Laser cutting scrap identification method and device, laser cutting machine and storage medium

Abstract

The application discloses a laser cutting excess material recognition method and device, a laser cutting machine and a storage medium. The laser cutting excess material recognition method comprises the following steps: using a calibration element to perform visual calibration on a line laser vision device, obtaining a conversion relationship δ between a coordinate system of the line laser vision device and a machine tool coordinate system, and determining a variable speed distance D; setting a scanning line segment length L0 of the line laser vision device, and determining a maximum scanning area according to the conversion relationship δ, the variable speed distance D and the scanning line segment length L0; controlling the line laser vision device to scan excess material in the maximum scanning area, obtaining point cloud data of the excess material, and processing the point cloud data to generate a contour image file of the excess material in the machine tool coordinate system. The scheme breaks through the technical prejudice that the existing laser cutting excess material recognition only needs to collect two-dimensional plane contour information, thereby effectively solving the technical problems of contour misjudgment and omission that are prone to occur in common working conditions such as plate warping, local deformation or multi-plate stacking.

Description

Technical Field

[0001] This application relates to the field of laser cutting technology, and in particular to a method, apparatus, laser cutting machine and storage medium for identifying laser cutting waste. Background Technology

[0002] In the field of laser cutting, the recycling and reuse of scrap materials is one way to improve material utilization and reduce production costs. However, it has long been generally believed in the field that the core objective of scrap material identification is only to obtain the outline shape within the plane of the sheet material for subsequent nesting and layout. Therefore, this two-dimensional planar outline information has fully met production needs.

[0003] Constrained by the aforementioned technological biases, the industry's scrap material identification technology has remained stagnant at the level of two-dimensional planar contour information acquisition. It typically employs two methods: manual point marking and area scan camera imaging. Manual point marking relies on operators recording planar coordinates point by point along the edge of the scrap material and fitting the contour, which has inherent defects such as low efficiency and varying accuracy depending on the individual. Area scan camera imaging, on the other hand, extracts edges by capturing planar images, achieving automated acquisition that replaces manual point marking. However, due to the inherent properties of two-dimensional imaging, it is prone to misjudgment and omission of contours in actual production conditions that are extremely common, such as board warping, local deformation, or multiple boards stacked. Summary of the Invention

[0004] This application provides a laser cutting waste material identification method, device, laser cutting machine, and storage medium, which breaks through the technical bias of existing laser cutting waste material identification methods that only require the collection of two-dimensional planar contour information. This effectively solves the technical problem that contour misjudgment and omission are prone to occur in common working conditions such as board warping, local deformation, or multiple boards stacked.

[0005] Firstly, the laser cutting waste identification method proposed in this application includes the following steps:

[0006] The line laser vision equipment is visually calibrated using calibration components to obtain the transformation relationship δ between the coordinate system of the line laser vision equipment and the machine tool coordinate system, and to determine the speed change distance D. Set the scanning line segment length L0 of the line laser vision device, and determine the maximum scanning area according to the conversion relationship δ, the speed change distance D and the scanning line segment length L0; The line laser vision device is controlled to scan the remaining material in the maximum scanning area, acquire the point cloud data of the remaining material, and process the point cloud data to generate a contour image file of the remaining material in the machine tool coordinate system.

[0007] The line laser vision device in this application is a three-dimensional information acquisition device. Visual calibration of the line laser vision device is performed using calibration components to obtain the transformation relationship δ between the coordinate system of the line laser vision device and the machine tool coordinate system. This ensures that the acquired data from the line laser vision device can accurately match the machine tool coordinate system used in subsequent nesting processes. Simultaneously, the start-stop acceleration characteristics of the scanning motion are considered, and the variable speed distance D is determined to ensure the stability of data acquisition during the uniform speed scanning segment. The scanning line segment length L0 of the line laser vision device is set, and the coordinate dimensions are unified and the motion reserved area of ​​the variable speed distance D is superimposed based on the transformation relationship δ. The effective acquisition range of the scanning line segment length L0 is determined to identify the maximum scanning area and ensure that the remaining material to be identified is within the effective scanning range. Based on this, the control line laser vision device scans the remaining material in the maximum scanning area to obtain point cloud data of the remaining material containing height dimension information. This distinguishes non-planar working conditions such as board warping, local protrusions, and multiple board stacking, thus solving the misjudgment defects of two-dimensional imaging. The point cloud data is then processed to convert the three-dimensional point cloud information into a contour image file of the remaining material in the machine tool coordinate system, so as to achieve direct adaptation with the coordinate system of the subsequent nesting process.

[0008] In a preferred embodiment, the specific steps for determining the maximum scanning area based on the conversion relationship δ, the speed change distance D, and the scanning line segment length L0 include: Using the positive limit of the X-axis of the machine tool coordinate system as a reference, the speed change distance D is offset in the negative X-axis direction, and using the negative limit of the X-axis of the machine tool coordinate system as a reference, the speed change distance D is offset in the positive X-axis direction to obtain the initial positive side boundary and the initial negative side boundary in the X-axis direction. Based on the orientation of the line laser vision device along the X-axis of the machine tool coordinate system, the X-axis offset obtained by the transformation relationship δ is superimposed onto the offset of the initial positive side boundary or the initial negative side boundary to determine the X-axis boundary of the maximum scanning area.

[0009] In a preferred embodiment, the specific steps for determining the maximum scanning area based on the conversion relationship δ, the speed change distance D, and the scanning line segment length L0 further include: Using the positive and negative Y-axis limits of the machine tool coordinate system as a reference, and based on the orientation of the line laser vision device along the Y-axis of the machine tool coordinate system, the Y-axis offset obtained by the transformation relationship δ and the offset of half the length L0 of the scanning line segment are superimposed on the offsets of the positive and negative Y-axis limits to determine the Y-axis boundary of the maximum scanning area.

[0010] In a preferred embodiment, the step of determining the speed change distance D includes: The coordinates of the start and end positions of the line laser vision device in the machine tool coordinate system are obtained when it accelerates or decelerates between a stationary state and a uniform scanning speed. The speed change distance D is calculated based on the coordinate difference between the start and end positions.

[0011] In a preferred embodiment, before controlling the line laser vision device to scan the remaining material, a scanning start and end position correction step is further included: Before controlling the line laser vision device to scan the remaining material, a scanning start and end position correction step is also included: Using the negative X-axis boundary of the maximum scanning area as a reference, and combining the speed change distance D, perform boundary limiting processing and offset calculation on the X-axis coordinate of the preset scanning starting point. Using the negative Y-axis boundary of the maximum scanning area as a reference, perform boundary limiting processing on the Y-axis coordinate of the preset scanning starting point to obtain the corrected scanning starting point coordinates. Based on the corrected scanning start coordinates, the scanning size parameters corresponding to the axial direction of the preset scanning area are superimposed. Then, boundary limiting processing is performed with the positive side boundary of the X-axis and the positive side boundary of the Y-axis corresponding to the maximum scanning area as the reference to obtain the corrected scanning end coordinates.

[0012] In a preferred embodiment, before controlling the line laser vision device to scan the remaining material, a scanning width correction step is further included: The total scan width W is calculated based on the corrected scan start coordinates and scan end coordinates. 扫 The number of scans n is calculated by combining the length L0 of the scan line segment and rounded up. Then, based on the number of scans n and the total scan width W, the result is calculated. 扫 Determine the width of a single scan.

[0013] In a preferred embodiment, the step of visually calibrating the line laser vision device using the calibration element further includes: Using the upper surface of the calibration component as the calibration reference surface, and combining its thickness N, the Z-axis zero-position height reference calibration is completed. The vertical distance parameter H from the line laser vision device to the calibration reference surface is determined, and the Z-axis coordinate value of the worktable surface in the machine tool coordinate system is derived to be -N.

[0014] In a preferred embodiment, after controlling the line laser vision device to scan and acquire point cloud data of the waste material, the step of processing the point cloud data includes: Based on the Z-axis zero-position height reference and the thickness parameter of the remaining material, the effective point cloud height threshold range is determined; The point cloud data is filtered in the Z direction according to the height threshold range to remove interfering point clouds outside the height threshold range, thereby obtaining the valid point cloud data corresponding to the remaining material.

