JP2018202599A - Control device for robot system and robot system - Google Patents

Abstract

To provide a technique negating the need of using a dedicated parts feeder, and to provide a technique capable of improving efficiency of work for picking up parts from the parts feeder.SOLUTION: A robot system comprises: a parts feeder having a parts storage part storing parts and a plurality of vibration actuators; and a robot having an end effector picking up the parts from the parts storage part. A control device for the robot system comprises a robot control part and a parts feeder control part. The parts feeder control part selects one or more control commands out of a plurality of control commands individually including control parameters of the plurality of vibration actuators, and causes the parts feeder to perform operation corresponding to the control commands by transmitting the control commands selected to the parts feeder.SELECTED DRAWING: Figure 1

Description

  The present invention relates to control of a robot system that handles parts.

  Patent Document 1 discloses a technique in which a robot picks up a part (part) from a parts feeder and performs an assembly operation. In this prior art, parts in a swivel-type parts feeder are imaged by a camera, and the presence / absence, position and orientation of the parts are recognized by image processing, and the robot performs part grasping and assembly work according to the recognition result.

Japanese Patent Application Laid-Open No. 60-200385

  However, the conventional technique has a problem that a dedicated parts feeder must be used according to the type and shape of the parts. Moreover, since a normal parts feeder can perform only a simple operation, there is a problem that it may be difficult to improve the efficiency of the work of picking up parts from the parts feeder.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following aspects.

(1) According to the first aspect of the present invention, a parts feeder having a parts accommodating part for accommodating parts and a plurality of vibration actuators for vibrating the parts accommodating part, and an end effector for picking up parts from the parts accommodating part And a control device for controlling a robot system comprising: The control device includes a parts feeder controller that controls the parts feeder, and a robot controller that controls the robot. The parts feeder control unit selects one or more control commands from a plurality of control commands each including control parameters of the plurality of vibration actuators, and transmits the selected control command to the parts feeder. The parts feeder is caused to perform an operation corresponding to the selected control command.
According to this control device, since one or more control commands are selected from a plurality of control commands including control parameters of the vibration actuator and transmitted to the parts feeder, the control parameters suitable for the operation of the parts feeder are selected. Can be sent to the feeder. As a result, the parts feeder can be appropriately operated according to the type and shape of the parts. Alternatively, it is possible to improve the efficiency of the work of picking up parts from the parts feeder.

(2) In the above control device, the plurality of control commands may include a separation command for causing the parts feeder to perform a separation operation for separating a plurality of parts assembled in the part storage unit.
According to this control device, it is possible to improve the efficiency of the work of picking up the parts from the parts feeder by separating the parts from each other using a separation command.

(3) In the control device, the plurality of control commands include a posture change command for causing the parts feeder to perform a posture changing operation for changing a posture of a part in the part accommodating portion. It may be a thing.
According to this control device, it is possible to improve the efficiency of the work of picking up a part by changing the attitude of the part using the attitude change command.

(4) In the control device, each of the separation command and the posture change command is a command for vibrating the plurality of vibration actuators, and in the separation command, the vibration duration of the vibration actuator is longer than the posture change command. May be set longer.
According to this control device, in the separation command, the vibration duration of the vibration actuator is set to be longer than that in the posture change command, so that it is possible to separate the parts well.

(5) The control device includes an image recognition unit that performs image recognition for recognizing a part in the part storage unit using an image acquired by a camera that captures an image of a part in the part storage unit. The parts feeder control unit may select one or more control commands from the plurality of control commands using the image recognition result and transmit the selected one to the parts feeder.
According to this control device, since the parts in the part storage unit are recognized by image recognition, it is possible to send a control command corresponding to the recognized result to the parts feeder to operate appropriately.

(6) In the control device, the image recognizing unit includes a replenishment section that receives replenishment of parts from a parts replenishment device, a picking section in which the end effector picks up parts, And when the image recognition recognizes that a part is present in the picking section, the robot controller picks up the recognized part by the end effector. If the robot recognizes that there is no part in the picking section by the image recognition, the parts feeder control unit moves the part from a section other than the picking section to the picking section. To send a feed command to the parts feeder Good.
According to this control device, the parts storage area is virtually divided into a plurality of sections including a supply section and a picking section, and the parts are picked up from the picking section, so that the efficiency of picking up the parts is improved. It is possible. In addition, when it is recognized by image recognition that no part is present in the picking section, the part is moved from the section other than the picking section to the picking section, so that the part can be appropriately moved to the picking section.

(7) In the control device, when the image recognition recognizes that there are only parts that cannot be picked up in the picking section, the part feeder control unit changes the attitude of the parts. The command may be transmitted to the parts feeder.
According to this control device, the posture of a part can be changed from a posture that cannot be picked up to a posture that can be picked up by a posture changing command.

(8) In the control device, the plurality of sections further include an intermediate section provided between the replenishment section and the picking section, and the feed command is configured to transfer parts existing in the supply section to the intermediate section. It is good also as what makes the said parts feeder perform the operation | movement which moves the part which exists in the said intermediate | middle division to the said picking division.
According to this control device, it is possible to appropriately move parts in the order of the replenishment section, the intermediate section, and the picking section using the feed command. In addition, it is possible to efficiently perform the work of picking up the parts in the picking section and the work of supplying the parts to the supply section.

(9) In the control device, the parts storage portion includes a parts storage area and an outer peripheral wall provided on an outer periphery of the parts storage area, and an outer peripheral portion of the parts storage area includes an end effector. There is an interference area where the gripping mechanism and the outer peripheral wall interfere with each other, and after the parts feeder controller separates the parts by a separation command, the parts existing in the interference area are directed to the inside of the parts receiving area. The centering command to be moved may be transmitted to the parts feeder.
According to this control device, since interference between the gripping mechanism of the end effector and the outer peripheral wall can be reduced, it is possible to improve the efficiency of picking up the parts.

(10) In the control device, the image recognition unit may include additional regions used by the end effector gripping mechanism to grip a part at a plurality of locations on an outer edge of each part in the image acquired by the camera. And a recognition process for recognizing a part in which the additional region does not overlap another part as a grippable part in the image. The robot control unit may control the robot so that the grippable part is gripped and picked up by the gripping mechanism of the end effector.
According to this control device, since the gripping mechanism of the end effector recognizes a grippable part in consideration of an additional region used for gripping the part, a part that cannot be gripped by the gripping mechanism is recognized as a part to be gripped. And the efficiency of picking up parts can be improved.

(11) In the control device, the image recognition unit uses an image update process for updating the image by deleting the grippable part from the image after the recognition process, and the updated image. , Executing the process of repeating the recognition process and the image update process, and registering the order in which individual parts are recognized as the grippable parts when the recognition process and the image update process are repeated. It may be a thing. The robot control unit may control the robot so that the parts are gripped and picked up by the gripping mechanism of the end effector according to the order.
According to this control device, a larger number of parts can be recognized as grippable parts by repeating the recognition process and the image update process, so that the efficiency of picking up the parts can be improved.

(12) In the control device, the end effector may include a first pickup mechanism and a second pickup mechanism. At this time, the image recognizing unit recognizes one of the parts existing in the picking section as a first pickable part that can be picked up by the first pickup mechanism; A process of recognizing a second pickable part that can be picked up by the second pick-up mechanism in a state where the pickable part is held by the first pick-up mechanism may be executed.
According to this control device, it is possible to improve the efficiency of the work of picking up parts using two pickup mechanisms.

(13) In the control device, in the process of recognizing the second pickable part, the image recognizing unit holds the first pickable part in the state of holding the first pickable part. For each of the one or more parts that can be picked up, a pickup cost may be calculated according to a predetermined calculation method, and the second pickable part may be selected according to the pickup cost.
According to this control device, it is possible to improve the efficiency of gripping parts by the second pickup mechanism.

(14) In the control device, the control parameter may include a frequency of a vibration signal supplied to the vibration actuator, an amplitude of the vibration signal, and a vibration duration time.
According to this control device, the parts feeder can be operated according to the frequency and amplitude of the vibration signal and the vibration duration time, so that the parts feeder can be operated, so that the efficiency of picking up the parts can be improved. is there.

(15) The control device includes a nonvolatile memory that preliminarily stores control parameters of the plurality of vibration actuators, and the control parameters stored in the nonvolatile memory include (a) each of the plurality of vibration actuators. A balance of vibration intensity, (b) a frequency of the vibration signal capable of activating the movement of the parts existing in the parts accommodating portion, and (c) a part existing in the parts accommodating portion is the parts accommodating portion. And the amplitude of the vibration signal that can be prevented from jumping out to the outside.
According to this control device, appropriate control parameters are stored in advance in a non-volatile memory, so the parts feeder can be operated efficiently using these control parameters, and the efficiency of picking up parts is improved. It is possible to make it.

(16) According to a second aspect of the present invention, there is provided a parts feeder having a parts accommodating part for accommodating parts and a plurality of vibration actuators for vibrating the parts accommodating part; and an end effector for picking up parts from the parts accommodating part. A robot system comprising: a robot; and the control device connected to the parts feeder and the robot.
Also with this robot system, the parts feeder can be appropriately operated according to the type and shape of the parts. Alternatively, it is possible to improve the efficiency of the work of picking up parts from the parts feeder.

  The present invention can be implemented in various forms other than the above. For example, the present invention can be realized in the form of a computer program for realizing the function of the control device, a non-transitory storage medium on which the computer program is recorded, or the like.

The conceptual diagram of the robot system in 1st Embodiment. The block diagram which shows the function of a control apparatus. The top view of a parts accommodating part. Explanatory drawing which shows the feed operation by a feed command. Explanatory drawing which shows the separation operation | movement by a separation command. Explanatory drawing which shows the flip operation | movement by a flip command. Explanatory drawing which shows centering operation | movement by a centering command. Explanatory drawing which shows the mode of the initial replenishment of the parts to a parts storage area. Explanatory drawing which shows the result of the centering operation | movement of parts. Explanatory drawing which shows the result of the separation operation | movement of parts. Explanatory drawing which shows the mode of picking up parts. Explanatory drawing which shows the result of the flip operation | movement of parts. Explanatory drawing which shows the mode of picking up parts. Explanatory drawing which shows the result of the feed operation of parts. Explanatory drawing which shows the state of picking up and replenishing parts. Explanatory drawing which shows the mode of picking up parts. Explanatory drawing which shows the mode of the feed operation of parts. Explanatory drawing which shows the state of picking up and replenishing parts. Explanatory drawing which shows the mode of picking up parts. The flowchart of the parts feeder control in 1st Embodiment. A graph showing the relationship between the number of parts and separation time. The flowchart of the robot control in 1st Embodiment. The flowchart of the parts feeder control in 2nd Embodiment. The conceptual diagram of the robot system in 3rd Embodiment. The flowchart of the parts feeder control in 3rd Embodiment. Explanatory drawing which shows the interference area | region in the outer periphery of parts accommodation area | region. Explanatory drawing which shows operation | movement of the centering command for interference area | region avoidance. Explanatory drawing of the image recognition process of the parts which provided the additional area | region for holding | grip. Explanatory drawing of the image updated so that a grippable part may be erase | eliminated. Furthermore, the explanatory drawing of the image updated. Furthermore, the explanatory drawing of the image updated. Explanatory drawing of a parts coordinate list. Explanatory drawing of the updated parts coordinate list. Furthermore, explanatory drawing of the updated part coordinate list. The flowchart of the parts feeder control in 4th Embodiment. Explanatory drawing which shows the state suitable for backfeed operation | movement. Explanatory drawing which shows the example of the platform state obtained by image recognition. The conceptual diagram of the robot system in 5th Embodiment. Explanatory drawing of the end effector of 5th Embodiment. Explanatory drawing of the recognition process of the parts which can be hold | gripped with two holding | grip mechanisms. Explanatory drawing of the recognition process of the parts which can be hold | gripped with two holding | grip mechanisms. Explanatory drawing of the recognition process of the parts which can be hold | gripped with two holding | grip mechanisms. Explanatory drawing of the recognition process of the parts which can be hold | gripped with two holding | grip mechanisms. The flowchart of the initial setting of the control parameter of a parts feeder. Explanatory drawing of the vibration frequency which activates the movement of parts. Explanatory drawing of the vibration amplitude which does not make parts jump out of a part accommodating part. Explanatory drawing of the number of parts in a feeder which increases the number of detection parts. Explanatory drawing which shows the determination process of the number of parts by simulation. Explanatory drawing of the continuation time of a separation operation | movement.

A. First Embodiment FIG. 1 is a conceptual diagram of a robot system according to a first embodiment. This robot system is installed on a gantry 700 and includes a robot 100, a control device 200, a teaching pendant 300, a parts feeder 400, a hopper 500, and a parts tray 600. The robot 100 is fixed under the top plate 710 of the gantry 700. The parts feeder 400, the hopper 500, and the parts tray 600 are placed on the table portion 720 of the gantry 700. The robot 100 is a teaching playback type robot. The operation using the robot 100 is executed according to teaching data created in advance. In this robot system, a system coordinate system Σs defined by three orthogonal coordinate axes X, Y, and Z is set. In the example of FIG. 1, the X axis and the Y axis are horizontal directions, and the Z axis is a vertical upward direction. The teaching point included in the teaching data and the attitude of the end effector are expressed by the coordinate value of the system coordinate system Σs and the angle around each axis.

  The robot 100 includes a base 120 and an arm 130. The arm 130 is sequentially connected by four joints J1 to J4. Of these joints J1 to J4, three joints J1, J2, and J4 are torsional joints, and one joint J3 is a translational joint. In this embodiment, a four-axis robot is illustrated, but a robot having an arbitrary arm mechanism having one or more joints can be used.

  An end effector 160 a is attached to the arm end 132 which is the tip of the arm 130. In the example of FIG. 1, the end effector 160a is a suction pickup mechanism having a suction nozzle 162 that vacuum-sucks parts. A camera 180 is further attached to the arm 130. This camera 180 is used when selecting a part to be picked up when picking up a part with the end effector 160a. However, the camera 180 can be omitted.

  The parts feeder 400 includes a parts accommodating part 410 that accommodates parts and a vibration part 420 that vibrates the parts accommodating part 410. Under the top plate 710 of the gantry 700, a camera 430 for capturing an image of the parts in the part storage unit 410 is installed.

  The hopper 500 is a parts supply device that supplies parts to the parts feeder 400. In this specification, the phrase “hopper” is used as a term indicating an apparatus for supplying parts, not limited to an apparatus having a funnel shape.

  The parts tray 600 is a tray having a large number of recesses for individually storing parts. In the present embodiment, the robot 100 performs an operation of picking up a part from the part storage unit 410 of the part feeder 400 and storing it in an appropriate position in the part tray 600. However, the robot system can also be applied when performing other work.

  The control device 200 includes a processor 210, a main memory 220, a non-volatile memory 230, a display control unit 240, a display unit 250, and an I / O interface 260. These units are connected via a bus. The processor 210 is, for example, a microprocessor or a processor circuit. The control device 200 is connected to the robot 100, the teaching pendant 300, the parts feeder 400, and the hopper 500 via the I / O interface 260. The control device 200 is further connected to the cameras 180 and 430 via the I / O interface 260.

  As the configuration of the control device 200, various configurations other than the configuration shown in FIG. 1 can be adopted. For example, the processor 210 and the main memory 220 may be deleted from the control device 200 of FIG. 1, and the processor 210 and the main memory 220 may be provided in another device that is communicably connected to the control device 200. In this case, the entire device including the other device and the control device 200 functions as the control device of the robot 100. In other embodiments, the controller 200 may have more than one processor 210. In still another embodiment, the control device 200 may be realized by a plurality of devices connected to be communicable with each other. In these various embodiments, the control device 200 is configured as a device or group of devices that includes one or more processors 210.

  The teaching pendant 300 is a kind of robot teaching device used when a human teaching worker teaches the operation of the robot 100. The teaching pendant 300 has a processor and a memory (not shown). Teaching data created by teaching using the teaching pendant 300 is stored in the nonvolatile memory 230 of the control device 200.

  FIG. 2 is a block diagram illustrating functions of the control device 200. The processor 210 of the control device 200 executes various program instructions 231 stored in advance in the non-volatile memory 230, whereby a robot control unit 211, a parts feeder control unit 212, a hopper control unit 213, and an image recognition unit 214 and the function of the control parameter setting unit 215 are realized. The functions of these units 211 to 215 will be described later. The parts feeder 400 includes a control unit 422 and a plurality of vibration actuators 424. The plurality of vibration actuators 424 are vibrators that vibrate the part housing portion 410 (FIG. 1).

  The nonvolatile memory 230 stores a control parameter 232 and a control command 233 of the vibration actuator 424 and a part coordinate list 234 in addition to the program command 231 and the teaching data 235. The control parameter 232, the control command 233, and the part coordinate list 234 will be described later. The robot control unit 211, the parts feeder control unit 212, and the hopper control unit 213 control the operation of each unit according to the teaching data 235.

  FIG. 3 is a plan view of the part accommodating portion 410. The parts accommodating portion 410 has a parts accommodating area 412 and an outer peripheral wall 414 provided on the outer periphery of the parts accommodating area 412 and extending in the Z direction. The parts accommodation area 412 is a flat, substantially rectangular area. In this example, the X direction is the longitudinal direction of the parts storage area 412, and the Y direction is the short direction of the parts storage area 412. In order to stably accommodate the parts, it is preferable that the surface of the part accommodating region 412 is maintained horizontal. The parts storage area 412 is also referred to as “platform”. A plurality of vibration actuators 424 a to 424 d are installed below the part housing region 412. The white circles indicating the vibration actuators 424a to 424d indicate the planar positions of the vibration actuators 424a to 424d below the part housing region 412. Actually, these vibration actuators 424 a to 424 d are installed in a state where they cannot be visually recognized under the part accommodation area 412. Here, the number of vibration actuators is four, and they are installed at the four corners of the part accommodating area 412. However, the number of vibration actuators is not limited to four, and an arbitrary number of vibration actuators may be provided.

  The lower case letters “a” to “d” added to the end of the reference numerals of the plurality of vibration actuators 424a to 424d are additional numerals attached to distinguish the vibration actuators. The “a” at the end of the reference numeral of the end effector 160a is also an additional reference numeral added to distinguish it from the end effector used in other embodiments. In the following description, when such additional codes are unnecessary, “a” to “d” are omitted.

  The parts accommodation area 412 is divided into three virtual sections RA, RB, and RC. It is preferable that the boundary line between adjacent sections is set in parallel to the short direction (Y direction) of the part storage area 412. The picking section RA is a section where the end effector 160a picks up parts. The replenishment section RC is a section that receives parts from the hopper 500. The intermediate section RB is a section provided between the picking section RA and the supply section RC. The widths (dimensions in the X direction) of these sections are preferably set to be equal to each other. Specifically, when the average value of the widths of the plurality of sections RA to RC is 100%, the width of each section is preferably in the range of 100% ± 10%.

  As the picking section RA, it is preferable to select a section having the shortest tact time of the work of the robot 100 among the plurality of sections RA to RC. In this way, the working efficiency of the robot 100 can be maximized. In the present embodiment, the “tact time” is the time required for one operation when the robot 100 repeats the operation of picking up one part from the parts feeder 400 and storing it in the parts tray 600 a plurality of times. . Normally, the section closest to the parts tray 600 among the plurality of sections RA to RC is selected as the picking section RA.

  The number of sections provided in the part storage area 412 is not limited to 3, and may be 2 or 4 or more. Alternatively, the parts storage area 412 may not be divided into a plurality of sections. However, if the parts storage area 412 is divided into a plurality of sections, the work efficiency of the robot 100 can be improved. In the example of FIG. 3, the boundary lines of a plurality of sections are set parallel to the short direction (X direction). Instead, the boundary lines of the plurality of sections are parallel to the longitudinal direction (Y direction). It may be set. Or you may set the boundary line parallel to a transversal direction (X direction) and the boundary line parallel to a longitudinal direction (Y direction) as a boundary line of a some division, respectively. Specifically, for example, a plurality of sections may be arranged in a 2 × 2 checkerboard shape.

The parts feeder 400 is configured to be able to perform various operations in accordance with various control commands transmitted from the parts feeder control unit 212 of the control device 200. Each control command is configured to include the following control parameters, for example.
(1) Frequency of vibration signal (2) Amplitude of vibration signal (3) Phase of vibration signal (4) Vibration duration “Vibration signal” is given to each vibration actuator 424 from the control unit 422 of the parts feeder 400. It is a signal, and each vibration actuator 424 vibrates according to this vibration signal.

  As a control parameter of the vibration actuator 424, there is a vibration signal waveform suitable for each operation. The waveform of the vibration signal selected by each control command is stored in advance in a nonvolatile memory (not shown) in the control unit 422 (FIG. 2) of the parts feeder 400, for example. In this case, the control unit 422 selects the waveform of the vibration signal according to the control command supplied from the control device 200. However, a control command including a parameter for selecting one waveform from among a plurality of types of vibration signal waveforms stored in the control unit 422 is transmitted from the control device 200 to the control unit 422 of the parts feeder 400. May be.

  Hereinafter, a feed command, a separation command, a flip command, and a centering command will be described as typical control commands. The control parameter of the vibration actuator 424 is also referred to as “vibration parameter”.

  FIG. 4A is an explanatory diagram showing a feed operation by a feed command. Among the plurality of vibration actuators 424a to 424d, vibration actuators in which black circles are added in white circles vibrate in this operation, and vibration actuators to which black circles are not added do not vibrate. The same applies to other operations described later. In the feed operation, the two vibration actuators 424c and 424d at the right end in the longitudinal direction (X direction) of the part storage area 412 vibrate, and the two vibration actuators 424a and 424b at the left end do not vibrate. Further, the two vibration actuators 424c and 424d that are operated vibrate at the same phase. As a result, the parts PP in the parts storage area 412 move in the direction from the right to the left (−X direction). For example, by executing the feed command once, the parts feeder 400 operates so that the parts PP existing in the intermediate section RB move to the picking section RA, and the parts PP present in the supply section RC move to the intermediate section RB. It is possible to make it.

The control parameters included in the feed command are as follows, for example.
(1) Frequency of vibration signal: Frequency that can activate the movement of the part PP (for example, the resonance frequency of the part housing region 412).
(2) Amplitude of vibration signal: An amplitude capable of moving the part PP so as to slide on the part housing area 412.
(3) Phase of the vibration signal: the same phase by the plurality of vibration actuators 424.
(4) Vibration duration: The time during which the part PP can be moved from one section to the next section.

  FIG. 4B is an explanatory diagram showing a separation operation by a separation command. In the separation operation, the plurality of vibration actuators 424a to 424d operate simultaneously. Further, the four vibration actuators 424a to 424d vibrate with the same phase. In this separation operation, it is possible to separate a plurality of parts PP that are gathered in the part storage area 412. By separating the parts PP from each other using such a separation command, it is possible to improve the efficiency of the work of picking up the parts from the parts feeder 400.

The control parameters included in the separation command are, for example, as follows.
(1) Frequency of vibration signal: Frequency that can activate the movement of the part PP (for example, the resonance frequency of the part housing region 412).
(2) Amplitude of vibration signal: Amplitude as large as possible as long as the part PP does not jump out of the part accommodating area 412.
(3) Phase of the vibration signal: the same phase by the plurality of vibration actuators 424.
(4) Vibration continuation time: The time during which the parts PP assembled in the intermediate section RB can be distributed almost uniformly in the plurality of sections RA to RC.

  FIG. 4C is an explanatory diagram showing a flip operation by a flip command. In the flip operation, the plurality of vibration actuators 424a to 424d operate simultaneously. Further, the four vibration actuators 424a to 424d vibrate with the same phase. In this flip operation, the parts PP in the part storage area 412 can be turned over. The sand-patterned part PPf is a front-facing part, and the hatched hatched part PPb is a back-facing part. The flip operation is an operation of turning over these parts PP.

The control parameters included in the flip command are, for example, as follows.
(1) Frequency of vibration signal: Frequency that can activate the movement of the part PP (for example, the resonance frequency of the part housing region 412).
(2) Amplitude of vibration signal: Amplitude as large as possible as long as the part PP does not jump out of the part accommodating area 412
(3) Phase of the vibration signal: the same phase by the plurality of vibration actuators 424.
(4) Vibration continuation time: a time that is as short as possible and increases the number of parts PP turned over.
In the flip command, the vibration duration period is set shorter than that in the separation command. That is, in the separation command, the vibration duration of the vibration actuator 424 is set longer than that of the flip command.

  The waveform of the vibration signal supplied to the vibration actuator 424 in accordance with the flip command is preferably a waveform that can flip the part PP.

  The flip command is a kind of posture change command for changing the posture of the part PP. As the posture change command other than the flip command, for example, a rotation command for rotating the part PP on the surface of the part receiving area 412 can be used. In the rotation operation by the rotation command, the part PP rotates around the vertical direction (Z direction). For example, as in other embodiments described later, when a part PP is picked up using a gripping mechanism that grips the part PP, a posture change command other than a flip command such as a rotation command may be effective. If the posture of the part PP is changed using the posture change command, the efficiency of the work of picking up the part PP can be improved.

  By using a separation command and a posture change command such as a flip command, it is possible to improve the efficiency of picking up parts. In the separation command, the vibration duration of the vibration actuator is set to be longer than that of a posture change command such as a flip command, so that the parts can be well separated.

  FIG. 4D is an explanatory diagram showing a centering operation by a centering command. In the centering operation, the plurality of vibration actuators 424a to 424d operate simultaneously. 4D moves the part PP toward the center in the longitudinal direction (X direction) of the part receiving area 412, so that the vibration actuators 424a and 424b at one end in the X direction are connected to the other end. The vibration actuators 424c and 424c in FIG. As another centering operation, an operation of moving the part PP toward the center in the short direction (Y direction) of the part housing region 412 is also possible. In this centering operation, the vibration actuators 424a and 424c at one end in the Y direction vibrate at a phase different by 180 degrees from the vibration actuators 424b and 424c at the other end.

The control parameters included in the centering command are as follows, for example.
(1) Frequency of the vibration signal: a frequency suitable for activating the movement of the part PP (for example, a resonance frequency of the part housing region 412).
(2) Amplitude of vibration signal: Amplitude as large as possible as long as the part PP does not jump out of the part accommodating area 412
(3) Phase of the vibration signal: the vibration actuator 424 at one end and the vibration actuator 424 at the other end have opposite phases.
(4) Vibration duration: A time suitable for the purpose of the centering operation.
As the “object of the centering operation”, for example, (a) assembling the parts PP in the center of the part receiving area 412 as preprocessing of the separation operation described in FIG. Two different purposes can be set: moving the parts PP existing in the interference area (described in the third embodiment) on the outer periphery of the housing area 412 toward the inside of the parts housing area 412. The vibration duration time of the centering operation with different purposes is set to an appropriate time according to each purpose.

  The parts feeder control unit 212 selects one or more control commands from among a plurality of control commands each including control parameters of the plurality of vibration actuators 424, and transmits the selected control commands to the parts feeder 400 to select them. The parts feeder 400 is caused to perform an operation corresponding to the control command. Therefore, control parameters suitable for the operation of the parts feeder 400 can be transmitted to the parts feeder 400. As a result, the parts feeder 400 can be appropriately operated according to the type and shape of the parts PP. Alternatively, the efficiency of the work of picking up the parts PP from the parts feeder 400 can be improved.

  Control parameters suitable for each control command are set in advance by the control parameter setting unit 215 and stored in the nonvolatile memory 230 before the robot 100 picks up the parts PP. The initial setting of such control parameters will be described after various embodiments relating to the picking up operation of the parts PP.

  FIG. 5A to FIG. 5H are explanatory diagrams showing how the parts move in the part accommodation area 412 in the picking-up operation of the parts PP by the robot 100. 5A to 5H, the illustration of the vibration actuator 424 is omitted.

  FIG. 5A shows an initial state in which the parts PP are first supplied from the hopper 500 to the parts accommodating area 412. The hopper 500 supplies a plurality of parts PP in the supply section RC. A part PP with a sand pattern is a face-up part, and a part PP with a hatched area is a face-down part. When the centering operation is performed in the state of FIG. 5A, as shown in FIG. 5B, the parts PP gather at the center of the part accommodating area 412 (that is, the intermediate section RB). Then, when the separation operation is performed, as shown in FIG. 5C, the parts PP are dispersed almost over the entire part accommodating area 412. As described above, by performing the centering operation before the separation operation, the parts PP can be more uniformly distributed. However, depending on the type and number of parts PP, the insertion method, and the insertion position, the parts PP can be obtained only by the separation operation. May be dispersed uniformly. In such a case, the centering operation may be omitted. In the state of FIG. 5C, normally, there are a front-facing part PPf and a back-facing part PPb in each of the sections RA, RB, RC. The operation of the robot 100 in the present embodiment is an operation of picking up the face-up parts PPf and storing them in the parts tray 600. Detection of the front-facing part PPf and the back-facing part PPb is performed by the image recognition unit 214 executing an image recognition process on the image acquired by the camera 430.

  FIG. 5D shows a state in which the robot 100 has picked up all the front parts PPf present in the picking section RA from the state of FIG. 5C. Next, when the flip operation is performed and the back-facing part PPb is turned face up, the state shown in FIG. 5E is obtained. In FIG. 5E, the parts PP existing in the sections RB and RC are simplified in the same state as in FIG. 5D, but actually, the parts PP of these sections RB and RC are turned over by the flip operation. The part PPf turned upside down by the flip operation is picked up from the picking section RA by the robot 100. Note that the flip operation and the pick-up operation of the parts PP are repeated until the pick-up of all the parts PP in the picking section RA is completed.

  FIG. 5F shows a state where the picking up of all the parts PP in the picking area RA has been completed. Next, when a feed operation is performed to move the parts PP in the parts storage area 412 in the direction of the picking section RA, the state shown in FIG. 5G is obtained. This feed operation is performed so as to move the parts PP from the respective sections RB and RC to the adjacent sections RA and RB. Then, with respect to the parts PP existing in the picking section RA, the pick-up operation and the flip operation of the parts PP described in FIGS. 5C to 5F are performed, and all the parts PP in the picking section RA are picked up. In the state of FIG. 5G, since there is no part PP in the replenishment section RC, even if the part PP is replenished from the hopper 500 to the replenishment section RC in parallel with the pick-up operation and the flip operation of the part PP in the picking section RA. Good.

  FIG. 5H shows a state where the picking up of the part PPf facing up in the picking section RA in FIG. 5G is completed and the part PP is supplied to the supply section RC. When all the parts PP existing in the picking section RA are picked up by repeating the flip operation and picking up operation of the parts PP as necessary from the state of FIG. 5H, the state of FIG. 5I is obtained. Next, when a feed operation is performed to move the parts PP in the parts storage area 412 in the direction of the picking section RA, the state shown in FIG. 5J is obtained. Then, with respect to the parts PP existing in the picking section RA, the pick-up operation and the flip operation of the parts PP described in FIGS. 5C to 5F are performed, and all the parts PP in the picking section RA are picked up. In the state of FIG. 5J, as in the state of FIG. 5G, since there is no part PP in the replenishment section RC, the replenishment section RC is supplied from the hopper 500 in parallel with the pick-up operation and the flip operation of the parts PP in the picking section RA. Parts PP may be replenished.

FIG. 5K shows a state where the picking up of the part PPf facing up in the picking section RA in FIG. 5J is completed and the part PP is supplied to the supply section RC. In FIG. 5K, the parts PP existing in the intermediate section RB are parts supplied to the supply section RC in FIG. 5I. The number of parts PP supplied to the supply section RC in FIGS. 5I and 5K is determined experimentally in advance and stored in the non-volatile memory 230. In the present embodiment, the replenishment number per time is set to a value that is ½ of the initial replenishment number in FIG. 5A, for example. In general, when the parts storage area 412 is divided into N ₄₁₂ sections (N ₄₁₂ is an integer of 2 or more), the replenishment quantity per second and subsequent times is 1 / (N it is preferably set to a value of ₄₁₂ -1).

  Instead of setting the number of replenished parts PP to a constant value, the number of parts PP picked up from the part storage area 412 may be used as the replenished number. Specifically, the number of parts PP picked up from the previous supply to the current supply may be used as the supply. In this way, since the number of parts PP existing in the part accommodation area 412 after supply is constant, the efficiency of picking up work is improved.

  When all the parts PP existing in the picking section RA are picked up by repeating the flip operation and picking up operation of the parts PP as necessary from the state of FIG. 5K, the state of FIG. 5L is obtained. After FIG. 5L, operations similar to those described in FIGS. 5B to 5K (that is, various operations after the centering operation) are executed. By doing so, the part PP can be picked up while the parts PP are appropriately supplied into the parts feeder 400, so that the part PP can be picked up efficiently.

  FIG. 6 is a flowchart of parts feeder control in the first embodiment. This control is executed by the parts feeder control unit 212 unless otherwise specified. Further, the control in FIG. 6 is repeatedly executed, for example, at regular intervals.

The parameters used in FIG. 6 are as follows.
nParts: Number of detected front parts PPf in the picking section RA.
nPartsBack: Number of detected parts PPb in the picking section RA.
syncLock: Synchronous control parameter for parts feeder control and robot control. When syncLock is true, the pick-up operation of the robot 100 is prohibited, and when syncLock is false, the pick-up operation of the robot 100 is permitted.
nFeed: Feed operation counter value.
N ₄₁₂ : The number of sections of the part accommodating area 412. In this embodiment, N ₄₁₂ = 3.

  In step S110, it is determined whether or not the detected number nParts of the facing parts PPf in the picking section RA is 1 or more. nParts is a value detected in step S240 described later, and its initial value (default value) is zero. When nParts is 1 or more, the processing of FIG. 6 is terminated, and the robot 100 picks up the parts PP. Since nParts is zero when the determination in step S110 is first executed, the process proceeds to step S120. In step S120, the synchronization control parameter syncLock is set to true, and the pick-up operation of the robot 100 is prohibited.

  In step S130, it is determined whether or not the detected number nPartsBack of the facing parts PPb in the picking section RA is 1 or more. nPartsBack is a value detected in step S250 to be described later, and its initial value (default value) is zero. If nPartsBack is 1 or more, a flip operation is executed in step S140, and the process proceeds to step S240 described later. In other words, when it is recognized by image recognition that there is only a part PP that cannot be picked up in the picking area RA, a flip operation for turning over the part PP is executed. In step S140, another type of posture changing operation may be performed instead of the flip operation. This point will be described in a third embodiment. Note that nPartsBack is zero when the determination in step S130 is executed for the first time, and the process proceeds to step S180.

  In step S180, it is determined whether the feed operation counter value nFeed is 1 or more. nFeed is a value that is set in step S230, which will be described later, and is changed in step S160. The initial value (default value) is zero. When nFeed is 1 or more, the process proceeds to step S150 described later. When the determination in step S180 is executed for the first time, nFeed is zero, and the process proceeds to step S190.

  In step S190, it is determined whether or not the processing after step S190 is executed for the first time. In the case of the first time, the process proceeds to step S200, and the initial parts supply from the hopper 500 to the parts feeder 400 is executed. This parts replenishment is the operation described with reference to FIG. 5A, and is executed when the hopper control unit 213 transmits a control command to the hopper 500. If the process after step S190 is not the first time, step S200 is skipped and the process proceeds to step S210.

In step S210, the centering operation described in FIG. 5B is executed, and in step S220, the separation operation described in FIG. 5C is executed. In step S230, nFeed is set to (N ₄₁₂ −1). In this embodiment, since N ₄₁₂ = 3, nFeed is 2. After step S230, the process proceeds to step S240 described later.

  Returning to step S180, if nFeed is 1 or greater, the process proceeds to step S150. nFeed is 1 or more in the state of FIGS. 5F and 5I described above. In this case, a feed operation is performed in step S150, nFeed is decremented by 1 in step S160, and parts are replenished in step S170. This parts replenishment is the operation described in FIGS. 5H and 5K. After step S170, the process proceeds to step S240. Note that step S170 may be executed after permitting the pick-up operation of the robot 100 in step S260 described later.

  In step S240, the number nParts of the facing parts PPf in the picking section RA is detected. This detection process is performed by performing image recognition in which the image recognition unit 214 recognizes the part PP existing in the picking section RA using the image captured by the camera 430. This image recognition can be realized, for example, by storing template images of the front part PPf and the back part PPb in the nonvolatile memory 230 in advance and executing template matching on the image captured by the camera 430. It is. When the facing part PPf is detected, the detected number is set as the value of nParts, and the coordinates of the detected facing part PPf are registered in the part coordinate list 234 (FIG. 2). Note that, for example, the coordinate value of the system coordinate system Σs (FIG. 1) is used as the coordinates of the front part PPf. In step S250, the number nPartsBack of the facing parts PPb in the picking section RA is detected. This detection process is also performed by executing image recognition in which the image recognition unit 214 recognizes the part PP existing in the picking section RA using the image captured by the camera 430. When the backward part PPb is detected, the detected number is set as the value of nPartsBack.

  In step S260, syncLock is set to false, the pick-up operation of the robot 100 is permitted, and the process of FIG.

  If it is determined in step S190 that it is not the first time, the part pick-up operation in the pickup section RA has been completed, and the number of parts in the part storage area 412 is small. In this case, the time of the separation operation in step S220 (referred to as “separation time”) may be shorter than the separation time in the case of executing step S220 for the first time.

  FIG. 7 shows an example of graphs G1 and G2 showing the relationship between the number of parts in the part storage area 412 and the separation time. Each of the graphs G1 and G2 shows a characteristic that the separation time is shortened as the number of parts is small. The graph G1 is an upwardly convex curve, and the graph G2 is stepped. The number of parts in the part storage area 412 can be acquired by an image recognition process using the camera 430 (FIG. 1). Thus, if the separation time is set to be shorter as the number of parts in the parts storage area 412 is smaller, the overall work time can be further shortened.

  FIG. 8 is a flowchart of the robot control in the first embodiment. This control is executed by the robot control unit 211. Further, the control of FIG. 8 is repeatedly executed at regular intervals, for example.

  In step S310, the process waits until syncLock changes from true (operation prohibited) to false (operation permitted). In step S320, one coordinate of the part PP to be picked up by the robot 100 is extracted from the part coordinate list 234. In step S330, the end effector 160 is used to pick up one part PP. In step S340, the picked-up part PP is moved to the target position. In the present embodiment, the target position is an empty position in the parts tray 600.

  The processing procedure described in FIGS. 6 and 8 is an example, and can be arbitrarily changed. For example, when there is no distinction between the front and back of the parts PP, steps S140 and S250 in FIG. 6 can be omitted. Further, as described above, step S170 may be executed after step S260.

  As described above, by executing the parts feeder control according to the procedure of FIG. 6 and the robot control according to the procedure of FIG. 8, while appropriately supplying the parts PP into the parts feeder 400 as described in FIGS. 5A to 5L, Picking up parts PP can be performed. As a result, the part PP can be picked up efficiently.

  In the above description, the parts feeder control unit 212 uses the result of image recognition by the image recognition unit 214 to perform a flip operation (step S140), a feed operation (step S150), a centering operation (step S210), a separation. Various operations such as the operation (step S220) are performed by the parts feeder 400, but other operations may be performed by the parts feeder 400. Also in this case, the parts feeder control unit 212 uses one or more control commands from among a plurality of control commands using the image recognition result for the image of the parts PP in the parts container 410 acquired by the camera 430. It is preferable to select and transmit to the parts feeder. In this way, since the parts in the part storage unit 410 are recognized by image recognition, it is possible to send an appropriate control command corresponding to the recognized result to the parts feeder 400 to operate appropriately.

B. Second Embodiment FIG. 9 is a flowchart of parts feeder control in the second embodiment. The only difference from the flowchart of the first embodiment shown in FIG. 6 is that steps S170 and S190 are deleted, and the other steps are the same as in FIG.

  In the second embodiment, when parts are initially replenished in step S200, new parts are not replenished until all of those parts are picked up, and after all the parts in the part accommodating area 412 are picked up, step S200 is performed. The replenishment is executed again at. According to the second embodiment, substantially the same effect as that of the first embodiment can be obtained.

C. Third Embodiment FIG. 10 is a conceptual diagram of a robot system according to a third embodiment. This robot system is the same as the robot system of the first embodiment (FIG. 1) except for the end effector 160b. The end effector 160 b is a gripper that grips and picks up a part using the gripping mechanism 164.

  FIG. 11 is a flowchart of parts feeder control in the third embodiment. The only difference from the flowchart of the first embodiment shown in FIG. 6 is that step S225 is added between step S220 and step S230, and the other steps are the same as in FIG. In the state where the separation operation in step S220 is completed, the parts PP are almost uniformly dispersed in the part accommodating area 412.

  FIG. 12A shows a state in which the separation operation in step S220 has been completed in the third embodiment. Three concave portions around the individual parts PP indicate portions to be gripped by the gripping mechanism 164 of the end effector 160b. However, as the gripping mechanism 164, a mechanism that grips at two points may be used instead of a mechanism that grips at three points. The part accommodating part 410 has an outer peripheral wall 414 provided on the outer periphery of the part accommodating area 412. An interference region Rint where the gripping mechanism 164 of the end effector 160b interferes with the outer peripheral wall 414 exists in the outer peripheral portion of the part accommodating region 412. The part PP that partially or entirely overlaps the interference region Rint may not be gripped due to physical interference between the gripping mechanism 164 and the outer peripheral wall 414. Therefore, when picking up the parts PP by the robot 100, it is preferable that there is no part PP that partially or entirely overlaps the interference region Rint.

  In step S225 of FIG. 11, the parts feeder control unit 212 transmits a centering command to the parts feeder 400 to execute a centering operation. FIG. 12B shows the result of this centering operation. As a result of the part PP existing in the interference area Rint in FIG. 12A moving toward the inside of the part accommodating area 412, a part or all of the parts PP are present in the interference area Rint. There is no part PP that overlaps. In this centering operation, the vibration duration time is set to a shorter value than the centering operation in step S210 described in FIG. 5B, and the moving distance of the parts PP is short. By performing such a centering operation, the interference between the gripping mechanism 164 of the end effector 160b and the outer peripheral wall 414 can be reduced, so that the efficiency of picking up the parts PP can be improved.

  In the third embodiment, since the part PP is picked up using the gripping mechanism 164, as will be described below, additional regions for gripping are set at a plurality of locations on the outer edge of the part PP, and the part PP It is preferable to execute detection (processing of step 240).

  FIG. 13A is an explanatory diagram of an image recognition process for a part PP provided with an additional area AD for gripping. Here, for convenience of illustration, it is assumed that only the picking section RA is drawn in a state rotated 90 degrees clockwise from the direction of FIG. 12B, and that all parts PP are face up. The interference region Rint is not shown. At the three gripping positions on the outer edge of each part PP, areas required when the gripping mechanism 164 grips are indicated by dotted lines as additional areas AD.

  In step S240 of FIG. 11, the image recognition unit 214 executes setting processing for setting each additional region AD in each part PP in the image acquired by the camera 430. In addition, the image recognition unit 214 assigns a part number to each part PP. In FIG. 13A, the number drawn at the center of each part PP indicates this part number. The image recognition unit 214 further executes a recognition process for recognizing, as a grippable part, a part PP in which the additional area AD does not overlap with another part PP among the parts PP existing in the picking section RA. The “other parts PP” at this time means the outer shape of the parts PP without the additional area AD. In the example of FIG. 13A, the grippable parts are eleven parts PP having part numbers 1 to 5, 7, 8, 12, 15, 16, and 18. The image recognition unit 214 registers the part PP recognized as a grippable part in the part coordinate list 234.

  FIG. 14A shows an example of the part coordinate list 234. Here, 11 parts PP recognized as grippable parts by the image recognition processing with respect to FIG. 13A are registered. The registered contents are a part number n and its coordinate value (Xn, Yn). The robot control unit 211 controls the robot 100 to grip and pick up grippable parts registered in the part coordinate list 234 with the gripping mechanism 164 of the end effector 160b. In this way, since the grippable part is recognized in consideration of the additional area AD necessary for gripping, it is possible to prevent the part PP that cannot be gripped by the gripping mechanism 164 from being recognized as the part to be gripped. It is possible to improve the efficiency of picking up work. Such recognition of the grippable part in consideration of the additional area AD may be performed only for the part PP in the picking section RA, or may be performed for the entire part accommodating area 412. However, the processing time can be shortened by recognizing the grippable parts in consideration of the additional area AD only for the parts PP in the picking section RA.

  Note that the image recognition unit 214 may perform an image update process for updating the image by deleting the grippable parts from the image after the recognition process in FIG. 13A. FIG. 13B shows the image thus updated. In this image, 11 parts PP recognized as grippable parts in FIG. 13A are erased, and their outer shapes are drawn with broken lines for convenience of illustration. The image recognition unit 214 performs the recognition process described with reference to FIG. 13A again using the updated image. The grippable parts recognized in FIG. 13B are five parts PP whose part numbers are 6, 9, 11, 13, and 19. The image recognizing unit 214 additionally registers the part PP recognized as a grippable part again in the part coordinate list 234.

  FIG. 14B shows a state in which five parts PP recognized as grippable parts by the image recognition processing for FIG. 13B are additionally registered. After the recognition processing in FIG. 13B, the image recognition unit 214 executes image update processing for updating the image by deleting the grippable parts from the image.

  FIG. 13C shows the updated image. In this image, the five parts PP recognized as grippable parts in FIG. 13B are deleted. As for the updated image, the recognition process and the image update process are executed as described above. Note that the grippable parts recognized in FIG. 13C are one part PP having a part number of 14. The image recognition unit 214 additionally registers the part PP recognized as a grippable part in the part coordinate list 234. FIG. 14C shows a state in which one part PP recognized as a grippable part by the image recognition processing for FIG. 13C is additionally registered. After the recognition processing in FIG. 13C, the image recognition unit 214 executes image update processing for updating the image by deleting the grippable parts from the image. FIG. 13D shows the image thus updated.

  As described with reference to FIGS. 13B to 13D, the image recognition unit 214 repeats the recognition process and the image update process, and registers the order in which individual parts are recognized as grippable parts in the part coordinate list 234. It is preferable to do. Thereafter, the robot control unit 211 can control the robot 100 so as to grip and pick up the parts PP in the picking section RA in accordance with the order registered in the part coordinate list 234. In this way, since a larger number of parts PP can be recognized as grippable parts, the efficiency of picking up the parts PP can be improved. However, the recognition process described in FIGS. 13B to 13D may not be repeated and the recognition process may be executed only once in FIG. 13A.

  In the third embodiment, in step S140 of FIG. 11, another type of posture changing operation may be performed instead of the flip operation. For example, as the posture changing operation, a rotation operation for rotating the parts PP on the surface of the part housing area 412 may be performed. In this way, by rotating the part PP that cannot be gripped by the gripping mechanism 164, it is possible to change the posture so that the part PP can be gripped. This rotation operation is also common to the flip operation used in the first embodiment in the sense that the posture of the part PP that cannot be picked up is changed to the part PP that can be picked up. As described above, generally, step S140 is a step of executing an attitude changing operation for changing the attitude of the part PP when it is recognized by the image recognition that only the part PP that cannot be picked up is present in the picking area RA. It is possible to think that there is.

D. Fourth Embodiment FIG. 15 is a flowchart of parts feeder control in a fourth embodiment. The only difference from the flowchart of the third embodiment shown in FIG. 11 is that step S155 is added after step S150, and the other steps are the same as in FIG. In step S155, a back feed operation is executed. This back feed operation is an operation of moving parts in the direction opposite to the feed operation executed in step S150. The back feed operation time is preferably shorter than the feed operation time in step S150.

  FIG. 16 is an explanatory diagram showing a state suitable for the back feed operation. In this example, the parts PP are unevenly distributed in the robot end region EA (shown with hatching) by the feed operation in step S150. In such a state, since the proportion of parts PP that are not suitable for picking up work by the robot is large, it is possible to make the state suitable for picking up work by back feeding to the hopper side.

Note that the necessity of the back feed operation in step S155 and the back feed time may be determined using the result of the image recognition process. This determination can be made using, for example, the uneven distribution ratio Ru of parts calculated according to the following equation.
Ru = Sp / Se (1)
Here, Sp is the total area of the parts PP in the end area EA, and Se is the area of the end area EA. For example, when the part PP is recognized as a black image, the part area Sp can be calculated as the number of black pixels in the end area EA.

  The width Wea of the end area EA is set smaller than the width of the pickup section RA. For example, the width Wea of the end area EA is preferably set to a value in the range of 1 to 2 times the width of the part PP. In the example of FIG. 16, the end area EA is set on the left side of the parts accommodating area 412, but the position of the end area EA is set according to the moving direction of the parts PP in the feed operation. That is, the end area EA is preferably set in the vicinity of a side that is the end of the direction of the feed operation among the four sides of the part housing area 412.

  When the parts uneven distribution ratio Ru is equal to or greater than a predetermined threshold, the parts PP are unevenly distributed as shown in FIG. 16, and therefore, it is preferable to perform the back feed operation in step S155. Further, the time for the back feed operation may be determined according to the uneven distribution rate Ru. Specifically, it is preferable to increase the back feed time as the uneven distribution ratio Ru increases.

  Various controls may be executed using image recognition results other than the uneven distribution rate Ru of parts.

  FIG. 17 shows examples of various platform states obtained by image recognition. “Platform” means the part receiving area 412. Here, the following seven states are illustrated.

<State 1: Pickable state>
The parts are dispersed in the platform so that they can be picked up.
<State 2: Empty state>
There are no parts in the platform.
<State 3: Pick position uneven distribution state>
The parts are unevenly distributed at the end of the platform. This state 3 corresponds to the state described with reference to FIG.
<State 4: excessive number of remaining parts>
The remaining number of parts is 20% more than the appropriate number.
<State 5: Number of remaining parts is too small>
The remaining number of parts is 20% or more less than the appropriate number.
<State 6: Excessive number of remaining back parts>
The number of back parts is 10% more than the appropriate number.
<State 7: No pickable parts>
There are parts in the platform, but there are no parts to be picked up, only other types of parts.

  These states can be used for executing various controls and adjusting control contents. For example, when the state 2 or the state 7 is recognized at an arbitrary time, the process may jump to step S200 in FIG. 15 to supply the parts. If state 4 or state 5 is recognized, the number of replenishments in step S170 may be changed according to the remaining number of parts. As described above, if the control contents are adjusted according to various image recognition results, the work efficiency can be further improved. Such adjustment of the control content according to the image recognition result can be applied to other embodiments. In addition, the point where the back feed operation of step S155 is performed after the feed operation of step S150 is also applicable to other embodiments.

E. Fifth Embodiment FIG. 18 is a conceptual diagram of a robot system in a fifth embodiment. This robot system is the same as the robot system of the first embodiment (FIG. 1) and the third embodiment (FIG. 10) except for the end effector 160c. The end effector 160c is a double-handed gripper capable of gripping and picking up two parts by using two gripping mechanisms 164.

  FIG. 19 is a plan view of the end effector 160c. The end effector 160c has two gripping mechanisms 164a and 164b and vertical movement mechanisms 166a and 166b. The gripping mechanisms 164a and 164b are grippers that grip the part PP at three points in this example. The vertical movement mechanisms 166a and 166b can move the gripping mechanisms 164a and 164b in the vertical direction (Z direction), respectively, and change the heights of the two gripping mechanisms 164a and 164b. Note that one of the two vertical movement mechanisms 166a and 166b may be omitted, and the relative height of the two gripping mechanisms 164a and 164b may be changed using one vertical movement mechanism.

  20A to 20D are explanatory diagrams illustrating a recognition process of the parts PP that can be gripped by the two gripping mechanisms 164a and 164b. First, the image recognition unit 214 recognizes a part PP1 that can be gripped by the first gripping mechanism 164a (FIG. 20A). This part PP1 is referred to as “first grippable part PP1”. Next, the image recognition unit 214 recognizes the position of the second gripping mechanism 164b while gripping the first grippable part PP1 by the first gripping mechanism 164a (FIG. 20B). At this time, the image recognition unit 214 uses the horizontal positional relationship between the two gripping mechanisms 164a and 164b to calculate the coordinates and gripping angle (angle around the Z axis) of the second gripping mechanism 164b. Then, the part PP2 that can be gripped by the second gripping mechanism 164b is recognized (FIG. 20C). This part PP2 is referred to as “second grippable part PP2”. As described above, when the process of recognizing the second grippable part PP2 that can be gripped by the second gripping mechanism 164b in the state where the first grippable part PP1 is gripped by the first gripping mechanism 164a, It is possible to improve the efficiency of the work of picking up the parts PP using the gripping mechanisms 164a and 164b.

  As the second grippable part PP2, it is preferable to select a part PP that is most easily gripped by the second gripping mechanism 164b in a state where the first grippable part PP1 is gripped by the first gripping mechanism 164a. This selection can be performed according to the pickup cost, for example. The “pickup cost” is a predetermined calculation for one or more parts PP that can be gripped by the second gripping mechanism 164b in a state where the first grippable part PP1 is gripped by the first gripping mechanism 164a. Calculated according to the method.

As a method for calculating the pickup cost, for example, the following various methods are conceivable.
(1) Pickup cost calculation method 1
To grip one or more parts PP in the vicinity of the second gripping mechanism 164b from the state of gripping the first grippable part PP1 by the first gripping mechanism 164a (FIG. 20A) with the second gripping mechanism 164b. The required trajectory of the robot 100 is calculated, and the time required to move the trajectory is taken as the pickup cost.
(2) Pickup cost calculation method 2
From the state in which the first grippable part PP1 is gripped by the first gripping mechanism 164a (FIG. 20A), one or more parts PP in the vicinity of the second gripping mechanism 164b are connected to the second gripping mechanism 164b and each part PP. The distance is calculated and the distance is set as the pickup cost.
(3) Pickup cost calculation method 3
To grip one or more parts PP in the vicinity of the second gripping mechanism 164b from the state of gripping the first grippable part PP1 by the first gripping mechanism 164a (FIG. 20A) with the second gripping mechanism 164b. The required rotation angle of the end effector 160c (rotation angle of the torsional joint J4) is calculated, and the rotation angle is used as the pickup cost.

  The second grippable part PP2 shown in FIG. 20C is a part that minimizes the pickup cost calculated by the calculation method 2. The second grippable part PP2 shown in FIG. 20D is a part that minimizes the pickup cost calculated by the calculation method 3. If the position for gripping the part PP is predetermined according to the shape of the part PP as in the part PP of this embodiment, the calculation method 1 (trajectory reference) or the calculation method 3 (rotation angle) Standard) is suitable. On the other hand, when the position for picking up by the end effector 160 does not depend on the shape of the part PP (for example, when using the suction pickup mechanism) as in the part PP of the first embodiment, the calculation method 1 (trajectory reference) or Calculation method 2 (distance reference) is suitable.

  As described above, one or more parts PP that can be gripped by the second gripping mechanism 164b in a state where the first grippable part PP1 is gripped by the first gripping mechanism 164a are picked up according to a predetermined calculation method. If the cost is calculated and the second grippable part PP2 is selected according to the pickup cost, the efficiency of gripping the part by the second gripping mechanism 164b can be improved.

  The selection of the two grippable parts as described above can also be applied to a robot including an end effector having two pickup mechanisms (for example, a suction pickup mechanism) other than the gripping mechanism 164. In this case, the image recognizing unit 214 performs processing for recognizing the second pickable part PP2 that can be picked up by the second pickup mechanism in a state where the first pickable part PP1 is held by the first pickup mechanism. To do. In this way, it is possible to improve the efficiency of picking up parts by the second pickup mechanism.

F. Initial Setting of Parts Feeder Control Parameters FIG. 21 is a flowchart of initial setting of control parameters of the parts feeder 400, and FIGS. 22A to 22E are explanatory diagrams showing processing contents of steps S420 to S450 of FIG. is there. This process is executed before the above-described robot 100 picks up the parts PP. Further, this process is executed by acquiring an image of the part PP in the part storage area 430 with the camera 430 and analyzing the image by the control parameter setting unit 215.

  In step S410, the balance adjustment of the vibration intensity of the plurality of vibration actuators 424 is performed. This adjustment is performed in order to compensate for the inclination of the part accommodating region 412 and the difference in the characteristics of the individual vibration actuators 424. Specifically, for example, a plurality of parts PP are accommodated in the part accommodating area 412, and the plurality of vibration actuators 424 are vibrated in the same phase to acquire the coordinates (XY coordinates) of the individual parts PP. Then, the amplitudes of the vibration signals supplied to the individual vibration actuators 424 are adjusted so that the coordinates of the plurality of parts PP are not biased and the average value of these coordinates is in the center of the part housing area 412. The balance of vibration intensity adjusted in this way is also used after step S420.

  In step S420, a frequency capable of activating the movement of the part PP is detected. In this detection process, for example, one part PP is accommodated in the part accommodating area 412, and a predetermined number of vibration actuators 424 are vibrated to acquire the movement amount of the part PP. Then, the frequency of the vibration signal that maximizes the amount of movement of the parts PP is adjusted.

  FIG. 22A shows an example of the relationship between the frequency and the part activity (movement amount of part PP) in step S420. In this example, the frequency Fc at the peak of the graph is detected as a frequency that can activate the movement of the part. This frequency Fc is, for example, a value equal to the resonance frequency of the part housing region 412. The appropriate frequency determined in this way is also used after step S430.

  In addition, as the number of the vibration actuators 424 used in step S420, an arbitrary number of 1 or more can be used. In addition, for each combination of the number of used vibration actuators 424 and the use location, a frequency that can activate the movement of the parts may be detected. For example, when the parts feeder 400 has four vibration actuators 424a to 424d, and one, two, or four of them are used, the number of vibration actuators 424 used and the number of combinations of use points. There are 11 ways at the maximum. In addition, when using two or four vibration actuators 424, even if a frequency capable of activating the movement of a part is detected for each phase difference value (for example, 0 degrees and 180 degrees). Good. As described above, it is preferable to set an appropriate control parameter for each combination of the number and use locations of the vibration actuators 424. The same applies to other control parameters described later.

  In step S430, an amplitude capable of preventing the part PP from jumping out is detected. This amplitude is as large as possible as long as the part PP does not jump out of the part housing area 412. In this detection process, for example, a plurality of parts PP are accommodated in the part accommodating region 412, and the plurality of vibration actuators 424 are vibrated to determine whether or not the parts PP jump out of the part accommodating portion 410 from the image of the camera 430. judge. This determination is performed while gradually increasing the amplitude of the vibration signal, and the maximum amplitude at which the protrusion of the part PP is not detected is obtained.

  FIG. 22B shows an example of the relationship between the amplitude and the part activity in step S430. In this example, the maximum amplitude Amax at which the part PP does not pop out is detected as an amplitude that can prevent the part PP from popping out. The appropriate amplitude determined in this way is also used after step S440.

  In step S440, an appropriate number of parts in the parts feeder 400 is determined. In this process, for example, the number of parts PP that can be picked up is analyzed by analyzing the image acquired by the camera 430 after accommodating a large number of parts PP in the parts accommodating area 412 and performing a separation operation (FIG. 4B). Ask. The determination as to whether or not pick-up is possible is preferably performed ignoring the front and back of the parts PP. In this case, for example, it is determined that a part PP that does not overlap at all with another part PP can be picked up. This process is executed under the condition that the number of parts PP accommodated in the part accommodating area 412 is sequentially changed, and the number of parts when the number of parts that can be picked up becomes the maximum is determined as the appropriate number of parts in the parts feeder 400. To do.

  FIG. 22C shows an example of the relationship between the number of parts in the parts feeder 400 in step S440 and the number of parts detected as being pickable. In this example, the number of parts in the parts feeder 400 when the number of detected parts reaches a peak is determined as an appropriate number of parts in the parts feeder 400. The appropriate number of parts determined in this way is also used after step S450. Note that, when the part PP is held using the gripping mechanism 164 as in the third to fifth embodiments, as shown in FIG. 22C, a window in which the gripping part by the gripping mechanism 164 is taken around the part PP. A PW may be provided, and a part PP whose window PW does not overlap with the outer shape of another part PP may be recognized as a “part PP that can be picked up”. Alternatively, the additional area AD (FIG. 13A) described in the third embodiment may be used instead of the window PW.

  The appropriate number of parts can be determined using simulation instead of performing an experiment of actually putting parts into the parts feeder 400.

  FIG. 22D is an explanatory diagram illustrating a process for determining the number of parts by simulation. At this time, first, one part is imaged using the camera 430, and a part image Mp is cut out. Then, an image in which the cut-out part image Mp is randomly arranged in the region R412 having the same shape as the part housing region 412 is created by simulation. Then, by analyzing this simulation image, the number of parts that can be picked up is obtained. If this process is executed a plurality of times by changing the number of part images Mp in the region R412, the same characteristics as in FIG. 22C can be obtained by simulation. Then, the number of parts in the parts feeder 400 when the number of detected parts reaches a peak can be determined as an appropriate number of parts in the parts feeder 400. In this way, if an appropriate number of parts is determined using simulation, it is possible to omit the labor of performing the experiment.

  In step S450, the control parameter of the separation command is adjusted. In this adjustment process, for example, parts that can be picked up by storing a plurality of parts PP in the part storage area 412, performing a separation operation by a separation command (FIG. 4B), and then analyzing an image acquired by the camera 430. The number of PP is obtained. The determination as to whether or not pick-up is possible is preferably performed ignoring the front and back of the parts PP. The number of parts PP housed in the part housing area 412 is preferably set to an appropriate number of parts determined in step S440, for example. This process is performed under the condition that the duration of the separation operation is sequentially changed, and the value that the number of parts that can be picked up is sufficiently large and the duration of the separation operation is not excessively large is set as the duration of the separation operation. decide.

  FIG. 22E shows an example of the relationship between the duration of the separation operation in step S450 and the number of parts detected as being pickable. In this example, for reference, results obtained in the case where the number of parts in the parts feeder 400 is 130, 65, and 33 are shown. As can be understood from these examples, the number of detected parts PP that can be picked up increases as the duration of the separation operation increases, but saturates after reaching a certain duration. Therefore, the duration (the time indicated by white circles in FIG. 22E) in which the number of parts that can be picked up is sufficiently large and does not become excessively long can be determined as the duration of the separation operation. Note that the duration of this separation operation can be automatically determined as, for example, the time to reach a value obtained by multiplying the peak value of the number of parts that can be picked up by a predetermined coefficient K. The coefficient K is preferably set to a value less than 1 and not less than 0.9, for example.

  In step S460, the control parameter of the centering command for avoiding the interference area Rint is adjusted. As described with reference to FIG. 12A in the third embodiment, the interference region Rint is a region where the gripping mechanism 164 and the outer peripheral wall 414 interfere with each other in the outer peripheral portion of the part housing region 412. In this adjustment process, for example, after a plurality of parts PP are accommodated in the part accommodating area 412 and a separation operation (FIG. 4B) is performed, a centering operation for the interference area Rint (steps S225 and 12B in FIG. 11) is performed. The number of parts PP that can be picked up is obtained by executing and analyzing the image acquired by the camera 430. The determination as to whether or not pick-up is possible is preferably performed ignoring the front and back of the parts PP. The number of parts PP housed in the part housing area 412 is preferably set to an appropriate number of parts determined in step S440, for example. This processing is executed under the condition that the duration of the centering operation is sequentially changed, and the duration that the number of parts that can be picked up is sufficiently large and not excessively long is set as the duration of the centering operation for avoiding the interference area. decide.

  In step S470, the control parameter of the flip command is adjusted. In this adjustment process, for example, one part PP is accommodated in the part accommodating area 412, and after performing the flip operation (FIG. 4C) by the flip command, the part PP is turned over by analyzing the image acquired by the camera 430. It is determined whether or not. This process is executed under the condition that the duration of the flip operation is sequentially changed, and the duration that the part PP has a high probability of being flipped over and is not excessively long is determined as the duration of the flip operation.

  In step S480, the control parameter of the feed command is adjusted. In this adjustment process, for example, a plurality of parts PP are accommodated in the part accommodating area 412, and after performing a feed operation (FIG. 4A) by a feed command, the movement of the parts PP is analyzed by analyzing an image acquired by the camera 430. Find the amount. This process is executed under the condition that the duration of the feed operation is sequentially changed, and the time when the amount of movement of the parts PP is appropriate is determined as the duration of the feed operation. Alternatively, the duration of the feed operation may be determined by obtaining the moving speed [mm / sec] of the part PP. For example, a plurality of parts PP are accommodated in the part accommodating area 412, and a feed operation (FIG. 4A) by a feed command is performed for a certain time (for example, 1 second), and then an image acquired by the camera 430 is analyzed to thereby analyze the parts. The moving speed [mm / sec] of PP can be obtained. Then, the duration of the feed operation can be determined by multiplying the distance to be moved by this moving speed.

In step S490, conditions for supplying parts by the hopper 500 are determined. In this process, for example, the hopper 500 is operated for a certain period of time to supply the parts PP to the parts accommodating area 412, and the image acquired by the camera 430 is analyzed to obtain the number of parts supplied. This process is executed under the condition that the replenishment time of the hopper 500 is sequentially changed, and the time when the replenishment number of parts PP is appropriate is determined as the replenishment time of the hopper 500. Note that it is preferable to determine both the initial supply number in step S200 of FIG. 6 and the second and subsequent supply numbers in step S170 as the supply number of parts PP. As described above, when the parts storage area 412 is divided into N ₄₁₂ sections (N ₄₁₂ is an integer equal to or greater than 2), the second and subsequent replenishment quantity is 1 / (N ₄₁₂ -1) may be set. Alternatively, the parts replenishment condition may be determined by obtaining the supply speed [pcs / sec] of the parts PP. For example, the hopper 500 is operated for a certain time (for example, 1 second), the parts PP is supplied to the part storage area 412, and the number of parts supplied is obtained by analyzing the image acquired by the camera 430. Obtain the supply rate of parts PP [pcs / sec]. Then, the supply time of the hopper 500 can be obtained by dividing the supply speed by the number of parts to be supplied.

  Various control parameters set as described above are stored in the non-volatile memory 230 (FIG. 2) of the control device 200. The control command transmitted from the parts feeder control unit 212 to the parts feeder 400 is configured to include control parameters related to the plurality of vibration actuators 424 among the control parameters set in this way. In other words, the parts feeder control unit 212 selects one or more control commands from among a plurality of control commands each including control parameters of the plurality of vibration actuators 424, and transmits the selected control command to the parts feeder 400. This causes the parts feeder 400 to perform an operation according to the selected control command. In this way, control parameters suitable for the operation of the parts feeder 400 can be transmitted to the parts feeder 400. As a result, the parts feeder 400 can be appropriately operated according to the type and shape of the parts PP. Alternatively, the efficiency of the work of picking up the parts PP from the parts feeder 400 can be improved.

  The present invention is not limited to the above-described embodiments, examples, and modifications, and can be realized with various configurations without departing from the spirit thereof. For example, the technical features in the embodiments, examples, and modifications corresponding to the technical features in each embodiment described in the summary section of the invention are to solve some or all of the above-described problems, or In order to achieve part or all of the above effects, replacement or combination can be performed as appropriate. Further, if the technical feature is not described as essential in the present specification, it can be deleted as appropriate.

  DESCRIPTION OF SYMBOLS 100 ... Robot, 120 ... Base, 130 ... Arm, 132 ... Arm end, 160a-160c ... End effector, 162 ... Suction nozzle, 164a-164b ... Grasping mechanism, 166a-166b ... Vertical movement mechanism, 180 ... Camera, 200 ... Control device, 210 ... Processor, 211 ... Robot control unit, 212 ... Part feeder control unit, 213 ... Hopper control unit, 214 ... Image recognition unit, 215 ... Control parameter setting unit, 220 ... Main memory, 230 ... Non-volatile memory 231 ... Program command, 232 ... Control parameter, 233 ... Control command, 234 ... Part coordinate list, 235 ... Teaching data, 240 ... Display control unit, 250 ... Display unit, 260 ... I / O interface, 300 ... Teaching pendant, 400 ... Parts 410, parts housing section, 412 ... parts housing area, 414 ... outer peripheral wall, 420 ... vibration section, 422 ... control section, 424a to 424d ... vibration actuator, 430 ... camera, 500 ... hopper, 600 ... parts tray, 700 ... frame, 710 ... top plate, 720 ... table section

Claims (16)

A control device for controlling a robot system, comprising: a parts feeder having a parts accommodating part for accommodating parts and a plurality of vibration actuators for vibrating the parts accommodating part; and a robot having an end effector for picking up parts from the part accommodating part. Because
A parts feeder controller for controlling the parts feeder;
A robot controller for controlling the robot;
With
The parts feeder control unit selects one or more control commands from a plurality of control commands each including control parameters of the plurality of vibration actuators, and transmits the selected control command to the parts feeder. A control device that causes the parts feeder to perform an operation according to a selected control command. The control device according to claim 1,
The control device, wherein the plurality of control commands include a separation command for causing the parts feeder to perform a separation operation for separating a plurality of parts assembled in the part housing unit. The control device according to claim 2,
The plurality of control commands include an attitude change command for causing the parts feeder to execute an attitude change operation for changing an attitude of a part in the part accommodating portion.
Control device. The control device according to claim 3,
Each of the separation command and the posture change command is a command to vibrate the plurality of vibration actuators, and in the separation command, the vibration duration of the vibration actuator is set longer than the posture change command. apparatus. The control device according to any one of claims 1 to 4,
An image recognition unit that performs image recognition for recognizing a part in the part storage unit using an image acquired by a camera that captures an image of a part in the part storage unit,
The said parts feeder control part is a control apparatus which selects one or more control commands from among these control commands using the result of the said image recognition, and transmits to the said parts feeder. The control device according to claim 5,
The image recognizing unit virtually divides a part storage area of the part storage unit into a plurality of sections including a supply section that receives supply of parts from a parts supply device and a picking section in which the end effector picks up parts. Divided into
When it is recognized by the image recognition that there is a part in the picking section, the robot control unit controls the robot to pick up the recognized part by the end effector,
When it is recognized by the image recognition that there is no part in the picking section, the parts feeder control unit sends a feed command to the parts feeder to move a part from a section other than the picking section to the picking section. Control device to send. The control device according to claim 6,
When it is recognized by the image recognition that there are only parts that cannot be picked up in the picking section, the parts feeder control unit transmits a posture change command for changing the posture of the parts to the parts feeder. ,Control device. The control device according to claim 6 or 7,
The plurality of sections further include an intermediate section provided between the supply section and the picking section,
The feed command causes the parts feeder to execute an operation of moving parts existing in the replenishment section to the intermediate section and moving parts existing in the intermediate section to the picking section. The control device according to claim 5,
The parts accommodating portion has a parts accommodating area and an outer peripheral wall provided on an outer periphery of the parts accommodating area, and a gripping mechanism of the end effector and the outer peripheral wall are provided on an outer peripheral portion of the parts accommodating area. There is an interference area that interferes,
The part feeder control unit transmits a centering command for moving a part existing in the interference area toward the inside of the part receiving area to the parts feeder after separating parts by a separation command. The control device according to claim 9,
The image recognition unit
In the image acquired by the camera, a setting process for setting additional regions used for gripping the part by the gripping mechanism of the end effector at a plurality of locations on the outer edge of each part;
In the image, a recognition process for recognizing a part that the additional area does not overlap with another part as a grippable part,
Run
The said robot control part is a control apparatus which controls the said robot so that the said holdable part may be hold | gripped and picked up by the said holding mechanism of the said end effector. The control device according to claim 10,
The image recognition unit
An image update process for updating the image by erasing the grippable part from the image after the recognition process;
A process of repeating the recognition process and the image update process using the updated image;
And run
Register the order in which the individual parts are recognized as the grippable parts during the repetition of the recognition process and the image update process,
The said robot control part is a control apparatus which controls the said robot so that the said parts may be hold | gripped and picked up by the said holding mechanism of the said end effector according to the said order. The control device according to any one of claims 6 to 8,
The end effector has a first pickup mechanism and a second pickup mechanism,
The image recognition unit
A process of recognizing one part as a first pickable part that can be picked up by the first pickup mechanism among the parts existing in the picking section;
A process of recognizing a second pickable part that can be picked up by the second pickup mechanism in a state where the first pickable part is held by the first pickup mechanism;
To execute the control device. The control device according to claim 12,
In the process of recognizing the second pickable part, the image recognition unit can pick up one or more parts by the second pickup mechanism in a state where the first pickable part is held by the first pickup mechanism. A control device that calculates a pickup cost for each part according to a predetermined calculation method and selects the second pickable part according to the pickup cost. The control device according to any one of claims 1 to 13,
The control device, wherein the control parameter includes a frequency of a vibration signal supplied to the vibration actuator, an amplitude of the vibration signal, and a vibration duration time. 15. The control device according to claim 14, wherein
A non-volatile memory that pre-stores control parameters of the plurality of vibration actuators;
The control parameter stored in the nonvolatile memory is
(A) a balance of vibration intensity of each of the plurality of vibration actuators;
(B) the frequency of the vibration signal capable of activating the movement of the parts present in the part housing part;
(C) the amplitude of the vibration signal capable of preventing the parts present in the parts accommodating portion from jumping out of the parts accommodating portion;
Including a control device. A parts feeder having a parts accommodating part for accommodating parts and a plurality of vibration actuators for vibrating the parts accommodating part;
A robot having an end effector that picks up a part from the parts container;
The control device according to any one of claims 1 to 15, connected to the parts feeder and the robot,
A robot system comprising:

Images