CN114013751A - Rectangular article boxing method, device, electronic device and storage medium - Google Patents
Rectangular article boxing method, device, electronic device and storage medium Download PDFInfo
- Publication number
- CN114013751A CN114013751A CN202210004025.3A CN202210004025A CN114013751A CN 114013751 A CN114013751 A CN 114013751A CN 202210004025 A CN202210004025 A CN 202210004025A CN 114013751 A CN114013751 A CN 114013751A
- Authority
- CN
- China
- Prior art keywords
- article
- boxing
- target
- height
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65B—MACHINES, APPARATUS OR DEVICES FOR, OR METHODS OF, PACKAGING ARTICLES OR MATERIALS; UNPACKING
- B65B57/00—Automatic control, checking, warning, or safety devices
- B65B57/10—Automatic control, checking, warning, or safety devices responsive to absence, presence, abnormal feed, or misplacement of articles or materials to be packaged
- B65B57/14—Automatic control, checking, warning, or safety devices responsive to absence, presence, abnormal feed, or misplacement of articles or materials to be packaged and operating to control, or stop, the feed of articles or material to be packaged
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65B—MACHINES, APPARATUS OR DEVICES FOR, OR METHODS OF, PACKAGING ARTICLES OR MATERIALS; UNPACKING
- B65B5/00—Packaging individual articles in containers or receptacles, e.g. bags, sacks, boxes, cartons, cans, jars
- B65B5/10—Filling containers or receptacles progressively or in stages by introducing successive articles, or layers of articles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65B—MACHINES, APPARATUS OR DEVICES FOR, OR METHODS OF, PACKAGING ARTICLES OR MATERIALS; UNPACKING
- B65B57/00—Automatic control, checking, warning, or safety devices
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
- Image Analysis (AREA)
Abstract
The application belongs to the field of robot control and discloses a rectangular article boxing method, a rectangular article boxing device, electronic equipment and a storage medium, wherein the size and top surface pose information of each article to be boxed are acquired through a first RGBD (red, green and blue) camera; the article to be boxed is a cuboid article; dimensions include length, width and height; generating a mask matrix; the size of the mask matrix is the same as the pixel size of the boxing area of the target box body in the image acquired by the second RGBD camera; sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of a boxing area with the height of the target article to update a mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; therefore, the target object can be accurately placed in the position capable of containing the target object, automatic boxing of the objects to be boxed is realized, and the method has better practicability and universality.
Description
Technical Field
The application relates to the field of robot control, in particular to a rectangular object boxing method, a rectangular object boxing device, electronic equipment and a storage medium.
Background
At present, some devices (such as electronic devices) are packaged by using a plurality of small rectangular parallelepiped boxes to package each part of the device, and then all the small rectangular parallelepiped boxes of the same device are packed into a large rectangular parallelepiped box. In the traditional method, a cuboid small box is manually packed into a cuboid large box, but the working efficiency is low. Therefore, some factories use robots to complete the boxing operation of loading the small rectangular box into the large rectangular box, but most of the current algorithms for picking and placing by the robots are not perfect and accurate, and the practicability and the universality are poor.
Disclosure of Invention
The present application aims to provide a rectangular parallelepiped article packing method, device, electronic apparatus, and storage medium, which are highly practical and versatile.
In a first aspect, the application provides a rectangular parallelepiped article boxing method, which is applied to a boxing robot, wherein the boxing robot comprises a first RGBD camera and a second RGBD camera, the first RGBD camera is arranged at the tail end of the boxing robot, the second RGBD camera is arranged right above a target box body and is used for acquiring an image of the target box body, and the target box body is a rectangular parallelepiped box body; the method comprises the following steps:
A1. acquiring the size and top surface pose information of each article to be boxed through the first RGBD camera; the article to be boxed is a cuboid article; the dimensions include length, width and height;
A2. generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the packing area of the target box body in the image collected by the second RGBD camera;
A3. sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of the boxing area with the height of the target article to update the mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item.
According to the cuboid object boxing method, the size and the top face pose information of an object to be boxed are determined through the first RGBD camera, the residual height information of each position in a boxing area of a target box body is obtained through the second RGBD camera, a mask matrix with the same pixel size as the boxing area is utilized to perform mask operation on a depth image of the boxing area to determine a candidate area with the residual height not less than the height of the target object, and finally an effective space capable of containing the target object is searched from the candidate area, so that the target object can be accurately placed at a position capable of containing the target object, automatic boxing of the object to be boxed is achieved, and the method has good practicability and universality.
Preferably, the initial value of each element value of the mask matrix is 1;
step a3 includes: sequentially taking each article to be boxed as a target article and executing the following steps:
A301. acquiring a depth image of the boxing area and residual height information of all parts of the boxing area through the second RGBD camera;
A302. in the mask matrix, changing an element value corresponding to a position where the remaining height of the boxing area is less than the height of the target article to 0;
A303. performing masking operation on the depth image of the boxing region by using the mask matrix to obtain a mask image, and taking a region with a non-zero pixel value in the mask image as the candidate region;
A304. searching for an effective space in the candidate area according to the length and the width of the target item;
A305. and grabbing the target object according to the top surface pose information of the target object and putting the target object into the effective space.
Preferably, before the step a3, the method further comprises the steps of:
calculating the top surface area of each article to be boxed according to the length and the width of each article to be boxed;
sorting each article to be boxed in a descending order according to the area of the top surface;
in step a3, each of the articles to be boxed is sequentially taken as a target article in the order of sorting.
Therefore, the articles with large top surface areas are always positioned below the articles with small top surface areas, and the articles are not easy to overturn after being stacked.
Preferably, step a301 comprises:
acquiring a second image of the target box acquired by the second RGBD camera; the second image comprises a second color image and a second depth image;
segmenting the boxed area of the target box body in the second color image to obtain the boundary of the boxed area;
segmenting the second depth image according to the boundary to obtain a depth image of the boxing area;
extracting real-time depth data of each pixel point from the depth image of the packing area;
and subtracting the real-time depth data from the initial depth data of each pixel point to obtain used height data of each pixel point, and subtracting the used height data of each pixel point from the height of the target box body to obtain residual height information of each pixel point in the boxing region.
