JP2013129005A - Gripping robot device, gripping operation control method and gripping operation control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for accurately recognizing the deviation of a gripping position of a workpiece as much as possible with a simple structure as compared with conventional art.SOLUTION: A gripping robot device 10 includes: a gripping part for gripping the workpiece placed on a reference surface; and a deviation amount calculation part for calculating the deviation amount of the gripping position of the workpiece by the gripping part from a reference position. The gripping part inclines the workpiece to bring one end of the workpiece into contact with the reference surface. At this time, the deviation amount calculation part calculates the deviation amount based on a positional relationship between the gripping part and the reference surface in a state that the workpiece inclined by the gripping part and the reference surface are brought into contact with each other.

Description

  The present invention relates to a gripping robot device, a gripping motion control method, and a gripping motion control device.

  Some industrial robots are equipped with a plurality of robot arms and perform fitting work and stacking work of objects (hereinafter referred to as “workpieces”).

  For example, Patent Document 1 discloses a technique for measuring the position and posture of a workpiece using an image sensor and controlling the hand based on the measurement result to grip the workpiece.

JP-A-8-224623

  However, even if the position and orientation of the workpiece are measured by an expensive measuring means (such as a camera) or complicated measurement processing as in the past, the position resolution is about 500 μm, and the position for gripping the workpiece (hereinafter referred to as “gripping”). A shift of several hundred μm may occur in the “position”. Of course, unless such a shift in the gripping position can be accurately recognized, it is not possible to accurately perform work fitting work and work stacking work that require accuracy.

  An object of this invention is to provide the technique which recognizes the shift | offset | difference of the holding position of a workpiece | work with high precision by a simpler structure than before.

  The present invention for solving the above-described problems is a gripping motion control device, wherein a gripping part that grips a workpiece placed on a reference surface, and a gripping position of the workpiece by the gripping part from a reference position are shifted. A displacement amount calculation unit that calculates an amount, wherein the gripping unit tilts the workpiece to bring one end of the workpiece into contact with the reference surface, and the displacement amount calculation unit is tilted by the gripping unit. In a state where the workpiece and the reference surface are in contact with each other, the deviation amount is calculated based on the positional relationship between the grip portion and the reference surface.

  Thereby, the shift | offset | difference of the holding position of a workpiece | work can be recognized with high precision by simpler structure than before.

  The gripper grips the workpiece with a direction parallel to the bottom surface of the workpiece (hereinafter referred to as “Y direction”) as a gripping direction, and the displacement amount calculation unit grips the workpiece by the gripper. For the position, the amount of deviation dx in the X direction orthogonal to the Y direction is expressed as dx = H / sin θ-L (where H is the distance from the reference surface of the gripping position, and L is the reference position of the workpiece) You may make it calculate using the numerical formula of the length to the said one end, (theta) represents the angle in which the said workpiece | work was inclined.

  In this way, it is possible to recognize the shift of the workpiece gripping position with high accuracy by a simple mathematical expression.

  The gripping portion includes a first gripping portion having a first gripping mechanism and a second gripping portion having a second gripping mechanism, and is gripped by the first gripping mechanism. The second gripping mechanism performs an operation of gripping one workpiece in a direction orthogonal to the gripping direction by the first gripping mechanism, and the shift amount calculation unit is configured such that the second gripping mechanism is the first gripping mechanism. In an operation of gripping in a direction orthogonal to the gripping direction of the first gripping mechanism, a direction orthogonal to the gripping direction of the first gripping mechanism generated in at least one of the first gripping part and the second gripping part The deviation amount may be calculated based on the force.

  Thereby, the shift | offset | difference of the holding position of a workpiece | work can be recognized with high precision by the method (2nd method) different from the above.

  Further, the deviation amount calculation unit may calculate the deviation amount using a mathematical formula of dx = Fx × β (where Fx is the force, dx is the deviation amount, and β is a coefficient). Good.

  In this way, even when the shift of the gripping position of the workpiece is calculated by the second method, the shift of the gripping position of the workpiece can be recognized with high accuracy by a simple mathematical formula.

  The gripping portion includes a first gripping portion having a first gripping mechanism and a second gripping portion having a second gripping mechanism, and is gripped by the first gripping mechanism. The second gripping mechanism performs an operation of gripping one workpiece in a direction orthogonal to the gripping direction by the first gripping mechanism, and the shift amount calculation unit is configured such that the second gripping mechanism is the first gripping mechanism. In an operation of gripping in a direction orthogonal to the gripping direction of the first gripping mechanism, a direction orthogonal to the gripping direction of the first gripping mechanism generated in at least one of the first gripping part and the second gripping part The second gripping mechanism is calculated in a state in which the displacement amount of the first workpiece is calculated based on the force of the second workpiece and the second workpiece tilted by the second gripping mechanism is in contact with the reference surface. And the second plane based on the positional relationship between the second reference plane and the reference plane. Calculates the displacement amount of the work, may be.

  In this way, even when a set of work pieces that can be fitted without changing their positions is placed on the work table (reference surface) in a state where the height of the gripping position cannot be specified, the fitting work can be performed accurately. Can do.

  The gripping portion includes a first gripping portion having a first gripping mechanism and a second gripping portion having a second gripping mechanism, and is gripped by the first gripping mechanism. 1 workpiece is transferred in a direction perpendicular to the gripping direction of the first gripping mechanism by the second gripping mechanism, and the shift amount calculation unit is controlled by the second gripping mechanism. In a state where the inclined second workpiece and the reference surface are in contact with each other, the deviation amount for the second workpiece is calculated based on the positional relationship between the second gripping mechanism and the reference surface. May be.

  Thus, even when a set of workpieces that can be fitted by changing the workpiece is placed on the work table (reference surface) in a state in which the height of the gripping position cannot be specified, the workpiece transfer operation is performed. Subsequently, the shift amount of the gripping position can be calculated, and the fitting operation can be executed efficiently.

  Moreover, you may make it further provide the correction | amendment part which correct | amends the operation | movement of the said holding part based on the said deviation | shift amount calculated by the said deviation | shift amount calculation part.

  Thus, the fitting operation can be accurately performed according to the deviation amount calculated by the deviation amount calculation unit.

  Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments.

2 is a diagram illustrating an example of an appearance of a robot 10. FIG. It is a figure for demonstrating the detail of the hand part 14 which the robot 10 has. 2 is a block diagram illustrating an example of a schematic configuration of a robot. FIG. (A) It is a figure explaining the method of inclining the workpiece | work 20a and calculating | requiring the shift | offset | difference of a holding position, when holding the workpiece | work 20a which has a recessed part (female) in a lower surface. (B) It is a figure explaining the method of calculating | requiring the shift | offset | difference of a holding position by inclining the workpiece | work 20b, when holding the workpiece | work 20b which has a convex part (male) in a lower surface. It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 1st Embodiment (first half process). It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 1st Embodiment (second half process). (A)-(C) In the workpiece | work fitting control concerning 1st Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (1st-3rd step). (A)-(C) In the workpiece | work fitting control concerning 1st Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (4th-6th steps). (A)-(C) In the workpiece | work fitting control concerning 1st Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (7th-9th steps). It is a figure explaining the method of pushing in the workpiece | work 20 and calculating | requiring the shift | offset | difference of a holding position. It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 2nd Embodiment (first half process). It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 2nd Embodiment (second half process). (A)-(C) In the workpiece fitting control concerning 2nd Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (1st-3rd step). (A)-(C) In the workpiece fitting control concerning 2nd Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (4th-6th stage). (A)-(B) In the workpiece fitting control concerning 2nd Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (7th-8th steps). It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 3rd Embodiment (first half process). It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 3rd Embodiment (second half process). (A)-(C) In the workpiece fitting control concerning 3rd Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (1st-3rd step). (A)-(C) In the workpiece fitting control concerning 3rd Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (4th-6th stage). (A)-(C) It is a figure which shows the state of the hand part 14 and the workpiece | work 20 in the workpiece | work fitting control concerning 3rd Embodiment (7th-9th stage). (A)-(B) In the workpiece fitting control concerning 3rd Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (10th-11th steps). It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 4th Embodiment (first half process). It is a flowchart for demonstrating an example of the workpiece | work fitting control concerning 4th Embodiment (second half process). (A)-(C) In the workpiece fitting control concerning 4th Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (1st-3rd step). (A)-(C) In the workpiece | work fitting control concerning 4th Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (4th-6th steps). (A)-(C) In the workpiece fitting control concerning 4th Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (7th-9th steps). (A)-(C) In the workpiece fitting control concerning 4th Embodiment, it is a figure which shows the state of the hand part 14 and the workpiece | work 20 (10th-11th steps). It is a figure explaining the modification of the said 1st Embodiment and the said 3rd Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<First Embodiment>
First, a first embodiment of the present invention will be described with reference to FIGS. 1st Embodiment is employ | adopted when one set of workpiece | work which can be fitted without replacement is mounted on the work bench | platform (reference surface). That is, one work is placed on the work table with the convex part facing the lower surface, and the other work is placed on the work table with the concave part facing the lower surface.

