Material identification and automatic loading and unloading system and method suitable for steel coil storage yard

Abstract

The invention discloses a material identification and automatic loading and unloading system and method suitable for a steel coil storage yard. The system comprises a first one-dimensional holder, a second one-dimensional holder, a first laser scanner, a second laser scanner, a first industrial personal computer and a second industrial personal computer. The industrial personal computer comprises an OPEN3D point cloud processing algorithm module, a holder and a laser scanning control module; the first one-dimensional holder and the first laser scanner are responsible for collecting point cloud information of a steel coil warehouse of a storage yard, and the first industrial personal computer processes the point cloud information to obtain the diameter and center coordinates of a steel coil. The second one-dimensional holder and the second scanner are responsible for collecting point cloud information of a storage yard truck parking area, and the second industrial personal computer processes the point cloud information to obtain the vehicle type, the diameter, the center coordinate and the placement position of the steel coil on the vehicle, so that automatic warehousing, loading and unloading of the steel coil are completed. The method can be applied to grabbing and positioning calculation of the steel coils on the vehicle during automatic warehousing of the steel coil storage yard and placement position calculation during delivery, and automatic warehousing, loading and unloading of the steel coils are achieved.

Description

Technical Field

[0001] The invention relates to a storage yard management system, and in particular to a material identification and automatic loading and unloading system and method suitable for a steel coil storage yard. Background Art

[0002] At present, most steel companies need to put the produced steel coils into the yard instead of using them directly. Due to the large number of steel coil specifications and the complex yard environment, it is easy to cause confusion in the management of steel coils and make them difficult to identify.

[0003] The existing methods of grabbing steel coils in the yard mainly rely on the following methods: relying on manual measurement by workers, telling the driving driver the specifications, grabbing information and placement information of the steel coils to be grabbed, and requiring the truck driver to maintain real-time communication with the driving driver. This grabbing and placing method has high labor costs, high risk factors, and too long working hours; relying on image processing recognition methods to obtain the two-dimensional information of the steel coils through image recognition, and telling the two-dimensional information to the workers to realize grabbing and placement. This method relies on the professionalism of the workers, and the image is easily affected by light, causing large errors in recognition. Summary of the invention

[0004] In view of this, the present invention provides a system and method for material identification and automatic loading and unloading in a steel coil yard, which can be used to identify and locate steel coils, saddles, and loading and unloading trucks in the steel coil yard. It uses point cloud to perform identification and positioning, and realizes the automatic entry and exit of steel coil grabbing and placement.

[0005] To this end, the present invention provides the following technical solutions, including:

[0006] A material identification and automatic loading and unloading system suitable for a steel coil yard, comprising:

[0007] Install the first one-dimensional pan / tilt head and laser scanner at the appropriate position of the vehicle to collect point cloud information of the yard;

[0008] Install a second 1D pan / tilt and laser scanner at an appropriate location on the vehicle to load and unload truck point cloud information;

[0009] The task instructions sent by the on-site WMS system are received through the industrial computer, the tasks are classified, the one-dimensional pan-tilt head and laser scanner are controlled to collect 3D point cloud data of the yard and loading and unloading trucks, and the POIINTNET++ model is used to segment the vehicles and saddles in the yard, and the regional growing algorithm is used to segment the steel coils.

[0010] A material identification and automatic loading and unloading method applicable to a steel coil yard comprises:

[0011] Sending instructions via the industrial computer to control the first and second pan / tilts and the first and second laser scanners to collect yard data;

[0012] Identification module based on z=γcosθ converts the two-dimensional polar coordinate point cloud data into three-dimensional point cloud data, where x, y, z are the three-dimensional coordinate values, r is the radial distance from the point in polar coordinates to the origin, provided by the laser radar message; θ is the polar angle, provided by the laser radar message; is the azimuth, the angular resolution of the gimbal;

[0013] For the stockyard point cloud information, the recognition module uses a straight-through filtering algorithm on the 3D point cloud data to extract the region of interest A, which contains all the steel coils in the stockyard;

[0014] Use the radius filtering algorithm to filter area A to remove noise points in the point cloud data, thereby improving processing speed and point cloud accuracy;

[0015] Use the K-nearest neighbor algorithm to traverse all points, and cooperate with the region growing algorithm to calculate the curvature change between each point. Points with a change less than the threshold are considered to be the same steel coil. After all points are traversed, the point cloud data of each steel coil can be obtained.

[0016] The cylindrical equation and RANSAC algorithm are used to perform cylindrical fitting on each steel coil point cloud to obtain the center coordinates, radius and orientation of each steel coil, thus completing the steel coil recognition task.

[0017] For the point cloud information of loading and unloading trucks, the recognition module uses a straight-through filtering algorithm on the three-dimensional point cloud data to extract the region of interest B, which includes the vehicle, saddle, and steel coils on the vehicle;

[0018] Use the radius filtering algorithm to filter area B to remove noise points in the point cloud data and improve the processing speed and point cloud accuracy;

[0019] Use the K-nearest neighbor algorithm to traverse all points, and cooperate with RANSAC coarse registration and ICP fine registration to align the vehicles in area B with the vehicles in the dataset to obtain the type and information of the current vehicle;

[0020] The point cloud segmentation model in POINTNET++ is used to segment the vehicle and obtain the loading and unloading area C, which contains the point cloud of each saddle and each steel coil.

