JPH07319506A - Autotuning controller - Google Patents

Abstract

PURPOSE:To set optimum control characteristics by automatically adjusting a control parameter so that cycle time measured by an automatic control parameter adjusting means can be desired cycle time. CONSTITUTION:An autotuning part 106 decides whether the present state of a controlled system 200 is a static state or a vibrating state by fetching the real position of the controlled system 200 fed back from a position sensor 23 and the estimated torque of the controlled system 200 estimated by a torque estimating part 107 and when the controlled system is vibrated, that vibration cycle is detected. Further, the autotuning part 106 successively samples and measures the real positions fed back from the position sensor 23 and the estimated torques estimated by the torque estimating part 107 during the execution of a user program and a control state is judged from the respective values so that velocity proportion gains 114 and 115 and velocity integration gains 116 and 117 as velocity control parameters and a position gain 113 as a position control parameter can be decided (adjusted).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic tuning controller for automatically adjusting a control state of a servo control device, particularly for adjusting position and speed control parameters.

[0002]

2. Description of the Related Art In a conventional auto-tuning controller, in auto-tuning for determining control parameters related to speed control, a servo motor is actually operated by a step command or the like to estimate the characteristics of a controlled object, and its operation characteristics are The control target is estimated from the (speed response) to determine the control parameter (Japanese Patent Laid-Open No. 3-2681).
No. 02).

[0003]

However, in the above-mentioned auto-tuning, even if the operation characteristic (speed response) to the command is optimum, when a series of operations for actually moving the assembly machine is performed, The operation may not be completed at the target cycle time by the user, and there is a problem that manual adjustment is performed again or the user command is repeatedly changed while measuring the cycle time.

Further, there are cases in which optimum control cannot be performed with a uniform control parameter due to load fluctuations during control of the controlled object. Furthermore, in the case of conventional auto-tuning, the drive control of the controlled object when performing auto-tuning has control parameters set only for certain specific operations, and various drive controls should be selected during automatic adjustment. I couldn't.

The present invention has been made in view of the above circumstances, and an object thereof is to provide an auto tuning controller capable of setting optimum control characteristics.

[0006]

In order to achieve the above object, the present invention provides an automatic tuning controller for automatically adjusting a control parameter for determining a control state of a control device according to claim 1, wherein: Employing a technical means of measuring a cycle time of an operation program to be executed and providing a control parameter automatic adjusting means for automatically adjusting the control parameter so that the measured cycle time becomes a desired cycle time. Is.

According to a second aspect of the present invention, in the auto tuning controller for automatically adjusting the control parameter for determining the control state of the control device, a plurality of the control parameters can be set and a switching means for switching the plurality of control parameters. In addition to the above, the technical means of switching control of the control parameter by the switching means based on the state of the load controlled by the control device is adopted. Further, the load controlled by the control device may include a transfer device that controls the transfer of the load, and the transfer device may be a vertical drive shaft whose load state varies due to gravity.

Further, according to a fourth aspect of the present invention, in an auto tuning controller for automatically adjusting a control parameter for determining a control state of a control device, at the time of automatically adjusting the control parameter, at least a single drive driven by the control device is used. A rough adjustment mode in which the control parameter is set by a driving operation, and a fine adjustment mode in which the control parameter is set by executing an operation program driven by the control device are provided.
The technical means of adjusting the control parameter by selecting both the adjustment modes is adopted.

[0009]

According to the first aspect of the present invention, the control parameter automatic adjusting means measures the cycle time of the operation program executed by the control device, and the measured cycle time becomes the desired cycle time. So that the control parameters are automatically adjusted. Therefore, even if the cycle time does not reach the target value even though the appropriate gain is set for the predetermined movement, the control parameters are automatically adjusted within the range in which the operating characteristics do not exceed the suitability. Therefore, the cycle time can be achieved to the target value without making complicated adjustments.

Further, according to a second aspect, the control parameter switching means controls the switching of the control parameters based on the state of the load controlled by the control device. As a result, even in the case where the load of the controlled object varies, the control parameter can be switched and controlled according to the variation of the load, so that it is possible to always control with the optimum control parameter. It is possible to further reduce the operation time. In particular, if this method is applied to the vertical drive shaft that is subjected to the action of gravity, the effect is higher.