Preferably, step a304 comprises:
and searching for an effective space capable of accommodating the window in the candidate area by a sliding window method by adopting the window with the same size as the top surface of the target object.
Preferably, step a304 further comprises:
and if a plurality of searched effective spaces exist, taking the effective space with the largest residual height as the final effective space.
Therefore, the lower layer space can be filled preferentially, the probability that the bottoms of the articles are not supported completely when the articles are stacked is reduced, and the probability that the articles are overturned after being stacked is further reduced; the probability that the lower-layer gap cannot be effectively utilized due to the fact that the upper-layer object covers the lower-layer gap can be reduced, and therefore the space utilization rate of the target box body is improved.
Preferably, step a305 comprises:
acquiring grabbing point position and posture information according to the height and the top surface position and posture information of the target object;
acquiring pose information of the effective space;
acquiring position and posture information of a placement point according to the height of the target object and the posture information of the effective space;
and grabbing the target object from the grabbing point and moving the target object to the placing point for placing according to the grabbing point position information and the placing point position information.
In a second aspect, the application provides a cuboid-shaped article boxing device applied to a boxing robot, wherein the boxing robot comprises a first RGBD camera and a second RGBD camera, the first RGBD camera is arranged at the tail end of the boxing robot, the second RGBD camera is arranged right above a target box body and is used for collecting images of the target box body, and the target box body is a cuboid-shaped box body; the method comprises the following steps:
the first acquisition module is used for acquiring the size and the top surface pose information of each article to be boxed through the first RGBD camera; the article to be boxed is a cuboid article; the dimensions include length, width and height;
the mask matrix generating module is used for generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the packing area of the target box body in the image collected by the second RGBD camera;
the first execution module is used for sequentially taking each article to be boxed as a target article, comparing the residual height information of each part of the boxing area with the height of the target article to update the mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item.
According to the cuboid-shaped object boxing device, the size and the top face pose information of an object to be boxed are determined through the first RGBD camera, the residual height information of each position in a boxing area of a target box body is obtained through the second RGBD camera, a mask matrix with the same pixel size as the boxing area is utilized to perform mask operation on a depth image of the boxing area to determine a candidate area with the residual height not less than the height of the target object, and finally an effective space capable of containing the target object is searched from the candidate area, so that the target object can be accurately placed in the position capable of containing the target object, automatic boxing of the object to be boxed is achieved, and the cuboid-shaped object boxing device has good practicability and universality.
In a third aspect, the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor executes the computer program to execute the steps of the rectangular parallelepiped object packing method as described above.
In a fourth aspect, the present application provides a storage medium having stored thereon a computer program which, when executed by a processor, executes the steps in the rectangular parallelepiped object packing method as described above.
Has the advantages that:
according to the cuboid-shaped object boxing method, the cuboid-shaped object boxing device, the electronic equipment and the storage medium, the size and the top face pose information of an object to be boxed are determined through the first RGBD camera, the residual height information of each position in a boxing area of a target box body is obtained through the second RGBD camera, a mask matrix with the same size as the pixel size of the boxing area is utilized to perform mask operation on a depth image of the boxing area to determine a candidate area with the residual height not less than the height of the target object, and finally an effective space capable of containing the target object is searched from the candidate area, so that the target object can be accurately placed in the position capable of containing the target object, automatic boxing of the object to be boxed is achieved, and the method has good practicability and universality.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application.
Drawings
Fig. 1 is a flowchart of a method for packing rectangular parallelepiped articles according to an embodiment of the present application.
Fig. 2 is a schematic structural diagram of a rectangular parallelepiped object packing device according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Referring to fig. 1, fig. 1 is a method for boxing a cuboid-shaped article, which is applied to a boxing robot in some embodiments of the present application, where the boxing robot includes a first RGBD camera and a second RGBD camera, the first RGBD camera is disposed at an end of the boxing robot, the second RGBD camera is disposed right above a target box and is used for acquiring an image of the target box, and the target box is a cuboid-shaped box; the method comprises the following steps:
A1. acquiring the size and top surface pose information of each article to be boxed through a first RGBD (red, green and blue) camera; the article to be boxed is a cuboid article; dimensions include length, width and height;
A2. generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the boxing region of the target box (i.e., the inner cavity region of the target box) in the image acquired by the second RGBD camera (for example, if the boxing region of the target box has n rows and m columns of pixel points in the image acquired by the second RGBD camera, that is, the pixel size is n × m, the mask matrix correspondingly has n rows and m columns of elements);
A3. sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of a boxing area with the height of the target article to update a mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item.
According to the cuboid object boxing method, the size and the top face pose information of an object to be boxed are determined through the first RGBD camera, the residual height information of each position in a boxing area of a target box body is obtained through the second RGBD camera, a mask matrix with the same pixel size as the boxing area is utilized to perform mask operation on a depth image of the boxing area to determine a candidate area with the residual height not less than the height of the target object, and finally an effective space capable of containing the target object is searched from the candidate area, so that the target object can be accurately placed at a position capable of containing the target object, automatic boxing of the object to be boxed is achieved, and the method has good practicability and universality.
Preferably, the initial value of each element value of the mask matrix is 1;
step a3 includes: sequentially taking each article to be boxed as a target article and executing the following steps:
A301. acquiring a depth image of a boxing area and residual height information of all parts of the boxing area by a second RGBD camera;
A302. in the mask matrix, changing the element value corresponding to the position where the remaining height of the boxing area is less than the height of the target object into 0;
A303. performing masking operation on the depth image of the boxing region by using the mask matrix to obtain a mask image, and taking a region with a non-zero pixel value in the mask image as a candidate region;
A304. searching an effective space in the candidate area according to the length and the width of the target object;
A305. and grabbing the target object according to the top surface pose information of the target object and putting the target object into an effective space.
In practical application, the object to be boxed can be vertically placed on a taking table (i.e. the object is placed close to the taking table according to the bottom surface of the object to be boxed), the target box body is vertically placed on a placing table (i.e. the opening of the target box body is placed upwards on the placing table, the bottom surface of the target box body is placed close to the placing table), the second RGBD camera is arranged downwards right above the placing table, and the optical axis of the second RGBD camera is perpendicular to the table top of the placing table. In order to prevent the inaccuracy of the depth image data collected by the RGBD camera due to reflection, a dark (e.g. black) background curtain can be arranged on the table top of the pickup table and the placing table.
