WO2013183190A1 - Système de robot, dispositif de commande de robot et procédé de commande de système de robot - Google Patents

Système de robot, dispositif de commande de robot et procédé de commande de système de robot Download PDF

Info

Publication number
WO2013183190A1
WO2013183190A1 PCT/JP2012/083299 JP2012083299W WO2013183190A1 WO 2013183190 A1 WO2013183190 A1 WO 2013183190A1 JP 2012083299 W JP2012083299 W JP 2012083299W WO 2013183190 A1 WO2013183190 A1 WO 2013183190A1
Authority
WO
WIPO (PCT)
Prior art keywords
robot
angular velocity
joint
velocity command
command
Prior art date
Application number
PCT/JP2012/083299
Other languages
English (en)
Japanese (ja)
Inventor
陽介 神谷
安田　賢一
慎悟 安藤
Original Assignee
株式会社安川電機
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社安川電機 filed Critical 株式会社安川電機
Publication of WO2013183190A1 publication Critical patent/WO2013183190A1/fr

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/0084—Programme-controlled manipulators comprising a plurality of manipulators
    • B25J9/0087—Dual arms
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1602—Programme controls characterised by the control system, structure, architecture
    • B25J9/1607—Calculation of inertia, jacobian matrixes and inverses
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/39—Robotics, robotics to robotics hand
    • G05B2219/39062—Calculate, jacobian matrix estimator