  FIG. 1 is a diagram illustrating an external appearance example of the robot 10 according to the present embodiment. As shown in the figure, the robot 10 is a double-arm robot having a plurality of robot arms (left arm portion 11a, right arm portion 11b). The robot 10 grips the workpiece W with each hand (left hand portion 14a, right hand portion 14b) attached to the tip of each robot arm (11a, 11b) and performs a fitting operation.

  Here, the left arm portion 11a functions as the left arm of the robot 10, and includes a plurality of links 12a and a plurality of joints (joints) 13a connecting the links 12a.

  Each joint 13a rotatably connects the links 12a or the body of the robot 10 and the link 12a. Therefore, by driving each joint 13a in conjunction with each other, the left hand portion 14a attached to the distal end portion of the left arm portion 11a can be moved freely (however, within a predetermined range) and in any direction. You can also go to.

  The right arm 11b functions as the right arm of the robot 10, and includes a plurality of links 12b and a plurality of joints (joints) 13b that connect the links 12b.

  Similar to the left arm portion 11a, each joint 13b of the right arm portion 11b rotatably connects the links 12b and the body of the robot 10 and the link 12b. Therefore, by driving each joint 13b in conjunction with each other, the right hand portion 14b attached to the tip portion of the right arm portion 11b can be freely moved (however, within a predetermined range), and in any direction. You can also go to.

  In addition, although the example of a biaxial arm is shown in the figure, of course, it is not limited to this. For example, the number of axes (number of joints) may be further increased. Further, the shapes of the links 12 (a, b) and the hand portions 14 (a, b) are not limited to this. The robot arm (11a, 11b) and the hand (14a, 14b) are also collectively referred to as a “gripping device”.

  FIG. 2 is a diagram for explaining details of the hands (the left hand unit 14a and the right hand unit 14b). As shown in the drawing, the hands (the left hand 14a and the right hand 14b) include a first finger 15 (a, b), a second finger 16 (a, b), and a force sensor 17 (a, b). And. The hands (the left hand 14a and the right hand 14b) can grip the workpiece W by moving the first finger 15 (a, b) and the second finger 16 (a, b) in the illustrated gripping direction. it can.

  Moreover, the hand (the left hand part 14a, the right hand part 14b) is provided with the force sensor 17 (a, b). The force sensor 17 (a, b) is a force sensor having at least one axis (for example, six axes), and a force F acting on the workpiece W (for example, a force pushing the workpiece W in the fitting direction) or moment. M is detected.

  By the way, the robot 10 having the robot arms (the left arm unit 11a and the right arm unit 11b) and the hands (the left hand unit 14a and the right hand unit 14b) as described above is controlled by the control unit 100. Although not shown in FIGS. 1 and 2, the control unit 100 may be built in the main body of the robot 10 or may be provided at a position where the robot 10 can be controlled by remote operation.

  FIG. 3 is a block diagram illustrating an example of a schematic configuration of the robot 10. As shown by the bold lines in the figure, the robot 10 includes a control unit 100, the above-described robot arms (left arm unit 11a, right arm unit 11b), and the above-described hands (left hand unit 14a, right hand unit 14b), Is provided.

  Although not described in FIGS. 1 and 2 above, each joint 13a included in the left arm portion 11a is provided with an actuator (motor) for driving each joint 13a. Similarly, each joint 13b included in the right arm portion 11b is also provided with an actuator (motor) for driving each joint 13b.

  The first finger 15a and the second finger 16a included in the left hand portion 14a are also provided with actuators (motors) for driving the fingers (15a, 16a). Similarly, the first finger 15b and the second finger 16b included in the right hand portion 14b are also provided with actuators (motors) for driving the fingers (15b, 16b).

  On the other hand, the control unit 100 includes a central control unit 101, a storage unit 102, an impedance control unit 103, a contact determination unit 104, an encoder reading unit 105, an arm control unit 106, a hand control unit 107, and a positional deviation. An estimation unit 108 and a force sensor control unit 109 are provided.

  Here, the central control unit 101 comprehensively controls each unit (102 to 109) included in the control unit 100.

  The storage unit 102 also stores data necessary for various calculations, data representing the shape of the workpiece W (hereinafter also referred to as “shape data”), a program for controlling the operation of the robot 10 (for example, fitting work), and the like. Remember. For example, the storage unit 102 includes a volatile memory such as a DDR-SDRAM, a non-volatile memory such as a flash memory, and the like.

  When the reaction force is applied from the external environment to the hand (left hand unit 14a, right hand unit 14b) of the robot arm (left arm unit 11a, right arm unit 11b), the impedance control unit 103 The robot arm 11 (a, b) and the hand 14 (a, b) are controlled so as to have impedance (position, velocity, acceleration) suitable for work. For example, in the present embodiment, the impedance control unit 103 contacts the work table (or the reference surface) when the work W gripped by the hand 14 (a, b) is lowered. Then, the speed and acceleration at which the workpiece W is lowered are adjusted so that the reaction force (drag) from the work table against the workpiece W (or hand) disappears. Such control can prevent the workpiece W from being dropped and the hand 14 (a, b) or the work table from being damaged.

  The contact determination unit 104 determines whether at least a part (for example, one end) of the workpiece W gripped by the hand 14 (a, b) has contacted the work table (reference surface). For example, the contact determination unit 104 determines whether one end of the workpiece W has contacted the work table (reference surface) based on the data obtained from the force sensor 17 (a, b).

  The encoder reading unit 105 reads the encoder value of the actuator provided in the joint 13 (a, b) of the robot arm (left arm unit 11a, right arm unit 11b). Although not shown, the encoder reading unit 105 can also read the encoder value of the actuator provided in the hand (left hand unit 14a, right hand unit 14b). These encoder values are used to grasp the position, moving direction, moving amount, speed, acceleration, etc. of the robot arm (left arm portion 11a, right arm portion 11b) and hand (left hand portion 14a, right hand portion 14b). Is done.

  The arm control unit 106 controls the robot arms (the left arm unit 11a and the right arm unit 11b). For example, the arm control unit 106 drives an actuator provided in the joint 13 (a, b) of the robot arm 11 (a, b) so that the hand (the left arm unit 14 a, the right hand unit 14 b) is the workpiece W. Control is performed so that the position, orientation, and speed are suitable for the fitting operation (an instruction is output).

