Object grasping method, electronic device, and storage medium

Abstract

The object grabbing method, the electronic device and the storage medium provided by the present disclosure, the method is applied to an electronic device, the position information of all objects to be grabbed in the current grabbing layer is obtained, and whether the first offset degree of the plurality of objects to be grabbed arranged continuously exceeds the first offset degree threshold is determined according to the position information. After that, the plurality of objects to be grabbed arranged continuously whose first offset degree does not exceed the first offset degree threshold is determined as a combination to be grabbed. In the technical solution, the combination capable of being grabbed is judged first, that is, the objects to be grabbed in the combination are close in position, which can improve the success rate of grabbing and improve the grabbing efficiency of the objects to be grabbed.

Description

Technical Field

[0001] This disclosure relates to the field of vision technology for electronic devices, and more particularly to a method for grasping objects, an electronic device, and a storage medium. Background Technology

[0002] In cardboard box grasping scenarios, robots often use cameras and other devices to take pictures of the cardboard boxes to determine whether they can be grasped. Then, they control the suction cups to grasp the cardboard boxes and place them at the target location.

[0003] In existing technologies, when gripping cardboard boxes, only one cardboard box can be gripped at a time. The simple method is to determine whether the size of the cardboard box is covered by the suction cup. If it is covered, it is gripped directly.

[0004] However, in the above implementation, the number of cardboard boxes grabbed each time is relatively small, which affects the grabbing efficiency. Therefore, how to improve the grabbing efficiency of cardboard boxes and other objects has become an urgent technical problem to be solved. Summary of the Invention

[0005] This disclosure provides a method for grasping objects, an electronic device, and a storage medium to solve the technical problem of how to improve the grasping efficiency of objects such as cardboard boxes.

[0006] Firstly, this disclosure provides a method for grasping an object, including:

[0007] Obtain the position information of all objects to be grabbed in the current grabbing layer;

[0008] Based on the location information, determine whether the first offset of the multiple consecutively arranged objects to be grasped exceeds the first offset threshold.

[0009] Multiple objects to be grasped that are arranged consecutively and whose first bias does not exceed the first bias threshold are identified as a grasping combination.

[0010] In one or more embodiments, the first offset includes a Z-axis deflection parameter, and the first offset threshold includes an angle deflection threshold.

[0011] Accordingly, obtaining the position information of all objects to be grasped in the current grasping layer includes:

[0012] Obtain the center pose of each object to be grasped in the current grasping layer;

[0013] Accordingly, determining whether the first offset of the multiple consecutively arranged objects to be grasped exceeds the first offset threshold based on the location information includes:

[0014] Accordingly, based on the center pose of the object to be grasped, starting from the initial object to be grasped among the multiple objects to be grasped arranged in succession, it is determined whether the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the angle deflection threshold.

[0015] If the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the angle deflection threshold, then continue to determine the Z-axis deflection parameter between the initial object to be grasped and the next consecutively arranged objects to be grasped in the combination direction; otherwise, stop.

[0016] Accordingly, determining a group of consecutively arranged objects to be grasped whose first bias does not exceed the first bias threshold as a grasping combination includes:

[0017] The initial object to be grasped and all objects to be grasped whose Z-axis deflection parameters in the combined direction do not exceed the angle deflection threshold are identified as the grasping combination.

[0018] In one or more embodiments, the first bias degree includes a first bias parameter or a second bias parameter, and correspondingly, the first bias degree threshold includes a first preset threshold or a second preset threshold. The first bias parameter is the distance value of two adjacent objects to be grasped in the combination direction that are biased out perpendicular to the combination direction, and the second bias parameter is the distance value between two adjacent objects to be grasped in the combination direction.

[0019] Accordingly, obtaining the position information of all objects to be grasped in the current grasping layer includes:

[0020] Obtain the center pose of each object to be grasped in the current grasping layer;

[0021] Accordingly, determining whether the first offset of the multiple consecutively arranged objects to be grasped exceeds the first offset threshold based on the location information includes:

[0022] Based on the center pose of the object to be grasped, starting from the initial object to be grasped among a plurality of continuously arranged objects to be grasped, it is determined whether the first offset parameter or the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the first preset threshold or the second preset threshold.

[0023] If the first offset parameter or the second offset parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the first preset threshold or the second preset threshold, then continue to determine the first offset parameter or the second offset parameter between the current object to be grasped that is adjacent to the starting object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction; otherwise, stop.

[0024] Accordingly, determining a group of consecutively arranged objects to be grasped whose first bias does not exceed the first bias threshold as a grasping combination includes:

[0025] The initial object to be grasped and all objects to be grasped whose first or second bias parameter in the combination direction does not exceed the first or second preset threshold are identified as the grasping combination.

[0026] In one or more embodiments, the location information includes the center pose of the object to be grasped, the first offset degree includes a first offset parameter and a second offset parameter, and correspondingly, the first offset degree threshold includes a first preset threshold and a second preset threshold, the first offset parameter is the distance value of two adjacent objects to be grasped in the combination direction that are offset perpendicular to the combination direction, and the second offset parameter is the distance value between two adjacent objects to be grasped in the combination direction.

[0027] Accordingly, obtaining the position information of all objects to be grasped in the current grasping layer includes:

[0028] Obtain the center pose of each object to be grasped in the current grasping layer;

[0029] Accordingly, determining whether the first offset of the multiple consecutively arranged objects to be grasped exceeds the first offset threshold based on the location information includes:

[0030] Based on the center pose of the object to be grasped, starting from the initial object to be grasped among a plurality of continuously arranged objects to be grasped, it is determined whether the first offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the first preset threshold, and whether the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the second preset threshold.

[0031] If the first offset parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the first preset threshold, and the second offset parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the second preset threshold, then continue to determine the first offset parameter and the second offset parameter between the current object to be grasped adjacent to the starting object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction; otherwise, stop.

[0032] Accordingly, determining a group of consecutively arranged objects to be grasped whose first bias does not exceed the first bias threshold as a grasping combination includes:

[0033] The initial object to be grasped and all objects to be grasped whose first bias parameter in the combination direction does not exceed the first preset threshold and whose second bias parameter does not exceed the second preset threshold are identified as the grasping combination.

[0034] In one or more embodiments, the object to be grasped is a single item or a combination of items to be grasped. When the object to be grasped is a combination of items to be grasped, obtaining the position information of all objects to be grasped in the current grasping layer includes:

[0035] Obtain the center pose of the entire group of items to be grabbed in the current grabbing layer;

[0036] Accordingly, determining whether the first offset of the multiple consecutively arranged objects to be grasped exceeds the first offset threshold based on the location information includes:

[0037] Based on the overall center pose of the combination of items to be grabbed, determine whether the first offset of the multiple consecutively arranged combinations of items to be grabbed exceeds the first offset threshold.

