JP6632224B2 - Holding control device, holding control method and program - Google Patents

Description

  The present invention provides a gripping control device and a gripping control method for controlling gripping by a robotic hand when gripping and removing a workpiece to be gripped from a plurality of workpieces with a robotic hand, and the gripping control method. It relates to a program to be executed by a computer.

  2. Description of the Related Art Conventionally, when a workpiece attached to a robot is gripped by a hand, a technique for controlling the robot so that the workpiece can be held in a stable state has been proposed (for example, see Patent Document 1 below). Specifically, Patent Literature 1 below describes a technique for controlling a robot hand so that the number of contact points between the workpiece and the robot hand increases.

JP 2007-125653 A JP 2011-22991 A

  However, in the technique described in Patent Document 1, for example, even when a workpiece to be gripped interferes with another workpiece in a workpiece supplied in a bulk state, the point of contact with the robot hand is increased. The workpiece to be gripped is gripped so as to increase the number. Therefore, when an operation of removing the gripped work from the box or the like is performed, a force based on the removal operation acts on another work and interferes with the work, causing a problem such as dropping or breakage of the gripped work and the other work. there is a possibility.

  The present invention has been made in view of such a problem, and when taking out a gripped work from a box or the like, it is possible to reduce a possibility that a problem such as dropping or breakage of the gripped work and another work occurs. The purpose is to provide a mechanism.

The gripping control device of the present invention is a gripping control device that controls gripping by the robot hand when gripping and taking out a workpiece to be gripped by a robot hand from a plurality of workpieces, Work information holding means for holding work information relating to a given work, holding form information holding means for holding holding form information relating to a holding form of the robot hand, and a work in the plurality of works obtained by an imaging device Based on the image and the work information, a recognition unit for recognizing grip position information on a grip position on the grip target work and surrounding information on a surrounding situation on the grip target work, and the grip position information, Based on the surrounding information and the movable direction of the work in the gripping form information, each gripping method in a plurality of gripping methods is And worth holding method evaluation unit, in accordance with the evaluation value of each of the gripping method is evaluated by the holding method evaluation unit, the holding method determining means for determining the holding method at the time of gripping a workpiece of the gripping target, the Control means for controlling gripping of the workpiece to be gripped by the robot hand based on the gripping method determined by the gripping method determining means .
The present invention also includes a grip control method by the grip control device described above, and a program for causing a computer to execute the grip control method.

  According to the present invention, when taking out a gripped work from a box or the like, it is possible to reduce the possibility that a problem such as dropping or breakage of the gripped work and other works occurs.

FIG. 1 is a schematic diagram illustrating an example of a schematic configuration of a grip control system according to a first embodiment of the present invention. FIG. 2 is a schematic diagram illustrating the first embodiment of the present invention and illustrating a case where a robot hand takes out a workpiece to be gripped from a box based on a gripping method determined by a gripping control device illustrated in FIG. 1. It is a block diagram showing an example of an internal configuration of a grasp control device concerning a 1st embodiment of the present invention. FIG. 4 illustrates the first embodiment of the present invention, and illustrates an example of a method of holding work information held in a work information holding unit illustrated in FIG. 3 and an example of items included in the work information. FIG. 4 is a diagram illustrating a holding method of holding form information held in a holding state information holding unit illustrated in FIG. 3, and an example of items included in the holding form information. FIG. 4 is a diagram for describing grip form information held by a grip form information holding unit illustrated in FIG. 3. FIG. 4 is a schematic diagram for describing a recognition process of grip position information by a grip position recognition unit illustrated in FIG. 3. FIG. 4 is a schematic diagram for explaining a surrounding information recognition process performed by an environment recognition unit illustrated in FIG. 3. FIG. 4 is a schematic diagram for explaining an evaluation process of each gripping method by a gripping method evaluation unit illustrated in FIG. 3. 5 is a flowchart illustrating an example of a processing procedure of a grip control method by the grip control device according to the first embodiment of the present invention. FIG. 5 is a diagram illustrating a method of holding work information held in a work information holding unit illustrated in FIG. 3 and an example of items included in the work information according to the second embodiment of this invention. FIG. 12 is a schematic diagram for explaining the work information shown in FIG. 11. FIG. 14 is a schematic diagram illustrating a third embodiment of the present invention and illustrating the processing content of an environment recognition unit and the processing content of a gripping method evaluation unit illustrated in FIG. 3. FIG. 14 is a diagram illustrating the fourth embodiment of the present invention, illustrating a method of holding work information held in the work information holding unit illustrated in FIG. 3, and an example of items included in the work information. FIG. 14 is a schematic diagram illustrating a fourth embodiment of the present invention and illustrating processing contents of a gripping method evaluation unit illustrated in FIG. 3. FIG. 14 is a schematic diagram illustrating a fifth embodiment of the present invention and illustrating processing contents of an environment recognition unit and processing contents of a gripping method evaluation unit illustrated in FIG. 3. FIG. 14 is a characteristic diagram illustrating a fifth embodiment of the present invention, in which forces acting on a workpiece to be grasped by another workpiece, which is a surrounding obstacle, are recorded in time series.

  Hereinafter, an embodiment (embodiment) for carrying out the present invention will be described with reference to the drawings.

(First embodiment)
First, a first embodiment of the present invention will be described.

  In the present embodiment, based on the gripping position information on the gripping position on the workpiece to be gripped, the surrounding information on the surrounding state of the workpiece to be gripped, and the gripping form information on the gripping form of the robot hand, Determine the gripping method when gripping the work. Thus, when the gripped work is taken out of the box or the like, the influence of the gripped work on other works can be reduced, so that a problem such as dropping or breakage of the gripped work and other works may occur. Properties can be reduced.

  FIG. 1 is a schematic diagram illustrating an example of a schematic configuration of a grip control system 100 according to the first embodiment of the present invention. The grip control system 100 includes a grip control device 110, a robot 120, and an imaging device 130, as shown in FIG.

  The gripping control device 110 is a device that controls the gripping by the hand 121 of the robot 120 when the workpiece 201 to be gripped is gripped by the hand 121 of the robot 120 and taken out of the box 210 from the piled works 200. is there.

  The robot 120 is a device that, based on the control of the grip control device 110, performs an operation of gripping the workpiece 201 to be gripped from the piled workpieces 200 and taking it out of the box 210 with the hand 121. In the work area of the robot 120, there are works 200 stacked in bulk and a box 210 for storing the works 200.

