Control Method, Device and Electronic Device of Sucker Robot

Abstract

The present disclosure provides a control method, device and electronic device for a suction cup robot. The control method of the suction cup robot includes: obtaining a first point cloud of adjacent objects of a target object; simulating the opening of a target partition among a plurality of partitions; determining a grasping result of a proposed execution pose of an end effector based on the target partition and the first point cloud; selectively determining the proposed execution pose as the actual grasping pose of the end effector according to the grasping result of the proposed execution pose, wherein in the actual grasping pose, the target partition does not contact the first point cloud; controlling the end effector to move to the actual grasping pose and controlling the opening of the target partition to grasp the target object, so as to achieve fast and accurate grasping of the target object.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular, to a control method, device, and electronic device for a suction cup robot. Background Art

[0002] A suction cup robot includes: an end effector, a suction cup fixed on the end effector, a robot base, and a robotic arm installed between the end effector and the robot base. The robot base can be fixed on the ground, the robotic arm drives the end effector to move, and the end effector drives the suction cup to grasp an object.

[0003] When grasping an object, it is required that the suction cup robot drives the suction cup of the end effector to grasp the target object, and then places the grasped target object at a specified position. However, there is a problem of simultaneously grasping non-target objects, which affects the efficiency of grasping the target object. Summary of the Invention

[0004] Multiple aspects of the present disclosure provide a control method, device, and electronic device for a suction cup robot to solve the problem of low efficiency in grasping a target object currently.

[0005] In a first aspect of an embodiment of the present disclosure, a control method for a suction cup robot is provided. A suction cup is fixed on the end effector of the suction cup robot. The suction cup includes multiple partitions, and each partition is independently controlled. The control method of the suction cup robot includes: obtaining a first point cloud of adjacent objects of the target object; simulating the opening of a target partition among the multiple partitions; determining a grasping result of a proposed execution pose of the end effector based on the target partition and the first point cloud; selectively determining the proposed execution pose as an actual grasping pose of the end effector according to the grasping result of the proposed execution pose, where, in the actual grasping pose, the target partition does not contact the first point cloud; controlling the end effector to move to the actual grasping pose, and controlling the target partition to open to grasp the target object.

[0006] In a second aspect of an embodiment of the present disclosure, a control device for a suction cup robot is provided, which is used to execute the control method of the suction cup robot in the first aspect. A suction cup is fixed on the end effector of the suction cup robot. The suction cup includes multiple partitions, and each partition is independently controlled. The control device of the suction cup robot includes:

[0007] An obtaining module, configured to obtain a first point cloud of adjacent objects of the target object;

[0008] A simulation module, configured to simulate the opening of a target partition among the multiple partitions;

[0009] A first determination module, configured to determine a grasping result of a proposed execution pose of the end effector based on the target partition and the first point cloud;

[0010] A second determination module, configured to selectively determine the pose to be executed as the actual grasping pose of the end effector according to the grasping result of the pose to be executed, wherein, in the actual grasping pose, the target partition does not contact the first point cloud.

[0011] A control module, configured to control the end effector to move to the actual grasping pose and control the target partition to open to grasp the target object.

[0012] A third aspect of the embodiments of the present disclosure provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, the control method of the suction cup robot in the first aspect is implemented.

[0013] A fourth aspect of the embodiments of the present disclosure provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, the control method of the suction cup robot in the first aspect is implemented.

[0014] A fifth aspect of the embodiments of the present disclosure provides a computer program product, the program product including: a computer program, the computer program is stored in a readable storage medium, at least one processor of the electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program to enable the electronic device to execute the control method of the suction cup robot in the first aspect.

