JP7670601B2 - Workpiece removal device, workpiece removal method, program and control device - Google Patents

Description

The present invention relates to a workpiece removal device, a workpiece removal method, a program, and a control device for removing a workpiece.

A workpiece removal device is known that grasps and removes workpieces one by one from among multiple workpieces placed in a three-dimensional space (see, for example, Patent Document 1).

JP 2009-128191 A

However, when the above-mentioned workpiece removal device grips a linear workpiece, there is a risk of failure to grip the workpiece, such as gripping multiple workpieces at the same time, for example, because the workpieces may become entangled.

The present invention has been made to solve these problems, and its main objective is to provide a workpiece removal device, workpiece removal method, program, and control device that can reduce failure to grip the workpiece and ensure that the workpiece can be removed.

In order to achieve the above object, one aspect of the present invention is to
A workpiece removal device that removes workpieces one by one from a plurality of linear workpieces placed in a three-dimensional space,
A gripping means for gripping the workpiece;
a control means for controlling the gripping means so that the workpiece gripped by the gripping means moves along a circular or rectangular trajectory after the gripping means grips the workpiece;
Equipped with
This is a workpiece removal device.
In one aspect of this invention, the gripping means may further include a judgment means for judging whether the gripping means has succeeded in gripping the workpiece, and the control means may control the gripping means so that the workpiece gripped by the gripping means traces a circular or rectangular trajectory only when the judgment means judges that gripping of the workpiece has failed.
In this one aspect, after the gripping means is controlled so that the workpiece gripped by the gripping means traces a circular trajectory, the judgment means again judges whether the gripping means has successfully gripped the workpiece, and the control means may control the gripping means so that the workpiece gripped by the gripping means traces a circular or rectangular trajectory depending on the judgment result by the judgment means.
In this one aspect, the control means may control the gripping means so that the workpiece gripped by the gripping means revolves around a circular or rectangular orbit and at least one of rotating the workpiece gripped by the gripping means and vibrating the workpiece gripped by the gripping means.
In one aspect of this invention, the workpiece may further include a torque detection means for detecting the torque when the workpiece moves in a circular or rectangular trajectory, and when the control means determines that the torque detected by the torque detection means is equal to or greater than a predetermined value, the control means may control the gripping means so that the workpiece gripped by the gripping means moves in a circular or rectangular trajectory and vibrates the workpiece gripped by the gripping means.
In one aspect of this invention, the apparatus further includes an acquisition means for acquiring three-dimensional information of the multiple linear workpieces, and the control means calculates the distance from the gripping point at which the gripping means grips the workpiece to the maximum height position of the multiple linear workpieces based on the three-dimensional information of the multiple linear workpieces acquired by the acquisition means, and may determine the operation of the workpiece after the gripping means grips the workpiece based on the calculated distance.
In this aspect, the control means may increase the amount of movement of the workpiece as the distance increases.
In this one aspect, the multiple linear workpieces in the three-dimensional space are arranged in a wound state, and the system further includes a radius information acquisition means for acquiring information on the radius of the workpiece in the wound state, and the control means may determine a radius of a circular orbit of the workpiece gripped by the gripping means based on the information on the radius of the workpiece acquired by the radius information acquisition means, and control the gripping means so that the workpiece describes a circular orbit based on the determined radius.
In order to achieve the above object, one aspect of the present invention is to
A method for removing workpieces one by one from a plurality of linear workpieces placed in a three-dimensional space, comprising the steps of:
A step of gripping the workpiece;
After gripping the workpiece, moving the gripped workpiece so that the gripped workpiece describes a circular or rectangular trajectory;
Including,
The present invention may also be directed to a workpiece removal method.
In order to achieve the above object, one aspect of the present invention is to
A process of grasping each of a plurality of linear workpieces placed in a three-dimensional space;
A process of gripping the workpiece and then moving the gripped workpiece so that the gripped workpiece describes a circular or rectangular trajectory;
to cause a computer to execute
It may be a program.
In order to achieve the above object, one aspect of the present invention is to
A control device that controls the picking out of a plurality of linear workpieces placed in a three-dimensional space one by one,
After the gripping means grips the workpiece, the gripping means is controlled so that the workpiece gripped by the gripping means moves along a circular or rectangular trajectory.
It may be a control device.

The present invention provides a workpiece removal device, a workpiece removal method, a program, and a control device that can reduce failure to grip the workpiece and ensure that the workpiece can be removed.

