Method for correcting object scan data, method for controlling a gripping system and gripping system

Abstract

The application provides a kind of object scanning data correction method, grabbing system control method, grabbing system and computer readable storage medium, wherein the object scanning data correction method includes: obtaining the scanning data of a feature surface in an object, the object has an ideal shape, under the ideal shape, the scanning shape of the feature surface is a closed figure, and there is no acute angle in the internal angle of the closed figure;The scanning data is converted into a plan view by vertical projection;Obtain the inscribed figure of the feature surface plan view of the object, the inscribed figure is similar to the shape of the feature surface in the ideal shape of the object;Calculate to obtain a correction model, the correction model is similar to the ideal shape of the object.The embodiments of the application utilize the inscribed figure similar to the ideal shape of the object to calculate and obtain the correction model of the object, and the correction model is used for subsequent processing process.The correction model can overcome the problem of inaccurate positioning and shape deformation edge corner loss, and improve the accuracy of object scanning data.

Description

Technical Field

[0001] This invention relates to the field of automated production technology, and in particular to a method for correcting object scanning data, a method for controlling a gripping system, and a gripping system. Background Technology

[0002] In industrial production and material handling, automated equipment is often used to grab workpieces, materials, or baskets containing materials. However, in practical applications, the shape of the workpiece, material, or basket may change, such as deformation or missing corners due to long-term use or collisions. The shape and pose of the object obtained by the scanning device may differ from the theoretical shape and pose of the object. In such cases, grabbing the object with the gripping device may result in inaccurate gripping, dropping, or tipping, affecting production safety and efficiency.

[0003] The content of the background section is merely the technology known to the inventor and does not necessarily represent the prior art in this field. Summary of the Invention

[0004] To address one or more deficiencies in the prior art, the present invention provides a method for correcting object scanning data, comprising:

[0005] The scanning data of a feature surface in an object is obtained. The object has an ideal shape. Under the ideal shape, the scanning shape of the feature surface is a closed figure, and the interior angle of the closed figure does not have a right angle.

[0006] The scanned data is converted into a planar image through vertical projection;

[0007] Obtain an inscribed graph of a planar image of a feature surface of an object, wherein the inscribed graph approximates the shape of the feature surface in the ideal shape of the object;

[0008] A correction model is calculated, which approximates the ideal shape of the object.

[0009] According to one aspect of the invention, the step of obtaining the inscribed graph of the surface planar image of the object feature further includes:

[0010] Obtain the smallest circumcircle of the planar image;

[0011] The inscribed figure is obtained by combining the planar image and the smallest circumcircle.

[0012] According to one aspect of the invention, the step of obtaining the inscribed graph of the surface planar image of the object feature further includes:

[0013] Obtain the bounding box of the minimum circumcircle;

[0014] The region of interest is obtained based on the bounding box, and the region of interest includes the entire area within the smallest circumcircle.

[0015] The inscribed graphic is obtained based on the blank range of the region of interest in the planar image.

[0016] According to one aspect of the invention, the step of obtaining the inscribed pattern includes: inverting the color of the region of interest in the planar image and obtaining the inscribed pattern within the blank area of ​​the region of interest, wherein the planar image is a point cloud image.

[0017] According to one aspect of the invention, wherein the inscribed graph is a maximum inscribed graph, and the step of calculating the correction model includes:

[0018] The maximum inscribed shape is used as the shape of the feature surface in the correction model, and the correction model is calculated based on the maximum inscribed shape. The correction model includes the geometry and pose of the object.

[0019] According to one aspect of the invention, the step of calculating the correction model includes:

[0020] Calculate the coordinates of the vertices of the inscribed figure within the plane of the planar image;

[0021] The coordinates of the vertices of the inscribed graphic in the plane of the planar image are mapped to three-dimensional space to obtain three-dimensional coordinates;

[0022] A correction model is calculated based on the three-dimensional coordinates of the vertices of the inscribed graph and the ideal shape of the object. The correction model includes the geometry and pose of the object.

