JP2013240858A - Part assembling device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a failure condition on the way of work without using a force sensor or an external observer, in assembling parts by a robot.SOLUTION: An actuator control unit 4 includes: transfer means (a switch 44, a driver 45, and a stop detection unit 46) for transferring parts gripped by a gripper 1 so as to assemble a pair of parts; and a state determination unit 49 configured to determine success or failure in assembling parts from a relative speed based on relative positions detected by a position detector 3 during a period of transfer by the transfer means.

Description

  The present invention relates to a component assembling apparatus using, for example, a robot.

  In recent years, image inspection based on machine vision has been widely used in production sites such as assembling electronic components, which has contributed to a reduction in manufacturing defect rate.

JP 2011-230245 A

  In this machine vision image inspection, the work target is often hidden behind the end effector such as a hand, so it is difficult to detect defects during work in real time, and generally after the assembly process is completed. Done.

On the other hand, in the assembly of parts by human hands, the work state is visually monitored, and at the same time, the tactile sensation that is the sensation of the fingertips, the force sensation that is the force of the finger, and in some cases the auditory sense is also used for comprehensive judgment To do. Therefore, it is possible to notice a defect during the work, and the work can be redone immediately.
Even in the assembly of parts by a robot, the failure rate can be further reduced if a failure state during the operation can be detected.

  Here, as the component assembly, for example, a case is assumed in which a groove and a claw of a pair of components (first and second components) are fitted. In this case, as one of the methods for determining the success or failure of parts assembly by the robot immediately after the work is finished, a work failure such as a part colliding in the middle of work due to a position error or the like, or a large pushing force is required. In order to make a determination, a method of detecting a change in force during component assembly and comparing it with the case where the work is normally completed can be considered.

In order to detect a change in force during the operation as described above, for example, between a device that grips a component and a device that transports the gripping device (for example, a commonly used industrial robot). A force sensor may be installed. However, the force sensor is expensive and has a problem of being easily damaged.
In particular, when an industrial robot having a large mass and high rigidity is used as a device for transferring the gripping device, rapid acceleration / deceleration in a short time is difficult due to its inertia. For this reason, the operation speed during work must be reduced so as not to damage the force sensor and the components, and there is also a problem that the work takes time.

  In addition, the change in the frictional force during work can be estimated by a disturbance observer without using a force sensor. In this disturbance observer, estimation is performed by inputting the relative speed of the actuator movable portion with respect to the actuator fixing portion and the drive current of the actuator. However, the calculation of the disturbance observer is complicated, and there are problems that the load on the controller increases and that it is easily affected by noise.

On the other hand, a system has been disclosed that accurately determines whether or not work depends on various causes in component assembly work (see, for example, Patent Document 1). In the system disclosed in Patent Document 1, a test operation is repeatedly performed in advance by a robot arm equipped with a hand, and a feature amount based on the force sensor output and / or each position detection signal of the robot arm is obtained. The feature amount existence range at the time of work success and at the time of work failure is calculated, and the quality of work and the cause of failure are estimated by the existence range of the feature amount in the actual work.
However, the system disclosed in Patent Document 1 has a problem that it is difficult to speed up the work due to the mass and rigidity of the robot arm because the parts are assembled by the robot arm equipped with a hand. Furthermore, since the force sensor is used, there is a problem that it is expensive and easily damaged.

  The present invention has been made to solve the above-described problems, and is a component assembling apparatus capable of detecting a defective state during work by a simple method without using a force sensor or an external observer in assembling components by a robot. The purpose is to provide.

  The component assembly apparatus according to the present invention includes a gripping unit that grips one of a pair of components, an actuator that can move the actuator movable unit to which the gripping unit is attached, relative to the actuator fixing unit, The actuator includes a position detector that detects a relative position of the movable part of the actuator with respect to the actuator fixed part, and an actuator controller that controls the actuator based on the relative position detected by the position detector. Transfer means for transferring the parts gripped by the gripping means so as to assemble, and success / failure determination means for determining success / failure of parts assembly from a relative speed based on a relative position detected by the position detector during a transfer period by the transfer means; It is equipped with.

  According to this invention, since it comprised as mentioned above, in the assembly of parts by the robot, it is possible to detect a defective state during the work by a simple method without using a force sensor or an external observer.

It is a figure which shows the structure of the component assembly apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the structure of the actuator control part in Embodiment 1 of this invention. It is a figure which shows the equivalent model of the actuator in Embodiment 1 of this invention. It is a figure showing the form of the component assembly by the component assembly apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the contact state of the 1st component and 2nd component in Embodiment 1 of this invention. It is a figure which shows the operation timing of the actuator control part in Embodiment 1 of this invention. It is a figure which shows the equivalent model of the speed control of the actuator movable part in Embodiment 1 of this invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
Embodiment 1 FIG.
1 is a diagram showing a configuration of a component assembling apparatus according to Embodiment 1 of the present invention.
The component assembly device is a device that assembles a pair of components (first component and second component). As shown in FIG. 1, a gripper (gripping means) 1, an actuator 2, a position detector 3, and an actuator control unit. It is composed of four. In the following, the frame 5 is used as the first component and the case 6 is used as the second component, and the claw 61 provided on the inner wall of the case 6 is fitted into the groove 51 provided on the outer wall of the frame 5. The description will be made assuming that the work to be performed is performed. The frame 5 is gripped by gripping means (not shown) and its position is fixed.

The gripper 1 is attached to the tip of the actuator 2 (actuator movable portion 21), and grips one of the pair of parts (case 6 in FIG. 1).
The actuator 2 moves the entire gripper 1 attached to the tip of the actuator movable portion 21 by moving the actuator movable portion 21 relative to the actuator fixing portion 22 in a predetermined direction. The actuator 2 is attached to a robot arm such as a multi-joint arm (not shown).

The position detector 3 is attached to the actuator fixing portion 22 and detects the relative position of the actuator movable portion 21 with respect to the actuator fixing portion 22. A position signal indicating the relative position detected by the position detector 3 is supplied to the actuator controller 4.
The actuator control unit 4 controls the operation (movement speed and thrust) of the actuator movable unit 21 based on the position signal from the position detector 3 and determines whether the component assembly is successful.

Next, the configuration of the actuator control unit 4 will be described with reference to FIG.
As shown in FIG. 2, the actuator control unit 4 includes a position / speed converting unit 41, a reference speed changing unit 42, a first subtractor 43, a switch 44, a driver 45, a stop detecting unit 46, a position holding unit (storage means) ) 47, a second subtracter (positional difference calculating means) 48 and a state determining unit (success / failure determining means) 49.

  The position / speed conversion unit 41 differentiates the position signal from the position detector 3 and converts it into a speed signal. Thereby, the relative speed with respect to the actuator fixing | fixed part 22 of the actuator movable part 21 can be obtained. The velocity signal converted by the position / velocity conversion unit 41 is supplied to the first subtracter 43, the stop detection unit 46, and the state determination unit 49.

  The reference speed changing unit 42 changes the value (reference speed) of the reference speed signal supplied to the first subtracter 43 based on the position signal from the position detector 3. Here, the reference speed changing unit 42 switches the reference speed to a low speed when the value of the position signal exceeds a preset speed switching position reference.

The low speed means that when an object having a mass M including the gripper 1 attached to the tip of the actuator movable portion 21 is moving at a speed V, and the case 6 is brought into contact with the frame 5 at a predetermined position, the actuator The positional deviation amount determined by the relationship between the force generated by the momentum or kinetic energy of the movable portion 21 and the frictional force at the contact point is not large, and the contact state can be maintained.
The reason why the reference speed is switched to a low speed when the speed switching position reference is exceeded is to shorten the work time, and the speed switching position reference is set immediately before the contact position in consideration of the error of the absolute position.

In order to move the actuator movable unit 21 including the gripper 1 at a low speed, the actuator control unit 4 first converts the position signal from the position detector 3 into a speed signal by the position / velocity conversion unit 41, The subtracter 43 subtracts the value of the speed signal from the value of the reference speed signal. Then, the subtraction result (speed difference) is fed back to the actuator 2 via the ac contact of the switch 44 and the driver 45, so that the moving speed of the actuator movable portion 21 is controlled.
When the case 6 is brought into contact with the frame 5 at a predetermined location, the speed starts to decrease due to the influence of the frictional force at the contact location, but the thrust generated by the actuator 2 increases because the speed is fed back so as to be constant. . Then, when the maximum value of the thrust is smaller than the frictional force at the contact location, the speed decreases and stops.
The amount of movement from contact to stop, that is, the amount of displacement at the contact point, is smaller if the maximum generated thrust is sufficiently smaller than the friction force at the contact point, but the maximum generated thrust is a speed as described later. Since it is proportional to the reference speed signal of the control loop, the value of the reference speed signal is set so that the maximum generated thrust is sufficiently smaller than the frictional force at the contact point.

  The first subtractor 43 subtracts the value of the reference speed signal output from the reference speed converter 42 from the value of the speed signal output from the position / speed converter 41. A speed difference signal indicating the value subtracted by the first subtractor 43 is supplied to the switch 44.

  The switch 44 outputs the speed difference signal from the first subtractor 43 to the driver 45 until the stop detection signal from the stop detection unit 46 is received. After receiving the stop detection signal, the switch 44 has a predetermined constant value. The connection path is switched so that a thrust command signal indicating thrust is output to the driver 45.

  The driver 45 feeds back the speed difference signal input via the switch 44 to the actuator 2 via the ac contact of the switch 44 and the driver 45, thereby controlling the moving speed of the actuator movable portion 21. Or the drive voltage according to a thrust command signal is generated, and the actuator movable part 21 is moved with the thrust according to this drive voltage.

  The stop detection unit 46 detects whether the value of the speed signal from the position / speed conversion unit 41 is within a range centered on zero for a preset time. Here, when the stop detection unit 46 determines that the value of the speed signal is within a predetermined range for a predetermined time, the stop detection unit 46 determines that the actuator movable unit 21 has stopped and indicates that fact. A stop detection signal is supplied to the switch 44 and the position holding unit 47.

  The position holding unit 47 holds the position signal input from the position detector 3 at the timing when the stop detection signal is received from the stop detection unit 46. Thereby, the relative position of the actuator movable part 21 when the case 6 first contacts the frame 5 can be stored. Then, the position holding unit 47 supplies the held position signal to the second subtracter 48.

  The second subtracter 48 subtracts the value of the position signal output from the position detector 3 from the value of the position signal output from the position holding unit 47. The position difference signal indicating the value subtracted by the second subtracter 48 is supplied to the state determination unit 49.

  The state determination unit 49 estimates disturbance torque during operation (dynamic frictional force due to contact between the frame 5 and the case 6) based on the speed signal from the position / speed conversion unit 41, and determines the success or failure of component assembly. It is. Note that the position difference signal from the second subtractor 48 may also be considered.

Note that the switch 44, the driver 45, and the stop detection unit 46 correspond to the “transfer means for transferring the parts held by the holding means so as to assemble the pair of parts” of the present invention.
In addition, the position / speed converting unit 41, the reference speed changing unit 42, the first subtractor 43, the switch 44, and the driver 45 maintain the contact state when a pair of components are brought into contact at a predetermined location. It corresponds to a “contact means for bringing a pair of parts into contact at a predetermined location at a speed capable of being applied”.

Here, in assembly of parts, disturbance torque such as dynamic friction force during work can be detected in order to detect work failures such as parts colliding during work due to position errors or the need for a large pushing force. May be detected and compared with a normal value.
FIG. 3 is a diagram showing an equivalent model of the actuator 2 used for component assembly. Here, E (s) is the driving voltage, I (s) is the driving current, τ (s) is the disturbance torque, T (s) is the driving torque, V (s) is the moving part speed, and X (s) is moving. Part position, L is inductance, R is resistance value, Kt is torque constant, Ke is counter electromotive force constant, J is mass of movable part, and D is viscous resistance.

In the equivalent model shown in FIG. 3, it is generally known that the disturbance torque τ (s) can be estimated by using a disturbance observer that receives the drive current I (s) and the moving part speed V (s). ing.
However, the calculation of the disturbance observer is complicated, and it is necessary to consider that the load on the controller increases.

Here, as described above, the disturbance observer inputs are the drive current I (S) and the moving part speed V (s). If the drive voltage E (s) is predetermined such as a constant value, the drive current I (S) is only affected by the moving part speed V (s). Therefore, the influence of disturbance torque τ (s) such as dynamic friction force during work appears as a direct speed change.
Therefore, in the present invention, the working state is estimated without performing complicated calculations by monitoring the change in the relative speed of the actuator movable portion 21 during the work.

Next, the component assembly / success / failure determination operation by the component assembly apparatus configured as described above will be described with reference to FIGS. In FIG. 5, only the vicinity of the groove 51 and the vicinity of the claw 61 that are contact portions are extracted from the frame 5 and the case 6. Further, the contact state (A) in FIG. 6 corresponds to the contact state in FIG.
In the component assembly by the component assembly apparatus of the present embodiment, as shown in FIG. 4, if the case 6 is slid with respect to the frame 5 and the claw 61 of the case 6 is fitted in the groove 51 of the frame 5. Work is successful.

In the component assembling apparatus of the present embodiment, the position detector 3 detects the relative position of the actuator movable portion 21 with respect to the actuator fixing portion 22 while performing the work, and a position signal indicating the relative position is sent to the actuator control portion 4. To supply.
In the actuator control unit 4, the position signal from the position detector 3 is first converted into a speed signal by the position / speed conversion unit 41, and the value of the speed signal is converted to the value of the reference speed signal by the first subtracter 43. Subtract with Then, the subtraction result (speed difference) is fed back to the actuator 2 via the ac contact of the switch 44 and the driver 45, thereby controlling the moving speed of the actuator movable portion 21.

Further, the reference speed conversion unit 42 switches the reference speed to a low speed when the value of the position signal from the position detector 3 exceeds the speed switching position reference.
For example, as shown in the current position (C) in FIG. 6, the speed switching position reference is set a few millimeters before the position where the case 6 contacts the frame 5 (stop position), and is shown as the reference speed (D). In addition, the reference speed is changed to a low speed with this speed switching position reference as a boundary. As a result, as shown at time t2 of the current speed (B), the speed rapidly decreases before the case 6 contacts the frame 5. In this state, when the case 6 approaches the frame 5 and the case 6 comes into contact with the frame 5 (contact state A in FIG. 5), the speed rapidly decreases, and the current speed (B) in FIG. It stops as shown at time t3 in C). At this time, since the thrust generated in the actuator movable portion 21 is proportional to the value of the reference speed (low speed) of the speed control, the case 6 is in contact with the frame 5 with a minute force and keeps its state.

Here, the relationship between the reference speed serving as a reference for the speed of the actuator movable portion 21 and the thrust generated when the contact is stopped will be described using an equivalent model of speed control in FIG.
In FIG. 7, the thrust (force applied to the case 6) F2 generated when the actuator movable portion 21 is moving at the speed V is expressed by the following equation (1).
F2 = Kt · Ki · (Kv · ΔV−KeV)
= Kt · Ki · (Kv · (Vref−Ve) −KeV) (1)
When the case 6 is in contact with the frame 5 and stopped (speed V = 0), the actual speed voltage is Ve = 0 and the actual speed V = 0, and the thrust at that time is expressed by the following equation (2). It is proportional to Vref.
F2 = Kt · Ki · Kv · Vref (2)
At this time, the frame 5 receives the reaction force F1 (= F2).
Therefore, if the reference speed is set to a sufficiently low speed, the case 6 can be brought into contact with the frame 5 with a minute force and kept in that state.

  Moreover, even if the inertia of the actuator movable part 21 including the gripper 1 is large by bringing the case 6 into contact with the frame 5 in a state where the speed is sufficiently reduced, the influence is hardly received. Therefore, the frame 5 and the case 6 are not damaged or deformed.

In this state, the stop detection unit 46 detects that the relative speed of the actuator movable unit 21 is within a predetermined range centered on zero for a preset time, and sends a stop detection signal to the switch 44. And supplied to the position holding unit 47.
The position holding unit 47 holds the position signal input from the position detector 3 and supplies it to the second subtracter 48 at the timing when the stop detection signal is received. The second subtracter 48 subtracts the position signal value of the position holding unit 47 from the position signal value from the position detector 3 and outputs a position difference signal. Immediately after the stop detection signal is output, the position difference signal is zero because the frame 5 and the case 6 are in contact with each other with a minute force and stopped.

On the other hand, when the switch 44 receives a stop detection signal from the stop detector 46, the switch 44 switches the connection path from the ac contact to the bc contact. Then, after a predetermined time has elapsed, a thrust required to fit the claw 61 of the case 6 into the groove 51 of the frame 5 is applied to the actuator 2 via the bc contact and the driver 45.
And in this invention, in order to determine whether the nail | claw 61 of the case 6 was engage | inserted in the groove | channel 51 of the flame | frame 5, the speed change of the actuator movable part 21 after a thrust is applied is monitored.

First, when a thrust is applied to the actuator 2, the claw 61 on the inner wall of the case 6 moves while contacting the outer wall of the frame 5, and contacts the wall 52 provided in front of the groove 51 on the outer wall of the frame 5 (see FIG. 5). Contact state B).
Thereafter, the case 6 moves further and gets over the wall 52. At this time, the case 6 is deformed so as to be spread outward, and the dynamic frictional force due to the contact increases with the movement, so that the speed decreases as shown in the current speed (B) of FIG.

Thereafter, from the state where the claw 61 reaches the apex of the wall 52 (contact state C in FIG. 5) to just before it starts to fall into the groove 51 (contact state D in FIG. 5), it can be estimated that the speed is substantially constant.
Thereafter, when the claw 61 further moves from the state (contact state D in FIG. 5) where the claw 61 is approaching the groove 51 and fits into the groove 51 (contact state E in FIG. 5), the case 6 that has been spread outward is the original. Since it returns to the state and the dynamic friction force also decreases, the speed increases again as shown at times t5 to t6 of the current speed (B) in FIG.

Then, when the claw 61 is further moved after being fitted in the groove 51 (contact state F in FIG. 5) and reaches the stopper position (contact state G in FIG. 5), the dynamic friction force increases rapidly, so that the current state of FIG. As shown at time t6 to t7 of speed (B), the speed decreases and stops.
Then, in the state determination unit 49, the speed change in the period (time t4 to t7 in FIG. 6) from when the thrust required to fit the claw 61 of the case 6 into the groove 51 of the frame 5 is started is stopped. Monitor. After that, the peak value and bottom value of the speed change during the period are detected, a threshold value is set between the peak value and the bottom value, and the state determination output is increased when a speed exceeding the threshold value is detected. To level.

Here, when the parts are assembled successfully, the claws 61 of the case 6 in the contact states D and E approach the groove 51 of the frame 5 in addition to the peak values that occur until the contact states A and B in FIG. Even when it fits into the groove 51 from the above state, the speed increases and exceeds the threshold value. Therefore, as shown in the state determination (G) in FIG. 6, the state determination output is output twice.
On the other hand, in the contact state B of FIG. 5, when the claw 61 of the case 6 contacts the wall 52 provided in front of the groove 51 of the frame 5 and stops without getting over the wall 52, the current state of FIG. As indicated by the dotted line portion of the speed (B), the speed decreases to zero. Therefore, as shown in the state determination (G) of FIG. 6, the state determination output is output only once.
Therefore, it is possible to determine the success or failure of component assembly by monitoring the speed change during the period.

  It should be noted that the position difference signal from the second subtractor 48 may be taken into consideration in the success / failure determination of component assembly by the state determination unit 49. That is, the value of the position difference signal when the speed exceeds the threshold value is detected during the period from when the thrust necessary for fitting the claw 61 of the case 6 into the groove 51 of the frame 5 is started until it stops. . Thereby, also in subsequent success / failure determination, since it is estimated that the speed exceeds the threshold value at the same position, the certainty for the success / failure determination can be improved.

  As described above, according to the first embodiment, during the assembly of the parts, the change in the speed of the actuator movable portion 21 is monitored to estimate the dynamic friction force during the operation. Therefore, the force sensor and the disturbance observer are used. It is possible to estimate the work state by a simple method without using.

  In the first embodiment, since the speed change is monitored after the case 6 is brought into contact with the frame 5 with a small force at a low speed, the position where the speed change should be monitored is accurately determined. It is possible to increase the certainty for success / failure determination.

  In the present invention, within the scope of the invention, any combination of the embodiments, or any modification of any component in each embodiment, or omission of any component in each embodiment is possible. .

1 Gripper 2 Actuator 3 Position detector 4 Actuator controller 5 Frame (first part)
6 Case (second part)
21 Actuator movable part 22 Actuator fixing part 41 Position / speed converting part 42 Reference speed changing part 43 First subtractor 44 Switch 45 Driver 46 Stop detecting part 47 Position holding part 48 Second subtractor 49 State determining part 51 Groove 52 Wall 61 nails

Claims (3)

In a component assembly device that assembles a pair of components,
Gripping means for gripping one of the pair of parts;
An actuator capable of moving the actuator movable part to which the gripping means is attached relative to the actuator fixing part;
A position detector for detecting a relative position of the actuator movable portion with respect to the actuator fixing portion;
An actuator controller for controlling the actuator based on the relative position detected by the position detector;
The actuator controller is
Transfer means for transferring the parts gripped by the gripping means to assemble the pair of parts;
A component assembly apparatus comprising: success / failure determination means for determining success / failure of component assembly from a relative speed based on a relative position detected by the position detector during a transfer period by the transfer means. The actuator controller is
Contact means for contacting the pair of components at a predetermined location at a speed capable of maintaining the contact state when the pair of components are contacted at a predetermined location;
The component assembling apparatus according to claim 1, wherein the transfer unit performs transfer from a state in which the pair of components are in contact with each other by the contact unit. Storage means for storing a relative position detected by the position detector when the pair of components are contacted by the contact means;
A position difference calculation means for calculating a position difference between the relative position detected by the position detector during the transfer period by the transfer means and the relative position stored in the storage means;
The component assembly apparatus according to claim 2, wherein the success / failure determination unit determines the success / failure of component assembly in consideration of the position difference detected by the position difference calculation unit.

Images