Definitions

  • the present invention relates to a robot system, a robot control device, and a control method of the robot system.
  • Patent Document 1 describes a robot control device that can reduce the power required for the operation of the robot when performing the work by the robot.
  • the robot control device includes an operation command generation unit that generates operation data for operating the robot, an energy calculation unit that calculates an energy amount necessary for the operation based on the operation data, and a reduction from the calculated energy amount.
  • a first path generation unit that generates a robot position command that can be generated, a second path generation unit that generates a robot position command based on motion data output from the motion command generation unit, And a control method switching means for switching the input from the first route generation means and the second route generation means and outputting a position command to the control means.
  • An object of the present invention is to provide a robot system, a robot control device, and a robot system control method capable of automatically generating a teaching position and an operation trajectory for reducing energy consumption of a redundant robot.
  • a robot having a plurality of joint axes that connect a plurality of links, including redundant axes, and wherein the plurality of joint axes are driven by a servo motor.
  • a robot control device for controlling the robot, The robot control device generates a first joint angular velocity command of the robot for reducing kinetic energy based on the input velocity command of the control point of the robot; A first optimizing operation command generation unit that generates a joint angular velocity command A for the robot to perform the arm lowering operation; A zero-space matrix calculator for calculating a zero-space matrix of a Jacobian matrix related to the control points; A second command generation unit that generates a second joint angular velocity command of the robot based on the joint angular velocity command A and the zero-space matrix generated by the first optimization operation command generation unit; A robot system including a control unit that controls the servo motor based on a third joint angular velocity command that is the sum of the first and second joint angular velocity commands is applied.
  • the robot has a plurality of joint axes including a redundant axis, and the plurality of joint axes are driven by a servo motor;
  • a robot control device for controlling the robot, The robot control device generates a first joint angular velocity command of the robot for reducing kinetic energy based on the input velocity command of the control point of the robot;
  • An optimization operation command generation unit for generating a joint angular velocity command for suppressing the posture holding torque of the robot;
  • a zero-space matrix calculator for calculating a zero-space matrix of a Jacobian matrix related to the control points;
  • a second command generation unit that generates a second joint angular velocity command of the robot based on the joint angular velocity command generated by the optimization operation command generation unit and the zero space matrix;
  • a robot system having a control unit that controls the servo motor based on a third joint angular velocity command that is the sum of the first and second joint angular velocity commands is applied.
  • the robot generates a first joint angular velocity command for reducing the kinetic energy based on a velocity command of a control point of a robot having a plurality of joint axes including redundant axes.
  • 1 command generator A first optimizing operation command generation unit that generates a joint angular velocity command A for the robot to perform the arm lowering operation;
  • a zero-space matrix calculator for calculating a zero-space matrix of a Jacobian matrix related to the control points;
  • a second command generation unit that generates a second joint angular velocity command of the robot based on the joint angular velocity command A and the zero space matrix;
  • a robot control device having a control unit for controlling a servo motor based on a third joint angular velocity command that is the sum of the first and second joint angular velocity commands is applied.
  • the step of generating a first joint angular velocity command for the robot for reducing kinetic energy based on a velocity command for a control point of a robot having a plurality of joint axes including redundant axes SA1 and Step SA2 for generating a joint angular velocity command A for performing an arm lowering operation in which the robot suppresses the extension of the redundant axis by using the redundancy degree of freedom while maintaining the position and orientation of the control point; Calculating SA3 of the Jacobian matrix for the control points; Generating a second joint angular velocity command for the robot based on the joint angular velocity command A and the zero-space matrix; and Step SA5 in which a servo motor for driving each joint axis is controlled based on a third joint angular velocity command that is the sum of the first and second joint angular velocity commands, and the robot performs the arm lowering operation;
  • the control method of the robot system that executes is applied.
  • a robot system 10 includes a robot 12, a robot controller 14 that controls the operation of the robot 12, and a teaching pendant for teaching the position and orientation of the robot 12. 16.
  • the robot coordinate system Cr shown in FIG. 1 has, for example, a point on the moving carriage 12a of the robot 12 as an origin, the X axis extending in the front-rear direction of the robot 12, the Y axis extending in the left-right direction, and the vertical direction. It is an orthogonal coordinate system consisting of the Z axis.
  • the robot 12 includes a moving carriage 12a, a waist 12b, a body 12c, a left arm LA, and a right arm RA.
  • the movable carriage 12a can travel on the floor surface using, for example, an omni wheel.
  • the movable carriage 12a has a travel axis Vx for translational movement in the X-axis direction, a travel axis Vy for translational movement in the Y-axis direction, and a travel axis Vz for rotational movement about the Z-axis. ing.
  • the waist part 12b has a waist joint axis L1 and a waist joint axis L2 which are arranged with a space between each other and are rotatable about axes extending in the direction along the floor surface.
  • the waist 12b has a waist joint axis L3 that can rotate around an axis extending in the longitudinal direction of the body 12c. Therefore, as shown in FIG. 2B, the robot 12 can drive the waist joint axes L1 to L3 to bend or twist the waist 12b.
  • the trunk 12c is provided on the upper side of the waist 12b. As shown in FIG. 2A, a head portion 12d is provided on the upper portion of the body 12c.
  • the left arm LA is provided on the left side of the body 12c, and a plurality of links LNK are connected via joint axes LJ1 to LJ7.
  • a left hand LH is provided at the tip of the left arm LA.
  • the right arm RA is provided on the right side of the trunk 12c, and a plurality of links LNK are connected via joint axes RJ1 to RJ7.
  • a right hand RH is provided at the tip of the right arm RA.
  • Each of the travel axes Vx, Vy, Vz, the waist joint axes L1 to L3, and each of the joint axes RJ1 to RJ7, LJ1 to LJ7 is driven by a servo motor (not shown) having an encoder built therein.
  • the traveling axes Vx, Vy, Vz and the waist joint axes L1 to L3 are simply referred to as joint axes Vx, Vy, Vz and joint axes L1 to L3, respectively.
  • the robot 12 can take any configuration as long as it has redundant joint axes with respect to the degree of freedom of the work space of the robot 12.
  • a redundant robot refers to a robot that has redundancy in the degree of freedom of movement of the joints relative to the degree of freedom of movement of the control point (for example, set at the hand of the robot 12).