[0023] According to one aspect of the present invention, the scan data of the object feature surface is point cloud data, and the step of converting the scan data into a planar image includes: transforming the point cloud data to the object coordinate system and obtaining a planar image by orthogonal projection;

[0024] The step of obtaining three-dimensional coordinates includes: mapping the coordinates of the vertices of the inscribed graphic in the plane of the planar image to three-dimensional space, and transforming them to the coordinate system of the sensor that acquires the point cloud data to obtain the three-dimensional coordinates.

[0025] According to one aspect of the invention, the object is a basket with a rectangular opening at the top, and the ideal shape is approximately an inverted frustum or cuboid; the characteristic surface of the object is the upper surface of the basket; and the inscribed shape is the largest inscribed rectangle.

[0026] According to one aspect of the invention, the ideal shape of the object includes the height and outer wall thickness of the basket, and the step of calculating the corrected model includes: calculating the geometry and pose of the basket based on the vertex coordinates of the largest inscribed rectangle and the height and thickness of the basket.

[0027] According to one aspect of the present invention, the present invention further includes a grasping system control method, the grasping system control method comprising:

[0028] Based on the object scanning data correction method described above, a correction model of the object to be grasped is obtained.

[0029] The gripping device is controlled to grip the object to be gripped according to the calibration model.

[0030] According to one aspect of the invention, the invention also includes a grasping system, the grasping system comprising:

[0031] A gripping device, wherein the gripping mechanism is configured to be driven to grip an object to be gripped;

[0032] An information acquisition module, configured to acquire scanning data of the object to be grasped or a preset surface of the object to be grasped; and

[0033] The control system is signal-connected to the grasping device and the information acquisition module, and is capable of executing the object scanning data correction method as described above.

[0034] According to one aspect of the invention, the invention also includes a computer-readable storage medium including computer-executable commands stored thereon, which, when executed by a processor, implement the object scanning data correction method and / or the grasping system control method as described above.

[0035] Compared with existing technologies, embodiments of the present invention provide a method for correcting object scanning data. After acquiring the object's scanning data, it is converted into a planar image, and then the inscribed graph of the planar image is obtained. Using the inscribed graph that approximates the ideal shape of the object, a corrected model of the object is calculated. The corrected model is used in subsequent processing, such as grasping. After obtaining the corrected model of the object, the problems of inaccurate positioning and shape deformation can be overcome, and missing edges and corners can be filled to a certain extent, improving the accuracy of the object scanning data. The present invention also includes a grasping system control method and an embodiment of the grasping system, which uses the aforementioned object scanning data correction method to obtain accurate object scanning data, improve grasping accuracy, and ensure production safety and efficiency. The present invention also includes a computer-readable storage medium capable of executing the aforementioned object scanning data correction method and / or grasping system control method. Attached Figure Description

[0036] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0037] Figure 1 This is a flowchart illustrating a method for correcting object scanning data in one embodiment of the present invention;

[0038] Figure 2 This is a flowchart illustrating a correction method comprising the process of acquiring an inscribed graphic of a feature surface planar image, according to one embodiment of the present invention.

[0039] Figure 3 This is a flowchart illustrating a correction method including the process of calculating a correction model in one embodiment of the present invention;

[0040] Figure 4 This is a schematic diagram of obtaining the smallest circumcircle of a planar image in one embodiment of the present invention;

[0041] Figure 5 This is a schematic diagram illustrating the inversion of the color of the region of interest in one embodiment of the present invention;

[0042] Figure 6 This is a schematic diagram of obtaining the inscribed pattern of a feature surface planar image in one embodiment of the present invention;

[0043] Figure 7 This is a schematic diagram of obtaining the correction model in one embodiment of the present invention;

[0044] Figure 8 This is a flowchart illustrating the capture system control method in one embodiment of the present invention;

[0045] Figure 9 This is a structural block diagram of a grasping system in one embodiment of the present invention. Detailed Implementation

[0046] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0047] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0048] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, an electrical connection, or a connection that allows for communication; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0049] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0050] The following disclosure provides many different embodiments or examples for implementing various structures of the invention. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the invention. Furthermore, reference numerals and / or letters may be repeated in different examples; such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. In addition, examples of various specific processes and materials are provided in this invention, but those skilled in the art will recognize the application of other processes and / or the use of other materials.