  The imaging device 130 is a device that, based on the control of the grip control device 110, captures images of the works 200 stacked in the box 210 and generates a work image of the stacked works.

Next, an operation flow in the grip control system 100 shown in FIG. 1 will be described.
First, the grip control device 110, based on the bulk-stacked work image captured by the imaging device 130 and the workpiece information on the workpiece 201 to be gripped, based on gripping position information on a gripping position on the workpiece 201 to be gripped, and Then, the surrounding information on the surrounding state of the workpiece 201 to be grasped is recognized.

  Thereafter, the gripping control device 110 performs a gripping method when gripping the workpiece 201 to be gripped based on the gripping position information described above, the surrounding information described above, and the gripping form information on the gripping form of the hand 121 of the robot 120. To determine.

Thereafter, the grip control device 110 controls the robot 120 and the hand 121 based on the determined gripping method of the workpiece 201 to be gripped, and controls gripping of the robot 201 by the hand 121 with respect to the workpiece 201 to be gripped. In response to this, the robot 120 grips the workpiece 201 to be gripped by the hand 121 and takes out the workpiece 201 to be gripped from the box 210.
In the present embodiment, a case will be described in which a workpiece to be grasped is taken out of a pile of workpieces. However, in the present invention, even when a plurality of aligned workpieces are taken out of the workpiece to be grasped, interference occurs between the workpieces. If applicable, it is applicable.

Next, an example in which the work 201 to be gripped is taken out of the box 210 by the determined gripping method will be described with reference to FIG.
FIG. 2 is a schematic diagram illustrating a first embodiment of the present invention, in which a robot hand 121 takes out a workpiece 201 to be gripped from a box 210 based on a gripping method determined by a gripping control device 110 illustrated in FIG. FIG. 2A, 2B, and 2C show a process in which the robot hand 121 takes out the workpiece 201 to be grasped from the box 210 in time series.

  At the time of FIG. 2A, the workpiece 201 to be gripped among the workpieces 200 stacked in bulk is gripped by the hand 121. Further, of the workpieces 200 stacked in bulk, another work 301 which is a work around the work 201 to be grasped is placed in the box 210 in a state where the work 301 interferes with the work 201 to be grasped.

  Subsequently, the hand 121 moves in the Z direction of the coordinate system 302 and attempts to take out the grasped work 201 to be grasped from the box 210. Here, when the hand 121 is moved in the Z direction in a state where the work 201 to be grasped is not relatively movable with respect to the hand 121, the other work 301 is lifted by the work 201 to be grasped, and It drops off from 210. However, in the example illustrated in FIG. 2, the gripping control device 110 selects a gripping method that allows the workpiece 201 to be gripped to move relative to the hand 121.

  Therefore, at the time of FIG. 2B, the work 201 to be grasped is relatively movable with respect to the hand 121 due to the weight of the other work 301.

  As a result, the work 201 to be gripped can be taken out of the box 210 without dropping the other work 301 from the box 210 at the time of FIG.

Next, an internal configuration of the grip control device 110 will be described with reference to FIG.
FIG. 3 is a block diagram illustrating an example of an internal configuration of the grip control device 110 according to the first embodiment of the present invention. FIG. 3 shows the robot 120 and its hand 121 and the imaging device 130 shown in FIG. 1 in addition to the grip control device 110. In addition, FIG. 3 does not illustrate the work 200 and the box 210 that are stacked in bulk as illustrated in FIG.

  As shown in FIG. 3, the grip control device 110 includes a control unit 111 and a holding unit 117. As shown in FIG. 3, the control unit 111 includes a grip position recognition unit 112, an environment recognition unit 113, a grip method evaluation unit 114, a grip method determination unit 115, and a robot control unit 116. As shown in FIG. 3, the holding unit 117 includes a work information holding unit 118 and a gripping form information holding unit 119.

  The work information holding unit 118 holds work information relating to the work 200 (including the work 201 to be gripped) that has been piled up in bulk. The grip mode information holding unit 119 holds grip mode information relating to the grip mode when the robot 120 grips the hand 121.

  The gripping position recognition unit 112 obtains gripping position information on a gripping position of the workpiece 201 to be gripped based on the bulk-stacked work image captured by the imaging device 130 and the work information held in the work information holding unit 118. Recognize. The environment recognizing unit 113 is configured to generate a bulk object image captured by the imaging device 130, the grip position information recognized by the grip position recognizing unit 112, and the work information held in the work information holding unit 118. Recognize surrounding information relating to the surrounding situation in the work 201. The grip position recognition unit 112 and the environment recognition unit 113 constitute a recognition unit in the present invention.

  The gripping method evaluation unit 114 is configured to generate a plurality of pieces of information based on the gripping position information recognized by the gripping position recognition unit 112, the surrounding information recognized by the environment recognition unit 113, and the gripping shape information held by the gripping shape information holding unit 119. The evaluation of each gripping method in the gripping method is performed. Specifically, in the present embodiment, the gripping method evaluation unit 114 determines each of the gripping methods in a plurality of gripping methods based on the gripping position information, the surrounding information, and the movable direction of the work in the gripping form information. Is evaluated.

  The gripping method determination unit 115 performs gripping based on the gripping position information recognized by the gripping position recognition unit 112, the surrounding information recognized by the environment recognition unit 113, and the gripping shape information held by the gripping shape information holding unit 119. A gripping method for gripping the target work 201 is determined. Specifically, the gripping method determination unit 115 determines a gripping method when gripping the workpiece 201 to be gripped, according to the evaluation value of each gripping method evaluated by the gripping method evaluation unit 114.

  The robot control unit 116 controls the robot 120 and the hand 121 based on the gripping method determined by the gripping method determination unit 115, and controls gripping of the robot 120 by the hand 121 on the workpiece 201 to be gripped.

  Next, each internal configuration of the grip control device 110 shown in FIG. 3 will be described in detail.

First, a method of holding work information held in the work information holding unit 118 and items included in the work information will be described with reference to FIG.
FIG. 4 illustrates the first embodiment of the present invention, and illustrates an example of a method of holding work information held in the work information holding unit 118 illustrated in FIG. 3 and an example of items included in the work information.