In some embodiments, step a1 includes:
acquiring a first image of each article to be boxed, which is acquired by a first RGBD camera vertically downwards; the first image comprises a first color image and a first depth image;
dividing a top surface image of each article to be boxed in the first color image, and acquiring pixel coordinate data of each pixel point of the top surface image;
extracting depth data of each pixel point of the top surface image from the first depth image according to the segmentation result;
acquiring the length and the width of a corresponding article to be packed and top face pose information of the top face of the article to be packed under a first camera coordinate system (namely a camera coordinate system of a first RGBD camera) according to pixel coordinate data of each pixel point of the top face image;
and acquiring the height of the corresponding article to be boxed according to the depth data of each pixel point of the top surface image.
In some preferred embodiments, the boxing robot drives the first RGBD camera to a preset shooting pose point to acquire the first image of each article to be boxed, so that when the length and the width of the article to be boxed are calculated, the actual length and the actual width can be obtained by directly multiplying the first pixel size (namely the pixel size in the first image) by a preset first conversion coefficient (obtained by actual measurement), and the method is convenient and fast. But is not limited thereto.
The step of dividing the top surface image of each article to be boxed in the first color image is not described in detail herein, as in the prior art (for example, the outer contour line of each article to be boxed is obtained by an edge detection method, and then the top surface image of each article to be boxed is divided according to the outer contour line).
The pixel point position of the first color image and the pixel point position of the first depth image are in one-to-one correspondence, so that the depth data of the top surface image can be directly extracted from the pixel point corresponding to the first depth image according to the pixel coordinate data of each pixel point of the divided top surface image.
The step of obtaining the length and the width of the corresponding article to be boxed according to the pixel coordinate data of each pixel point of the top surface image comprises the following steps:
calculating a first pixel length and a first pixel width of the top surface image according to the pixel coordinates of the corner points of the top surface image;
acquiring a first pixel size (namely the pixel size in a first image) and a first conversion coefficient of an actual size (for the condition that the first image of each article to be boxed is collected at a preset shooting pose point, the first conversion coefficient is a preset value, for the condition that the first image of each article to be boxed is not collected at the preset shooting pose point, the first conversion coefficient is calculated according to the mean value of depth data of each pixel point of a top surface image, wherein the specific calculation method is the prior art);
and multiplying the first pixel length and the first pixel width of the top surface image by the first conversion coefficient respectively to obtain the length and the width of the corresponding article to be boxed.
The specific method for acquiring the top surface pose information of the top surface of the article to be boxed under the first camera coordinate system according to the pixel coordinate data of each pixel point of the top surface image is the prior art, and the detailed description is not given here.
The step of obtaining the height of the corresponding article to be boxed according to the depth data of each pixel point of the top surface image comprises the following steps:
calculating a first mean value of depth data of each pixel point of the top surface image;
acquiring a second average value of depth data of pixel points of a background part in the first depth image;
and subtracting the first average value from the second average value to obtain the corresponding height of the articles to be boxed.
In practical application, the top surface image of each article to be packed may be segmented according to the first depth image (for example, the top surface image of each article to be packed is clustered by using a clustering algorithm according to the depth data of each pixel point, so as to implement segmentation), and then the pixel coordinate data of each pixel point of the top surface image is extracted from the first color image according to the segmentation result.
In this embodiment, step a2 includes:
acquiring a third color image of the target box body acquired by the second RGBD camera;
segmenting the boxing area of the target box body by using the third color image to obtain a third pixel size of the boxing area;
and generating a mask matrix according to the third pixel size of the boxed area, and assigning the value of each element of the mask matrix to be 1.
The number and the positions of the elements of the mask matrix correspond to the number and the positions of the pixel points of the partitioned boxing region image one to one.
In some preferred embodiments, step a3 is preceded by the steps of:
A4. calculating the top surface area of each article to be boxed according to the length and the width of each article to be boxed;
A5. sorting the articles to be boxed in a descending order according to the area of the top surface;
further, in step a3, the target articles are sequentially sorted by the sort order.
Therefore, the articles with large top surface area are always positioned below the articles with small top surface area, and the articles are not easy to overturn after being superposed, so that the practicability of the boxing method is further improved.
In some embodiments, step a301 comprises:
acquiring a second image of the target box body acquired by a second RGBD camera; the second image comprises a second color image and a second depth image;
segmenting the boxed area of the target box body in the second color image to obtain the boundary of the boxed area (the specific segmentation method can adopt the prior art, and the detailed description is not given here);
segmenting the second depth image according to the boundary to obtain a depth image of the packing area;
extracting real-time depth data of each pixel point from the depth image of the packing area;
and subtracting the real-time depth data from the initial depth data of each pixel point to obtain the used height data of each pixel point, and subtracting the used height data of each pixel point from the height (which can be measured in advance) of the target box body to obtain the residual height information of each pixel point in the boxing area.
The residual height refers to the height of an available space at the position of the corresponding pixel point, the residual height information includes the information of the residual height, and the initial depth data is the depth data of each pixel point in the packing area when no article is packed. The second color image may be the same image as the third color image (i.e., the third color image may be used as the second color image corresponding to the first target article) or may not be the same image as the first color image for the first target article, and for the subsequent other target articles, the second color image and the third color image may not be the same image.
In step a302, for example, if the remaining height of the (i, j) th pixel point in the depth image of the binning area is less than the height of the target object, the value of the corresponding (i, j) th element in the mask matrix is changed to 0.
Wherein, step a303 includes:
and multiplying each element value in the mask matrix by the pixel value of the pixel point corresponding to the depth image of the packing area respectively to update the pixel value of the pixel point of the depth image of the packing area according to the calculation result, so as to obtain the mask image.
Therefore, in the mask image, the pixel values of the pixel points with the residual height smaller than the height of the target object are all 0, and the pixel values of other pixel points are kept unchanged and are all nonzero values. Therefore, the region with the non-zero pixel value is a region which meets the height requirement of the target object on the placement space (namely, the candidate region is a region which meets the height requirement of the target object on the placement space), only the effective space needs to be searched in the candidate region subsequently, and compared with a mode of searching the effective space in the whole boxing region every time, the method can reduce the search range, reduce the data processing amount and improve the search efficiency.
