JPH07160323A - Operation waveform diagnostic device for industrial robot - Google Patents

Abstract

PURPOSE:To provide the operation waveform diagnostic device for the industrial robot which automatically diagnoses the waveform of operation data and displays the result. CONSTITUTION:A robot system has a robot main body which performs specific operation and a controller 2 which controls this robot main body, and a work station 10 having a data file 16a which contains the history of operation data on the robot main body by having a communication with the controller 2, a waveform feature extraction part 18 which extracts feature quantities of the waveform of the operation data inputted by being compared with a reference waveform, a standard data generation part 19 which generates modeled standard data from the reference waveform and feature quantities, a standard data adjustment part 20 which adjusts the standard data according to the operation data, and a waveform diagnostic part 21 which diagnoses the waveform of the operation data while comparing the adjusted standard data with the operation data and looking up a previously set comment table 21a for waveform diagnosis is connected to the controller 2.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion waveform diagnosing device for an industrial robot, and more particularly to a device for automatically diagnosing the waveform of motion data and displaying the result.

[0002]

2. Description of the Related Art Today, industrial robots (hereinafter simply referred to as robots) are being widely introduced in various production plants including automobile factories. Such a robot system is generally shown in FIG. 13, for example. In addition, the robot main body 1 (for example, a multi-axis robot having a plurality of drive axes including a servo motor), a controller 2 for controlling the robot main body, and an amplifier 3 (for example, a digital servo amplifier) that amplifies a signal, in this case,
The robot body 1 is connected to the amplifier 3 via the D / A converter 4 and the A / D converter 5. The speed command or the position command from the controller 2 is amplified by the amplifier 3 and then sent to each axis of the robot body 1 via the D / A converter 4. As a result, the robot body 1 performs a predetermined operation, and the operation data of each axis at this time (for example, speed feedback data from the encoder, current data, etc.)
Is input to the amplifier 3 via the A / D converter 5 and then sent to the controller 2.

Conventionally, depending on the system, the operation data is further stored in the memory in the controller 2 and the signal waveform thereof is displayed (see FIG. 14). Thus, the waveform of each robot body 1 during operation is judged to be good or bad, and the cause of the malfunction is investigated, the contents of secular change are checked, and tuning is provided. Note that FIG. 14 shows a display example of the signal waveform, and FIG. 14A shows each waveform of the speed command (dotted line) and the speed feedback signal (solid line) with respect to time.
FIG. 6B shows the waveform of the current value with respect to time.

[0004]

However, in such a conventional operation waveform display method, the data when each axis of the robot body 1 operates is stored in the memory in the controller 2 via the amplifier 3. Since it is only displayed in advance, the data is rewritten and the previous data does not remain when the next operation is performed. Therefore, in order to retain the previous operation data, it is necessary to perform data communication and store it in an external data file, or print it on paper and leave it.

In evaluating the waveform of the operation data, whether it is a method of reading out from a data file and displaying it on the screen or a method of printing it on paper, in any case, a person having specialized knowledge needs to know the past. It was necessary to prepare data and compare it with the waveform on the screen or on the paper to judge whether the waveform is good or bad, and further to determine which part of the waveform is the problem.

The present invention has been made in view of the above problems of the prior art, and an operation waveform diagnostic device for an industrial robot capable of automatically diagnosing the waveform of the operation data of the robot and displaying the result. The purpose is to provide.

[0007]

The present invention for achieving the above object provides a robot main body for performing a predetermined work, a controller for controlling the robot main body, and operation data of the robot main body by communicating with the controller. Storage means for storing the history of the operation waveform, operation waveform characteristic extraction means for extracting a characteristic amount of the waveform of the operation data by comparison with a reference waveform, and standard data modeled on the basis of the reference waveform and the characteristic amount. And standard data adjusting means for adjusting the standard data in accordance with the operation data, and a waveform diagnostic comment preset by comparing the adjusted standard data with the operation data. Waveform diagnosis means for performing a waveform diagnosis of the operation data while looking up a table, and display means for displaying the result of the waveform diagnosis. Characterized in that it.