According to a fourth aspect of the present invention, the operation mode of the controlled object for performing auto-tuning includes a rough adjustment mode in which the control parameters are set by a single drive operation, and an operation program driven by the control device. A fine adjustment mode in which the control parameter is set by execution is provided, and the control parameter is adjusted by selecting the both adjustment modes. As a result, the rough adjustment mode is selected when the operation program cannot be executed or rough control parameter adjustment is performed, and the fine adjustment mode is selected when high-precision control parameters are set based on the entire operation program. The adjustment mode can be selected, and the optimum adjustment mode can be selected according to the operating state of the controlled object.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, one embodiment of the auto tuning controller of the present invention will be described with reference to FIGS.
FIG. 1 is a diagram showing an overall configuration of a system controlled by an auto tuning controller.

Reference numeral 100 is an auto-tuning controller for controlling a uniaxial robot 40 that reciprocates an object in a linear direction. Auto tuning controller 1
The AC servomotor 20 driven and controlled by 00 is
A motor power cable 21 that is installed in the uniaxial robot 40 to drive the drive axis of the uniaxial robot 40 and supplies a rotational drive current of the AC servomotor 20 and an encoder that detects the position of the AC servomotor 20. It is connected to the auto tuning controller 100 by an encoder cable 22 for inputting a signal. Further, the auto tuning controller 100 is provided with a parallel input / output port for connecting an external input / output device such as a programmable controller.

This auto tuning controller 10
0 is connected to a personal computer 30 via a line cable 32 of RS232C type or the like. A user program for driving and controlling the uniaxial robot 40 and a driving command for manual control (so-called jog Feed) etc. can be supplied to the auto tuning controller 100. An automatic adjustment program, which is responsible for an interface with a user for automatically adjusting the control parameters in the automatic tuning controller 100, should be stored in a memory medium such as the floppy disk 31 or a hard disk in the personal computer 30. It is possible to activate this program by the personal computer 30 and set a user program in the adjustment process.

Next, the auto tuning controller 10
The detailed configuration of 0 will be described. FIG. 2 is a block diagram showing the configuration of the auto tuning controller 100. Auto tuning controller 10 of the present embodiment
0 is an interface unit (I / F unit) 101, an input / output unit (I / O unit) 102, an external device command decoding unit 103, I
/ O control unit 104, position command unit 105, auto tuning unit 106, torque estimation unit 107, position control unit 10
8, a speed control unit 109, a current control unit 110, etc. are built in. The controlled object 200 is the AC servomotor 20.
And a load attached to the motor shaft of the AC servomotor 20.

The position sensor 23 is a position detecting means for detecting the actual position of the controlled object 200, and specifically, the AC servomotor 20 or a rotary encoder attached to the drive shaft can be used. It should be noted that this rotary encoder can determine the rotation direction by generating pulses having different phases by 90 °, and can determine the origin by obtaining an output of one pulse for one rotation. .

An input / output unit (I / O unit) 102 is a port for connecting an external device such as a sequencer, and its input / output signal is controlled by the I / O control unit and sent to the external device command analysis unit 103. Be analyzed. Personal computer 30
The robot operation panel or the like is connected to the auto tuning controller 100 via the interface unit 101, and supplies a jog operation command or a user program to the external device command decoding unit 103. External device command decoding unit 1
Reference numeral 03 denotes a user program (operation command program of the controlled object 200) in a high-level language or a specialized language created by the user using the personal computer 30 or the operation panel for the robot.
Is converted into a language or a signal that can be interpreted by the auto tuning controller 100, and a command (position command) of a position to be moved in a time unit at a certain fixed sampling time.
To the position control unit 108 via the position command unit 105. The position control unit 108 uses the position gain 113, which is a position control parameter, from the deviation between the position command output from the position command unit 105 at every certain sampling time and the actual position of the control target 200 fed back from the position sensor 23. Is used for P control, and a speed command is given to the speed control unit 109. The position control unit 108 is further provided with a changeover switch SW1 controlled by the auto-tuning unit 106, and a route for sending the position command from the position command unit 105 directly to the speed control unit 109 and a normal position loop control. It is configured to switch between the route for performing.