In this embodiment, step a304 includes:
and searching an effective space capable of accommodating the window in the candidate area by a sliding window method by adopting the window with the same size as the top surface of the target object.
Specifically, the steps include:
s1, calculating a second pixel size of the top surface of the target object according to a second conversion coefficient (which can be measured in advance) of the second pixel size and the actual size of the second image and the length and width of the target object;
s2, generating a window according to the second pixel size of the top surface of the target object, and enabling the pixel size of the window to be equal to the second pixel size of the top surface of the target object (for example, if the second pixel size of the top surface of the target object is n × m, namely n rows and m columns exist, the pixel size of the window is n × m);
s3, searching line by line in the candidate area in the sequence from top to bottom, wherein the step length between lines is a pixel point; in each row, taking one pixel point as a step length, and performing sliding window search from left to right;
and S4, in the process of sliding the window, marking the window area as an effective space when the pixel points with non-zero pixel values are filled in the window.
In a further preferred embodiment, when no valid space is searched after completing a search, the window may be rotated by 90 ° and the search may be repeated (i.e., steps S3-S4 are repeated).
When a plurality of effective spaces are searched, the first searched effective space can be directly used as a final effective space, and the target object can be placed in the final effective space.
In some preferred embodiments, when a plurality of effective spaces are searched, the effective space with the largest remaining height may be used as a final effective space, and the target item may be placed in the final effective space; thus, step a304 further comprises:
if there are a plurality of searched effective spaces, the effective space with the largest remaining height is taken as the final effective space.
Therefore, the lower layer space can be filled preferentially, the probability that the bottoms of the articles are not supported completely when the articles are stacked is reduced, and the probability that the articles are overturned after being stacked is further reduced; the probability that the lower-layer gap cannot be effectively utilized due to the fact that the upper-layer object covers the lower-layer gap can be reduced, and therefore the space utilization rate of the target box body is improved.
In practical applications, there may be more than one effective space with the largest remaining height, and in this case, the first searched effective space in the effective space with the largest remaining height may be the final effective space.
Preferably, step a305 comprises:
acquiring grabbing point position and posture information according to the height and the top surface position and posture information of the target object;
acquiring pose information of an effective space;
acquiring position and posture information of a placement point according to the height of the target object and the posture information of the effective space;
and grabbing the target object from the grabbing point and moving the target object to the placing point for placing according to the grabbing point position information and the placing point position information.
The specific method for acquiring the pose information of the grabbing point according to the height and the top face pose information of the target object is the prior art, and the detailed description is not provided here.
The pose information of the effective space is pose information of the effective space in a second camera coordinate system (i.e. a camera coordinate system of a second RGBD camera), and can be acquired by the prior art (details on an acquisition method thereof are not described here); the specific method for acquiring the position and pose information of the placement point according to the height of the target object and the position and pose information of the effective space is the prior art, and the detailed description is not provided here.
After the grabbing point position information and the placing point position information are obtained, a path from the grabbing point to the placing point can be planned, so that the boxing robot moves along the path.
In some preferred embodiments, the step of grabbing the target object from the grabbing point and moving the target object to the placing point for placing according to the grabbing point location information and the placing point location information includes:
moving to a first grabbing transition pose point; the first grabbing transition position point is located right above the target object;
moving to a grabbing point to grab a target object according to the grabbing point position information;
moving to a second grabbing transition pose point; the second grabbing transition position point is located right above the target object;
moving to a first placement transition pose point; the first placement transition position point is positioned right above the effective space;
moving to a placing point for placing according to the pose information of the placing point;
moving to a second placement transition pose point; the second placement transition position point is located directly above the effective space.
The heights of the first grabbing transition pose point, the second grabbing transition pose point, the first placing transition pose point and the second placing transition pose point can be preset according to actual needs.
By setting these transition points, the advantages are: the barrier can be avoided, the object can be conveniently and vertically placed in the target box body, and the object is prevented from colliding with other objects outside the effective space when the object is placed.
According to the method for packing the cuboid-shaped objects, the size and the top surface pose information of each object to be packed are obtained through the first RGBD camera; the article to be boxed is a cuboid article; dimensions include length, width and height; generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the boxing area of the target box body in the image acquired by the second RGBD camera; sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of a boxing area with the height of the target article to update a mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target object; therefore, the target object can be accurately placed in the position capable of containing the target object, automatic boxing of the objects to be boxed is realized, and the method has better practicability and universality.
Referring to fig. 2, the present application provides a rectangular article boxing device, which is applied to a boxing robot, wherein the boxing robot includes a first RGBD camera and a second RGBD camera, the first RGBD camera is disposed at a tail end of the boxing robot, the second RGBD camera is disposed right above a target box and is used for collecting an image of the target box, and the target box is a rectangular box; the method comprises the following steps:
the first acquisition module 1 is used for acquiring the size and the top surface pose information of each article to be boxed through a first RGBD (red, green and blue) camera; the article to be boxed is a cuboid article; dimensions include length, width and height;
the mask matrix generating module 2 is used for generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the boxing area of the target box body in the image acquired by the second RGBD camera;
the first execution module 3 is used for sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of the boxing area with the height of the target article to update the mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item.
According to the cuboid-shaped object boxing device, the size and the top face pose information of an object to be boxed are determined through the first RGBD camera, the residual height information of each position in a boxing area of a target box body is obtained through the second RGBD camera, a mask matrix with the same pixel size as the boxing area is utilized to perform mask operation on a depth image of the boxing area to determine a candidate area with the residual height not less than the height of the target object, and finally an effective space capable of containing the target object is searched from the candidate area, so that the target object can be accurately placed in the position capable of containing the target object, automatic boxing of the object to be boxed is achieved, and the cuboid-shaped object boxing device has good practicability and universality.
