Robot, remote teaching system, teaching robot, control method, and program

Abstract

To provide a robot, a remote teaching system, a teaching robot, a control method, and a program that prevent an excessive force from being applied during teaching.SOLUTION: And a control unit configured to, when it is determined that the workpiece and the object have come into contact with each other based on the force detected by the detection unit, execute impedance control of the robot arm in which a position of the workpiece at the time of contact is set as a neutral point of a spring in a control model of a spring mass damper system.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present disclosure relates to a robot, a remote teaching system, a teaching robot, a control method, and a program. [Background technology]

[0002] Robots are used in a variety of work sites. In some cases, operators directly operate the robot and teach it how to move. This includes teaching it to follow a certain motion. A follow-up motion is a motion in which the robot moves while coming into contact with an object. This follow-up motion requires careful consideration of safety and stability, as it involves contact with the object.

[0003] For example, Patent Document 1 discloses a control device that switches modes during teaching and causes a robot to memorize its movements. This technology requires that mode changes be specified in advance, and does not address cases where teaching is performed to an unknown object by directly operating the robot. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2017-164876 Summary of the Invention [Problem to be solved by the invention]

[0005] Furthermore, when teaching the above-mentioned tracing motion, the operator may not be able to feel the force being applied to the workpiece due to the robot's inertia and the influence of the high gear ratio joints, and the operator may unintentionally apply excessive force.

[0006] Non-limiting examples of the present disclosure contribute to providing a robot, a remote teaching system, a teaching robot, a control method, and a program that suppress the application of excessive force during teaching. [Means for solving the problem]

[0007] A robot according to one embodiment of the present disclosure includes a detection unit that detects a force from an object acting on a workpiece set on a robot arm, and a control unit that, when it determines that the workpiece and the object have come into contact based on the force detected by the detection unit, performs impedance control of the robot arm such that the position of the workpiece at the time of contact is the neutral point of the spring in a control model of a spring-mass-damper system.

[0008] A remote teaching system according to one embodiment of the present disclosure is a remote teaching system including a teaching robot to which an operator gives instructions, and a taught robot that operates based on a target position notified by the teaching robot and has a workpiece set on its robot arm, wherein the taught robot has a detection unit that detects a force from an object acting on the workpiece, and a taught robot control unit that sends a contact notification to the teaching robot when it determines that the workpiece and the object have come into contact based on the force detected by the detection unit, and the teaching robot has a teaching robot control unit that, upon receiving the contact notification, performs impedance control such that the position of the tip of the robot arm of the teaching robot at the time of contact is the neutral point of the spring in a control model of a spring-mass-damper system.

[0009] A teaching robot according to one embodiment of the present disclosure is a teaching robot that is taught by an operator, and is equipped with a detection unit that detects contact between a workpiece set on the robot arm of a taught robot that operates in synchronization with the teaching robot and an object, and a teaching robot control unit that, when the detection unit detects the contact, performs impedance control such that the position of the tip of the robot arm of the teaching robot is the neutral point of the spring in a control model of a spring-mass-damper system.

[0010] A control method according to one embodiment of the present disclosure is a robot control method that detects a force from an object acting on a workpiece set on a robot arm, and when it is determined that the workpiece and the object have come into contact based on the detected force, performs impedance control of the robot arm, with the position of the workpiece at the time of contact set as the neutral point of the spring in a control model of a spring-mass-damper system.

[0011] A program according to one embodiment of the present disclosure causes a computer to detect a force from an object acting on a workpiece set on a robot arm, and when it determines that the workpiece and the object have come into contact based on the detected force, executes impedance control of the robot arm, with the position of the workpiece at the time of contact set as the neutral point of the spring in a control model of a spring-mass-damper system.

[0012] These comprehensive or specific aspects may be realized as a system, an apparatus, a method, an integrated circuit, a computer program, or a recording medium, or may be realized as any combination of a system, an apparatus, a method, an integrated circuit, a computer program, and a recording medium. [Effects of the Invention]

[0013] Non-limiting examples of the present disclosure prevent excessive force from being applied during teaching.

[0014] Further advantages and benefits of an embodiment of the present disclosure will become apparent from the specification and drawings. Such advantages and / or benefits may be provided by some of the embodiments and features described in the specification and drawings, respectively, but not necessarily all of them may be provided to obtain one or more identical features. [Brief explanation of the drawings]

[0015] [Figure 1] A diagram showing an example of a robot [Figure 2A] A diagram showing an example of a peg fitting operation [Figure 2B] A diagram showing an example of a peg fitting operation [Figure 2C] A diagram showing an example of a peg fitting operation [Figure 2D] A diagram showing an example of a peg fitting operation [Figure 2E] A diagram showing an example of a peg fitting operation [Figure 3] Block diagram showing the configuration for performing impedance control [Figure 4] Flowchart showing the control content of the restriction determination unit [Figure 5] A diagram showing an example of the configuration of a remote teaching system. [Figure 6] Sequence diagram showing the processing flow in the remote teaching system DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functions are designated by the same reference numerals, and redundant description will be omitted.