  As an example of a method of holding work information by the work information holding unit 118, the work information is held in an XML (Extensible Markup Language) format as shown in FIG. FIG. 4 shows work information of the work 201 to be grasped.

  In the work information held in the work information holding unit 118, the work information of one work is described in an item surrounded by a tag <object>. Further, as items included in the tag <object>, a tag <name>, a tag <shape>, and a tag <size> are shown.

  The tag <name> describes information on the name of the work.

  The tag <shape> describes information on the shape of the workpiece. That is, the shape information of the workpiece 201 to be grasped is described in the tag <shape> of the workpiece information. In the present embodiment, since the work 201 to be grasped is a rectangular parallelepiped, the shape information is a box.

  The tag <size> describes dimension information of the work. Specifically, in the tag <xyz> included in the tag <size>, the dimensions of each side of the above-described rectangular parallelepiped are described.

Next, with reference to FIGS. 5 and 6, a description will be given of a holding method of holding form information held in the holding form information holding unit 119 and items included in the holding form information.
FIG. 5 is a diagram illustrating an example of a holding method of holding form information held in the holding state information holding unit 119 illustrated in FIG. 3 and an example of items included in the holding form information. FIG. 6 is a diagram for describing gripping form information held in the gripping form information holding unit 119 shown in FIG.

First, the definition of the holding form in the present invention will be described.
The holding form is the position and posture of the hand 121 of the robot 120 with respect to the work 201 to be held when the hand 121 of the robot 120 holds the work 201 to be held, and the movable direction of the hand 121 of the robot 120.

  As an example of a holding method of the holding form information held in the holding state information holding unit 119, the holding form information is held in an XML format as shown in FIG.

  In the holding form information held in the holding state information holding unit 119, one holding form is described in an item surrounded by a tag <grasp>. Further, as the items included in the tag <grasp>, a tag <name>, a tag <object name>, a tag <xyz>, a tag <rpy>, and a tag <move xyz> are shown.

  In the tag <name>, information on the name in the gripping form is described.

  The tag <object name> describes information on the name of the workpiece 201 to be grasped. In the example shown in FIG. 5, a work 1 representing the work 201 to be grasped is described.

  In the tag <xyz> and the tag <rpy>, the position / posture of the hand coordinate system (right hand system) based on the work coordinate system (right hand system) of the work 201 to be gripped described in <object name>. Information is described. Specifically, the tag <xyz> represents the position of the hand coordinate system based on the work coordinate system, and the tag <rpy> represents the posture of the hand coordinate system based on the work coordinate system. In the example shown in FIG. 5, three numbers separated by spaces are described in the tag <xyz>, each corresponding to XYZ in the work coordinate system, and the unit is m. Further, in the example shown in FIG. 5, three numbers separated by spaces are also described in the tag <rpy>, which represent roll, pitch, and yaw, respectively, and have units of radians.

  In the tag <move xyz>, information on the movable direction of the workpiece 201 described in <object name> with respect to the workpiece coordinate system of the workpiece 201 to be grasped is described. In the example shown in FIG. 5, the movable direction of the work of grasp1 is the Y-axis direction, and the movable direction of the work of grasp2 is the X-axis direction.

  Here, FIG. 6A is a diagram showing the information of grasp1 in FIG. 5, and FIG. 6B is a diagram showing the information of grasp2 in FIG. In FIG. 6, a work coordinate system 601 and a hand coordinate system 602 are defined, and arrows 603 and 604 indicate the movable directions of the work, respectively.

Next, the grip position recognition unit 112 shown in FIG. 3 will be described with reference to FIG.
FIG. 7 is a schematic diagram for explaining the recognition processing of the grip position information by the grip position recognition unit 112 shown in FIG.

  The grip control device 110 illustrated in FIG. 7 (the grip position recognition unit 112 in FIG. 3) performs, based on the bulk-stacked work image captured by the imaging device 130 and the work information stored in the work information storage unit 118, The grip position information relating to the grip position of the workpiece 201 to be gripped is recognized. Here, as a method of recognizing the position / posture of an object, many methods have been conventionally proposed, and for example, the method described in Patent Document 2 described above may be used.

  Subsequently, the grip position recognition unit 112 illustrated in FIG. 3 determines whether or not the robot 120 can generate a path for moving the hand 121 to a grippable position set on the workpiece 201 to be gripped. Here, as a method of calculating the route, a general RRT (Rapidly exploring Random Tree) can be used as a route searching method of the robot. In this case, the reason why the RRT is used is that even a complicated route can be searched for in a short time. Then, when a path for the robot 120 to move the hand 121 to a grippable position set on the workpiece 201 to be gripped can be generated, the gripping position recognition unit 112 determines the position at which the workpiece 201 to be gripped can be gripped. 701 is recognized as grip position information. Then, the grip position recognition unit 112 outputs the grip position information to the environment recognition unit 113 and the grip method evaluation unit 114.

Next, the environment recognition unit 113 shown in FIG. 3 will be described with reference to FIG.
FIG. 8 is a schematic diagram for explaining the surrounding information recognition process performed by the environment recognition unit 113 illustrated in FIG.

  When taking out the work 201 to be grasped recognized by the grasp position recognizing unit 112, the environment recognizing unit 113 recognizes an obstacle (a surrounding work or the like) around which the work 201 to be grasped is placed. Hereinafter, a method of recognizing a surrounding obstacle (surrounding work or the like) on which the work 201 to be grasped is placed will be described with reference to FIG.

  FIG. 8 illustrates an example of a bulk-stacked work image obtained by capturing images of the workpieces 200 stacked in the box 210 by the imaging device 130. In the present embodiment, the environment recognizing unit 113 holds the bottom surface 801 of the box 210 in advance before capturing the work 201 to be gripped and the other works 802, 803, 804, and 805 in the box 210. I do. Then, the environment recognition unit 113 calculates a difference between the bulk-stacked work image and the bottom surface image. Then, based on the difference image, the environment recognizing unit 113 masks a part of the work 201 to be grasped, the position and orientation of which is recognized by the grasping position recognizing unit 112. The arrangement information of the other works 802 to 805 as objects is recognized. That is, in the present embodiment, the environment recognizing unit 113 recognizes, as the surrounding information relating to the surrounding state of the work 201 to be grasped, the arrangement information of the surrounding work and the like in the work 201 to be grasped.