[0015] In a preferred embodiment, the step of controlling the line laser vision device to scan the scrap material includes: Plan a line-by-line scanning path based on the corrected scan start coordinates, scan end coordinates, scan count n, and single scan width; The line laser vision device is driven to move along the scanning path of each row, and the point cloud data of the remaining material is collected within the uniform speed movement range.

[0016] Secondly, the laser cutting apparatus proposed in this application includes a laser cutting head, a line laser vision device disposed on one side of the laser cutting head, and a control processing unit capable of performing the above-mentioned laser cutting waste identification method. The control processing unit is electrically connected to the laser cutting head and the line laser vision device to control the laser cutting head to drive the line laser vision device to move along the machine tool coordinate system.

[0017] Thirdly, the laser cutting machine proposed in this application includes the aforementioned laser cutting device.

[0018] Fourthly, the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the laser cutting waste identification method described above. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the first process of the laser cutting waste identification method according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the generation of the maximum scanning area by the laser cutting head and line laser vision device along the N-shaped scanning path in an embodiment of the present invention. Figure 3 This is a schematic diagram of the second process of the laser cutting waste identification method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the third process of the laser cutting waste identification method according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the Z-axis zero-position height reference calibration using a calibration component in an embodiment of the present invention; Figure 6This is a schematic diagram of the fourth process of the laser cutting waste identification method according to an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the effect of Z-axis filtering of point cloud data according to a height threshold range; Figure 8 This is a schematic diagram illustrating the height relationship between the remaining material and the calibration part in an embodiment of the present invention.

[0021] Explanation of icon numbers: 100 - Line laser vision equipment, 200 - Calibration parts, 300 - Excess material, 400 - Laser cutting head, 500 - Worktable, 600a - Maximum scanning area.

[0022] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0024] It should be noted that all directional indicators (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicator will also change accordingly.

[0025] Furthermore, the use of terms such as "first" and "second" in this application is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed in this application.

[0026] In the laser cutting scrap identification scheme using area array camera imaging, the area array camera is mounted directly above the laser cutting machine tool's worktable. Based on the planar pixel array photosensitive imaging principle of the area array camera, a single orthophoto is taken to obtain a 2D projection image of the scrap within the worktable area. Specifically, the area array camera first captures a planar grayscale image of the scrap. After preprocessing operations such as filtering, denoising, and thresholding to eliminate image noise and background interference, the outline pixel points of the scrap are extracted using an edge detection operator. Finally, the pixel coordinates are converted into 2D outline data in the machine tool worktable's planar coordinate system according to the pre-calibrated object-image ratio.

[0027] Therefore, due to the inherent characteristics of two-dimensional projection imaging by area scan cameras, this technical solution can only collect the projection contour information of the scrap material on the imaging plane, completely lacking the spatial dimension data of the board height direction. When the scrap material is warped, the actual edge of the warped area will cause the projection contour to shift and geometrically distort due to the tilt of the spatial position, resulting in a significant deviation from the true planar contour of the scrap material, which in turn leads to misjudgment of the contour size and shape. When the board has local convex or concave deformation, the light and dark boundary line formed by the deformed area will be misidentified by the edge detection algorithm as the outer contour boundary of the scrap material, causing misjudgment of the contour edge. In the stacking situation that is common in production, the edges of multiple scrap materials at different height layers overlap and intertwine in the two-dimensional projection image. The area scan camera cannot distinguish the true boundaries of the upper and lower layers of scrap materials through height information, which easily leads to the problem of the upper layer of scrap material obscuring the lower layer of scrap material, or the misjudgment of the multiple layers of stacked edges as a single scrap material contour.

[0028] To address the limitations of 2D vision solutions using area array cameras in acquiring residual material height information and their lack of perception of material warping, stacking, or localized deformation, line laser 3D vision, capable of acquiring height dimension information, can be employed as a data acquisition device to overcome the technical bottleneck of 2D imaging. By mounting the line laser vision device on the crossbeam of a laser cutting machine, a full-area scan of the machine tool's working area can be completed. To ensure the matching accuracy between point cloud data and the machine tool coordinate system, existing technologies typically use at least two calibration plates positioned at different locations with height differences for visual calibration. This calibration is achieved by collecting the machine tool coordinate system coordinates of the corner points of the calibration plates. Based on the corresponding visual point cloud coordinates, the transformation relationship δ between the line laser vision device coordinate system and the machine tool coordinate system is obtained. Then, based on the statistical mean and standard deviation of the point cloud Z-axis coordinates, an adaptive height threshold is set for point cloud segmentation. This can automatically remove point clouds of structures such as machine tool blades and support teeth. Even under complex working conditions such as board warping, sinking, and multi-layer stacking, the effective point cloud dataset of residual material can still be stably extracted. Finally, the effective point cloud is projected onto the XY plane of the machine tool after coordinate transformation. After edge point extraction and segmented fitting, a closed two-dimensional contour smoothly connected by straight line segments and arc segments is generated and packaged into a standard DXF vector graphic file, which can be directly imported into the nesting and layout system.

[0029] In the aforementioned existing technologies, line laser vision equipment is typically mounted directly on the crossbeam of a laser cutting machine and uses the entire working area of ​​the machine tool as the scanning range for data acquisition. This approach addresses the shortcomings of area scan cameras in 2D vision, with research and development generally focusing on core technical aspects such as the calibration accuracy of point cloud data, segmentation algorithm optimization, and improvement of contour fitting effects. However, the planning of the scanning path and the setting of the scanning range, which superficially only involve the engineering arrangement of the equipment's movement stroke, have long been neglected as technical issues affecting the feasibility of the solution and have generally not received significant attention.

[0030] In the conventional approach of this field, since line laser vision equipment needs to acquire complete worktable data, it is considered a natural and inevitable choice to scan the entire working area with the crossbeam. Under this understanding, those skilled in the art often directly use the geometric boundaries of the worktable as the start and end points of the scanning stroke, without fully considering the acceleration and deceleration process and position control deviation of the line laser vision equipment during the start and end phases of the scanning motion, nor designing the stroke margin in conjunction with the equipment's own installation dimensions, moment of inertia, and the spatial avoidance relationship between the machine tool's edge structure and supporting components. As a result, during actual scanning, the line laser vision equipment is prone to collisions or scratches with the machine tool's edge structure and supporting components due to the scanning stroke exceeding the safe movement range and position control deviations during the start and end phases of the motion, causing wear, optical path deviation, and even hardware damage to the line laser vision equipment.

[0031] To effectively solve the problems existing in the aforementioned prior art, please refer to Figure 1 and Figure 2 The laser cutting waste 300 identification method provided by the present invention includes the following steps: Step 1: Use calibration component 200 to perform visual calibration on line laser vision device 100. In the process of solving and obtaining the transformation relationship δ between the coordinate system of line laser vision device 100 and machine tool coordinate system, the speed change distance D is determined through machine tool motion test. The variable speed distance D is determined by combining the installation dimensions and motion inertia of the line laser vision device 100 itself, as well as the difference between the measured motion coordinates of the line laser vision device 100 and the position coordinates fed back by the CNC system of the machine tool. It corresponds to the minimum motion displacement required for the device to start from a stationary state and reach a uniform and stable scanning state, and also covers the minimum motion displacement required for the device to decelerate from a uniform and stable scanning state to a stationary state. The variable speed distance D is used to offset the position control deviation caused by the acceleration and deceleration process during the start and end of the scanning motion, and also to reserve sufficient space clearance between the line laser vision device 100 and the edge structure of the machine tool and the surrounding support components.

[0032] It should be noted that the calibration component 200 in this embodiment is a single calibration component with a standard circular outer edge. During calibration, its center can be used as the calibration reference feature, or multiple sets of sampling points on the circular contour can be used as the calibration reference feature. That is, the complete point cloud data of the circular contour is collected by the line laser vision device 100, the three-dimensional coordinates of each sampling point on the circumference in the visual coordinate system are extracted, and the actual spatial coordinates of each sampling point in the machine tool coordinate system are collected accordingly. Compared with the prior art, which requires at least two calibration plates with height differences, the present invention can complete the visual calibration using only a single circular calibration component 200. While ensuring the stability of feature point extraction and the accuracy of calibration solution by utilizing its center or circular contour, the calibration process is simplified, the number of calibration components 200 is reduced, and the adverse effects of the accumulation of positional errors between multiple calibration plates on the calibration results are avoided.