1 is a diagram showing a schematic system configuration of a workpiece removal device according to an embodiment of the present invention; 1 is a block diagram showing a schematic system configuration of a control device according to an embodiment of the present invention; 4 is a flowchart showing a flow of a workpiece unloading method according to the present embodiment. 1 is a block diagram showing a schematic system configuration of a control device according to an embodiment of the present invention; FIG. 13 is a diagram showing the distance from the gripping point of the workpiece to the maximum height position of the workpiece. FIG. 4 is a flowchart showing a flow of a workpiece removal method according to the present embodiment. 10A to 10C are diagrams for explaining a workpiece removal method according to the present embodiment. 1 is a block diagram showing a schematic system configuration of a control device according to an embodiment of the present invention; FIG. 1 shows a circular trajectory corresponding to the radius of the cable.

EMBODIMENT 1
Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Fig. 1 is a diagram showing a schematic system configuration of a workpiece extraction device according to this embodiment. The workpiece extraction device 1 according to this embodiment can grip and move each of a plurality of linear workpieces arranged in a three-dimensional space. The linear workpiece is, for example, a cable-like workpiece having a connector, a branching portion, etc. The plurality of workpieces are linear parts randomly piled up in a box, on a plane, etc.

The workpiece removal device 1 of this embodiment includes a robot arm 2, a control device 3, and a 3D vision sensor 4.

The robot arm 2 is a specific example of a gripping means. The robot arm 2 is configured as a multi-joint arm having, for example, a plurality of links 21, joints (wrist joints, elbow joints, shoulder joints, etc.) 22 that rotatably connect the links 21, and an end effector 23 provided at the tip of the links 21 to grip a workpiece.

Each joint 22 is provided with a rotation sensor such as an encoder that detects rotation information of the joint 22, an actuator such as a servo motor that drives the joint 22, and a force sensor that detects the operating force of the joint 22. The force sensor is, for example, a torque sensor that detects the torque of the joint 22. Each joint 22 is provided with a reduction mechanism, etc.

The end effector 23 has, for example, multiple fingers. The fingers are provided with joints, actuators, etc., and each finger is configured to be able to grip a workpiece.

The control device 3 performs various calculations and control processes for the robot arm 2. The control device 3 has a hardware configuration of a normal computer, including, for example, a processor 3a such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), an internal memory 3b such as a RAM (Random Access Memory) or a ROM (Read Only Memory), a storage device 3c such as a HDD (Hard Disk Drive) or an SSD (Solid State Drive), an input/output I/F 3d for connecting peripheral devices such as a display, and a communication I/F 3e for communicating with devices outside the device.

The three-dimensional vision sensor 4 is a specific example of an acquisition means. The three-dimensional vision sensor 4 acquires three-dimensional information of multiple workpieces. The three-dimensional information of the workpieces includes information such as the shape, position (three-dimensional coordinates, etc.), and posture of each workpiece.

The three-dimensional vision sensor 4 is provided, for example, on the end effector 23 or link 21 of the robot arm 2. The three-dimensional vision sensor 4 is composed of a camera, a laser sensor, etc. The three-dimensional vision sensor 4 outputs the acquired three-dimensional information of the workpiece to the control device 3.

Figure 2 is a block diagram showing a schematic system configuration of a control device according to this embodiment. The control device 3 according to this embodiment has a candidate calculation unit 31 and a robot control unit 32.

The candidate calculation unit 31 calculates at least one candidate gripping point, which is a candidate for the gripping point of the workpiece when the end effector 23 of the robot arm 2 grips the workpiece, based on the three-dimensional information of the workpiece acquired by the three-dimensional vision sensor 4.

For example, the candidate calculation unit 31 calculates depth information indicating the height position of each workpiece based on the three-dimensional information of the workpiece acquired by the three-dimensional vision sensor 4. The candidate calculation unit 31 detects convex parts protruding in the height direction based on the calculated depth information of each workpiece.

The candidate calculation unit 31 performs binarization based on the calculated depth information, for example, by setting points at or above a predetermined height as 1 and points below the predetermined height as 0, and detects points that are 1 as convex parts. The candidate calculation unit 31 may set the convex parts detected as described above as candidate gripping points. This allows convex parts that are easy for the end effector 23 to grip to be the candidate gripping points.

The candidate calculation unit 31 outputs the multiple gripping candidate points calculated as described above to the robot control unit 32.