  The hand control unit 107 controls the hands (the left hand unit 14a and the right hand unit 14b). For example, the hand control unit 106 drives an actuator provided on the first finger 15 (a, b) and the second finger 16 (a, b) of the hand 14 (a, b) to move the workpiece W to the first. Controls such as control to pinch (that is, grip) the first finger 15 (a, b) and the second finger 16 (a, b), control to release the workpiece W, and the like (output instructions).

  The position deviation estimation unit 108 estimates a deviation amount dx from the reference position for a position where the workpiece W is gripped (hereinafter referred to as “gripping position”). The reference position is a position where the work W can be reliably gripped by the first finger 15 (a, b) and the second finger 16 (a, b). For example, one side of a quadrangle that forms the lower surface of the work W May be the midpoint position or the center of gravity of the workpiece W.

  If the gripping direction in which the workpiece W is gripped is the Y direction, the shift amount dx is a shift amount in the X direction, which is a direction substantially orthogonal to the gripping direction.

  In the present embodiment (first embodiment), the misregistration estimation unit 108 is based on the positional relationship between the work W in a state where it is tilted at an angle θ and brought into contact with the work table (reference surface) and the work table (reference surface). Then, the shift amount dx of the gripping position is estimated.

  FIG. 4 is a diagram for explaining a method for obtaining a grip position shift (shift amount dx) in the present embodiment (first embodiment).

  4A is a view showing an example of a work W having a recess (female) on the lower surface (that is, the fitting surface) when viewed from the side, and FIG. 4B is a lower surface (fitting). It is a figure which shows the example at the time of seeing the workpiece | work W which has a convex part (male) in a surface from the side. Hereinafter, a workpiece W having a concave portion on the lower surface is referred to as “work (concave) W”, and a workpiece W having a convex portion on the lower surface is referred to as “work (convex) W”.

  As shown in FIG. 4A, a work (concave) W having a recess on the lower surface is inclined at an angle θ1, and one end of the work (concave) W (in the illustrated example, the lower surface of the work (concave) W is configured. In the state where the rectangular “short side” is in contact with the work table (reference surface), the actual gripping position by each finger (first finger 15 and second finger 16) is the work table (reference surface). To the position (height) of the distance Hs. As shown in the drawing, assuming that the actual gripping position is shifted by a shift amount dx from the reference position (in the illustrated example, “the midpoint position of the long side constituting the lower surface of the workpiece W”), The distance from the position (contact position) at which one end of the (concave) W is in contact with the work table (reference surface) to the actual gripping position is “dx + L”. However, L is the distance from the contact position to the reference position. In the present embodiment, the length of the long side of the workpiece W is “2 × L”, and L is half the length of the long side constituting the lower surface of the workpiece W. Dx is a value in the range of “−L ≦ dx ≦ L”.

  In this state, the misregistration estimation unit 108 of the present embodiment (first embodiment) substitutes the values “Hs” and “L” for the relational expression Hs = {dx + (L / 2) × sin (θ1). Then, the shift amount “dx” is calculated. Here, “Hs” is a value obtained based on an encoder value obtained from an actuator provided in the robot arm (the left arm portion 11a and the right arm portion 11b). “L” is a value obtained from the shape data of the workpiece (concave) W stored in the storage unit 102.

  Also, as shown in FIG. 4B, a workpiece (convex) W having a convex portion on the lower surface is inclined at an angle θ2, and one end of the workpiece (convex) W (in the example shown, the workpiece (convex) W In a state where the rectangular “short side” constituting the lower surface is in contact with the work table (reference surface), the actual gripping position by each finger (the first finger 15 and the second finger 16) is the work table ( The position (height) is a distance Hs from the reference plane. As shown in the drawing, assuming that the actual gripping position is shifted by a shift amount dx from the reference position (in the illustrated example, “the midpoint position of the long side constituting the lower surface of the workpiece W”), The distance from the position (contact position) at which one end of the (convex) W is in contact with the work table (reference surface) to the actual gripping position is “dx + (L / 2)”. However, L is the length of the long side constituting the lower surface of the workpiece W. Dx is a value in the range of “− (L / 2) ≦ dx ≦ (L / 2)”.

  In this state, the misregistration estimation unit 108 of the present embodiment (first embodiment) substitutes the values “Hs” and “L” for the relational expression Hs = {dx + (L / 2) × sin (θ2). Then, the shift amount “dx” is calculated. Here, “Hs” is a value obtained from an encoder value obtained from an actuator provided in the robot arm (left arm portion 11a, right arm portion 11b). “L” is a value obtained from the shape data of the workpiece (convex) W stored in the storage unit 102.

  Returning to FIG. 3, the force sensor control unit 109 controls the left force sensor 17 a provided in the left hand part 14 a and the right force sensor 17 b provided in the right hand part 14 b. For example, the force sensor control unit 109 acquires the force F (force F acting on the workpiece W) and the moment M detected by the left force sensor 17a and the right force sensor 17b.

  As indicated by a dotted line in the figure, the control unit 100 includes an imaging unit 110 for determining the direction (up and down) of the workpiece W placed on the work table (reference surface). It may be.

  The main components of the control unit 100 described above are control of input / output to / from the CPU, which is an arithmetic unit, a ROM in which a program is recorded, a RAM that temporarily stores data as a main memory, and a host. It can be achieved by a general computer provided with an interface and a system bus serving as a communication path between each component. An ASIC (Application Specific Integrated Circuit) designed to perform a specific process exclusively may be included, or may be configured by an ASIC.

  The robot 10 to which this embodiment is applied has the above configuration. However, this configuration is not limited to the above configuration because the main configuration has been described in describing the features of the present invention. In addition, other configurations included in a general dual-arm robot are not excluded. The robot 10 may be a robot having more robot arms, hands, and fingers.

  In addition, each of the constituent elements described above is classified according to main processing contents in order to facilitate understanding of the configuration of the control unit 100. The present invention is not limited by the way of classification and names of the constituent elements. The configuration of the control unit 100 can be classified into more components depending on the processing content. Moreover, it can also classify | categorize so that one component may perform more processes. Further, the processing of each component may be executed by one hardware or may be executed by a plurality of hardware.

  Next, a characteristic operation of the robot 10 configured as described above in the first embodiment will be described.

<Work Fitting Control According to First Embodiment>
5 and 6 are examples of control (hereinafter referred to as “work fitting control”) related to the operation of fitting the workpiece (convex) W and the workpiece (concave) W, which is executed by the robot 10 of the first embodiment. It is a flowchart for demonstrating.

  FIGS. 7 to 9 show the left and right hand fingers (first finger 15 (a, b), second finger 16 (a, b)) and workpiece W at each control stage of the flow. It is a figure which shows a position and attitude | position.

  As described above, the workpiece fitting control according to the first embodiment is a pair of a workpiece (concave) W1 and a workpiece (convex) W2 on a work table (reference surface) as shown in FIG. It may be executed when exists.

  Therefore, when the main flow shown in FIG. 5 is started, the central control unit 101 first determines the position (for example, the X and Y coordinates on the reference surface) of the workpiece W placed on the work table (reference surface). The workpiece W is detected, and a workpiece (concave) W having a concave portion on the lower surface and a workpiece (convex) W having a convex portion on the lower surface are distinguished from each other (step S101).

  However, the position of each workpiece W may be prepared in the storage unit 102 in advance, or may be obtained using the imaging unit 110 or various sensors. In addition, as a method of distinguishing between the workpiece (concave) W and the workpiece (convex) W, a method of distinguishing using data prepared in the storage unit 102 in advance, an image captured by the imaging unit 110 is used. There are a method of analyzing and distinguishing images, and a method of measuring and distinguishing the height of the lower surface of the workpiece W and the height of the reference surface.