  Next, the environment recognition unit 113 calculates a contact point 806 between the workpiece 201 to be grasped and other workpieces 802, 803, 804, and 805 as surrounding obstacles. Here, the contact point 806 between the workpiece 201 to be gripped and the other workpieces 802 to 805 as surrounding obstacles in FIG. 8 is a representative point, and a point having a similar shape shown in FIG. Contact points between the work 201 to be grasped and other works 802 to 805 which are obstacles around the work 201. In the present embodiment, the environment recognition unit 113 determines, as surrounding information relating to the surrounding state of the work 201 to be grasped, contact points between the work 201 to be held and the other works 802 to 805 as surrounding obstacles. It is also possible to apply a mode of recognizing the information of 806.

Next, the gripping method evaluation unit 114 shown in FIG. 3 will be described with reference to FIG.
FIG. 9 is a schematic diagram for explaining an evaluation process of each gripping method by the gripping method evaluation unit 114 shown in FIG.

  As described above, the gripping method evaluation unit 114 includes the gripping position information recognized by the gripping position recognition unit 112, the surrounding information recognized by the environment recognition unit 113, and the gripping shape information held by the gripping shape information holding unit 119. Based on the above, each gripping method in a plurality of gripping methods is evaluated. Here, in the present invention, the gripping method means a combination of a gripping position and a gripping form.

  FIGS. 9A and 9B are images of a work piece pile obtained by imaging the work pieces 200 stacked in the box 210 by using the imaging device 130, respectively. FIGS. 9A and 9B show states in which two gripping forms input from the gripping form information holding unit 119 are applied to the gripping position information recognized by the gripping position recognizing unit 112, respectively. Is shown.

  9 (c) and 9 (d) are diagrams showing the evaluation method of the gripping method in FIGS. 9 (a) and 9 (b), respectively. 9C and 9D, a contact point 806 is a contact point between the work 201 to be grasped and another work which is a surrounding obstacle, calculated by the environment recognition unit 113 described above. 9 (c) and 9 (d) show an evaluation range 901 and an evaluation range 902 which are set when the evaluation is performed by the gripping method evaluation unit 114, respectively. Is defined for each gripping method. In addition, in the case of the examples illustrated in FIGS. 9C and 9D, the evaluation range 901 and the evaluation range 902 are, for each gripping method, the work 201 to be gripped that exists in the movable direction of the work 201 to be gripped. Is set on the surface.

  Next, the gripping method evaluation unit 114 sets the reciprocal of the sum of the numbers of the contact points 806 belonging to the evaluation range 901 and the evaluation range 902 as the evaluation value of the gripping method related to the evaluation range. Specifically, the evaluation value V is represented by the following equation (1). In the following equation (1), “p” is the number of contact points 806 in the evaluation range.

  Here, the evaluation value in the case of FIG. 9C is V = １ ／, and the evaluation value in the case of FIG. 9D is V = １ ／. That is, the smaller the number of contact points 806 existing in the evaluation range 901 and the evaluation range 902, the higher the evaluation value.

  Then, the gripping method evaluation unit 114 outputs the calculated evaluation value of each gripping method to the gripping method determination unit 115.

  Then, the gripping method determination unit 115 determines a gripping method based on the evaluation value input from the gripping method evaluation unit 114. Specifically, in the present embodiment, the gripping method determination unit 115 determines a gripping method having a high evaluation value input from the gripping method evaluation unit 114 as a gripping method when actually gripping the workpiece 201 to be gripped. I do. In the example illustrated in FIG. 9, the gripping method in FIG. 9A is determined as a gripping method when the workpiece 201 to be gripped is actually gripped.

  Then, the gripping method determination unit 115 outputs information on the determined gripping method to the robot control unit 116.

  Then, the robot control unit 116 controls the robot 120 and the hand 121 based on the gripping method input from the gripping method determination unit 115, and controls the robot 120 to grip the workpiece 201 to be gripped by the hand 121.

Next, a processing procedure of a grip control method by the grip control device 110 according to the present embodiment will be described with reference to FIG.
FIG. 10 is a flowchart illustrating an example of a processing procedure of a grip control method by the grip control device 110 according to the first embodiment of the present invention.

  First, in step S <b> 1001, the gripping position recognition unit 112 recognizes the work 201 to be gripped based on the bulky work image input from the imaging device 130 and the work information held in the work information holding unit 118. I do.

  Subsequently, in step S1002, the gripping position recognition unit 112 sets the hand 121 to the gripping target based on the bulk work image input from the imaging device 130 and the work information held in the work information holding unit 118. The position where the work 201 can be gripped is recognized. That is, the grip position recognition unit 112 recognizes grip position information on the grip position of the workpiece 201 to be gripped.

  Subsequently, in step S1003, the environment recognizing unit 113 based on the bulky work image input from the imaging device 130, the grip position information recognized in step S1002, and the work information held in the work information holding unit 118. Then, the surrounding information on the surrounding state of the workpiece 201 to be grasped is recognized.

  Subsequently, in step S1004, the gripping method evaluation unit 114 determines the gripping position information recognized in step S1002, the surrounding information recognized in step S1003, and the gripping form information held in the gripping form information holding unit 119. The evaluation of each gripping method among a plurality of gripping methods is performed, and the above-described evaluation value is calculated for each gripping method.

  In step S1005, first, the gripping method determination unit 115 determines a gripping method having the highest evaluation value among the gripping methods evaluated in step S114 as a gripping method when the workpiece 201 to be gripped is actually gripped. (Gripping method determination step). Next, the robot control unit 116 controls the robot 120 and the hand 121 based on the gripping method input from the gripping method determining unit 115, and the gripping method input from the gripping method determining unit 115 to the hand 121 of the robot 120. The workpiece 201 to be gripped is gripped with.

  Subsequently, in step S1006, the control unit 111 determines whether or not the transfer of the workpiece 201 to be grasped by the hand 121 of the robot 120 has been completed.

As a result of the determination in step S1006, when the transfer of the workpiece 201 to be grasped by the hand 121 of the robot 120 is not completed (S1006 / NO), the process proceeds to step S1007.
In step S1007, the robot control unit 116 controls the robot 120 and the hand 121 to perform a process of transporting the workpiece 201 to be grasped. Thereafter, the process returns to step S1006.

