JPH1011121A - Robot control device and its controlling method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a robot control device and its controlling method capable of reducing power consumption by reducing holding torque while holding waiting posture at teached posture when the operation of a robot is stopped. SOLUTION: In the case of holding a coating robot at fixed waiting posture on a certain fixed teaching point, minimum motor torque necessary for holding respective connection parts (1st to 6th joints) between 1st to 6th motors and a revolution base 7, a 1st arm 9, a 2nd arm 11 and 1st to 3rd wrist units at fixed rotational angles is found out and rotation speed commands θr and torque commands τr are outputted to respective motors so as to hold the waiting posture by the minimum motor torque. In addition, the motor torque on the waiting posture is reduced by driving each joint only by a slight angle at a prescribed rotational speed or slightly changing the objective angle of each joint.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a robot control apparatus and a control method suitable for use in a teaching-reproduction type industrial robot.

[0002]

2. Description of the Related Art In conventional industrial robot control,
As a method of reducing power consumption, a method of converting thermal energy generated when the operation of a robot is stopped into electric power has been proposed. Further, Japanese Patent Application Laid-Open No. 5-138562 discloses a robot control method for reducing power consumption by setting the standby posture of the robot to a posture having a small gravitational torque in a stationary state.

[0003]

However, in the above-described heat energy conversion apparatus, a generator must be provided in each of the motor amplifiers in order to actively generate electric power. For this reason, there has been a problem that the size of the controller is increased and the cost is increased.

On the other hand, in the robot control method for changing the standby posture, since the standby posture is set to a posture having a small gravitational torque, it may interfere with the environment of the work, its supply device, or other peripheral devices. Therefore, in such a case, the method disclosed in JP-A-5-138562 cannot be applied. Further, such a method has a problem that the robot may take a posture different from the posture taught by the instructor.

Further, the reduction gear used for general industrial robots has a problem that the starting torque becomes extremely large due to frictional resistance at the start of rotation.

The present invention has been made in view of such circumstances, and uses the frictional resistance of the speed reducer in reverse to maintain the standby posture when the operation of the robot is stopped in the taught posture while holding the robot at that time. An object of the present invention is to provide a robot control device and a control method thereof that can reduce torque, reduce power consumption, and reduce running cost.

[0007]

According to the first aspect of the present invention,
In a robot control device that controls a robot that operates a movable part by torque of a motor transmitted through a speed reducer based on teaching data, a first command signal based on the teaching data is output to the robot. First command signal output means, and during the operation of the robot based on the first command signal, judgment processing means for judging whether or not the operation of the robot is in a standby state in which a constant standby posture is maintained; A second command to decrease the torque of the motor in a range where the movable portion is kept stationary by the torque and the friction torque of the speed reducer when it is determined by the determination processing means that the motor is in the standby state; And second command signal output means for outputting a signal.

According to a second aspect of the present invention, there is provided a robot control method using the robot control device according to the first aspect, wherein a first step of moving the robot to a teaching point of the teaching data, and When you move to
A second step of determining whether the teaching point is at a position to take a standby posture; and, when it is determined that the teaching point is at a position to take a standby posture in the second step, the torque of the motor is increased in a certain direction. A third step of outputting a third command signal that changes by a small amount, and a fourth command signal that stores the third command signal immediately before the movable portion starts operating from a stationary state by the third command signal. A fifth step of outputting a fourth command signal for changing the torque of the motor by a small amount in a direction opposite to the constant direction in the third step, and a step of outputting the fourth command signal. A sixth step of storing the fourth command signal immediately before the part operates from the stationary state, and the fourth and fifth command signals stored in the fourth and sixth steps. Indicate less torque It is characterized by controlling the robot a command signal based on the seventh second command signal determined by a process of which the second instruction signal in the teaching point.

According to a third aspect of the present invention, there is provided the robot control method according to the second aspect, wherein the robot is moved to each of the teaching points, and the second to seventh steps are repeated to hold all the teachings in a standby posture. Determining and storing the second command signal at a point, the teaching data and the stored second command signal;
The robot is controlled based on the command signal of (1).

[0010] The invention according to claim 4 is the first and second inventions.
2. The robot control method according to claim 1, wherein the command signal is a speed command signal for instructing an operation speed of the movable portion with respect to the motor, wherein the determination processing unit determines that the motor is in a standby state. When
A tenth step of obtaining a torque of the motor required to maintain the movable part in a stationary state based on an operation amount of the movable part, and the necessary torque of the motor obtained in the tenth step is zero. In this case, the second command signal is output to instruct the operating speed of the movable portion in the operation in the minute range, in which the movable portion is operated in the minute range and the torque of the motor in the stationary state is set to 0. An eleventh step of operating the movable part in the minute range, and if the required torque of the motor obtained in the tenth step is not zero, the torque of the motor decreases and the movable part And a twelfth step of changing the second command signal by a small amount until immediately before the operation of the second command signal.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION

<Structure> Overall Configuration An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration of a painting robot system to which a robot control device according to an embodiment of the present invention is applied.

In FIG. 1, reference numeral 1 denotes a controller, which receives instruction data from the teaching pendant 2 and exchanges operation signals and currents with the manipulator 4 via the cable 3, thereby controlling the manipulator 4. The controller 1 has a storage device for storing a control program for the manipulator 4, teaching data, various control parameters, and the like, and an arithmetic processing device for performing various arithmetic processes based on the programs and the instruction data. It supplies various operation command signals and current to the manipulator 4 based on the result of the arithmetic processing by the arithmetic processing device. It should be noted that the specific contents of the arithmetic processing here will be clarified in the operation description to be described later.