[0051] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0052] Figure 1 The following is a detailed flowchart of a method 100 for correcting object scanning data according to a preferred embodiment of the present invention. The object scanning data can be obtained using devices such as industrial cameras; preferably, the object scanning data is point cloud data. In this embodiment, the object has an ideal shape, that is, the shape of the object without any interference. For example, the object is a cuboid or other shaped workpiece, or the object is a cuboid, frustum, or cylindrical basket. The ideal shape of the object can be obtained through the previous process or factory parameters. Taking a cuboid-shaped basket as an example, the ideal shape includes the known cuboid shape and the specific values ​​of its length, width, and height.

[0053] In step S101, scanning data of a feature surface in the object is acquired. For example, scanning data of a feature surface in the object is acquired using an industrial camera or LiDAR. Under the ideal shape of the object, the feature surface has a clearly defined closed shape. Depending on the selected feature surface and the ideal shape of the object, the closed shape of the feature surface is different. Taking a slightly open basket as an example, the top surface of the opening can be a rectangle, a circle, or other specific shapes. In this embodiment, it is limited that the closed shape of the feature surface of the object does not have a right angle, that is, the shape of the feature surface under the ideal shape of the object is a convex polygon or an outwardly convex arc shape, without containing concave interior angles. In actual production, workpieces or baskets with concave shapes are rarely used and are not considered in this embodiment.

[0054] In step S102, the scanned data is converted into a planar image through vertical projection. The acquired scanned data includes normal information. In this step, the scanned data is converted into a planar image through vertical projection. Furthermore, the projection plane is a plane perpendicular to the optical axis of the scanning device or a plane perpendicular to the axial direction of the object. Taking a horizontally placed basket as an example, in order to provide a basis for grasping, the open side is used as the feature surface in this embodiment. The scanning device scans the feature surface and projects it vertically on the horizontal plane to obtain a planar image of the feature surface, which facilitates the subsequent processing of the planar image.

[0055] In step S103, the inscribed graph of the planar image of the object's feature surface is obtained, and the inscribed graph is similar to the shape of the feature surface under the ideal shape of the object. For example, if the object is an open basket, the feature surface is the open surface of the basket. Under the ideal shape of the basket, the feature surface is, for example, a rectangular open surface. In this step, the inscribed rectangle of the planar image of the basket's feature surface is obtained. Furthermore, according to a preferred embodiment of the present invention, the aspect ratio of the inscribed rectangle is the same as or close to the aspect ratio of the open rectangle under the ideal shape of the basket, so as to obtain an inscribed rectangle close to the edge of the planar image. Preferably, the inscribed rectangle is the largest inscribed rectangle of the planar image of the feature surface.

[0056] In step S104, a correction model is calculated, wherein the correction model is approximately the ideal shape of the object. In specific applications, the ideal shape of the object is a three-dimensional shape. In this embodiment, the correction shape of the feature surface is obtained by using the shape of the inscribed graphic. By combining the correction shape of the feature surface with the ideal shape of the object, a three-dimensional correction model of the object is obtained. Taking an open basket as an example, after obtaining the inscribed graphic of the planar image of the feature surface, other known data of the ideal shape of the object, such as the opening angle and height, are substituted into the inscribed graphic to obtain the correction model of the object. The correction model is approximately the ideal model of the object and is used for subsequent processing, such as grabbing the object according to the correction model.