  Then, the central control unit 101 determines whether or not a pair of a workpiece (concave) W and a workpiece (convex) W exists on the work table (reference surface) based on the processing result of step S101 (step S102). ).

  Here, when there is no pair of workpiece (concave) W and workpiece (convex) W as one set, the central control unit 101 determines that there is no pair of workpiece (concave) W and workpiece (convex) W. (Step S102; No), an error process is executed (Step S118), and the workpiece fitting control according to the first embodiment is stopped.

  Thus, when there is no workpiece (concave) W and workpiece (convex) W pair on the work table (reference surface), the workpiece fitting control according to the first embodiment can be immediately terminated. Note that the error processing in step S118 includes processing for notifying the user that there is no pair of workpiece (concave) W and workpiece (convex) W.

  On the other hand, if there is at least one pair of workpiece (concave) W and workpiece (convex) W, the central control unit 101 determines that there is a pair of workpiece (concave) W and workpiece (convex) W. (Step S102; Yes), the workpiece fitting control according to the first embodiment is continued.

  In this case, the arm control unit 106 and the hand control unit 107 work together to move the work (concave) W1 placed on the work table (reference surface) by the fingers 15a and 16a of the left hand unit 14a. Grasping (step S103). 7A shows the hand (first finger 15a of the left hand, first finger 15b of the right hand), the work W (work (concave) W1, work ( The state of convex) W2) is shown.

  Then, the arm control unit 106 and the hand control unit 107 raise the work (concave) W1 gripped by the left hand unit 14a (step S104). FIG. 7B shows a state in which the workpiece (concave) W1 has risen when step S104 is completed.

  By the way, while the workpiece (concave) W1 is rising, the encoder reading unit 105 reads the encoder value from the actuator provided in the left arm unit 11a (each joint 13a) and stores it in the storage unit 102. Thus, the distance Z1 that the workpiece (concave) W1 has been raised in step S104 can be specified later.

  Next, the arm control unit 106 and the hand control unit 107 incline the work (concave) W1 held by the left hand unit 14a by an angle θ1 (step S105). FIG. 7C shows a state in which the workpiece (concave) W1 is tilted at an angle θ1 with respect to the reference plane when step S105 is completed.

  Even while the workpiece (concave) W1 is tilted, the encoder reading unit 105 reads the encoder value from the actuator provided in the left arm unit 11a (each joint 13a) and stores it in the storage unit 102. Thus, the angle θ1 at which the workpiece (concave) W1 is tilted in step S105 can be specified later.

  Then, the arm control unit 106 and the hand control unit 107 start control to lower the work (concave) W1 inclined by the angle θ1. At this time, assuming that one end of the work (concave) W1 is in contact with the workbench (reference surface), the impedance control unit 103 applies a reaction force from the workbench to the work (concave) W1 at the time of contact. Impedance control (for example, control for repeatedly lowering and raising the left arm portion 11a) is performed.

  Here, whether or not one end of the workpiece (concave) W1 has contacted the work table (reference surface) is determined by the contact determination unit 104 (step S106).

  Then, until one end of the work (concave) W1 comes into contact with the work table (reference surface) (step S106; No), the control for lowering the work (concave) W1 is continued under impedance control.

  Of course, even while the workpiece (concave) W1 is descending, the encoder reading unit 105 reads the encoder value from the actuator provided in the left arm unit 11a (each joint 13a) and stores it in the storage unit 102. Thus, the distance Z2 where the workpiece (concave) W1 is lowered can be specified later.

  By the way, when one end of the work (concave) W1 comes into contact with the work table (reference surface) (step S106; Yes), the arm control unit 106 and the hand control unit 107 stop the lowering control of the work (concave) W1. FIG. 8A shows a state of the workpiece (concave) W1 at this stage (a state in which the angle θ1 is inclined with respect to the reference surface and the rectangular short side constituting the lower surface is in contact with the work table). ing.

  In the state shown in FIG. 8A, the positional deviation estimation unit 108 determines the height of the actual gripping position by each finger (the first finger 15 and the second finger 16a) (that is, the distance from the reference plane) Hs1. Is calculated (step S107). Specifically, the positional deviation estimation unit 108 calculates Hs1 by subtracting the distance Z2 lowered by impedance control from the distance Z1 obtained by raising the workpiece (concave) W1 in step S104 (Hs1 = Z1). -Z2).

  Next, the misregistration estimation unit 108 determines the reference position (here, the long side L of the rectangle that forms the lower surface of the workpiece W) with respect to the actual gripping position by each finger (the first finger 15 and the second finger 16a). A deviation amount dx1 from the middle point is calculated (step S108). Specifically, the position deviation estimation unit 108 first reads the length L of the long side of the rectangle that forms the lower surface of the workpiece (concave) W1 from the storage unit 102 (shape data). The positional deviation estimation unit 108 then sets the relational expression Hs1 = {dx1 + (L / 2) × sin (θ1) to “Hs1” calculated in step S107 and the value “L” read from the storage unit 102. Is substituted to calculate the shift amount “dx1”.

  Note that the calculation in step S108 is based on the principle described with reference to FIG. In this way, the positional deviation estimation unit 108 of the present embodiment can easily obtain the accurate deviation amount dx1 for the gripping position.

  After the displacement amount dx1 is calculated, the arm control unit 106 and the hand control unit 107 return the workpiece (concave) W1 gripped by the left arm 11a to a horizontal posture with respect to the work table (reference surface), Increase (step S109).

  When the control in steps S103 to S109 is performed and the shift amount dx1 of the grip position of the workpiece (concave) W1 by the left hand unit 14a is calculated, the control unit 100 then determines the workpiece (concave) W1. The work (convex) W2 to be fitted is gripped by the right hand portion 14b, and the grip position deviation amount dx2 for the work (convex) W2 is calculated according to the same control as in the case of the work (concave) W1 (FIG. Steps S110 to S116 shown in FIG. In FIGS. 8B, 8C, 9A, and 9B, the right hand portion 14b of the right arm 11b and the workpiece (convex) W2 are controlled according to the control in steps S110 to S116. A state transition is shown.

  When the process of step S116 is completed, the arm control unit 106 and the hand control unit 107 determine that the workpiece (concave) W1 gripped by the left hand unit 14a and the workpiece (convex) W2 gripped by the right hand unit 14b. Are fitted (step S117). However, in step S117, the arm control unit 106 and the hand control unit 107 perform the grip position displacement amount dx1 for the workpiece (concave) W1 calculated in step S108 and the workpiece (convex) W2 calculated in step S115. The positions of the left hand portion 14a and the right hand portion 14b are corrected so that the concave portion of the workpiece (concave) W1 and the convex portion of the workpiece (convex) W2 are aligned with each other on the basis of the shift amount dx2 of the gripping position. Then, both works are fitted (FIG. 9C).

  Thus, even when the gripping position of the workpiece W by the left hand portion 14a and the right hand portion 14b is shifted, the fitting operation can be performed accurately.

  The above control is the workpiece fitting control according to the first embodiment of the present invention. As described above, the workpiece fitting control according to the first embodiment is suitable when the fitting operation can be performed while the workpiece W placed on the work table is held first without turning over. .

  In addition, according to the workpiece fitting control according to the first embodiment, it is possible to recognize the shift of the gripping position only by bringing the tilted workpiece W into contact with the reference surface. For this reason, it can be said that the shift of the gripping position can be recognized with a simpler configuration than in the past. Further, according to the first embodiment, since the positional relationship between the gripped workpiece W and the reference plane can be accurately specified (measured), it is possible to recognize the shift of the gripping position with higher accuracy than before.