The robot control unit 32 is a specific example of a control means. The robot control unit 32 controls the operation of the robot arm 2. For example, the robot control unit 32 feedback-controls the robot arm 2 by controlling the actuators of each joint based on rotation information (such as rotation angle) from a rotation sensor of each joint and the operating force from a force sensor. This allows the end effector 23 of the robot arm 2 to be moved to a desired position.

The robot control unit 32 controls the actuators of the fingers of the end effector 23 to control the gripping and release of the workpiece by the end effector 23. In this way, the robot control unit 32 can grip and move the workpiece by controlling the robot arm 2.

The robot control unit 32 may select one gripping candidate point from the multiple gripping candidate points calculated by the candidate calculation unit 31. Based on the three-dimensional information of the workpiece acquired by the three-dimensional vision sensor 4, the robot control unit 32 controls the robot arm 2 so that the end effector 23 of the robot arm 2 grips the workpiece at the gripping candidate point.

However, when conventional workpiece removal devices grip linear workpieces, there is a risk that the workpieces may become entangled, for example, and therefore fail to grip the workpieces, such as gripping multiple workpieces at the same time.

In contrast, the workpiece removal device 1 according to this embodiment controls the robot arm 2 so that after the end effector 23 of the robot arm 2 grasps the workpiece, the workpiece grasped by the end effector 23 traces a circular trajectory. By controlling the robot arm 2 so that the end effector 23 of the robot arm 2 traces a circular trajectory, the workpiece grasped by the end effector 23 also traces a circular trajectory, as described above.

The workpiece removal device 1, for example, controls the robot arm 2 so that the end effector 23 of the robot arm 2 grasps the workpiece and lifts it to a predetermined height, and then the robot arm 2 is controlled so that the workpiece grasped by the end effector 23 moves in a circular trajectory.

For example, linear workpieces such as cables are often wound in a circle. Therefore, by rotating the workpiece from the outside in a circular orbit as if to untie the circle, the entanglements between the workpieces can be easily untangled. This reduces the chance of the workpiece failing to be grasped and allows the workpiece to be removed reliably.

The circular trajectory may include not only a perfect circle but also an ellipse. Furthermore, the robot control unit 32 may control the robot arm 2 so that the workpiece grasped by the end effector 23 of the robot arm 2 traces a rectangular trajectory.

The robot control unit 32 may control the robot arm 2 so that after the robot arm 2 grasps the workpiece, the workpiece grasped by the robot arm 2 rotates while revolving around a circular orbit. This makes it easier to untangle the entangled workpieces by the revolution and rotation of the workpieces.

The robot control unit 32 may control the robot arm 2 so that after the robot arm 2 grasps the workpiece, the workpiece grasped by the robot arm 2 revolves in a circular orbit and vibrates the workpiece grasped by the robot arm 2. This makes it easier to disentangle the entangled workpieces by revolving and vibrating the workpieces.

For example, the end effector 23 of the robot arm 2 may be provided with a vibrator that generates the above vibration. It is preferable to set the magnitude of the above vibration according to the size of the parts where entanglement of workpieces is likely to occur, such as connectors and branching parts.

When the robot control unit 32 determines that the torque detected by the torque sensor is equal to or greater than a predetermined value, it may control the robot arm 2 so that the workpiece grasped by the robot arm 2 revolves in a circular orbit and the workpiece grasped by the robot arm 2 is vibrated by a vibrator.

The torque sensor is a specific example of a torque detection means. The torque sensor is provided, for example, on the end effector 23 of the robot arm 2. The torque sensor detects the rotational torque when the workpiece moves in a circular orbit. When the workpieces held by the robot arm 2 are caused to revolve in a circular orbit, if the workpieces are tightly entangled, the torque value of the torque sensor becomes high and exceeds a predetermined value. At this time, the workpieces held by the robot arm 2 are vibrated by a vibrator, so that the tight entanglement between the workpieces can be more easily undone.

The robot control unit 32 may control the robot arm 2 so that the workpiece grasped by the robot arm 2 rotates while revolving around a circular orbit and vibrates the workpiece grasped by the robot arm 2. This makes it easier to disentangle the entanglement between the workpieces by revolving, rotating and vibrating the workpieces.

The robot control unit 32 may control the robot arm 2 so that the workpiece grasped by the end effector 23 of the robot arm 2 moves back and forth in a circular trajectory.