[0017] First Embodiment The first embodiment is an embodiment in which an operation of fitting a workpiece (peg) into an object (insertion target object) (hereinafter referred to as "peg fitting operation") is used as an example of an operation to be taught. FIG. 1 is a diagram showing an example of a robot 100 according to an embodiment of the present invention. The robot 100 includes a robot control unit 50, a first force sensor 110, a second force sensor 120, and a handle 130. A peg 200 is set on the robot arm of the robot 100. The insertion target object 300 has a hole in which the peg 200 is fitted.

[0018] The robot control unit 50 is composed of a central processing unit (CPU), storage devices such as read-only memory (ROM) and random access memory (RAM), communication devices, etc., and controls the entire robot 100. The handle 130 is an operation unit operated by the operator 10 during the peg fitting operation. The first force sensor 110 is a force sensor provided on the robot arm, which detects a force generated by the operation of the operator 10 and outputs the detected force to the robot control unit 50. The first force sensor 110 is an example of another detection unit. The second force sensor 120 detects a force acting on the peg 200 and outputs the detected force to the robot control unit 50. The second force sensor 120 is an example of a detection unit. An example of a force acting on the peg 200 is a force generated when the peg 200 comes into contact with the insertion target object 300. In addition to these sensors, torque sensors (not shown) provided at each joint of the robot arm serve as other detection units for detecting forces acting in response to instruction. The torque detected by this torque sensor is recorded as a teaching operation.

[0019] 2A to 2E are diagrams showing an example of a peg fitting operation. During the peg fitting operation, the peg 200 is set in the robot 100, but the robot 100 is omitted from FIGS. 2A to 2E. For ease of explanation, only a cross section of the hole into which the peg 200 is fitted is shown for the insertion target object 300 in FIGS. 2A to 2E.

[0020] Fig. 2A shows the initial position of the peg 200 in this operation example. Fig. 2B shows the peg 200 that has been moved close to the hole. Fig. 2C shows the peg 200 that has been brought into contact with the insertion target object 300. Fig. 2D shows the peg 200 that has been moved to a position where it can be inserted into the hole. Fig. 2E shows the peg 200 that has been fitted into the hole. In the series of operations shown in Fig. 2, the force generated by the operation of the operator 10 is detected by the first force sensor 110.

[0021] As shown in Fig. 2A, the operator 10 moves the peg 200 from the initial position to the vicinity of the hole. When the operator 10 moves the peg 200 to the vicinity of the hole as shown in Fig. 2B, the operator 10 lowers the peg 200 and brings the tip of the peg 200 into contact with the insertion target object 300 as shown in Fig. 2C. At this time, a force acts on the peg 200, and the second force sensor 120 detects this force.

[0022] After contacting the peg 200, the operator 10 moves the peg 200 toward the hole. This action is an example of a tracing action. Since the insertion target object 300 and the peg 200 remain in contact while the peg 200 is moving, the second force sensor 120 detects a force. When the peg 200 moves directly above the hole through the operation of the operator 10, as shown in FIG. 2D, the peg 200 and the insertion target object 300 are no longer in contact. Therefore, no force acts on the peg 200, and no force is detected by the second force sensor 120. As described above, a force is detected by the second force sensor 120 from the time the insertion target object 300 and the peg 200 come into contact as shown in FIG. 2C until the peg 200 moves directly above the hole as shown in FIG. 2D.

[0023] When the peg 200 moves to a position directly above the hole, the operator 10 fits the peg 200 into the hole. This completes the teaching. In this manner, the operator 10 performs a series of peg fitting operations, such as moving the peg 200 close to the hole, bringing the peg 200 into contact with the insertion target object 300, moving the peg 200 to a position directly above the hole, and fitting the peg 200.

[0024] 3 is a block diagram showing a configuration for executing impedance control (admittance control) according to this embodiment. The robot control unit 50 includes an impedance control unit 400 and a restriction determination unit 500 as components of the control system.

[0025] The impedance control unit 400 applies a control model of a spring-mass-damper system. In Fig. 3, F1 indicates the force detected by the first force sensor 110. F2 indicates the force detected by the second force sensor 120. K indicates the elastic coefficient of the spring. D indicates the viscosity coefficient of the damper. M indicates the mass.