Second Embodiment
Next, a second embodiment of the present invention will be described with reference to FIGS. Since the robot 10 according to the second embodiment has basically the same configuration as the robot 10 according to the first embodiment, the following description will focus on differences.

  In the second embodiment, a set of workpieces that can be fitted by moving the workpiece W (that is, passing the workpiece W gripped by one robot arm to the other robot arm) is placed on the workbench (reference Adopted when placed on the surface). That is, one work is placed on the work table with the convex part facing the lower surface, and the other work is placed on the work table with the convex part facing the lower surface. On the contrary, one work may be placed on the work table with the concave portion facing the lower surface, and the other work may be placed on the work table with the concave portion facing the lower surface. .

  In the second embodiment, a method different from that of the first embodiment is adopted as a method of estimating (calculating) the grip position shift amount dx.

  FIG. 10 is a diagram for explaining a method of obtaining a grip position shift (shift amount dx) in the second embodiment.

  As shown in the drawing, the workpiece W having a convex portion on the lower surface is gripped by the fingers (15a, 16a) provided in the left hand portion 14a and transferred to the fingers (15b, 16b) provided in the right hand portion 14b. In such a delivery operation, the workpiece W is held by both the left hand portion 14a and the right hand portion 14b.

  In such a state, for example, as illustrated, a force F (force in the direction of the right hand portion 14b) to be pushed from the left hand portion 14a is applied to the workpiece W. At this time, the actual gripping position by each finger (the first finger 15a and the second finger 16a) of the left hand portion 14a is the reference position (in the example shown, “in the long side constituting the lower surface of the workpiece W Assuming that there is a deviation dx from the “point position”), the workpiece W is given a rotational motion around the reference position by the moment M of the force F pushed into the reference position.

  Therefore, the displacement estimation unit 108 of the second embodiment calculates the displacement amount “dx” by substituting the values of “M”, “α”, and “F” into the relational expression of M = α × F × dx. To do. Here, “M” and “F” are values detected by the force sensor (the left force sensor 17a and the right force sensor 17b). “Α” is a constant for adjusting the relationship between the moment M of the force F pushed into the workpiece W and the displacement dx, and is a value obtained from data stored in the storage unit 102 in advance.

  According to the method as described above, since the shift amount dx of the gripping position can be calculated following the transfer operation of the workpiece W, the fitting operation can be performed efficiently.

  Hereinafter, a characteristic operation of the robot 10 in the second embodiment will be described.

<Work Fitting Control According to Second Embodiment>
FIGS. 11 and 12 are flowcharts for explaining an example of workpiece fitting control according to the operation of fitting two workpieces (convex) W, which is executed by the robot 10 of the second embodiment.

  FIGS. 13 to 15 show the left and right hand fingers (first finger 15 (a, b), second finger 16 (a, b)) and workpiece W for each control stage of this flow. It is a figure which shows a position and attitude | position.

  As mentioned above, the workpiece fitting control according to the second embodiment is a pair of workpiece (convex) W3 and workpiece (convex) W4 on the work table (reference surface) as shown in FIG. May be executed when both are present with the convex part facing down. Of course, on the contrary, it may be executed when a pair of a workpiece (concave) W and a workpiece (concave) W is present on the work table (reference surface) with the concave portion facing downward.

  When the main flow shown in FIG. 11 is started, the central control unit 101 first detects the position (for example, the X and Y coordinates on the reference surface) of the workpiece W placed on the work table (reference surface). Among the workpieces W, a workpiece (concave) W having a concave portion on the lower surface and a workpiece (convex) W having a convex portion on the lower surface are distinguished (step S201).

  Here, when there is at least one pair of workpiece (concave) W and workpiece (convex) W, the central control unit 101 determines that there is a pair of workpiece (concave) W and workpiece (convex) W. Since it determines (step S202; Yes) and the workpiece | work fitting control concerning 1st Embodiment is more suitable, control of step S103-S117 is performed.

  On the other hand, when there is no pair of workpiece (concave) W and workpiece (convex) W as one set, the central control unit 101 determines that there is no pair of workpiece (concave) W and workpiece (convex) W. (Step S202; No), the process proceeds to Step S203.

  Next, the central control unit 101 turns a pair of workpieces W that can be fitted by turning over one workpiece W (for example, passing the workpiece W gripped by one robot arm to the other robot arm). It is determined whether or not it can be created (step S203).

  Here, when two or more workpieces (convex) W are not placed on the work table (reference surface) and two or more workpieces (concave) W are not placed, The central control unit 101 determines that a pair of workpieces W that can be fitted even if one workpiece W is turned over cannot be created (step S203; Yes), executes error processing (step S220), and performs the second process. The workpiece fitting control according to the embodiment is stopped.

  On the other hand, when there are two or more workpieces (convex) W or two or more workpieces (convex) W on the work table (reference surface), the central control unit 101 has one workpiece W. It is determined that a pair of workpieces W that can be fitted can be created by turning over (Step S203; Yes), and the process proceeds to Step S204.

  When the process proceeds to step S204, the arm control unit 106 and the hand control unit 107 work in conjunction with the workpiece (convex) W3 placed on the work table (reference surface) to each finger 15a of the left hand unit 14a. , 16a (step S204). In FIG. 13A, the hand (first finger 15a, second finger 16a, first finger 15b, second finger 16b of the right hand) at the stage where step S204 is completed and The state of the work W (work (convex) W3, work (convex) W4) is shown.

  Then, the arm control unit 106 and the hand control unit 107 raise the work (concave) W3 gripped by the left hand unit 14a (step S205). FIG. 13B shows a state in which the workpiece (convex) W3 has risen when step S205 is completed.

  Next, the arm control unit 106 and the hand control unit 107 cause the workpiece (convex) W3 gripped by the left hand unit 14a to be gripped also by the right hand unit 14b (step S206). Thus, as shown in FIG. 13C, at the stage where step S206 is completed, the workpiece (convex) W3 includes the left hand portion 14a (first finger 15a and second finger 16a) and the right hand portion 14b. It is in a state of being gripped by both hands (first finger 15b and second finger 16b).

  Then, the arm control unit 106 and the hand control unit 107 perform control to apply a force F in the direction of the left hand unit 14a to the work (convex) W3 by the right hand unit 14b (first finger 15b, second finger 16b). Start (step S207). Here, the arm control unit 106 and the hand control unit 107 fix the work (convex) W3 by the left hand unit 14a so that the work (convex) W3 does not move (for example, the left hand unit 14a is also a work ( Convex) Apply force in the opposite direction to W3).

  While the force F is applied to the workpiece (convex) W3, the left force sensor 17a provided in the left hand portion 14a determines the force F applied to the grip position of the workpiece (convex) W3 and the reference position. The moment M when the center of rotation is used is detected.

  At this time, the force sensor control unit 109 acquires the force F detected by the left force sensor 17a and confirms the operation of the left force sensor 17a (step S208).

  Specifically, if the force F acquired from the left force sensor 17a is less than a certain value, the force sensor control unit 109 determines that reconfirmation of the operation of the left force sensor 17a is necessary (step S208; No), after waiting for a predetermined time, the process of step S208 is executed again.

  On the other hand, when the force F acquired from the left force sensor 17a is greater than or equal to a certain value, the force sensor control unit 109 determines that the left force sensor 17a is operating normally (step S208; Yes). ), The force F acquired here is stored in the storage unit 102.