[0015] In the embodiments of the present disclosure, in the scenario of palletizing and depalletizing of a suction cup robot, by obtaining the first point cloud of the adjacent object of the target object; simulating the opening of the target partition among multiple partitions; determining the grasping result of the pose to be executed of the end effector based on the target partition and the first point cloud; selectively determining the pose to be executed as the actual grasping pose of the end effector according to the grasping result of the pose to be executed, wherein, in the actual grasping pose, the target partition does not contact the first point cloud; controlling the end effector to move to the actual grasping pose and controlling the target partition to open to grasp the target object, it is possible to achieve fast and accurate grasping of the target object. Description of the Drawings

[0016] The drawings described herein are used to provide a further understanding of the present disclosure, and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation to the present disclosure. In the drawings:

[0017] Figure 1 It is an application scenario diagram of a control method of a suction cup robot provided by an exemplary embodiment of the present disclosure;

[0018] Figure 2 It is a step flow chart of a control method of a suction cup robot provided by an exemplary embodiment of the present disclosure;

[0019] Figure 3 Schematic diagram of a simulated suction cup and a first point cloud provided by an exemplary embodiment of the present disclosure;

[0020] Figure 4 Flowchart of steps of another control method for a suction cup robot provided by an exemplary embodiment of the present disclosure;

[0021] Figure 5 Another schematic diagram of a simulated suction cup and a first point cloud provided by an exemplary embodiment of the present disclosure;

[0022] Figure 6 Schematic diagram of a second point cloud provided by an exemplary embodiment of the present disclosure;

[0023] Figure 7 Block diagram of the structure of a control device for a suction cup robot provided by an exemplary embodiment of the present disclosure;

[0024] Figure 8 Schematic diagram of the structure of an electronic device provided by an exemplary embodiment of the present disclosure. Detailed implementation manners

[0025] To make the objectives, technical solutions and advantages of the present disclosure clearer, the technical solutions of the present disclosure will be clearly and completely described below in conjunction with the specific embodiments of the present disclosure and the corresponding drawings. Apparently, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present disclosure.

[0026] In the case of unpacking and palletizing scenarios, it is necessary for a suction cup robot to drive the fixture of the end effector to grab a box and move it for unpacking and palletizing. Generally, the fixture of the end effector adopts a suction cup type fixture. This type of suction cup fixture includes at least one suction cup, and each suction cup can be set to suck or release independently, so as to suck boxes of an expected quantity or size through different suction areas. When sucking a box for unpacking and palletizing, since the order of disassembling or stacking boxes according to the determined stack type is determined, it is necessary to first determine which one or which several target boxes to suck and transfer during this suction process, and then stack the target boxes to a specified position or remove the target boxes from the specified position through the suction cup. When sucking a target box each time, other non-target boxes are required not to be sucked and moved, so as to be left for subsequent sucking. However, currently when using a suction cup to grab a target box, the suction cup may contact an adjacent box, and then grab the adjacent box as well, thereby affecting the grabbing efficiency of the target box.

[0027] Based on the above problems, the control method of the suction cup robot provided by the embodiments of the present disclosure obtains the first point cloud of the adjacent objects of the target object; simulates the opening of the target partition among multiple partitions; determines the grasping result of the planned execution pose of the end effector based on the target partition and the first point cloud; and selectively determines the planned execution pose as the actual grasping pose of the end effector according to the grasping result of the planned execution pose, wherein, in the actual grasping pose, the target partition does not contact the first point cloud; controls the end effector to move to the actual grasping pose, and controls the opening of the target partition to grasp the target object, so as to achieve fast and accurate grasping of the target object.

[0028] In this embodiment, the overall control method of the suction cup robot can be implemented by means of a server. In addition, the server executing the control method of the suction cup robot can be a cloud server. In addition, the control method of the suction cup robot can also be applied to server-side devices such as a conventional server or a server array, which is not limited herein.

[0029] In addition, an application scenario of the embodiments of the present disclosure is as Figure 1 , Figure 1 which includes: a suction cup robot 10, and the suction cup robot 10 includes: an end effector 11 and a suction cup 12 arranged on the end effector 11. The end effector 11 can move to drive the suction cup 12 to move, and the suction cup 12 can perform depalletizing, that is, suck and move the boxes A1 to A9 to the moving placement area X in sequence.

[0030] Among them, Figure 1 this is only an exemplary application scenario, and the embodiments of the present disclosure can be applied to any object grasping scenario. The embodiments of the present disclosure do not limit the specific application scenario.