[0026] The impedance control unit 400 calculates a target position rd that indicates the behavior of the robot 100 in response to the force F1 as if it were an object having a virtual mass connected to a spring and a damper. In this embodiment, the neutral point of the spring is used as one of the target positions, and this neutral point is set to the position of the workpiece when it is determined that the workpiece and the object have come into contact. Note that, in order for the robot 100 to move in the direction of the force applied by the operator 10 to the handle 130, if there is a restoring force of the spring, the robot 100 will return to the neutral point, so the elastic coefficient K under normal conditions is set to 0.

[0027] Impedance control unit 400 includes subtractors 410, 420, 460, proportional controllers 430, 470, 480, and integrators 440, 450 as components for calculating rd. Force F1 is input to impedance control unit 400 from first force sensor 110, and elastic coefficient K and neutral point r0 are input to impedance control unit 400 from limit determination unit 500. Note that while the second and first derivatives of rd output from proportional controller 430 and integrator 440 are both expressed using dot notation in FIG. 3, they will be expressed using a prime symbol (') in the following explanation.

[0028] The force F1 input from the first force sensor 110 is input to the subtractor 410. In the subtractor 410, the output of the proportional controller 480 is subtracted from the force F1. To explain the output of the proportional controller 480, first, the output rd of the integrator 450 is subtracted by r0 in the subtractor 460. The output rd-r0 of the subtractor 460 is multiplied by K in the proportional controller 480. The proportional controller 480 outputs K(rd-r0) to the subtractor 410. As a result, the subtractor 410 outputs F1-K(rd-r0) to the subtractor 420.

[0029] F1-K(rd-r0) input from subtractor 410 is input to subtractor 420. In subtractor 420, the output of proportional controller 470 is subtracted from F1-K(rd-r0). To explain the output of proportional controller 470, first, output rd' of integrator 440 is multiplied by D in proportional controller 470. Proportional controller 470 outputs Drd' to subtractor 420. As a result, subtractor 420 outputs F1-K(rd-r0)-Drd' to proportional controller 430.

[0030] The output F1-K(rd-r0)-Drd' of the subtractor 420 is multiplied by 1 / M in the proportional controller 430. The proportional controller 430 outputs rd''=(F1-K(rd-r0)-Drd') / M to the integrator 440. In the integrator 440, rd'' is integrated, and rd' is output to the integrator 450 and the proportional controller 470. In the integrator 450, rd' is integrated, and rd is output to the subtractor 460 and the limit determination unit 500.

[0031] The limit determination unit 500 receives the force F2 and rd and outputs the elastic coefficient K and the neutral point r0. Figure 4 is a flowchart showing the control contents of the limit determination unit 500. This flowchart is executed when the peg fitting operation starts.

[0032] The limit determination unit 500 sets the elastic coefficient K to 0 [N / m] and the neutral point r0 to 0 as initial values ​​(step S101), and outputs K and r0 to the impedance control unit 400 (step S102). The limit determination unit 500 acquires the force F2 detected by the second force sensor 120 (step S103). In the state shown in FIGS. 2A and 2B above, the peg 200 and the insertion target object 300 are not in contact with each other, so the force F2 is 0. From step S103 until a positive determination is made in step S107, the limit determination unit 500 continues to acquire F2.

[0033] The restriction determination unit 500 determines whether the force F2 is greater than a threshold value fa (step S104). This threshold value fa is a threshold value that can determine whether the peg 200 and the insertion target object 300 have come into contact with each other. Therefore, although the threshold value fa is theoretically 0, an appropriate value is determined in advance by experiment or the like, taking into consideration the accuracy of the second force sensor 120 and the occurrence of noise.

[0034] If the force F2 is equal to or smaller than the threshold value fa (step S104: NO), the limit determination unit 500 proceeds to step S103. If the force F2 is greater than the threshold value fa (step S104: YES), the limit determination unit 500 determines that the peg 200 and the insertion target object 300 have come into contact, sets the elastic coefficient K to 1000 [N / m], sets the neutral point r0 to the target position rd (step S105), and outputs K and r0 to the impedance control unit 400 (step S106). In this way, when it is determined that the workpiece (peg 200) and the object (insertion target object 300) have come into contact based on the detected force, impedance control is executed with the position of the workpiece at the time of contact as the target position.

[0035] The elastic coefficient K substituted in step S105 need only be a value that causes the peg 200 to move very little in the normal direction of the contact surface (hereinafter also referred to as the "z direction"), and is not limited to 1000 [N / m]. Furthermore, with regard to the neutral point r0, the position of the peg 200 at the time of contact is set to the neutral point of the spring, so rd at that time is set to r0. In this way, the control executed by the impedance control unit 400 with the elastic coefficient K set to 1000 [N / m] and rd set to the neutral point r0 may be referred to as "spring simulation control."