[0038] In one or more embodiments, before obtaining the center pose of the entire assembly of items to be grasped in the current grasping layer, the method further includes:

[0039] Obtain the position information of each item to be grabbed in the current grabbing layer;

[0040] Based on the position information of each of the items to be grabbed, determine whether the second offset of the multiple consecutively arranged items to be grabbed exceeds the second offset threshold.

[0041] A group of consecutively arranged items to be grabbed, whose second bias does not exceed the second bias threshold, is defined as the combination of items to be grabbed.

[0042] In one or more embodiments, the second offset includes at least one of the Z-axis deflection parameter of the object to be grasped, a first offset parameter, and a second offset parameter, wherein the first offset parameter is the distance value of two adjacent objects to be grasped in the combined direction that are offset perpendicular to the combined direction, and the second offset parameter is the distance value between two adjacent objects to be grasped in the combined direction.

[0043] In one or more embodiments, determining whether the first offset of a plurality of consecutively arranged objects to be grasped exceeds a first offset threshold based on the location information includes:

[0044] Identify the two sets of objects to be grasped that are currently compared by the first bias.

[0045] Based on the position information corresponding to the two sets of objects to be grasped, determine whether the first offset between the two sets of objects to be grasped exceeds the first offset threshold.

[0046] If the first offset between the two groups of objects to be grasped does not exceed the first offset threshold, then determine whether the total number of objects to be grasped whose first offset does not exceed the first offset threshold has reached the set threshold.

[0047] If the total number of objects to be grasped whose first bias does not exceed the first bias threshold does not reach the set threshold, then the next two groups of objects to be grasped will be determined for comparison of the bias, wherein the set threshold is less than or equal to the maximum grasping amount of the object grasping structure.

[0048] In a second aspect, this disclosure provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0049] The memory stores computer-executed instructions;

[0050] The processor executes computer execution instructions stored in the memory to implement the method as described in the first aspect or any of the above methods.

[0051] Thirdly, this disclosure provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in the first aspect or any of the above methods.

[0052] Fourthly, this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect or any of the above methods.

[0053] The object grasping method, electronic device, and storage medium disclosed herein acquire position information of all objects to be grasped in the current grasping layer. Based on the position information, the method determines whether a first offset of multiple consecutively arranged objects to be grasped exceeds a first offset threshold. Then, multiple consecutively arranged objects to be grasped whose first offset does not exceed the first offset threshold are identified as a grasping combination. In this technical solution, the ability to grasp combinations is first determined, ensuring that the objects to be grasped in the combination are geographically close, which improves the success rate of grasping and thus increases the grasping efficiency of the objects to be grasped. Attached Figure Description

[0054] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0055] Figure 1 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 1 ;

[0056] Figure 2 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 2 ;

[0057] Figure 3 This is a schematic diagram illustrating the determination of Z-axis deflection parameters provided in the disclosed embodiments;

[0058] Figure 4 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 3 ;

[0059] Figure 5 A schematic diagram illustrating the determination of the first bias parameter provided in the disclosed embodiments;

[0060] Figure 6 A schematic diagram illustrating the determination of the second bias parameter provided in the disclosed embodiments;

[0061] Figure 7 A schematic diagram illustrating the determination of the first bias parameter and the second bias parameter provided in the disclosed embodiments;

[0062] Figure 8 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 4 ;

[0063] Figure 9 A schematic diagram illustrating the determination of whether the object to be grasped is a combination of items to be grasped, provided in the disclosed embodiment;

[0064] Figure 10 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 5 ;

[0065] Figure 11 A schematic diagram illustrating the determination of two sets of objects to be grasped, provided for a disclosed embodiment;

[0066] Figure 12 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 6 ;

[0067] Figure 13 A schematic diagram showing the placement of 10 boxes according to an embodiment of this disclosure;

[0068] Figure 14 A schematic diagram of the structure of the object grasping device provided in the embodiments of this disclosure;

[0069] Figure 15 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure.

[0070] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0072] Before introducing the embodiments of this disclosure, the application background of the embodiments of this disclosure will first be explained:

[0073] When destacking and stacking existing cardboard boxes, only one cardboard box can be grabbed and moved at a time. Specifically, it is simple to determine whether the size of the cardboard box is covered by the suction cup. If it is covered, it is grabbed directly.

[0074] The above-mentioned method has extremely low grasping efficiency. When a large number of boxes need to be unpacked and stacked, grasping them one by one results in high time and power costs. Therefore, in view of the technical problems existing in the current technology, how to improve the grasping efficiency of objects such as cardboard boxes is a key issue.

[0075] To address the technical problems existing in the prior art, the inventors of this disclosure have the following concept: In the case of multiple stacked objects that need to be grasped and moved in a destacking and palletizing scenario, the actual placement of objects in each layer of the stack can be used to determine multiple groups of objects that can be grasped simultaneously for each layer. Each group of objects can be one or more. Then, each group of objects is treated as a combination that can be grasped and moved at one time, so that the fixture can reliably grasp multiple objects at one time. That is, while ensuring the success rate of object grasping, the efficiency of destacking and palletizing is optimized by increasing the number of objects grasped by the fixture at one time.

[0076] The technical solutions of this disclosure will now be described in detail through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0077] It is worth noting that the application fields of the object grasping method, electronic device and storage medium disclosed herein are not limited.

[0078] In this disclosure, the executing entity is an electronic device, which may be a control unit in the electronic device or a controller that controls the electronic device. The electronic device may be a robot, a computing device, a mobile phone, a computer, or other such devices.

[0079] Figure 1 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 1 ,like Figure 1 As shown, the method for grasping this object may include the following steps:

[0080] Step 11: Obtain the position information of all objects to be grabbed in the current grabbing layer.

[0081] In this step, before determining the combination of objects to be grabbed that can be grabbed at once, it is necessary to first obtain the position information of each object to be grabbed in the current grabbing layer.

[0082] Optionally, the position information of the object to be grasped can be the point cloud pose obtained by the camera taking pictures of the scene including the object to be grasped and processing the acquired point cloud data; it can also be input by the user; or it can be obtained by the camera taking pictures and then calibrating with a bounding box.

[0083] It should be understood that there are other ways to obtain the position information of the object being grasped besides the examples above.

[0084] Optionally, the position information of the object to be grasped can represent the placement parameters of the object in the actual scene, including physical location, spatial occupancy, etc.

[0085] Step 12: Based on the location information, determine whether the first offset of multiple consecutively arranged objects to be grabbed exceeds the first offset threshold.