[0031] Figure 2 is the step flow chart of a control method of a suction cup robot provided by an exemplary embodiment of the present disclosure; specifically includes the following steps:

[0032] S201, obtain the first point cloud of the adjacent objects of the target object.

[0033] Among them, referring to Figure 1 , a suction cup 12 is fixed on the end effector 11 of the suction cup robot 10, and the suction cup 12 includes multiple partitions (such as B1 to B4), and each partition is independently controlled. Specifically, each partition can be independently controlled means that the partition can be independently controlled to have the ability to suck and not have the ability to suck. For example, in Figure 1 , it can be controlled that partitions B1 and B2 have the ability to suck, and partitions B3 and B4 do not have the ability to suck.

[0034] In the present disclosure, the target object is the object to be grasped this time, and the target object can be one object or two objects. The adjacent object is arranged adjacent to the target object. For example, in Figure 1 if the boxes A1 and A2 are the target objects, then the box A3 is the adjacent object. If the box A2 is the target object, then the boxes A1 and A3 are the adjacent objects.

[0035] In the present disclosure, a point cloud acquisition device is arranged in the scene where the suction cup robot is located. The point cloud acquisition device includes a camera. The point cloud acquisition device can perform point cloud acquisition on the adjacent object to obtain the first point cloud of the adjacent object.

[0036] S202, simulate turning on the target partition among multiple partitions.

[0037] Specifically, in a computer or a server, a suction cup is simulated. The simulated suction cup corresponds one-to-one with the actual suction cup, and the simulated suction cup has the pose information of the actual suction cup. Among them, simulating turning on the target partition among multiple partitions can be understood as simulating turning on the target partition of the simulated suction cup.

[0038] Exemplarily, referring to Figure 3 , Figure 3 the suction cup P in Figure 1 is a simulation of the suction cup 12 in

[0039] The suction cup P includes four partitions D1 to D4, which respectively correspond to the partitions B1 to B4 in the suction cup 12. Figure 3 Among them, the target partition is at least one partition among multiple partitions, and the target partition is determined according to the number of target objects, the grasping area, and the scene position. For example, the target partitions for simulating turning on the suction cup 12 are partitions B3 and B4, corresponding to

[0040] S203, based on the target partition and the first point cloud, determine the grasping result of the proposed execution pose of the end effector.

[0041] Among them, the initial proposed execution pose of the end effector is determined by the suction cup robot according to the calculation model of the suction cup robot and the second point cloud of the target object. The suction cup can grasp the target object in the proposed execution pose.

[0042] Furthermore, in the simulated scene, simulate the end effector in the proposed execution pose, and based on the target partition and the first point cloud, determine the grasping result of the suction cup. As Figure 3 shown, P is the simulated suction cup, partitions D3 and D4 are the simulated target partitions turned on, Figure 1 A2 in

[0043] In addition, the grasping result is used to determine whether adjacent objects will be grasped, referring to Figure 3 If, in the pose to be executed of the end effector, the pose of the simulated suction cup P is as shown in 31, the simulated opened target partitions (partition D3 and partition D4) contact the first point clouds (C1 and C2). It can be seen that if this pose to be executed is adopted to control the suction cup to grasp the target object A2 in the actual scenario, adjacent objects A1 and A3 will be grasped. If, in the pose to be executed of the end effector, the pose of the simulated suction cup P is as shown in 32, the simulated opened target partitions (partition D3 and partition D4) do not contact the first point clouds (C1 and C2). It can be seen that if this pose to be executed is adopted to control the suction cup to grasp the target object A2 in the actual scenario, adjacent objects A1 and A3 will not be grasped.

[0044] S204. According to the grasping result of the pose to be executed, selectively determine the pose to be executed as the actual grasping pose of the end effector.

[0045] Specifically, if the grasping result of the pose to be executed indicates that adjacent objects will not be grasped, this pose to be executed can be determined as the actual grasping pose of the end effector. If the grasping result of the pose to be executed indicates that adjacent objects will be grasped, this pose to be executed cannot be determined as the actual grasping pose of the end effector. Further, the pose to be executed can be adjusted until the grasping result of the pose to be executed indicates that adjacent objects will not be grasped.

