JP6565622B2 - Robot system and robot control method - Google Patents

Description

  The present invention relates to a robot system and a robot control method.

  In a robot having a multi-joint arm, inertial force generated by movement of a certain axis may affect other axes as interference force, and may adversely affect positioning and tracking accuracy. For example, in Patent Document 1, a multi-inertia interference model is used to obtain a non-interference torque signal and add it to the torque signal, thereby eliminating the influence of the interference force and controlling each axis to operate as instructed. A robot controller is described.

  On the other hand, a so-called end effector attached to the tip of the robot arm can also be a source of interference force. Until now, control capable of effectively reducing the interference force by the end effector has been disclosed in Patent Document 2 and Patent Document 3. As described in the above, what compensates for the machining reaction force obtained by actual measurement is known.

JP 2005-186235 A JP 2011-2116050 A JP 2012-176478 A

  As described above, a generally applicable control that can effectively reduce the interference force by the end effector attached to the tip of the arm of the articulated robot has not been known. Therefore, if an end effector that has a large inertial force during the operation of the end effector, such as a large moving part mass, is used, even if it is within the load capacity of the robot, interference during operation of the end effector will occur. The force cannot be neglected, and vibrations may occur, which may take time for positioning and reduce accuracy.

  The problem to be solved by the present invention is to provide a robot system and a robot control method capable of effectively reducing the interference force by the end effector in a robot having a multi-joint arm.

  A robot system according to one aspect of the present invention includes a robot having an articulated arm having a plurality of drive shafts, an end effector having an end effector shaft attached to the tip of the arm and serving as a drive shaft, A robot controller that controls a drive shaft, and the robot controller acts on each of the plurality of drive shafts by the inertia force, and an inertia force acquisition unit that obtains an inertia force generated by driving the end effector shaft A disturbance torque obtaining unit for obtaining a disturbance torque, a drive shaft control unit for controlling the plurality of drive shafts based on a torque command obtained from a control target value for each axis of the plurality of drive shafts and the disturbance torque; Have.

  In the robot system according to one aspect of the present invention, the drive shaft control unit may compensate for the disturbance torque by feedforward control.

  In the robot system according to one aspect of the present invention, the robot controller may include an end effector axis control unit that controls the end effector axis.

  In the robot system according to one aspect of the present invention, the robot controller includes a reverse inertia force acquisition unit that obtains a reverse inertia force that is an inertia force generated in the end effector by driving the plurality of drive shafts. The end effector axis control unit may control the end effector axis based on a target force and the reverse inertia force for the end effector axis.

  In the robot system according to one aspect of the present invention, the robot controller may include an inertial mass estimation unit that drives the end effector shaft and estimates the inertial mass of the end effector.

  In the robot system according to one aspect of the present invention, the robot controller may include an end effector information reading unit that reads end effector information including at least information related to the inertial mass of the end effector.

  In the robot system according to one aspect of the present invention, the articulated arm has at least six drive shafts, and the drive shaft control unit is configured to perform Jacobian for at least three axes on the base side of the arm. The disturbance torque may be obtained from the inertial force using a transposed matrix, and the disturbance torque may be obtained from the inertial force using a geometric relationship for at least three axes on the tip side of the robot.

  According to another aspect of the present invention, there is provided a robot control method, wherein the end of an end effector having an end effector shaft that is attached to the arm tip of a robot having an articulated arm having a plurality of drive shafts and that is a drive shaft. Obtaining an inertia force generated by driving the effector shaft, obtaining a disturbance torque acting on each of the plurality of drive shafts from the inertia force, and obtaining a torque command obtained from a control target value for each of the plurality of drive shafts; The plurality of drive shafts are controlled based on the disturbance torque.

It is a schematic diagram showing a robot system concerning an embodiment of the present invention. It is a functional block diagram of the robot system concerning this embodiment. It is a figure which shows the example of a control loop. It is the schematic diagram which looked at the flange from the axial direction of the T-axis. It is the schematic diagram which looked at the flange from the axial direction of the T-axis. It is the schematic diagram which looked at the flange from the axial direction of the B-axis. It is the schematic diagram which looked at the flange from the axial direction of the T-axis. It is the schematic diagram which looked at the flange from the axial direction of the B-axis. It is a flowchart which shows operation | movement of a robot controller.