[0086] In this step, based on the position information of each object to be grabbed, the first offset of the multiple objects to be grabbed arranged in succession is determined. After each first offset is determined, it is determined whether the first offset exceeds the first offset threshold.

[0087] Optionally, the first offset of the object to be grasped includes at least one of the following: Z-axis deflection parameter, first offset parameter, and second offset parameter.

[0088] Among them, the Z-axis deflection parameter represents the rotation angle of the object to be grasped relative to the reference object to be grasped along the Z-axis. For example, when the long sides of the two objects to be grasped are parallel, the rotation angle of one of the objects to be grasped along the Z-axis is used. The first offset parameter is the distance value of the two adjacent objects to be grasped in the combination direction that are offset perpendicular to the combination direction. The second offset parameter is the distance value between the two adjacent objects to be grasped in the combination direction.

[0089] That is, the first offset can be determined based on the pose information in the position information. It should be understood that the first offset can represent the distance between the objects to be grasped, the offset value in the combination direction, and the difference in the placement angle, etc.

[0090] Step 13: Determine the combination of multiple objects to be grasped that are arranged continuously with the first bias degree not exceeding the first bias degree threshold as the grasping combination.

[0091] In this step, during the above process, it is continuously determined whether the first offset of the multiple consecutively arranged objects to be grasped exceeds the first offset threshold. Then, the multiple consecutively arranged objects to be grasped whose first offset does not exceed the first offset threshold are determined as the grasping combination.

[0092] For example, if the first bias threshold is 0.5, the objects to be grabbed can be box 1, box 2, box 3, box 4, and box 5. Taking box 1 as the starting position, the bias of box 2 is determined to be 0.4, the bias of box 3 is 0.45, and the bias of box 4 is 0.6. Then, when the bias of box 4 is greater than 0.5, the judgment stops, and box 1, box 2, and box 3 are taken as the objects to be grabbed in the grabbing combination.

[0093] The object grasping method provided in this disclosure obtains the position information of all objects to be grasped in the current grasping layer, and determines whether the first offset of multiple consecutively arranged objects to be grasped exceeds a first offset threshold based on the position information. Then, multiple consecutively arranged objects to be grasped whose first offset does not exceed the first offset threshold are identified as a grasping combination. This technical solution first determines which combinations can be grasped, ensuring that the objects to be grasped in the combination are geographically close, which improves the grasping success rate and thus increases the grasping efficiency of the objects to be grasped.

[0094] Based on the above embodiments, the first offset degree includes the Z-axis deflection parameter, and the first offset degree threshold includes the angle deflection threshold. Figure 2 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 2 ,like Figure 2 As shown, steps 11-13 above can each include the following implementation steps:

[0095] in, Figure 3 This is a schematic diagram illustrating the determination of Z-axis deflection parameters provided in a disclosed embodiment. Figure 3 As shown, taking two adjacent objects among multiple objects to be grasped as an example, the center pose of object 201 is A and the center pose of object 202 is B; the Z-axis deflection parameter can be understood as: for example, with the center of the two objects to be grasped as the origin, and the length and width as the X-axis and Y-axis of the coordinate system respectively, the minimum angle value when one of the objects to be grasped is rotated along its own Y-axis to be parallel to the X-axis and Y-axis of the other object to be grasped.

[0096] Step 21: Obtain the center pose of each object to be grasped in the current grasping layer;

[0097] Optionally, in the embodiments involved in this solution, in addition to the center pose, the comparison can also be the poses of the same position in the two objects to be grasped, for example, the poses of the corners of the lower left edges of the two objects to be grasped, which are implemented in the same way.

[0098] Step 22: Based on the center pose of the object to be grasped, starting from the initial object to be grasped among the multiple objects to be grasped arranged in succession, determine whether the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combined direction exceeds the angle deflection threshold.

[0099] In this step, starting from the initial object to be grasped, along the combination direction, the Z-axis deflection parameters between other objects to be grasped and the initial object to be grasped are determined one by one, and it is determined whether the corresponding Z-axis deflection parameters exceed the angle deflection threshold.

[0100] Optionally, the combination direction can be the general direction in which a row or column of objects to be grabbed is arranged. In real-world scenarios, Figure 3 The diagram shows one direction. If there is also an object to be grabbed on the right side of object 201, the combined direction can also be to the right side of object 201. The principle is similar for other sides.

[0101] Optionally, the starting object to be grabbed can be the object to be grabbed at the default position in the current grabbing layer, the object to be grabbed can be set by the user, or the object to be grabbed at a certain edge point or center point of the current grabbing layer.

[0102] Step 23: If the Z-axis deflection parameter between the starting object to be grabbed and its adjacent objects to be grabbed in the combination direction does not exceed the angle deflection threshold, then continue to determine the Z-axis deflection parameter between the starting object to be grabbed and the next consecutively arranged objects to be grabbed in the combination direction; otherwise, stop.

[0103] In this step, during the sequential judgment, if the Z-axis deflection parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the angle deflection threshold, then the judgment continues to be made on the Z-axis deflection parameter between the starting object to be grasped and the adjacent objects to be grasped in the previous object to be grasped in the combination direction; if the Z-axis deflection parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the angle deflection threshold, then the judgment stops.

[0104] For example, if the angle deflection threshold is 15 degrees and the Z-axis deflection parameter of object 201 and object 202 is 10 degrees, then the Z-axis deflection parameter of subsequent object 203 and object 201 will continue to be judged; if the Z-axis deflection parameter of object 201 and object 202 is 16 degrees, then the judgment will stop.

[0105] Step 24: Determine the starting object to be grabbed and all objects to be grabbed whose Z-axis deflection parameters in the combination direction do not exceed the angle deflection threshold as the grabbing combination.

[0106] In this step, the initial object to be grabbed and all objects to be grabbed whose Z-axis deflection parameters in the combined direction do not exceed the angle deflection threshold are identified as objects to be grabbed in the grabbing combination.

[0107] For example, when the Z-axis deflection parameter of object 201 and object 202 is 10 degrees, and the Z-axis deflection parameter of object 203 and object 201 is determined to be 19 degrees, then object 201 and object 202 are identified as objects to be grasped in the grasping combination.

[0108] It should be understood that the X, Y, and Z directions shown in this scheme are only for illustrative purposes and do not restrict the specific directions. The directions are explained in conjunction with the attached diagram. As long as the directions of the parameters being compared are the actual directions defined in this scheme, the names of the X, Y, and Z directions can be adjusted according to the defined directions of the coordinate system.