Preferably, the initial value of each element value of the mask matrix is 1;
the first execution module 3 is used for sequentially taking each article to be boxed as a target article and executing the following steps:
acquiring a depth image of a boxing area and residual height information of all parts of the boxing area by a second RGBD camera;
in the mask matrix, changing the element value corresponding to the position where the remaining height of the boxing area is less than the height of the target object into 0;
performing masking operation on the depth image of the boxing region by using the mask matrix to obtain a mask image, and taking a region with a non-zero pixel value in the mask image as a candidate region;
searching an effective space in the candidate area according to the length and the width of the target object;
and grabbing the target object according to the top surface pose information of the target object and putting the target object into an effective space.
In practical application, the object to be boxed can be vertically placed on a taking table (i.e. the object is placed close to the taking table according to the bottom surface of the object to be boxed), the target box body is vertically placed on a placing table (i.e. the opening of the target box body is placed upwards on the placing table, the bottom surface of the target box body is placed close to the placing table), the second RGBD camera is arranged downwards right above the placing table, and the optical axis of the second RGBD camera is perpendicular to the table top of the placing table. In order to prevent the inaccuracy of the depth image data collected by the RGBD camera due to reflection, a dark (e.g. black) background curtain can be arranged on the table top of the pickup table and the placing table.
In some embodiments, the first acquiring module 1 is configured to, when acquiring the size and top surface pose information of each article to be boxed through the first RGBD camera, perform:
acquiring a first image of each article to be boxed, which is acquired by a first RGBD camera vertically downwards; the first image comprises a first color image and a first depth image;
dividing a top surface image of each article to be boxed in the first color image, and acquiring pixel coordinate data of each pixel point of the top surface image;
extracting depth data of each pixel point of the top surface image from the first depth image according to the segmentation result;
acquiring the length and the width of a corresponding article to be packed and top face pose information of the top face of the article to be packed under a first camera coordinate system (namely a camera coordinate system of a first RGBD camera) according to pixel coordinate data of each pixel point of the top face image;
and acquiring the height of the corresponding article to be boxed according to the depth data of each pixel point of the top surface image.
In some preferred embodiments, the boxing robot drives the first RGBD camera to a preset shooting pose point to acquire the first image of each article to be boxed, so that when the length and the width of the article to be boxed are calculated, the actual length and the actual width can be obtained by directly multiplying the first pixel size (namely the pixel size in the first image) by a preset first conversion coefficient (obtained by actual measurement), and the method is convenient and fast. But is not limited thereto.
The step of dividing the top surface image of each article to be boxed in the first color image is not described in detail herein, as in the prior art (for example, the outer contour line of each article to be boxed is obtained by an edge detection method, and then the top surface image of each article to be boxed is divided according to the outer contour line).
The pixel point position of the first color image and the pixel point position of the first depth image are in one-to-one correspondence, so that the depth data of the top surface image can be directly extracted from the pixel point corresponding to the first depth image according to the pixel coordinate data of each pixel point of the divided top surface image.
When the first obtaining module 1 obtains the length and the width of the corresponding article to be boxed according to the pixel coordinate data of each pixel point of the top surface image, the following steps are executed:
calculating a first pixel length and a first pixel width of the top surface image according to the pixel coordinates of the corner points of the top surface image;
acquiring a first pixel size (namely the pixel size in a first image) and a first conversion coefficient of an actual size (for the condition that the first image of each article to be boxed is collected at a preset shooting pose point, the first conversion coefficient is a preset value, for the condition that the first image of each article to be boxed is not collected at the preset shooting pose point, the first conversion coefficient is calculated according to the mean value of depth data of each pixel point of a top surface image, wherein the specific calculation method is the prior art);
and multiplying the first pixel length and the first pixel width of the top surface image by the first conversion coefficient respectively to obtain the length and the width of the corresponding article to be boxed.
The specific method for acquiring the top surface pose information of the top surface of the article to be boxed under the first camera coordinate system according to the pixel coordinate data of each pixel point of the top surface image is the prior art, and the detailed description is not given here.
The first obtaining module 1 executes the following steps when obtaining the height of the corresponding article to be boxed according to the depth data of each pixel point of the top surface image:
calculating a first mean value of depth data of each pixel point of the top surface image;
acquiring a second average value of depth data of pixel points of a background part in the first depth image;
and subtracting the first average value from the second average value to obtain the corresponding height of the articles to be boxed.
In practical application, the top surface image of each article to be packed may be segmented according to the first depth image (for example, the top surface image of each article to be packed is clustered by using a clustering algorithm according to the depth data of each pixel point, so as to implement segmentation), and then the pixel coordinate data of each pixel point of the top surface image is extracted from the first color image according to the segmentation result.
In this embodiment, the mask matrix generating module 2 is configured to, when generating a mask matrix from an image acquired by the second RGBD camera, perform:
acquiring a third color image of the target box body acquired by the second RGBD camera;
segmenting the boxing area of the target box body by using the third color image to obtain a third pixel size of the boxing area;
and generating a mask matrix according to the third pixel size of the boxed area, and assigning the value of each element of the mask matrix to be 1.
The number and the positions of the elements of the mask matrix correspond to the number and the positions of the pixel points of the partitioned boxing region image one to one.
In some preferred embodiments, the rectangular parallelepiped article packing device further includes:
the first calculation module is used for calculating the area of the top surface of each article to be boxed according to the length and the width of each article to be boxed;
the sorting module is used for sorting the articles to be boxed in a descending order according to the area of the top surface;
furthermore, the first execution module 3 sequentially takes the articles to be boxed as target articles according to the sequencing order.
Therefore, the articles with large top surface area are always positioned below the articles with small top surface area, and the articles are not easy to overturn after being superposed, so that the practicability of the boxing device is further improved.
In some embodiments, the first performing module 3 performs, when acquiring the depth image of the binning area and the remaining height information throughout the binning area by the second RGBD camera:
acquiring a second image of the target box body acquired by a second RGBD camera; the second image comprises a second color image and a second depth image;
segmenting the boxed area of the target box body in the second color image to obtain the boundary of the boxed area (the specific segmentation method can adopt the prior art, and the detailed description is not given here);
segmenting the second depth image according to the boundary to obtain a depth image of the packing area;
extracting real-time depth data of each pixel point from the depth image of the packing area;
and subtracting the real-time depth data from the initial depth data of each pixel point to obtain the used height data of each pixel point, and subtracting the used height data of each pixel point from the height (which can be measured in advance) of the target box body to obtain the residual height information of each pixel point in the boxing area.