[0036] With the spring simulation control, even if the operator 10 applies force in a direction pressing the peg 200 against the insertion object 300, the restoring force of the spring is used to control the peg 200 to return to the neutral point. This makes it possible to prevent excessive force from being generated between the peg 200 and the insertion object 300. Note that the restoring force of the spring does not act in directions other than the z direction, for example, a direction perpendicular to the z direction (hereinafter also referred to as the "y direction"), so the operator 10 can move the peg 200 in the y direction, etc.

[0037] After the peg 200 and the insertion target object 300 come into contact, the operator 10 moves the peg 200 to directly above the hole while maintaining contact with the insertion target object 300, as shown in Fig. 2C. Because the peg 200 remains in contact with the insertion target object 300 until just before it moves directly above the hole, the force F2 detected until just before the peg 200 moves directly above is greater than 0. On the other hand, once the peg 200 moves directly above the hole, the contact is released, and the detected force F2 becomes 0.

[0038] The limit determination unit 500 determines whether the force F2 that continues to be acquired even after contact is less than a threshold value fb (step S107). This threshold value fb is a threshold value that can determine whether the contact between the peg 200 and the insertion target object 300 has been released. In this case, theoretically it is sufficient to determine whether the force F2=0, but an appropriate value is determined in advance by experiment or the like, taking into consideration the accuracy of the second force sensor 120 and the occurrence of noise.

[0039] If the force F2 is equal to or greater than the threshold value fb (step S107: NO), the limit determination unit 500 executes step S107 again. If the force F2 is smaller than the threshold value fb (step S107: YES), the limit determination unit 500 determines that the contact between the peg 200 and the insertion target object 300 has been released, sets the elastic coefficient K to a smaller value (for example, 0 [N / m]), sets rd to the neutral point r0 (step S108), outputs K and r0 to the impedance control unit 400 (step S109), and ends the process. When the elastic coefficient K becomes 0 [N / m], the restoring force of the spring disappears, and the operator 10 can insert the peg 200.

[0040] As described above, when teaching a peg fitting operation to the robot 100 that performs impedance control on the force detected by the first force sensor 110, the limit determination unit 500 determines whether or not the peg 200 and the insertion target object 300 have come into contact with each other using the second force sensor 120. If the limit determination unit 500 determines that the peg 200 and the insertion target object 300 have come into contact with each other, it outputs the above-mentioned parameters (1000 [N / m], r0) for starting the spring simulation control. If the limit determination unit 500 determines that the contact has been released, it outputs parameters (0 [N / m], r0) for ending the spring simulation control.

[0041] This spring simulation control provides the following effect. That is, due to the inertia of the robot 100 and the influence of the high gear ratio joints, the operator 10 may not be able to feel the force acting on the peg 200 and may unintentionally apply excessive force to the peg 200. Even in such a case, the spring simulation control performs control to return the peg 200 to the neutral point using the restoring force of the spring. As a result, it is possible to prevent excessive force from being generated between the peg 200 and the insertion target object 300.

[0042] Furthermore, since the spring simulation control in this embodiment is not a control that stops the peg fitting operation, it has the effect of allowing the peg fitting operation to continue even if a temporary force such as contact occurs.

[0043] In the above-described embodiment, the elastic coefficient K is set to 1000 [N / m] when it is determined that the peg 200 and the insertion object 300 have come into contact. However, the elastic coefficient K may be set to a value greater than 0. If the elastic coefficient K is a value other than 0, a restoring force acts, which can prevent excessive force from occurring between the peg 200 and the insertion object 300 compared to when no restoring force acts. On the other hand, if the elastic coefficient K is too small, the restoring force is weak. Therefore, an appropriate value may be prepared in advance through experiments or the like depending on the inertia of the robot, the influence of joints with high gear ratios, the arm strength of the operator 10, and the like.

[0044] Second Embodiment In the first embodiment, an embodiment has been described in which an operator directly operates a robot to be taught, whereas in the second embodiment, an operator operates a robot different from the robot to be taught.

[0045] 5 is a diagram showing an example of the configuration of a remote teaching system 1000 according to the second embodiment. The remote teaching system 1000 is composed of a teaching robot 600 and a taught robot 700. The teaching robot 600 and the taught robot 700 are capable of communicating various types of information. Note that, for example, in this embodiment, the influence of the inertia of the robots and joints with high gear ratios is assumed to be the same for the teaching robot 600 and the taught robot 700.