[0109] The object grasping method provided in this embodiment obtains the center pose of each object to be grasped in the current grasping layer. Based on the center pose of the object to be grasped, starting from the initial object to be grasped among a plurality of continuously arranged objects to be grasped, it determines whether the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the angle deflection threshold. If the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the angle deflection threshold, the method continues to determine the Z-axis deflection parameter between the initial object to be grasped and the next continuously arranged objects to be grasped in the combination direction; otherwise, it stops. Finally, the initial object to be grasped and all objects to be grasped whose Z-axis deflection parameter in the combination direction does not exceed the angle deflection threshold are determined as a grasping combination. In this technical solution, by determining the placement angle between the initial object to be grasped and the object to be grasped that needs to be judged, when the placement angle meets the placement requirements (i.e., when the placement angle is less than the threshold required based on the actual scenario), multiple objects to be grasped that meet the placement angle requirements are arranged consecutively as a combination that can be grasped, so as to ensure the success rate of subsequent grasping.

[0110] Based on the above embodiments, the bias degree includes a first bias parameter and / or a second bias parameter, and correspondingly, the first bias degree threshold includes a first preset threshold and / or a second preset threshold; then Figure 4 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 3 ,like Figure 4 As shown, steps 11-13 above can each include the following implementation steps:

[0111] It should be understood that this embodiment includes three cases, namely, the first bias degree includes: a first bias parameter; a second bias parameter; and a first bias parameter and a second bias parameter; wherein, the bias parameter can represent the distance between objects or the offset value between objects in any direction, depending on the scene.

[0112] Right now, Figure 5 A schematic diagram illustrating the determination of the first bias parameter provided in the disclosed embodiments; Figure 6 A schematic diagram illustrating the determination of the second bias parameter provided in the disclosed embodiments; Figure 7 The second bias parameter and the second bias parameter determination diagram provided in the disclosed embodiment will be explained in the following steps in conjunction with the corresponding determination diagram, wherein the initial object to be grasped can be object 201.

[0113] Step 41: Obtain the center pose of each object to be grabbed in the current grabbing layer;

[0114] Step 42: Based on the center pose of the object to be grasped, starting from the initial object to be grasped among the multiple objects to be grasped arranged in a continuous manner, determine whether the first offset parameter and / or the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceed the first preset threshold and / or the second preset threshold.

[0115] The first offset parameter is the distance between two adjacent objects to be grasped in the combination direction, which is perpendicular to the combination direction. The second offset parameter is the distance between two adjacent objects to be grasped in the combination direction.

[0116] In this step, there are three possible implementations depending on the scenario requirements: First, based on the center pose of the object to be grasped, starting from the initial object among the multiple continuously arranged objects to be grasped, determine whether the first offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combined direction exceeds a first preset threshold; Second, based on the center pose of the object to be grasped, starting from the initial object among the multiple continuously arranged objects to be grasped, determine whether the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combined direction exceeds a second preset threshold; Third, based on the center pose of the object to be grasped, starting from the initial object among the multiple continuously arranged objects to be grasped, determine whether the first offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combined direction exceeds the first preset threshold, and whether the second offset parameter exceeds the second preset threshold.

[0117] Optionally, depending on the actual needs, different judgment methods can be selected. The following explanation uses the combination direction along the Y-axis as an example (it should be understood that if the combination direction is the X-axis, the implementation principle is similar. The X and Y directions mentioned above are only for the purpose of explaining this solution. In actual scenarios, the X and Y directions can also be named according to the definition of the coordinate system):

[0118] For example, such as Figure 5 As shown, the first preset threshold is 6, and along the combination direction, the first offset parameter between object 202 and object 201 is 5;

[0119] For example, such as Figure 6 As shown, the second preset threshold is 5, and along the combination direction, the second offset parameter between object 202 and object 201 is 4;

[0120] For example, such as Figure 7As shown, the first preset threshold is 6, the second preset threshold is 5, and along the combination direction, the second offset parameter between object 202 and object 201 is 4, and the first offset parameter between object 202 and object 201 is 5 (in this implementation, the order of judging the first offset parameter and the second offset parameter is arbitrary, or they can be judged simultaneously).

[0121] The center pose of object 201 is (5,4), and the center pose of object 202 is (10,12). It should be understood that the second bias parameter is the center distance between object 201 and object 202 minus half of the side length (e.g., 4) of their respective combined directions, i.e., the second bias parameter is 4; the first bias parameter is the center distance between object 201 and object 202 on the X-axis, i.e., the first bias parameter is 5.

[0122] Step 43: If the first offset parameter or the second offset parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the first preset threshold and / or the second preset threshold, then continue to determine the first offset parameter and / or the second offset parameter between the current object to be grasped adjacent to the starting object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction; otherwise, stop.

[0123] Following the example above, such as Figure 5 As shown, the first bias parameter between object 202 and object 201 is 5, which is less than the first preset threshold 6. Then, the first bias parameter between the next object and object 201 in the combination direction is judged and compared. When it is determined that the first bias parameter exceeds the first preset threshold 6, the judgment is stopped.

[0124] like Figure 6 As shown, the second bias parameter between object 202 and object 201 is 4, which is less than the second preset threshold 5. Then, the second bias parameter between the next object and object 201 in the combination direction is judged and compared. When it is determined that the second bias parameter exceeds the second preset threshold 5, the judgment is stopped.

[0125] like Figure 7 As shown, the first bias parameter between object 202 and object 201 is 5, and the second bias parameter between object 202 and object 201 is 4. 5 is less than the first preset threshold 6, and 4 is less than the second preset threshold 5. Then, the first bias parameter and the second bias parameter between the next object and object 201 in the combination direction are judged and compared. When it is determined that the first bias parameter exceeds the first preset threshold 6 or the second bias parameter exceeds the second preset threshold 5, the judgment is stopped.

[0126] It should be understood that the order of judgment for the first bias parameter and the second bias parameter can be either sequential or simultaneous.

[0127] Step 44: Determine the starting object to be grasped and all objects to be grasped whose first bias parameter and / or second bias parameter in the combination direction do not exceed the first preset threshold and / or the second preset threshold as the grasping combination.

[0128] In this step, under the different offset values ​​mentioned above, the following implementations are corresponding: First: all objects to be grasped, including the initial object to be grasped and all objects to be grasped whose first offset parameter in the combination direction does not exceed the first preset threshold, are determined as a grasping combination; Second: all objects to be grasped, including the initial object to be grasped and all objects to be grasped whose second offset parameter in the combination direction does not exceed the second preset threshold, are determined as a grasping combination; Third: all objects to be grasped, including the initial object to be grasped and all objects to be grasped whose first offset parameter in the combination direction does not exceed the first preset threshold and whose second offset parameter does not exceed the second preset threshold, are determined as a grasping combination.