  On the other hand, as a result of the determination in step S1006, when the transfer of the workpiece 201 to be grasped by the hand 121 of the robot 120 is completed (S1006 / YES), the processing of the flowchart in FIG. 10 ends.

In the present embodiment, grip position information regarding a grip position on the workpiece 201 to be gripped, surrounding information regarding a surrounding state of the workpiece 201 to be gripped, and grip mode information regarding a grip mode of the hand 121 of the robot 120 (for example, The gripping method when gripping the workpiece 201 to be gripped is determined based on the information on the movable direction of the workpiece 201 to be gripped in the gripping mode of the hand 121).
According to this configuration, when removing the gripped work from the box or the like, the influence of the gripped work on other works can be reduced. As a result, it is possible to reduce the possibility that a problem such as dropping or breakage of the gripped work and other works occurs.

(Second embodiment)
Next, a second embodiment of the present invention will be described.

  In the first embodiment described above, the form in which the work information held in the work information holding unit 118 includes the shape information of the work 200 (including the work 201 to be gripped) that has been piled up is described. In the second embodiment, as the work information held in the work information holding unit 118, in addition to the above-described shape information, information on slipperiness, information on fragility, and entanglement in the work 201 to be gripped and the like are included. It shall include information on ease and information on flexibility. Then, by using the above-described work information held in the work information holding unit 118 to determine a gripping method when gripping the work 201 to be gripped, it is possible to determine whether the gripped work or another work is dropped or damaged. It is possible to reduce the possibility that a failure occurs.

  The schematic configuration of the grip control system according to the second embodiment is the same as the schematic configuration of the grip control system 100 according to the first embodiment shown in FIG. The internal configuration of the grip control device according to the second embodiment is the same as the internal configuration of the grip control device 110 according to the first embodiment shown in FIG.

  Although the internal configuration of the grip control device 110 according to the second embodiment is as shown in FIG. 3, the first embodiment differs from the first embodiment in the content of the work information held in the work information holding unit 118 and the grip. Since the processing contents of the method evaluation unit 114 are different, the different parts will be described below. The description of the other parts other than the different part is omitted.

A method of holding work information held in the work information holding unit 118 according to the present embodiment and items included in the work information will be described with reference to FIGS.
FIG. 11 illustrates the second embodiment of the present invention, and is a diagram illustrating an example of a method of holding work information held in the work information holding unit 118 illustrated in FIG. 3 and an example of items included in the work information. FIG. 12 is a schematic diagram for explaining the work information shown in FIG.

  As shown in FIG. 11, in addition to the work information items in the first embodiment described with reference to FIG. 4, the work information holding unit 118 of the present embodiment includes a tag <plane>, a tag <slipperiness>, Work information including items of a tag <flimness>, a tag <complicateness>, and a tag <flexibleness> is held.

The tag <plane> is a tag representing one surface constituting the shape of the workpiece 201 to be grasped. The tag <plane> will be described with reference to FIG.
Surfaces 1201, 1202, 1203 and 1204 shown in FIG. 12 are surfaces constituting the work 201 to be grasped, respectively. The planes 1201, 1202, 1203, and 1204 shown in FIG. 12 correspond to the plane <-1, plane_2, plane_3, and plane_4 of the tag <plane> shown in FIG. 11, respectively. The tag <plane> has a tag indicating the state of the surface of the workpiece 201 to be grasped, as shown in FIG. The state of the work surface indicated by each tag will be described below.

  The tag <slippersines> indicates information on slipperiness, the tag <flimscenes> indicates information on breakability, the tag <complicateness> indicates information on ease of entanglement, and the tag <flexibleness> indicates information on flexibility. . The values set for these tags will be described later in the description of the gripping method evaluation unit 114.

Next, the gripping method evaluation unit 114 according to the present embodiment will be described.
The gripping method evaluation unit 114 according to the present embodiment is configured to calculate the evaluation value in the gripping method evaluation unit 114 according to the first embodiment. This further considers the information of entanglement easiness and the information of flexibility. Specifically, the evaluation value V is represented by the following equation (2).

  This evaluation value V is calculated for each gripping method. In this equation (2), “slp” is a coefficient representing slipperiness, “flm” is a coefficient representing fragility, “com” is a coefficient representing entanglement, and “flx” is a coefficient representing flexibility. is there. In the equation (2), the variable in which “v” follows each coefficient name is a value assigned to the tag <slippersines>, the tag <flimness>, the tag <complicateness>, and the tag <flexibleness> in FIG. It is. With this, for each surface of the workpiece 201 to be gripped, slipperiness, fragility, entanglement, and flexibility can be set. In the equation (2), the variable in which n follows each coefficient name is the number of contact points described with reference to FIG. In the equation (2), “other” is the number of contact points on a surface on which no tag indicating the surface state exists.

  In the present embodiment, as the work information held in the work information holding unit 118, in addition to the shape information described in the first embodiment, information on slipperiness, information on breakability, information on ease of entanglement. , And flexibility information. Then, by using the above-described work information held in the work information holding unit 118 to determine a gripping method when gripping the work to be gripped, a problem such as dropping or breakage of the gripped work and another work is determined. Can be reduced.

(Third embodiment)
Next, a third embodiment of the present invention will be described.

  In the above-described first embodiment, as the surrounding information recognized by the environment recognition unit 113, the arrangement information of the surrounding work and the like of the holding target work 201 and the information of the contact point between the holding target work 201 and the surrounding work and the like are included. Was applied. In the third embodiment, as the surrounding information recognized by the environment recognizing unit 113, information on the force acting on the work 201 to be grasped from the surrounding work or the work 201 to be grasped using the hand 121 of the robot 120 is used. This is a form in which information on force generated by moving is applied. Then, by using the above-described surrounding information recognized by the environment recognition unit 113 to determine a gripping method when gripping the workpiece 201 to be gripped, the workpiece gripped when the workpiece 201 to be gripped is taken out is movable. By doing so, the influence of the gripped work on other works can be reduced. As a result, it is possible to reduce the possibility that a problem such as dropping or breakage of the gripped work or another work occurs.

  The schematic configuration of the grip control system according to the third embodiment is the same as the schematic configuration of the grip control system 100 according to the first embodiment shown in FIG. The internal configuration of the grip control device according to the third embodiment is the same as the internal configuration of the grip control device 110 according to the first embodiment shown in FIG.