Next, the workpiece removal method according to this embodiment will be described. Figure 3 is a flowchart showing the flow of the workpiece removal method according to this embodiment.

The three-dimensional vision sensor 4 acquires three-dimensional information of multiple workpieces and outputs the acquired three-dimensional information of the workpieces to the control device 3 (step S101).

The candidate calculation unit 31 of the control device 3 calculates candidate gripping points of the workpiece when the end effector 23 of the robot arm 2 grips the workpiece based on the three-dimensional information of the workpiece acquired by the three-dimensional vision sensor 4 (step S102). The candidate calculation unit 31 outputs the calculated candidate gripping points to the robot control unit 32.

The robot control unit 32 controls the robot arm 2 so that the end effector 23 of the robot arm 2 grasps and lifts the workpiece at the candidate grasping point based on the three-dimensional information of the workpiece acquired by the three-dimensional vision sensor 4 (step S103).

The robot control unit 32 controls the robot arm 2 so that the workpiece grasped by the end effector 23 of the robot arm 2 follows a circular trajectory (step S104).

As described above, the workpiece removal device 1 according to this embodiment controls the robot arm 2 so that after the robot arm 2 grasps a workpiece, the workpiece grasped by the robot arm 2 moves in a circular trajectory. This makes it possible to easily untangle the workpieces, suppress failure to grasp the workpiece, and reliably remove the workpiece.

EMBODIMENT 2
4 is a block diagram showing a schematic system configuration of the control device according to this embodiment. The control device 30 according to this embodiment further includes a success/failure determination unit 33 that determines whether the robot arm 2 has successfully grasped the workpiece.

The robot control unit 32 controls the robot arm 2 so that the workpiece grasped by the robot arm 2 traces a circular trajectory only when the success/failure determination unit 33 determines that grasping of the workpiece has failed. As a result, the operation of tracing the circular trajectory of the workpiece is performed only when it is determined that grasping of the workpiece has failed, and therefore only the minimum necessary operation is required.

The success/failure determination unit 33 is a specific example of a determination means. When the robot arm 2 grasps a workpiece, the success/failure determination unit 33 determines whether the workpiece has been successfully grasped. For example, after the workpiece has been grasped by the robot arm 2 and lifted up to a predetermined height, the success/failure determination unit 33 determines whether the workpiece has been successfully grasped.

The success/failure determination unit 33 determines that the gripping of the workpiece has failed when the end effector 23 of the robot arm 2 grips multiple (two or more) workpieces at the same time. The success/failure determination unit 33 may also determine that the gripping of the workpiece has failed when the end effector 23 of the robot arm 2 grips zero workpieces, i.e., when it is unable to grip the workpieces. In particular, as described above, when gripping linear workpieces, the workpieces are likely to become entangled and fail to grip multiple workpieces at the same time, and it is necessary to reliably determine this failure.

For example, a force sensor 5 is provided between the link tip of the robot arm 2 and the end effector 23. When the end effector 23 grips a workpiece, the weight of the workpiece is applied to the end effector 23. The force sensor 5 detects the weight of the workpiece.

The success/failure determination unit 33 may determine whether the workpiece has been successfully grasped based on the weight of the workpiece detected by the force sensor 5. The weight of the workpiece may be set in advance in the success/failure determination unit 33. Based on the weight of the workpiece detected by the force sensor 5, the success/failure determination unit 33 determines that grasping of the workpiece has failed if the end effector 23 of the robot arm 2 grasps multiple workpieces simultaneously, if it grasps zero workpieces, or if the workpiece cannot be lifted.

The success/failure determination unit 33 may determine whether or not the workpiece has been successfully grasped based on the weight of the workpiece detected by the torque sensor. The success/failure determination unit 33 may determine whether or not the workpiece has been successfully grasped based on an image of the workpiece captured by a camera. The success/failure determination unit 33 outputs the result of the determination of whether or not the workpiece has been successfully grasped, determined as described above, to the robot control unit 32.

As described above, when the success/failure determination unit 33 determines that the gripping of the workpiece has failed, the robot control unit 32 controls the robot arm 2 so that the workpiece gripped by the robot arm 2 follows a circular trajectory.

Furthermore, after the workpiece grasped by the robot arm 2 has been controlled to trace a circular trajectory, the success/failure determination unit 33 may again determine whether the robot arm 2 has successfully grasped the workpiece. Then, the robot control unit 32 may control the robot arm 2 so that the workpiece grasped by the robot arm 2 traces a circular trajectory, depending on the determination result by the success/failure determination unit 33. This makes it possible to more reliably untangle the workpieces and more reliably prevent failure to grasp the workpiece.