[0129] The object grasping method provided in this embodiment obtains the center pose of each object to be grasped in the current grasping layer, and based on the center pose of the object to be grasped, starting from the initial object to be grasped among a plurality of continuously arranged objects to be grasped, determines whether the first offset parameter and / or the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds a first preset threshold and / or a second preset threshold. If the first offset parameter or the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the first preset threshold and / or the second preset threshold, then the method continues to determine the first offset parameter and / or the second offset parameter between the object to be grasped currently adjacent to the initial object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction; otherwise, the method stops. Finally, the initial object to be grasped and all objects to be grasped whose first offset parameter and / or the second offset parameter in the combination direction does not exceed the first preset threshold and / or the second preset threshold are determined as a grasping combination. In this technical solution, multiple objects that meet the placement requirements of the corresponding scenario are identified by judging the spacing and / or deviation value between objects in different scenarios. The objects that meet the requirements are then used as a combination to be grabbed to ensure the success rate of subsequent grabbing.

[0130] Based on the above embodiments, the object to be grasped can be a single item or a combination of items. In the case where the object to be grasped is a combination of items, then... Figure 8 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 4 ,like Figure 8 As shown, steps 11-13 above can each include the following implementation steps:

[0131] in, Figure 9 This is a schematic diagram illustrating the determination of whether the object to be grasped is a combination of items to be grasped, provided in a disclosed embodiment.

[0132] Step 81: Obtain the center pose of the entire combination of items to be grabbed in the current grabbing layer;

[0133] In this step, when the object to be grasped is a combination of items to be grasped, that is, the combination of items to be grasped includes multiple items to be grasped, the center pose of the entire combination of items to be grasped can be determined first, that is, the center pose of the area involved by the multiple items to be grasped.

[0134] For example, such as Figure 9 As shown, the center pose C of the combination of items to be grabbed 901 and the center pose D of the combination of items to be grabbed 902.

[0135] It should be understood that before step 81, it is necessary to determine multiple items to be grabbed as a combination of items to be grabbed, which includes the following steps: obtaining the position information of each item to be grabbed in the current grabbing layer; determining whether the second offset of multiple consecutively arranged items to be grabbed exceeds the second offset threshold based on the position information of each item to be grabbed; determining the multiple consecutively arranged items to be grabbed whose second offset does not exceed the second offset threshold as a combination of items to be grabbed, wherein the second offset includes at least one of a first offset parameter, a second offset parameter, and a Z-axis deflection parameter, the first offset parameter being the distance value of two adjacent items to be grabbed in the combination direction perpendicular to the combination direction, and the second offset parameter being the distance value between two adjacent items to be grabbed in the combination direction. The implementation method before step 81 is the same as or similar to the technical solution and effect of the above embodiment, and will not be repeated here.

[0136] The parameter types for the first bias and the second bias can be the same or different. Furthermore, if the parameter types are the same, the values ​​for the first bias threshold and the second bias threshold can be set identically or differently based on the actual data capture requirements.

[0137] Step 82: Based on the overall center pose of the combination of objects to be grabbed, determine whether the first offset of the multiple continuously arranged objects to be grabbed exceeds the first offset threshold.

[0138] In this step, after determining the overall center pose of the current combination of items to be grabbed, the overall center pose of the combination of items to be grabbed is compared with the overall center pose of the next adjacent combination of items to be grabbed to determine the corresponding first offset. Then, the first offset is compared with the first offset threshold to determine whether it exceeds the first offset threshold.

[0139] Optionally, this step can be implemented as described in the above embodiments. The first offset can include at least one of the Z-axis deflection parameter, the first offset parameter, and the second offset parameter. Correspondingly, the first offset threshold can also include at least one of the angle deflection threshold, the first preset threshold, and the second preset threshold. The technical solutions for their implementation are similar and will not be described in detail here.

[0140] For example, at least one of the Z-axis deflection parameter, the first offset parameter, and the second offset parameter between the center pose C of the object assembly 901 and the center pose D of the object assembly 902.

[0141] Furthermore, combinations of items to be grabbed that do not exceed the first bias threshold are also considered as objects to be grabbed in the grabbing combinations.

[0142] The object grasping method provided in this disclosure involves determining whether the object to be grasped is a single item or a combination of items. When the object is a combination of items, the method obtains the center pose of the entire combination of items in the current grasping layer. Based on the center pose of the combination of items, it determines whether the first offset of multiple consecutively arranged objects exceeds a first offset threshold. This technical solution achieves the goal of ensuring that multiple consecutively arranged combinations of items meet the corresponding requirements, significantly improving the success rate and efficiency of grasping multiple boxes in subsequent operations.

[0143] Based on the above embodiments, Figure 10 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 5 ,like Figure 10 As shown, step 12 above may include the following steps:

[0144] in, Figure 11 This is a schematic diagram illustrating the determination of two sets of objects to be grasped, provided for a disclosed embodiment.

[0145] Step 101: Determine the two sets of objects to be grasped that are currently being compared based on the first bias.

[0146] This step involves determining whether two or more groups of objects to be grasped can be grasped simultaneously. First, the two groups of objects to be grasped are identified and compared based on the first bias.

[0147] In one implementation, multiple groups of objects to be grasped are determined in the above combination direction. Each group of objects to be grasped can be an object to be grasped in the combination to be grasped. At this time, two adjacent groups of objects to be grasped are selected as the two groups of objects to be grasped for comparison of the first bias.

[0148] For example, it could be Figure 11 The first group of objects to be grabbed and the second group of objects to be grabbed are shown.

[0149] Step 102: Based on the position information corresponding to the two sets of objects to be grasped, determine whether the first offset between the two sets of objects to be grasped exceeds the first offset threshold.

[0150] In this step, after determining the two sets of objects to be grasped for comparison of the first bias, the first bias between the two sets of objects to be grasped is determined based on the position information corresponding to the two sets of objects to be grasped, and then it is determined whether the bias exceeds the first bias threshold.

[0151] Optionally, the position information corresponding to the two sets of objects to be grasped can be the overall center pose of the two sets of objects to be grasped, and it can be determined whether the offset between the center poses exceeds the first offset threshold.

[0152] Similar to the above embodiments, the first offset degree may include at least one of the following: Z-axis deflection parameter, first offset parameter and second offset parameter; correspondingly, the first offset degree threshold may also include at least one of the following: angle deflection threshold, first preset threshold and second preset threshold, which will not be elaborated here.

[0153] Step 103: If the first offset between the two groups of objects to be grabbed does not exceed the first offset threshold, then determine whether the total number of objects to be grabbed whose first offset does not exceed the first offset threshold has reached the set threshold.

[0154] In this step, the threshold can be set by the user according to actual needs; for example, the user may want to grab 10 objects at a time.

[0155] Once it is determined that the first offset between the two groups of objects to be grasped does not exceed the first offset threshold, the total number of objects to be grasped whose first offset does not exceed the first offset threshold can be further determined, and this total number can be compared with the set threshold.