The speed control unit 109 divides an actual speed obtained by differentiating the actual position of the controlled object 200 fed back from the position sensor 23 with a differentiation operator 111, and a speed command output from the position control unit 108. The deviation is input, and the switches SW2 and S in the speed control unit 109 are input.
Each is supplied to W3. SW2 switches and selects two speed proportional gains 114 and 115 of the speed control parameters, and SW3 switches and selects two speed integral gains 116 and 117 of the speed control parameters.

Thus, the deviation amount is PD-controlled through either the speed proportional gain or the speed integral gain, and the torque command is given to the current control unit 110. Current control unit 110
Is converted from the torque command from the speed control unit 109 into a current command, and the deviation between the current command and the actual current fed back from the current sensor 112 that detects the current actually flowing to the servo motor is controlled by the control target 200 in advance. PI control is performed using the set current control parameter, and a current is given to the servo motor.

The auto-tuning unit 106 takes in the actual position of the controlled object 200 fed back from the position sensor 23 and the estimated torque of the controlled object 200 estimated by the torque estimation section 107, and the present state of the controlled object 200 remains stationary. If it is vibrating, the vibration cycle is detected. However, it is necessary to consider the quantum error of the position sensor 23 (rotary encoder) when determining whether the controlled object 200 is in a stationary state or a vibrating state. Therefore, when the vibration amplitude α of the actual speed or the estimated torque is smaller than the preset value, it is determined that the vibration does not occur.

Further, the auto-tuning unit 106 sequentially samples and measures the actual position fed back from the position sensor 23 during execution of the user program and the estimated torque estimated by the torque estimation unit 107, and determines the control state from those values. Judgment, speed proportional gains 114, 115 and speed integral gains 116, 1 which are speed control parameters
17, position gain 11 which is the position control parameter
Determine (adjust) 3.

The torque estimation unit 107 includes a current control unit 110.
The driving torque and stationary torque of the controlled object 200 are estimated from the actual current from Next, the actual operation of the auto tuning according to the present invention for determining the speed control parameter and the position control parameter will be described based on the flowcharts shown in FIGS.

First, in step S1, the system is set up. The personal computer 30 is connected to the auto tuning controller 100 via the line cable 32. A program for performing auto-tuning is installed in advance in the personal computer 30, and the program is selected and executed, and thereafter, an operator operates in accordance with a menu displayed on the display screen of the personal computer 30 in order. Auto tuning is performed.

Normally, when the 1-axis robot 40 is repeatedly operating in a factory or the like by automatic operation, the operation program is downloaded to the auto tuning controller 100 itself, so the personal computer 3
0 is not connected to the auto tuning controller 100 and is stored in another place. Next, the power of the auto tuning controller 100 is turned on. At this time, the AC servomotor 20 is additionally provided with a switch (not shown) for supplying or not supplying power, and this switch is kept in an off state. This prevents the drive from vibrating at the same time when the power is turned on.

Next, the power switch for the AC servo motor is turned on, and the personal computer 30 commands and drives the jog feed operation for temporarily moving the drive shaft at a constant low speed, and the rough operating state of the system is visually confirmed. To do. In step S2, initial values of various parameters such as axis parameters and servo parameters are downloaded to the auto tuning controller 100 side. As a result, the position gain 113, the velocity proportional gains 114 and 115, and the velocity integral gains 116 and 117 are also set to predetermined initial values.

Next, the "automatic adjustment" menu for auto tuning is selected from among several operation menus displayed on the screen of the personal computer 30. Then, the menus of "rough adjustment", "fine adjustment", and "stiffness adjustment" are displayed on the screen. "Rough adjustment" and "fine adjustment" are two types of modes for automatically adjusting the position control parameter and the speed control parameter. "Rigidity adjustment" indicates the rigidity of the control target, that is, low response, medium response, and high response. It can be selected from three levels of response, and any one of them is selected in step S4. If it is known in advance that the rigidity is high, the rigidity is set to "high response", so that the adjustment width of the control parameter can be reduced from the initial stage of the adjustment, and the automatic adjustment can be performed faster.