[0008]

According to the present invention having such a configuration, the operation data of the robot main body is sequentially fetched into the storage means by communication with the controller and accumulated in the history file of the operation data. The operation waveform feature extraction means reads out a required number of operation data from the storage means, compares each with the reference waveform, and extracts the feature amount of the waveform of each operation data. The standard data creating means creates standard data modeled based on the reference waveform and the extracted feature amount. The standard data adjusting means adjusts the created standard data to the scale of operation data for waveform diagnosis from now on.
The standard data is adjusted to the operational data.
The waveform diagnosing means compares the adjusted standard data with the operation data and performs a waveform diagnosis of the operation data while looking up a preset waveform diagnosis comment table. The result of this waveform diagnosis is displayed on the display means.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. 1 is a schematic configuration diagram of this embodiment, FIG. 2 is a system configuration diagram of a main part of FIG. 1, FIG. 3 is a diagram showing an example of a comment table, FIG. 4 is an operation flowchart of the same embodiment, and FIGS. 4 is a flow chart showing the contents of the subroutine of FIG. 4, FIGS. 8 to 10 are waveform diagrams for explaining FIGS. 5 to 7, respectively, FIG. 11 is a diagram for explaining a standard data adjusting method, and FIG. 12 is this embodiment. It is a figure which shows the example of a display by. The same parts as those in FIG. 13 are designated by the same reference numerals.

The operation waveform diagnosing device for a robot shown in FIG. 1 has a robot body 1 for performing a predetermined work as in the conventional case.
And a controller 2 for controlling the same, and an amplifier 3 for amplifying signals and the like. Here, for example, a multi-axis robot having a plurality of drive axes including a servo motor is used as the robot main body 1, As the amplifier 3, for example, a digital servo amplifier is used. Robot body 1
Via the D / A converter 4 and the A / D converter 5
Connected with. As a result, the speed command or the position command from the controller 2 is amplified by the amplifier 3 and then D
Is sent to each axis of the robot body 1 via the / A converter 4,
As a result, the robot body 1 performs a predetermined operation, and the operation data of each axis at this time (for example, speed feedback data from the encoder, current data, etc.) is A / D.
After being input to the amplifier 3 via the converter 5, it is sent to the controller 2. The D / A converter 4 and the A / D converter 5 may be integrated with the amplifier 3.

Further, in this embodiment, the controller 2
Further, for example, the data terminal device of the workstation 10 is connected via the data communication interface 7 of RS-232C, for example. As a result, the operation data of the robot body 1 input to the controller 2 is transferred to the workstation 10 by data communication. A personal computer or the like may be used as the data terminal device instead of the workstation 10.

As shown in FIG. 2, the workstation 10 comprises a main body 11, a keyboard 12, and a display device 13 as a display means, and further the main body 1
Reference numeral 1 denotes a communication processing unit 15 for performing data communication between the input / output interface 14 and the controller 2, a data file 16 for storing various data, and a simulation unit for performing simulation-related processing to create simulation data. 17, a waveform feature extraction unit 18 as an operation waveform feature extraction unit that extracts a feature amount of the waveform of the action data that is captured by comparing with a reference waveform (specifically, simulation data or standard data as described later), As a standard data creating unit 19 as a standard data creating unit for creating standard data modeled by adding the above-mentioned feature amount to a reference waveform, and as a standard data adjusting unit for adjusting the standard data to the scale of operation data to be subjected to waveform diagnosis. Standard data adjusting unit 20, the adjusted standard data and the operation data Compared to and a preset waveform diagnostic comment table 21a to look-up while the waveform diagnosing section 21 serving as a waveform diagnosis means for performing a waveform diagnosis of the operation data. The data file 16 includes a waveform data history data file 16a as a storage unit for accumulating operation data taken in from the controller 2 and a failure history data file 16b for storing a history of operation data at the time of failure. The waveform diagnosis comment table 21a is a list of comments for the operation data added in the waveform diagnosis (see FIG. 3), and is stored in the memory in the waveform diagnosis unit 21, for example. This comment table is shown in FIG.
As shown in, it consists of a comment (qualitative or causal) for the velocity waveform and a comment (qualitative or causal) for the current waveform. These comments are related to the feature quantity of the operating waveform and appropriate by experiment. Set to.

The present apparatus thus constructed operates as follows in accordance with the flowcharts of FIGS. This flow chart is executed for each program of each axis of each robot.

First, when the operator performs simulation modeling in the simulation unit 17 in the workstation 10 (S1), the simulation unit 17 executes the simulation (S).
2) It is determined whether or not the initial model is appropriate by comparing it with the operation data of the robot body 1 (hereinafter referred to as an actual machine) that is actually operated (S3). If the result of this determination is not proper, the procedure returns to step 1 and steps 1 to 3
Repeat the operation of.