[0046] Exemplarily, referring to Figure 3 if, in the pose to be executed 31 of the end effector, the target partitions (D3 and D4) contact the first point clouds (C1 and C3), adjust the pose to be executed of the end effector to obtain a new pose to be executed 32. In the pose to be executed 32, the target partitions (D3 and D4) do not contact the first point clouds (C1 and C3), then determine the pose to be executed 32 as the actual grasping pose.

[0047] In the present disclosure, in the actual grasping pose, the target partition does not contact the first point cloud. Among them, in the simulated scenario, if the target partition does not contact the first point cloud, it can be determined that adjacent objects will not be grasped in this actual grasping pose.

[0048] S205. Control the end effector to move to the actual grasping pose and control the target partition to open to grasp the target object.

[0049] Specifically, in the actual scenario, referring to Figure 1 control the end effector 11 to move to the actual grasping pose. In this actual grasping pose, opening the target partitions (B3 and B4) can accurately grasp the target object and will not grasp adjacent objects.

[0050] The present disclosure can avoid the problem that when the end effector first moves to the execution pose and grasps the target object, it may grasp adjacent objects. If adjacent objects are misgrasped, the grasping process of the entire suction cup robot will go wrong. For example, in the scenario of palletizing and depalletizing, it is necessary to stack 3 target boxes on a pallet. If mis-suction occurs, 4 boxes may be grasped together, and it is very likely that these boxes cannot fit on this pallet, resulting in errors in palletizing and depalletizing. The present disclosure simulates and then controls the end effector to move to the actual grasping pose to grasp the target object after determining that the suction cup will not grasp adjacent objects, so as to achieve fast and accurate grasping of the target object.

[0051] Figure 4 FIG. 4 is a flowchart of steps of another control method for a suction cup robot provided by an exemplary embodiment of the present disclosure, which specifically includes the following steps:

[0052] S401, obtain the first point cloud of adjacent objects of the target object.

[0053] For the specific implementation process of this step, refer to S201, which will not be elaborated here.

[0054] S402, obtain the robot model.

[0055] Among them, the robot model includes: a partition model corresponding to the partition and an end effector model corresponding to the end effector. Specifically, the partition model includes the pose information of the corresponding partition, and the model of the end effector includes the pose information of the end effector.

[0056] S403, simulate the robot using the robot model, simulate adjacent objects according to the first point cloud, and simulate the target object according to the second point cloud.

[0057] Among them, in the simulation scenario, the robot model is used to simulate the robot, the first point cloud is used to simulate adjacent objects, and the second point cloud of the target object is used to simulate the target object. The acquisition method of the second point cloud of the target object is the same as that of the first point cloud, which will not be elaborated here.

[0058] S404, increase the thickness of the partition model on the side of the partition model facing away from the end effector model to obtain a collision detection layer.

[0059] Refer to Figure 5 , P is the partition model corresponding to the suction cup 12. A collision detection layer Z is added to the side of the partition model P facing the target object. The collision detection layer Z is simulated and has a certain thickness. Among them, there are also multiple pieces of the collision detection layer Z, which correspond to the partitions one by one. When simulating the activation of the target partition, only the collision detection layer corresponding to the target partition needs to be simulated.

[0060] Among them, the collision detection layer is used to detect whether the partition contacts the first point cloud. Specifically, in the pose to be executed, there may only be surface contact between the partition model and the first point cloud, and it is impossible to accurately determine the possibility of whether the partition model contacts the first point cloud only based on the existence of the point cloud. If the thickness of the partition model is increased, when there may be surface contact between the partition model and the first point cloud, the collision detection layer will coincide with the first point cloud in pose. Refer to Figure 5 , so the collision detection layer can more accurately detect whether the partition model contacts the first point cloud.

[0061] S405. Determine the target area of the grasping surface of the target object according to the second point cloud of the target object.