[0033] Step 2: Set the scanning line length L0 of the line laser vision device 100, and determine the maximum scanning area 600a according to the transformation relationship δ, the speed change distance D and the scanning line length L0. That is, set the scanning line length L0 of the line laser vision device 100, and then complete the coordinate system mapping based on the calibrated transformation relationship δ. Combine the margin configuration of the speed change distance D and the effective acquisition range of the scanning line length L0 to determine the maximum scanning area 600a. Wherein, the scanning line segment length L0 is the effective data acquisition stroke length of the laser line of the line laser vision device 100 along its lateral extension direction (i.e., the line segment length direction). The maximum scanning area 600a is the coverage area of ​​the worktable 500 surface where the line laser vision device 100 can stably output high-precision point cloud data. Its boundary in the direction of beam movement is constrained by the speed change distance D, and its boundary in the direction of line segment length is constrained by the scanning line segment length L0. That is, when the line laser vision device 100 makes lateral reciprocating motion along the Y-axis of the beam, both ends of the scanning line segment length L0 are converged inward to within the boundary of the effective processing area of ​​the worktable 500 surface, so that the entire effective acquisition field of view of the line laser vision device 100 in the direction of line segment length corresponds to the space above the worktable 500 surface, avoiding the line laser beam from irradiating non-target structures such as the machine tool edge frame and side support components when extending outward to the worktable 500.

[0034] Understandably, if the aforementioned non-target structures are included in the scanning range, a large number of interfering point clouds unrelated to the identification of the scrap 300 will be introduced into the point cloud data. This will not only significantly increase the computational burden of subsequent point cloud filtering and segmentation algorithms, but also easily lead to missegmentation of the point cloud due to confusion between the height features of the edge structures and the boundaries of the scrap 300, thereby interfering with the accurate extraction of the scrap 300 contour. Therefore, strictly limiting the maximum scanning area 600a based on the scanning line segment length L0 can effectively suppress the point cloud interference introduced by non-target structures, improve the signal-to-noise ratio and purity of the original point cloud dataset, and not only fully cover the conventional placement area of ​​the scrap 300 on the workbench 500 to ensure comprehensive acquisition of the scrap 300 data, but also avoid the risk of the scanning stroke exceeding the safe movement boundary of the equipment.

[0035] Step 3: The control line laser vision device 100 scans and acquires the remaining material 300 within the maximum scanning area 600a. The point cloud data of the remaining material 300 acquired under uniform motion is then processed by point cloud filtering and segmentation, coordinate transformation, edge point extraction and segment fitting to generate a contour image file of the remaining material 300 adapted to the machine tool coordinate system.

[0036] This method not only effectively eliminates the potential for collisions or scratches between the line laser vision device 100 and the edge structure and supporting components of the machine tool, avoiding wear, optical path deviation, and even hardware damage caused by collisions, but also effectively extends the service life and optical path stability of the line laser vision device 100. Simultaneously, it avoids position control deviations caused by acceleration and deceleration during the scanning start and end stages from the perspective of data acquisition quality, and effectively shields the structural interference point cloud generated by the edge structure and supporting components of the machine tool. It provides dual protection for the integrity, purity, and coordinate transformation calibration accuracy of the point cloud acquisition data from both the motion and field-of-view dimensions, reducing the algorithmic complexity and misidentification probability of subsequent point cloud segmentation and contour fitting, thereby improving the accuracy of the contour recognition of the remaining material 300.

[0037] In the above embodiments, please refer to Figure 2 and Figure 3 The specific steps in Step 2 above, "determining the maximum scanning area based on the conversion relationship δ, the speed change distance D, and the scanning line segment length L0," include: Using the positive limit of the X-axis of the machine tool coordinate system as a reference, offset the speed change distance D in the negative X-axis direction, and use the negative limit of the X-axis of the machine tool coordinate system as a reference to offset the speed change distance D in the positive X-axis direction, to obtain the initial positive side boundary and the initial negative side boundary in the X-axis direction. That is, by offsetting the speed change distance D to the inner side of the worktable 500 on the basis of the positive and negative limits of the X-axis, the initial positive side boundary and the initial negative side boundary together constitute the basic safe scanning range of the X-axis. This basic safe scanning range will eliminate the acceleration and deceleration section when the crossbeam moves to the limit position, ensuring that the line laser vision device 100 is in a uniform and stable scanning state throughout the movement within this basic safe scanning range. At the same time, sufficient safety avoidance margin is reserved for the line laser vision device 100 and the edge structure and support components of the machine tool, thus avoiding the risk of collision and scratch from the stroke level.

[0038] Based on the orientation of the line laser vision device 100 along the X-axis of the machine tool coordinate system, the X-axis offset obtained by the transformation relationship δ is superimposed onto the offset of the initial positive or negative side boundary to determine the X-axis boundary of the maximum scanning area 600a.

[0039] That is, based on the difference in installation position between the line laser vision device 100 and the laser cutting head 400, an X-axis offset is introduced to correct the orientation of the initial boundary. Specifically, the cutting center point of the laser cutting head 400 is the reference control point for the interpolation motion of the machine tool CNC system. Its coordinate value in the machine tool coordinate system directly corresponds to the motion command coordinate of the CNC system. The laser cutting head 400 moves longitudinally along the X-axis with the crossbeam and moves laterally along the Y-axis of the crossbeam with the slide. The full range of motion of the laser cutting head 400 is strictly constrained by the limits of each axis of the machine tool coordinate system. In this embodiment, the line laser vision device 100 can be installed on the laser cutting head 400 or on the slide assembly where it is located. Therefore, the line laser vision device 100 is not coaxially arranged with the cutting center point. The X-axis coverage of the maximum scanning area 600a will be shifted as a whole depending on the installation orientation.

[0040] If the initial positive and negative side boundaries of the basic safe scanning range are directly used as the X-axis boundary of the maximum scanning area 600a, the scanning range on one side will exceed the effective processing area of ​​the worktable 500 or the coverage of the scanning on one side will be insufficient. Therefore, the X-axis offset amount needs to be superimposed on the offset amount of the corresponding side boundary according to the axial orientation of the line laser vision device 100. This ensures that the X-axis boundary of the final maximum scanning area 600a meets the stroke accuracy requirements of uniform speed scanning and is compatible with the actual installation position of the line laser vision device 100. It also ensures that the effective scanning field of view completely covers the area where the remaining material 300 is placed on the worktable 500 and does not exceed the safe movement range, and also avoids non-target structures from entering the scanning field of view and introducing interference point clouds.

[0041] It needs to be clarified that the positive X-axis limit is the maximum safe travel endpoint of the X-axis in the positive direction, preset by the machine tool CNC system. It corresponds to the extreme value of the X-axis coordinate of the cutting center point when the crossbeam moves to the positive limit position of the bed, and is the positive safety constraint boundary for the X-axis movement of the machine tool. The negative X-axis limit is the maximum safe travel endpoint of the X-axis in the negative direction, preset by the machine tool CNC system. It corresponds to the extreme value of the X-axis coordinate of the cutting center point when the crossbeam moves to the negative limit position of the bed, and is the negative safety constraint boundary for the X-axis movement of the machine tool. The negative X-axis direction is the direction of movement in which the X-axis coordinate value decreases in the machine tool coordinate system, corresponding to the crossbeam driving the cutting head towards the negative X-axis limit. The direction of the positional movement; the positive X-axis direction is the direction of movement in which the X-axis coordinate value increases in the machine tool coordinate system, corresponding to the direction in which the crossbeam drives the cutting head to move toward the positive X-axis limit position; the X-axis offset is the coordinate difference between the center line of the effective scanning area of ​​the line laser vision device 100 and the cutting center point of the laser cutting head 400 in the X-axis of the machine tool coordinate system. This difference needs to be converted and obtained by mapping the installation position of the line laser vision device 100 in the vision coordinate system to the machine tool coordinate system through the calibrated transformation relationship δ, and is used to quantitatively characterize the fixed spatial position offset relationship between the line laser vision device 100 and the laser cutting head 400 in the X-axis.