[0057] This embodiment obtains scan data by scanning feature surfaces and obtains a correction model through inscribed graph calculation. This can fill in any missing point cloud areas that may occur during object scanning and can correct deformed objects. In a specific embodiment of the invention, to ensure the normal use of the workpiece or basket, the degree of deformation or missing parts is relatively minor, avoiding situations where inscribed graphs cannot be used or where inscribed graph deviations are large. The correction method provided in this embodiment can obtain a relatively accurate object correction model, providing a foundation for grasping. Furthermore, according to a preferred embodiment of the invention, the correction model includes the object's geometry and pose. The geometry approximates the object's ideal shape, and the object's pose reflects the object's current position and orientation through the scan data.

[0058] Figure 2The following is a detailed flowchart of a method 200 for correcting object scan data according to a preferred embodiment of the present invention, which specifically includes the process of obtaining an inscribed pattern of a planar image of an object feature surface. Figure 4 and Figure 5 The diagram below illustrates the process of obtaining the inscribed graph. Figure 2 , Figure 4 and Figure 5 Please elaborate.

[0059] In correction method 200, steps S201, S202, and S207 are basically the same as steps S101, S102, and S104 in the aforementioned correction method 100, and will not be described again here. This embodiment elaborates on the process of obtaining the inscribed graphic of the planar image of the surface feature of an object.

[0060] like Figure 2 As shown, in step S203, the minimum circumcircle of the planar image is obtained. In practical applications, there are cases where scan data is missing. For example, the object may have defects compared to its ideal shape, or some scan data may be missing due to reflections on the object's feature surface or scanning errors. When scan data is missing, a closed shape cannot be formed, or the planar image obtained from the scan data may have a large error compared to the object's ideal shape, affecting the accuracy of the inscribed shape. Therefore, in the preferred embodiment of the present invention, the minimum circumcircle of the planar image is constructed to supplement the missing scan data, thereby obtaining a more accurate inscribed shape. For example, if the feature surface of an object in its ideal shape is rectangular, some scan data may be missing when obtaining the feature surface scan data. For example, the rectangle may lack scan data for one vertex. By constructing the minimum circumcircle from the planar image obtained from the remaining scan data, the missing vertex can be filled in. Of course, if there are many missing scan data, the correction method in this embodiment cannot be used to obtain the inscribed shape.

[0061] like Figure 4 As shown, taking a rectangular-opening basket as an example, the rounded rectangle in the middle is a planar image converted from the scan data of the feature surface, and the outer circle represents the smallest circumcircle of the planar image. For objects in practical applications, the feature surfaces under their ideal shape are generally regular shapes. When there are partial missing parts in the actual scan data, after obtaining the smallest circumcircle, a more accurate inscribed figure can be obtained by combining the planar image of the scan data with the smallest circumcircle. This is suitable for situations with large deformations or missing point clouds.

[0062] Furthermore, according to a preferred embodiment of the present invention, a region of interest can be selected to improve the accuracy of obtaining the inscribed pattern. In step S204, the bounding box of the smallest circumcircle is obtained; in step S205, the region of interest is obtained based on the bounding box. Specifically, the region of interest includes the entire area within the smallest circumcircle, for example... Figure 5 As shown. The bounding box can be obtained using an algorithm to solve for the optimal bounding space of a discrete point set. The basic idea is to approximate a complex geometric object with a slightly larger volume and simpler properties (called the bounding box), for example, a representation of... Figure 5 The positive direction range outside the minimum circumcircle shown can improve computational efficiency and reduce the impact on the inscribed figure when the actual scan data deviates significantly from the closed shape of the feature surface under the ideal shape of the object, thereby further improving the accuracy of the inscribed figure and optimizing the correction model.

[0063] The region of interest (ROI) can also reduce the impact of scan data obtained outside the ROI on the inscribed graph. When scanning an object, not only the scan data of the object's feature surface is obtained, but also the scan data of the surrounding environment. Preliminary filtering can be performed based on information such as depth and normal. After obtaining the ROI, scan data outside the ROI can be further excluded, improving the accuracy of obtaining the inscribed graph. The ROI includes the entire area within the smallest circumcircle. Preferably, the ROI is a square area in the planar image, and the sides of the square are tangent to the smallest circumcircle, further narrowing the ROI and reducing the computational difficulty.