EMBODIMENT 3
In this embodiment, the robot control unit 32 may calculate a distance d from a gripping point at which the robot arm 2 grips the workpiece to a maximum height position of the multiple linear workpieces, as shown in Fig. 5, based on three-dimensional information of the multiple linear workpieces acquired by the three-dimensional vision sensor. The robot control unit 32 may determine the operation of the workpiece after the robot control unit 32 grips the workpiece, based on the calculated distance d.

The greater the distance d from the gripping point of the workpiece to the maximum height position of the workpiece, the more likely it is that the workpiece at that gripping point is strongly entangled with other workpieces. Therefore, by determining the operation of the workpieces taking into account the distance d, which reflects the degree of entanglement, the workpieces can be more effectively untangled.

It is preferable that the robot control unit 32 increases the amount of movement of the workpiece as the distance d from the gripping point of the workpiece to the maximum height position of the workpiece increases. This makes it possible to more reliably untangle the workpieces by increasing the amount of movement of the workpiece when the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is large and the degree of entanglement is high.

For example, the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is divided into six levels, from (1) to (6), which increase in level in that order. (1) has the shortest distance d, the lowest degree of entanglement, and the lowest level. On the other hand, (6) has the longest distance d, the highest degree of entanglement, and the highest level.

The movements of the workpiece are pre-associated with the levels (1) to (6), for example as shown in FIG. 6. The amount of movement of the workpiece increases gradually in this order from (1) to (6). Note that in FIG. 6, the workpiece is rotated clockwise, but it may be rotated counterclockwise. The range of levels (1) to (6) is set to an optimal value determined in advance through experiments.

When the robot control unit 32 determines that the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is within the range of level (1), the robot arm 2 grips and lifts the workpiece, and then controls the robot arm 2 to lift the workpiece gripped by the robot arm 2 straight up.

When the robot control unit 32 determines that the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is within the range of level (2), the robot arm 2 grips and lifts the workpiece, and then controls the robot arm 2 so that the workpiece gripped by the robot arm 2 revolves half a revolution in a circular orbit.

When the robot control unit 32 determines that the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is within the range of level (3), the robot arm 2 grips and lifts the workpiece, and then controls the robot arm 2 so that the workpiece gripped by the robot arm 2 rotates once while revolving half a revolution in a circular orbit.

When the robot control unit 32 determines that the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is within the range of level (4), the robot arm 2 grips and lifts the workpiece, and then controls the robot arm 2 so that the workpiece gripped by the robot arm 2 revolves around the workpiece in a circular orbit once.

When the robot control unit 32 determines that the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is within the range of level (5), the robot arm 2 grips and lifts the workpiece, and then controls the robot arm 2 so that the workpiece gripped by the robot arm 2 revolves once around a circular orbit while rotating once on its own axis.

When the robot control unit 32 determines that the distance d from the gripping point of the workpiece to the maximum height position of the workpiece is within the range of level (6), the robot arm 2 grips and lifts the workpiece, and then controls the robot arm 2 so that the workpiece gripped by the robot arm 2 revolves around the workpiece in a circular orbit for two revolutions.

The correspondence information that associates the distance d from the gripping point of the workpiece to the maximum height position of the workpiece as described above with the operation of the workpiece may be set in advance in the robot control unit 32, etc.

The robot control unit 32 may use a learning device to determine the behavior of the workpiece after the robot control unit 32 grasps the workpiece. The learning device learns the correspondence between multiple linear workpiece images acquired by a camera or the like and the behavior of the workpiece after the robot control unit 32 grasps the workpiece. The robot control unit 32 may then input an image of the workpiece including a grasping point to the learned learning device, and determine the behavior of the workpiece after the robot control unit 32 grasps the workpiece based on the results output from the learning device.

The learning device is composed of a neural network such as a recurrent neural network (RNN). This RNN may have a long short term memory (LSTM) in the intermediate layer. Instead of a neural network, the learning device may be composed of another learning device such as a support vector machine (SVM).

If the success/failure determination unit 33 determines that gripping of the workpiece has failed, the robot control unit 32 may calculate the distance d from the gripping point of the workpiece to the maximum height position of the workpiece, and determine the operation of the workpiece after the robot control unit 32 grips the workpiece based on the calculated distance d.