  Although the internal configuration of the grip control device 110 according to the third embodiment is as shown in FIG. 3, the first embodiment differs from the first embodiment in the processing contents of the environment recognition unit 113 and the processing contents of the grip method evaluation unit 114. Since the contents are different, the different parts will be described below. The description of the other parts other than the different part is omitted.

The processing content of the environment recognition unit 113 and the processing content of the gripping method evaluation unit 114 according to the present embodiment will be described with reference to FIG.
FIG. 13 shows the third embodiment of the present invention, and is a schematic diagram for explaining the processing content of the environment recognition unit 113 and the processing content of the gripping method evaluation unit 114 shown in FIG. Here, FIGS. 13A and 13B show a state in which the hand 121 of the robot 120 is gripping the workpiece 201 to be gripped by different gripping methods, respectively.

  13A and 13B, the environment recognition unit 113 of the present embodiment recognizes another work 804 that is an obstacle around the work 201 to be grasped. Further, the environment recognition unit 113 of the present embodiment calculates and recognizes information on the force 1301 generated at the contact point 806 as surrounding information.

  Next, a method of calculating an evaluation value by the gripping method evaluation unit 114 of the present embodiment will be described. Specifically, in the present embodiment, the gripping method evaluation unit 114 calculates the evaluation value V using the following equation (3). The evaluation value V expressed by the expression (3) indicates the reciprocal of the force generated in the hand 121 by the force acting on the work 201 to be grasped from the surrounding work.

  In the expression (3), “free” is a coefficient determined by the degree of freedom of each gripping method in FIGS. 13A and 13B. In FIG. 13A, a force is applied in a direction in which the workpiece 201 to be grasped moves, so that the value of free is lower than 1. For example, in the gripping method of FIG. 13A, the value of free is set to 0.4. In FIG. 13B, a force is applied in a direction different from the direction in which the workpiece 201 to be grasped moves, and thus the value of free is 1. Further, in this equation (3), “r” is from the contact points 1304 and 1305 with the hand 121 to the contact point 806 in the gripping method of the hand 121 in each of FIGS. 13A and 13B. Is the length of In the equation (3), “F” is a force F generated at the contact point 806. Further, in the equation (3), “θ” is an angle formed between a direction perpendicular to the direction of the force F and r.

  Further, in the present embodiment, while the hand 121 is actually being gripped by the hand 121, the hand 121 is slightly moved as shown in directions 1306 and 1307 shown in FIG. A force may be applied to the work 201 to measure and evaluate the generated force. That is, in this case, the environment recognizing unit 113 recognizes information on force generated by moving the workpiece 201 to be grasped using the hand 121 of the robot 120 as surrounding information. When this mode is adopted, assuming that a force generated when the hand 121 is minutely moved is F, the gripping method evaluation unit 114 calculates the evaluation value V using the following equation (4).

  In the present embodiment, as the surrounding information recognized by the environment recognizing unit 113, information on force acting on the work 201 to be grasped from the surrounding work or the hand 121 of the robot 120 is used to move the work 201 to be grasped. We apply the information of the force that arises. Then, by using this surrounding information to determine a gripping method when gripping the workpiece 201 to be gripped, the workpiece that is gripped when the workpiece 201 to be gripped is taken out is moved so that the workpiece that is gripped can be moved to another location. The effect on the work can be reduced. As a result, it is possible to reduce the possibility that a problem such as dropping or breakage of the gripped work or another work occurs.

(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described.

  In the first embodiment described above, the form in which the work information held in the work information holding unit 118 includes the shape information of the work 200 (including the work 201 to be gripped) that has been piled up is described. In the fourth embodiment, it is assumed that the work information held in the work information holding unit 118 includes information on the priority of the gripping method in the work 201 to be gripped. Then, using the above-described work information held in the work information holding unit 118, a gripping method when gripping the work 201 to be gripped is determined. Accordingly, when the gripped work is taken out, the gripped work moves, so that the influence of the gripped work on other works can be reduced. As a result, it is possible to reduce the possibility that a problem such as dropping or breakage of the gripped work or another work occurs.

  The schematic configuration of the grip control system according to the fourth embodiment is the same as the schematic configuration of the grip control system 100 according to the first embodiment shown in FIG. The internal configuration of the grip control device according to the fourth embodiment is the same as the internal configuration of the grip control device 110 according to the first embodiment shown in FIG.

  Although the internal configuration of the grip control device 110 according to the fourth embodiment is as shown in FIG. 3, the first embodiment differs from the first embodiment in the content of the work information held in the work information holding unit 118 and the grip. Since the processing contents of the method evaluation unit 114 are different, the different parts will be described below. The description of the other parts other than the different part is omitted.

First, a method of holding work information held in the work information holding unit 118 of the present embodiment and items included in the work information will be described with reference to FIG.
FIG. 14 illustrates the fourth embodiment of the present invention, and illustrates an example of a method of holding work information held in the work information holding unit 118 illustrated in FIG. 3 and an example of items included in the work information.

  The work information of the work information holding unit 118 according to the present embodiment is different from the work information according to the first embodiment described with reference to FIG. 4, and a tag <shape> illustrated in FIG. 14 expresses a three-dimensional shape. The file name of the file saved in the data format is specified. By specifying the shape of the work in this way, work information of a work having an arbitrary shape can be held.

  Further, as shown in FIG. 14, the work information holding unit 118 of the present embodiment holds work information including an item of a tag <priority>. This tag <priority> indicates information on the priority of the gripping method. In the example shown in FIG. 14, the smaller the value described, the higher the priority of the gripping method.

  As shown in FIG. 14, the tag <priority> includes a tag <xyz> indicating position information that can be held, and a tag <grasp_name> indicating grip mode information. At this time, the position of the workpiece coordinate system is described in the tag <xyz>, and the name given to the gripping form information held in the gripping form information holding unit 119 is described in the tag <grasp_name>.

Next, the gripping method evaluation unit 114 of the present embodiment will be described with reference to FIG.
FIG. 15 is a schematic diagram illustrating the fourth embodiment of the present invention and illustrating the processing contents of the gripping method evaluation unit 114 illustrated in FIG. 3. Here, FIGS. 15A and 15B show a state in which the hand 121 of the robot 120 grips the workpiece 1501 to be gripped by different gripping methods, respectively.