[0064] In step S206, an inscribed pattern is obtained based on the blank area of ​​the region of interest in the planar image. Specifically, according to a preferred embodiment of the present invention, the color of the region of interest in the planar image is inverted, and the inscribed pattern within the blank area of ​​the region of interest is obtained. In this embodiment, the planar image is a point cloud image. In practical applications, the planar image obtained from the scanning data is as follows: Figure 4 As shown, the scan data of the feature surface and the minimum circumcircle are displayed in a color different from the background color. For example, if the background color is black, the scan data and the minimum circumcircle are displayed in white when obtaining a planar image for distinction. Of course, other colors can also be selected when obtaining a planar image, and there are no restrictions here.

[0065] After obtaining the region of interest, the colors of the region of interest in the planar image are inverted, such as... Figure 5 As shown, changing the background color of the region of interest (within the square frame) to white, and displaying the scan data and the smallest circumcircle in black, then... Figure 5 The white area represents the valid region, and the black point cloud data constitutes a connected image. The inscribed graph can then be obtained from the connected image within the white valid region. A connected image also reduces the impact of discrete noise on the inscribed graph, improving its accuracy. A planar image is a point cloud image, composed of multiple points obtained from scanning. When scanning the feature surfaces of an object, these surfaces need to be represented by multiple points. Figure 4 and Figure 5 Even with minor point cloud gaps, the shape of a connected region can be supplemented using the minimum circumcircle. The connected shape represents the actual scan data of the object's feature surface, allowing for clearer boundary ranges through color inversion, while further reducing interference from scan data outside the region of interest on blank areas.

[0066] Figure 3 The following is a detailed flowchart of the correction method 300 according to a preferred embodiment of the present invention, which includes the process of calculating and obtaining the correction model. In this embodiment, steps S301, S302 and S303 in the correction method 300 are basically the same as steps S101, S102 and S103 in the correction method 100 in the foregoing embodiment. Step S303 in this embodiment, which obtains the inscribed graph of the surface planar image of the object feature, can also be performed according to the steps in the correction method 200, and will not be described again.

[0067] like Figure 3 As shown, in step S304, the coordinates of the vertices of the inscribed graphic in the plane of the planar image are calculated, specifically for example... Figure 6 As shown, the white connected shape represents the planar image obtained from the scan data, and the gray boundary inside forms an inscribed figure. In this embodiment, the object feature surface is rectangular, and the inscribed figure includes four vertices. After determining the position of the inscribed figure, the coordinates of the four vertices in the inscribed figure within the plane of the planar image can be calculated. This step is performed within the plane of the planar image, and the obtained vertex coordinates are also planar coordinates.

[0068] In step S305, the coordinates of the vertices of the inscribed graphic in the plane of the planar image are mapped to three-dimensional space to obtain three-dimensional coordinates. Since the object has a three-dimensional shape in actual space, obtaining only the planar coordinates of the vertices of the inscribed graphic cannot truly reflect the geometric shape and pose of the object. Therefore, in this embodiment, the three-dimensional coordinates of the vertices in the inscribed graphic are obtained through mapping. Specifically, for example, the planar coordinates of the vertices of the inscribed graphic are assigned a normal distance and converted to the coordinate system of the scanning device.

[0069] Without being subject to S306, a correction model is calculated based on the 3D coordinates of the vertices of the inscribed graphic and the ideal shape of the object. Specifically, the 3D coordinates of the vertices of the inscribed graphic are used to represent the 3D coordinates of the vertices of the object's feature surfaces. Then, the known geometric constraints in the ideal shape of the object are substituted to obtain the geometric shape representing the object's scan data. For example... Figure 7As shown, the three-dimensional coordinates of the vertices of the object's feature surface can also reflect the object's pose under the scanning device. According to a preferred embodiment of the present invention, the scanning data of the object's feature surface is point cloud data. By transforming the point cloud data to the object coordinate system and orthogonally projecting it, a planar image of the aforementioned object feature surface can be obtained. After obtaining the inscribed graphic, the three-dimensional coordinates of the vertices of the inscribed graphic can be obtained through the planar coordinates of the vertices of the inscribed graphic in the plane of the planar image. Specifically, the planar coordinates of the vertices of the inscribed graphic are mapped to three-dimensional space and transformed to the coordinate system of the sensor that acquires the point cloud data to obtain the three-dimensional coordinates of the vertices of the inscribed graphic.

[0070] The geometric constraints of the ideal shape of the object include, for example, the dimensional and angular relationships of the geometric shape. In this embodiment, such as... Figure 7 As shown, the object has a rectangular opening at the top, and its ideal shape is approximately an inverted frustum or a cuboid with vertical sides. The upper surface of the basket (the open side) is selected as the feature surface. After obtaining the three-dimensional coordinates of the inscribed figure of the feature surface, the corrected model of the object is obtained based on the length and tilt angle of the basket's side edges in the ideal shape. In the gripping operation, only the shape correction of the open side needs to be considered. The possible deformation of the sides and bottom of the basket does not affect the gripping process. Therefore, in this embodiment, it is directly calculated based on the geometric constraints of the ideal shape without additional scanning and correction.

[0071] Furthermore, according to a preferred embodiment of the present invention, the geometric constraints of the ideal shape of the object also include the outer wall thickness of the basket. Based on the vertex coordinates of the largest inscribed rectangle and combined with the geometric constraints of the ideal shape of the basket, a correction model of the basket, i.e., the geometry and position of the basket, can be calculated. For cases where other surfaces besides the feature surface also require correction, the aforementioned correction method can be used to correct these other surfaces.

[0072] The present invention also includes an embodiment of a grasping system control method, such as... Figure 8 As shown, in the specific flow of the grasping system control method 400, in step S401, a corrected model of the object to be grasped is obtained according to the object scanning data correction method in the aforementioned embodiment. During the grasping process, for example, the object is transported to the bottom of the grasping system by a conveyor belt. A sensor capable of acquiring scanning data obtains scanning data of the side of the object facing the sensor. This side of the object facing the sensor is the feature surface. After obtaining the scanning data of the feature surface, the corrected model of the object can be obtained according to the aforementioned object scanning data correction method. In step S402, the grasping device is controlled to grasp the object to be grasped according to the corrected model. In this embodiment, the grasping device communicates with the sensor or the processing control system. After obtaining the corrected model of the object, the grasping device can be controlled to grasp the object. This embodiment can improve the accuracy of grasping and reduce the risk of grasping failure.

[0073] like Figure 9 As shown, the present invention also includes an embodiment of a gripping system 1, which includes a gripping device 10, an information acquisition module 20, and a control system 30. The gripping device 10 can be driven to grip an object to be gripped, such as a gripper type, an internal support type, or an adsorption type structure, which is not limited in this embodiment.

[0074] The information acquisition module 20 can acquire scanning data of the object to be grasped or a preset surface of the object to be grasped, such as an industrial camera, depth sensor, lidar, etc., or multiple sensors can work together to acquire scanning data of the object to be grasped.

[0075] The control system 30 is connected to the grasping device 10 and the information acquisition module 20 by signal, and can execute the object scanning data correction method in the aforementioned embodiment. Based on the scanning data of the object to be grasped obtained by the information acquisition module 20, a correction model of the object to be grasped is obtained, and the grasping device 10 is controlled to perform a grasping action based on the correction model of the object to be grasped.

[0076] The present invention also includes an embodiment of a computer-readable storage medium comprising computer-executable commands stored thereon, which, when executed by a processor, implement the object scanning data correction method as described in the foregoing embodiments and / or the grasping system control method as described in the foregoing embodiments.

[0077] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for correcting object scan data, comprising: The scanning data of a feature surface in an object is obtained. The object has an ideal shape. Under the ideal shape, the scanning shape of the feature surface is a closed figure, and the interior angle of the closed figure does not have a right angle. The scanned data is converted into a planar image through vertical projection; Obtain an inscribed graph of a planar image of a feature surface of an object, wherein the inscribed graph approximates the shape of the feature surface in the ideal shape of the object; A correction model is calculated, which approximates the ideal shape of the object. In this step, the correction shape of the feature surface is obtained using the inscribed graph, and the correction model is obtained by combining the correction shape of the feature surface with the ideal shape. The step of calculating and obtaining the correction model includes: Calculate the coordinates of the vertices of the inscribed figure within the plane of the planar image; The coordinates of the vertices of the inscribed graphic in the plane of the planar image are mapped to three-dimensional space to obtain three-dimensional coordinates; A correction model is calculated based on the three-dimensional coordinates of the vertices of the inscribed graph and the ideal shape of the object. The correction model includes the geometry and pose of the object.

2. The correction method according to claim 1, wherein the step of obtaining the inscribed pattern of the surface planar image of the object feature further comprises: Obtain the smallest circumcircle of the planar image; The inscribed figure is obtained by combining the planar image and the smallest circumcircle.

3. The correction method according to claim 2, wherein the step of obtaining the inscribed graph of the surface planar image of the object feature further includes: Obtain the bounding box of the minimum circumcircle; The region of interest is obtained based on the bounding box, and the region of interest includes the entire area within the smallest circumcircle. The inscribed graphic is obtained based on the blank range of the region of interest in the planar image.

4. The correction method according to claim 3, wherein the step of obtaining the inscribed pattern includes: The color of the region of interest in the planar image is inverted, and the inscribed shape within the blank area of ​​the region of interest is obtained. The planar image is a point cloud image.

5. The correction method according to any one of claims 1-4, wherein the inscribed graph is the maximum inscribed graph, and the step of calculating the correction model includes: The maximum inscribed shape is used as the shape of the feature surface in the correction model, and the correction model is calculated based on the maximum inscribed shape. The correction model includes the geometry and pose of the object.

6. The correction method according to claim 1, wherein the scan data of the object feature surface is point cloud data, and the step of converting the scan data into a planar image through vertical projection includes: The point cloud data is transformed to the object coordinate system, and a planar image is obtained through vertical projection. The step of obtaining three-dimensional coordinates includes: mapping the coordinates of the vertices of the inscribed graphic in the plane of the planar image to three-dimensional space, and transforming them to the coordinate system of the sensor that acquires the point cloud data to obtain the three-dimensional coordinates.

7. The correction method according to any one of claims 1-4, wherein the object is a basket with a rectangular opening at the top, and the ideal shape is approximately an inverted frustum or cuboid; the characteristic surface of the object is the upper surface of the basket; and the inscribed shape is the largest inscribed rectangle.

8. The correction method according to claim 7, wherein the ideal shape of the object includes the height of the basket and the thickness of its outer wall, and the step of calculating the correction model includes: Calculate the geometry and pose of the basket based on the vertex coordinates of the largest inscribed rectangle and the height and thickness of the basket.

9. A method for controlling a capture system, comprising: According to the object scanning data correction method as described in any one of claims 1-8, a correction model of the object to be grasped is obtained; The gripping device is controlled to grip the object to be gripped according to the calibration model.

10. A crawling system, comprising: A gripping device, wherein the gripping device is capable of being driven to grip an object to be gripped; An information acquisition module, configured to acquire scanning data of the object to be grasped or a preset surface of the object to be grasped; and A control system is connected to the grasping device and the information acquisition module by signal, and is capable of executing the object scanning data correction method as described in any one of claims 1-8.

11. A computer-readable storage medium comprising computer-executable commands stored thereon, the executable commands, when executed by a processor, implementing the object scan data correction method as claimed in any one of claims 1-8 and / or the grasping system control method as claimed in claim 9.

Images