[0042] For example, the orientation of the line laser vision device 100 relative to the cutting center point can be divided into positive X-axis orientation and negative X-axis orientation based on the machine tool coordinate system. The positive X-axis orientation refers to the line laser vision device 100 being located on the side of the cutting center point closer to the larger X-axis coordinate value, including positive X-axis collinear orientation, positive X-axis Y-axis offset orientation, and positive X-axis Y-axis offset orientation. The negative X-axis orientation refers to the line laser vision device 100 being located on the side of the cutting center point closer to the smaller X-axis coordinate value, including negative X-axis collinear orientation, negative X-axis Y-axis offset orientation, and negative X-axis Y-axis offset orientation.

[0043] Understandably, the X-axis offset in both the positive X-axis and positive Y-axis offset configurations, and the X-axis offset in both the positive X-axis and negative Y-axis offset configurations, is ΔX. This X-axis offset is determined solely by the installation position of the line laser vision device 100 in the X-axis direction and is unaffected by Y-axis offset. Therefore, the calculation method for the X-axis boundary of the maximum scanning area 600a is completely consistent with the collinear configuration in the positive X-axis direction. Similarly, the X-axis offset in both the negative X-axis and positive Y-axis offset configurations, and the X-axis offset in both the negative X-axis and negative Y-axis offset configurations, is ΔX. This X-axis offset is determined solely by the installation position of the line laser vision device 100 in the X-axis direction and is unaffected by Y-axis offset. Therefore, the calculation method for the X-axis boundary of the maximum scanning area 600a is completely consistent with the collinear configuration in the negative X-axis direction.

[0044] Assume the positive limit coordinate of the X-axis of the laser cutting machine's coordinate system is X. 正 The negative limit coordinate of the X-axis is X 负 The absolute value of the X-axis offset of the line laser vision device 100 relative to the cutting center point, obtained by calibration and conversion relationship δ, is ΔX. When the line laser vision device 100 is located on the positive X-axis side, the offset is negative; when located on the negative X-axis side, the offset is positive. When the line laser vision device 100 is arranged collinearly on the positive X-axis, the entire effective scanning area of ​​the line laser vision device 100 is offset by a distance ΔX in the positive X-axis direction. In this case, the initial positive boundary is first calculated as X. 正 -D, The initial negative boundary is X 负 +D, Since the effective scanning area of ​​the line laser vision device 100 is offset towards the positive X-axis, if the initial positive side boundary is directly used as the positive endpoint of the scanning motion, the positive endpoint of the scanning field of view of the line laser vision device 100 will enter the acceleration / deceleration section or even exceed the safe travel range. Therefore, the X-axis offset ΔX needs to be added to the offset of the initial positive side boundary, and finally the positive side boundary along the X-axis is X. 正 -D-ΔX, the negative boundary along the X-axis remains X. 负+D, thus ensuring that the entire effective scanning field of view of the line laser vision device 100 is within the uniform and safe range throughout the entire scanning stroke.

[0045] When the line laser vision device 100 adopts a collinear arrangement along the negative X-axis, the entire effective scanning area of ​​the line laser vision device 100 is shifted by a distance ΔX along the negative X-axis. At this time, the initial positive side boundary and the initial negative side boundary are still X. 正 -D and X 负 +D, because the scanning area of ​​the line laser vision device 100 is offset in the negative X-axis direction, if the initial negative side boundary is directly used as the negative endpoint of the scanning motion, the negative endpoint of the scanning field of view of the line laser vision device 100 will enter the acceleration / deceleration section or even exceed the safe travel range. Therefore, the absolute value of the X-axis offset ΔX needs to be superimposed on the offset of the initial negative side boundary, and finally the X-axis negative side boundary is X. 负 +D+ΔX, the positive lateral boundary along the X-axis remains X. 正 -D ensures that the negative endpoint of the 100 scanning field of view of the line laser vision device is always within the uniform and safe range, avoiding the risk of data accuracy degradation or collision.

[0046] Further, please refer to Figure 2 and Figure 3 The specific steps in Step 2 above, "determining the maximum scanning area 600a based on the conversion relationship δ, the speed change distance D, and the scanning line segment length L0," also include: Based on the positive and negative Y-axis limits of the machine tool coordinate system, and according to the orientation of the line laser vision device 100 along the Y-axis of the machine tool coordinate system, the Y-axis offset obtained by the transformation relationship δ and the offset of half the scanning line segment length L0 are superimposed on the offsets of the positive and negative Y-axis limits to determine the Y-axis boundary of the maximum scanning area 600a.

[0047] In this way, this step completes the boundary constraint of the maximum scanning area 600a in the online laser emission direction, which, together with the aforementioned X-axis boundary, constitutes the closed maximum scanning area 600a. Specifically, the effective acquisition line segment of the line laser vision device 100 extends along the Y-axis, with a scanning line segment length of L0. Based on the center line of the effective scanning area, it covers half of the acquisition range on both the positive and negative sides of the Y-axis. Since the line laser vision device 100 is mounted on the laser cutting head 400 or its slide assembly, there is a Y-axis installation position offset relative to the cutting center point. The overall coverage of the effective acquisition line segment will be shifted as a whole with the installation orientation.

[0048] If the positive and negative Y-axis limits are directly used as the boundaries of the maximum scanning area 600a, the effective acquisition line segments on one side will exceed the effective processing area of ​​the worktable 500, irradiating non-target structures such as the machine tool edge frame and side support components, introducing a large number of structural interference point clouds. Therefore, it is necessary to use the boundaries of the worktable 500 corresponding to the positive and negative Y-axis limits as a reference, and combine the Y-axis layout orientation of the line laser vision device 100 to directionally superimpose the Y-axis offset and the offset of the half-segment scanning line segment length into the offset calculation of the corresponding side boundary. This ensures that both ends of the scanning line segment length L0 converge inward to within the Y-axis boundary of the effective processing area of ​​the worktable 500, so that all effective acquisition line segments of the line laser correspond to the space above the worktable 500, reducing the introduction of structural interference point clouds, improving the signal-to-noise ratio and purity of the original point cloud data, and ensuring the matching accuracy between the scanning coverage and the processing area of ​​the worktable 500, avoiding scanning blind spots or invalid scanning areas.

[0049] It needs to be clarified that the positive Y-axis limit is the maximum safe travel endpoint of the Y-axis in the positive direction, preset by the machine tool CNC system. It corresponds to the extreme value of the Y-axis coordinate of the cutting center point when the slide moves to the positive limit position along the extension direction of the crossbeam. It is the positive safety constraint boundary for the Y-axis movement of the machine tool, corresponding to the positive Y-axis boundary of the effective machining area of ​​the 500-face worktable. The negative Y-axis limit is the maximum safe travel endpoint of the Y-axis in the negative direction, preset by the machine tool CNC system. It corresponds to the extreme value of the Y-axis coordinate of the cutting center point when the slide moves to the negative limit position along the extension direction of the crossbeam. It is the positive safety constraint boundary for the Y-axis movement of the machine tool, corresponding to the positive Y-axis boundary of the effective machining area of ​​the 500-face worktable. The negative safety constraint boundary for axial movement corresponds to the negative boundary of the effective processing area of ​​the worktable 500 surface in the Y direction; the Y-axis offset is the coordinate difference between the center line of the effective scanning area of ​​the line laser vision device 100 and the cutting center point of the laser cutting head 400 in the Y-axis of the machine tool coordinate system. This difference needs to be converted by mapping the installation position of the line laser vision device 100 in the vision coordinate system to the machine tool coordinate system through the calibrated transformation relationship δ, and is used to quantitatively characterize the fixed spatial position offset relationship between the line laser vision device 100 and the laser cutting head 400 in the Y-axis.

[0050] For example, assume that the positive limit coordinate of the Y-axis of the laser cutting machine's coordinate system is Y. 正 The negative limit coordinate of the Y-axis is Y 负 The absolute value of the Y-axis offset of the line laser vision device 100 relative to the cutting center point, obtained by conversion using the calibration relationship δ, is ΔY. The X-axis layout is explained in detail below.

