Picking system, picking device, and picking method

The picking system addresses interference issues by using point cloud data to determine container attitude and frame, ensuring precise workpiece extraction and reducing costs, thereby improving productivity.

JP2024169846AActive Publication Date: 2024-12-06TOYOTA JIDOSHA KK
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2023086653
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-26
Publication Date
2024-12-06
Estimated Expiration
2043-05-26

AI Technical Summary

Technical Problem

Existing article retrieval devices face issues with interference with container frames and internal materials due to incorrect detection of container attitude, leading to decreased gripping success rates and potential collisions.

Method used

A picking system and device that utilize point cloud data to determine the container's attitude and outer frame, enabling precise workpiece extraction by comparing stored and captured data to minimize interference and optimize gripping.

Benefits of technology

The system effectively suppresses interference with container frames and internal materials, improving gripping success rates and reducing calculation costs while enhancing productivity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024169846000001_ABST
    Figure 2024169846000001_ABST
Patent Text Reader

Abstract

To provide a picking system, a picking device, and a picking method capable of appropriately picking up a work by suppressing interference with an outer frame of a container and an inner member other than the work.SOLUTION: A picking system 10 according to the present disclosure includes: a storage unit 103 in which first point group data indicating a shape of a container 110 is stored; a recognition unit 104 for recognizing an opening end outer frame of the container as second point group data from an image obtained by imaging the container; and a specification unit 105 for specifying an attitude of the container by comparing the first point group data with the second point group data.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a picking system, a picking device, and a picking method, and more particularly to a picking system, a picking device, and a picking method that use an arm to pick up a workpiece stored in a container. [Background technology]

[0002] Patent Document 1 discloses an article picking device that recognizes the position of an article in a container and picks up the identified article using an arm. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2014-028415 A Summary of the Invention [Problem to be solved by the invention]

[0004] The item removal device disclosed in Patent Document 1 photographs a container with a camera, detects the opening end face of the container, particularly the inner wall of the container, recognizes the position of the item inside the container using a coordinate measuring machine, and removes the item using an arm.

[0005] However, if the container's posture, especially its tilt angle, is not detected correctly, the arm may interfere with objects other than the workpiece, such as cushioning material, reducing the number of parts that can be grasped. On the other hand, if the arm attempts to grasp the workpiece while ignoring objects other than the workpiece, it may collide with other objects, lowering the success rate of grasping.

[0006] The present disclosure has been made to solve such problems, and aims to provide a picking system, picking device, and picking method that can suppress interference between the outer frame of the container and internal materials other than the work, and properly pick up the work. [Means for solving the problem]

[0007] The picking system according to the present disclosure includes a storage unit in which first point cloud data indicating the shape of a container is stored, a recognition unit that recognizes the outer frame of the open end of the container as second point cloud data from an image obtained by capturing an image of the container, and an identification unit that identifies the attitude of the container by comparing the first point cloud data with the second point cloud data. In this way, a picking system can be provided that can appropriately pick up works by suppressing interference between the outer frame of the container and internal materials other than the workpieces.

[0008] In addition, the workpieces may be picked from the workpieces closest to the outer frame of the open end of the container. In this way, a picking system that can reduce calculation costs and improve productivity can be provided.

[0009] The picking device according to the present disclosure includes a control unit that is equipped with the picking system, and an arm that picks out the workpieces stored in the container based on the information obtained by the identification unit. In this way, it is possible to provide a picking device that can appropriately pick up the workpieces by suppressing interference between the outer frame of the container and the internal materials other than the workpieces.

[0010] The picking method according to the present disclosure implements the picking system to pick up the workpieces from the container, thereby providing a picking method that can reduce calculation costs and improve productivity. Effect of the Invention

[0011] The present disclosure makes it possible to provide a picking system, picking device, and picking method that can suppress interference between the outer frame of a container and internal materials other than the work, and can properly pick up the work. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram of a picking system according to the present disclosure. [Diagram 2] 13A-13C are diagrams illustrating a process for extracting the top surface area of ​​a container according to the present disclosure. [Diagram 3] 1 is a flowchart illustrating a picking system according to the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] <Embodiment 1> Hereinafter, the present disclosure will be described with reference to the drawings. Fig. 1 is a schematic diagram of a picking system according to the present disclosure. As shown in Fig. 1(a), a picking system 10 is mounted on and executed by a control unit 101 provided in a picking device 100 that controls an arm 102. The control unit 101 includes a storage unit 103, a recognition unit 104, and an identification unit 105.