A detailed adjustment method in each of the "rough adjustment" and "fine adjustment" modes will be described later.
In the rough adjustment mode, each control parameter is used to adjust the gain within a rough range for the jog feed operation. In the fine adjustment mode, the robot operation program is actually operated, and the optimum gain is obtained in the entire operation. To make adjustments.

In this way, the two adjustment modes are provided in order to avoid taking a long time until the optimum gain value is obtained when an accurate gain adjustment is performed from the beginning of the adjustment stage. During a certain work, there is a case where it is desired to adjust the gain within a certain range or only in a specific operation. In such a case, a mode capable of adjusting a single operation is required.

Next, the operation when the "rough adjustment" mode is selected will be described. When the rough adjustment mode is selected from the menu on the screen in step S5, the initial operation is performed in step S6. This is to drive a small amount of axis movement in the same manner as jog feed, at which time load inertia is estimated and speed loop gain is adjusted. That is, when the initial operation is selected,
The auto tuning unit 106 has switches SW1 and SW.
2 and SW3 are tilted to the contacts 1a, 1b, 1c side, respectively.
As a result, the position command sent from the position command unit 105 to the position control unit 108 does not pass through the position loop gain 113, and is directly sent to the speed control unit 109 as a speed command in a so-called open loop control state ( Step S
7).

In this state, the axis is moved using the initial value of the speed control parameter set in step S2 (step S8), and the acceleration / deceleration torque during the movement, the acceleration / deceleration time, etc. are calculated by the torque estimation unit 107 and the automatic tuning. Part 10
6 obtains the feedback signals from the current control unit 110 and the position sensor 23, and estimates the load torque and the load inertia in the auto tuning unit 106 from these values (step S9).

Next, the position and speed control parameters are calculated from the estimated load torque and load inertia, and the position gain 113 and each speed control gain 114, 1 are obtained.
15, 116, 117, (At this time, the same value is set to 114 and 115, and the same value is set to 116 and 117.)
Set to. Further, at the position where the movement ends, the velocity proportional gain 11 is set so that the controlled object 200 does not vibrate and stands still.
4, the speed integral gain 116, the automatic tuning unit 1
Automatic adjustment is performed according to 06 (step S100). That is, the signal from the position sensor 23 is sent to the auto tuning unit 106.
Then, the vibration frequency x is calculated from the vibration amplitude α of the actual speed and the cycle of the peak value (step S101).

Then, the auto-tuning unit 106 determines whether the current state of the controlled object 200 is a stationary state or a vibrating state. In this determination, when the vibration amplitude α of the actual speed (or the actual position) is smaller than the preset value, it is judged that the vibration is not occurring, and when the vibration amplitude α is equal to or larger than the preset value, the vibration state is determined. It is determined that
(Step S102) Then, when the determination result of step S102 is YES,
That is, when it is determined that the control target 200 is in the vibrating state, it is determined whether the vibration frequency x is smaller than a certain ratio with respect to the frequency fs of a certain sampling time (step S103). If the determination result is YES, that is, if the vibration frequency x is smaller than a certain ratio with respect to the frequency fs (x / fs <A), the vibration amplitude α of the actual speed becomes less than or equal to the set value (step Until it is determined to be in a stationary state in S102), the auto tuning unit 1
At 06, the speed control parameters 114 and 116 are gradually increased (step S105).

Further, the determination result of step S103 is NO.
In other words, when the vibration frequency x is equal to or higher than a certain ratio with respect to the frequency fs (x / fs ≧ A), the vibration amplitude α of the actual speed becomes equal to or lower than the set value (step S10).
The automatic tuning unit 106 gradually reduces the speed control parameter until the determination unit 2 determines that the vehicle is in the stationary state (step S104).

When the result of the determination in step S102 is NO, that is, when it is determined that the controlled object 200 is in a stationary state, the speed control parameters 114 and 116 determined by the auto tuning unit 106 at that time are set to the speed control parameters. It is updated as a speed control parameter used in the control unit 109, and the switch SW1 in the position control unit 108 is set to the contact 1b.
Then, the control loop is returned to the normal control loop state to end the initial operation (step S10).

Next, the jog operation is commanded from the personal computer 30 (step S11). This jog operation command is sent to the position command unit 105 through the external device command decoding unit 103, and the position command unit 105 sends the position command to the position control unit 108 to perform axis movement.
During this movement, the auto tuning unit 106 measures various data such as the actual position of the controlled object fed back from the position sensor 23 at each sampling time and the estimated torque value from the torque estimation unit 107, and the external device command decoding unit. From the command state of 103, an overshoot amount, an undershoot amount with respect to the command position, a load torque of the controlled object 200, an acceleration / deceleration torque during acceleration / deceleration, an acceleration / deceleration time, etc. are calculated (step S12).

Subsequently, in the "rough adjustment", the process proceeds to step S13, and it is determined whether or not various data measured and calculated in step S12 meet some determination conditions. The contents of the judgment are as follows. Position overshoot (exceeding the final destination of one motion pattern beyond in-position)
Is there Judgment 2. Is there vibration when the controlled object is stopped? Judgment 3. Speed overshoot and undershoot are below 10%.

According to the above judgment results, if all the conditions are satisfied, the adjustment completion is displayed on the screen (step S1).
6) If at least one of the conditions is not satisfied, it is displayed on the screen as incomplete. If the measurement adjustment is to be performed again when it has not been completed, the jog operation is designated again from the menu on the screen of the personal computer 30 and the same operation as described above is performed. At this time, in step S14, it is determined whether or not the number of repetitions of the operation is a predetermined number of times N times or less, and if it is N times or less, the control parameter is adjusted according to the following rule (step 15), and then, again. Perform an action.

Rule 1. When the overshoot is 10% or more and the damping ratio is 0.2 or more, the speed proportional gain and the speed integral gain are increased according to a fixed rule. Rule 2. When the overshoot is 10% or more and the damping ratio is 0.2 or less, the position gain is reduced. Rule 3. If the determination 2 (whether the controlled object is vibrating when stopped) is not satisfied, the speed proportional gain and the speed integral gain are lowered according to a fixed rule.

If the above determination is not satisfied within N times, the incomplete adjustment is displayed on the screen together with the reason (that is, which of the above determination conditions is not met) (step S18), and the "rough" Adjustment ”ends. Next, the operation of the "fine adjustment" mode will be described. After "rough adjustment" is completed and each data is displayed on the screen (step S16), it is judged whether "fine adjustment" is to be executed (step S17), and "fine adjustment" mode is selected. , Step S5, and selects the "fine adjustment" mode. When the "fine adjustment" mode is selected, the operation program requiring automatic adjustment is read out on the personal computer 30, and the target cycle time of the operation program and its margin are set (step S200). FIG. 6A is a diagram showing a state of the command speed of the set operation program. The horizontal axis is the time axis and the vertical axis is the speed. In FIG. 6A, section A,
B, C, D, and E form one cycle of the program, and C and E form a stop section. In addition, the allowance is a percentage set with respect to the target time as to how much the target time may be subtracted.

Next, the value set in step S200 is downloaded to the auto tuning controller 100 (step S201). The program is operated in this state (step 11), as in “rough adjustment”, overshoot and undershoot amount with respect to the command position, load torque of the control target 200, acceleration / deceleration torque and acceleration / deceleration time during acceleration / deceleration, cycle The time, the amplitude attenuation ratio, etc. are calculated (step S12).

Next, from these calculated values, the drive axis is
It is determined whether it is the vertical axis (step S20).
2). That is, at motor startup (cycle startup) and in FIG.
If the stationary torque when the controlled object is stopped in the sections C and E of (a) is equal to or larger than the predetermined torque Ta of the rated torque of the motor, the auto-tuning unit 106 determines that the axis is the vertical axis, and the speed proportional gain 114, Different values are set to 115 and different values to the velocity integral gains 116 and 117 (step S205).

A method of controlling switching of these gains will be described. For example, as shown in FIG. 7, the switch SW2 is controlled by the auto tuning unit 106 so as to be controlled by the velocity proportional gain 114 and the velocity integral gain 116 during acceleration in the gravity direction (downward) and deceleration in the antigravity direction (upward).
SW3 is switched to the contact points 2a and 3a, respectively, and is switched so as to be controlled by the velocity proportional gain 115 and the velocity integral gain 117 during deceleration in the gravity direction (downward) and during movement and during acceleration in the antigravity direction (upward). SW2, SW
3 is switched to the contacts 2b and 3b side, respectively. This makes it possible to control the gain with an appropriate gain when the load changes significantly depending on the moving direction such as the vertical axis.