[0046] The teaching robot 600 also performs admittance control, moving in response to the force applied by the operator, and when it receives a contact notification from the teachable robot 700 indicating that the peg has come into contact with the non-insertion target, it sets that position to the neutral position of the spring. The teachable robot operates in response to the operation of the teaching robot.

[0047] Torque sensors are provided at each of the joints 610A, 610B, 610C, and 610D of the robot arm of the teaching robot 600. The teaching robot control unit 650 is composed of a CPU, storage devices such as ROM and RAM, a communication device, and the like, and controls the entire teaching robot 600. The teaching robot control unit 650 is capable of performing admittance control and is also equipped with the impedance control unit 400 and limit determination unit 500 described in the first embodiment, and is capable of performing spring simulation control, as in the first embodiment. The detection unit 651 of the teaching robot 600 detects contact between a workpiece set on the robot arm of the teachable robot 700, which operates in synchronization with the teaching robot 600, and an object by receiving a contact notification (described later) from the teachable robot 700. The detection unit 651 notifies the teaching robot control unit 650 that contact has been detected.

[0048] The torque sensors provided at the joints 610A, 610B, 610C, and 610D detect the torque at each joint of the teaching robot 600 and output the detected torque to the teaching robot control unit 650.

[0049] The teachable robot 700 includes a force sensor 720 and a teachable robot control unit 750. A peg 800 is set on the robot arm of the teachable robot 700. A hole into which the peg 800 fits is provided in the insertion target object 900. Joints 710A, 710B, 710C, and 710D of the teachable robot 700 correspond to joints 610A, 610B, 610C, and 610D of the teaching robot 600, respectively.

[0050] The trainee robot control unit 750 is composed of a CPU, storage devices such as ROM and RAM, a communication device, etc., and controls the entire trainee robot 700. The force sensor 720 detects a force acting on the peg 800 and outputs the detected force to the trainee robot control unit 750. An example of the force acting on the peg 800 is a force that occurs when the peg 800 comes into contact with the insertion target object 900.

[0051] When the teachable robot control unit 750 determines that the peg 800 and the insertion target object 900 have come into contact, it transmits a contact notification indicating that the peg 800 and the insertion target object 900 have come into contact to the teaching robot 600. When the teaching robot control unit 650 receives the contact notification at the detection unit 651, it executes impedance control with the position of the tip of the robot arm of the teaching robot 600 at the time of contact (for example, the position of the joint 610A) as the target position.

[0052] When the teachable robot control unit 750 determines that the contact between the peg 800 and the insertion target object 900 has been released, it transmits a contact release notification indicating that the contact between the peg 800 and the insertion target object 900 has been released to the teaching robot 600. When the teaching robot control unit 650 receives the contact release notification, it sets the elastic modulus of the spring to 0.

[0053] This remote teaching system 1000 realizes a remote teaching method in which an operator 10 operates a teaching robot 600 different from a teachable robot 700, and moves the teachable robot 700 in response to the operation of the teaching robot 600, thereby teaching the teachable robot 700 to perform an action. When teaching remotely, the operator 10 cannot directly see the peg 800, making it more difficult to operate than when directly operating the teachable robot 700. Therefore, the operator 10 is likely to unintentionally apply excessive force, and therefore the spring simulation control according to this embodiment is particularly effective when teaching remotely.

[0054] FIG. 6 is a sequence diagram showing the processing flow in the remote teaching system 1000. When the teaching robot 600 starts a peg fitting operation, the teaching robot 600 operates using admittance control based on the torque detected by each torque sensor 610 at control periodic intervals (e.g., 1 ms). At the same time, the target position of the teaching robot 600 during operation, or the target position of the teachee robot 700 based on the target position, is transmitted to the teachee robot 700 at predetermined teaching intervals (step S201). An example of the predetermined interval is 10 ms. In this case, the teaching robot 600 transmits the target positions of the teachee robot 700, sampled every 10 ms, to the teachee robot 700 one after another. Upon receiving the target position, the teachee robot 700 operates the teachee robot 700 based on the received target position each time it receives the target position (step S202). From this point onward, the robot 700 will operate the corresponding joint each time it receives a calculated target position, but this is omitted in FIG. 6. Here, an example of a target position based on the target position of the teaching robot 600 during operation will be described. For example, when the teaching robot 600 and the taught robot 700 are similar to each other, and the target position of the teaching robot 600 is x, if a matrix A that performs a similarity transformation (similar enlargement transformation, similarity contraction transformation) is used, the target position y based on the target position x of the teaching robot 600 is Ax. Furthermore, when the teaching robot 600 and the taught robot 700, which are of the same size, are taught facing each other, if a matrix B that represents symmetric movement with respect to a virtual boundary line between the teaching robot 600 and the taught robot 700 is used, the target position y based on the target position x of the teaching robot 600 is Bx.