[0051] When the line laser vision device 100 adopts a collinear layout along the positive X-axis, the center line of the effective scanning area and the cutting center point perfectly coincide in the Y-axis coordinates, with a Y-axis offset of 0. The effective acquisition line segment extends L0 / 2 on both the positive and negative sides of the Y-axis, centered on the Y-coordinate of the cutting center point. Therefore, using the positive Y-axis limit as a reference, offsetting half the scanning line segment length L0 towards the inside of the worktable yields the positive lateral boundary of the Y-axis. 正 -L0 / 2, using the negative Y-axis limit as a reference, offsetting half the scan line segment length L0 towards the inside of the worktable to obtain the negative Y-axis boundary. 负 +L0 / 2, under this boundary, the cutting center point can ensure that both ends of the scanning line length L0 do not exceed the upward boundary of the worktable 500Y, and together with the aforementioned X-axis boundary, they form the maximum scanning area 600a.

[0052] When the line laser vision device 100 adopts a positive X-axis and positive Y-axis offset configuration, the center line of the effective scanning area is offset by ΔY relative to the cutting center point in the positive Y-axis direction. The entire effective acquisition line segment is translated by ΔY in the positive Y-axis direction along with the installation position. To ensure that the positive endpoint of the effective acquisition line segment does not exceed the boundary of the worktable 500 corresponding to the positive Y-axis limit, it is necessary to add the total offset of the Y-axis offset and half the length of the scanning line segment to the inside based on the positive Y-axis limit, so as to obtain the positive lateral boundary of the Y-axis as Y. 正 -ΔY-L0 / 2, while the effective acquisition line segment's negative endpoint is shifted in the positive direction, resulting in a corresponding increase in the safe distance from the negative limit of the Y-axis. Therefore, the negative side boundary of the Y-axis is Y. 负 -ΔY+L0 / 2, this boundary setting ensures that the entire effective acquisition line segment is within the effective processing area of ​​the worktable 500, which, together with the aforementioned X-axis boundary, constitutes the maximum scanning area 600a.

[0053] When the line laser vision device 100 adopts a positive X-axis and negative Y-axis offset configuration, the center line of the effective scanning area is offset by ΔY relative to the cutting center point in the negative Y-axis direction. The entire effective acquisition line segment is translated by ΔY in the negative Y-axis direction along with the installation position. To ensure that the negative endpoint of the effective acquisition line segment does not exceed the boundary of the worktable 500 corresponding to the negative Y-axis limit, it is necessary to superimpose the absolute value of the Y-axis offset and half the length of the scanning line segment to the inner side on the basis of the negative Y-axis limit, so as to obtain the negative Y-axis boundary as Y. 负 +ΔY+L0 / 2, while the effective acquisition line segment's positive endpoint is shifted negatively overall, increasing the safe distance from the positive limit of the Y-axis. Therefore, the positive lateral boundary of the Y-axis is Y. 正 +ΔY-L0 / 2, this boundary setting ensures that the entire effective acquisition line segment is within the effective processing area of ​​the worktable 500, which, together with the aforementioned X-axis boundary, constitutes the maximum scanning area 600a.

[0054] Understandably, when the line laser vision device 100 adopts a collinear arrangement in the negative X-axis direction, the center line of the effective scanning area and the cutting center point completely coincide in the Y-axis direction, and the Y-axis offset is 0. Therefore, the calculation method of the Y-axis boundary is completely consistent with the collinear arrangement in the positive X-axis direction, that is, the positive lateral boundary of the Y-axis is Y... 正 -L0 / 2, the negative boundary of the Y-axis is Y 负 +L0 / 2, the constraint logic of the Y-axis boundary is not affected by the X-axis installation position; when the line laser vision device 100 adopts the X-axis negative and Y-axis positive offset layout, the center line of the effective scanning area is offset by ΔY in the positive Y-axis direction relative to the cutting center point. Therefore, the calculation method for the Y-axis boundary is completely consistent with the X-axis positive and Y-axis positive offset layout, i.e., the positive Y-axis lateral boundary is Y. 正 -ΔY-L0 / 2, the negative boundary along the Y-axis is Y 负 -ΔY+L0 / 2, the installation offset along the X-axis only changes the position and boundary of the maximum scanning area 600a along the X-axis, without changing the calculation rules and constraint effect of the Y-axis boundary; when the line laser vision device 100 adopts a negative X-axis and negative Y-axis offset layout, the center line of the effective scanning area is offset ΔY in the negative Y-axis direction relative to the cutting center point. Therefore, the calculation method for the Y-axis boundary is completely consistent with the positive X-axis and negative Y-axis offset layout, i.e., the negative Y-axis boundary is Y... 负 +ΔY+L0 / 2, the positive lateral boundary along the Y-axis is Y 正 +ΔY-L0 / 2, the difference in installation position along the X-axis does not affect the constraint logic and safety effect of the Y-axis boundary.

[0055] In the above embodiment, the step of "determining the shift distance D" in Step 1 includes: The coordinates of the start and end positions of the line laser vision device 100 in the machine tool coordinate system when it accelerates or decelerates between a stationary state and a uniform scanning speed are obtained, and the speed change distance D is calculated based on the coordinate difference between the start and end positions.

[0056] This step obtains the speed change distance parameter by measuring the difference in motion coordinates. It can directly adapt to the different characteristics of the drive performance of different models, the actual load inertia of the crossbeam and the equipment installation status, eliminate the deviation between theoretical calculation values ​​and actual operating conditions, and avoid the problem of the effective scanning range being included in the acceleration and deceleration section due to the theoretical calculation value being too small, or the effective scanning stroke being wasted due to the calculation value being too large.

[0057] For example, the instantaneous coordinates and corresponding speed values ​​during the motion process are collected in real time through the position loop feedback data of the machine tool CNC system. When the running speed of the line laser vision device 100 is detected to reach the target uniform scanning speed for the first time, the X-axis coordinate of the machine tool coordinate system corresponding to the current position is recorded as the end position of the acceleration process. The X-axis coordinate of the stationary starting point of the line laser vision device 100 is taken as the starting position of the acceleration process. The difference between the two coordinates is the speed change distance D of the positive acceleration segment. Then, the line laser vision device 100 is controlled to decelerate along the X-axis until it comes to a complete stop at the same uniform speed. The difference between the coordinates of the deceleration starting position and the deceleration ending position is recorded to obtain the speed change distance D of the deceleration segment.

[0058] Alternatively, please refer to Figure 4 Before the step of "controlling the laser vision device 100 to scan the remaining material 300" in Step 3 above, there is also a step of correcting the start and end positions of the scan: Using the negative X-axis boundary of the maximum scanning area 600a as a reference, and combining the speed change distance D, the X-axis coordinate of the preset scanning start point (hereinafter referred to as the preset X coordinate) is subjected to boundary limiting processing and offset calculation. Using the negative Y-axis boundary of the maximum scanning area 600a as a reference, the Y-axis coordinate of the preset scanning start point (hereinafter referred to as the preset Y coordinate) is subjected to boundary limiting processing to obtain the corrected scanning start point coordinate. Specifically, since the line laser vision device 100 needs to start from a stationary state and undergo an acceleration process before entering a uniform scanning state, an additional speed change distance D needs to be reserved in the negative direction as an acceleration stroke space based on the negative boundary of the X-axis of the maximum scanning area 600a or the preset X coordinate within the maximum scanning area 600a. That is, the theoretical starting position of the acceleration segment is obtained by offsetting the speed change distance D in the negative X-axis direction based on the negative boundary of the X-axis or the preset X coordinate. This position also corresponds to the machine tool's safe travel limit in the negative X-axis direction. The preset X coordinate needs to be uniformly corrected to this position through boundary limiting processing and offset calculation to avoid exceeding the machine tool's safe travel limit and to ensure that the acceleration stroke length is exactly equal to the speed change distance D, so that the line laser vision device 100 has fully entered a uniform and stable scanning state when it moves to the negative boundary of the X-axis or the preset X coordinate.