Figure 7 is a flowchart showing the flow of the workpiece removal method according to the embodiment. As shown in Figure 8 (a), the 3D vision sensor 4 acquires 3D information of multiple workpieces and outputs the acquired 3D information of the workpieces to the control device 3 (step S201).

The candidate calculation unit 31 of the control device 3 calculates candidate gripping points of the workpiece when the end effector 23 of the robot arm 2 grips the workpiece based on the three-dimensional information of the workpiece acquired by the three-dimensional vision sensor 4 (step S202). The candidate calculation unit 31 outputs the calculated candidate gripping points to the robot control unit 32.

The robot control unit 32 controls the robot arm 2 so that the end effector 23 of the robot arm 2 grasps and lifts the workpiece at the candidate grasping point based on the three-dimensional information of the workpiece acquired by the three-dimensional vision sensor 4 (step S203).

As shown in FIG. 8(b), the success/failure determination unit 33 determines whether the workpiece has been successfully grasped based on the weight of the workpiece detected by the force sensor 5 (step S204).

If the success/failure determination unit 33 determines that gripping of the workpiece has failed due to the above-mentioned multiple workpiece gripping (NO in step S204), the robot control unit 32 calculates the distance from the gripping point of the workpiece to the maximum height position of the workpiece based on the three-dimensional information of the multiple workpieces acquired by the three-dimensional vision sensor 4 (step S205).

As shown in FIG. 8(c), the robot control unit 32 determines the movement of the workpiece based on the calculated distance and the correspondence information, controls the robot arm 2 so that the workpiece performs that movement (step S206), and returns to the above processing (step S204).

If the success/failure determination unit 33 determines that the workpiece has been successfully grasped by grasping the multiple workpieces as shown in FIG. 8(d) (YES in step S204), the robot control unit 32 controls the robot arm 2 to move the workpiece to the desired position (step S207).

If the success/failure determination unit 33 determines that the gripping of the workpieces has failed due to gripping of zero workpieces (NO in step S204), the process may return to the above (step S201).

EMBODIMENT 4
9 is a block diagram showing a schematic system configuration of the control device according to the present embodiment. A linear workpiece such as a cable may be arranged in a wound state, for example. In response to this, the control device 40 according to the present embodiment further includes a radius information acquisition unit 34 that acquires information on the radius of the workpiece in the wound state.

The radius information acquisition unit 34 is a specific example of a radius information acquisition means. The radius information acquisition unit 34 calculates the radius of the workpiece based on, for example, an image of the workpiece acquired by the 3D vision sensor 4. Information on the radius of the workpiece may be input to the radius information acquisition unit 34, or may be set in advance in the radius information acquisition unit 34.

The robot control unit 32 determines the radius of the circular trajectory of the workpiece gripped by the robot arm 2 based on the information on the radius of the workpiece acquired by the radius information acquisition unit 34. The robot control unit 32 controls the robot arm 2 so that the workpiece describes a circular trajectory with the determined radius. This makes it possible to set the optimal size of the circular trajectory of the workpiece according to the size of the radius of the linear workpiece arranged in a wound state, thereby making it possible to more effectively untangle the workpieces from each other.

It is preferable that the robot control unit 32 increases the circular trajectory of the workpiece grasped by the robot arm 2 as the radius of the workpiece acquired by the radius information acquisition unit 34 increases.

For example, as shown in FIG. 10, the radius Ra of cable A is larger than the radius Rb of cable B. Therefore, the robot control unit 32 determines that the circular trajectory Sa of cable A is larger than the circular trajectory Sb of cable B.

Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. These embodiments and their modifications are included within the scope and gist of the invention, and are included in the scope of the invention and its equivalents as set forth in the claims.

The present invention can also be realized, for example, by having a processor execute a computer program to perform the processes shown in Figures 3 and 7.

The program can be stored and provided to a computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R/Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (random access memories)).

The program may be provided to the computer by various types of transitory computer readable media. Examples of transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer readable medium may provide the program to the computer via a wired communication path, such as an electrical wire or optical fiber, or via a wireless communication path.

Each unit constituting the control device 3 according to each of the above-described embodiments can be realized not only by a program, but also in part or in whole by dedicated hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array).

