a concrete mixing station based on point cloud. Background Art

[0002] In concrete production, it is often necessary to monitor aggregate inventory regularly. The aggregate inventory is calculated by subtracting the weight used in the aggregate production process from the weight of the incoming aggregate weighed in the weighing room. Generally, there is a cumulative error, and the error is large.

[0003] When using manual measurement, the surveyor needs to go to the aggregate silo, use a tape measure to measure the length, width, and height of the aggregate pile, and then manually estimate the remaining amount of aggregate. This measurement method has two disadvantages:

[0004] (1) Since personnel need to be present at the aggregate silo, measurements are usually taken once a month, for example, which does not allow real-time measurement.

[0005] (2) The measurement accuracy depends on the estimation results of the surveyors. Since the aggregate pile is not a strict rectangular block, the top of the aggregate pile is uneven, and sometimes the forklift fails to pile the materials in time, resulting in irregular empty storage space in the pile, it is difficult to estimate and the estimation accuracy is poor.

[0006] For example, the 3D point cloud segmentation method based on multi-line laser radar with announcement number CN106204705B extracts ground objects for autonomous driving through point cloud data segmentation, and does not involve the volume of specific detected objects. However, the calculation of aggregate inventory not only requires more specific volume detection, but also, since the on-site aggregate warehouse is an irregular polygon, the ground of the warehouse site is not absolutely horizontal. Sometimes it is built according to the terrain, and the warehouse ground plane will have a certain inclination angle with the absolute horizontal plane, which brings great difficulties to data detection. Summary of the invention

[0007] The purpose of the present invention is to provide a method for detecting the inventory balance of aggregates in a concrete mixing station based on point cloud. The method solves the problem that the aggregate silo is not of a conventional shape and has an inclination angle due to the terrain by preprocessing the data, excludes the inclined ground plane factor from the calculation of the inventory balance, and further improves the accuracy of the inventory balance measured in this application scenario by polygon cutting in a Cartesian coordinate system.

[0008] The present invention is achieved through the following technical solutions:

[0009] A method for detecting the aggregate inventory balance of a concrete mixing plant based on point cloud comprises the following steps:

[0010] S1. Install the two-dimensional radar at the geometric center of two connected aggregate bins. The two-dimensional radar is driven by a motor to perform rotation scanning. The radar controller receives and temporarily stores the data sent by the radar, including distance and angle information.

[0011] S2. The computer obtains the temporarily stored distance and angle information read by the radar controller. This is the point cloud data collected in the clearing state, including the following steps:

[0012] 1) Calculate the plane model of the ground by randomly sampling 3 points. The formula is as follows:

[0013] ax+by+cz+d=0

[0014] Calculate a set of a1, b1, c1, d1;

[0015] 2) Calculate n groups of a through self-learning iteration n times n , b n , c n , d n ;

[0016] 3) Calculate the distance from all points to the plane model under different a, b, c, d. When the distance is less than the threshold h, the point is inside the plane model. Select a group of a, b, c, d with the largest number of points entering the plane model from the point cloud data as the plane model coefficients; pre-process the data to exclude the inclined ground plane from the inventory balance, thereby improving the accuracy of the measured inventory balance;

[0017] 4) Calculate the angle α between the ground plane and the horizontal plane according to the plane model coefficient;

[0018] 5) In the Cartesian coordinate system, the collected point cloud data is rotated in the x-axis, y-axis and z-axis directions according to the calculated angle α, and the three-dimensional point cloud data is polygonally cut according to the shape of the aggregate warehouse to obtain the point cloud data of each warehouse;

[0019] S3. When the aggregate bin is in a stock state, the point cloud data of each aggregate bin is calculated to calculate the remaining weight of each aggregate bin.

[0020] Furthermore, step 4) also includes

[0021] According to the angle formula between two planes:

[0022]

[0023] In this application:

[0024]

[0025] And get the rotation matrix:

[0026]

[0027] All data are calculated with the rotation matrix to obtain new point cloud data, and at this time the point cloud data on the ground coincides with the x0y plane in the Cartesian coordinate system.

[0028] Furthermore, step S3 also includes the following steps:

[0029] S31, finding the boundary of the point cloud data, and subdividing the point cloud data into multiple cuboids according to a set threshold h;

[0030] S32. Calculate the volume of each small cuboid using the following formula:

[0031] V i =S i *H i

[0032] H i is the height of the cuboid;

[0033] S33, accumulating the volume of each small cuboid in each aggregate bin to obtain the volume of the remaining material in each aggregate bin,

[0034] The formula is as follows:

[0035]

[0036] S34. The volume of the remaining material in each aggregate bin is multiplied by the density of the remaining material in the aggregate bin to obtain the weight of the remaining material in each aggregate bin. The formula is as follows:

[0037] m=V*ρ

[0038] ρ is the density of the material.

[0039] Furthermore, starting from the coordinate origin, taking the square area with the set threshold as the side length as the unit, it is divided into small rectangles along the positive direction of the x-axis and the positive direction of the y-axis. The z coordinate of the points in each small rectangle is the average height of the rectangle, and the volume of each small rectangle is calculated. In this way, the points within the warehouse are divided and calculated.

[0040] Furthermore, it also includes a two-dimensional radar, a motor and a radar controller. The radar controller controls the rotation of the motor, and the rotation of the motor drives the two-dimensional radar to rotate. The controller is connected to the two-dimensional radar through Ethernet for Ethernet communication, and the host computer is connected to the radar controller through Ethernet for Ethernet communication.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. Through data preprocessing, the problem of the aggregate silo's irregular shape and inclination angle built according to the terrain was solved. The inclined ground plane factor was excluded from the calculation of inventory balance, and then polygon cutting under the Cartesian coordinate system was used to further improve the accuracy of the inventory balance measured in this application scenario.

[0043] 2. When calculating the angle between the ground plane and the absolute horizontal plane, a self-learning calibration method is used to conveniently and concisely calculate the angle between the ground plane and the absolute horizontal plane.

[0044] 3. Compared with the previous scheme of subtracting the weight of aggregate used from the weight of aggregate incoming material and the scheme of manual measurement, the measurement result will not have cumulative errors, will not be affected by human factors, and has high measurement accuracy.

[0045] 4. Real-time measurement of aggregate pile volume. By measuring the pile shape of aggregate residue, the volume of aggregate residue can be directly output with high real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of the aggregate warehouse inventory surplus calculation of the present invention;

[0047] Figure 2 It is a sub-flow chart of the self-learning iteration of the present invention;

[0048] Figure 3 It is the preprocessing and segmentation sub-flow chart of the present invention;

[0049] Figure 4 is a volume calculation sub-flow chart of the present invention;

[0050] Figure 5 It is a topological structure diagram of the system of the present invention. DETAILED DESCRIPTION

[0051] The present invention will be further described below in conjunction with the accompanying drawings.

[0052] like Figure 1 – Figure 4 As shown, in Example 1, a method for detecting the aggregate inventory balance of a concrete mixing station based on point cloud comprises the following steps:

[0053] S1. Install the two-dimensional radar at the geometric center of two connected aggregate bins. The two-dimensional radar is driven by a motor to perform rotation scanning. The radar controller receives and temporarily stores the data sent by the radar, including distance and angle information.

[0054] S2. The computer obtains the temporarily stored distance and angle information read by the radar controller. This is the point cloud data collected in the clearing state, including the following steps:

[0055] 1) Calculate the plane model of the ground by randomly sampling 3 points. The formula is as follows:

[0056] ax+by+cz+d=0

[0057] Calculate a set of a1, b1, c1, d1;

[0058] 2) Calculate n groups of a through self-learning iteration n times n , b n , c n , d n ;

[0059] 3) Calculate the distance from all points to the plane model under different a, b, c, d. When the distance is less than the threshold h, the point is inside the plane model. Select a group of a, b, c, d with the largest number of points entering the plane model from the point cloud data as the plane model coefficients; pre-process the data to exclude the inclined ground plane from the inventory balance, thereby improving the accuracy of the measured inventory balance;

[0060] 4) Calculate the angle α between the ground plane and the horizontal plane according to the plane model coefficient;

[0061] 5) In the Cartesian coordinate system, the collected point cloud data is rotated in the x-axis, y-axis and z-axis directions according to the calculated angle α, and the three-dimensional point cloud data is polygonally cut according to the shape of the aggregate warehouse to obtain the point cloud data of each warehouse;

[0062] S3. When the aggregate bin is in a stock state, the point cloud data of each aggregate bin is calculated to calculate the remaining weight of each aggregate bin.

[0063] Embodiment 2: A method for detecting the aggregate inventory balance of a concrete mixing station based on point cloud, wherein step 4) further comprises:

[0064] According to the angle formula between two planes:

[0065]

[0066] In this application:

[0067]

[0068] And get the rotation matrix:

[0069]

[0070] All data are calculated with the rotation matrix to obtain new point cloud data, and the point cloud data on the ground now coincides with the x0y plane in the Cartesian coordinate system;

[0071] Step S3 also includes the following steps:

[0072] S31, finding the boundary of the point cloud data, and subdividing the point cloud data into a plurality of cuboids according to a set threshold h;

[0073] S32. Calculate the volume of each small cuboid using the following formula:

[0074] V i =S i *H i

[0075] H i is the height of the cuboid;

[0076] S33, accumulating the volume of each small cuboid in each aggregate bin to obtain the volume of the remaining material in each aggregate bin,

[0077] The formula is as follows:

[0078]

[0079] S34. The volume of the remaining material in each aggregate bin is multiplied by the density of the remaining material in the aggregate bin to obtain the weight of the remaining material in each aggregate bin. The formula is as follows:

[0080] m=V*ρ

[0081] ρ is the density of the material;

[0082] Starting from the origin of the coordinate system, the area of ​​a square with the set threshold value h as the side length is taken as the unit, and it is divided into small rectangles in the positive direction of the x-axis and the positive direction of the y-axis in sequence. The z coordinates of the points in each small rectangle are averaged as the height of the rectangle, and the volume of each small rectangle is calculated. In this way, the points within the range of this warehouse are divided and calculated. For example, the threshold value h is a parameter set according to the site. Starting from the origin of the coordinate system, the threshold value is set to 0.5m, and the square surrounded by the coordinates (0, 0), (0, 0.5), (0.5, 0.5) and (0.5, 0) is taken as the bottom area, and the average z coordinate of all points in this range is taken as the height to calculate the volume of this rectangle. Similarly, the square surrounded by (0.5, 0), (0.5, 0.5), (1, 0.5) and (1, 0) is taken as the bottom area, and the average z coordinate of all points in this range is taken as the height to calculate the volume of this rectangle. In this way, the points within the range of this warehouse are divided and calculated. It can also be set to 0.03m, 0.01m, etc. The principle is the same as above and will not be repeated.

[0083] like Figure 5 As shown, it also includes a two-dimensional radar, a motor and a radar controller. The radar controller controls the rotation of the motor, and the rotation of the motor drives the two-dimensional radar to rotate. The controller is connected to the two-dimensional radar through Ethernet for Ethernet communication. The host computer is connected to the radar controller through Ethernet for Ethernet communication. The rest is the same as Example 1.

Claims

1. A method for detecting aggregate inventory surplus in a concrete mixing plant based on point cloud, characterized in that: The steps include: S1. Install the two-dimensional radar at the geometric center of two connected aggregate bins. The two-dimensional radar is driven by a motor to perform rotation scanning. The radar controller receives and temporarily stores the data sent by the radar, including distance and angle information. S2. The computer obtains the temporarily stored distance and angle information read by the radar controller. This is the point cloud data collected in the clearing state, including the following steps: 1) Calculate the plane model of the ground by randomly sampling 3 points. The formula is as follows: ax+by+cz+d=0 Calculate a set of a1, b1, c1, d1; 2) Calculate n groups of a through self-learning iteration n times n , b n , c n , d n ; 3) Calculate the distance from all points to the plane model under different a, b, c, d. When the distance is less than the threshold h, the point is inside the plane model. Select the set of a, b, c, d with the largest number of points entering the plane model from the point cloud data as the plane model coefficients; 4) Calculate the angle α between the ground plane and the horizontal plane based on the plane model coefficient: According to the angle formula between two planes: In this application: And get the rotation matrix: All data are calculated with the rotation matrix to obtain new point cloud data, and the point cloud data on the ground now coincides with the x0y plane in the Cartesian coordinate system; 5) In the Cartesian coordinate system, the collected point cloud data is rotated in the x-axis, y-axis and z-axis directions according to the calculated angle α, and the three-dimensional point cloud data is polygonally cut according to the shape of the aggregate warehouse to obtain the point cloud data of each warehouse; S3. When the aggregate bin is in a stock state, the point cloud data of each aggregate bin is calculated to calculate the remaining weight of each aggregate bin.

2. The method for detecting aggregate inventory surplus of a concrete mixing plant based on point cloud according to claim 1 is characterized in that: Step S3 also includes the following steps: S31, finding the boundary of the point cloud data, and subdividing the point cloud data into a plurality of cuboids according to a set threshold h; S32. Calculate the volume of each small cuboid using the following formula: V i =S i *H i H i is the height of the cuboid; S33, accumulating the volume of each small cuboid in each aggregate bin to obtain the volume of the remaining material in each aggregate bin, The formula is as follows: S34. The volume of the remaining material in each aggregate bin is multiplied by the density of the remaining material in the aggregate bin to obtain the weight of the remaining material in each aggregate bin. The formula is as follows: m=V*ρ ρ is the density of the material.

3. The method for detecting aggregate inventory surplus of a concrete mixing plant based on point cloud according to claim 1, characterized in that: Starting from the coordinate origin, the square area with the set threshold h as the side length is divided into small rectangles along the positive direction of the x-axis and the positive direction of the y-axis. The z coordinate of the points in each small rectangle is the average height of the rectangle. The volume of each small rectangle is calculated. In this way, the points within the bin are divided and calculated.

4. The method for detecting aggregate inventory surplus of a concrete mixing plant based on point cloud according to claim 1, characterized in that: It also includes a two-dimensional radar, a motor and a radar controller. The radar controller controls the rotation of the motor, and the rotation of the motor drives the two-dimensional radar to rotate. The controller is connected to the two-dimensional radar through Ethernet for Ethernet communication, and the host computer is connected to the radar controller through Ethernet for Ethernet communication.

Images