[0059] Based on the corrected scanning start coordinates, the scanning dimensions corresponding to the preset scanning area along the axis are superimposed. Then, boundary limiting processing is performed with the positive and negative boundaries of the X-axis and Y-axis corresponding to the maximum scanning area 600a as the reference to obtain the corrected scanning end coordinates. This ensures that the scanning end will not exceed the positive and negative boundaries of the X-axis and Y-axis of the maximum scanning area 600a. Combined with the corrected scanning start coordinates, the effective data acquisition range of the entire scanning stroke will always be within the dual constraints of the uniform speed safe scanning range and the maximum scanning area 600a. This can not only adapt to the personalized scanning path planning needs of different sizes and specifications of 300mm scrap material, but also effectively prevent various accuracy and safety hazards caused by out-of-boundary scanning.

[0060] It should be noted that the aforementioned boundary limiting processing refers to a numerical correction process that determines and constrains the relationship between preset coordinate values ​​and preset boundary thresholds. In this embodiment, when the coordinate value of the preset scanning start point exceeds the boundary threshold range of the axis corresponding to the maximum scanning area 600a, the coordinate value of the preset scanning start point is forcibly constrained to the corresponding boundary threshold. When the preset scanning start point is within the boundary threshold range of the axis corresponding to the maximum scanning area 600a, the original preset coordinate value is retained unchanged. This ensures that the final output scanning start and end coordinates are always constrained within the range of the maximum scanning area 600a, avoiding risks such as data distortion in acceleration and deceleration sections, introduction of non-target structure interference point clouds, or equipment collision and scratches caused by the preset scanning path exceeding the boundary of the maximum scanning area 600a.

[0061] It should also be noted that the preset X-coordinate and preset Y-coordinate together constitute the preset scanning starting point of the preset scanning area, corresponding to the corner point where the X-coordinate and Y-coordinate of the preset scanning area are minimized in the machine tool coordinate system. The preset scanning area is a rectangular scanning area pre-planned according to the placement and dimensions of the scrap material 300 to be identified.

[0062] For example, when the preset scanning start point of the preset scanning area is within the maximum scanning area 600a, and the initial scanning end point coordinates obtained by superimposing the scanning size parameters of the corresponding axis of the preset scanning area do not exceed the maximum scanning area 600a, firstly, boundary limiting processing is performed on the preset scanning start point. The preset X coordinate is greater than the negative boundary of the X-axis, so no limiting correction is needed. The preset Y coordinate is greater than the negative boundary of the Y-axis, so no limiting correction is needed. Then, the preset X coordinate is shifted by a speed change distance D in the negative X-axis direction to obtain the corrected X-axis coordinate of the scanning start point. The preset Y coordinate remains unchanged, thus obtaining the complete corrected scanning start point coordinates. Then, the scanning size of the corresponding axis of the preset scanning area is superimposed on the corrected scanning start point coordinates to obtain the initial scanning end point coordinates. Positive boundary limiting processing is performed on them. The X and Y axis values ​​are both less than the corresponding positive boundary thresholds, so no limiting correction is needed. Finally, the corrected scanning end point coordinates are consistent with the initial calculated values.

[0063] When the preset X-coordinate is less than the negative boundary of the X-axis and the preset Y-coordinate is less than the negative boundary of the Y-axis, boundary limiting processing is first performed on the preset scan start point. The preset X-coordinate breaks through the negative boundary constraint of the X-axis, triggering limiting and clamping to the negative boundary of the X-axis. The preset Y-coordinate breaks through the negative boundary constraint of the Y-axis, triggering limiting and clamping to the negative boundary of the Y-axis. Then, the X-axis coordinate after limiting is offset by a variable speed distance D in the negative X-axis direction to obtain the corrected X-axis coordinate of the scan start point. The preset Y-coordinate retains the value after limiting, and finally, the complete corrected scan start point coordinate is obtained. Then, the scan size of the corresponding axis of the preset scan area is superimposed on the corrected scan start point coordinate to obtain the initial scan end point coordinate. Positive boundary limiting processing is performed on it. Its X and Y axis values ​​are both less than the corresponding positive boundary threshold, so no limiting correction is required. Finally, the corrected scan end point coordinate is consistent with the initial calculated value.

[0064] When the preset X-coordinate is less than the negative boundary of the X-axis and the preset Y-coordinate is greater than the negative boundary of the Y-axis, boundary limiting processing is first performed on the preset scan start point. The preset X-coordinate breaks through the negative boundary constraint of the X-axis, triggering limiting and clamping to the negative boundary of the X-axis. The preset Y-coordinate meets the constraint requirements and does not need to be corrected. Then, the X-axis coordinate after limiting is offset by a variable speed distance D in the negative X-axis direction to obtain the corrected X-axis coordinate of the scan start point. The preset Y-coordinate remains unchanged, thus obtaining the corrected scan start point coordinate. The scan size of the corresponding axis of the preset scan area is superimposed on the corrected scan start point coordinate to obtain the initial scan end point coordinate. Positive boundary limiting processing is performed on it. Its X and Y axis values ​​are both less than the corresponding positive boundary threshold, and no limiting correction is needed. Finally, the corrected scan end point coordinate is consistent with the initial calculated value.

[0065] In the above, if the initial scanning endpoint coordinates are obtained by superimposing the scanning size of the preset scanning area along the corresponding axis on the corrected scanning start coordinates, and the positive and side boundary limiting processing is performed on them, the initial X-axis endpoint coordinates exceed the positive and side boundary threshold, triggering the limiting processing and clamping to the positive and side boundary of the X-axis. The initial Y-axis endpoint coordinates exceed the positive and side boundary of the Y-axis, triggering the limiting processing and clamping to the positive and side boundary of the Y-axis. Finally, the corrected scanning endpoint coordinates are obtained.

[0066] Further, please refer to Figure 4 Before the control line laser vision equipment 100 scans the remaining material 300, a scanning width correction step is also included: The total scan width W is calculated based on the corrected scan start and end coordinates. 扫 The number of scans n is calculated by combining the length of the scan line segment L0 and rounded up. Then, based on the number of scans n and the total scan width W, the result is calculated. 扫 Determine the width of a single scan.

[0067] Specifically, the total scan width W扫 To correct the coordinate difference between the scanning endpoint and the scanning start point along the Y-axis, the actual effective coverage span of the preset scanning area along the Y-axis is determined. Since the effective acquisition range of a single transverse scan by a line laser vision device is the length L0 of the scanning line segment, when the total scanning width W... 扫 When the length of the scan line segment is greater than L0, full width coverage needs to be achieved through multiple reciprocating scans.

[0068] As a preferred embodiment, the total scanning width W is first... 扫 Divide by half the scan line segment length L0 to obtain the ratio parameter i;

[0069] Substituting the ratio parameter i into the preset formula F(n) yields the theoretical value of the number of scans n. Then, the theoretical value is rounded up to obtain the final number of scans n. This rounding process ensures that the total coverage of the number of scans n is always not less than the total scan width W. 扫 To meet coverage requirements and avoid blind spots due to insufficient scan counts (n);

[0070] Based on this, the total scan width W 扫 After deducting the length L0 of half-scan line segment, the result is divided equally according to the number of scans n to obtain the step width W of a single scan. 单 This makes the step width W of each scan channel... 单 Maintaining uniformity and consistency ensures that the point cloud acquisition density is uniform and stable throughout the entire scanning area. At the same time, it ensures that there is always sufficient overlap between adjacent scanning channels, providing sufficient feature benchmarks for subsequent splicing and matching of multiple point cloud data, thereby improving the accuracy of point cloud splicing and the reliability of residual material contour recognition.

[0071]

[0072] For example, when the total scan width W 扫 When the scan width is exactly an integer multiple of the half-width scan segment length L0, the theoretical value of the calculated scan count n is an integer. After rounding up, the scan count remains unchanged. The width of a single scan is equal to the half-width scan segment length, and a fixed half-width overlap ratio is maintained between adjacent scan paths. When the total scan width W... 扫 When the scan length L0 is not an integer multiple of the half-scan line segment length, the number of scans n after rounding up is greater than the theoretically calculated value, corresponding to a step width W for a single scan. 单 When the length of the scan line segment is less than half the width of the scan, the overlap ratio of adjacent scan paths increases accordingly, but it can still completely cover the entire total scan width range.