  In the gripping methods in FIGS. 15A and 15B, information on the priority is held by the work information holding unit 118. The work information holding unit 118 sets the priority of the gripping method illustrated in FIG. 15B to 2 and the priority of the gripping method illustrated in FIG. That is, in this example, the gripping method shown in FIG. 15A is set to have a higher priority than the gripping method shown in FIG.

Here, the reason why the gripping method shown in FIG. 15A has a higher priority than the gripping method shown in FIG. 15B will be described.
The work 1501 to be grasped is L-shaped as shown in FIG. Then, when a force is generated by another work on the L-shaped projection 1502 of the work 1501 to be grasped, the work 1501 to be grasped can be moved by the grasping method shown in FIG. On the other hand, in the gripping method shown in FIG. 15B, the workpiece 1501 to be gripped cannot move. That is, the hand 121 of the robot 120 grips the workpiece 1501 to be gripped so that when a force is applied to the workpiece 1501 to be gripped in a certain direction, the workpiece 1501 to be gripped can operate in the direction of the force. I do. By gripping in this manner, the force applied to the workpiece 1501 to be gripped can be released, and the influence on surrounding objects (including surrounding workpieces) can be reduced.

  Then, in the present embodiment, the gripping method evaluation unit 114 assigns a high evaluation value to the gripping method with a high priority and outputs the evaluation method to the gripping method determination unit 115.

  In the present embodiment, the work information held in the work information holding unit 118 includes information on the priority of the gripping method in the work 201 to be gripped. Then, using the above-described work information held in the work information holding unit 118, the holding method evaluation unit 114 evaluates the holding method, and determines the holding method when holding the work to be held based on the evaluation result. I am trying to do it. Accordingly, when the workpiece 1501 to be gripped is taken out, the gripped workpiece moves, and thus the influence of the gripped workpiece on other workpieces can be reduced. As a result, it is possible to reduce the possibility that a problem such as dropping or breakage of the gripped work or another work occurs.

(Fifth embodiment)
Next, a fifth embodiment of the present invention will be described.

  In the above-described first embodiment, as the surrounding information recognized by the environment recognition unit 113, the arrangement information of the surrounding work and the like of the grasping target work 201 and the information of the contact point between the holding target work 201 and the surrounding work and the like are included. Was applied. In the fifth embodiment, information on the force acting on the work 201 to be grasped from the surrounding work is applied as the surrounding information recognized by the environment recognition unit 113 when the work 201 to be grasped is taken out after being grasped. It is a form. The gripping method determination unit 115 determines a gripping method when gripping the workpiece 201 to be gripped, using the above-described surrounding information recognized by the environment recognition unit 113. Then, the robot control unit 116 determines the position and posture of the hand 121 so that the resistance against the direction of the force applied to the workpiece 201 by the hand 121 after the hand 121 grips the workpiece 201 is minimized. Control. Thereby, when taking out the work 201 to be grasped, the influence of the work 201 to be grasped on other works can be reduced. As a result, it is possible to reduce the possibility that a problem such as dropping or breakage of the gripped work or another work occurs.

  The schematic configuration of the grip control system according to the fifth embodiment is the same as the schematic configuration of the grip control system 100 according to the first embodiment shown in FIG. The internal configuration of the grip control device according to the fifth embodiment is the same as the internal configuration of the grip control device 110 according to the first embodiment shown in FIG.

  Although the internal configuration of the grip control device 110 according to the fifth embodiment is as shown in FIG. 3, the first embodiment differs from the grip control device 110 in the processing content of the environment recognition unit 113 and the processing content of the grip method evaluation unit 114. Since the contents are different, the different parts will be described below. The description of the other parts other than the different part is omitted.

The processing contents of the environment recognition unit 113 and the processing contents of the gripping method evaluation unit 114 according to the present embodiment will be described with reference to FIG.
FIG. 16 shows the fifth embodiment of the present invention, and is a schematic diagram for explaining the processing content of the environment recognition unit 113 and the processing content of the gripping method evaluation unit 114 shown in FIG. Here, FIGS. 16A and 16B show a state in which the hand 121 of the robot 120 is gripping the workpiece 201 to be gripped by different gripping methods, respectively.

  16A and 16B, the environment recognition unit 113 of the present embodiment recognizes another work 1605 which is an obstacle around the work 201 to be grasped. Further, the environment recognizing unit 113 of the present embodiment transmits information on the force acting on the work 201 to be grasped from another work 1605 which is a surrounding obstacle when the work 201 to be grasped is taken out after being grasped. Calculate and recognize as information.

The processing content of the present embodiment will be specifically described.
The transport path 1601 and the transport path 1602 in FIG. 16A and the transport path 1603 and the transport path 1604 in FIG. 16B respectively correspond to the gripping method in FIG. 16A and the gripping method in FIG. This is a transport path by the hand 121 set in advance for the gripping method.

  Here, the environment recognizing unit 113 simulates the operations of the transport path 1601, the transport path 1602, the transport path 1603, and the transport path 1604, and generates a time series of a force generated by a nearby obstacle acting on the workpiece 201 to be grasped. Is calculated. In the present embodiment, the environment recognizing unit 113 calculates the force of the surrounding obstacle acting on the work 201 to be grasped in a time-series manner. It may be.

  FIG. 17 shows a fifth embodiment of the present invention, and is a characteristic diagram in which forces acting on a work 201 to be grasped by another work 1605 as a surrounding obstacle are recorded in chronological order. Here, in FIG. 17, the horizontal axis indicates time (seconds), and the vertical axis indicates force (Newton). Also, in FIG. 17, a graph 1701 shows the characteristics of the transport route 1601 in FIG. 16A, a graph 1702 shows the characteristics of the transport route 1602 in FIG. 16A, and a graph 1703 shows the transport in FIG. The characteristic of the path 1603 is shown, and the graph 1704 shows the characteristic of the transport path 1604 in FIG.

  Then, the gripping method evaluation unit 114 of the present embodiment calculates the evaluation value V using the following equation (5). That is, the evaluation value V is the reciprocal of the maximum value of the time-series force value x.

  In the example shown in FIG. 17, the maximum value 1705 of the force in the graph 1704 is the minimum, and the evaluation value is the highest. That is, in the present embodiment, when a force is applied in a certain direction on the transport path after the work 201 to be grasped is gripped, the work 201 to be gripped is movable in the direction of the force. The work 201 is gripped. Further, by gripping in this manner, the force applied to the workpiece 201 to be gripped can be released on the transport path after gripping the workpiece 201 to be gripped, and the force applied to the other workpiece 1605 can be released. The effect can be reduced.