  • a three-axis robot having only three joint axes that rotate about the Y axis of the robot coordinate system Cr has redundancy with respect to the XZ plane. Therefore, this three-axis robot is a robot having redundant joint axes with respect to the movement on the XZ plane.
  • the robot may be a redundant robot having a redundancy degree of freedom for a required operation regardless of the number of axes of the joints and the axis configuration.
  • the control point refers to the origin of the coordinate system set at a location to be controlled by the robot, and is generally set at the hand.
  • the control point may be used to include not only the position but also the posture.
  • the position and posture of the hand coordinate system based on the robot coordinate system Cr are expressed as “control point position and posture”. .
  • the robot control device 14 is connected to the robot 12 via a cable.
  • the robot control device 14 may be mounted on the robot 12.
  • the robot controller 14 includes a first command generation unit 14a, a first optimization operation command generation unit 14b, a second optimization operation command generation unit 14c, and a zero space matrix calculation. 14d, 2nd command generation part 14e, and position speed control part (an example of a control part) 14f.
  • the robot control device 14 is equipped with a CPU (not shown), and the function of each block shown in FIG. 3 is realized by a software program or hardware executed by the CPU.
  • a speed command (vector) V of the hand (an example of a control point) of the robot 12 is input to the first command generation unit 14a.
  • the first command generation unit 14a can generate a joint angular velocity command (first joint angular velocity command) ⁇ TASK for moving the hand based on the input hand speed command V.
  • This joint angular velocity command ⁇ TASK is a command for improving the efficiency of the movement of the robot 12 and reducing the kinetic energy necessary for the movement of each joint.
  • the speed command V of the hand of the robot 12 is simply referred to as “hand speed command V”.
  • the first optimization operation command generation unit 14b can output a joint angular velocity command ⁇ ELB (joint angular velocity command A) for the robot 12 to perform the arm lowering operation.
  • the arm lowering operation is an operation for suppressing the extension of the arm elbow (redundant axis) caused by the movement of each joint of the robot 12 to the inside using the redundancy degree of freedom while maintaining the position and orientation of the control point.
  • the calculation method of the joint angular velocity command ⁇ ELB will be described later.
  • the second optimization operation command generation unit 14c is configured to reduce the torque necessary for holding the posture of the robot 12 (posture holding torque), and the joint angular velocity command ⁇ GRV (for improving the motion of the robot 12).
  • the joint angular velocity command B) can be output.
  • the calculation method of the joint angular velocity command ⁇ GRV will be described later.
  • the robot controller 14 can automatically generate a redundant axis joint angular velocity command ⁇ ELB or a joint angular velocity command ⁇ GRV in conjunction with the joint angular velocity command ⁇ TASK that realizes the hand velocity command V.
  • the zero-space matrix calculation unit 14d calculates the Jacobian matrix J related to the hand of the robot 12 based on the joint angles ⁇ CUR (vector) of the joint axes Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, and LJ1 to LJ7 of the robot 12. Can be calculated. Specifically, the zero space matrix calculation unit 14d calculates the position and orientation of the hand of the robot 12 from the joint angle ⁇ CUR and the directions of the joint axes Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, and LJ1 to LJ7. The zero space matrix N of the Jacobian matrix J related to the hand of the robot 12 can be calculated.
  • the joint angle ⁇ CUR either a current value ⁇ fb of the joint angle calculated from the encoder of the servo motor that drives the robot 12 or a command value ⁇ ref of the joint angle calculated on software is used. .
  • the current value ⁇ fb of the joint angle and the command value ⁇ ref of the joint angle are values before one control period (one sampling period).
  • the current value ⁇ fb of the joint angle of the robot 12 is obtained based on the result of multiplying the encoder value fed back from the encoder of each servo motor by the reduction ratio.
  • the zero-space matrix N is an operation matrix for generating an operation command with a redundancy degree of freedom while maintaining the position and orientation of the hand in the redundant robot.
  • This motion command is not a command for a specific axis as a redundant axis, but a command for a redundancy degree of freedom (joint degree of freedom that does not affect the position and orientation of the hand), and the control target axis includes all redundant axes. This is an axis or a part of the axis, and differs depending on the posture and operation direction of the robot for each control cycle.
  • this control target axis has only three joint axes (first to third joint axes) that rotate about the Y axis of the robot coordinate system Cr, and arbitrarily moves the position on the XZ plane.
  • a three-axis robot will be described as an example.
  • the control target axis as the redundancy degree of freedom is not limited to one of the first to third joint axes of the three-axis robot, and the robot coordinate system Cr is maintained while maintaining the hand position. At least one of the first to third joint axes that can be rotated about the Y axis (the posture can be changed). Which joint axis is the control target axis is calculated using a zero-space matrix.
  • the control target axis as the redundancy degree of freedom is the first joint axis and the first joint axis. 2 joint axes.
  • control point When a part other than the hand of the robot 12 is set as a control target part (control point), another Jacobian matrix corresponding to the control target part (control point) is used.
  • the second command generator 14e can generate a joint angular velocity command (second joint angular velocity command) ⁇ ref2 based on the joint angular velocity command ⁇ ELB , the joint angular velocity command ⁇ GRV , and the zero space matrix N.
  • the generated joint angular velocity command ⁇ ref2 is added to the joint angular velocity command ⁇ TASK generated by the first command generation unit 14a to become a joint angular velocity command (third joint angular velocity command) ⁇ ref3 (see FIG. 3).
  • the joint angle command value ⁇ ref is obtained by adding the joint angular velocity command ⁇ ref3 to the current joint angle value ⁇ fb of the robot 12 when the power is turned on (a value obtained by multiplying the encoder value by the reduction ratio), and is integrated every control cycle. And updated.
  • the position speed control unit 14f can control the servo motor based on the joint angular speed command ⁇ ref3. Specifically, the position / speed control unit 14f controls the servo motor based on the joint angle command value ⁇ ref which is the sum of the integral value of the joint angular velocity command ⁇ ref3 and the joint angle ⁇ CUR and the current value ⁇ fb of the joint angle. Feedback control is possible. That is, the position / velocity control unit 14f can control each servo motor to control the position / speed of each joint axis Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, and LJ1 to LJ7 of the robot 12.
  • the joint angular velocity command ⁇ ref3 is used as a command to the position / speed control unit 14f, and the joint angular velocity command ⁇ ref3 is obtained based on the result of multiplying the encoder value fed back from the encoder of each servo motor by the reduction ratio.
  • the feedback control may be performed based on the current value.