[0073] In the above embodiments, please refer to Figure 5The step "using calibration component 200 to visually calibrate the line laser vision device 100" in Step 1 above also includes: Using the upper surface of calibration component 200 as the calibration reference surface, and combining its thickness N, the zero-position height reference calibration in the Z direction is completed. The vertical distance parameter H between the line laser vision device 100 and the calibration reference surface is determined, and the Z-axis coordinate value of the worktable 500 surface in the machine tool coordinate system is derived as -N.

[0074] Specifically, this step, based on solving the XY plane coordinate system transformation relationship of the aforementioned circular contour features, simultaneously completes the Z-axis zero-position height reference calibration, so that the three-dimensional point cloud data collected by the line laser vision device 100 can be completely and accurately mapped into the three-dimensional space system of the machine tool coordinate system. The calibration component 200 is a single standard component with a known standard thickness. When in use, it is placed flat on the worktable 500. Its upper surface is parallel to the worktable 500 and the distance between them is constant at a thickness N. Using this upper surface as the calibration reference surface, the flatness characteristics of its high-precision machining can be used to improve the acquisition and calibration accuracy of the Z-axis height data. The determined vertical distance parameter H can be used as the inherent calibration reference value of the height direction of the line laser vision device 100, which is used for the height normalization processing of the subsequent original point cloud data and the extraction and calculation of the thickness feature of the scrap 300. The Z-axis coordinate value of the worktable 500 is derived in reverse through the thickness N of the calibration component 200.

[0075] In the above embodiments, please refer to Figure 6 The step in Step 1 above, "after the control line laser vision device 100 scans and acquires the point cloud data of the remaining material 300, the point cloud data is processed," includes: Based on the Z-axis zero-position height reference and the thickness parameter of the remaining material 300, the effective point cloud height threshold range is determined. The point cloud data is filtered in the Z direction according to the height threshold range. Interference point clouds outside the height threshold range are removed to obtain the valid point cloud data corresponding to the remaining material 300.

[0076] This step, combined with the aforementioned maximum scanning area 600a and the field-of-view constraints of the preset scanning area, forms a spatial three-dimensional joint filtering mechanism. This mechanism can further eliminate various interfering point clouds within the field of view that do not match in height characteristics. The lower limit of the height threshold interval is based on the worktable 500 surface, and the upper limit is determined by the thickness parameter of the worktable 500 surface plus the remaining material 300. Of course, the upper limit of the height threshold interval can reserve an appropriate upper tolerance range based on the actual deformation characteristics of the remaining material 300, ensuring that all effectively acquired point clouds on the upper surface of the remaining material 300 are included in the filtering interval. This Z-axis filtering method does not rely on complex point cloud segmentation algorithms; it can quickly complete the batch removal of interfering point clouds simply by determining the height dimension, resulting in high computational efficiency. Figure 6As shown, the logic of Z-axis screening is stable and reliable, which can effectively reduce the computational load of subsequent point cloud segmentation, edge extraction and contour fitting. It also avoids confusion between interference structures with large height differences and the boundary features of the leftover material 300, further improving the accuracy and stability of the contour recognition of the leftover material 300.

[0077] For example, such as Figure 8 As shown, firstly, based on the Z-axis zero-position height reference, the reference height value of the 500-level worktable surface is determined, corresponding to the Z-axis coordinate in the machine tool coordinate system as -N. Then, combined with the remaining material of 300 mm for the current thickness parameter M, the lower limit of the effective point cloud height threshold range is determined to be Z. min =-N, upper limit is Z max =-N+M forms a closed effective height threshold range. After Z-axis filtering, all system noise points with heights below the lower limit of this range, laser scattering noise points in the air, and interference point clouds such as machine tool side support structures and surrounding protective components with heights above the upper limit of this range are quickly filtered out, and the point cloud coordinate data corresponding to the remaining material 300 is accurately extracted.

[0078] Further, in Step 3 above, after obtaining the effective three-dimensional point cloud data corresponding to the scrap material 300 through Z-axis filtering, the inner and outer contour point cloud sets of the scrap material 300 are extracted as a whole. The Z-axis coordinate dimension of each contour point cloud is removed, and the three-dimensional point cloud data of each contour is converted into two-dimensional planar point cloud data. Then, the coordinate mapping is completed according to the transformation relationship δ between the coordinate system of the line laser vision device 100 and the machine tool coordinate system. The coordinate correction is performed by combining the X-axis offset and Y-axis offset of the line laser vision device 100 relative to the cutting center point, and the actual coordinate point cloud of the scrap material 300 contour under the machine tool coordinate system is obtained. After connecting and fitting the point cloud coordinates of each contour in the spatial topology order, the scrap material 300 contour based on the machine tool coordinate system can be reconstructed. Finally, the scrap material 300 contour image file that can be directly adapted to the machine tool CNC system and called by the subsequent nesting process is generated.

[0079] It should be noted that the step of "controlling the laser vision device 100 to scan the scrap material 300" in Step 1 above includes: Plan the line-by-line scanning path based on the corrected scan start coordinates, scan end coordinates, scan count n, and single scan width; The drive line laser vision device 100 moves along the scanning path of each row, collecting point cloud data of the remaining material 300 within the uniform speed movement interval. The row-by-row scanning path extends along the X-axis and advances sequentially along the Y-axis according to the width of a single scan. Acceleration and deceleration sections corresponding to the speed difference are reserved at both ends of the X-axis stroke of each scan. The data acquisition action is initiated only after the line laser vision device 100 enters the uniform speed movement state and terminates before entering the deceleration section.

[0080] If the scanning path is N-shaped, all scanning lines start from the Y-axis position of the corrected scanning start point, complete a single scan along the X-axis, and then step along the Y-axis by one single scan width to start the next scan, and so on to complete all n scans. If an S-shaped scanning path is used, when the number of scans n is even, reciprocating continuous scanning can be achieved by swapping the start and end coordinates of the even-numbered scan lines, so that the end position of the previous scan is directly used as the start position of the next scan.

[0081] This application also proposes a laser cutting apparatus, which includes a laser cutting head 400, a line laser vision device 100 disposed on one side of the laser cutting head 400, and a control processing unit capable of executing the above-described laser cutting waste 300 identification method. The control processing unit is electrically connected to the laser cutting head 400 and the line laser vision device 100 to control the laser cutting head 400 to drive the line laser vision device 100 to move along the machine tool coordinate system. The specific structure of the laser cutting waste 300 identification method is as described in the above embodiments. Since this vision-based laser cutting apparatus adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be elaborated here.

[0082] The laser cutting device also includes a bed, a crossbeam drive assembly arranged along the X-axis of the bed, and a slide assembly slidably mounted on the crossbeam drive assembly. The laser cutting head 400 and the line laser vision device 100 are both mounted on the slide assembly. The slide assembly can drive the laser cutting head 400 and the line laser vision device 100 to reciprocate synchronously along the Y-axis of the crossbeam. The crossbeam drive assembly drives the slide assembly, the laser cutting head 400, and the line laser vision device 100 to reciprocate synchronously along the X-axis. It can be understood that the laser cutting head 400 and the line laser vision device 100 maintain a fixed relative position in the machine tool coordinate system, and the X-axis offset and Y-axis offset between the laser cutting head 400 and the line laser vision device 100 can be obtained by conversion δ obtained through calibration.

[0083] Furthermore, a worktable 500 is set above the machine bed, which is used to hold the cutting scrap 300 to be identified and the single circular calibration part 200 for calibration. The aforementioned control and processing unit can be equipped with a coordinate calibration calculation module, a scanning boundary planning module, a path generation module and a point cloud processing module. It can sequentially complete the coordinate system transformation relationship δ solution, speed change distance D measurement, maximum scanning area 600a boundary calculation, scanning start and end coordinate limit correction, N-shaped or S-shaped line-by-line scanning path planning, as well as Z-axis height screening of the acquired point cloud data, interference point cloud removal, scrap 300 contour extraction, coordinate mapping and contour fitting reconstruction, and finally output the scrap 300 contour data adapted to the machine tool coordinate system. It can be directly connected to the subsequent nesting and cutting processes to realize the automated connection between scrap 300 identification and cutting operations.