[0055] When the teachable robot control unit 750 of the teachable robot 700 determines that the peg 800 and the insertion object 900 have come into contact (step S203), it transmits a contact notification to the teaching robot 600 (step S204). In step S203, similar to step S104 in Fig. 4, the teachable robot control unit 750 determines whether the force F2 detected by the force sensor 720 is greater than the threshold value fa, and determines that the peg 800 and the insertion object 900 have come into contact.

[0056] When the teaching robot control unit 650 of the teaching robot 600 receives the contact notification, the detection unit 651 detects contact, and the teaching robot control unit 650 starts spring simulated movement control (step S205). As a result, in the teaching robot 600, control is performed to return the height at the time of contact by the restoring force of the spring through the spring simulated movement control, as in the first embodiment. As a result, even when remotely operated, it is possible to prevent excessive force from being generated between the peg 800 and the insertion target object 900.

[0057] Even after the spring simulation control starts, the teaching robot 600 performs admittance control based on the torque detected by each torque sensor 610, and operates based on the calculated position control target value. At the same time, the target position of the teachable robot 700 is transmitted to the teachee robot 700 one after another at predetermined intervals (step S206). Every time the teachee robot 700 receives a transmitted target position, it operates the teachee robot 700 based on the received target position.

[0058] When the teachable robot control unit 750 of the teachable robot 700 determines that the contact between the peg 800 and the insertion object 900 has been released (step S207), it transmits a contact release notification to the teaching robot 600 (step S208). In step S207, similar to step S107 in Fig. 4, the teachable robot control unit 750 determines whether the force F2 detected by the force sensor 720 is smaller than the threshold value fb, and determines that the contact between the peg 800 and the insertion object 900 has been released if the force F2 is smaller than the threshold value fb.

[0059] When the teaching robot 600 receives the contact release notification, it detects that contact has been released, and the teaching robot control unit 650 ends the spring simulation control (step S209). At this time, the elastic coefficient of the spring is set to 0, so the spring's restoring force no longer acts. Thereafter, the operator 10 performs an operation to fit the peg 800 into the hole, and since the target position is also transmitted to the teachable robot 700 at this time, it is possible to teach the teachable robot 700 the peg fitting operation even when the operator is operating remotely.

[0060] In the above embodiment, the spring simulated operation was initiated with the position of the peg at the time when the force F2 exceeded fa set as the neutral position of the spring simulated operation. However, the robot's target position at that time may be changed as the neutral position instead of the peg position. This is because the position at the time when the force F2 exceeded fa may be slightly different from the robot's target position at that time due to sensor errors, the robot's response speed, and threshold settings. If the robot's response speed is slow, and the position at the moment of contact is set as the neutral point, the actual robot position may move further in the direction of contact due to a response delay. On the other hand, if there is a large error between the robot's target position and its actual position, setting the robot's actual position as the neutral point may result in the neutral point being different from the position where contact actually occurred. For these reasons, the neutral point may be changeable.

[0061] Furthermore, regarding how to return to the neutral point after setting it, in the example (z direction) where the spring simulation operation is terminated when the workpiece moves to just above the hole (see Figure 2D) after contacting the workpiece, K = 0 is sufficient. However, if the workpiece moves while being subjected to frictional force (see Figure 2C), the frictional force will exceed the threshold, and once the spring simulation operation has begun, the workpiece will gradually become unable to move unless the neutral point changes. Therefore, if the workpiece moves away from the neutral point to a certain extent, it is considered not to be in a state of hitting something, so the neutral point is gradually shifted.

[0062] Specifically, the position of the peg when the force F2 exceeds fa is first set as the target position. If the peg is then pushed further downward, the target position is changed, and there are two ways to do this. One is to change the target position to the current peg position when the difference in distance between the peg position when the force F2 exceeds fa and the current peg position exceeds a predetermined value. The other is to change the target position to the current peg position in response to changes in the current peg position. Specifically, this method acquires the current peg position at predetermined intervals and changes the acquired position to the new target position.

[0063] <Variation 1> In the above embodiment, the workpiece (peg) receives force from an object (insertion target object), but it may also receive force from other objects (obstacles around the insertion target object). An obstacle around the insertion target object is, for example, the wall of the concave surface when the insertion target object is located on the concave surface of a concave-shaped object. When performing a peg fitting operation on the insertion target object, if the peg is first inserted into the concave surface, there is a possibility that the peg will come into contact with the wall of the concave surface.