[0062] Refer to Figure 6 , if C2 is the second point cloud corresponding to the target object B2, and the point cloud of the grasping surface of the target object B2 is Q, then the target area of the grasping surface can be determined according to Q.

[0063] S406. Determine the target number of target partitions according to the target area.

[0064] Among them, the ratio of the total area of the target partitions with the target number to the target area is greater than the preset ratio threshold.

[0065] In the present disclosure, the area of each partition can be obtained first, and then the target number of target partitions can be determined according to the target area of the grasping surface. Among them, the ratio of the total area of the target partitions to the target area is greater than the preset ratio, such as two-thirds. For example, if the target area of the target object is 1 square meter, the preset ratio is two-thirds, and the area of each partition is 0.4 square meter, then it can be determined that the number of target partitions is greater than or equal to 2.

[0066] S407. Simulate turning on the target number of target partitions among multiple partitions.

[0067] Exemplarily, referring to the above, it is possible to simulate turning on more than 2 or equal to 2 target partitions.

[0068] In addition, according to the scene position information of the target object, determine the first preset combination method; according to the first preset combination method of the partitions, determine the target partitions among multiple partitions, and simulate turning on the target partitions. If there is no actual grasping pose that does not contact the first point cloud in the first preset combination method, then according to the second preset combination method of the partitions, execute determining the target partitions among multiple partitions and simulating turning on the target partitions.

[0069] Among them, the scene position information of the target object refers to the position information of the target object relative to the adjacent object when there is an adjacent object. For example, refer to Figure 1, if the target object is box A2, the scene position information of the target object is the middle position. If the target object is box A1, the scene position information is the left edge position. If the target object is box A3, the scene position information is the right edge position.

[0070] Further, the scene position information can correspond to multiple combination ways of target partitions. For example, if the scene position information of the target object is the middle position, and the box A2 as shown in Figure 1 is the target object, the target partition is two partitions, and the combination ways of the target partitions can be (B1, B2), (B2, B3), and (B3, B4). Take any one of the combination ways as the first preset combination way. For example, take (B2, B3) as the first preset combination way and simulate opening the target partition (B2, B3).

[0071] In addition, in the first preset combination way, if there is an actual grasping pose that does not contact the first point cloud and contacts the second point cloud after subsequent steps, the partition of the first preset combination way can be determined as the target partition. If in the first preset combination way, there is no actual grasping pose that does not contact the first point cloud and contacts the second point cloud after subsequent steps, the target partition corresponding to the second preset combination way (B3, B4) of the partition can be simulated to be opened according to the second preset combination way (B3, B4).

[0072] In summary, the combination way of the target partition that can grasp the target object and will not grasp the adjacent object can be determined, and then the target partition corresponding to the corresponding combination way to be simulated and opened can be selected from multiple partitions.

[0073] S408. Based on the target partition and the first point cloud, determine the grasping result of the proposed execution pose of the end effector.

[0074] The specific implementation process of this step refers to S203 and will not be elaborated here.

[0075] S409. If the grasping result indicates that no adjacent object is grasped, set the proposed execution pose as the actual grasping pose.

[0076] S410. If the grasping result indicates that an adjacent object is grasped, adjust the proposed execution pose according to the contact information between the target partition and the first point cloud, and execute S408.

[0077] Among them, refer to Figure 3, in the planned execution pose 31, the contact information between the target partition and the first point cloud is that both the left and right sides of the target partition (D3 and D4) are in contact with the first point cloud (C1 and C3). Based on this, the planned execution pose can be adjusted. After adjusting the planned execution pose, a new planned execution pose 32 is obtained. In the new planned execution pose 32, the grasping result of the planned execution pose of the end effector is determined. The above steps can be looped until the actual grasping pose is determined.

[0078] S411, control the end effector to move to the actual grasping pose, and control the target partition to open to grasp the target object.