  • the position / speed control unit 14f is provided in an amount corresponding to the number of joint axes Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, and LJ1 to LJ7 of the robot 12.
  • control target portion control point
  • control target part is, for example, a specific part on the left hand LH or a specific part on the right hand RH.
  • Other examples of the control target location include a specific part of the joint of the robot 12 and the tip of an end effector attached to the hand.
  • a control object location can be switched as left arm LA, right arm RA, or both arms LA and RA.
  • the designated coordinate system is, for example, a robot coordinate system Cr.
  • the robot coordinate system Cr is selected as the designated coordinate system
  • the hand of the robot 12 is moved in the X-axis direction, the Y-axis direction, and the Z-axis direction, the rotation direction around the X axis, and the rotation around the Y axis.
  • the posture can be changed in the rotation direction and the rotation direction around the Z axis.
  • Other examples of the designated coordinate system include a joint coordinate system and a coordinate system arbitrarily set by the operator of the robot 12.
  • the teaching pendant 16 is provided with a jog operation button, an operation mode switching button A, an operation mode switching button B, and a playback button (not shown).
  • the jog operation button is a button for the operator to jog the robot 12.
  • a hand speed command V preset for the jog operation button is input to the first command generation unit 14a.
  • Jog operation buttons are, in detail, X + button, X- button, Y + button, Y- button, Z + button, Z- button, Rx + button, Rx- button, Ry + button, Ry- button, Rz + button, Rz- button It is.
  • the hand speed command V is input to the first command generator 14a so that the hand moves in the positive direction of the X axis of the designated coordinate system at a preset speed.
  • the hand speed command V is sent to the first command generator 14a so that the hand posture changes in the positive direction around the X axis of the designated coordinate system at a preset speed. Entered.
  • Operation mode switching buttons A and B are buttons for the operator to switch the operation mode of the robot 12.
  • operation mode A the arm switching operation mode
  • operation mode B the operation mode is switched to an operation mode (hereinafter sometimes referred to as operation mode B) for reducing the posture holding torque.
  • the signal A shown in FIG. 3 is input to the first optimization operation command generation unit 14b, and the signal B is input to the second optimization operation command generation unit 14c. Is input.
  • the signal A is input to the first optimization operation command generation unit 14b
  • the first optimization operation command generation unit 14b operates.
  • the signal B is input to the second optimization operation command generation unit 14c, the second optimization operation command generation unit 14c operates.
  • the playback button is a button for performing the playback operation of the taught work program.
  • An operator of the robot 12 can jog the robot 12 using a jog operation button to teach a work program for the robot 12. Further, the operator can perform a reproduction operation of the work program using the reproduction button.
  • the operation of the robot system 10 will be described with reference to FIG.
  • the operator presses the jog operation button and as shown in the upper diagram of FIG. 6A, the robot 12 grips the article OBJ placed on the table 40 between the hands LH and RH of both arms LA and RA.
  • the operation when teaching the program will be described.
  • the operation of the robot 12 is controlled in the operation mode A so that the arm 12 performs the arm lowering operation when the robot 12 grips the article OBJ placed on the table 40.
  • Step SA1 The operator presses the operation mode switching button A to set the operation mode to the operation mode A (arm lowering mode). By this operation, the signal A shown in FIG. 3 is input to the first optimization operation command generation unit 14b.
  • the operator presses a jog operation button (for example, X + button) to move the hand of the robot 12 to a position and posture for gripping the article OBJ. That is, the operator jogs the robot 12.
  • a jog operation button for example, X + button
  • the hand speed command V is input to the first command generator 14a shown in FIG.
  • the robot 12 may move autonomously based on information obtained from a sensor (not shown). For example, it may be an operation of recognizing the article OBJ using a vision sensor attached to the robot 12 and autonomously moving the hand from the current position toward the article OBJ.
  • the hand speed command V input to the first command generation unit 14a is obtained, for example, by the following procedure. First, a speed command is calculated by dividing the distance between the current control point of the robot 12 and the position of the article OBJ as a recognition result by a preset operation time. Next, the calculated speed command is filtered to obtain a hand speed command V to which a predetermined acceleration / deceleration is applied. For the filtering, for example, a moving average filter or a low-pass filter is used. By filtering, rapid acceleration / deceleration of the robot is avoided on the software.
  • the first command generation unit 14a, the first optimization operation command generation unit 14b, the zero space matrix calculation unit 14d, the second command generation unit 14e, and the position / speed control unit 14f of the robot control device 14 are as follows. To work. The first command generation unit 14a generates a joint angular velocity command ⁇ TASK for suppressing kinetic energy using the equations (2) and (3). That is, the weighted Jacobian inverse matrix is obtained using the weight matrix W, and the joint angular velocity command ⁇ TASK is obtained by multiplying the obtained weighted Jacobian inverse matrix by the speed command of the control point.
  • the weight matrix W is a weight matrix.
  • the weight matrix W is, for example, 1) an inertia matrix (weight matrix W1) obtained by dynamic calculation, 2) a constant matrix set based on the load weight and load inertia of the drive target such as the link LNK (see FIG. 1). (Weight matrix W2) or 3) Constant matrix (weight matrix W3) set based on the maximum torque and maximum speed of each joint axis Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, LJ1 to LJ7 .
  • the weight matrix W1 is an inertia matrix of the joint coordinate system.
  • the robot control basic theory written by Tsuneo Yoshikawa, Corona, 1988
  • the weight matrices W2 and W3 are set as follows, respectively.
  • the weight matrices W2 and W3 are not limited to the following setting examples.
  • the weight of the driving target of the joint axes Vx, Vy, and Vz of the robot 12 is 10 kg
  • the weight of the driving target of the joint axes L1 to L3 is 5 kg
  • the joint axes RJ1 to RJ7
  • the diagonal term is set to a diagonal matrix that is a value obtained by multiplying the weight of each link by a positive constant K2. That is, the weight matrix W2 is set as follows, for example.
  • W2 diag [10 ⁇ K2, 10 ⁇ K2, 10 ⁇ K2, 5 ⁇ K2, 5 ⁇ K2, 5 ⁇ K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2, K2]
  • the maximum translational thrusts of the joint axes Vx and Vy of the robot 12 are each 10 N
  • the maximum torques of the joint axes Vz and L1 to L3 are 10 Nm, 5 Nm, 5 Nm, and 5 Nm, respectively.
  • the diagonal term is the maximum translational thrust or maximum torque of each joint axis Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, LJ1 to LJ7 It is set to a diagonal matrix that is a value multiplied by a positive constant K3. That is, the weight matrix W3 is set as follows, for example.
  • W3 diag [10 ⁇ K3, 10 ⁇ K3, 10 ⁇ K3, 5 ⁇ K3, 5 ⁇ K3 5 ⁇ K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3, K3 K3, K3, K3]
  • the weight matrix W1 is determined.