[0064] Furthermore, an obstacle around the object to be inserted is, for example, a component located next to the object to be inserted and a hole in the object to be inserted that are close to each other. If the distance between the hole and the component is close, there is a possibility that the component will come into contact with the component when performing the peg fitting operation.

[0065] By detecting the force from obstacles around the object to be inserted with the detector, excessive force is prevented from being applied to the peg, and the peg can be inserted while avoiding the surrounding obstacles. Specifically, when the peg and the obstacle collide in the left-right direction, the neutral point can be used to prevent excessive left-right force, such as frictional force, from being applied to the peg.

[0066] <Variation 2> In the above-described embodiment, the spring simulated motion control is initiated upon contact with the insertion target object. However, the spring simulated motion control may also be initiated upon contact with an object other than the insertion target object. In this way, even when the peg comes into contact with the object, the point at which the peg comes into contact is set as the neutral point, preventing further force from being applied to the peg. Furthermore, when the spring simulated motion control is initiated upon contact with an object other than the insertion target object, a means for allowing the operator to cancel the spring simulated motion control may be provided. By providing a means for canceling the spring simulated motion control, the operator can cancel the spring simulated motion control if it is initiated at a position unintended by the operator, allowing the operator to flexibly use the spring simulated motion control according to the situation.

[0067] <Summary of the embodiment> A robot according to an embodiment of the present disclosure includes a detection unit (second force sensor 120) that detects a force from an object acting on a workpiece (peg 200) set on a robot arm, and a control unit (robot control unit 50) that, when it is determined that the workpiece and the object (insertion target 300) have come into contact based on the force detected by the detection unit, executes impedance control of the robot arm by setting the position of the workpiece at the time of contact as the neutral point (r0) of the spring in a control model of a spring-mass-damper system. This makes it possible to prevent excessive force from being applied during teaching.

[0068] A remote teaching system (remote teaching system 1000) according to one embodiment of the present disclosure includes a teaching robot (teaching robot 600) taught by an operator and a teachable robot (teachable robot 700) that operates in synchronization with the teaching robot and has a workpiece (peg 800) set on its robot arm. The teachable robot includes a detection unit (force sensor 720) that detects a force from an object acting on the workpiece, and a teachable robot control unit (teachable robot control unit 750) that sends a notification indicating the force detected by the detection unit to the teaching robot. The teaching robot includes a teaching robot control unit (teaching robot control unit 650) that, upon receiving the contact notification, performs impedance control such that the position of the tip of the robot arm of the teaching robot at the time of contact is the neutral point (r0) of the spring in a spring-mass-damper control model. This makes it possible to prevent excessive force from being applied during teaching.

[0069] A teaching robot according to an embodiment of the present disclosure is a teaching robot (teaching robot 600) that is taught by an operator, and includes a detection unit (detection unit 651) that detects contact between an object and a workpiece set on a robot arm of a teachable robot (teachable robot 700) that operates in synchronization with the teaching robot, and a teaching robot control unit (teaching robot control unit 650) that, when the detection unit detects the contact, executes impedance control with the position of the tip of the robot arm of the teaching robot set to the neutral point (r0) of the spring in a control model of a spring-mass-damper system. This makes it possible to prevent excessive force from being applied during teaching.

[0070] In a control method according to an embodiment of the present disclosure, a force from an object acting on a workpiece (peg 200) set on a robot arm is detected (step S103), and if it is determined that the workpiece and the object (insertion target 300) have come into contact based on the detected force (F2) (step S104: YES), impedance control of the robot arm is executed with the position of the workpiece at the time of contact set as the neutral point (r0) of the spring in a control model of a spring-mass-damper system (step S106). This makes it possible to prevent excessive force from being applied during teaching.

[0071] A program according to an embodiment of the present disclosure causes a computer to detect a force from an object acting on a workpiece (peg 200) set on a robot arm (step S103), and if it determines that the workpiece and the object have come into contact based on the detected force (F2) (step S104: YES), execute impedance control of the robot arm with the position of the workpiece at the time of contact set as the neutral point (r0) of the spring in a control model of a spring-mass-damper system (step S106). This makes it possible to prevent excessive force from being applied during teaching.

[0072] The present disclosure can be realized in software, hardware, or software in conjunction with hardware.

[0073] Each functional block used in the description of the above embodiments may be partially or entirely realized as an LSI, which is an integrated circuit, and each process described in the above embodiments may be partially or entirely controlled by a single LSI or a combination of LSIs. The LSI may be composed of individual chips, or may be composed of a single chip that includes some or all of the functional blocks. The LSI may have data input and output. Depending on the degree of integration, the LSI may be called an IC, system LSI, super LSI, or ultra LSI.