If the initial model is appropriate, the simulation section 17 calculates the values of various parameters of the block diagram for simulation for realizing the initial model, and the operation data for waveform diagnosis (for example, speed feedback signal). , Current value, position feedback signal, etc.)
Create simulation data corresponding to (S
4).

Then, the communication processing unit 15 in the workstation 10 causes various data when the actual machine 1 is operating (for example, a speed command output from the controller 2 to the actual machine 1,
Data such as velocity feedback signal and position feedback signal from encoders attached to each axis of the actual machine 1 and current values flowing through the servo motor of each axis are acquired by performing data communication with the controller 2 and waveform data history data. It is stored in the file 16a (S5). then,
It is judged whether or not the waveform data of the actual machine 1 has been fetched by the number required for creating the standard data (S6). If the required number has not been reached as a result of this judgment, the procedure returns to step 5, and steps 5 and 6 are executed. Repeat the operation of.

If the required number of waveform data of the actual machine 1 have been fetched, the waveform feature extraction unit 18 in the workstation 10 judges whether it is the initial standard data creation mode (7), and as a result of this judgment. If it is the initial standard data creation mode, proceed to step 8 to create standard data using the simulation data as a reference waveform. On the other hand, if it is not the initial standard data creation mode, that is, standard data has already been created. If so, step 1
In step 0, the standard data is used as a reference waveform to create new standard data, that is, the standard data is updated.

That is, in the initial standard data creation mode, the waveform feature extraction unit 18 compares the data of the actual machine 1 that has been taken in with the simulation data as the reference waveform, using the simulation data as the reference waveform. The characteristic amount of the waveform of the data is extracted (S8). The contents of this subroutine depend on the type of operation data shown in FIG.
~ As shown in Fig. 7.

FIG. 5 is a flow chart when the operation data is a velocity waveform. In this case, the waveform feature extraction unit 18
First, as shown in FIG. 8A, the speed command (dotted line)
The amount of delay of the speed feedback signal (solid line) with respect to, for example, the amount of the cross start point a, the target speed reaching point b, the cross end point c, etc. is extracted (S15), and then, as shown in FIG. The characteristic amount of the feedback waveform at the time of speed, for example, the amount of peak-peak size (average value and maximum value) d, vibration center position e, vibration period (average value) f, etc. is extracted (S16).

FIG. 6 is a flow chart when the operation waveform is a current waveform. In this case, as shown in FIG. 9A, the waveform feature extraction unit 18 first approximates the rising waveform g by a quadratic equation using the least squares method, and then extracts each coefficient ( S17), in order, as shown in FIG. 9B, extraction of the maximum current value h and the required time i until then (S18), extraction of the current value j at constant speed (S19), current during deceleration The value k is taken out (S20) and the current value m in the steady state is taken out (S21).

Further, FIG. 7 is a flow chart when the operation data is a position waveform. In this case, as shown in FIG. 10, the waveform feature extraction unit 18 first approximates the waveform p at the time of rising (at the time of acceleration) with a quadratic equation by using the least square method, and extracts each coefficient (S22). ), And also a minimum of 2
Each coefficient is extracted by approximating the waveform q at the time of deceleration by a quadratic equation using the multiplication method (S23).

When the characteristic amount of the waveform of each operation data of the actual machine 1 is extracted in this way, the standard data creating unit 19 in the workstation 10 adds the characteristic amount to the simulation data to generate the modeled standard data. Create (S9). Specifically, standard data is created by changing the parameters of the block diagram for simulation and reproducing the waveform of the operation data of the actual machine 1.
For example, when the delay in the rising of the speed feedback signal is large, the parameter value is changed to increase the static friction resistance.

On the other hand, when the mode is not the initial standard data creation mode, the standard data has already been created,
The secular change in the characteristics of the actual machine 1 becomes a problem. Therefore, in this case, the waveform feature extraction unit 18 compares the already-created standard data as the reference waveform with the captured data of the actual device 1 and the standard data as the reference waveform, and changes the operation data with time. Is extracted (S10), and the extracted secular change is added to the standard data to update the contents of the standard data (S11). The extraction of the secular change is specifically performed by extracting the characteristic amount of the waveform of each captured operation data, as in step 8, so that the contents of step 10 and step 11 are the simulation data as the reference waveform. The steps are the same as step 8 and step 9 except that they are used or standard data, and therefore detailed description thereof will be omitted.