The teaching pendant 2 is an input device for inputting instruction data for an operator to instruct the operation of the manipulator 4, and outputs the input instruction data to the controller 1. The cable 3 is a member that connects the controller 1 and the manipulator 4, and transmits the operation signal and the current.

The manipulator 4 is a main part of the painting robot in the present painting robot system, and is attached to a predetermined position on the floor surface by a fixed base 5 fixed to the floor surface. Showing swivel base,
It is composed of components such as a motor and an arm.
Hereinafter, each of these components will be described. In this figure, the speed reducer provided at the connection portion (joint portion of the manipulator 4) of the output shaft of each motor is omitted, and the details of the structure of the connection portion provided with the speed reducer will be described later. I do.

Reference numeral 6 denotes a first motor mounted on the fixed base 5, the output shaft of which is connected to a turning base 7 via a speed reducer.
Is linked to The first motor 6 is driven based on an operation command signal supplied from the controller 1, and rotates the turning base 7 in the direction of the arrow θ1 in the figure.

A second motor 8 is attached to an upper portion of the turning base 7 in the figure, and an output shaft of the second motor 8 is connected to one end of a first arm 9 via a speed reducer. I have. Then, the second motor 8 is driven based on an operation command signal supplied from the controller 1 so that the first arm 9 rotates around the rotation axis of the second motor 8 in the direction of the arrow θ2 in the drawing. Has become.

On the other hand, the third end of the first arm 9
The motor 10 is attached, and the third motor 10
Is connected to the second arm 11 via a speed reducer. Then, the third motor 10 is driven based on an operation command signal supplied from the controller 1, and the second arm 11 is driven by an arrow θ3 in FIG.
It rotates in the direction.

The second arm 11 has the third motor 10
A wrist motor case 12 is attached to a portion connected to the rotation shaft of the wrist, and a first wrist unit 13 and a second wrist unit
A wrist unit 14 and a third wrist unit 15 are attached.

Here, fourth, fifth and sixth motors (not shown) are housed in the wrist motor case 12.
A transmission mechanism (not shown) for transmitting each driving force of the fourth, fifth, and sixth motors to the first wrist unit 13, the second wrist unit 14, and the third wrist unit 15, respectively, is provided inside the second arm 11. Is provided. Also, the first wrist unit 13, the second wrist unit 14, the third wrist unit 15
Are sequentially attached via connecting shafts rotatable in directions of arrows θ4, θ5, and θ6 in the drawing. And
Similarly to the above, the fourth, fifth, and sixth motors are driven based on the operation command signal supplied from the controller 1, whereby the first wrist unit 13, the second wrist unit 14,
The third wrist unit 15 rotates in the θ4, θ5, and θ6 directions, respectively.

Reference numeral 16 denotes a bracket attached to the third wrist unit 15, and a coating gun 17 is attached via the bracket 16. The painting gun 17 is a painting tool that sprays paint in a certain direction and with a certain spread, and includes the first and second arms and the first, second, and third arms.
The paint is sprayed at a predetermined position and a predetermined direction based on the rotation angles θ1 to θ6 of the wrist unit to paint a work (not shown).

As described above, in the present painting robot system, the painting is performed by the rotation angles θ1 to θ6 of the first arm 9, the second arm 11, the first wrist unit 13, the second wrist unit 14, and the third wrist unit 15. The position (x) of the gun 17 and the posture (R) of the manipulator 4 are determined. The position, posture, rotation angle, and the like are determined by arithmetic processing in the controller 1, and are controlled by operation command signals to the first to sixth motors based on the determination.

B. Structure of Joint Part Here, the connection part of the motor output shaft provided with the speed reducer, that is, the structure of the joint part of the manipulator 4 will be described. FIG. 2 shows the structure of the joint between the second motor 8 and the first arm 9 which rotates in the θ2 direction as an example of the joint. In this figure, the manipulator 4 is indicated by an arrow A in FIG.
It is equivalent to one seen from the direction.

In FIG. 2, reference numeral 18 denotes an output shaft of the second motor 8. Reference numeral 19 denotes a speed reducer attached to the output shaft 18. The speed reducer 19 transmits the amplified torque (for example, 100 times) to the first arm 9 by reducing the rotational speed (angular speed) of the output shaft 18 at a predetermined ratio (for example, reducing it to 1/100), and As a result, the rotation angle θ2 of the first arm 9 is displaced by a predetermined amount. In addition, the manipulator 4
Reduction gears are similarly provided for other joint portions (joint portions rotating in the directions θ1, θ3 to θ6).

<Holding Torque> First, the starting torque and running torque of the speed reducer will be described with reference to FIG. In this figure, θ 'on the horizontal axis represents the rotational speed transmitted through the speed reducer, and τ on the vertical axis represents the motor torque required at that time.

As shown in this graph, the motor torque (holding torque) at the transmission rotation speed θ ′ = 0 (at rest) may be any value between + τa and −τa. here,
τa is the magnitude of the friction torque due to the frictional resistance of the speed reducer. When θ ′ exceeds a predetermined range + θ′c to −θ′c, the running torque τ becomes

(Equation 1) It can be expressed as.

Next, the holding torque of the arm and the wrist unit at the joint of the manipulator 4 will be described. Here, as shown in FIG. 4, in a state where the first arm 9 and the second arm 11 form 90 ° (θ3 = 90 °), the rotation angle θ2 of the first arm 9 becomes −20 ° (dashed line in the figure). Ι) to + 90 ° (dashed-dotted line Π), the holding torque of the second motor 8 will be described as an example.
Here, the rotation angle from the vertical direction in the figure to the left is +, and the rotation angle to the right is-.