[0074] The integrated circuit method is not limited to LSI, but may be realized by a dedicated circuit, a general-purpose processor, or a dedicated processor. Also, a field programmable gate array (FPGA) that can be programmed after LSI manufacturing, or a reconfigurable processor that can reconfigure the connections and settings of circuit cells within the LSI, may be used. The present disclosure may be realized as digital processing or analog processing.

[0075] Furthermore, if an integrated circuit technology that can replace LSI emerges due to advances in semiconductor technology or other derivative technologies, it is natural that such technology may be used to integrate functional blocks. The application of biotechnology, etc. is also a possibility.

[0076] Although various embodiments have been described above with reference to the drawings, it goes without saying that the present disclosure is not limited to such examples. It is clear that a person skilled in the art can conceive of various modifications or alterations within the scope of the claims, and it is understood that these also naturally fall within the technical scope of the present disclosure. Furthermore, the components of the above-described embodiments may be combined in any manner without departing from the spirit of the disclosure.

[0077] Although specific examples of the present disclosure have been described in detail above, these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and alterations of the specific examples exemplified above. [Industrial Applicability]

[0078] An embodiment of the present disclosure is suitable for teaching an operation in which a workpiece is moved while in contact with an object. [Explanation of symbols]

[0079] 10 Operators 50 Robot control unit 100 robots 110 First force sensor 120 Second force sensor 200, 800 pegs 300, 900 Inserted object 400 Impedance control section 500 Restriction Judgment Unit 600 Teaching Robot 650 Teaching robot control unit 700 Taught robot 750 Taught robot control unit

Claims

1. a detection unit that detects a force from an object acting on a workpiece set on the robot arm; a control unit that, when it is determined that the workpiece and the object have come into contact based on the force detected by the detection unit, executes impedance control of the robot arm such that the position of the workpiece at the time of contact is set as a neutral point of a spring in a control model of a spring-mass-damper system; A robot equipped with

2. The robot according to claim 1 , wherein the control unit sets a value of the elastic coefficient of the spring in a control model of a spring-mass-damper system to a larger value when the contact occurs.

3. The robot according to claim 2 , wherein the control unit reduces the elastic coefficient of the spring when it determines that the contact between the workpiece and the object has been released based on the force detected by the detection unit.

4. The robot according to claim 1 , wherein the control unit changes the neutral point from the position of the workpiece at the time of contact.

5. The robot according to claim 4 , wherein the control unit changes the neutral point to the current position of the workpiece when a difference in distance between the position of the workpiece at the time of contact and the current position of the workpiece exceeds a predetermined value.

6. The robot according to claim 4 , wherein the control unit changes the neutral point to the current position of the workpiece in response to a change in the current position of the workpiece.

7. another detection unit that detects a force acting by teaching; the other detection unit is a torque sensor, The robot according to claim 1 , wherein the torque detected by the torque sensor is recorded as a teaching operation.

8. A remote teaching system including a teaching robot to which an operator gives instructions, and a taught robot that operates based on a target position notified from the teaching robot and has a workpiece set on its robot arm, The robot to be taught is a detection unit that detects a force from an object acting on the workpiece; a teachable robot control unit that transmits a contact notification to the teaching robot when it determines that the workpiece and an object have come into contact with each other based on the force detected by the detection unit; Equipped with The teaching robot is a teaching robot control unit that, upon receiving the contact notification, executes impedance control with the position of the tip of the robot arm of the teaching robot at the time of contact set to a neutral point of a spring in a control model of a spring-mass-damper system; Equipped with a remote teaching system.

9. A teaching robot to which an operator gives instruction, a detection unit that detects contact between a workpiece set on a robot arm of a taught robot that operates in synchronization with the teaching robot and an object; a teaching robot control unit that, when the detection unit detects the contact, executes impedance control with the position of the tip of the robot arm of the teaching robot set to a neutral point of a spring in a control model of a spring-mass-damper system, A teaching robot equipped with

10. A method for controlling a robot, comprising: Detects the force from an object acting on a workpiece set on the robot arm, When it is determined that the workpiece and the object have come into contact based on the detected force, impedance control of the robot arm is performed with the position of the workpiece at the time of contact being set as a neutral point of a spring in a control model of a spring-mass-damper system. Control method.

11. On the computer, The robot arm detects the force acting on the workpiece from an object, when it is determined that the workpiece and the object have come into contact based on the detected force, an impedance control of the robot arm is executed, with the position of the workpiece at the time of contact being set as a neutral point of a spring in a control model of a spring-mass-damper system. program.

Images