  Next, the force sensor control unit 109 acquires the moment M detected by the left force sensor 17a and stores it in the storage unit 102 (step S209).

  Then, the misregistration estimation unit 108 determines the reference position (here, the lower surface of the workpiece (convex) W3) with respect to the actual gripping position by each finger (the first finger 15b and the second finger 16b) of the right hand unit 14b. The amount of deviation dx2 from the long side L of the rectangle to be constructed is calculated (step S210). Specifically, the positional deviation estimation unit 108 first reads from the storage unit 102 the force F, moment M, and constant α applied to the workpiece (convex) W3. Then, the displacement estimation unit 108 calculates the displacement amount “dx2” by substituting the values of “M”, “α”, and “F” into the relational expression M = α × F × dx.

  Note that the calculation in step S210 is based on the principle described in FIG.

  After the shift amount dx2 is calculated, the arm control unit 106 and the hand control unit 107 release the gripping by the left hand unit 14a on the workpiece (convex) W3 gripped by the left hand unit 14a and the right hand unit 14b. (Step S211). FIG. 14A shows a state in which the left hand unit 14a has released the workpiece (convex) W3 in step S211.

  Thus, the workpiece (convex) W3 is transferred from the left hand portion 14a to the right hand portion 14b. As described above, in the robot 10 of the second embodiment, an accurate deviation amount dx2 with respect to the gripping position can be obtained following the workpiece (convex) W3 transfer operation.

  When the control in steps S204 to S211 is performed and the shift amount dx2 of the gripping position of the workpiece (concave) W3 by the right hand unit 14b is calculated, the control unit 100 then determines the workpiece (convex) W3. The workpiece (convex) W4 to be mated is gripped by the left hand portion 14a, and the gripping position deviation amount dx1 for the workpiece (convex) W4 is calculated according to the same control (S103 to S109) as in the first embodiment. (Steps S212 to S218 shown in FIG. 12). FIGS. 14B, 14C, and 15A show how the states of the left hand portion 14a and the workpiece (convex) W4 transition according to the control in steps S212 to S218. ing.

  When the control in step S218 ends, the arm control unit 106 and the hand control unit 107 determine that the workpiece (convex) W4 gripped by the left hand unit 14a and the workpiece (convex) W3 gripped by the right hand unit 14b. Are fitted (step S219). However, in step S219, the arm control unit 106 and the hand control unit 107 perform the grip position deviation amount dx1 for the workpiece (convex) W4 calculated in step S217 and the workpiece (convex) W3 calculated in step S210. The positions of the left hand portion 14a and the right hand portion 14b are corrected so that the convex portion of the workpiece (convex) W4 and the concave portion of the workpiece (convex) W3 are aligned on the basis of the gripping position shift amount dx2. Then, both works are fitted (FIG. 15B).

  Thus, the fitting operation can be performed accurately also in the second embodiment.

  The above control is the workpiece fitting control according to the second embodiment of the present invention. As described above, the workpiece fitting control according to the second embodiment is suitable when a set of workpieces that can be fitted by changing the workpiece W is placed on the work table (reference surface). Yes.

  Further, according to the workpiece fitting control according to the second embodiment, it is possible to recognize the shift of the gripping position by simply applying the force F to the workpiece W gripped by both hands. For this reason, it can be said that the shift of the gripping position can be recognized with a simpler configuration than in the past. Further, since the moment M of the force F applied to the workpiece W can be accurately identified (detected) using the force sensor 17 (a, b), the displacement of the gripping position can be detected with higher accuracy than before. Can be recognized.

<Third Embodiment>
Next, a third embodiment of the present invention will be described with reference to FIGS. Since the robot 10 according to the third embodiment has basically the same configuration as the robot 10 according to the first embodiment or the second embodiment, the difference from the first embodiment or the second embodiment is mainly described. This will be described below.

  The third embodiment is employed when a set of workpieces that can be fitted without being held is placed on a work table (reference surface) in a state where the height of the gripping position cannot be specified. For example, one gripping target workpiece (concave) W is placed on another object (for example, another workpiece W) placed on the work table, and the other gripping target workpiece (convex) W is placed. This is also the case where the object is placed on another object (for example, another work W) placed on the work table.

  In the third embodiment, as in the workpiece fitting control according to the second embodiment, the grip position displacement amount dx1 is estimated according to the principle described with reference to FIG. (Calculation) For the other workpiece (convex) W to be gripped, the grip position displacement amount dx2 is estimated (calculated) according to the principle described with reference to FIG.

  Therefore, the work fitting control according to the third embodiment has a flow substantially similar to the work fitting control according to the second embodiment.

<Work Fitting Control According to Third Embodiment>
FIGS. 16 and 17 are flowcharts for explaining an example of workpiece fitting control according to the operation of fitting the workpiece (concave) W and the workpiece (convex) W, which is executed by the robot 10 of the third embodiment. is there.

  18 to 21 show the left and right hand fingers (first finger 15 (a, b), second finger 16 (a, b)) and workpiece W at each control stage of this flow. It is a figure which shows a position and attitude | position.

  As described above, the workpiece fitting control according to the third embodiment is stacked on the workpiece W7 and the workpiece W8 placed on the work table (reference surface) as shown in FIG. It may be executed when there is a pair such as a work (concave) W5 and a work (convex) W6.

  When the flow shown in FIG. 16 is started, the central control unit 101 executes the same control as steps S101 to S102 and S118 of the first embodiment (steps S301 to S302 and S320).

  In the work fitting control according to the third embodiment, the same control as steps S204 to S210 of the second embodiment is executed (steps S303 to S309), and each finger (first finger) of the left hand unit 14a is executed. For the actual gripping position by 15a and the second finger 16a), a deviation dx1 from the reference position (here, the midpoint of the long side L of the rectangle constituting the lower surface of the workpiece (concave) W5) is calculated ( FIG. 18 (A) to (C)).

  Next, the arm control unit 106 and the hand control unit 107 release the gripping by the right hand 14b for the work (concave) W5 gripped by the left hand 14a and the right hand 14b (step S310). FIG. 19A shows a state in which the right hand 14b has released the work (concave) W5 in step S320.

  Thus, the workpiece (concave) W5 is not transferred from the left hand 14a to the right hand 14b, but is held by the left hand 14a. This point is different from the second embodiment. That is, in the second embodiment, the gripping position shift amount dx is obtained after the workpiece (convex) W3 transfer operation, whereas in the third embodiment, the gripping position shift amount dx is calculated. As a main purpose, the workpiece (concave) W5 is held by the left hand 14a and the right hand 14b.

  When the above step S310 is completed, the arm control unit 106 and the hand control unit 107 grip the workpiece (convex) W6, which is a mating partner of the workpiece (concave) W5, with the right hand unit 14b (step S311). The work (convex) W6 is placed on the work table (reference surface) and released (step S312).

  The purpose of this is to place a workpiece (convex) W6 whose height of the gripping position (distance from the reference plane) is unknown on the reference plane, clarify the height of the gripping position, and grip it again (see FIG. 19 (B), (C)).

  If the height of the gripping position becomes clear, the same control as steps S111 to S116 of the first embodiment can be executed (steps S313 to S318), and each finger (first finger 15b and second finger) of the right hand unit 14b can be executed. The deviation dx2 from the reference position (here, the midpoint of the long side L of the rectangle constituting the lower surface of the workpiece (convex) W6) is calculated (see FIG. 20A). ) To (C), FIG. 21 (A)).

  When the control in step S318 is finished, the arm control unit 106 and the hand control unit 107 determine that the workpiece (concave) W5 gripped by the left hand portion 14a and the workpiece (convex) W6 gripped by the right hand portion 14b. Are fitted (step S319). However, in step S319, the arm control unit 106 and the hand control unit 107 perform the grip position displacement amount dx1 for the workpiece (concave) W5 calculated in step S309 and the workpiece (convex) W6 calculated in step S317. The positions of the left hand portion 14a and the right hand portion 14b are corrected so that the concave portion of the workpiece (concave) W5 and the convex portion of the workpiece (convex) W6 are aligned with each other on the basis of the grip amount deviation dx2 of Then, both works are fitted (FIG. 21B).

  Thus, the fitting operation can be accurately performed also in the third embodiment.

  The above control is the workpiece fitting control according to the third embodiment of the present invention. As described above, the workpiece fitting control according to the third embodiment is configured such that a set of workpieces that can be fitted without being moved is placed on the work table (reference surface) in a state where the height of the gripping position cannot be specified. Suitable if you are.

<Fourth embodiment>
Next, a fourth embodiment of the present invention will be described with reference to FIGS. Since the robot 10 according to the fourth embodiment has basically the same configuration as that of the robot 10 according to the first to third embodiments, the following description will be focused on differences from the first to third embodiments.

  In the fourth embodiment, a set of workpieces that can be fitted by changing the workpiece W (that is, transferring the workpiece W gripped by one robot arm to the other robot arm) has a height of the gripping position. It is adopted when it is placed on the work table (reference plane) in a state where it cannot be specified. For example, one workpiece (convex) W to be gripped is placed on another object (for example, another workpiece W) placed on the work table, and the other workpiece (convex) W to be gripped is placed. This is also the case where the object is placed on another object (for example, another work W) placed on the work table. Of course, one workpiece (concave) W to be grasped is placed on another object placed on the work table, and the other work (concave) W to be grasped is also placed on the work table. It may be a case where it is placed on another object.

  In the fourth embodiment, similarly to the workpiece fitting control according to the second embodiment, the gripping position shift amount dx2 is estimated according to the principle described with reference to FIG. (Calculation) For the other workpiece (convex) W to be gripped, the grip position displacement amount dx1 is estimated (calculated) according to the principle described in FIG.

  For this reason, the work fitting control according to the fourth embodiment has substantially the same flow as the work fitting control according to the second embodiment.

<Work Fitting Control According to Fourth Embodiment>
FIG. 22 and FIG. 23 are flowcharts for explaining an example of workpiece fitting control according to the operation of fitting two workpieces (convex) W, which is executed by the robot 10 of the fourth embodiment.

  24 to 27 show the left and right hand fingers (first finger 15 (a, b), second finger 16 (a, b)) and workpiece W at each control stage of this flow. It is a figure which shows a position and attitude | position.

  As described above, the workpiece fitting control according to the fourth embodiment is stacked on the workpiece W11 and the workpiece W12 placed on the work table (reference surface) as shown in FIG. It may be executed when a pair such as a workpiece (convex) W9 and a workpiece (convex) W10 exists with the convex portion facing downward. Of course, on the contrary, it may be executed when a pair of a workpiece (concave) W and a workpiece (concave) W is present on the work table (reference surface) with the concave portion facing downward.

  When the main flow shown in FIG. 22 is started, the control unit 100 executes the same control as steps S201 to S203 and S220 of the second embodiment (steps S401 to S403 and S421).

  Further, in the work fitting control according to the fourth embodiment, the same control as steps S204 to S210 of the second embodiment is executed (steps S404 to S410), and each finger (first finger) of the left hand unit 14a is executed. With respect to the actual gripping position by 15a and the second finger 16a), a deviation amount dx2 from the reference position (here, the midpoint of the long side L of the rectangle constituting the lower surface of the workpiece (convex) W9) is calculated ( FIG. 24 (A) to (C)).

  Next, the arm control unit 106 and the hand control unit 107 release the gripping by the left hand 14a for the work (convex) W9 gripped by the left hand 14a and the right hand 14b (step S411). FIG. 25A shows a state in which the left hand 14a has released the workpiece (convex) W9 in step S411.

  Thus, the workpiece (convex) W9 is transferred from the left hand 14a to the right hand 14b. This point is the same as in the second embodiment, and the grip position displacement amount dx2 can be calculated following the delivery operation.

  When step S411 is completed, the arm control unit 106 and the hand control unit 107 grip the workpiece (convex) W10, which is a mating partner of the workpiece (convex) W9, with the left hand unit 14a (step S412). The work (convex) W10 is placed on the work table (reference surface) and released (step S413).

  The purpose of this is to place a workpiece (convex) W10 whose height of the gripping position (distance from the reference plane) is unknown on the reference plane, reveal the height of the gripping position, and grip again (see FIG. 25 (B), (C)).

  If the height of the gripping position becomes clear, the same control as steps S111 to S116 of the first embodiment can be executed (steps S414 to S419), and each finger (first finger 15a and second finger of the left hand unit 14a) can be executed. The displacement dx1 from the reference position (here, the midpoint of the long side L of the rectangle constituting the lower surface of the workpiece (convex) W10) is calculated (see FIG. 26A). ) To (C), FIG. 27 (A)).

  When the control of step S419 is finished, the arm control unit 106 and the hand control unit 107 determine that the workpiece (convex) W10 gripped by the left hand unit 14a and the workpiece (convex) W9 gripped by the right hand unit 14b. Are fitted (step S420). However, in step S420, the arm control unit 106 and the hand control unit 107 perform the grip position deviation amount dx2 for the workpiece (convex) W10 calculated in step S418 and the workpiece (convex) W9 calculated in step S410. The positions of the left hand portion 14a and the right hand portion 14b are corrected so that the convex portion of the workpiece (convex) W10 and the concave portion of the workpiece (convex) W9 are aligned with each other based on the grip position deviation amount dx2. Then, both works are fitted (FIG. 27B).

  Thus, the fitting operation can be accurately performed also in the fourth embodiment.

  The above control is the workpiece fitting control according to the fourth embodiment of the present invention. As described above, the workpiece fitting control according to the fourth embodiment is performed on the work table (reference standard) in a state in which a set of workpieces that can be fitted by changing the workpiece W cannot be specified with respect to the height of the gripping position. Suitable when placed on the surface).

  Each processing unit of each flow described above is divided according to the main processing contents in order to make the robot 10 easy to understand. The invention of the present application is not limited by the method of classification of the processing steps and the names thereof. The processing performed by the robot 10 can be divided into more processing steps. One processing step may execute more processes.

  Moreover, said embodiment intends to illustrate the summary of this invention, and does not limit this invention. Many alternatives, modifications, and variations will be apparent to those skilled in the art.

  For example, although not described in each of the above embodiments, depending on the case of gripping a workpiece (concave) W having a concave portion on the lower surface and the case of gripping a workpiece (convex) W having a convex portion on the lower surface, Change the hand grip position. For example, when gripping a workpiece (concave) W, the grip position by the hand is set to the same height (position) as the reference surface, and when gripping the workpiece (convex) W, the grip position by the hand is set to the reference plane. It is made higher by a predetermined length (the length of the convex portion).

  Further, the right hand part 14a and the left hand part 14b in each of the above embodiments may be interchanged.

  In the first embodiment and the third embodiment, the grip position shift amount dx may be estimated (calculated) instead of using another method.

  For example, FIG. 28 is a diagram illustrating a modification of the first embodiment and the third embodiment.

  As shown in the figure, in the modification, an operation is performed in which the workpiece W held by each finger (15a, 16a) provided in the left hand portion 14a is held by each finger (15b, 16b) provided in the right hand portion 14b. Is done.

  At this time, each finger (15b, 16b) provided in the right hand unit 14b has a position (gripping position) held by each finger (15a, 16a) of the left hand unit 14a as a center line (dotted line in the drawing). An attempt is made to grip the workpiece W.

  Temporarily, the actual gripping position by each finger (the first finger 15a and the second finger 16a) of the left hand portion 14a is a reference position (in the illustrated example, “the midpoint of the long side constituting the upper surface of the workpiece W”). If it corresponds to “position”), each finger (the first finger 15b and the second finger 16b) reaches the workpiece W almost simultaneously in the gripping operation by the right hand portion 14b.

  However, as shown in the drawing, when the actual gripping position by each finger (the first finger 15a and the second finger 16a) of the left hand portion 14a is deviated from the reference position (deviation amount dx), the right One of the fingers (first finger 15b or second finger 16b) of the hand portion 14b reaches the workpiece W first.

  At this time, a reaction force Fx opposite to the gripping direction is applied to the finger (the first finger 15b or the second finger 16b) that has reached the workpiece W.

  Therefore, the positional deviation estimation unit 108 of the modification example calculates the deviation amount “dx” by substituting the values of “β” and “Fx” into the relational expression dx = β × Fx. Here, “Fx” is a value detected by a force sensor (at least one of the left force sensor 17a and the right force sensor 17b). “Β” is a constant for adjusting the relationship between the reaction force Fx and the shift amount dx, and is a value obtained from data stored in the storage unit 102 in advance.

  Therefore, in the illustrated modification, the misalignment estimation unit 108 is configured so that the right hand unit 14b holds at least one of the left hand unit 14a and the right hand unit 14b in an operation in which the right hand unit 14b grips in a direction orthogonal to the gripping direction by the left hand unit 14a. Based on the force Fx in the direction orthogonal to the gripping direction by the left hand portion 14a generated on one side, the displacement amount dx for the workpiece W is calculated.

  According to the modified example as described above, the grip position shift amount dx can be calculated by a simpler method than in the first embodiment and the third embodiment.

DESCRIPTION OF SYMBOLS 10 ... Robot, 11a ... Left arm part, 11b ... Right arm part, 12a ... Link (left), 12b ... Link (right), 13a ... Joint (left), 13b ... joint (right), 14a ... left hand part, 14b ... right hand part, 15a ... first finger (left), 15b ... first finger (right), 16a .. second finger (left), 16b ... second finger (right), 17a ... force sensor (left), 17b ... force sensor (right), 100 ... control unit 101 ... Central control unit, 102 ... Storage unit, 103 ... Impedance control unit, 104 ... Contact determination unit, 105 ... Encoder reading unit, 106 ... Arm control unit, 107 ..Hand control unit, 108 ... Position deviation estimation unit, 109 ... Force sensor Control unit, 110 ... imaging unit, W ··· work.

Claims (9)

A gripping robot device,
A gripping part for gripping a workpiece placed on a reference surface;
A displacement amount calculation unit that calculates a displacement amount from a reference position with respect to a gripping position of the workpiece by the gripping unit,
The gripping part is
Tilt the work to bring one end of the work into contact with the reference surface,
The deviation amount calculation unit
In a state where the workpiece tilted by the gripping portion and the reference surface are in contact with each other, the deviation amount is calculated based on a positional relationship between the gripping portion and the reference surface.
A gripping robot device characterized by that. The gripping robot device according to claim 1,
The gripping part is
Gripping the workpiece with a direction parallel to the bottom surface of the workpiece (hereinafter referred to as “Y direction”) as a gripping direction;
The deviation amount calculation unit
With respect to the gripping position of the workpiece by the gripping portion, the amount of deviation dx in the X direction orthogonal to the Y direction is
dx = H / sin θ-L
(Where H is the distance of the gripping position from the reference surface, L is the length of the workpiece from the reference position to the one end, and θ is the angle at which the workpiece is tilted)
Using the formula of
A gripping robot device characterized by that. The gripping robot device according to claim 1 or 2,
The gripping part is
A first gripping part having a first gripping mechanism and a second gripping part having a second gripping mechanism;
An operation in which the second gripping mechanism grips the first workpiece gripped by the first gripping mechanism in a direction perpendicular to the gripping direction by the first gripping mechanism;
The deviation amount calculation unit
In the operation in which the second gripping mechanism grips in a direction perpendicular to the gripping direction by the first gripping mechanism, the first generated in at least one of the first gripping part and the second gripping part. Calculating the amount of deviation based on the force in the direction orthogonal to the gripping direction by the gripping mechanism of
A gripping robot device characterized by that. The gripping robot device according to claim 3,
The deviation amount calculation unit
dx = Fx × β
(Where Fx is the force, dx is the displacement, and β is a coefficient)
The amount of deviation is calculated using the following formula:
A gripping robot device characterized by that. The gripping robot device according to any one of claims 1 to 4,
The gripping part is
A first gripping part having a first gripping mechanism and a second gripping part having a second gripping mechanism;
An operation in which the second gripping mechanism grips the first workpiece gripped by the first gripping mechanism in a direction perpendicular to the gripping direction by the first gripping mechanism;
The deviation amount calculation unit
In the operation in which the second gripping mechanism grips in a direction perpendicular to the gripping direction by the first gripping mechanism, the first generated in at least one of the first gripping part and the second gripping part. Based on the force in the direction perpendicular to the gripping direction by the gripping mechanism of the first workpiece, calculating the deviation amount for the first workpiece,
In a state where the second workpiece tilted by the second gripping mechanism and the reference surface are in contact with each other, the displacement of the second workpiece is determined based on the positional relationship between the second gripping mechanism and the reference surface. Calculate the quantity,
A gripping robot device characterized by that. A gripping robot device according to any one of claims 1 to 5,
The gripping part is
A first gripping part having a first gripping mechanism and a second gripping part having a second gripping mechanism;
The first work gripped by the first gripping mechanism is transferred by the second gripping mechanism being gripped in a direction perpendicular to the gripping direction by the first gripping mechanism,
The deviation amount calculation unit
In a state where the second workpiece tilted by the second gripping mechanism and the reference surface are in contact with each other, the displacement of the second workpiece is determined based on the positional relationship between the second gripping mechanism and the reference surface. Calculate the quantity,
A gripping robot device characterized by that. The gripping robot device according to any one of claims 1 to 6,
A correction unit that corrects the operation of the gripping unit based on the shift amount calculated by the shift amount calculation unit;
A gripping robot device characterized by that. A gripping motion control method performed by a gripping robot device having a gripping mechanism,
A gripping step of gripping the workpiece placed on a reference surface by the gripping mechanism;
A shift amount calculating step of calculating a shift amount from a reference position with respect to a gripping position of the workpiece in the gripping step;
In the gripping step,
Tilt the work to bring one end of the work into contact with the reference surface,
In the deviation amount calculating step,
In a state where the workpiece tilted in the gripping step and the reference surface are in contact with each other, the deviation amount is calculated based on a positional relationship between the gripping mechanism and the reference surface.
A gripping motion control method characterized by the above. A gripping motion control device for controlling a gripping mechanism,
An instruction unit that outputs an instruction about a gripping operation of the workpiece to the gripping mechanism;
A displacement amount calculation unit for calculating a displacement amount from a reference position with respect to a workpiece gripping position by the gripping mechanism,
The instruction unit includes:
Grip the work placed on the reference surface, tilt the work, and output an instruction to contact one end of the work with the reference surface to the gripping mechanism,
The deviation amount calculation unit
In a state where the workpiece tilted by the gripping mechanism and the reference surface are in contact with each other, the deviation amount is calculated based on a positional relationship between the gripping mechanism and the reference surface.
A gripping motion control device characterized by the above.

Images