In the drawing, reference numeral 20 denotes a joint of a connecting portion between the first motor 6 and the turning base 7, and reference numeral 21 denotes a second motor 8 and the first motor 6.
Reference numeral 22 denotes a joint of a connecting portion of the arm 9, and reference numeral 22 denotes a joint of a connecting portion of the third motor 10 and the second arm 11. In the following, for the sake of convenience, these joints are firstly assigned according to the subscripts attached to the respective rotation angles θ1, θ2, θ3.
The joint, the second joint, and the third joint will be referred to as joints. Also, the connection portions of the fourth to sixth motors and the first to third wrist units are similarly referred to as fourth joint, fifth joint, and sixth joint.

Reference numeral 23 denotes the center of mass of the arm between the second and third joints. Here, the distance between the center of mass 23 and the second joint is g2, the mass of this arm is m2, and the length is l2. 24 represents the center of mass of the arm between the third and fourth joints. Here, the position of this center of mass 24 in the x and z directions is xm3 and zm3, respectively, and the distance between the third joint is g.
The arm has a mass of m3 and a length of l3. 2
5 represents the center of mass of the arm in front of the fourth joint. The position of the center of mass 25 in the x and z directions is xm4 and z, respectively.
m4.

When each distance, mass and the like are set in this way,
The gravitational torque τg required to maintain the attitude of the manipulator 4 is

(Equation 2) It can be expressed as. Here,

(Equation 3) And G = −9.8 m / s ² (gravitational acceleration).

A graph of the gravitational torque τg of Equation 2 is shown by a solid line in FIG. When the above-described friction torques + τa and −τa (see FIG. 3) are added to this solid line, a torque curve shown by a dotted line in FIG. 5 is obtained. That is, the area surrounded by these two dotted lines and is the holding torque required for the second motor 8 to stop the first arm 9 and hold the attitude of the manipulator 4. The minimum value of the holding torque is a curve as shown by a solid line.

Here, the current required for the operation of the manipulator by the motor and for maintaining the attitude is proportional to the torque τ, so that the smaller the absolute value or the effective value of the torque τ, the lower the power consumption. Therefore, when the manipulator 4 is at rest, driving the second motor 8 with the torque on the solid line described above minimizes power consumption.

Here, the holding torque of the second motor 8 when the rotation angles θ2 and θ3 are within the predetermined range has been described. However, the same rotation angle range and other motors have the same geometrical shape as described above. The minimum necessary holding torque at rest can be determined by the chemical analysis.

<Operation> A. Operation Teaching Data and Target Position Next, the operation of the present painting robot system will be described. First, the teaching data of the robot, which is the premise of the operation description, will be described. Normally, the teaching data of the robot is determined by a turning point or a representative point of the trajectory, an interpolation condition between a plurality of continuous teaching points, and a moving speed. For example,
In the trajectory in FIG. 6, points indicated by P0, P1, P2,... Are teaching points. In the present painting robot system, the teaching pendant 2
And stores it in the storage device of the controller 1 as a teaching point.

In the actual operation of the painting robot, the operation control is performed by calculating the target positions indicated by k, k + 1,... In FIG. Will be In the discrete-time control system, the unit time refers to a sampling time at which a signal such as a current position and a command signal are transmitted and received. In the present coating robot system, the processing time of the controller 1 is 5 ms. The target position is calculated for each sampling time.

In the following, the counter of the teaching point is i (i = 0, 1, 2,..., N) and the counter of the target position is k (k = 0, 1, 2,..., N). Teaching points Pi and Pi + 1
Is determined by the arithmetic processing unit of the controller 1 every 5 ms. Note that the maximum value N of the counter k is represented by T, the moving time from the teaching point Pi to Pi + 1.
Assuming that the sampling time is Ts, N = T / Ts. This corresponds to the total number of target positions that divide between teaching points (hereinafter, referred to as the total number of divisions).

B. First Operation Mode Next, an operation by a method of actually reducing the holding torque of the motor in the present painting robot system will be described. First, as the first operation mode, the controller 1 sends the current position rotation angle θ of each joint from each of the first to sixth motors.
f (θf = [θf1, θf2,..., θf6] ^T ) and the torque of each motor τf (τf = [τf1, τf2,..., τf6] ^T ), and the controller 1 sends the first to sixth signals. The rotation speed command θr '(θr' = [θr1 ', θr2',
.., Θr6 ′] ^T ) and torque command τr (τr = [τr1, τr
2,..., Τr6] ^T ) will be described.

(1) Mode of Operation Control FIG. 7 shows a control block diagram of the painting robot in the first operation mode. As shown in this figure, in this operation mode, the target position rotation angle θr (θr = [θr1, θr2,
.., Θr6] ^T ) and the current position rotation angle θf (θf = [θf1,
θf2,..., θf6] ^T ) with the target torque command τr (τr =
[Τr1, τr2, ..., τr6] ^T ) is the current motor torque τf
Are respectively fed back, and an operation command signal and a current value to each motor are calculated as follows.

First, based on the deviation of the current position rotation angle θf from the target position rotation angle θr, the position compensator Gp (z) calculates a rotation speed command θr ′ that holds each joint at the target position rotation angle θr. I do. The current value to be supplied to each motor to rotate each joint at the rotation speed θr ′ is calculated by the speed compensator Gv (z) based on the deviation of the current rotation speed θf ′ from the rotation speed command θr ′. And outputs the result to a control object including a motor and an arm.

On the other hand, the torque compensator in the figure calculates a current command based on the deviation of the current motor torque τf from the target torque τr as required, and outputs the current command to each motor. In addition, in the integral calculator z / z-1, the rotation speed θf ′ from each motor is used.
(Θf ′ = [θf1 ′, θf2 ′,..., Θf6 ′] ^T ) is integrated and fed back as the current position rotation angle θf. In this operation mode, the operation of the painting robot is controlled in this way.

(2) Calculation of Target Holding Torque Here, a processing method for calculating the target torque τr in the standby posture at the time of standby for the teaching data will be described. Here, the standby posture refers to maintaining a constant posture for a predetermined time, and is a state in which the position feedback value does not change during the predetermined time. Hereinafter, the minimum target torque required to hold the standby posture is referred to as a target holding torque, and is denoted by τgr (τgr = [τgr1, τgr2,..., Τgr6] ^T ).
Expressed by The processing for obtaining the target holding torque τgr is performed based on the teaching data given by the instructor while controlling the movement between the teaching points and the standby posture during standby as described below by the above-described control mode.

FIG. 8 shows the procedure of a method for calculating the target holding torque τgr. As shown in the figure, first, the arithmetic processing unit of the controller 1 reads the teaching data in the storage device (step S1), and determines the total number n of teaching points (step S2). Each teaching point Pi (i = 0, 1, 2,..., N) from the initial position (P0) to the final teaching point (Pn)
Are sequentially processed according to the procedure shown in step S3 and subsequent steps.

In step S3, the painting robot is moved to the position of the teaching point Pi and the posture at the position. This movement is performed according to the above-described control mode by calculating the target position xr at every sampling time of 5 ms. For details, the determined operation control after the calculation of the target holding torque τgr ((3) described later)
Will be described. If there is no change in the posture of the painting robot at the teaching point Pi for a certain period of time, step S
In step 4, it is determined that the position is a position where the painting robot waits, and the process proceeds to step S5.

In step S5, the posture of the coating robot is slowly moved back and forth (positive direction and negative direction) from the stationary state by changing the target position rotation angle of each joint by a small amount, and the operation at this time is performed. The posture holding torque τa ⁺ (τa ⁺ = [τa1 ⁺ , τa
^{2 +, ..., τa6 +]} T) and ^{τa - (τa - = [τa1} -, τa2 -,
..., τa6 ^- seek] ^T). This posture holding torque τa ⁺ , τ
a ^- is a positive direction, the motor torque limit capable of retaining the posture of the negative direction.

Here, the posture holding torques τa ⁺ and τa ⁻
Will be described in detail with reference to FIG. still,
In FIG. 9, the first, second,...
(J = 1, 2,..., 6).

First, in step S10, the position of the teaching point Pi and the posture at the position are set as target values xr0 and Rr0, and the target position rotation angles θr1, θr2,.
Is calculated (step S11). Then, the processing of steps S12 to S15 is sequentially performed for each of the first to sixth joints.

First, the target position rotation angle θrj is increased by a small amount dθj until the j-th joint starts operating in the forward direction (until θf> 0) (step S12), and the holding torque τfj immediately before the operation is increased. The orientation holding torque τaj ^{+ in the} direction is stored (step S13).

Next, the target position rotation angle θrj is decreased by a small amount dθj until the joint starts operating in the negative direction (until θf <0) (step S14), and the holding torque τfj immediately before the operation is reduced in the negative direction. posture holding torque Tauaj ^- stored as (step S15). In this way, the first to sixth
The posture holding torques τa ⁺ and τa ⁻ in the positive direction and the negative direction are obtained for each joint, and the process proceeds to step S6 in FIG.

Next, the magnitudes of the posture holding torques τa ⁺ and τa ⁻ obtained as described above are compared in step S6. That is, the absolute values | τaj ⁺ | and | τaj ⁻ | of the posture holding torque in the positive direction and the negative direction at each joint are compared. Then, | τaj ⁺ | if the larger of the target holding torque τgr (i) j of the j-th joint at this point (a taught point Pi) τaj ^- and then (step S7), | τaj ^-
Is larger than the target holding torque τgr (i) j
aj ⁺ (step S8). Where the target holding torque τ
The suffix (i) of gr (i) j indicates the target holding torque at the teaching point Pi. Thus, the target holding torque τgr (i) at the teaching point Pi ([τgr (i) 1, τgr (i) 2,.
τgr (i) 6] ^T ).

When the posture of the painting robot changes at the teaching point Pi in step S4, it is determined that the painting robot is not in a standby state, and the processing after step S5 is not performed.

By the above-described processing, the target holding torque τgr at all the teaching points where the painting robot stands by is obtained. The above processing is performed every predetermined time (for example, 100 hours), and the value of τgr (i) is updated as appropriate. In this way, a correction is made for a decrease in friction torque (τa) due to wear of the member.

(3) Operation Control Incorporating Target Holding Torque τgr Next, an operation control method in this operation mode incorporating the target holding torque τgr will be described with reference to FIG. This figure shows a processing procedure of the operation control performed every sampling time of 5 ms.

First, the counter k at the target position obtained by dividing the teaching point in step S20 is incremented by 1, and the process proceeds to step S21. If the value of k becomes equal to the total number of divisions N between the teaching points at this time, the process proceeds to step S22. In this case, the manipulator 4 has moved to the teaching point Pi + 1 following the teaching point Pi that has already passed.

In step S22, the counter i corresponding to the teaching point is incremented by one. Next, in step S23, an interpolation formula for the target position xr connecting the two teaching points Pi and Pi + 1 is determined based on the interpolation information and the target positions of the currently passing teaching point Pi and the next teaching point Pi + 1. I do. This interpolation formula is usually a polynomial at time t.

(Equation 4) It is represented as

Further, at step S23, the next teaching point Pi +
The moving time T up to 1 is divided by the unit time in the present control system, that is, the sampling time Ts, and the value is set as a new total division number N. In step S24, 0 is substituted for k, and the counter k of the target position between the teaching points is reset.

On the other hand, if it is determined in step S21 that the value of k has not yet reached the total number of divisions N, the process proceeds to step S25 without performing the processes in steps S22 to S24. This means that the manipulator 4 moves between the teaching point Pi that has already passed and the next teaching point Pi + 1.
Is moving. In this case, step S
25, the time t is set to k in the interpolation formula between the teaching points that are currently passing.
The target position xr is calculated as / N × T.

Subsequently, in step S26, the painting robot is moved to the target position and the posture at the position. Here, after the processing in steps S22 to S24, the posture is held as the position of the teaching point Pi, and after the processing in step S25, the target position and the position derived from the interpolation formula are determined. The position of the painting gun 17 and the posture of the painting robot are moved to the posture of.

Next, in step S27, it is determined whether or not the value of k is 0, that is, whether or not the manipulator 4 is on the teaching point. Then, when the value of k is 0, it is determined whether or not the teaching point is at a position where a standby posture is to be taken (step S28). move on.

In step S29, the torque compensator uses the torque compensator so that the torque of each motor at each joint becomes the target holding torque τgr (i) at the teaching point Pi among the target holding torque τgr obtained above. The command τr is changed so that the holding torque at all the joints is set to the minimum necessary holding torque. Thus, the standby posture at the teaching point Pi is maintained with the minimum power consumption. If the value of k is not 0, step S28,
The processing of S29 is not performed.

Thus, the operation control for one sampling time is completed. Thereafter, the operation control according to the processing procedure of step S20 and subsequent steps is repeatedly performed for each sampling time Ts, whereby the operation control of the coating robot is performed.

C. Second operation mode Next, as a second operation mode, the controller 1
To the sixth motor, a feedback signal of only the rotation angle θf of each joint is received, and the controller 1
A case where a command signal of only the rotation speed command θr ′ is output to each of the sixth to sixth motors will be described.

(1) Mode of Operation Control FIG. 11 is a control block diagram of the painting robot in the second operation mode. As shown in this figure, even in this operation mode, the target position rotation angle θr is changed to the current position rotation angle θf.
Is fed back, and the supply current value to each motor is calculated and output by the position compensator Gp (z) and the speed compensator Gv (z) as in the first operation mode. The control mode specific to this operation mode is the same as the control in the first operation mode. In addition, only when the coating robot is in the standby posture, the control mechanism f (θr, θ
The control is performed by adding the rotation angle calculated in f) to the command signal.

(2) Operation Control in Second Operation Mode FIG. 12 shows a processing procedure of an operation control method in the second operation mode. This figure shows a processing procedure of operation control performed for each sampling time, as in FIG.

First, in step S30 of FIG. 12, the counter k of the target position is incremented by 1, and the process proceeds to step S31. At this time, if the value of k is equal to the total number of divisions N,
Proceed to step S32. This corresponds to steps S20 to S22 described above.
As in the case of the above, the teaching point Pi + 1 following the teaching point Pi already passed
Is the case where the manipulator 4 has moved.

In step S32, the counter i corresponding to the teaching point is incremented by one. Next, the teaching point P currently passing
The positional relationship between i and the next teaching point Pi + 1 is determined (step S
33) If the positions of the two teaching points are different, the process proceeds to step S34. In step S34, the target position xr connecting the two teaching points is obtained based on the positions of the teaching points Pi and Pi + 1 and the interpolation information.
Is determined. Note that this interpolation formula is represented by a polynomial at time t, similar to the above equation (4).

On the other hand, if the positions of the teaching points Pi and Pi + 1 are the same in step S33, it is determined that the posture is held at these positions (step S35), and the flow advances to step S36 to set k to 0. And reset the counter k.

Subsequently, in step S37, the target position xr and posture Rr of the painting robot corresponding to the counter k are calculated. The calculation of the target position and the posture here is performed in step S34.
Is calculated based on the above interpolation formula (in the case of movement between teaching points), and after the processing of step S35 (in the case of holding the posture at the teaching point), the teaching point Pi is set to the target position. And Then, in step S38, the posture of the painting robot is moved so that the position of the painting gun 17 becomes the calculated target position.

Next, if the current painting robot is in the standby posture, that is, if it is determined that the processing of steps S33 to S35 is to be performed and the posture is to be maintained, the process proceeds from step S39 to S40, The standby posture is maintained by the processing described below. The processing here is
This corresponds to the control mechanism f (θr, θf).

First, in step S40, the rotation angles θ1 to θ6 of each joint are calculated based on the current position xf of the coating gun 17 and the attitude Rf of the manipulator 4, and based on the calculated values, the above equations (2) and (3) are calculated. By each joint (j)
Is calculated with respect to gravity torque τgj. Next, for each joint, it is determined whether the required motor torque τj is 0, that is, whether the anti-gravity torque τgj is within the range of the friction torque of the speed reducer (step S41).
The process proceeds to the following process according to the determination result (step S
42 or S43).

Here, the motor torque τj of the j-th joint is 0
Is satisfied, the process A of step S42 is performed.
FIG. 13 shows the procedure of the process A. As shown in this figure, in the process A, first, the coating is performed on the basis of the target position rotation angle (referred to as θrj ^(k-1)) one sampling time ago and the current target position rotation angle (referred to as θrj ^(k) ). Obtain the robot's motion braking direction. In this calculation process, the target position rotation angle θ
θrj ^{(k )} is subtracted from rj ^(k-1), and the sign of the value obtained by the subtraction is stored as a variable sign (step S50). Here, the sign of the sign is the direction of the angular acceleration (braking direction) for stopping the painting robot. The target position rotation angle θrj ^(k-1) one sampling time before is stored in such a manner as to be sequentially overwritten at one address of a storage device of the controller 2 so that it can be appropriately referred to.
Used in the above arithmetic processing.

Next, the process proceeds to step S51, in which the braking direction is returned in the opposite direction by a minute amount with the target position rotation angle being θrj−sign Δθrj, and the rotation angle of the j-th joint is returned by the minute angle Δθrj in the direction opposite to the braking direction. Proceed to step S52.

In step S52, the motor torque τj not including the friction torque (τa) when the j-th joint is moved from the state where the rotation angle is changed in step S51 to the true target position rotation angle in the standby posture is calculated. The angular acceleration and angular velocity (rotational speed) during braking are calculated from the calculated motor torque value. Then, in step S53, the angular velocity is integrated to calculate a temporal change in the target position rotation angle, and the movement to the standby posture is performed according to the calculated target position rotation angle (step S54). As a result, the motor torque τj in the state where it has moved to the true standby posture is set to 0.

On the other hand, assuming that the motor torque τj of the j-th joint is not 0 in step S41 of FIG. 12, the process B of step S43 is performed. FIG. 14 shows the procedure of the process B. As shown in this figure, in the process B, first, the sign of the gravitational torque (the sign of 0−τgj) is obtained, and the variable
It is stored as a sign (step S60).

Next, in step S61, it is determined whether the sign of the value ^obtained by subtracting θrj ^(k) from the target position rotation angle θrj ^(k-1) is the same as sign. That is, it is determined whether the gravitational torque is the same as the braking direction. If the two are different, the process proceeds to step S62, and the integral operation by the integral operation unit of the control system is omitted.

After omitting the integration operation in step S62, until the current position rotation angle fluctuates (step S62).
63), the target position angle θrj is changed by a small amount Δθj in the direction opposite to the gravitational torque (step S64). As a result, the motor torque τ is gradually reduced, and the standby posture is maintained with the minimum motor torque that can maintain the rotation angle of the j-th joint. Note that when such torque control is performed, the motor torque τ may change beyond 0, so that the motor torque τ is monitored and τ =
The processing B may be ended when it becomes 0.

In this way, the operation control for one sampling time is completed. Thereafter, the operation control according to the same processing procedure as described above is repeatedly performed for each sampling time Ts, and the operation control of the coating robot is performed.

C. Third Operation Mode Next, as a third operation mode, similar to the second operation mode, the case where the controller 1 receives a feedback signal of only the rotation angle θf of each joint and outputs a command signal of only the rotation speed command θr ′ An operation mode in which the holding torque is reduced by minutely changing the target position rotation angle in the standby posture will be described (see the control block diagram in FIG. 11).

(1) Determination of Target Position Rotation Angle Here, a method of determining the target position rotation angle when the holding torque is reduced by slightly changing the rotation speed command during standby will be described. The target position rotation angle determined here is based on the posture of the painting robot based on the teaching data.
It becomes the target position rotation angle in the target posture slightly shifted. Hereinafter, this shifted target position rotation angle is referred to as θrpaus.
(Θrpaus = [θrpaus1, θrpaus2, ..., θrpaus6] ^T )
Expressed by

FIG. 15 shows a processing procedure for determining the target position rotation angle. As shown, first, the arithmetic processing unit of the controller 1 reads the teaching data in the storage device (step S70), and determines the total number n of teaching points (step S71). Then, from the initial position (P0) to the final teaching point (P
For each teaching point Pi up to n), processing is sequentially performed according to the procedure shown in step S72 and subsequent steps. The flow of these processes is the same as the calculation of the target holding torque in the first operation mode.

In step S72, the painting robot is moved to the position of the teaching point Pi and the posture at the position. Note that this movement is performed by calculating a target position for each sampling time. The details will be described in the determined operation control ((2) described later) after the determination of the target position rotation angle θr. If there is no change in the posture of the painting robot at the teaching point Pi for a certain period of time, it is determined in step S73 that the painting robot is at a position where the painting robot waits, and the process proceeds to step S74.

In step S74, the posture of the coating robot is slowly moved back and forth from the stationary state by changing the rotation angle of the target position of each joint by a small amount, and the starting time immediately before the start of this operation is started. T ⁺ (positive) and T ^- Request (negative direction).

[0081] Here, the start time T ⁺ and T ^- for the obtaining process will be described in detail with reference to FIG. 16. Note that FIG.
, The first, second,..., Sixth joint numbers are j (j
= 1, 2,..., 6), and C is a counter for time measurement.

First, in step S80, the position of the teaching point Pi and the posture at the position are set as target values xr0 and Rr0, and the target position rotation angles θr1, θr2,.
Is calculated (step S81). Then, the first to sixth Each respective joints, sequentially performs steps S82～S92, startup time of the positive and negative directions Tj ⁺ and Tj ^- the finding. In this process, it is assumed that the integration operation by the integration calculator is not performed.

First, the target position rotation angle θrj is increased by a small amount dθj until the j-th joint starts operating in the forward direction (step S82). Thereafter, the j-th joint is held at the posture immediately before the operation, and the counter C is reset by substituting 0 (step S83).

Next, the target position rotation angle θrj is decreased by a small amount dθj until the joint starts operating in the negative direction (step S84), and the counter C is incremented by 1 each time the joint is decreased (step S85). Then, the value of the counter C when the joint is operated in the negative direction of the start time Tj ^- stored as, thereafter, resets the counter C to 0 (step S
86). Further, the target position rotation angle of the operation immediately before (one step before the minute change in the amount step) θrj ^- to.

Subsequently, the target position rotation angle θrj is increased by a small amount dθj by a small amount dθj until the joint starts operating in the positive direction from the state immediately before the operation in the negative direction (step S87). Increment by 1 (step S88).
Then, the value of the counter C when the joint is operated is stored as the start time Tj ^{+ in the} forward direction (step S89).
The target position rotation angle immediately before this operation is set to θrj ⁺ .
In this way, the positive and negative directions of the start time T ⁺ and T for each of the first to sixth joints ^- the determined, the process proceeds to step S75 in FIG. 15.

Next, the starting time T obtained as described above
performs magnitude comparison at step S75 for the size ^- j ⁺ and Tj. When greater in startup time Tj ^+, the point target position of the j-joints in (teaching point Pi) rotation angle θrpaus (i) j a negative direction operation immediately before the rotational angle Shitarj ^- and then (step S76) startup time Tj ^- when the larger of, the target position rotational angle θrpaus (i) j is a positive direction operation immediately before θrj ⁺ (step S77), if both the start-up time is equal to the target position rotational angle Shitarpaus (i) j is the target position rotation angle θr (i) j based on the teaching data (step S7
8). Here, when adding the target position rotation angle θrpaus (i) j
(i) represents the target position rotation angle at the teaching point Pi. Thus, the target position rotation angle θrpaus (i) at the teaching point Pi is obtained.

When the posture of the painting robot changes at the teaching point Pi in step S73, it is determined that the painting robot is not in a standby state, and the processing after step S74 is not performed. By the processing described above, the target position rotation angle θrp at all the teaching points where the painting robot stands by
Ask for aus.

(2) Operation Control Incorporating Target Position Rotation Angle θrpaus Next, an operation control method in this operation mode incorporating the target position rotation angle θrpaus will be described with reference to FIG. This figure shows a processing procedure of the operation control performed at each sampling time.
The flow of the processing is similar to the operation control in the first operation mode.

First, the counter k of the target position obtained by dividing the teaching point in step S90 is incremented by 1, and the process proceeds to step S91. If the value of k at this time is equal to the total number of divisions N, the process proceeds to step S92. Proceed to. In this case, the manipulator 4 has moved to the next teaching point Pi + 1.

In step S92, the counter i corresponding to the teaching point is incremented by one, and the teaching point Pi currently passing through is incremented.
And the target position of the next teaching point Pi + 1 and the interpolation information,
An interpolation formula for the target position xr connecting the two teaching points Pi and Pi + 1 is determined. This interpolation formula is represented by a polynomial at time t, similar to the above equation (4). In step S93, 0 is substituted for k, and the counter k of the target position between the teaching points is reset.

On the other hand, if it is determined in step S91 that the value of k has not yet reached the total number of divisions N, the process proceeds to step S94 without performing the processes in steps S92 and S93. This is the case where the manipulator 4 is moving between the teaching point Pi that has already passed and the next teaching point Pi + 1. In this case, the target position xr is calculated by an interpolation formula between the teaching points currently passing at step S94.

Subsequently, in step S95, the painting robot is moved to the target position and the posture at the position. Here, after the processing in steps S92 and S93, the attitude is held as the position of the teaching point Pi, and after the processing in step S94, the target position derived from the interpolation formula and the position The position of the painting gun 17 and the posture of the painting robot are moved to the posture of.

Next, in step S96, it is determined whether or not the value of k is 0, that is, whether or not the manipulator 4 is on the teaching point. Then, when the value of k is 0, it is determined whether or not the teaching point is at a position where a standby posture is to be taken (step S97). move on.

In step S98, a rotation speed command based on the target position rotation angle θrpaus (i) at the teaching point Pi among the target position rotation angles θrpaus obtained above is output to the motor of each joint. . That is, after removing the integral calculator in the control system, the target position rotation angle θr of each joint is set to θrpaus (i), and the posture of the coating robot is moved at a very small speed. This allows
The standby posture at the teaching point Pi is maintained with the minimum power consumption. If the value of k is not 0, the processes in steps S97 and S98 are not performed.

Thus, the operation control for one sampling time is completed. Thereafter, the operation control according to the processing procedure of step S90 and subsequent steps is repeatedly performed for each sampling time Ts, whereby the operation control of the painting robot is performed.

[0096]

As described above, according to the present invention, when it is determined by the determination processing means that the apparatus is in the standby state,
Since the torque of the motor is reduced by using the frictional resistance in the movable portion, the torque of the motor that maintains the standby state can be reduced without changing the position and posture of the robot based on the teaching data. As a result, the power consumption can be reduced, and the running cost for operating the robot to perform various processes can be reduced.

In particular, according to the second or fourth aspect of the present invention, the standby posture is controlled by obtaining a necessary and minimum motor torque when the robot is actually operated.
The robot can be controlled by the power consumption and running cost in the limit that can be reduced.