[0014] The storage unit 103 pre-stores first point cloud data indicating the shape of the container 110 placed on the transport platform 120. The recognition unit 104 recognizes the outer frame of the open end of the container 110 as second point cloud data from an image of the container 110 acquired by an imaging unit 106 described below. The identification unit 105 compares the first point cloud data stored in the storage unit 103 with the second point cloud data acquired by the recognition unit 104 to identify the posture of the container 110.

[0015] The imaging unit 106 captures an image of the container 110 and transmits information about the image obtained to the control unit 101 or the recognition unit 104. The imaging unit 106 preferably includes a camera, a three-dimensional measuring machine, or the like. In FIG. 1(b), the imaging unit 106 is illustrated as being provided in the control unit 101, but the present disclosure is not limited to this. For example, the imaging unit 106 may be provided on the upper part of the conveyance platform 120 or the like to capture an image of the container 110. In this case, the imaging unit 106 preferably includes a communication function for transmitting information about the image of the container 110 to the control unit 101.

[0016] 1(c) shows a state when a deviation θ occurs between a normal vector A of the container 110 and a vector B in the direction in which the arm 102 is inserted. The arm 102 is preferably equipped with a two-fingered hand.

[0017] The first point cloud data and the second point cloud data represent a set of points having position information (X, Y, Z) and color information (R, G, B). Acquisition of the second point cloud data of the outer frame of the opening end of the container 110 will be described with reference to Figs. 2 and 3.

[0018] First, the imaging unit 106 acquires image data and three-dimensional point cloud data of the container 110 shown in FIG. 2(a) (S11).

[0019] Next, the recognition unit 104 performs semantic segmentation (area classification) on the image data of the container 110 (S12) and extracts three-dimensional point cloud data of the opening end outer frame 111 of the container 110 shown in FIG. 2(b) (S13).

[0020] Thereafter, the identification unit 105 compares the second point cloud data, which is the extracted three-dimensional point cloud data, with the first point cloud data, which is the three-dimensional point cloud data of the registered container 110 stored in the memory unit 103 (S14), and calculates the relative position of the container 110 and the arm 102 (S15).

[0021] After the calculation, a gripping point of the arm 102 is searched for based on the obtained 3D point cloud information (S16). The gripping point may be determined using machine learning that introduces graspability as a parameter. This makes it possible to bring the deviation θ between the normal vector A of the container 110 and the vector B of the direction in which the arm 102 is inserted, as shown in FIG. 1(c), closer to zero.

[0022] In this way, it becomes possible to appropriately remove a workpiece even from a container 110 that contains workpieces and other internal materials such as cushioning materials. In addition, in the gripping point search in S16, the open end outer frame 111 of the container 110 can be excluded to narrow the calculation area, thereby reducing the calculation cost.

[0023] In addition, when removing the workpiece, it is preferable to remove the workpiece from the area stored in the area close to the open end outer frame 111. By doing so, the calculation area can be narrowed down every time a workpiece is successfully removed, which reduces the calculation cost and improves productivity.

[0024] In this way, it is possible to provide a picking system, picking device, and picking method that can suppress interference between the outer frame of the container and internal materials other than the workpieces, and can properly pick up the workpieces.

[0025] The present disclosure is not limited to the above, and can be modified as appropriate without departing from the spirit and scope of the present disclosure. [Explanation of symbols]

[0026] 10 Picking System 100 Picking device 101 Control section 102 Arm 103 Storage section 104 Recognition part 105 Specific section 106 Imaging unit 110 Container 111 Open end outer frame 120 Transport platform

Claims

1. a storage unit in which first point cloud data indicating a shape of a container is stored; a recognition unit that recognizes an outer frame of an opening end of the container from an image obtained by capturing an image of the container as second point cloud data; an identification unit that identifies the orientation of the container by comparing the first point cloud data with the second point cloud data. Picking system.

2. Pick the workpieces from the closest to the outer frame of the open end of the container. The picking system according to claim 1 .

3. A control unit including the picking system according to claim 1 or 2; A picking device comprising: an arm that removes a workpiece contained in the container based on the information obtained by the identification unit.

4. The picking system according to claim 1 or 2 is implemented to pick up the workpiece from the container. Picking method.

Citation Information

Patent Citations

  • Method and apparatus for taking out bulked article by using robot

    JP2015089590A

  • Picking hand and picking transfer device

    JP2017019100A

  • Device and method for determining action of robot, and program

    JP2021020266A

  • Method and computing system for object detection or robot interaction planning based on image information generated by a camera

    JP2022036243A

  • Part conveyance system

    JP2022159887A