  FIG. 1 is a schematic diagram showing a robot system according to an embodiment of the present invention. The robot system includes a robot 1 having an articulated arm 10 and an end effector 11 attached to the tip of the arm, and a robot controller 2 that controls the robot 1.

  Here, the robot 1 may be a general industrial robot. In this embodiment, the robot 1 is a 6-axis vertical articulated robot. However, the type and number of axes of the robot 1 are not limited, It may be of seven or more axes, or may be of other types such as a horizontal articulated type. In this example, the drive shafts provided on the arm 10 are called S-axis, L-axis, U-axis, R-axis, B-axis, and T-axis in order from the base side, and the rotation directions of the respective drive shafts are shown in FIG. As shown in FIG.

  Further, in this specification, the three axes from the base side of the arm 10, the S axis, the L axis, and the U axis are collectively referred to as a basic axis, and the three axes from the distal end side of the arm 10 are T The axis, the B axis, and the R axis are collectively referred to as a wrist axis. When the arm 10 has seven or more axes, the axis may be included in the basic axis or the wrist axis depending on the redundant axis arrangement. In any case, the basic axis includes at least three axes from the base side of the arm 10, and the wrist axis includes at least three axes from the wrist side of the arm 10.

  The end effector 11 has at least one end effector shaft that is a drive shaft, and the end effector 11 is operated by driving the end effector shaft. Hereinafter, when simply referred to as a drive shaft, it refers to each axis (S-axis to T-axis) provided in the arm 10, and the axis (Z-axis) of the end effector 11 is referred to as an end effector axis to distinguish both.

  Although there are no particular limitations on the use, function, and type of the end effector 11, in this example, the end effector 11 is a spot welding gun, and the end effector axis is shown as the Z axis in the figure. The end effector 11 of this example opens and closes the jaw-shaped member by driving the Z-axis, sandwiches the target member with welding electrodes provided so as to face each other at the tip of the jaw-shaped member, and applies a high voltage. Thus, spot welding is performed. Therefore, the movable member (in this case, the jaw-shaped member) of the end effector 11 moves along the Z-axis, and generates an inertial force in the Z-axis direction along with the movement. Although the end effector 11 itself has a weight within the payload of the robot 1, the inertial force generated by the operation of the end effector 11 cannot be ignored as the interference force with respect to the robot 1 because the mass of the movable member is large and the size is large. It will be about the size.

  The robot controller 2 controls the drive shaft and the end effector shaft. The drive source of the drive shaft and the end effector shaft is not particularly limited, but a servo motor is used in this embodiment. Each servo motor is controlled by the current output from the robot controller 2 so that its position and speed become desired.

  Further, in the following description, the base of the robot is the origin, the coordinates that take the vertical information in the Z direction are the robot coordinates Obot, the center of the flange to which the end effector 11 is attached is the origin, and the flange method The coordinates that take the line direction in the Z direction are flange coordinates Offrange (not shown), the point P that is the action reference point of the end effector 11 is the origin, and the coordinates that take the drive direction of the end effector axis in the Z direction are tool coordinates Otool ( (Not shown).

  FIG. 2 is a functional block diagram of the robot system according to the present embodiment. The robot controller 2 includes a drive axis control unit 20 and an end effector axis control unit 21, and the drive target 13 (S axis to T axis in FIG. 1) and the end effector axis 14 are input by inputting control target values, respectively. A control output (here, Z axis in FIG. 1), a current for each servo motor, is output, and the drive shaft 13 and the end effector shaft 14 are controlled.

  Further, the robot controller 2 is configured to compensate for the interference force that is applied to the arm 10 by the operation of the end effector 11 and the reverse interference that is an interference force that is applied to the end effector 11 by the operation of the arm 10. It has a configuration for inverse interference force compensation that compensates for force. Hereinafter, the interference force generated in the arm 10 due to the operation of the end effector 11 is simply referred to as interference force, and the interference force generated in the end effector 11 due to the operation of the arm 10 is referred to as reverse interference force.

[Configuration for interference force compensation]
The inertial force acquisition unit 22 acquires drive information of the end effector shaft 14, for example, current acceleration from the end effector axis control unit 21, and calculates and acquires the generated inertial force from the known inertial mass of the end effector 11. . This inertial force acts as an interference force on each drive shaft 13 of the arm 10.