[0156] Step 104: If the total number of objects to be grasped whose current first bias does not exceed the first bias threshold does not reach the set threshold, then continue to determine the next two sets of objects to be grasped with different biases.

[0157] Among them, the threshold is set to be less than or equal to the maximum amount of the object grasping structure.

[0158] In this step, if the total number of objects to be grasped whose current first bias does not exceed the first bias threshold does not reach the set threshold, then the next two sets of objects to be grasped with different biases can be determined, and the above process can be repeated.

[0159] Optionally, the next two sets of objects to be grasped for comparing the bias can be the two sets of objects to be grasped other than those involved in step 101 above, or the third set of objects to be grasped that is adjacent to the second set of objects to be grasped in the combination direction, that is, to determine the first bias between the second set of objects to be grasped and the third set of objects to be grasped.

[0160] Optionally, since the object grasping structure has limited grasping capacity, that is, the object grasping structure has a maximum grasping capacity.

[0161] The maximum gripping capacity is related to factors such as the size of the object gripping structure and the suction force of the suction cup.

[0162] For example, taking a suction cup as the object grasping structure, the suction cup can only grasp all the objects in the grasping combination if the difference between the length of the area involved in the grasping combination and the length of the suction cup is less than a first threshold, and the difference between the width of the area involved in the grasping combination and the width of the suction cup is less than a second threshold.

[0163] In practice, the number of objects grabbed at one time is less than or equal to the maximum number of objects that can be grabbed by the object grabbing structure. In other words, the threshold can be set to be less than or equal to the maximum number of objects that can be grabbed by the object grabbing structure.

[0164] In other implementations, the threshold can be set to be less than or equal to the maximum number of objects to be grasped in the object grasping structure without limitation. The user sets how many objects to grasp, and then the corresponding number of objects to be grasped are grasped from the grasping combination. The grasping combination that meets the requirements can be determined in advance.

[0165] The object grasping method provided in this embodiment determines two sets of objects to be grasped based on their current offset. Then, based on the position information of the two sets of objects, it determines whether the first offset between them exceeds a first offset threshold. If the first offset does not exceed the first offset threshold, it determines whether the total number of objects whose first offset does not exceed the first offset threshold reaches a set threshold. Finally, if the total number of objects whose first offset does not exceed the first offset threshold does not reach the set threshold, it continues to determine the next two sets of objects to be grasped based on their offset. The set threshold is less than or equal to the maximum grasping capacity of the object grasping structure. This technical solution, by comparing the offsets of the objects to be grasped between two sets, enables the grasping determination of multiple rows and columns of objects, improving the grasping efficiency and success rate of subsequent objects.

[0166] Based on the above embodiments, Figure 12 Flowchart of the object grasping method provided in the embodiments of this disclosure Figure 6 ,like Figure 12 As shown, step 12 above may include the following steps:

[0167] in, Figure 13 This is a schematic diagram showing the placement of the 10 boxes provided in an embodiment of the present disclosure, as shown below. Figure 12 As shown, the boxes include: Box 1, Box 2, Box 3, Box 4, Box 5, Box 6, Box 7, Box 8, Box 9, and Box 10.

[0168] Step 1: Select starting box 1 (After the judgment of starting from box 1 is completed, it is necessary to traverse other boxes to perform similar steps as follows);

[0169] Step 2: Along the combination direction, determine whether the first offset between box 3 and box 1 exceeds the first offset threshold.

[0170] The first offset can be at least one of the Z-axis deflection parameter, the first offset parameter, and the second offset parameter mentioned above; that is, depending on the scenario, when determining the combination to be grasped, the boxes in the combination to be grasped can be implemented in the following ways:

[0171] 1. Boxes whose Z-axis deflection parameters do not exceed the angle deflection threshold;

[0172] 2. Boxes whose first bias parameter does not exceed the first preset threshold;

[0173] 3. Boxes whose second bias parameter does not exceed the second preset threshold;

[0174] 4. Boxes whose Z-axis deflection parameters do not exceed the angle deflection threshold and whose first offset parameters do not exceed the first preset threshold;

[0175] 5. Boxes whose Z-axis deflection parameters do not exceed the angle deflection threshold and whose second offset parameters do not exceed the second preset threshold;

[0176] 6. A box whose Z-axis deflection parameter does not exceed the angle deflection threshold, whose second offset parameter does not exceed the second preset threshold, and whose first offset parameter does not exceed the first preset threshold;

[0177] 7. Boxes whose first bias parameter does not exceed the first preset threshold and whose second bias parameter does not exceed the second preset threshold.

[0178] Step 3: If the threshold is not exceeded, determine that the difference between the length of the area involved in box 3 and box 1 and the length of the suction cup is less than the first threshold, and the difference between the width of the area involved and the width of the suction cup is less than the second threshold; if the threshold is exceeded, then end.

[0179] Step 4: If both are less than the specified value, then box 3 and box 1 are confirmed as the combination to be grabbed; if any condition is not met, the process ends.

[0180] Step 5: Determine whether the offset between box 5 and box 1 or box 2 (refer to the example above for specific situations) exceeds the first offset threshold.

[0181] Step 6: If not exceeded, determine that the difference between the length of the area involved by boxes 5, 3, and 1 and the length of the suction cup is less than the first threshold, and the difference between the width of the area involved and the width of the suction cup is less than the second threshold; if exceeded, end.

[0182] Step 7: If all conditions are less than, then box 5, box 3, and box 1 are confirmed as the combination to be grabbed; if any condition is not met, then the combination to be grabbed is not updated.

[0183] Step 8: Add box 7 and repeat the above judgment process;

[0184] Step 9: You can use the box 2 that was not mentioned above as the new starting box and continue judging along the combination direction; you can also use the box 9 that was not mentioned as the new starting box and continue judging along the combination direction; or you can use any box other than the previous starting box as the new starting box.

[0185] In the above implementation, the offset between two groups of boxes can also be determined. For example, boxes 1, 3, 5, and 7 are one group; boxes 2, 4, 6, and 8 are another group; and boxes 9 and 10 are yet another group.

[0186] This possible implementation is similar to the technical solution and technical effect of the above embodiments, and will not be described again here.

[0187] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein. For details not disclosed in the apparatus embodiments of this disclosure, please refer to the embodiments of the method disclosed herein.

[0188] Figure 14 This is a schematic diagram of the structure of the object-grabbing device provided in an embodiment of this disclosure. Figure 14 As shown, the object-grabbing device is used in electronic devices and includes:

[0189] The acquisition module 141 is used to acquire the position information of all objects to be captured in the current capture layer;

[0190] The first determining module 142 is used to determine, based on the position information, whether the first offset of a plurality of consecutively arranged objects to be grasped exceeds the first offset threshold.