[0079] Furthermore, it also includes: obtaining the third point cloud of the obstacles in the environment where the target object is located; simulating the movement of the end effector between the target object and the target position along the first path, and determining whether the suction cup robot collides with the third point cloud; if not, determining the first path as the target path, and controlling the end effector to move between the target object and the target position along the target path; if so, adjusting the first path to obtain a new first path, and performing the step of simulating the movement of the end effector between the target object and the target position along the first path and determining whether the suction cup robot collides with the third point cloud.

[0080] Among them, the end effector needs to move between the target position and the target object to move the grasped target object to the target position and move from the target position to the target object to grasp the target object.

[0081] In the present disclosure, the third point cloud of the obstacles can be obtained in advance, so that in the simulation scenario, the target path can be determined, and when the end effector moves along the target path, the suction cup robot (robotic arm, end effector and suction cup) will not collide with the obstacles, and thus will not affect the movement of the end effector. Among them, in the palletizing and depalletizing scenario, the obstacles are such as walls, other pallets or other machinery and equipment.

[0082] Furthermore, it can also be detected whether there is a collision with the obstacles through the collision detection layer. That is, whether the obstacle interferes with the movement of the suction cup in the current path.

[0083] Specifically, when the end effector moves from the target position to the target object, the suction cup robot will not collide with the obstacles. When the suction cup robot grasps the target object and moves from the target object to the target position, both the suction cup robot and the target object will not collide with the obstacles.

[0084] Refer to Figure 1When the end effector 11 moves between the box A1 and the target position X along the path L1, it will collide with the obstacle W. If the path is adjusted to L2, no collision will occur, and the quick grasping and transfer of the target object can be achieved.

[0085] In the present disclosure, the target partition for grasping the target object and the actual grasping pose of the target object can be determined by computer simulation, so as to accurately grasp the target object with one control operation of the suction cup robot. In addition, by simulating the obstacles in the simulated environment and the moving path of the end effector, a target path that will not collide with the obstacles is obtained, so that the suction cup robot can also move between the target object and the target position with one movement. In summary, the present disclosure mainly determines the accurate actual grasping pose and the target path by simulation to control the suction cup robot to efficiently grasp and transfer the target object.

[0086] In the embodiment of the present disclosure, refer to Figure 7 In addition to providing the control method of the suction cup robot, a control device 70 of the suction cup robot is also provided. A suction cup is fixed on the end effector of the suction cup robot. The suction cup includes a plurality of partitions, and each partition is independently controlled. The control device 70 of the suction cup robot includes:

[0087] An acquisition module 71, configured to acquire the first point cloud of the adjacent object of the target object;

[0088] A simulation module 72, configured to simulate opening the target partition among the plurality of partitions;

[0089] A first determination module 73, configured to determine the grasping result of the proposed execution pose of the end effector based on the target partition and the first point cloud;

[0090] A second determination module 74, configured to selectively determine the proposed execution pose as the actual grasping pose of the end effector according to the grasping result of the proposed execution pose, where, in the actual grasping pose, the target partition does not contact the first point cloud.

[0091] A control module 75, configured to control the end effector to move to the actual grasping pose and control the target partition to open to grasp the target object.

[0092] In an alternative embodiment, the second determination module 74 is specifically configured to: if the grasping result indicates that the adjacent object is not grasped, take the proposed execution pose as the actual grasping pose; if the grasping result indicates that the adjacent object is grasped, adjust the proposed execution pose according to the contact information between the target partition and the first point cloud, and execute the step of determining the grasping result of the proposed execution pose of the end effector.

[0093] In an alternative embodiment, the simulation module 72 is specifically configured to: determine the target area of the grasping surface of the target object according to the second point cloud of the target object; determine the target number of target partitions according to the target area, where the ratio of the total area of the target number of target partitions to the target area is greater than a preset ratio threshold; and simulate the activation of the target number of target partitions among multiple partitions.

[0094] In an alternative embodiment, the simulation module 72 is specifically configured to: determine target partitions among multiple partitions according to a first preset combination mode of partitions, and simulate the activation of the target partitions.

[0095] In an alternative embodiment, the simulation module 72 is further configured to determine the first preset combination mode according to the scene position information of the target object.

[0096] In an alternative embodiment, the simulation module 72 is further configured to, if there is no actual grasping pose that does not contact the first point cloud in the first preset combination mode, determine target partitions among multiple partitions according to a second preset combination mode of partitions, and simulate the activation of the target partitions.

[0097] In an alternative embodiment, the acquisition module 71 is configured to: acquire the third point cloud of the obstacles in the environment where the target object is located; the simulation module 72 is configured to determine whether the suction cup robot collides with the third point cloud when simulating the movement of the end effector between the target object and the target position along a first path; the control module 75 is configured to, if not, determine the first path as the target path, and control the end effector to move between the target object and the target position along the target path; if so, adjust the first path to obtain a new first path, and execute the step of determining whether the suction cup robot collides with the third point cloud when simulating the movement of the end effector between the target object and the target position along the first path.

[0098] In an alternative embodiment, the acquisition module 71 is further configured to acquire a robot model before simulating the activation of the target partitions among multiple partitions; the robot model includes: a partition model corresponding to the partition and an end effector model corresponding to the end effector; the simulation module 72 is further configured to simulate the robot using the robot model, simulate adjacent objects according to the first point cloud, and simulate the target object according to the second point cloud; increase the thickness of the partition model on the side facing away from the end effector model to obtain a collision detection layer, where whether the partition contacts the first point cloud is detected through the collision detection layer.

[0099] The control device of the suction cup robot provided by the present disclosure can obtain the first point cloud of the adjacent object of the target object; simulate the opening of the target partition among multiple partitions; determine the grasping result of the proposed execution pose of the end effector based on the target partition and the first point cloud; according to the grasping result of the proposed execution pose, selectively determine the proposed execution pose as the actual grasping pose of the end effector, wherein, in the actual grasping pose, the target partition does not contact the first point cloud; control the end effector to move to the actual grasping pose, and control the opening of the target partition to grasp the target object, which can achieve fast and accurate grasping of the target object. For specific implementation, please refer to the above method embodiments and will not be elaborated here.

[0100] In addition, in some of the processes described in the above embodiments and the accompanying drawings, there are multiple operations that appear in a specific order. However, it should be clearly understood that these operations can be executed not in the order in which they appear in this document or in parallel, and are only used to distinguish different operations. The serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and these operations can be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0101] Figure 8 It is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present disclosure. As Figure 8 shown, the electronic device 80 includes: a processor 81, and a memory 82 communicatively connected to the processor 81, and the memory 82 stores computer-executable instructions.

[0102] Wherein, the processor executes the computer-executable instructions stored in the memory to implement the control method of the suction cup robot provided by any of the above method embodiments. The specific functions and technical effects that can be achieved are not elaborated here.

[0103] The embodiment of the present disclosure also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the control method of the suction cup robot provided by any of the above method embodiments.

[0104] The embodiment of the present disclosure also provides a computer program product, which includes: a computer program, the computer program is stored in a readable storage medium, and at least one processor of the electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program to enable the electronic device to execute the control method of the suction cup robot provided by any of the above method embodiments.

[0105] In several embodiments provided by the present disclosure, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of systems or units can be in electrical, mechanical or other forms.

[0106] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0107] In addition, in each embodiment of the present disclosure, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of a combination of hardware and software functional units.

[0108] The above-mentioned integrated units implemented in the form of software functional units can be stored in a computer-readable storage medium. The above-mentioned software functional units stored in a storage medium include several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute some steps of the methods in each embodiment of the present disclosure. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0109] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the system is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system described above can refer to the corresponding process in the foregoing method embodiments and will not be repeated here.

[0110] Other embodiments of the present disclosure will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known common general knowledge or conventional technical means in the technical field not disclosed by the present disclosure. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0111] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A control method for a sucker robot, characterized in that, A suction cup is fixed on the end effector of the suction cup robot. The suction cup includes multiple partitions, each partition is independently controlled. The control method of the suction cup robot includes: Obtain the first point cloud of the adjacent object of the target object; Simulate opening the target partition among the multiple partitions; Based on the target partition and the first point cloud, determine the grasping result of the proposed execution pose of the end effector; According to the grasping result of the proposed execution pose, selectively determine the proposed execution pose as the actual grasping pose of the end effector, wherein, in the actual grasping pose, the target partition does not contact the first point cloud; Control the end effector to move to the actual grasping pose, and control the target partition to open to grasp the target object.

2. The control method of the sucker robot according to claim 1, characterized in that, The step of selectively determining the proposed execution pose as the actual grasping pose of the end effector according to the grasping result of the proposed execution pose includes: If the grasping result indicates that the adjacent object is not grasped, then take the proposed execution pose as the actual grasping pose; If the grasping result indicates that the adjacent object is grasped, then adjust the proposed execution pose according to the contact information between the target partition and the first point cloud, and execute the step of determining the grasping result of the proposed execution pose of the end effector.

3. The control method of the sucker robot according to claim 1 or 2, characterized in that, The step of simulating opening the target partition among the multiple partitions includes: According to the second point cloud of the target object, determine the target area of the grasping surface of the target object; According to the target area, determine the target number of the target partitions, and the ratio of the total area of the target number of target partitions to the target area is greater than a preset ratio threshold; Among the multiple partitions, simulate opening the target number of target partitions.

4. The control method of the sucker robot according to claim 1 or 2, characterized in that, The step of simulating opening the target partition among the multiple partitions includes: Determine the target partition among the multiple partitions according to the first preset combination mode of the partitions, and simulate opening the target partition.

5. The control method of the sucker robot according to claim 4, characterized in that It further includes: Determine the first preset combination mode according to the scene position information of the target object.

6. The control method of the sucker robot according to claim 4, wherein It further includes: If there is no actual grasping pose that does not contact the first point cloud in the first preset combination mode, then execute the step of determining the target partition among the multiple partitions and simulating opening the target partition according to the second preset combination mode of the partitions.

7. The control method of the sucker robot according to any one of claims 1 or 2, characterized in that It further includes: Obtain the third point cloud of the obstacle in the environment where the target object is located; Simulate the case where the end effector moves between the target object and the target position along the first path, and determine whether the suction cup robot collides with the third point cloud; If not, determine the first path as the target path, and control the end effector to move between the target object and the target position along the target path; If so, adjust the first path to obtain a new first path, and execute the step of simulating the case where the end effector moves between the target object and the target position along the first path and determining whether the suction cup robot collides with the third point cloud.

8. The control method of the sucker robot according to claim 3, characterized in that, Before the step of simulating opening the target partition among the multiple partitions, it further includes: Obtain a robot model; the robot model includes: a partition model corresponding to the partition and an end effector model corresponding to the end effector, Simulate the robot using the robot model, simulate the adjacent object according to the first point cloud, and simulate the target object according to the second point cloud; On the side of the partition model facing away from the end effector model, increase the thickness of the partition model to obtain a collision detection layer, wherein whether the partition contacts the first point cloud is detected through the collision detection layer.

9. A control device for a sucker robot, characterized in that, For implementing the control method of the suction cup robot according to any one of claims 1 to 8, a suction cup is fixed on the end effector of the suction cup robot, the suction cup includes a plurality of partitions, each partition is independently controlled, and the control device of the suction cup robot includes: An acquisition module, configured to acquire a first point cloud of an adjacent object of the target object; A simulation module, configured to simulate turning on a target partition among the plurality of partitions; A first determination module, configured to determine a grasping result of a proposed execution pose of the end effector based on the target partition and the first point cloud; A second determination module, configured to selectively determine the proposed execution pose as an actual grasping pose of the end effector according to the grasping result of the proposed execution pose, wherein, in the actual grasping pose, the target partition does not contact the first point cloud; A control module, configured to control the end effector to move to the actual grasping pose and control the target partition to be turned on to grasp the target object.

10. An electronic device, characterized in that, Comprising: A processor, a memory, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, it implements the control method of the suction cup robot according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, Computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed by the processor, they are used to implement the control method of the suction cup robot according to any one of claims 1 to 8.

Images