  • weight matrix W By using the weight matrix W, it is possible to give an angular velocity command according to the load weight and load inertia, and to suppress the necessary kinetic energy. Specifically, a smaller angular velocity command is issued for a joint axis (for example, joint axes Vx, Vy, Vz, L1 to L3, LJ1, LJ2, RJ1, RJ2) having a large load weight or load inertia of a link or the like to be driven.
  • a weight matrix W for giving a larger angular velocity command to a joint axis for example, joint axes LJ3 to LJ7, RJ3 to RJ7) having a smaller load weight or load inertia to be driven is given.
  • Step SA2 When the signal A is input, the first optimization operation command generation unit 14b calculates a joint angular velocity command ⁇ ELB for the robot 12 to perform the arm lowering operation.
  • the joint angular velocity command ⁇ ELB can be calculated by at least the following three methods. These methods include, for example, 1) a calculation method A1 that uses the gradient of the norm of gravity torque, 1) a calculation method A2 that uses the gradient of potential energy, or 3) a calculation method A3 that positions the arm in a lowered posture. is there. Which calculation method is used is determined according to the situation such as the calculation capability of the robot controller to be used and the dynamic calculation parameters of the target robot.
  • Which of the left arm LA, the right arm RA, or both the arms LA, RA is to be subjected to the arm lowering operation is determined by the location to be controlled. For example, in a certain teaching process, when the control target location is the right arm RA, the arm lowering operation of only the right arm RA is executed.
  • a joint angular velocity command ⁇ ELB for moving each joint axis in a direction to reduce the norm of the gravitational torque is calculated using the gradient of the norm of the gravitational torque.
  • the norm of the gravity torque for example, the Euclidean norm or the maximum norm of the gravity torque (torque applied to each joint axis by gravity) ⁇ GRV of each joint is used.
  • K is a gain that restricts the control of the arm lowering operation. If the gain K is set to a large value, the operation of the robot 12 can be speeded up. However, the gain K is set to a value suppressed to such an extent that the robot 12 does not oscillate.
  • the first optimization operation command generation unit 14b suppresses the posture maintenance torque (torque against the gravity torque).
  • a posture can be generated.
  • the posture P1 in which the body 12c of the robot 12 stands upright and the posture P2 in which the head 12d (see FIG. 2A) faces directly below in the vertical direction can reduce the posture holding torque.
  • the posture P2 in which the head 12d faces directly below the vertical direction cannot be taken.
  • the posture holding torque is selected for which of the two postures to suppress the posture holding torque. That is, for joint axes that have two forms of postures that suppress posture holding torque, such as the hip joint shaft, considering the joint range of motion, either the positive or negative direction of the joint axis It is determined whether to reduce the posture holding torque in one direction.
  • * is a subscript for specifying which link among a plurality of links constituting both arms LA and RA.
  • M represents the mass of each link of the robot 12.
  • g represents gravitational acceleration.
  • Cr T * is a homogeneous transformation matrix up to the * th link (a link specified from a plurality of links) in the robot coordinate Cr.
  • lg * is the barycentric position of the * link on the * link coordinates (link coordinates of links specified from a plurality of links).
  • the redundant axis By operating the redundant axis using the gradient of potential energy P, a posture that suppresses potential energy can be generated.
  • the posture in which the head 12d faces downward in the vertical direction has the smallest potential energy.
  • the calculation method A2 is not applied to the hip joint axes L1 and L2, and the body 12c is held in an upright posture.
  • this calculation method A2 is applied to joints other than the hip joint axes L1 and L2, and the posture holding torque is suppressed.
  • the joint angular velocity command ⁇ ELB is calculated using a difference between the current angle ⁇ CUR of the joint axis and the optimum angle ⁇ OPT (vector) of the joint axis based on the following equation (4b). .
  • the optimal joint axis angle ⁇ OPT is 1) that the right arm RA or left arm LA of the robot 12 projects to the side of the elbow, and 2) that the right arm RA or left arm LA interferes with the torso, It is a value determined by the operator empirically in consideration of 3) the movable range of each joint axis and 4) the specific posture of the robot 12. This calculation method A3 is applied only when the optimum joint axis angle ⁇ OPT is explicitly determined.
  • ⁇ CUR either the current value ⁇ fb of the joint angle calculated from the encoder of the servo motor that drives the robot 12 or the command value ⁇ ref of the joint angle calculated on software is used.
  • the current value ⁇ fb of the joint angle and the command value ⁇ ref of the joint angle are values before one control cycle.
  • the robot 12 can perform the arm lowering operation (see FIG. 5A). All of the calculation methods A1 to A3 of the joint angular velocity command ⁇ ELB described above are applied only to the joint related to the arm lowering operation.
  • the second command generation unit 14e can perform the arm lowering operation for the left arm LA by calculating the joint angular velocity command ⁇ ELB related to the joint axis LJ2. Further, for example, the second command generation unit 14e can perform the arm lowering operation for the right arm RA by calculating the joint angular velocity command ⁇ ELB regarding the joint axes RJ1 and RJ2.
  • the arms LA and RA may be moved to a position higher than the arm position that moves based on the joint angular velocity command ⁇ ELB .
  • the movement amount at that time is preferably a movement amount that does not become an obstacle in securing the work space or does not become an obstacle in reducing the intimidation.
  • Step SA3 the zero-space matrix calculation unit 14d determines the zero space of the Jacobian matrix J related to the hand of the robot 12 based on the joint angle ⁇ CUR of the robot 12 (the current value ⁇ fb of the joint angle or the command value ⁇ ref of the joint angle).
  • the matrix N is calculated.
  • Step SA4 The second command generation unit 14e calculates the joint angular velocity command ⁇ ref2 based on the joint angular velocity command ⁇ ELB and the zero space matrix N as shown in the following equation (5).
  • Step SA5 The joint angular velocity command ⁇ ref3 is calculated based on the following equation (6).
  • ⁇ ref3 ⁇ TASK + ⁇ ref2 Equation (6)
  • the position / speed control unit 14f executes feedback control based on the command value ⁇ ref of the joint angle and the current value ⁇ fb of the joint angle.
  • the position / speed controller 14f controls each servo motor and controls each joint axis Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, and LJ1 to LJ7 of the robot 12. As a result, the robot 12 performs an arm lowering operation.
  • FIG. 6A is a diagram showing a comparison between the posture of the robot 12 in the operation mode A (lower right diagram) and the posture of the robot 12 when the operation mode A is not selected (lower left diagram).
  • the joint shaft with a small load weight or load inertia of the drive target moves more greatly.
  • the joint shaft having a large load weight and load inertia does not move much.
  • the robot 12 is controlled by steps SA1 to SA5 so as to reduce the necessary kinetic energy.
  • Step SB1 Subsequently, the operator causes the robot 12 to hold the posture holding the article OBJ for a certain period of time.