The residual height refers to the height of an available space at the position of the corresponding pixel point, and the initial depth data is the depth data of each pixel point in the packing area when no article is packed. The second color image may be the same image as the third color image (i.e., the third color image may be used as the second color image corresponding to the first target article) or may not be the same image as the first color image for the first target article, and for the subsequent other target articles, the second color image and the third color image may not be the same image.
When the first execution module 3 changes the element value corresponding to the position where the remaining height of the packing region is smaller than the height of the target object to 0 in the mask matrix, for example, the remaining height of the (i, j) th pixel point in the depth image of the packing region is smaller than the height of the target object, the value of the corresponding (i, j) th element in the mask matrix is changed to 0.
When the first execution module 3 performs a masking operation on the depth image of the boxing region by using the mask matrix to obtain a mask image, the following steps are performed:
and multiplying each element value in the mask matrix by the pixel value of the pixel point corresponding to the depth image of the packing area respectively to update the pixel value of the pixel point of the depth image of the packing area according to the calculation result, so as to obtain the mask image.
Therefore, in the mask image, the pixel values of the pixel points with the residual height smaller than the height of the target object are all 0, and the pixel values of other pixel points are kept unchanged and are all nonzero values. Therefore, the region with the non-zero pixel value is a region which meets the height requirement of the target object on the placement space (namely, the candidate region is a region which meets the height requirement of the target object on the placement space), only the effective space needs to be searched in the candidate region subsequently, and compared with a mode of searching the effective space in the whole boxing region every time, the method can reduce the search range, reduce the data processing amount and improve the search efficiency.
In this embodiment, the first execution module 3, when searching for an effective space in the candidate area according to the length and width of the target item, executes:
and searching an effective space capable of accommodating the window in the candidate area by a sliding window method by adopting the window with the same size as the top surface of the target object.
Specifically, the first execution module 3, when searching for an effective space that can accommodate a window in the candidate area by a sliding window method using a window having the same size as the top surface of the target item, executes:
s1, calculating a second pixel size of the top surface of the target object according to a second conversion coefficient (which can be measured in advance) of the second pixel size and the actual size of the second image and the length and width of the target object;
s2, generating a window according to the second pixel size of the top surface of the target object, and enabling the pixel size of the window to be equal to the second pixel size of the top surface of the target object (for example, if the second pixel size of the top surface of the target object is n × m, namely n rows and m columns exist, the pixel size of the window is n × m);
s3, searching line by line in the candidate area in the sequence from top to bottom, wherein the step length between lines is a pixel point; in each row, taking one pixel point as a step length, and performing sliding window search from left to right;
and S4, in the process of sliding the window, marking the window area as an effective space when the pixel points with non-zero pixel values are filled in the window.
In a further preferred embodiment, when no valid space is searched after completing a search, the window may be rotated by 90 ° and the search may be repeated (i.e., steps S3-S4 are repeated).
When a plurality of effective spaces are searched, the first searched effective space can be directly used as a final effective space, and the target object can be placed in the final effective space.
In some preferred embodiments, when a plurality of effective spaces are searched, the effective space with the largest remaining height may be used as a final effective space, and the target item may be placed in the final effective space; thus, the first execution module 3, when searching for an effective space that can accommodate a window in the candidate area in a sliding window method using a window having the same size as the top surface of the target item, executes:
if there are a plurality of searched effective spaces, the effective space with the largest remaining height is taken as the final effective space.
Therefore, the lower layer space can be filled preferentially, the probability that the bottoms of the articles are not supported completely when the articles are stacked is reduced, and the probability that the articles are overturned after being stacked is further reduced; the probability that the lower-layer gap cannot be effectively utilized due to the fact that the upper-layer object covers the lower-layer gap can be reduced, and therefore the space utilization rate of the target box body is improved.
In practical applications, there may be more than one effective space with the largest remaining height, and in this case, the first searched effective space in the effective space with the largest remaining height may be the final effective space.
Preferably, the first executing module 3 executes, when grabbing the target object according to the pose information of the target object and placing the target object in the effective space:
acquiring grabbing point position and posture information according to the height and the top surface position and posture information of the target object;
acquiring pose information of an effective space;
acquiring position and posture information of a placement point according to the height of the target object and the posture information of the effective space;
and grabbing the target object from the grabbing point and moving the target object to the placing point for placing according to the grabbing point position information and the placing point position information.
The specific method for acquiring the pose information of the grabbing point according to the height and the top face pose information of the target object is the prior art, and the detailed description is not provided here.
The pose information of the effective space is pose information of the effective space in a second camera coordinate system (i.e. a camera coordinate system of a second RGBD camera), and can be acquired by the prior art (details on an acquisition method thereof are not described here); the specific method for acquiring the position and pose information of the placement point according to the height of the target object and the position and pose information of the effective space is the prior art, and the detailed description is not provided here.
After the grabbing point position information and the placing point position information are obtained, a path from the grabbing point to the placing point can be planned, so that the boxing robot moves along the path.
In some preferred embodiments, the first executing module 3 executes, when grabbing the target object from the grabbing point and moving to the placing point for placing according to the grabbing point location information and the placing point location information:
moving to a first grabbing transition pose point; the first grabbing transition position point is located right above the target object;
moving to a grabbing point to grab a target object according to the grabbing point position information;
moving to a second grabbing transition pose point; the second grabbing transition position point is located right above the target object;
moving to a first placement transition pose point; the first placement transition position point is positioned right above the effective space;
moving to a placing point for placing according to the pose information of the placing point;
moving to a second placement transition pose point; the second placement transition position point is located directly above the effective space.
The heights of the first grabbing transition pose point, the second grabbing transition pose point, the first placing transition pose point and the second placing transition pose point can be preset according to actual needs.
By setting these transition points, the advantages are: the barrier can be avoided, the object can be conveniently and vertically placed in the target box body, and the object is prevented from colliding with other objects outside the effective space when the object is placed.