[0191] The second determining module 143 is used to determine a group of objects to be grasped that are arranged continuously and whose first bias does not exceed the first bias threshold as a grasping combination.

[0192] In one or more embodiments, the first offset includes a Z-axis deflection parameter, and the first offset threshold includes an angle deflection threshold.

[0193] Accordingly, module 141 is used for:

[0194] Obtain the center pose of each object to be grasped in the current grasping layer;

[0195] Accordingly, the first determining module 142 is used for:

[0196] Based on the center pose of the object to be grasped, starting from the initial object to be grasped among multiple continuously arranged objects to be grasped, determine whether the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combined direction exceeds the angle deflection threshold.

[0197] If the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the angle deflection threshold, then continue to determine the Z-axis deflection parameter between the initial object to be grasped and the next consecutively arranged objects to be grasped in the combination direction; otherwise, stop.

[0198] Accordingly, the second determining module 143 is used for:

[0199] The initial object to be grabbed and all objects to be grabbed whose Z-axis deflection parameters in the combination direction do not exceed the angle deflection threshold are identified as the combination to be grabbed.

[0200] In one or more embodiments, the first bias degree includes a first bias parameter or a second bias parameter, and correspondingly, the first bias degree threshold includes a first preset threshold or a second preset threshold. The first bias parameter is the distance value of two adjacent objects to be grasped in the combination direction that are offset perpendicular to the combination direction, and the second bias parameter is the distance value between two adjacent objects to be grasped in the combination direction.

[0201] Accordingly, the acquisition module 141 is used for:

[0202] Obtain the center pose of each object to be grasped in the current grasping layer;

[0203] Accordingly, the first determining module 142 is used for:

[0204] Based on the center pose of the object to be grasped, starting from the initial object to be grasped among a series of continuously arranged objects to be grasped, it is determined whether the first offset parameter or the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combined direction exceeds the first preset threshold or the second preset threshold.

[0205] If the first offset parameter or the second offset parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the first preset threshold or the second preset threshold, then continue to determine the first offset parameter or the second offset parameter between the current object to be grasped that is adjacent to the starting object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction; otherwise, stop.

[0206] Accordingly, the second determining module 143 is used for:

[0207] The starting object to be grasped and all objects to be grasped whose first or second offset parameter in the combination direction does not exceed the first or second preset threshold are identified as the grasping combination.

[0208] In one or more embodiments, the location information includes the center pose of the object to be grasped, the first offset degree includes a first offset parameter and a second offset parameter, and correspondingly, the first offset degree threshold includes a first preset threshold and a second preset threshold, the first offset parameter is the distance value of two adjacent objects to be grasped in the combined direction that are offset perpendicular to the combined direction, and the second offset parameter is the distance value between two adjacent objects to be grasped in the combined direction.

[0209] Accordingly, module 141 is used for:

[0210] Obtain the center pose of each object to be grasped in the current grasping layer;

[0211] Accordingly, the first determining module 142 is used for:

[0212] Based on the center pose of the object to be grasped, starting from the initial object to be grasped among the multiple objects to be grasped arranged in a continuous manner, it is determined whether the first offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the first preset threshold, and whether the second offset parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds the second preset threshold.

[0213] If the first offset parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the first preset threshold, and the second offset parameter between the starting object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the second preset threshold, then continue to determine the first offset parameter and the second offset parameter between the current object to be grasped adjacent to the starting object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction; otherwise, stop.

[0214] Accordingly, the second determining module 143 is used for:

[0215] The starting object to be grasped and all objects to be grasped whose first bias parameter in the combination direction does not exceed the first preset threshold and whose second bias parameter does not exceed the second preset threshold are identified as the grasping combination.

[0216] In one or more embodiments, the object to be grasped is a single item or a combination of items. When the object to be grasped is a combination of items, the acquisition module 141 is configured to:

[0217] Obtain the center pose of the entire combination of items to be grabbed in the current grabbing layer;

[0218] Accordingly, the first determining module 142 is used for:

[0219] Based on the overall center pose of the combination of items to be grabbed, determine whether the first offset of multiple consecutively arranged combinations of items to be grabbed exceeds the first offset threshold.

[0220] In one or more embodiments, before obtaining the center pose of the entire group of items to be grabbed in the current grabbing layer, the acquisition module 141 is further configured to:

[0221] Obtain the location information of each item to be grabbed in the current grabbing layer;

[0222] Based on the location information of each item to be grabbed, determine whether the second offset of multiple consecutively arranged items to be grabbed exceeds the second offset threshold.

[0223] Multiple consecutively arranged items to be grabbed that do not exceed the second bias threshold are identified as a combination of items to be grabbed.

[0224] In one or more embodiments, the second offset includes at least one of the Z-axis deflection parameter of the object to be grasped, a first offset parameter, and a second offset parameter, wherein the first offset parameter is the distance value of two adjacent objects to be grasped in the combination direction that are offset perpendicular to the combination direction, and the second offset parameter is the distance value between two adjacent objects to be grasped in the combination direction.

[0225] In one or more embodiments, the first determining module 142 is further configured to:

[0226] Identify the two sets of objects to be grasped that are currently compared by the first bias.

[0227] Based on the position information corresponding to the two sets of objects to be grasped, determine whether the first offset between the two sets of objects to be grasped exceeds the first offset threshold.

[0228] If the first offset between the two groups of objects to be grasped does not exceed the first offset threshold, then determine whether the total number of objects to be grasped whose first offset does not exceed the first offset threshold has reached the set threshold.

[0229] If the total number of objects to be grasped that do not exceed the first bias threshold does not reach the set threshold, then the next two sets of objects to be grasped with different biases will be determined, wherein the set threshold is less than or equal to the maximum grasping quantity of the object grasping structure.

[0230] The object grasping device provided in this disclosure can be used to execute the object grasping method in any of the above embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0231] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing element calls; they can be fully implemented in hardware; or some modules can be implemented by processing element calls to software, while others are implemented in hardware. Additionally, these modules can be fully or partially integrated together, or implemented independently. The processing element mentioned here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through the integrated logic circuits in the hardware of the processor element or through software instructions.

[0232] Figure 15 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this disclosure, such as... Figure 15 As shown, the electronic device may include: a processor 151, a memory 152, and computer program instructions stored in the memory 152 and executable on the processor 151, wherein the processor 151 executes the computer program instructions to implement the method provided in any of the foregoing embodiments.

[0233] Optionally, the various devices described above in this electronic device can be connected via a system bus. The memory 152 can be a separate storage unit or a storage unit integrated into the processor 151. The number of processors 151 can be one or more.

[0234] It should be understood that processor 151 can be a Central Processing Unit (CPU), or other general-purpose processor 151, digital signal processor 151 (DSP), application-specific integrated circuit (ASIC), etc. General-purpose processor 151 can be a microprocessor 151, or any conventional processor 151. The steps of the method disclosed in this disclosure can be directly manifested as execution by hardware processor 151, or execution by a combination of hardware and software modules in processor 151.

[0235] The system bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus. Memory 152 may include random access memory (RAM) 152, and may also include non-volatile memory (NVM) 152, such as at least one disk storage device 152.

[0236] All or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable memory 152. When the program is executed, it performs the steps of the above method embodiments; and the aforementioned memory 152 (storage medium) includes: read-only memory 152 (ROM), RAM, flash memory 152, hard disk, solid-state hard disk, magnetic tape, floppy disk, optical disk, and any combination thereof.

[0237] The electronic device provided in this disclosure can be used to execute the object grasping method provided in any of the above method embodiments. Its implementation principle and technical effects are similar and will not be repeated here. This disclosure also provides a computer-readable storage medium storing computer instructions. When these computer instructions are executed on a computer, the computer causes the computer to execute the object grasping method described above.

[0238] The aforementioned computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0239] Optionally, a readable storage medium can be coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Alternatively, the readable storage medium can be an integral part of the processor. Both the processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components within the device.

[0240] This disclosure also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when the at least one processor executes the computer program, it can implement the above-described object grasping method.

[0241] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for grasping an object, characterized in that, include: Obtain the center pose of each object to be grasped in the current grasping layer; Based on the center pose of the object to be grasped, starting from the initial object to be grasped among a plurality of continuously arranged objects to be grasped, it is determined whether the first offset between the initial object to be grasped and its adjacent objects to be grasped in the combined direction exceeds a first offset threshold; the first offset includes a Z-axis deflection parameter, and the first offset threshold includes an angle deflection threshold. If the Z-axis deflection parameter between the initial object to be grasped and its adjacent objects to be grasped in the combination direction does not exceed the angle deflection threshold, then continue to determine whether the Z-axis deflection parameter between the initial object to be grasped and the next consecutively arranged objects to be grasped in the combination direction exceeds the angle deflection threshold; otherwise, stop. The initial object to be grasped and all objects to be grasped whose Z-axis deflection parameters in the combined direction do not exceed the angle deflection threshold are identified as the grasping combination.

2. A method for grasping an object, characterized in that, include: Obtain the center pose of each object to be grasped in the current grasping layer; Based on the center pose of the object to be grasped, starting from the initial object among a plurality of continuously arranged objects to be grasped, it is determined whether the first offset between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds a first offset threshold; wherein, the first offset includes a first offset parameter or a second offset parameter, the first offset parameter being the distance value of two adjacent objects to be grasped in the combination direction perpendicular to the combination direction, and the second offset parameter being the distance value between two adjacent objects to be grasped in the combination direction; the first offset threshold includes a first preset threshold or a second preset threshold; If the first bias parameter or the second bias parameter does not exceed the first preset threshold or the second preset threshold, then continue to determine whether the first bias parameter or the second bias parameter between the object to be grasped currently adjacent to the starting object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction exceeds the first preset threshold or the second preset threshold; otherwise, stop. The initial object to be grasped and all objects to be grasped whose first or second bias parameter in the combination direction does not exceed the first or second preset threshold are identified as the grasping combination.

3. A method for grasping an object, characterized in that, include: Obtain the center pose of each object to be grasped in the current grasping layer; Based on the center pose of the object to be grasped, starting from the initial object among a plurality of continuously arranged objects to be grasped, it is determined whether the first offset between the initial object to be grasped and its adjacent objects to be grasped in the combination direction exceeds a first offset threshold; wherein, the first offset includes a first offset parameter and a second offset parameter, the first offset parameter being the distance value of two adjacent objects to be grasped in the combination direction perpendicular to the combination direction, and the second offset parameter being the distance value between two adjacent objects to be grasped in the combination direction; the first offset threshold includes a first preset threshold and a second preset threshold. If the first bias parameter does not exceed the first preset threshold and the second bias parameter does not exceed the second preset threshold, then continue to determine whether the first bias parameter and the second bias parameter between the current object to be grasped adjacent to the starting object to be grasped in the combination direction and its next adjacent object to be grasped in the combination direction exceed the first preset threshold and the second preset threshold; otherwise, stop. The initial object to be grasped and all objects to be grasped whose first bias parameter in the combination direction does not exceed the first preset threshold and whose second bias parameter does not exceed the second preset threshold are identified as the grasping combination.

4. The method according to any one of claims 1-3, characterized in that, The object to be grasped is a single object or a combination of objects to be grasped. When the object to be grasped is a combination of objects to be grasped, the center pose of the object to be grasped is the center pose of the combination of objects to be grasped as a whole. Accordingly, the step of determining whether the first bias exceeds the first bias threshold includes: Based on the overall center pose of the combination of items to be grabbed, determine whether the first offset of the starting combination of items to be grabbed and its adjacent combination of items to be grabbed in the combination direction exceeds the first offset threshold.

5. The method according to claim 4, characterized in that, Before obtaining the center pose of the entire group of items to be grabbed in the current grabbing layer, the method further includes: Obtain the position information of each item to be grabbed in the current grabbing layer; Based on the position information of each of the items to be grabbed, determine whether the second offset of the multiple consecutively arranged items to be grabbed exceeds the second offset threshold. A group of consecutively arranged items to be grabbed, whose second bias does not exceed the second bias threshold, is defined as the combination of items to be grabbed.

6. The method according to claim 5, characterized in that, The second offset includes at least one of the Z-axis deflection parameter of the object to be grasped, the first offset parameter, and the second offset parameter. The first offset parameter is the distance value of two adjacent objects to be grasped in the combination direction that are offset perpendicular to the combination direction. The second offset parameter is the distance value between two adjacent objects to be grasped in the combination direction.

7. The method according to any one of claims 1-3, characterized in that, The step of determining whether the first bias exceeds the first bias threshold includes: Identify the two sets of objects to be grasped that are currently compared by the first bias. Based on the position information corresponding to the two sets of objects to be grasped, determine whether the first offset between the two sets of objects to be grasped exceeds the first offset threshold. If the first offset between the two groups of objects to be grasped does not exceed the first offset threshold, then determine whether the total number of objects to be grasped whose first offset does not exceed the first offset threshold has reached the set threshold. If the total number of objects to be grasped whose first bias does not exceed the first bias threshold does not reach the set threshold, then the next two groups of objects to be grasped will be determined for comparison of the bias, wherein the set threshold is less than or equal to the maximum grasping amount of the object grasping structure.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 7.

Images