  • the robot 12 keeps the position and orientation of the hand of the robot 12 (while holding the article OBJ), and the robot 12 has the joint axes Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, LJ1 to
  • the operation of the robot 12 is controlled in the operation mode B so as to take a posture to suppress the torque of the LJ7.
  • the load torque generally increases as the joint shafts of the left arm LA and the right arm RA are closer to the body 12c.
  • the robot 12 is moved closer to the table 40 by the movable carriage 12a, and the posture is maintained as a result of taking the posture in which the left arm LA and the right arm RA are bent using the redundancy degree of freedom while maintaining the position and posture of the hand. Torque is suppressed.
  • the load torque of the joint axes L1 and L2 on the movable carriage 12a side increases. Therefore, the posture holding torque is suppressed as a result of taking the posture in which the waist 12b is extended using the redundancy degree of freedom while maintaining the position and posture of the hand.
  • the signal A is not input to the first optimization operation command generation unit 14b shown in FIG. 3, and the signal B is input to the second optimization operation command generation unit 14c instead. Since the signal A is not input, the first optimization operation command generation unit 14b does not operate.
  • the hand speed command V is input to the first command generation unit 14a shown in FIG.
  • the first command generation unit 14a, the second optimization operation command generation unit 14c, the zero space matrix calculation unit 14d, the second command generation unit 14e, and the position / speed control unit 14f of the robot control device 14 are as follows. To work.
  • the first command generation unit 14a generates a joint angular velocity command ⁇ TASK for suppressing kinetic energy using the aforementioned equations (2) and (3).
  • ⁇ TASK joint angular velocity command
  • Step SB2 When the signal B is input, the second optimization operation command generation unit 14c calculates the joint angular velocity command ⁇ GRV .
  • the joint angular velocity command ⁇ GRV can be calculated by at least the following three methods. These methods include, for example, 1) a calculation method B1 that uses the gradient of the norm of gravitational torque, 1) a calculation method B2 that uses the gradient of potential energy, or 3) a calculation that positions to a posture that suppresses posture holding torque. There is method B3.
  • a joint angular velocity command ⁇ GRV for moving each joint axis in the direction of decreasing the norm of the gravitational torque is calculated using the gradient of the norm of the gravitational torque.
  • the norm of the gravity torque for example, the gravity torque of each joint (torque applied to each joint axis by gravity) ⁇ GRV or the maximum norm is used.
  • the joint angular velocity command ⁇ GRV is calculated using the difference between the current angle ⁇ CUR of the joint axis and the optimum angle ⁇ OPT of the joint axis.
  • the optimal joint axis angle ⁇ OPT is, for example, a posture in which the robot 12 stands upright and the right arm RA and the left arm LA extend downward in the vertical direction. This posture is a posture that suppresses the posture holding torque of the robot 12 from a static viewpoint.
  • Step SB3 The zero space matrix calculation unit 14d calculates the zero space matrix N of the Jacobian matrix J related to the hand of the robot 12 based on the joint angle ⁇ CUR of the robot 12 as shown in the above-described equation (1).
  • Step SB4 The second command generation unit 14e calculates the joint angular velocity command ⁇ ref2 based on the joint angular velocity command ⁇ GRV and the zero space matrix N as shown in the following equation (8).
  • Step SB5 The joint angular velocity command ⁇ ref3 is calculated based on the above equation (6).
  • the position / speed control unit 14f performs feedback control based on the command value ⁇ ref of the joint angle and the current value ⁇ fb of the joint angle.
  • the position / speed controller 14f controls each servo motor and controls each joint axis Vx, Vy, Vz, L1 to L3, RJ1 to RJ7, and LJ1 to LJ7 of the robot 12. As a result, the robot 12 suppresses the torque necessary for maintaining the posture.
  • FIG. 6B compares the posture of the robot 12 in the state where the operation in the operation mode A is completed (the state in which the execution of step SA5 is completed) (upper diagram) and the posture of the robot 12 in the operation mode B (lower diagram).
  • FIG. In this operation mode B the posture of the robot 12 is the posture in which the operation in the operation mode A is completed (refer to the upper diagram in FIG. 6B), and the left arm LA and the right arm RA are bent and the waist 12b is extended (in FIG. 6B). (See the figure below). That is, the posture of the robot 12 is held in a state where the torque necessary for holding the posture is suppressed.
  • steps SB1 to SB5 do not necessarily have to be executed. If the robot 12 does not hold the posture of gripping the article OBJ, the process proceeds from step SA5 to the next step SC. For example, when the robot 12 operates continuously, such as an operation of gripping the article OBJ, moving the article OBJ, and releasing the gripped article OBJ, steps SB1 to SB5 are not executed. It is possible to reduce energy consumption by executing a predetermined next operation. However, when the robot 12 holds the article OBJ with the right arm RA and then works with the left arm LA while holding the position of the right arm RA, or when performing vision recognition processing or voice communication processing, By executing steps SB1 to SB5 to suppress the holding torque, energy consumption can be suppressed.
  • Step SC As shown in FIG. 4, when the teaching is completed, the operation of the robot system 10 ends. If the teaching has not been completed, step SA1 is executed.
  • the work program can be taught by repeatedly executing Steps SA1 to SA5, Steps SB1 to SB5, and Step SC.
  • the taught work program is reproduced by pressing a reproduction button provided on the teaching pendant 16 (see FIG. 1).
  • the series of steps SA1 to SA5, steps SB1 to SB5, and step SC may be executed in parallel or in a different order, if possible.
  • the robot system 10 According to the present embodiment, it is possible to automatically generate the teaching position and the motion trajectory that reduce the energy consumption of the redundant robot.
  • the present invention is not limited to the above-described embodiments, and modifications can be made without changing the gist of the present invention.
  • a case where the invention is configured by combining some or all of the above-described embodiments and modifications is also included in the technical scope of the present invention.
  • the robot controller 14 may be provided with only one of the first optimization operation command generation unit 14b and the second optimization operation command generation unit 14c. That is, it is possible to control the robot 12 so as to perform the arm lowering operation by providing only the first optimization operation command generation unit 14b, or provide only the second optimization operation command generation unit 14c.
  • the robot 12 can also be controlled so as to suppress the holding torque. That is, the robot 12 may be controlled by only one of the operation modes A and B.
  • Each button is not limited to a hardware button.
  • the button may be a button realized by software.
  • the teaching device is not limited to the teaching pendant.
  • Another example of the teaching device is a remote operation device having a communication function with the robot control device 14 such as a personal computer.
  • the present invention can be applied not only to the case of using an actual robot as shown in the above-described embodiments, but also to the case of using a robot simulation model. That is, it can also be used to confirm the validity of a work program using an offline simulator.