Further, according to the present invention, since the posture does not change when the standby posture is maintained, the present invention can be applied without any interference even when the operating environment of the robot is restricted. As a result, an effect is obtained that low-cost and flexible robot control according to various operation environments becomes possible. Here, according to the third aspect of the invention,
In accordance with various operation environments, the second command signal for the entire operation range is stored, and the standby state is controlled.

Furthermore, since the robot control according to the present invention can be realized without newly requiring a special device, the cost of the device itself does not increase.

In addition, according to the present invention, since the power consumption is reduced, the calorific value is also reduced, and the temperature rise of the motor can be suppressed, so that the effect of improving the explosion-proof property can be obtained. At the same time, an increase in the temperature of the amplifier in the control device that supplies the current to the motor can be suppressed, thereby preventing malfunction of the CPU of the control device and improving the durability of the hardware components. Is obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a painting robot system to which a robot control device according to an embodiment of the present invention is applied.

FIG. 2 is a view showing a structure of a joint portion between a second motor 8 and a first arm 9;

FIG. 3 is a diagram showing a relationship between a rotation speed θ ′ transmitted via a speed reducer and a motor torque τ.

FIG. 4 is a diagram showing a case where the first arm 9 changes from −20 ° to + 90 ° in a state where the first arm 9 and the second arm 11 form 90 °.

FIG. 5 is a diagram showing a relationship between a rotation angle θ2 and a gravitational torque τg (holding torque).

FIG. 6 is a diagram illustrating an example of a trajectory of a target position and a teaching point.

FIG. 7 is a control block diagram of the painting robot in a first operation mode.

FIG. 8 shows a target holding torque τgr in the first operation mode.
It is a figure showing the calculation processing procedure of.

[9] attitude holding torque .tau.a at step S5 in FIG. 8 ⁺ and .tau.a ^- is a diagram illustrating a processing procedure for obtaining the.

FIG. 10 shows a target holding torque τ in the first operation mode.
It is a figure showing the operation control method which adopted gr.

FIG. 11 is a control block diagram of the painting robot in the second and third operation modes.

FIG. 12 is a diagram illustrating an operation control method in a second operation mode.

13 is a diagram showing a process A in step S42 of FIG.

FIG. 14 is a diagram showing a process B in step S43 of FIG.

FIG. 15 is a diagram showing a procedure for determining a target position rotation angle shifted by a small amount in the third operation mode.

FIG. 16 shows a startup time T in step S74 of FIG.
⁺ And T ^- it is a diagram illustrating a processing procedure for obtaining the.

FIG. 17 is a diagram illustrating an operation control method incorporating a target position rotation angle θrpaus in a third operation mode.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Controller 2 Teaching pendant 4 Manipulator 6 1st motor 7 Revolving base 8 2nd motor 9 1st arm 10 3rd motor 11 2nd arm 12 Motor case 13 1st wrist unit 14 2nd wrist unit 15 3rd wrist unit 17 Painting gun

Claims (4)

[Claims] 1. A robot control device for controlling a robot that operates a movable part by torque of a motor transmitted through a speed reducer based on teaching data, wherein a first control based on the teaching data is provided to the robot based on the teaching data. First command signal output means for outputting a command signal; and determining whether or not the operation of the robot is in a standby state in which a predetermined standby posture is maintained during the operation of the robot based on the first command signal. Processing means; and when the determination processing means determines that the motor is in a standby state, the torque of the motor is reduced by the torque and the friction torque of the speed reducer in a range where the movable portion is kept stationary. Output the second command signal
And a command signal output means. 2. A robot control method by the robot control device according to claim 1, wherein a first step of moving the robot to a teaching point of the teaching data, and when the robot moves to the teaching point, A second step of determining whether or not the teaching point is in a position to take a standby posture
And a third step of outputting a third command signal for changing the torque of the motor by a small amount in a certain direction when it is determined in the second step that the position should be in the standby posture. A fourth step of storing the third command signal immediately before the movable portion operates from the stationary state by the third command signal, and changing the torque of the motor in a direction opposite to a fixed direction in the third step. A fifth step of outputting a fourth command signal for changing the direction by a small amount in a direction, and a fourth command signal for storing the fourth command signal immediately before the movable portion starts operating from a stationary state by the fourth command signal. Step 6, and among the fourth and fifth command signals stored in the fourth and sixth steps, the second command at the teaching point is a command signal that indicates a small torque to the motor. Signal The determined by a seventh step second
And controlling the robot based on the command signal. 3. The robot control method according to claim 2, wherein the robot is moved to each of the teaching points, and the second to seventh steps are repeated, and the second and the seventh steps are performed at all the teaching points that hold a standby posture. A robot control method comprising: determining and storing a command signal; and controlling the robot based on the teaching data and the stored second command signal. 4. The robot control method according to claim 1, wherein the first and second command signals are speed command signals for instructing an operation speed of the movable part with respect to the motor. A tenth process of obtaining a torque of the motor necessary to maintain the movable part in a stationary state based on an operation amount of the movable part when the determination processing part determines that the movable part is in a standby state; If the necessary torque of the motor obtained in the process of step 10 is 0, the movable portion is operated in a minute range, and the torque of the motor in a stationary state is set to 0. An eleventh step of outputting the second command signal for instructing the operation speed of the movable part and operating the movable part in the minute range; A twelfth step of, if the required torque of the motor is not zero, changing the second command signal by a small amount until immediately before the movable part operates by reducing the torque of the motor. A robot control method characterized by the above-mentioned.