According to the cuboid-shaped object boxing device, the size and the top surface pose information of each object to be boxed are obtained through the first RGBD camera; the article to be boxed is a cuboid article; dimensions include length, width and height; generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the boxing area of the target box body in the image acquired by the second RGBD camera; sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of a boxing area with the height of the target article to update a mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target object; therefore, the target object can be accurately placed in the position capable of containing the target object, automatic boxing of the objects to be boxed is realized, and the method has better practicability and universality.
Referring to fig. 3, fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure, where the present disclosure provides an electronic device, including: the processor 301 and the memory 302, the processor 301 and the memory 302 are interconnected and communicate with each other through the communication bus 303 and/or other types of connection mechanisms (not shown), the memory 302 stores a computer program executable by the processor 301, and when the electronic device runs, the processor 301 executes the computer program to execute the cuboid-shaped object packing method in any optional implementation manner of the above embodiments to realize the following functions: acquiring the size and top surface pose information of each article to be boxed through a first RGBD (red, green and blue) camera; the article to be boxed is a cuboid article; dimensions include length, width and height; generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the boxing area of the target box body in the image acquired by the second RGBD camera; sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of a boxing area with the height of the target article to update a mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item.
The present application provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for packing a rectangular parallelepiped article in any one of the optional implementations of the foregoing embodiments is executed, so as to implement the following functions: acquiring the size and top surface pose information of each article to be boxed through a first RGBD (red, green and blue) camera; the article to be boxed is a cuboid article; dimensions include length, width and height; generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the boxing area of the target box body in the image acquired by the second RGBD camera; sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of a boxing area with the height of the target article to update a mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item. The storage medium may be implemented by any type of volatile or nonvolatile storage device or combination thereof, such as a Static Random Access Memory (SRAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), an Erasable Programmable Read-Only Memory (EPROM), a Programmable Read-Only Memory (PROM), a Read-Only Memory (ROM), a magnetic Memory, a flash Memory, a magnetic disk, or an optical disk.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
In addition, units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
Furthermore, the functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (10)
1. A cuboid-shaped article boxing method is applied to a boxing robot, the boxing robot comprises a first RGBD camera and a second RGBD camera, the first RGBD camera is arranged at the tail end of the boxing robot, the second RGBD camera is arranged right above a target box body and is used for collecting images of the target box body, and the target box body is a cuboid-shaped box body; the method is characterized by comprising the following steps:
A1. acquiring the size and top surface pose information of each article to be boxed through the first RGBD camera; the article to be boxed is a cuboid article; the dimensions include length, width and height;
A2. generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the packing area of the target box body in the image collected by the second RGBD camera;
A3. sequentially taking each article to be boxed as a target article, comparing the residual height information of each position of the boxing area with the height of the target article to update the mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item.
2. The rectangular parallelepiped object boxing method according to claim 1, wherein an initial value of each element value of the mask matrix is 1;
step a3 includes: sequentially taking each article to be boxed as a target article and executing the following steps:
A301. acquiring a depth image of the boxing area and residual height information of all parts of the boxing area through the second RGBD camera;
A302. in the mask matrix, changing an element value corresponding to a position where the remaining height of the boxing area is less than the height of the target article to 0;
A303. performing masking operation on the depth image of the boxing region by using the mask matrix to obtain a mask image, and taking a region with a non-zero pixel value in the mask image as the candidate region;
A304. searching for an effective space in the candidate area according to the length and the width of the target item;
A305. and grabbing the target object according to the top surface pose information of the target object and putting the target object into the effective space.
3. The rectangular parallelepiped object boxing method according to claim 1, wherein before the step a3, the method further comprises the steps of:
calculating the top surface area of each article to be boxed according to the length and the width of each article to be boxed;
sorting each article to be boxed in a descending order according to the area of the top surface;
in step a3, each of the articles to be boxed is sequentially taken as a target article in the order of sorting.
4. The rectangular parallelepiped object boxing method according to claim 2, wherein the step a301 comprises:
acquiring a second image of the target box acquired by the second RGBD camera; the second image comprises a second color image and a second depth image;
segmenting the boxed area of the target box body in the second color image to obtain the boundary of the boxed area;
segmenting the second depth image according to the boundary to obtain a depth image of the boxing area;
extracting real-time depth data of each pixel point from the depth image of the packing area;
and subtracting the real-time depth data from the initial depth data of each pixel point to obtain used height data of each pixel point, and then subtracting the used height data of each pixel point from the height of the target box body to obtain residual height information of each pixel point in the boxing area.
5. The rectangular parallelepiped object boxing method according to claim 2, wherein the step a304 comprises:
and searching for an effective space capable of accommodating the window in the candidate area by a sliding window method by adopting the window with the same size as the top surface of the target object.
6. The method for boxing rectangular parallelepiped-shaped articles as claimed in claim 5, wherein the step A304 further comprises:
and if a plurality of searched effective spaces exist, taking the effective space with the largest residual height as the final effective space.
7. The rectangular parallelepiped object boxing method according to claim 2, wherein the step a305 comprises:
acquiring grabbing point position and posture information according to the height and the top surface position and posture information of the target object;
acquiring pose information of the effective space;
acquiring position and posture information of a placement point according to the height of the target object and the posture information of the effective space;
and grabbing the target object from the grabbing point and moving the target object to the placing point for placing according to the grabbing point position information and the placing point position information.
8. A cuboid-shaped article boxing device is applied to a boxing robot, the boxing robot comprises a first RGBD camera and a second RGBD camera, the first RGBD camera is arranged at the tail end of the boxing robot, the second RGBD camera is arranged right above a target box body and is used for collecting images of the target box body, and the target box body is a cuboid-shaped box body; it is characterized by comprising:
the first acquisition module is used for acquiring the size and the top surface pose information of each article to be boxed through the first RGBD camera; the article to be boxed is a cuboid article; the dimensions include length, width and height;
the mask matrix generating module is used for generating a mask matrix according to the image acquired by the second RGBD camera; the size of the mask matrix is the same as the pixel size of the packing area of the target box body in the image collected by the second RGBD camera;
the first execution module is used for sequentially taking each article to be boxed as a target article, comparing the residual height information of each part of the boxing area with the height of the target article to update the mask matrix, determining a candidate area of the boxing area by using the updated mask matrix, and searching an effective space in the candidate area to place the target article in the effective space; the remaining height of the candidate area is not less than the height of the target item.