1 Workpiece removal device, 2 Robot arm, 3 Control device, 4 3D vision sensor, 5 Force sensor, 21 Link, 22 Joint, 23 End effector, 30 Control device, 31 Candidate calculation unit, 32 Robot control unit, 33 Success/failure determination unit, 34 Radius information acquisition unit, 40 Control device

Claims (10)

A workpiece removal device that removes workpieces one by one from a plurality of linear workpieces placed in a three-dimensional space,
A gripping means for gripping the workpiece;
a control means for controlling the gripping means so that the workpiece gripped by the gripping means moves along a circular or rectangular trajectory after the gripping means grips the workpiece;
Equipped with
The control means controls the gripping means so that the workpiece gripped by the gripping means revolves around a circular or rectangular orbit and at least one of the following is performed: the workpiece gripped by the gripping means is rotated on its axis; and the workpiece gripped by the gripping means is vibrated.
Work removal device. The workpiece removal device according to claim 1,
The gripping means further includes a determination means for determining whether the gripping means has successfully gripped the workpiece,
The control means controls the gripping means so that the workpiece gripped by the gripping means moves in a circular or rectangular trajectory only when the determination means determines that gripping of the workpiece has failed.
Work removal device. The workpiece removal device according to claim 2,
The determination means determines again whether the gripping means has successfully gripped the workpiece after the gripping means has controlled the workpiece to move along a circular trajectory,
The control means controls the gripping means so that the workpiece gripped by the gripping means moves in a circular or rectangular trajectory according to the determination result by the determination means.
Work removal device. The workpiece removal device according to any one of claims 1 to 3 ,
Further, a torque detection means is provided for detecting a torque when the workpiece moves in a circular or rectangular trajectory,
When the control means determines that the torque detected by the torque detection means is equal to or greater than a predetermined value, the control means controls the gripping means so that the workpiece gripped by the gripping means moves in a circular or rectangular trajectory and vibrates the workpiece gripped by the gripping means.
Work removal device. The workpiece removal device according to any one of claims 1 to 3 ,
The method further includes acquiring three-dimensional information of the plurality of linear workpieces,
The control means
Calculating a distance from a gripping point at which the gripping means grips the workpiece to a maximum height position of the plurality of linear workpieces based on the three-dimensional information of the plurality of linear workpieces acquired by the acquisition means;
determining an operation of the workpiece after the gripping means grips the workpiece based on the calculated distance;
Work removal device. The workpiece removal device according to claim 5 ,
The control means increases the amount of movement of the workpiece as the distance increases.
Work removal device. The workpiece removal device according to any one of claims 1 to 3 ,
The plurality of linear workpieces in the three-dimensional space are arranged in a wound state,
Further comprising a radius information acquisition means for acquiring information on the radius of the workpiece in the wound state,
The control means determines a radius of a circular orbit of the workpiece gripped by the gripping means based on information of the radius of the workpiece acquired by the radius information acquisition means, and controls the gripping means based on the determined radius so that the workpiece describes a circular orbit. A method for removing workpieces one by one from a plurality of linear workpieces placed in a three-dimensional space, comprising the steps of:
A step of gripping the workpiece by a gripping means;
After gripping the workpiece, moving the gripped workpiece so that the gripped workpiece describes a circular or rectangular trajectory;
Including,
The gripping means is controlled so that the workpiece gripped by the gripping means revolves around a circular or rectangular orbit and at least one of the following is performed: the workpiece gripped by the gripping means is rotated on its axis; and the workpiece gripped by the gripping means is vibrated.
How to remove the workpiece. A process of gripping each of a plurality of linear workpieces placed in a three-dimensional space by a gripping means ;
A process of moving the gripped workpiece so that the gripped workpiece describes a circular or rectangular trajectory after the gripping means grips the workpiece;
A process of controlling the gripping means so that the workpiece gripped by the gripping means revolves around a circular or rectangular orbit and at least one of rotating the workpiece gripped by the gripping means and vibrating the workpiece gripped by the gripping means is performed;
to cause a computer to execute
program. A control device that controls the picking out of a plurality of linear workpieces placed in a three-dimensional space one by one,
After the gripping means grips the workpiece, the gripping means is controlled so that the workpiece gripped by the gripping means moves along a circular or rectangular trajectory ;
The gripping means is controlled so that the workpiece gripped by the gripping means revolves around a circular or rectangular orbit and at least one of the following is performed: the workpiece gripped by the gripping means is rotated on its axis; and the workpiece gripped by the gripping means is vibrated.
Control device.

Images