When the standard data is newly created or updated, the standard data adjusting unit 20 in the workstation 10 uses, for example, the speed command as a reference among the operation data of the actual machine 1 to be subjected to waveform diagnosis. The various waveforms of the standard data are adjusted so as to match the scale of the waveform of the speed command as a reference (S12). Specifically, for example, first, the difference r in the time axis direction is based on the velocity waveform.
Then, the standard data is multiplied by a constant based on the difference r, then the difference s of the maximum values is taken, and the standard data is multiplied by a constant based on the difference s (see FIG. 11). Then, the current waveform and the like are also adjusted using the values of the differences r and s. 12
Shows standard data before adjustment (dotted line) and standard data after adjustment (dotted line).

When the standard data is adjusted in accordance with the actual data in this way, the waveform diagnostic section 21 in the workstation 10 compares the operation data of the actual machine 1 with the standard data scaled accordingly. Then, the waveform of the operation data of the actual machine 1 is diagnosed while looking up the comment table 21a (see FIG. 3), and a comment is made on the data such as whether the waveform state is good or bad, and which part is what causes the problem. Attach (S13). The comparison is performed by comparing the various characteristic amounts extracted in step 8 or step 10 with standard data.

When the waveform diagnosis is completed, the operation data of the actual machine 1 (solid line) and the adjusted standard data (dotted line) are commented via the display interface in the input / output interface 14, as shown in FIG. It is displayed on the screen of the display device 13 (S14).

Therefore, in this embodiment, since the workstation 10 is connected to the controller 2, the operation data of each axis of the robot body 1 can be stored in the data file 16a in the workstation 10 by data communication. it can. In addition, since standard data is created from these data and this standard data and actual data are automatically compared, it is possible to automatically diagnose the waveform of actual operation data and display the result. As a result, the knowledge of a waveform diagnosis expert is shared, and the efficiency of the operation waveform diagnosis work of the robot is dramatically improved.

[0028]

As described above, according to the present invention, the waveform of the actual operation data of the robot body can be automatically diagnosed and the result can be displayed. Therefore, the efficiency of the operation of diagnosing the operation waveform of the robot can be improved. Is dramatically improved.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of the present embodiment.

FIG. 2 is a system configuration diagram of a main part of FIG.

FIG. 3 is a diagram showing an example of a comment table.

FIG. 4 is an operation flowchart of the embodiment.

5 is a flowchart showing the contents of the subroutine of FIG.

FIG. 6 is a flowchart showing the contents of the subroutine of FIG.

FIG. 7 is a flowchart showing the contents of the subroutine of FIG.

FIG. 8 is a waveform diagram used to explain FIG.

FIG. 9 is a waveform diagram used to explain FIG.

FIG. 10 is a waveform diagram used to explain FIG. 7.

FIG. 11 is a diagram for explaining a method of adjusting standard data.

FIG. 12 is a diagram showing a display example according to the present embodiment.

FIG. 13 is a schematic configuration diagram showing an example of a conventional robot system.

FIG. 14 is a diagram showing a display example of operation data.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Robot main body 2 ... Controller 3 ... Amplifier 10 ... Workstation 13 ... Display device (display means) 16a ... Waveform data history data file (storage means) 18 ... Waveform feature extraction part (motion waveform feature extraction means) 19 ... Standard Data creation unit (standard data creation means) 20 ... Standard data adjustment unit (standard data adjustment means) 21 ... Waveform diagnosis unit (waveform diagnosis means)

Claims (1)

[Claims] 1. A robot main body that performs a predetermined work, a controller that controls the robot main body, a storage unit that communicates with the controller and stores a history of operation data of the robot main body, and compares the reference waveform with a reference waveform. An operation waveform feature extraction unit that extracts a feature amount of a waveform of the operation data, a standard data creation unit that creates standard data modeled on the basis of the reference waveform and the feature amount, and the standard data as the action. Standard data adjusting means for adjusting to the data, and a waveform for performing waveform diagnosis of the operation data while comparing the adjusted standard data with the operation data and looking up a preset waveform diagnosis comment table. A waveform of an industrial robot, comprising: a diagnostic means, and a display means for displaying a result of the waveform diagnosis. Cross-sectional devices.