9. An electronic device comprising a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor executes the computer program to execute the steps of the rectangular parallelepiped-shaped article packing method according to any one of claims 1 to 7.
10. A storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, executes the steps of the rectangular parallelepiped object packing method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210004025.3A CN114013751B (en) | 2022-01-05 | 2022-01-05 | Rectangular article boxing method, device, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210004025.3A CN114013751B (en) | 2022-01-05 | 2022-01-05 | Rectangular article boxing method, device, electronic device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114013751A true CN114013751A (en) | 2022-02-08 |
| CN114013751B CN114013751B (en) | 2022-04-01 |
Family
ID=80069518
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210004025.3A Active CN114013751B (en) | 2022-01-05 | 2022-01-05 | Rectangular article boxing method, device, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114013751B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102923578A (en) * | 2012-11-13 | 2013-02-13 | 扬州华泰特种设备有限公司 | Automatic control system of efficient handing operation of container crane |
| US20170267461A1 (en) * | 2014-08-04 | 2017-09-21 | Panasonic Intellectual Property Management Co., Ltd. | Package loading instruction method and loading instruction system |
| CN107804511A (en) * | 2016-09-09 | 2018-03-16 | 广州汽车集团乘用车有限公司 | A kind of plate packing method and device |
| CN108009675A (en) * | 2017-11-28 | 2018-05-08 | 上海量科电子科技有限公司 | Cargo packing method, apparatus and system |
| CN108698224A (en) * | 2017-08-23 | 2018-10-23 | 深圳蓝胖子机器人有限公司 | The method of robot store items, the system and robot of control robot store items |
| CN110443549A (en) * | 2019-07-26 | 2019-11-12 | 清华大学 | The vanning scheme generation method and system of article in a kind of logistics warehouse |
| CN110853004A (en) * | 2019-11-04 | 2020-02-28 | 华东交通大学 | Intelligent shape recognition and optimized packing method and device for part load goods |
| CN112967283A (en) * | 2021-04-22 | 2021-06-15 | 上海西井信息科技有限公司 | Target identification method, system, equipment and storage medium based on binocular camera |
| CN113060353A (en) * | 2021-02-26 | 2021-07-02 | 深圳市华盛控科技有限公司 | Boxing equipment and boxing method thereof |
| CN113592855A (en) * | 2021-08-19 | 2021-11-02 | 山东大学 | Heuristic deep reinforcement learning-based autonomous grabbing and boxing method and system |
-
2022
- 2022-01-05 CN CN202210004025.3A patent/CN114013751B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102923578A (en) * | 2012-11-13 | 2013-02-13 | 扬州华泰特种设备有限公司 | Automatic control system of efficient handing operation of container crane |
| US20170267461A1 (en) * | 2014-08-04 | 2017-09-21 | Panasonic Intellectual Property Management Co., Ltd. | Package loading instruction method and loading instruction system |
| CN107804511A (en) * | 2016-09-09 | 2018-03-16 | 广州汽车集团乘用车有限公司 | A kind of plate packing method and device |
| CN108698224A (en) * | 2017-08-23 | 2018-10-23 | 深圳蓝胖子机器人有限公司 | The method of robot store items, the system and robot of control robot store items |
| CN108009675A (en) * | 2017-11-28 | 2018-05-08 | 上海量科电子科技有限公司 | Cargo packing method, apparatus and system |
| CN110443549A (en) * | 2019-07-26 | 2019-11-12 | 清华大学 | The vanning scheme generation method and system of article in a kind of logistics warehouse |
| CN110853004A (en) * | 2019-11-04 | 2020-02-28 | 华东交通大学 | Intelligent shape recognition and optimized packing method and device for part load goods |
| CN113060353A (en) * | 2021-02-26 | 2021-07-02 | 深圳市华盛控科技有限公司 | Boxing equipment and boxing method thereof |
| CN112967283A (en) * | 2021-04-22 | 2021-06-15 | 上海西井信息科技有限公司 | Target identification method, system, equipment and storage medium based on binocular camera |
| CN113592855A (en) * | 2021-08-19 | 2021-11-02 | 山东大学 | Heuristic deep reinforcement learning-based autonomous grabbing and boxing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114013751B (en) | 2022-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10974394B2 (en) | Robot assisted object learning vision system | |
| Schwarz et al. | Fast object learning and dual-arm coordination for cluttered stowing, picking, and packing | |
| JP7433609B2 (en) | Method and computational system for object identification | |
| WO2017201490A1 (en) | Method for automatically generating a planogram that assigns products to shelving structures within a store | |
| CN111199562A (en) | System and method for rapid object detection while robotic picking | |
| JP2022514757A (en) | Systems and methods for object sizing based on partial visual information | |
| CN112802105A (en) | Object grabbing method and device | |
| CN108789499B (en) | Article retrieval system | |
| JP2004160567A (en) | Article taking-out device | |
| Bormann et al. | Towards automated order picking robots for warehouses and retail | |
| US20190354924A1 (en) | Work support device and work support method | |
| CN106887006B (en) | Method, device and machine sorting system for identifying stacked objects | |
| EP3611117A1 (en) | Article management assistance device, article management assistance system, and article management assistance method | |
| CN113264302A (en) | Control method and device of warehousing robot, robot and warehousing system | |
| CN114013751B (en) | Rectangular article boxing method, device, electronic device and storage medium | |
| Li et al. | Using Kinect for monitoring warehouse order picking operations | |
| CN115249324A (en) | Method and device for determining position to be stacked in stack shape and computing equipment | |
| US12017368B2 (en) | Mix-size depalletizing | |
| CN113111899A (en) | Object recognition or object registration method based on image classification and computing system | |
| US20230121334A1 (en) | Method and System for Efficiently Packing a Transport Container with Items Picked from a Transport Structure | |
| CN111191083B (en) | Method and computing system for object identification | |
| US11900652B2 (en) | Method and computing system for generating a safety volume list for object detection | |
| CN112288038B (en) | Object recognition or object registration method based on image classification and computing system | |
| JP2014174628A (en) | Image recognition method | |
| CN115445963A (en) | Disordered wire sorting method, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |