JP2010214573A - Robot and its control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a productive robot capable of shortening time for part assembly and a control method thereof. <P>SOLUTION: In the case of gripping a first part 210 by a finger part 166 of a hand part 165, the gripping operation can be conducted while applying a perturbation torque by a perturbation actuator 164, the finger part 166 of the hand part 165 and the first part 210 are appropriately fit and thus it is possible to enhance gripping precision of the first part 210. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a robot including a force sensor that detects a force applied to an end effector such as a gripping part, a nut fastener, a welding gun, or a spray gun, and a control method thereof.

RCC (Remote Center Compliance) as a compliance element that absorbs the displacement in the parallel movement direction and the angular direction between the chuck part and the arm part that grip the part as a robot for automating the assembly work of the part, 2. Description of the Related Art An automatic component assembly apparatus that includes a force sensor that detects force and torque applied from a chuck portion is known (see, for example, Patent Document 1).
In the operation in which the robot inserts the first component, which is the object, into the second component, the dimensions and position of the component, the cycle time of the operation, and the like are input as commands to the robot control unit in advance. When the operation is started, the arm unit and hand unit of the robot move from the initial position in order to pick up the first component, the hand unit descends, and the hand unit grips the first component. Next, the hand part is raised while holding the first part, the arm part is moved to the insertion position of the first part, and the hand part is lowered until it comes into contact with the second part.

JP-A-7-241733 (pages 5-6, FIG. 1)

However, at this time, if the relative position between the first part and the second part is within the assumption, the insertion operation will be successful, but there will be a variation in the relative position between them for various reasons. As a result, the insertion operation may not be successful. Once such a failure occurs, humans will need to perform recovery work, and productivity will be significantly reduced.
FIG. 15 shows a diagram for explaining the state of the defect. (A) is a state before the first part 210 is inserted into the second part 220, and (b) is a state where the insertion operation is good and the first part 210 is successfully inserted into the second part 220. (C) is a state in which the position of the first part 210 and the second part 220 are slightly shifted, and the two parts are in contact with each other to cause a slip between the hand part 160 and the first part 210, (D) is a view showing a state in which the first component 210 is inserted into the second component 220 at a slight angle.

In the conventional robot operation with respect to the positional shift of (c), the position accuracy has been improved and the position feedback has been repeatedly performed. However, there is a limit to improving the positional accuracy for conveying parts and improving the positional accuracy for arranging parts on a pallet or the like. In addition, since the gripping position when the robot hand unit 160 grips the component also varies, there may be a difference between the center position of the assumed part and the center position of the actual part.
In the case of (d) being inclined, the force sensor signal is fed back if the robot's hand unit 160 is provided with a force sensor and a compliance element for absorbing the displacement and the angle shift. This makes it possible to perform insertion while correcting the inclined state. As a similar operation, the same control can be performed in the case of a press-fitting operation in which the gap between the first component 210 and the second component 220 is small.

  However, it takes time to determine whether it can be inserted due to misalignment or whether it can not be inserted at an angle, and it takes time to correct it. It is not high.

  The present invention has been made to solve the above-described problems, and can be realized as the following forms or application examples.

[Application Example 1]
A hand unit that grips an object, a perturbation actuator that vibrates the hand unit, a force sensor that detects a force applied to the hand unit, an arm unit that moves the hand unit, and an output of the force sensor And a control means for controlling the perturbation actuator simultaneously with the driving of the hand unit or the arm unit.

According to this application example, when the object is gripped by the hand unit, it is possible to perform a gripping operation while giving a perturbation torque by the perturbation actuator, and the hand unit and the target object are well adapted to each other. The gripping accuracy is improved.
Also, when aligning the target object, force applied to the hand part is detected by a force sensor with a compliance function while giving perturbation torque by the perturbation actuator, and the perturbation actuator is driven by the hand part or arm part. By simultaneously controlling the drive, the object can be easily aligned.
Accordingly, a robot with a short assembly time and improved productivity can be obtained.

[Application Example 2]
In the robot, the force sensor includes a first base having a magnetic circuit formed by a magnetic field generation unit having a yoke, a second base having a magnetic sensor crossing the magnetic circuit, and the first sensor. An elastic member interposed between the first base and the second base and elastically deformed by a load applied between them; and at least one of a position adjusting means of the magnetic sensor or a permeance adjusting means of the yoke. Robot characterized by having.
In this application example, since at least one of the magnetic sensor position adjusting means or the yoke permeance adjusting means constituting the magnetic circuit is provided, the relative position between the magnetic circuit and the magnetic sensor after assembly or at least one of the magnetic circuits is provided. Can be adjusted. When the first base and the second base are assembled through the elastic member, the relative position between the first base and the second base is shifted due to elastic deformation of the elastic member, etc. Even when the magnetic sensor is assembled with a relative position shifted, at least one of the position of the magnetic sensor in the magnetic circuit and the magnetic circuit can be adjusted. Therefore, a force sensor with little variation in output according to sensitivity and applied force can be obtained. Since the force applied to the hand unit and the perturbation actuator can be detected more accurately by such a force sensor, the positioning of the object can be performed more easily, the assembly time of parts can be shortened, and the robot with higher productivity can be obtained. can get.

[Application Example 3]
In the robot, the shape of the yoke is such that the magnetic circuit is formed such that the output of the magnetic sensor and the moving distance of the magnetic sensor moving in the magnetic circuit by the load are in a proportional relationship. A robot characterized by being.
In this application example, the shape of the yoke is formed so that the magnetic circuit has a proportional relationship between the output of the magnetic sensor and the moving distance of the magnetic sensor. Therefore, a force sensor having a constant output according to sensitivity and force can be obtained at any position in the magnetic circuit formed by the yoke. Since the force applied to the hand unit and the perturbation actuator can be detected more accurately by such a force sensor, the positioning of the object can be performed more easily, the assembly time of parts can be shortened, and the robot with higher productivity can be obtained. can get.

[Application Example 4]
The robot according to claim 1, wherein the magnetic field generator includes a permanent magnet.
In this application example, since a permanent magnet that does not require wiring is used for the magnetic field generator, a force sensor with a simple structure and a reduced manufacturing cost can be obtained. Since such a force sensor is provided, a robot with low productivity and improved productivity can be obtained.

[Application Example 5]
A hand unit that grips an object, a perturbation actuator that vibrates the hand unit, a force sensor that detects a force applied to the hand unit and the perturbation actuator, and an arm unit that moves the hand unit. A method for controlling a robot, comprising: determining a contact state of the object based on a drive signal of the perturbation actuator and an output signal of the force sensor; A robot control method characterized by performing drive control.

  According to this application example, the contact state of the object is determined based on the drive signal of the perturbation actuator and the output signal of the force sensor, and the gripping of the hand unit and the drive control of the arm unit are performed. The object can be easily aligned. Therefore, it is possible to obtain a robot control method in which parts assembly time is short and productivity is improved.

[Application Example 6]
The robot control method according to claim 1, wherein the drive control is performed by changing at least one of a drive frequency and an amplitude of the perturbation actuator.
In this application example, by changing at least one of the drive frequency or amplitude of the perturbation actuator, information for controlling the contact position, the stable point, etc. of the object increases due to the difference between the drive signal and the output signal. Therefore, it is possible to easily align the target object, shorten the time for assembling the parts, and obtain a robot control method with improved productivity.

The schematic perspective view of the robot in 1st Embodiment. The whole robot block diagram. (A) is a side view as an example of the hand mechanism part which has a force sensor, (b) and (c) are AA schematic sectional drawings in (a) which show the example of a finger | toe part. The schematic side view of the perturbation actuator using the eccentric gravity center weight. The schematic side view of the perturbation actuator using a piezoelectric element. Schematic of a force sensor. (A) is a schematic plan view, (b) is an AA schematic sectional drawing in (a). (A) is a schematic expanded sectional view near the magnetic field generation part of a force sensor and a magnetic sensor, (b) is a related figure of the output voltage of a magnetic sensor, and the distance from a magnetic field generation part. The circuit diagram in case a magnetic sensor is a Hall element. The operation | movement flowchart figure of a robot. In the modified example, (a) is a schematic enlarged cross-sectional view in the vicinity of the magnetic field generation unit and the magnetic sensor of the force sensor, and (b) is a relationship diagram between the output voltage of the magnetic sensor and the distance from the magnetic field generation unit. The figure which showed the compression characteristic of the elastic member. The schematic diagram of the force sensor in a 2nd embodiment. (A) is a schematic plan view, (b) is a schematic cross-sectional view taken along the line A-B-C-D-F in (a). The figure which showed the relationship between the center of an axial part, and the contact position of an object. The schematic diagram of the force sensor in a 3rd embodiment. (A) is a schematic plan view, (b) is a BB schematic sectional drawing in (a). (A) is a diagram showing a state before insertion, (b) is a diagram showing a state where the insertion operation is successful and parts are inserted successfully, and in (c), the positions of the first component and the second component are a little. The figure which shows the state which has slip | deviated and both have contacted and produced the slip between a hand part and a 1st component, In (d), a 1st component and a 2nd component are inserted slightly diagonally. FIG.

Hereinafter, embodiments will be described in detail with reference to the drawings.
(First embodiment)
FIG. 1 is a schematic perspective view of the robot 100 according to this embodiment, and FIG. 2 is an overall configuration diagram of the robot 100.

1 and 2, the robot 100 includes a command input unit 110, a calculation unit 120, a control unit 130 as a control means, a support unit 140, an arm unit 150, and a hand mechanism unit 160.
The arm unit 150 includes an actuator 151 and an arm 152. The hand mechanism unit 160 includes an actuator 161, a slide shaft 162, a force sensor 163, a perturbation actuator 164, and a hand unit 165. The hand unit 165 includes two finger units 166 as gripping units.

In FIG. 1, the support part 140 as a base is fixed to, for example, an assembling apparatus or a floor. The robot 100 is a horizontal articulated robot, and the arm 152 is moved in the horizontal direction by an actuator 151.
A hand mechanism unit 160 is provided at the tip of the arm unit 150. The slide shaft 162 of the hand mechanism unit 160 is moved in the vertical direction (vertical direction) by the actuator 161. Further, the slide shaft 162 may be rotated by the actuator 161.
A force sensor 163 is provided on the slide shaft 162, and a perturbation actuator 164 and a hand portion 165 are provided at the tip thereof. The hand unit 165 is located at the tip, and a perturbation actuator 164 is provided between the hand unit 165 and the force sensor 163.

For example, the robot 100 performs an operation of inserting the first component 210 into the second component 220. The second component 220 is positioned by the positioning 221 provided in the transport unit 230, and sequentially sent to the insertion position 240 where the insertion operation is performed.
FIG. 1 shows orthogonal coordinates composed of the X, Y, and Z axes, where the direction in which the second component 220 is conveyed by the conveyance unit 230 is the X axis and the movement direction of the slide shaft 162 is the Z axis. It was.
On the other hand, the arm unit 150 and the hand unit 165 of the robot 100 move in the horizontal direction in the XY plane, for example, in order to obtain the first component 210 arranged on the pallet 211, and the hand unit 165 The finger part 166 of the hand part 165 grips the first component 210 as it descends in the Z-axis direction. The robot 100 lifts the hand unit 165 in the Z-axis direction while holding the first component 210, moves the arm 152 in the horizontal direction to the insertion position 240 of the first component 210, and moves the hand unit 165 in the Z-axis direction. To lower.

The robot operation will be described in detail below.
In FIG. 2, the command input unit 110 inputs the position of the first component 210 and the second component 220 that are objects and robot operation commands from an operator or another system.
The calculation unit 120 determines the operation target values of the arm unit 150 and the hand unit 165 of the robot 100 based on a command or a signal obtained by the force sensor 163.
The control unit 130 performs servo control of the arm unit 150 and the hand mechanism unit 160 based on the operation target value.

Hereinafter, the operation of the calculation unit 120 will be described. The calculation unit 120 includes an operation target value calculation unit 121, an operation target value command unit 122, a comparison unit 123, and a contact position calculation unit 124.
Based on the command from the command input unit 110, the operation target value calculation unit 121 calculates the movement amount and angle of the arm unit 150 and the hand mechanism unit 160 to the object.
The operation target value command unit 122 outputs the calculated movement amount to the servo control units 131 and 132 of the arm unit 150 and the hand mechanism unit 160, respectively, and servo control of the arm unit 150 and the hand mechanism unit 160 is performed.
Further, the contact position calculation unit 124 calculates the contact position between the object and the hand unit 165 by a method described later based on the signal obtained by the force sensor 163. The comparison unit 123 determines whether or not the contact position matches the center of the hand unit 165 and inputs the result to the operation target value command unit 122. Based on the comparison result, the operation target value command unit 122 sends a new operation target value to the control unit 130 to cause the arm unit 150 and the hand mechanism unit 160 to move.

Next, the operation of the hand mechanism unit 160 of the robot 100 will be described in detail.
FIG. 3A shows a side view as an example of the hand mechanism unit 160 having the force sensor 163.
3B and 2C, two examples of the AA schematic cross-sectional view in FIG. The hand unit 165 and the finger unit 166 are controlled by the control unit 130 to move, hold an object, and control other operations.

In FIG. 3A, the two finger portions 166 are opened and closed in a direction approaching and separating from each other, and grip the first component 210 that is the object to be sandwiched from both sides.
FIG. 3B is a schematic cross-sectional view of the finger portion 166 when the first component 210 is cylindrical. The finger portion 166 is formed with a V-groove 167 for gripping the cylindrical portion, so that the first component 210 can be easily gripped vertically.

  FIG. 3C is a schematic cross-sectional view of the finger 166 when the first component 210 is rectangular. On the finger portion 166, convex portions 169 having tapered portions 168 are formed at four corners in order to grasp the rectangular portion.

A well-known actuator can be used as the perturbation actuator 164.
FIG. 4 shows a schematic side view of a perturbation actuator 300 as an example of the perturbation actuator 164.
In FIG. 4, a perturbation actuator 300 accommodates a rotary shaft 310, a cylindrical eccentric weight 320, a tooth wheel train 330, a first radial bearing 340, a second radial bearing 350, a thrust bearing 360, and a motor 370. A container 380.

The eccentric gravity center weight 320 and the gear 331 of the toothed wheel train 330 are attached to one side of the rotating shaft 310. The other side of the rotating shaft 310 is supported by a first radial bearing 340. Further, the side surface of the eccentric gravity center weight 320 is supported by the second radial bearing 350. Further, the bottom surface of the eccentric gravity center 320 is supported by a thrust bearing 360.
A gear 332 is attached to the motor 370 and meshes with the gear 331. The gear wheel 331 and the gear wheel 332 constitute a toothed wheel train 330. Here, the toothed wheel train 330 may be composed of three or more gears.

A part 321 having a different weight is provided in a part of the eccentric gravity center weight 320 so that the gravity center position is biased.
The perturbation actuator 300 causes vibration by rotating the eccentric gravity center weight 320 with the motor 370 via the toothed wheel train 330 and generates perturbation torque.
The frequency of vibration is determined by the number of rotations of the motor 370 and the toothed wheel train 330, and an appropriate value is selected according to the size and weight of the hand portion 165 shown in FIG. The driving voltage is the same.

  In FIG. 3, the hand portion 165 attached to the perturbation actuator 300 vibrates due to the perturbation torque generated by the perturbation actuator 300 (164), and the fitting portion between the first component 210 and the second component 220 also vibrates. When the center of the fitting is deviated, the signal is generated in the force sensor 163. If the signal is quite large, the position of the hand unit 165 is corrected by feedback. If the signal is within a range that can be absorbed by the compliance of the force sensor 163, the process can proceed to the next fitting (insertion) step.

FIG. 5 shows a schematic side view of a perturbation actuator 400 as an example of the perturbation actuator 164.
In FIG. 5, the perturbation actuator 400 includes a piezoelectric element 410, upper and lower plates 421 and 422, a drive circuit 420, and an elastic body 430.
The perturbation actuator 400 is caused to vibrate by driving the piezoelectric element 410 with the AC voltage of the drive circuit 420, and generates a perturbation torque in the same manner as the perturbation actuator 300.

  The upper and lower plates 421 and 422 are coupled by an elastic body 430, and the piezoelectric element 410 is disposed between the upper and lower plates 421 and 422. The piezoelectric element 410 is vibrated by a drive signal from the drive circuit 420. The piezoelectric element 410 has a larger amount of displacement and a greater force generated than the stacked piezoelectric element 410 than a single layer. The displacement of the stacked piezoelectric elements 410 increases as the number of stacked layers and the thickness of the stacked layers increase. In consideration of the size and weight of the hand portion 165, it is selected so that a sufficient amount of displacement can be obtained.

  FIG. 6 shows a schematic diagram of the force sensor 10 as an example of the force sensor 163 in the present embodiment. (A) is a schematic top view of the force sensor 10, (b) is an AA schematic sectional drawing in (a).

In FIG. 6, the force sensor 10 includes a shaft portion 1 as a first base, an elastic member 2, a housing 3 as a second base, four magnetic field generators 4, four magnetic sensors 5, and a magnetic sensor 5. Position adjusting means 50 is provided.
A ring-shaped convex portion 11 is formed in the middle of the shaft portion 1. The housing 3 has a ring shape, and a groove 31 is formed on the inner surface thereof. The convex portion 11 of the shaft portion 1 is fitted into the groove 31 of the housing 3 via the elastic member 2. Therefore, the elastic member 2 is interposed between the shaft portion 1 and the housing 3.
In the figure, the axial direction of the shaft portion 1 is shown as the Z axis, and the direction in which the force is detected is shown as the X axis direction and the Y axis direction.

In addition, thrust bearings 6A and 6B are disposed in the Z-axis direction between the convex portion 11 and the groove 31, and the shaft portion 1 is not deformed so that the shaft portion 1 moves in the Z-axis direction or the shaft portion 1 is twisted. It has become difficult.
Therefore, the force sensor 10 in this embodiment is a force sensor that does not detect a moment of force.

  On the outer periphery 111 of the convex portion 11 of the shaft portion 1, two recesses 12 are formed in the X-axis direction and two locations in the Y-axis direction with the center 1C of the shaft portion 1 interposed therebetween. A magnetic field generator 4 is provided. Further, a recess 32 is formed at the position of the groove 31 of the housing 3 facing the magnetic field generator 4, and the magnetic sensor 5 is disposed on the bottom surface 325 of the recess 32.

The shaft portion 1 and the housing 3 are less deformable than the elastic member 2 and are made of a nonmagnetic material. For example, as the metal, aluminum, nonmagnetic stainless steel, brass, zinc, magnesium, and other ceramics and hard resins can be used.
Examples of the elastic member 2 include silicon rubber, fluorine rubber, isoprene rubber, nitrile rubber, butyl rubber, chloroprene rubber, acrylic rubber, urethane rubber, ethylene propylene rubber, butadiene rubber, styrene butadiene rubber, natural rubber, and the like, and foam molding thereof. It is possible to appropriately select the one that can take a larger deformation amount according to the required load and deformation amount.

An electromagnet, a permanent magnet, or the like can be used for the magnetic field generating unit 4. However, a permanent magnet is preferable because wiring is not required, the structure is simple, and the size can be reduced.
Suitable permanent magnets include ferrite magnets, NdFeB magnets, SmCo magnets, and alnico magnets. The number of permanent magnets may be one or two.
As the magnetic sensor 5, a Hall element, a magnetoresistive element semiconductor type, a flux gate type, or the like can be used.

In the force sensor 10, when the shaft portion 1 receives a force in the X-axis direction or the Y-axis direction, the shaft portion 1 acts so as to deform the elastic member 2, and an output corresponding to the deformation amount is generated. It has become.
For example, as indicated by the white arrow shown in FIG. 6B, when a force Fx is applied in the + X-axis direction of the shaft portion 1, a force is also applied to the elastic member 2 in the + X-axis direction. The member 2 is deformed, and the shaft portion 1 approaches the magnetic sensor 5 in the + X direction. When the magnetic field generator 4 moves with the movement of the shaft 1, the magnetic field surrounding the magnetic sensor 5 also changes, and the output voltages of the four magnetic sensors 5 change.

Here, the magnetic sensor 5 in the + X axis direction is the magnetic sensor (+ X), the magnetic sensor 5 in the −X axis direction is the magnetic sensor (−X), the magnetic sensor 5 in the + Y axis direction is the magnetic sensor (+ Y), The magnetic sensor 5 in the −Y-axis direction is defined as a magnetic sensor (−Y).
For example, assuming that the output voltage before the force Fx is applied in the + X-axis direction is V ₀ , the output voltages V _{+ X} , V _-X , V _{+ Y} , V of the magnetic sensor 5 after the force Fx is applied in the + X-axis direction. _-Y changes as follows:
Magnetic sensor (+ X) output voltage V _{+ X} > V ₀
Magnetic sensor (−X) output voltage V _−X <V ₀
Magnetic sensor (+ Y) output voltage V _{+ Y} = V ₀
Magnetic sensor (−Y) output voltage V _−Y = V ₀
From the comparison of these output voltages, it is determined that the force sensor 10 has applied the force Fx in the + X-axis direction. The magnitude of the force Fx can be determined by measuring the magnitude of the output voltage V in advance according to the magnitude of the force, and knowing the output voltage V.

The same applies to the force Fy in the + Y-axis direction, and changes as follows.
Magnetic sensor (+ X) output voltage V _{+ X} = V ₀
Magnetic sensor (−X) output voltage V _−X = V ₀
Magnetic sensor (+ Y) output voltage V _{+ Y} > V ₀
Magnetic sensor (−Y) output voltage V _−Y <V ₀
An oblique force can also be calculated from the composition of the output voltages in the X-axis direction and the Y-axis direction.

FIG. 7A shows a schematic enlarged cross-sectional view of the vicinity of the magnetic field generator 4 and the magnetic sensor 5 of the force sensor 10.
FIG. 7B shows a relationship diagram between the output voltage of the magnetic sensor 5 and the relative movement distance with respect to the magnetic field generation unit 4 when the initial position L ₀ of the magnetic sensor 5 is set. Here, the direction toward the magnetic field generator 4 is represented as a positive direction. Hereinafter, the detection principle will be described with reference to the drawings.
In FIG. 7A, the magnetic field generation unit 4 and the magnetic sensor 5 are disposed to face each other with the elastic member 2 interposed therebetween. The magnetic field generation unit 4 includes two permanent magnets 41 and 42 and a yoke 43, and a magnetic circuit 45 is formed so that the magnetic flux 44 flows so as to spread toward the outer peripheral side of the convex portion 11 of the shaft portion 1 and returns again. Has been.

This magnetic flux 44 is detected by the magnetic sensor 5 and output as a voltage.
When the shaft 1 receives a force toward the magnetic sensor 5, the elastic member 2 is deformed, the distance between the magnetic field generator 4 and the magnetic sensor 5 is reduced, and the output of the magnetic sensor 5 is increased by increasing the magnetic flux 44. Become. As a result, force can be detected as voltage.
In FIG. 7B, when the relative movement distance from the initial position L ₀ of the magnetic sensor 5 to the magnetic field generation unit 4 increases (the distance between the magnetic field generation unit 4 and the magnetic sensor 5 decreases), The output voltage increases.

FIG. 8 shows a circuit example when the magnetic sensor 5 is a Hall element as a circuit diagram. OPAmp represents an operation amplifier, Tr represents a transistor, and R represents a resistance.
The circuit includes a constant current unit, an instrumentation amplifier unit, and a balance adjustment unit. As these circuits, well-known circuits can be used, and the circuit is not limited to the circuits shown in the drawings.

The Hall element measures the strength of the magnetic field, which is the density of the magnetic flux 44, according to the following principle.
In a p-type or n-type semiconductor sample, for example, a current is passed in the x direction and a magnetic field is applied in the y direction. At this time, charged particles (carriers) flowing through the semiconductor sample move in the z direction under the Lorentz force by the magnetic field. As a result, an electric field (Hall electric field) appears in a direction orthogonal to both the current and the magnetic field. This is the Hall effect, and the Hall element can measure the strength of the magnetic field using the Hall effect.
Examples of semiconductors used in the Hall element include gallium arsenide (GaAs), indium arsenic (InAs), and indium antimony (InSb). Indium antimony is highly sensitive but easily affected by temperature. Arsenic has characteristics such as being hardly affected by temperature, and can be used properly depending on the application.

In order to measure the magnetic field with high accuracy, it is better that the Hall current is constant, and the Hall current is made constant by the constant current portion. The Hall current can be determined by the resistance Rh.
A Hall voltage is generated at the output section of the Hall element, but since the voltage is small, it is amplified by the instrumentation amplifier section. The instrumentation amplifier unit has a single-ended output with a pair of differential input terminals and a reference terminal as a potential reference, high input impedance, and a common mode rejection ratio (CMR: COMMON MODE REJECTION) of 70 dB to 100 dB. Excellent to some extent.
The output voltage Vout of the instrumentation amplifier unit is expressed as shown in Expression (1) by the Hall voltage Vh and the gain of the instrumentation amplifier unit.

As described above, the Hall voltage Vh proportional to the magnetic field is obtained as the output. Further, when the zero point of the Hall voltage Vh amplified by the instrumentation amplifier unit is shifted, it can be adjusted by the balance adjusting unit.
By appropriately designing the material, dimensions, and shape of the permanent magnets 41 and 42, the gradient of the magnetic field gradient, which is a magnetic circuit in the space between the two permanent magnets 41 and 42, can be adjusted.
Further, the position of the magnetic sensor 5 can be moved with respect to the distribution of the magnetic flux 44 by the position adjusting means 50 of the magnetic sensor 5. As the position adjusting means 50, for example, the position of the magnetic sensor 5 can be mechanically adjusted in the X, Y, and Z axis directions using a screw, a micrometer, or the like.

Next, regarding the control method of the robot 100, the control flow will be described based on the operation flowchart of the robot 100 shown in FIG. 1, FIG. 2, and FIG.
1, 2, and 9, when the operation is started, the arithmetic unit 120 receives a command input (target setting) from the command input unit 110, analyzes the contents, and uses the operation amount up to the first component 210. A certain operation target value is calculated.
Next, the arm 152 is moved to the gripping position of the pallet 211 according to the determined operation target value in accordance with the gripping command of the first component 210. This operation is performed by the servo control unit 132 of the arm unit 150.

  Next, the hand unit 165 performs a gripping operation for the first component 210. This operation is performed by the servo control unit 131 of the hand mechanism unit 160. Specifically, the hand unit 165 is lowered and the first component 210 is gripped. At this time, gripping is performed while being perturbed by the perturbation actuator 164. The gripping by the hand unit 165 is controlled by changing at least one of the driving frequency or the amplitude of the perturbation actuator 164.

Next, according to the movement command to the insertion position 240 of the 1st component 210, the hand part 165 is raised, moved to the insertion position 240, and the hand part 165 is lowered. Also at this time, at least one of the drive frequency or amplitude of the perturbation actuator 164 is changed.
A force sensor 163 provided in the hand mechanism unit 160 outputs a sensor signal. The output sensor signal is input to the contact position calculation unit 124, and the contact position and angle between the first component 210 and the second component 220 are calculated.

  Next, the calculated contact position and angle value are input to the comparison unit 123, and it is determined whether or not the center position of the first component 210 and the center position of the second component 220 are the same. . Here, when the center position of the first component 210 matches the center position of the second component 220, the process proceeds to the insertion operation.

On the other hand, when the center position of the first component 210 and the center position of the second component 220 do not match, the correction value of the angle of the arm 152 for the alignment of the hand unit 165 is calculated and the correction is performed. The value is sent to the servo control unit 132 of the arm 152 to align the hand unit 165, and the actuator 151 is operated to correct the angle of the arm 152.
Thereafter, the moving operation of the hand unit 165 is also performed. At that time, the signal from the force sensor 163 is received, and the contact position calculation unit 124 again detects the contact position and angle between the first component 210 and the second component 220. Is calculated. If the result is good, the process proceeds to the next insertion operation. This routine is performed until the center position of the first part 210 matches the center position of the second part 220.

  A command for moving the hand unit 165 in the insertion direction is sent via the servo control unit 131 of the hand mechanism unit 160, and the actuator 161 of the hand mechanism unit 160 is operated. Also at that time, the signal of the force sensor 163 is received, the insertion position is calculated, and it is determined whether the insertion operation is properly performed.

  If the insertion operation has not been completed, the movement amount of the hand unit 165 is corrected, a movement amount correction command for the hand unit 165 is issued, and the movement operation of the hand unit 165 is performed. This routine is repeated until the insertion operation is completed.

When the insertion operation is finished, the grip is released, the arm 152 and the hand unit 165 are initialized, and the operation is finished.
Note that after the gripping operation is completed, operations such as movement and placement of the object to the target position are performed. Further, depending on the object to be grasped, the operation until the arm angle is moved may be repeated a plurality of times until the signal from the force sensor 163 is lessened.
The gripping by the hand unit 165 or the drive control of the arm 152 is controlled by changing at least one of the drive frequency or the amplitude of the perturbation actuator 164.

(Modification)
FIG. 10A shows a schematic enlarged cross-sectional view in the vicinity of the magnetic field generator 40 and the magnetic sensor 5 in the modification. FIG. 10B shows a relational diagram between the output voltage of the magnetic sensor 5 and the relative movement distance with respect to the magnetic field generating unit 40 when the initial position L ₀ of the magnetic sensor 5 is set, as in the first embodiment. The same components as those of the first embodiment are denoted by the same reference numerals as those of the first embodiment.

The magnetic field generator 40 includes a permanent magnet 49 and two yokes 46 and 47, and a magnetic circuit 48 is configured so that the magnetic flux 44 flows so as to spread toward the outer peripheral side of the shaft 1 and returns again.
In this case, the magnetic field gradient in the space between the two 46 and 47 can be designed to be relatively uniform, and the gradient of the magnetic field gradient can be adjusted depending on the shape of the yokes 46 and 47. For example, the inclination of the magnetic field gradient can be changed by tapering the yokes 46 and 47 so that the yokes 46 and 47 are thinned away from the permanent magnet and changing the shape of the taper.

In FIG. 10A, the magnetic sensor 5 is disposed in a substantially uniform magnetic field gradient in the space between the yokes 46 and 47.
Further, depending on the shape of the yokes 46 and 47, it is possible to change the shape of the elastic member 2 so that the yokes 46 and 47 do not contact the elastic member 2. In the modified example, the elastic member 2 is not disposed in a portion of the yokes 46 and 47 that protrudes from the convex portion 11 of the shaft portion 1.

  The yokes 46 and 47 are provided with permeance adjusting means 60 so that the gradient of the magnetic field gradient and the like can be adjusted. Similar to the position adjusting means 50, the permeance adjusting means 60 can mechanically adjust the positions of the yokes 46 and 47 in the X, Y, and Z axis directions using, for example, a screw, a micrometer, or the like.

When the magnetic field gradient in the space between the yokes 46 and 47 is uniform, as shown in FIG. 10B, the magnetic field generator 40 when the output voltage of the magnetic sensor 5 and the initial position L ₀ of the magnetic sensor 5 are set. The relative movement distance with respect to is proportional.

  In the first embodiment and the modification, as the yokes 43, 46, 47, iron-based materials having high saturation magnetic flux density, electromagnetic soft iron, low carbon steel, silicon steel plate, etc. are suitable. Rust plating or painting is applied. The shapes of the yokes 43, 46 and 47 are preferably such that the magnetic resistance is reduced so that the magnetic flux from the permanent magnet passes through the magnetic sensor 5 as much as possible. A suitable shape can be selected from the permeance of the permanent magnet, and the shapes of the yokes 43, 46, and 47 can be designed accordingly.

In FIG. 11, the example of the compression characteristic of the elastic member 2 was shown. The horizontal axis indicates the load, and the vertical axis indicates the amount of displacement.
In FIG. 11, the amount of displacement is not proportional to the load and becomes a gently upwardly convex curve, but the shape of this curve also changes depending on the type of elastic member 2 selected and the manner of foaming.
The output characteristics (load-output voltage) of the force sensor 10 are both the compression characteristics (load-displacement amount) of the elastic member 2 and the output characteristics of the magnetic sensor 5 shown in FIGS. 7B and 10B. It depends on.
In order to select the type of the elastic member 2, the elastic member 2 is selected from a comprehensive point of view such as compression characteristics, stability against repeated loads, temperature characteristics, etc. In particular, compression characteristics are important as an end effector of a robot. . First, the compression characteristics of the elastic member 2 are determined, the type and grade of the elastic member 2 are determined, and then the magnetic field gradients of the magnetic field generators 4 and 40 are designed so that the output characteristics of the magnetic sensor 5 have a proportional relationship. .
If the output characteristics of the magnetic sensor 5 are proportional, the load can be easily obtained from the output voltage, and no means for converting in consideration of the compression characteristics of the elastic member 2 is required.

  Further, it is sufficient that at least one of the position adjusting means 50 of the magnetic sensor 5 or the permeance adjusting means 60 of the yokes 46 and 47 is provided, and the magnetic field gradient is also provided by the position adjusting means 50 or the permeance adjusting means 60 of the yokes 46 and 47. Can be adjusted.

According to such this embodiment and modification, there exist the following effects.
(1) When the first part 210 is gripped by the finger part 166 of the hand part 165, the gripping operation can be performed while giving a perturbation torque by the perturbation actuator 164, and the finger part 166 of the hand part 165 and the first part 166 The part 210 becomes familiar and can improve the gripping accuracy of the first part 210.
As for a specific example, since the relative position between the pallet 211 that supports the first part 210 and the first part 210 is not so accurate, when the first part 210 is gripped, it does not necessarily fit in the V-groove 167. Therefore, when the first part 210 is gripped by the finger part 166, if the gripper is gripped by the perturbation actuator 164, the cylindrical first part 210 fits and grips the V groove 167 of the finger part 166. Can do.
When the perturbation actuator 164 is gripped while perturbing, the rectangular first component 210 can be gripped by being familiar with the tapered portion 168 of the finger portion 166.

Further, when the first component 210 and the second component 220 are aligned, a force sensor 163 having a compliance function is applied to the hand unit 165 and the perturbation actuator 164 while a perturbation torque is applied by the perturbation actuator 164. By detecting the force and drivingly controlling the perturbation actuator 164 simultaneously with the driving of the hand unit 165 and the arm unit 150, the first component 210 and the second component 220 can be easily aligned.
Therefore, it is possible to obtain the robot 100 with a short part assembly time and improved productivity.

  (2) Since at least one of the position adjusting means 50 of the magnetic sensor 5 or the permeance adjusting means 60 of the yokes 46 and 47 constituting the magnetic circuit 48 is provided, the magnetic circuits 45 and 48 and the magnetic sensor 5 Adjustment of the relative position or at least one of the magnetic circuits 45, 48 can be made possible. When the shaft portion 1 and the housing 3 are assembled via the elastic member 2, the relative positions of the shaft portion 1 and the housing 3 are shifted due to elastic deformation of the elastic member 2, etc. Even when the relative position with respect to the sensor 5 is shifted and assembled, the position of the magnetic sensor 5 in the magnetic circuits 45 and 48 or at least one of the magnetic circuits 48 can be adjusted. Therefore, it is possible to obtain the force sensor 10 with little variation in output according to sensitivity and applied force. Since the force applied to the hand unit 165 and the perturbation actuator 164 can be detected more accurately by such a force sensor 10, the first component 210 and the second component 220 can be more easily aligned, It is possible to obtain the robot 100 with a short part assembly time and higher productivity.

  (3) Since permanent magnets that do not require wiring are used for the magnetic field generators 4 and 40, the force sensor 10 having a simple structure and a reduced manufacturing cost can be obtained. Since such a force sensor 10 is provided, it is possible to obtain the robot 100 with low productivity and improved productivity.

  (4) Based on the drive signal of the perturbation actuator 164 and the output signal of the force sensor 10, the contact state between the first component 210 and the second component 220 is determined, and the gripping and arm of the hand portion 165 are determined. Since the drive control of the unit 150 is performed, the first component 210 and the second component 220 can be easily aligned. Therefore, it is possible to obtain a control method of the robot 100 with a short part assembly time and improved productivity.

  (5) By changing at least one of the drive frequency or amplitude of the perturbation actuator 164, the contact position, the stable point, etc. between the first component 210 and the second component 220 are controlled by the difference between the drive signal and the output signal. More information for Therefore, it is possible to easily align the first component 210 and the second component 220, and to obtain a control method of the robot 100 with a short time for assembling components and improved productivity.

According to the modification, the following effects are obtained.
(6) The shapes of the yokes 46 and 47 are formed so that the magnetic circuit 48 has a proportional relationship between the output of the magnetic sensor 5 and the moving distance of the magnetic sensor 5. Therefore, the force sensor 10 having a constant output according to sensitivity and force can be obtained at any position in the magnetic circuit 48 formed by the yokes 46 and 47. Since the force applied to the hand portion 165 and the perturbation actuator 164 can be detected more accurately by the force sensor 10 as described above, the first component 210 and the second component 220 can be more easily aligned with each other. The robot 100 can be obtained with a short time and improved productivity.

(Second Embodiment)
FIG. 12A shows a schematic plan view of the force sensor 20 as an example of the force sensor 163 in the present embodiment, and FIG. 12B shows an ABCDE in FIG. F schematic sectional view is shown.
In this embodiment, a large number of magnetic sensors 5 and magnetic field generators 4 are arranged, and the force sensor 20 is multi-axial. The same components are denoted by the same reference numerals, and only those necessary for explanation are labeled. Components not described have the same functions and structures as those of the first embodiment even if the sizes and shapes are different. Further, the position adjusting means 50 and the permeance adjusting means 60 of the magnetic sensor 5 are omitted.
In the present embodiment, twelve pairs of the magnetic sensor 5 and the magnetic field generator 4 are provided. Of these, eight are provided in the Z-axis direction.

The elastic member 21 is provided in a ring shape inside a pair of the eight magnetic sensors 5 and the magnetic field generator 4 provided in the Z-axis direction. Also, the thrust direction receivers 6A and 6B in the first embodiment are not provided. Therefore, in addition to the X-axis direction, the Y-axis direction, and the Z-axis direction, moments Mx, My, and Mz in the X, Y, and Z-axis directions can be measured.
As an example, the moment M is indicated by a hatched arrow in the figure. The moment M has an X component and a Y component.
Note that a through hole 14 is formed in the shaft portion 13 of the present embodiment.

FIG. 13 is a diagram showing the relationship between the center of the shaft portion 13 and the contact position of the object.
The origin 0 of the coordinate is the center 1C of the shaft portion 13, and N ′ (X1, Y1) is the contact position between the shaft portion 13 and the object. The combined moment M of the position N ′ of the object with the origin 0 as the center is expressed as M = (Mx ² + My ² ) ^(1/2) where Mx is the X component of the moment and My is the Y component of the moment.
If the distance from the origin 0 to the point N ′ is r, the moment M around the Z axis is expressed by the force F as M = F × r.
From the sign of the moments Mx and My, it can be seen in which region (one of the first to fourth quadrants on the XY plane) the point N ′ is. Further, from the ratio of the moments Mx and My, the angle θ is tan θ = My / Mx.
From the distance r, the coordinates X1 and Y1 of the point N ′ are X1 = r × cos θ and Y1 = r × sin θ, respectively.

According to this embodiment, there are the following effects.
(7) Not only the forces in the X-axis, Y-axis, and Z-axis directions but also the components Mx, My, and Mz of the moment M can be measured.

(Third embodiment)
FIG. 14 is a schematic plan view of a force sensor 70 as an example of the force sensor 163 in the present embodiment, and FIG. 14B is a schematic cross-sectional view taken along line BB in FIG. The same components are denoted by the same reference numerals, and only those necessary for explanation are labeled. Components not described have the same functions and structures as those of the first embodiment even if the sizes are different.
Also in this embodiment, the position adjusting means 50 and the permeance adjusting means 60 of the magnetic sensor 5 are omitted.