[0084] This application also proposes an electronic device that can be integrated into the CNC system or host computer of a laser cutting device, including one or more processors and a memory electrically connected to the processors. The memory is used to store computer program instructions that can be called and executed by the processor. When the processor calls and runs the computer program stored in the memory, it can implement all the steps of the laser cutting waste 300 identification method described in the above embodiments, and complete the entire process of line laser vision calibration, scanning boundary calculation, scanning path planning, point cloud data processing, and waste 300 contour reconstruction.

[0085] The processor can be a central processing unit (CPU), a digital signal processor, or other processing units with data processing and instruction execution capabilities. In addition to executing program instructions related to the identification of the residual material 300, it can also generate corresponding motion control instructions based on the calculation results to regulate the other functional components in the electronic device to operate in coordination according to preset logic.

[0086] The memory may include one or more forms of computer-readable storage media, specifically covering two categories: volatile memory and non-volatile memory. Volatile memory may include random access memory (RAM), cache, etc., for temporarily storing point cloud acquisition data, intermediate calculation results, and real-time running instructions to improve data retrieval and processing efficiency. Non-volatile memory may include read-only memory (ROM), solid-state storage units, hard disks, etc., for persistently storing the program code, calibration parameter files, historical scrap 300 contour data, and equipment configuration parameters of the laser cutting scrap 300 identification method.

[0087] In one embodiment, the electronic device may further include an input device and an output device. These components are interconnected and interact with each other via a bus system or other adapted connection architecture. The input device may include an external interactive terminal (keyboard, mouse, etc.), a data acquisition interface, and a command input peripheral. The input device is used to receive raw point cloud data transmitted by the line laser vision device 100, manually entered parameters of the scrap material 300, and scanning control commands. The output device may include a display terminal (such as a monitor, printer, etc.), a data interaction interface, and a communication transmission module. The output device is used to output the reconstructed outline file of the scrap material 300, scanning operation status information, and recognition result parameters. It can also synchronize the recognition results to a remote nesting system or production management platform through a communication network to achieve automated data connection between the scrap material 300 recognition process and the subsequent cutting and nesting process.

[0088] In addition to the methods and apparatus described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0089] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0090] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0091] The computer-readable storage medium may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. Readable storage media may include, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, and portable compact disk read-only memory (CD). ROM, optical storage device, magnetic storage device, or any suitable combination thereof.

[0092] The above description is merely an optional embodiment of this application and does not limit the patent scope of this application. Any equivalent structural transformations made based on the inventive concept of this application and the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included within the patent protection scope of this application.

Claims

1. A method for identifying laser-cut waste material, characterized in that, Includes the following steps: The line laser vision equipment is visually calibrated using calibration components to obtain the transformation relationship δ between the coordinate system of the line laser vision equipment and the machine tool coordinate system, and to determine the speed change distance D. Set the scanning line segment length L0 of the line laser vision device, and determine the maximum scanning area according to the conversion relationship δ, the speed change distance D and the scanning line segment length L0; The line laser vision device is controlled to scan the remaining material in the maximum scanning area, acquire the point cloud data of the remaining material, and process the point cloud data to generate a contour image file of the remaining material in the machine tool coordinate system.

2. The laser cutting waste identification method as described in claim 1, characterized in that, The specific steps for determining the maximum scanning area based on the conversion relationship δ, the speed change distance D, and the scanning line segment length L0 include: Using the positive limit of the X-axis of the machine tool coordinate system as a reference, the speed change distance D is offset in the negative X-axis direction, and using the negative limit of the X-axis of the machine tool coordinate system as a reference, the speed change distance D is offset in the positive X-axis direction to obtain the initial positive side boundary and the initial negative side boundary in the X-axis direction. Based on the orientation of the line laser vision device along the X-axis of the machine tool coordinate system, the X-axis offset obtained by the transformation relationship δ is superimposed onto the offset of the initial positive side boundary or the initial negative side boundary to determine the X-axis boundary of the maximum scanning area.

3. The laser cutting waste identification method as described in claim 2, characterized in that, The specific steps for determining the maximum scanning area based on the conversion relationship δ, the speed change distance D, and the scanning line segment length L0 further include: Using the positive and negative Y-axis limits of the machine tool coordinate system as a reference, and based on the orientation of the line laser vision device along the Y-axis of the machine tool coordinate system, the Y-axis offset obtained by the transformation relationship δ and the offset of half the length L0 of the scanning line segment are superimposed on the offsets of the positive and negative Y-axis limits to determine the Y-axis boundary of the maximum scanning area.

4. The laser cutting waste identification method as described in claim 1, 2, or 3, characterized in that, The steps for determining the speed change distance D include: The coordinates of the start and end positions of the line laser vision device in the machine tool coordinate system are obtained when it accelerates or decelerates between a stationary state and a uniform scanning speed. The speed change distance D is calculated based on the coordinate difference between the start and end positions.

5. The laser cutting waste identification method as described in claim 1, 2, or 3, characterized in that, Before controlling the line laser vision device to scan the remaining material, a scanning start and end position correction step is also included: Using the negative X-axis boundary of the maximum scanning area as a reference, and combining the speed change distance D, perform boundary limiting processing and offset calculation on the X-axis coordinate of the preset scanning starting point. Using the negative Y-axis boundary of the maximum scanning area as a reference, perform boundary limiting processing on the Y-axis coordinate of the preset scanning starting point to obtain the corrected scanning starting point coordinates. Based on the corrected scanning start coordinates, the scanning size parameters corresponding to the axial direction of the preset scanning area are superimposed. Then, boundary limiting processing is performed with the positive side boundary of the X-axis and the positive side boundary of the Y-axis corresponding to the maximum scanning area as the reference to obtain the corrected scanning end coordinates.

6. The laser cutting waste identification method as described in claim 5, characterized in that, Before controlling the line laser vision device to scan the remaining material, a scanning width correction step is also included: The total scan width W is calculated based on the corrected scan start coordinates and scan end coordinates. 扫 The number of scans n is calculated by combining the length L0 of the scan line segment and rounded up. Then, based on the number of scans n and the total scan width W, the result is calculated. 扫 Determine the width of a single scan.

7. The laser cutting waste identification method as described in claim 6, characterized in that, The step of visually calibrating the line laser vision device using the calibration component further includes: Using the upper surface of the calibration component as the calibration reference surface, and combining its thickness N, the Z-axis zero-position height reference calibration is completed. The vertical distance parameter H from the line laser vision device to the calibration reference surface is determined, and the Z-axis coordinate value of the worktable surface in the machine tool coordinate system is derived to be -N.

8. The laser cutting waste identification method as described in claim 7, characterized in that, After controlling the line laser vision device to scan and acquire point cloud data of the waste material, the steps of processing the point cloud data include: Based on the Z-axis zero-position height reference and the thickness parameter of the remaining material, the effective point cloud height threshold range is determined; The point cloud data is filtered in the Z direction according to the height threshold range to remove interfering point clouds outside the height threshold range, thereby obtaining the valid point cloud data corresponding to the remaining material.

9. The laser cutting waste identification method as described in claim 6, characterized in that, The steps of controlling the line laser vision device to scan the scrap material include: Plan a line-by-line scanning path based on the corrected scan start coordinates, scan end coordinates, scan count n, and single scan width; The line laser vision device is driven to move along the scanning path of each row, and the point cloud data of the remaining material is collected within the uniform speed movement range.

10. A laser cutting device, characterized in that, include: Laser cutting head; A line laser vision device is located on one side of the laser cutting head; as well as A control processing unit capable of performing the laser cutting waste identification method as described in any one of claims 1 to 9, wherein the control processing unit is electrically connected to the laser cutting head and the line laser vision device to control the laser cutting head to drive the line laser vision device to move along the machine tool coordinate system.

11. An electronic device, characterized in that, include: A processor is used to store computer programs; as well as The memory, electrically connected to the processor, is used to execute the program stored in the memory to implement the steps of the laser cutting waste identification method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the laser cutting waste identification method according to any one of claims 1 to 9.

Images