[0021] For the steel coil point cloud data in area C, the cylinder equation and RANSAC algorithm are used to perform cylindrical fitting on each steel coil point cloud to obtain the center coordinates, radius and orientation of each steel coil, and complete the steel coil identification.

[0022] For the saddle point cloud data in area C, the plane equation Ax+By+Cz+D=0 and the RANSAC algorithm are used for plane fitting to obtain the two plane equations of the saddle V-groove and the normal vectors L1 and L2 of the plane;

[0023] Find the intersection point A of L1 and L2, which is the estimated placement position of the steel coil, and complete the task of loading and unloading the truck;

[0024] In addition, the yard material identification and automatic loading and unloading system needs to be calibrated before use;

[0025] Place three reflectors under the PTZ and laser scanner, place plumb bobs on the vehicle and use the Graeme line system to measure the X1Y1Z1, X2Y2Z2, and X3Y3Z3 coordinates of the reflectors, use the laser scanner to collect the point cloud data of the reflectors, process the point cloud data to find the three point sets with the highest reflectivity, find the centroids x1y1z1, x2y2z2, and x3y3z3 of the three point sets, transform the three XYZ coordinates and the three xyz coordinates, and get the calibrated transformation matrix to complete the calibration task.

[0026] Combined with the above, material identification and automatic loading and unloading of the steel coil yard are completed.

[0027] Advantages and positive effects of the present invention: The present invention provides a material identification and automatic loading and unloading system and method for a steel coil yard. Without changing the original site and equipment, a laser scanner, a pan / tilt head and an industrial computer are added to reduce labor costs, improve work efficiency, shorten work cycles, increase safety factors, and achieve fast and precise steel coil identification, positioning, and grasping position placement, thereby realizing intelligent management of the steel coil yard. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the real-time examples of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some real-time examples of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 This is a schematic diagram of the 3D vision installation for warehousing and grabbing in an example of the present invention;

[0030] Figure 2 This is a schematic diagram of the 3D vision installation for outbound grabbing in an example of the present invention;

[0031] Figure 3 The present invention is a flow chart of scanning a steel coil yard in an example;

[0032] In the figure, 1 is the first one-dimensional pan-tilt head; 2 is the first laser scanner; 3 is the on-site driving; 4 is the second one-dimensional pan-tilt head; 5 is the second laser scanner; and 6 is the on-site driving. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] The present invention is a scanning recognition system and a scanning method designed for the material handling equipment of the steel coil yard, the "intelligent electromagnetic overhead crane", in order to realize unmanned and fully automatic operation and autonomously perceive the actual on-site status during production operations.

[0036] like Figure 1 and Figure 2 As shown, the present invention proposes a material identification and automatic loading and unloading system suitable for steel coil yard, including a first one-dimensional pan-tilt, a second one-dimensional pan-tilt, a first laser scanner, a second laser scanner, a first and a second industrial computer; the industrial computer includes an OPEN3D point cloud processing algorithm module, a pan-tilt and laser scanning control module; the first one-dimensional pan-tilt and the first laser scanner are responsible for collecting the point cloud information of the steel coil warehouse in the yard, and the first industrial computer processes the point cloud data to obtain the diameter and center coordinates of the steel coil; the second one-dimensional pan-tilt and the second scanner are responsible for collecting the point cloud information of the truck parking area in the yard, and the second industrial computer processes the point cloud data to obtain the vehicle type, the diameter and center coordinates of the steel coil on the vehicle, and the placement position, to complete the automatic storage and loading and unloading of the steel coil. The scanner needs to be installed in a place with a wide driving field of view to ensure that the vehicles in the yard and parking area can be completely scanned during scanning.

[0037] The one-dimensional gimbal is a stabilizer that turns the laser scanner into three-dimensional data. It stabilizes the laser scanner and generates three-axis data. The horizontal position of the laser scanner can be adjusted after the laser scanner is installed.

[0038] The industrial computer includes an identification module, which can communicate with the on-site PLC. The OPEN3D point cloud processing algorithm module of the first industrial computer processes the 3D point cloud data to give the steel coil diameter and center coordinates; the OPEN3D point cloud processing algorithm module of the second industrial computer processes the 3D point cloud data to give the vehicle type, steel coil diameter, center coordinates and placement coordinates.

[0039] The above embodiment provides a material identification and automatic loading and unloading system suitable for steel coil yards. Without changing the original site and equipment, by adding equipment such as laser scanners, one-dimensional pan-tilt heads and industrial computers, labor costs can be reduced, work efficiency can be improved, work cycles can be shortened, and safety factors can be increased. Rapid and accurate steel coil identification, positioning, grasping and placement can be achieved, thereby realizing intelligent management of steel plate yards.