In FIG. 14, the force sensor 70 includes a shaft portion 71 as a first base and a base 73 as a second base.
A disc-shaped support plate 72 is provided at one end of the shaft portion 71. The base 73 has a cylindrical shape having a bottom 731 and includes a ring-shaped lid 74. The lid 74 has an opening 741 at the center. A circuit portion 75 is provided on the bottom portion 731 of the base 73, and a support portion 76 is provided thereon.
On the support plate 72, the magnetic field generators 4 </ b> A, 4 </ b> B, 4 </ b> C, 4 </ b> D are arranged so as to be rotationally symmetric with respect to the center 7 </ b> C of the shaft part 71. On the other hand, magnetic sensors 5A, 5B, 5C, 5D are arranged on the support portion 76 so as to face the magnetic field generating portions 4A, 4B, 4C, 4D.

  The support plate 72 of the shaft portion 71 is held by the elastic member 22 between the lid portion 74 and the support portion 76 so that the support plate 72 and the support portion 76 of the shaft portion 71 are parallel to each other. Therefore, the elastic member 22 is interposed between the shaft portion 71 and the base 73. The shaft portion 71 protrudes from the opening 741.

For example, when a force Fx in the + X direction indicated by a white arrow in the drawing is applied to the shaft portion 71, the shaft portion 71 is displaced in the + X direction with respect to the base 73. Then, the outputs of the four magnetic sensors 5A, 5B, 5C, and 5D deviate from the equilibrium point, and the output voltage changes.
Here, the direction of the magnetic field gradient of the four magnetic field generators 4A, 4B, 4C, and 4D is changed to change the sign of the output voltage so that the direction of the force can be seen on the XY plane. For example, the output voltage of each magnetic sensor is set as shown in Table 1.

  Here, with respect to the force in the first quadrant direction (+ X + Y), 5A is +, 5B is ˜0, 5C is −5, and 5D is ˜0, and the force in the second quadrant direction is (−X + Y). On the other hand, 5A is ˜0, 5B is -5, -5C is ˜0, and 5D is an output of +, and in contrast to the force in the third quadrant (+ X−Y), 5A is ˜0, 5B is + 5C Is 0 and 5D is an output of-, and 5A is -5B is 0, 5C is + 5D, and 5D is 0 output with respect to the force in the fourth quadrant (-XY).

According to this embodiment, there are the following effects.
(8) The direction and magnitude of the force applied on the XY plane can be known from the amount of change in the output voltage of the magnetic sensors 5A, 5B, 5C, 5D.

Various changes can be made in addition to the above-described embodiments and modifications.
The shape of the yokes 46 and 47 is not limited to the shape surrounded by the plane shown in the modified example, but may be a shape having a curved surface. For example, when designing the magnetic field gradient of the magnetic field generators 4 and 40 so that the output characteristics of the magnetic sensor 5 have a proportional relationship, the shape of the yokes 46 and 47 is changed to a curved surface according to the compression characteristics of the elastic member 2. It is good also as a shape with.

Further, the arrangement of the magnetic field generation unit and the magnetic sensor is not limited to being arranged at a symmetric position with respect to the origin, the X axis, the Y axis, and the Z axis, and may be arranged at an asymmetric position.
Further, the first base is not limited to the one having the cylindrical shaft portions 1, 13, 71. For example, a triangular prism or a quadrangular prism may be used.

  DESCRIPTION OF SYMBOLS 1,13,71,72 ... Shaft part and support plate as 1st base | substrate, 2, 21, 22 ... Elastic member, 3 ... Housing as 2nd base | substrate, 4, 4A, 4B, 4C, 4D, 40 ... Magnetic field generator, 5, 5A, 5B, 5C, 5D ... Magnetic sensor 10, 20, 70, 163 ... Force sensor, 41, 42, 49 ... Permanent magnet, 43, 46, 47 ... Yoke, 45, 48 DESCRIPTION OF SYMBOLS ... Magnetic circuit, 50 ... Position adjustment means, 60 ... Permeance adjustment means, 73 ... Base as 2nd base | substrate, 100 ... Robot, 130 ... Control part as control means, 150 ... Arm part, 164 ... Perturbation actuator, 165 ... hand part, 210 ... first part as object, 220 ... second part as object.

Claims (6)

A hand unit for gripping an object;
A perturbation actuator that vibrates the hand part;
A force sensor for detecting the force applied to the hand part;
An arm part for moving the hand part;
A robot comprising: control means for controlling driving of the perturbation actuator simultaneously with driving of the hand unit or the arm unit based on an output of the force sensor. The robot according to claim 1, wherein
The force sensor is
A first base body having a magnetic circuit formed by a magnetic field generation unit including a yoke;
A second substrate having a magnetic sensor across the magnetic circuit;
An elastic member interposed between the first base and the second base and elastically deformed by a load applied between the first base and the second base;
A robot comprising: at least one of a position adjusting unit of the magnetic sensor and a permeance adjusting unit of the yoke. The robot according to claim 2, wherein
The shape of the yoke is a shape that forms the magnetic circuit such that the output of the magnetic sensor and the moving distance of the magnetic sensor that moves in the magnetic circuit by the load are in a proportional relationship. robot. In the robot according to claim 2 or claim 3,
The magnetic field generating unit includes a permanent magnet. A hand unit that grips an object, a perturbation actuator that vibrates the hand unit, a force sensor that detects a force applied to the hand unit and the perturbation actuator, and an arm unit that moves the hand unit. A robot control method,
Based on the drive signal of the perturbation actuator and the output signal of the force sensor,
Judging the contact state of the object,
A robot control method comprising: gripping the hand unit or controlling driving of the arm unit. The robot control method according to claim 5, wherein
The robot control method, wherein the drive control is performed by changing at least one of a drive frequency or an amplitude of the perturbation actuator.

Images