Abstract

L'invention porte sur un système de robot (10), un dispositif de commande de robot (14) et un procédé de commande de système de robot (10) apte à générer automatiquement des positions d'apprentissage et des trajectoires de mouvement qui réduisent l'énergie consommée par un robot redondant. Le système de robot (10) comprend un robot (12) ayant de multiples axes d'articulation, qui comprennent un axe redondant, et un dispositif de commande de robot (14). Le dispositif de commande de robot (14) comprend : une première unité génératrice de commande (14a) destinée à générer une commande de vitesse angulaire pour une première articulation du robot (12) de façon à réduire l'énergie de mouvement ; une première unité génératrice de commande d'action optimisée (14b), qui génère une commande de vitesse angulaire d'articulation (A) pour que le robot (12) exécute un mouvement d'abaissement de bras ; une unité de calcul de matrice à espace zéro (14d) destinée à calculer une matrice à espace zéro d'une matrice de Jacobi par rapport à un point de commande ; une seconde unité génératrice de commande (14e) destinée à générer une commande de vitesse angulaire pour une seconde articulation du robot (12) sur la base de la commande de vitesse angulaire d'articulation (A) et de la matrice à espace zéro ; et une unité de commande (14f) destinée à commander les servomoteurs.
PCT/JP2012/083299 2012-06-04 2012-12-21 Système de robot, dispositif de commande de robot et procédé de commande de système de robot WO2013183190A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2012064391 2012-06-04
JPPCT/JP2012/064391 2012-06-04