[0040] like Figure 3 As shown, a method for material identification and automatic loading and unloading in a steel coil yard is provided, and the specific steps are as follows:

[0041] Step S1, accepting the task sent by the PLC through the industrial computer, controlling the pan / tilt and laser scanner to collect on-site 3D point cloud data;

[0042] Step S2, the recognition module of the industrial computer processes the 3D point cloud data collected in the yard, gives the direct size and center coordinate information of the target steel coil, identifies the vehicle type and saddle position information, and gives the steel coil placement information;

[0043] Step S3: The industrial computer sends the above information to the on-site PLC, controls the crane to move to the specified position to start grabbing and placing.

[0044] The industrial computer recognition module aims to identify the diameter, center coordinates, vehicle type and placement of the steel coil. It includes: using direct point cloud filtering to obtain the area of ​​interest, using radius filtering to remove noise in the data, using RANSAC to segment and obtain the steel coil warehouse point cloud, vehicle point cloud and saddle point cloud, using regional growing algorithm to segment the steel coil warehouse point cloud to obtain the point cloud of each steel coil, using cylinder equation to fit the point cloud of each steel coil to obtain the diameter and center coordinates of the steel coil, using RANSAC plane fitting to fit the saddle point cloud to obtain the plane equation and normal vector on both sides of the saddle, and calculating the intersection of the normal vectors of the two planes, which is the center coordinate of the steel coil.

[0045] In the embodiment of the present invention, the RANSAC method is used to perform straight line fitting on the contour. RANSAC is the abbreviation of Random Sample Consensus. It is an algorithm that calculates the mathematical model parameters of the data based on a set of sample data sets containing abnormal data to obtain valid sample data. Compared with other methods for plane fitting, the RANSAC method can well eliminate cluttered points, select the global optimal point for fitting, and reduce the error of the fitted straight line.

[0046] 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 material identification and automatic loading and unloading system suitable for steel coil yard, characterized in that: include: The first one-dimensional pan / tilt and the first laser scanner installed on the crane above the steel coil warehouse in the yard are used to collect point cloud data of the steel coil warehouse; Install a second one-dimensional pan-tilt head and a second laser scanner for vehicles traveling above the parking area of ​​the yard to collect point cloud data of the vehicles; An industrial computer installed next to the first one-dimensional pan-tilt platform and the first laser scanner, used to obtain the diameter and center position of the steel coil and control the first one-dimensional pan-tilt platform and the first laser scanner; The industrial computer installed next to the second one-dimensional pan-tilt platform and the second laser scanner is used to obtain the vehicle type, the diameter and center coordinates of the steel coil on the vehicle, the placement position and control the second one-dimensional pan-tilt platform and the second laser scanner.

2. The material identification and automatic loading and unloading system for steel coil yard according to claim 1 is characterized in that: A dustproof and waterproof protective cover is installed above the first laser scanner and the second laser scanner.

3. A method for identifying and automatically loading and unloading materials in a steel coil yard, applied to the system for identifying and automatically loading and unloading materials in a steel coil yard as described in any one of claims 1-2, characterized in that: The method comprises: The industrial computer receives the on-site PLC task instructions, and the pan / tilt and laser scanning control module of the industrial computer controls the pan / tilt and laser scanner to collect data according to the task; The first industrial computer OPEN3D point cloud processing algorithm module processes the 3D point cloud data to give the diameter and center coordinates of the steel coil; The OPEN3D point cloud processing algorithm module of the second industrial computer processes the 3D point cloud data to give the vehicle type, steel coil diameter, center coordinates and placement coordinates.

4. The material identification and automatic loading and unloading method applicable to the steel coil yard according to claim 3 is characterized in that: Based on the point cloud data, the steel coil information and location information are identified, including: The first pan / tilt and the first laser scanner for obtaining the steel coil warehouse in the yard need to be moved to the preset parking area to calibrate the coordinates of the target area; Place the reflector under the pan / tilt and laser scanner, place a plumb bob on the crane to measure the XYZ coordinates of the reflector, use the laser scanner to collect the point cloud data of the reflector, and use the industrial computer to process the point cloud data to find the xyz of the point with the highest reflectivity, transform the XYZ coordinates and the xyz coordinates to obtain the calibrated transformation matrix.

5. The material identification and automatic loading and unloading method applicable to the steel coil yard according to claim 4 is characterized in that: The RANSA segmentation method is used to extract objects of interest in the steel coil warehouse and parking area.

6. The material identification and automatic loading and unloading method applicable to a steel coil yard according to claim 4 is characterized in that: Based on the scanned 3D point cloud information, combined with the point cloud extracted from the target of interest, the coil diameter, center coordinates, vehicle type and placement are determined, including: The 3D point cloud data of the steel coil warehouse and parking area is downsampled and pre-processed for denoising; According to the extracted region of interest, RANSAC and cylindrical equation fitting are performed to obtain the diameter and center coordinates of the steel coil. The saddle information in the vehicle is obtained through straight-through filtering and segmentation algorithm, the two plane equations of the saddle are fitted, and the intersection of the normal vectors of the two plane equations is calculated to obtain the placement position; Convert the grab center coordinates and placement position into driving Graham line coordinates through the transformation matrix; Combined with the above, material identification and automatic loading and unloading of the steel coil yard are completed.

Images