  In the present embodiment, information on the force acting on the work 201 to be grasped from the surrounding work when the work 201 to be grasped is taken out after grasping is applied as the surrounding information recognized by the environment recognition unit 113. . Then, using the surrounding information, a gripping method when gripping the workpiece 201 to be gripped is determined. Then, by controlling the position and posture of the hand so that the resistance against the direction of the force applied by the hand 121 to the work 201 to be gripped after the grip is performed, the gripped work and other works may be dropped. It is possible to reduce the possibility that a problem such as breakage occurs.

(Other embodiments)
The present invention supplies a program for realizing one or more functions of the above-described embodiments to a system or an apparatus via a network or a storage medium, and one or more processors in a computer of the system or the apparatus read and execute the program. This processing can be realized. Further, it can also be realized by a circuit (for example, an ASIC) that realizes one or more functions.
This program and a computer-readable storage medium storing the program are included in the present invention.

  It should be noted that the above-described embodiments of the present invention are merely examples of specific embodiments for carrying out the present invention, and the technical scope of the present invention should not be interpreted in a limited manner. It is. That is, the present invention can be implemented in various forms without departing from the technical idea or the main features.

Reference Signs List 100 grip control devices 110, 111 control unit, 112 grip position recognition unit, 113 environment recognition unit, 114 grip method evaluation unit, 115 grip method determination unit, 116 robot control unit, 117 holding unit, 118 work information storage unit, 119 grip Form information storage

Claims (15)

A gripping control device that controls gripping by the robot hand when gripping and extracting a workpiece to be gripped by a robot hand from a plurality of workpieces,
Work information holding means for holding work information related to the work to be gripped,
Gripping form information holding means for holding gripping form information relating to a gripping form in the robot hand,
Based on the work images of the plurality of works and the work information obtained by the imaging device, based on the work information, grasp position information on a grasp position on the work to be grasped and surrounding information on a surrounding situation in the work on the grasp target. Recognition means for recognizing
Gripping method evaluation means for evaluating each gripping method in a plurality of gripping methods, based on the movable direction of the workpiece in the gripping position information, the surrounding information and the gripping form information,
A gripping method determining unit that determines a gripping method when gripping the workpiece to be gripped , according to the evaluation value of each gripping method evaluated by the gripping method evaluating unit ,
A control unit that controls gripping of the workpiece to be gripped by the hand of the robot based on the gripping method determined by the gripping method determination unit . The recognition means ,
A grip position recognition unit that recognizes the grip position information based on the work image and the work information;
An environment recognizing means for recognizing the surrounding information based on the work image, the grip position information recognized by the grip position recognizing means , and the work information. Item 2. The grip control device according to Item 1. The work information, the gripping control device according to claim 1 or 2, characterized in that the information containing information of the shape of the gripping target work. The work information, the gripping control device according to claim 1 or 2, characterized in that the information including information on slipperiness of the gripping target work. The work information, the gripping control device according to claim 1 or 2, characterized in that the information including information of fragility in the gripping target work. The work information, the gripping control device according to claim 1 or 2, characterized in that the information including information of tangling ease in the gripping target work. The work information, the gripping control device according to claim 1 or 2, characterized in that the information including information of flexibility in the gripping target work. 3. The grip control device according to claim 1, wherein the work information is information including priority information of a gripping method of the grip target work. 4. The grip control device according to any one of claims 1 to 8 , wherein the surrounding information is arrangement information of a surrounding work in the work to be gripped. The grip control device according to any one of claims 1 to 8 , wherein the surrounding information is information on a point of the work to be gripped that is in contact with a surrounding work. The grip control device according to any one of claims 1 to 8 , wherein the surrounding information is information on a force acting on the work to be gripped from a surrounding work. Said peripheral information, the gripping control system according to any one of claims 1 to 8, characterized in that the information of the force caused by moving the workpiece of the gripping target using a hand of the robot. The said surrounding information is the information of the force which acts on the said to-be-held-object workpiece | work from the surrounding workpiece | work at the time of taking-out after the said to-be-held-target workpiece | work, The one of Claim 1 thru | or 8 characterized by the above-mentioned. A gripping control device according to Item. When grasping and taking out a workpiece to be grasped by a robot hand from a plurality of workpieces, the robot performs grasping control by the hand of the robot, and workpiece information holding workpiece information on the workpiece to be grasped. a holding unit, a holding pattern information holding means and the grasping control method by the gripping controller having a holding the holding pattern information according to the holding pattern in a hand of the robot,
Based on the work images of the plurality of works and the work information obtained by the imaging device, based on the work information, grasp position information on a grasp position on the work to be grasped and surrounding information on a surrounding situation in the work on the grasp target. A recognition step of recognizing
A gripping method evaluation step of evaluating each gripping method in a plurality of gripping methods based on the movable direction of the workpiece in the gripping position information, the surrounding information and the gripping form information,
According to the evaluation value of each gripping method evaluated in the gripping method evaluation step, a gripping method determining step of determining a gripping method when gripping the workpiece to be gripped,
A control step of controlling the gripping of the workpiece to be gripped by the hand of the robot based on the gripping method determined in the gripping method determining step. When grasping and taking out a workpiece to be grasped by a robot hand from a plurality of workpieces, the robot performs grasping control by the hand of the robot, and workpiece information holding work information relating to the workpiece to be grasped. a holding unit, a program for executing a gripping control method by grasping the control device provided with a gripping configuration information holding means for holding the holding pattern information according to the holding pattern in a hand of the robot into the computer,
Based on the work images of the plurality of works and the work information obtained by the imaging device, based on the work information, grasp position information on a grasp position on the work to be grasped and surrounding information on a surrounding situation in the work on the grasp target. A recognition step of recognizing
A gripping method evaluation step of evaluating each gripping method in a plurality of gripping methods based on the movable direction of the workpiece in the gripping position information, the surrounding information and the gripping form information,
According to the evaluation value of each gripping method evaluated in the gripping method evaluation step, a gripping method determining step of determining a gripping method when gripping the workpiece to be gripped,
And a control step of controlling the gripping of the workpiece to be gripped by the hand of the robot based on the gripping method determined in the gripping method determining step.