  Further, the disturbance torque acquisition unit 23 calculates and acquires the disturbance torque generated in each drive shaft 13 from the inertia force. The obtained disturbance torque is used for control in the drive shaft control unit 20 and compensated. More specifically, the drive shaft control unit 20 performs control so that the disturbance torque is canceled with respect to the torque command for each drive shaft 13 obtained from the control target value.

In general position / speed control of a servo motor, a control loop as shown in FIG. 3 is used. That is, the speed command v and the acceleration command a are obtained in order from the difference e between the position target u that is a control target and the current position X of the motor, converted into a torque command T for the motor, and current output according to the torque command T Is made to the motor. Current position X and the motor speed v _e estimated by the calculation of the motor is fed back as shown.

In the present embodiment, in addition to such a general control loop, the disturbance torque T _t acquired by the disturbance torque acquisition unit 23 is added to the torque command T with an opposite sign. Thereby, the interference force generated in the arm 10 by the operation of the end effector 11 is canceled. Therefore, in terms of the disturbance torque T _t , this means that feedforward control is performed on the electric motor. Alternatively, it can be said that the inertial force acquisition unit 22 and the disturbance torque acquisition unit 23 in FIG. 2 function as a kind of disturbance observer with respect to the drive shaft control unit 20.

  With this configuration, the interference force by the end effector 11 is effectively reduced, and the vibration caused by the operation of the end effector 11 is suppressed, so that the positioning accuracy is improved and the speed is increased. In addition, since the interference force is reduced by feedforward control, it is excellent in quick response compared to the case of performing this by feedback control, and vibration is suppressed promptly.

[Configuration for compensating for reverse interference force]
The reverse inertia force acquisition unit 24 acquires drive information of each drive shaft 13 from the drive shaft control unit 20, for example, torque generated in each current drive shaft, and sends the inertial mass of the known end effector 11 to the end effector 11. Calculate and obtain the resulting reverse inertial force. This calculation is an inverse calculation using the torque T generated in each drive shaft in contrast to the calculation for obtaining the disturbance torque _Tt . This reverse inertia force acts on the end effector shaft 14 of the end effector 11 as a reverse interference force.

The obtained reverse inertial force is used for control in the end effector shaft control unit 21 and compensated. More specifically, the end effector shaft control unit 21 performs control so that the reverse inertia force is canceled with respect to the target force obtained from the control target value. In order to perform such control, the disturbance torque T _t in FIG. 3 may be read as the torque generated in the end effector shaft 14 due to the reverse inertia force, that is, the feed forward with respect to the motor of the end effector shaft 14 with respect to the reverse inertia force. Control is done. Further, the inverse inertia force acquisition unit 24 functions as a kind of disturbance observer with respect to the end effector shaft control unit 21.

  With this configuration, the reverse interference force due to the operation of the arm 10 is effectively reduced, and the vibration generated in the end effector 11 is suppressed, so that the positioning accuracy is improved and the speed is increased. In addition, since the reverse interference force is reduced by feedforward control, it is excellent in quick response compared to the case of performing this by feedback control, and vibration is suppressed promptly.

[Configuration for obtaining inertial mass]
In order to perform the above control, the inertial mass related to the end effector axis of the end effector 11 needs to be known. Therefore, when the inertial mass estimation unit 25 does not know the inertial mass, for example, when end effector information described later cannot be used, the inertial mass estimation unit 25 instructs the end effector axis control unit 21 to perform a known operation with respect to the end effector shaft 14. Let Due to the interference force generated at this time, a current for canceling the interference force is generated in each drive shaft 13, and the inertia mass is calculated and estimated from the value of the current value (or command torque).

  In this calculation, the inertial force generated in the end effector 11 by the calculation for the reverse interference force compensation described above is obtained from the disturbance torque estimated from the current value (or command torque) generated in each drive shaft 13, and the end effector is obtained. What is necessary is just to divide by the acceleration commanded to the axis 14. Thereby, interference force compensation and inverse interference force compensation for any end effector 11 are possible without measuring the inertial mass in advance.

  Alternatively, since the value of the inertial mass is unique to the end effector 11, it may be measured in advance for each end effector 11 and prepared as end effector information. The end effector information reading unit 26 reads the end effector information from, for example, an external device and obtains an inertial mass when such end effector information is available. The end effector information may include information other than the inertial mass, for example, information related to the attachment angle of the end effector 11 with respect to the flange 12. Thereby, even when the end effector 11 is replaced with a different one, interference force compensation and inverse interference force compensation can be performed using an appropriate inertial mass.

  In the above description, the robot controller 2 has the end effector axis control unit 21 and controls the end effector axis 14 as well. In such a configuration, the robot controller 2 can directly obtain the drive information of the end effector shaft 14, so that the above-described interference force compensation and inverse interference force compensation can be easily performed. However, the present invention is not limited to this, and the end effector axis control unit 21 may be prepared as a separate device for the robot controller 2 and may be controlled synchronously by communicating with each other. It is necessary to prepare a communication line so that necessary information can be transmitted and received between the effector axis control unit 21 and the robot controller 2 and to pay attention to the influence of lag caused by communication.

  In the above-described example, the inertial force is obtained timely by calculating from the drive information of the end effector shaft 14 and the known inertial mass of the end effector 11, but this is given in advance as a known one. When the effector shaft 14 is driven, it may be obtained by referring to a given inertial force. This is effective when the operation of the end effector shaft 14 is limited to a predetermined one and the inertial force generated is limited. At this time, the inertia force acquisition unit 22 may refer to the inertia force according to the operation of the end effector shaft 14. In the case where the inertial force is given as a known one, the inertial force may be measured in advance for each assumed operation of the end effector 11, or the inertial force acquisition unit 22 calculates the inertial force as described above. It is also possible to store the result after obtaining the above and refer to the stored inertial force when the end effector shaft 14 is operated under the same conditions. For example, the inertial force given as a known one is included in the above-described end effector information, and may be read by the end effector information reading unit 26.

[Calculation of interference force compensation]
Next, a specific calculation method when performing interference force compensation will be described.

  In the present embodiment, the calculation by the disturbance torque acquisition unit 23 is divided into a calculation for the basic axis and a calculation for the wrist axis, and the basic axis acts on each axis by inertia force using a Jacobian transpose matrix. For the wrist axis, the disturbance torque acting on each axis is obtained from the inertial force using a geometrical relationship.

  In general, the computation using the Jacobian transpose matrix has a large computation amount, and the computation amount increases rapidly as the order of the Jacobian transpose matrix increases. Therefore, by performing calculations based on a geometrical relationship with a small amount of calculation for some axes, the number of axes that require calculation using the Jacobian transpose matrix can be reduced and the order can be reduced. it can.

[Calculation for basic axis]
First, the rotation matrix rot from the flange coordinate Offrange to the tool coordinate Otool when the attachment angle of the end effector 11 to the flange 12 is the rotation angle Rx, Ry, Rz with respect to the X axis, Y axis, and Z axis of the flange coordinate Offrange. _{The flag} is as follows.

  Here, the tool coordinates Otool are rotated with respect to the flange coordinates Offrange in the order of the Z axis, the Y axis, and the X axis.

The rotation matrix rot _tool from the tool coordinates Otool to the flange coordinates Offrange is as follows.

Furthermore, assuming that the matrix indicating the posture of the point P viewed from the origin of the robot coordinates Orobot is Inrepos and the matrix indicating the posture of the flange 12 is rot, the rotation matrix rot _robot from the robot coordinates Obot to the tool coordinates Otool is as follows. .

When this rotation matrix rot _robot is used and the inertial force generated in the Z direction of the tool coordinate Otool is Fz _tool , the force generated in the robot coordinate Orobot is as follows.

Furthermore, using a Jacobian transpose matrix ^{J T,} it converts the external force of the robot coordinate Orobot S axis, L-axis, the disturbance torque tau _S with respect to the U-axis, tau _L, the tau _U.

By the above calculation, the disturbance torques τ _S , τ _L , τ _U with respect to the S axis, the L axis, and the U axis generated by the inertial force Fz_tool are obtained.

[Calculation for wrist axis]
For the wrist axis, disturbance torques τ _T , τ _B , and τ _R acting on the T-axis, B-axis, and R-axis are obtained based on the geometric relationship.

At this time, due to the inertial force Fz _tool generated in the Z direction of the _tool coordinate Otool, the force generated in the flange coordinate Offrange is as follows.

  In the flange coordinates Offrange, the coordinates of the point P are set to (Lx, Ly, Lz).

FIG. 4 is a schematic view of the flange as viewed from the axial direction of the T axis. Here, the center of rotation of the T-axis coincides with the center of the flange and is the origin of the flange coordinates Offrange. As is clear from the figure, the disturbance torque τ _T acting on the T-axis is as follows.

FIG. 5A is a schematic view of the flange as viewed from the axial direction of the T axis, and θ _T is the rotation angle of the T axis. FIG. 5B is a schematic view of the flange as viewed from the axial direction of the B axis, and d6 is a distance from the rotation center of the B axis to the flange center.

As is clear from the figure, the disturbance torque τ _B acting on the B-axis is as follows.

For the R axis, FIG. 6A is a schematic view of the flange as viewed from the axial direction of the T axis, and θ _T is the rotation angle of the T axis. FIG. 6B is a schematic view of the flange as viewed from the axial direction of the B axis, and θ _B is the rotation angle of the B axis. At this time, the force acting in the direction perpendicular to the paper surface of FIG. 6B appears as a force in the horizontal direction of the paper surface in FIG. 6A, so that the disturbance torque τ _R acting on the R axis is as follows.

In this way, the disturbance torques τ _T , τ _B , τ _R acting on the wrist shaft can be easily obtained using the geometric relationship.

[Calculation of reverse interference force compensation]
Next, a specific calculation method when performing reverse interference force compensation will be described.

  In this calculation, the torque generated by the drive shaft 13 may be calculated as a disturbance torque, and the calculation reverse to that performed in the above-described interference force compensation may be performed. However, here, the torque is relatively close to the end effector 11. The influence of reverse interference force due to the driving of a wrist shaft is smaller than the influence of reverse interference force due to the driving of the basic shaft at a relatively far position, so only the influence of reverse interference force due to the driving of the basic shaft is considered. And Of course, it is also possible to consider the influence of reverse interference force due to driving of the wrist shaft.

Here, assuming that the torques generated in the basic axis S axis, L axis, and U axis are τ _S , τ _L , and τ _U , respectively, the inverse inertia force in the robot coordinates Orobot generated by these torques is the inverse matrix of the Jacobian transpose matrix J ^T. Using (J ^T ) ^{−1, the result} is as follows.

Further, by multiplying the above-described rotation matrix rot _robot ^T by the transpose matrix rot _robot ^T , the inverse inertia force Fz _tool in the z-axis direction at the tool coordinates Otool can be obtained as follows.

[Robot controller operation]
The operation of the robot controller will be described with reference to the flowchart shown in FIG.

  First, in step ST1, it is determined whether or not the inertial mass of the end effector 11 is known. If it is known (affirmed), the process proceeds to step ST5. If not (no), the process proceeds to step ST2, and it is determined whether or not end effector information is available.

  If the determination in step ST2 is affirmative, that is, if end effector information is available, the process proceeds to step ST3, where the end effector information reading unit 26 reads the end effector information and acquires the inertial mass included in the end effector information. If NO in step ST2, that is, if the end effector information is not available, the process proceeds to step ST4, where the inertial mass estimating unit 25 estimates the inertial mass by driving the end effector shaft. In any case, after acquiring the inertial mass, the process proceeds to step ST5.

  After step ST5, the control is performed during normal operation of the robot 1, so that the arm 10 and the end effector 11 of the robot 1 perform a desired operation based on a general motion program or a command from the host controller. Is made. In FIG. 7, in particular, the operations related to the interference force and inverse interference force compensation performed in parallel with these normal controls are shown.

  In step ST5, it is determined whether or not the interference force compensation is effective. In the robot system according to the present embodiment, it is assumed that the presence / absence of interference force compensation and inverse interference force compensation can be switched by a user's designation or the like. In step ST5, a determination is made to switch presence / absence of interference force compensation. Yes.

  If the determination in step ST5 is affirmative, that is, if the interference force compensation is valid, the process proceeds to step ST6, where the inertia force acquisition unit 22 acquires the inertia force. Furthermore, it progresses to step ST7 and the disturbance torque with respect to each drive shaft 13 is acquired by the disturbance torque acquisition part 23 from an inertia force. And it progresses to step ST8, feedforward control is performed to the drive shaft 13 by making the obtained disturbance torque into the correction value with respect to a torque command, and interference force compensation is performed.

  If NO in step ST5, that is, if the interference force compensation is invalid, and after step ST8 ends, the process proceeds to step ST9, where it is determined whether the reverse interference force compensation is valid. In step ST9, it is determined to switch the presence / absence of inverse interference force compensation.

  If the determination in step ST9 is affirmative, that is, if the reverse interference force compensation is valid, the process proceeds to step ST10, where the reverse inertia force acquisition unit 24 acquires the reverse inertia force. Further, the process proceeds to step ST11, where the obtained reverse inertia force is used as a correction value for the driving force command, or the reverse effect force is converted into torque, and then the feed effect control is performed on the end effector shaft 14 as the correction value for the torque command. Interference force compensation is performed.

  After the end of the series of operations, that is, in the case of negative in step ST9 and after the end of step ST11, the process returns to step ST5 and is repeated thereafter. Thereby, in this invention, according to operation | movement of the robot system 1, the interference force and reverse interference force which change every moment are acquired and compensated in real time. Therefore, it is not necessary to measure the inertial force and reverse inertial force generated in the end effector 11 in advance, and even when the operation of the robot system 1 is newly created or changed, the interference force and reverse interference force are immediately detected. Compensation can be performed, the positioning accuracy can be ensured, and the operation can be speeded up.

  As mentioned above, although embodiment concerning this invention was described, the specific structure shown in this embodiment was shown as an example, and it is not intending limiting the technical scope of this invention to this. Those skilled in the art may appropriately modify these disclosed embodiments, and it should be understood that the technical scope of the invention disclosed herein includes such modifications.

  DESCRIPTION OF SYMBOLS 1 Robot, 2 Robot controller, 10 Arm, 11 End effector, 12 Flange, 13 Drive axis, 14 End effector axis, 20 Drive axis control part, 21 End effector axis control part, 22 Inertial force acquisition part, 23 Disturbance torque acquisition part , 24 Inverse inertia force acquisition unit, 25 Inertial mass estimation unit, 26 End effector information reading unit.

Claims (8)

A robot having an articulated arm with a plurality of drive axes;
An end effector comprising a movable member attached to the arm tip and driven along one end effector axis to perform a predetermined operation ;
A robot controller for controlling the plurality of drive axes,
The robot controller is
An inertial force acquisition unit that repeatedly calculates and acquires the inertial force in the end effector axial direction generated in the end effector with the predetermined operation of the movable member ;
A disturbance torque obtaining unit for obtaining a disturbance torque acting on each of the plurality of drive shafts from the inertial force repeatedly obtained by the inertial force obtaining unit;
A drive shaft control unit for controlling the plurality of drive shafts based on a torque command obtained from a control target value for each of the plurality of drive shafts and the disturbance torque;
A robot system.   The robot system according to claim 1, wherein the drive shaft control unit performs compensation for the disturbance torque by feedforward control. The robot system according to claim 1, wherein the robot controller includes an end effector axis control unit that controls driving of the movable member along the end effector axis. The robot controller has a reverse inertia force acquisition unit that repeatedly calculates and acquires a reverse inertia force that is an inertia force generated in the end effector by driving the plurality of drive shafts,
The robot system according to claim 3, wherein the end effector axis control unit controls the end effector axis based on a target force and a reverse inertia force in driving the movable member along the end effector axis. The robot controller B over La, upon driving along the end effector axis of the movable member, the robot system according to claim 3 or 4 having an inertial mass estimation unit for estimating an inertial mass of the end effector.   The robot system according to claim 1, wherein the robot controller includes an end effector information reading unit that reads end effector information including at least information related to an inertial mass of the end effector. The articulated arm has at least six drive axes.
The drive axis control unit obtains the disturbance torque from the inertia force using a Jacobian transposition matrix for at least three axes on the base side of the arm, and has a geometric relationship with respect to at least three axes on the tip side of the robot. The robot system according to any one of claims 1 to 6, wherein the disturbance torque is obtained from the inertial force by using a motor. The movable member of the end effector having a movable member attached to the tip of the robot having a multi-joint arm having a plurality of drive shafts and driven along one end effector shaft to perform a predetermined operation . Repeatedly determining the inertial force generated by driving along the end effector axis;
Obtain disturbance torque acting on each of the plurality of drive shafts from the inertial force repeatedly obtained ,
Controlling the plurality of drive shafts based on a torque command obtained from a control target value for each of the plurality of drive shafts and the disturbance torque;
Robot control method.

Images