Next, when it is determined in step S202 that the axis is not the vertical axis, it is determined in step S203 whether there is a load change. That is, if the difference between the acceleration torque and the deceleration torque is equal to or larger than the predetermined torque Tb of the rated torque of the AC servomotor 20 in each of the moving sections A, B, and D in FIG. , The torque difference is T
Different values are set to the speed proportional gains 114 and 115, and different values are set to the speed integral gains 116 and 117, respectively, by distinguishing those of b or more and those of Tb or less (step S205). Thus, for example, in the section where a heavy object is loaded and moving, the switches SW2 and SW3 are respectively set to contact points 2a so as to be controlled by the speed proportional gain 114 and the speed integral gain 116 set when the torque is Tb or more. , 3a side, and switches SW2, S to be controlled by the speed proportional gain 115 and the speed integral gain 117 in the case where the section is moving in the state of
W3 can be switched to the contact points 2b and 3b, respectively, and can be controlled with optimum control parameters.

In step S203, if there is no torque equal to or greater than the predetermined torque Tb in any section, the same value is set in each speed proportional gain 114, 115, and similarly, each speed integral gain 116, 117 is set. The same value is set for each. (Step S204) Next, it is determined whether the value measured in step S12 matches each determination condition (step S13).

The determination condition at this time is "rough adjustment"
In addition to the judgment conditions at the time, judgment 4. 10% of maximum commanded speed of AC servo motor
From 90% to 90% of the maximum commanded rotational speed of the AC servo motor (fall time) is calculated. Judgment whether it is shorter than a predetermined value tb 5. Determine if the cycle time is less than or equal to the specified target value.

If these judgment conditions are not satisfied, the process proceeds to step S14, and the number of repetitions of the operation program is N.
It is determined whether the number of times is less than or equal to the number of times (step S14). If the number of times is less than or equal to N, the parameters are adjusted according to the following rules. (Step S15) Rule 1. If the determination 4 (rise time is ta or less, fall time is tb or less) is not satisfied, the position gain is increased.

Rule 2. When the overshoot is 10% or more and the damping ratio is 0.2 or more, the speed proportional gain and the speed integral gain are increased according to a fixed rule. Rule 3. When the overshoot is 10% or more and the damping ratio is 0.2 or less, the position gain is reduced. Rule 4. If the determination 2 (whether the controlled object is vibrating when stopped) is not satisfied, the speed proportional gain and the speed integral gain are lowered according to a fixed rule.

Rule 5. When the time to enter the in-position is long (exceeds the target cycle time), the speed proportional gain and the speed integral gain are increased according to a fixed rule. Further, the position gain 113 gradually increases every time the cycle adjustment is repeated, but if the slope of acceleration / deceleration at the actual speed does not rise even if the value is increased by a certain value or more, the position gain 113 has a minimum value for maintaining the current slope. The value is the position gain 113. Since this adjustment is performed in parallel with the above rule, if it overlaps with the adjustment of rule 3, for example, the position gain will not change because the adjustment directions thereof are canceled.

When all the judgment conditions are satisfied, the adjustment completion is displayed on the screen (step S16), and when any of the judgment conditions is not satisfied even if the operation program is repeated N times, the adjustment is not completed. Is displayed on the screen (step S18), and the process ends. 6 (a), 6 (b)
6 shows the command speed of the user program and the state of the actual speed at that time, and FIGS. 6C and 6D show the command speed after the fine adjustment and the state of the actual speed thereof. In FIG. 6B, the cycle time target value is exceeded before the automatic adjustment. After the automatic adjustment, each gain is adjusted to the optimum value to shorten each moving section A, B, D, and the cycle time is shortened as a whole, and the command speed is set within the target value (Fig. 6 (c)). ), And the actual operation is within the target value. (FIG.6 (d)). In the sections C and D, which are movement stop sections, the stop time is the minimum time required for each work, so this time is left as the setting value of the first operation program.

The "fine adjustment" may be performed after the "rough adjustment" has been performed as described above, or the "fine adjustment" mode may be selected and executed from the beginning.
In that case, the initial operation performed by the rough adjustment (step S6
~ Step S10) is not performed, but once operates based on the initial value downloaded in Step S2, the load torque and the load inertia of the system are estimated from the measurement during the operation of the first operation program, and the estimated values are obtained. The value calculated based on the above is set, and the following "fine adjustment" is executed.

[Other Embodiments] Although the auto-tuning unit 106 is provided in the auto-tuning controller 10, it may be separately provided. Further, although the present embodiment is applied to the one for controlling the one-axis moving type robot, the present invention is not limited to the robot, but the machine tool,
It can also be applied to drive control of an assembly machine, a carrier, and the like.

The torque estimation unit 107 includes a current sensor 112.
Although the torque was estimated only with the actual current fed back, the actual position from the position sensor 23 was directly taken in and A
Current electrical angle and current sensor 112 of the C servo motor 20
The torque required to make the controlled object 200 stand still may be estimated based on the actual current from. The torque can be calculated based on the following formula.

[0053]

## EQU1 ## TRQ = KT × ia / sin θ where TRQ = estimated torque, KT = torque constant, ia =
Real current, θ = motor electrical angle. Further, in this embodiment, the optimum values of the speed control parameter and the position control parameter are obtained from the vibrations of the actual speed and the actual position, but the torque ripple γ is calculated based on the data obtained by the torque estimation unit 107 at each stage. It is also possible to obtain and determine the speed control parameter and the position control parameter so that this ripple γ becomes the minimum.

Further, in the present embodiment, in the "rough adjustment" mode, it is not judged whether or not there is a vertical axis and load fluctuation, but the jog operation is carried out in both positive and negative directions, and a fine operation is performed from the operating state at that time. As in the adjustment mode, it is possible to determine the vertical axis and whether or not there is a load change, and set speed control parameters corresponding to them.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall configuration of a system controlled by an auto tuning controller.

FIG. 2 is a block diagram showing a configuration of an auto tuning controller 100.

FIG. 3 is a flowchart showing the operation of this embodiment.

FIG. 4 is a flowchart showing the operation of this embodiment.

FIG. 5 is a flowchart showing the operation of this embodiment.

6A and 6B are diagrams showing states of a command speed of a user program and an actual speed at that time, and FIGS.
(D) is a diagram showing a state of a command speed after fine adjustment and its actual speed.

FIG. 7 is a diagram showing a speed control state when speed control parameters are switched and controlled on a vertical axis.

[Explanation of symbols]

 20 AC Servo Motor 23 Position Sensor 30 Personal Computer 40 1-axis Robot 100 Auto Tuning Controller 103 External Device Command Decoding Unit 105 Position Command Unit 106 Auto Tuning Unit 107 Torque Estimating Unit 108 Position Control Unit 109 Speed Control Unit 110 Current Control Unit 113 Position Gain 114, 115 Speed proportional gain 116, 117 Speed integral gain 200 Control target

Claims (4)

[Claims] 1. An auto-tuning controller for automatically adjusting a control parameter for determining a control state of a control device, wherein a cycle time of an operation program executed by the control device is measured, and the measured cycle time is set to a desired value. An auto-tuning controller characterized in that a control-parameter automatic adjusting means for automatically adjusting the control parameter so as to obtain a cycle time is provided. 2. An auto tuning controller for automatically adjusting a control parameter for determining a control state of a control device, wherein the control parameter is provided so that a plurality of the control parameters can be set, and the control device has a switching means for switching the plurality of control parameters. An auto-tuning controller characterized in that the control parameter is switched and controlled by the switching means on the basis of a state of a load controlled by. 3. The load controlled by the control device includes a transfer device that controls the transfer of the load, and the transfer device is a vertical drive shaft whose load state varies due to gravity. The auto tuning controller described in. 4. An auto-tuning controller for automatically adjusting a control parameter for determining a control state of a control device, wherein when the control parameter is automatically adjusted, the control parameter is adjusted by at least a single drive operation driven by the control device. A rough adjustment mode to be set and a fine adjustment mode in which the control parameter is set by execution of an operation program driven by the control device are provided, and the adjustment of the control parameter can be performed by selecting both adjustment modes. An auto tuning controller that is characterized by what it does.