Publications (1)

Publication Number Publication Date
WO2013183190A1 true WO2013183190A1 (fr) 2013-12-12

Family

ID=49711601

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2012/083299 WO2013183190A1 (fr) 2012-06-04 2012-12-21 Système de robot, dispositif de commande de robot et procédé de commande de système de robot

Country Status (1)

Country Link
WO (1) WO2013183190A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106335064A (zh) * 2016-11-29 2017-01-18 合肥工业大学 一种柔性关节机器人系统的控制器设计方法
CN107901041A (zh) * 2017-12-15 2018-04-13 中南大学 一种基于图像混合矩的机器人视觉伺服控制方法
CN110883770A (zh) * 2019-05-08 2020-03-17 丁亚东 一种基于位置和图像的机器人混合视觉伺服控制方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000354988A (ja) * 1999-06-14 2000-12-26 Matsushita Electric Ind Co Ltd ロボットの制御方法と制御装置
JP2006334774A (ja) * 2005-05-31 2006-12-14 Honda Research Inst Europe Gmbh エフェクタの軌道を制御するための方法
JP2009521751A (ja) * 2005-12-22 2009-06-04 本田技研工業株式会社 多関節システムの姿勢の再構築、リターゲット、軌道追従及び推定

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000354988A (ja) * 1999-06-14 2000-12-26 Matsushita Electric Ind Co Ltd ロボットの制御方法と制御装置
JP2006334774A (ja) * 2005-05-31 2006-12-14 Honda Research Inst Europe Gmbh エフェクタの軌道を制御するための方法
JP2009521751A (ja) * 2005-12-22 2009-06-04 本田技研工業株式会社 多関節システムの姿勢の再構築、リターゲット、軌道追従及び推定

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
"Jacobian o Mochiita Gyaku Undogaku", TMPSWIKI, 1 February 2008 (2008-02-01) *
MINORU ASADA, ROBOTICSHANDBOOK-2ND ED., CORONA PUBLISHING CO.,LTD., GORAI,, pages 285 - 287 *
SHUGEN MA: "Efficient Redundancy Control of Redundant Manipulators", JOURNAL OF THE ROBOTICS SOCIETY OF JAPAN, vol. 14, no. 5 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106335064A (zh) * 2016-11-29 2017-01-18 合肥工业大学 一种柔性关节机器人系统的控制器设计方法
CN107901041A (zh) * 2017-12-15 2018-04-13 中南大学 一种基于图像混合矩的机器人视觉伺服控制方法
CN110883770A (zh) * 2019-05-08 2020-03-17 丁亚东 一种基于位置和图像的机器人混合视觉伺服控制方法

Similar Documents

Publication Publication Date Title
US9002519B2 (en) Robot control method, robot control device, and robot control system
JP4896276B2 (ja) ロボット、ロボットの制御装置、制御方法、及び制御プログラム
Williams et al. Planar translational cable‐direct‐driven robots
JP6083145B2 (ja) ロボットの制御装置、およびロボット
JP6255901B2 (ja) ロボット制御装置、ロボットおよびロボットシステム
JP5916583B2 (ja) 多関節ロボットのウィービング制御装置
KR20180059888A (ko) 로봇교시방법 및 로봇 암 제어장치
JP5144035B2 (ja) 産業用ロボットのツールの位置・姿勢制御方法及び制御システム
KR20180069031A (ko) 로봇의 다이렉트 교시방법
Scimmi et al. Implementing a vision-based collision avoidance algorithm on a UR3 Robot
WO2013183190A1 (fr) Système de robot, dispositif de commande de robot et procédé de commande de système de robot
You et al. Semi-autonomous bilateral teleoperation of hexapod robot based on haptic force feedback
Bergamasco et al. Exoskeletons as man-machine interface systems for teleoperation and interaction in virtual environments
Muscolo et al. A comparison between two force-position controllers with gravity compensation simulated on a humanoid arm
Yang et al. Dynamic compensation control of flexible macro–micro manipulator systems
JPH08161015A (ja) 多関節型ロボットの駆動制御方法
Nahavandi et al. Automated robotic grinding by low-powered manipulator
Sekimoto et al. Skilled-motion plannings of multi-body systems based upon Riemannian distance
WO2020149020A1 (fr) Dispositif de commande de robot, procédé de commande de robot et programme de commande de robot
JP2016022559A (ja) 多関節ロボットアームの重力補償方法
Zhou et al. Neural network-based variable stiffness impedance control for internal/external forces tracking of dual-arm manipulators under uncertainties
JP4647919B2 (ja) 制御方法および制御装置
Yoon et al. Adaptive control for cable driven parallel robots
JPH11333771A (ja) 冗長運動機構の制御方法およびその制御装置並びに冗長運動機構および２自由度並進駆動機構
Liu et al. Mechanisms and design of a humanoid robot for two-handed manipulation

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12878396

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12878396

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP