Life prediction device, method for using the life prediction device, life prediction method, life prediction program, anomaly detection device, anomaly detection method, and anomaly detection program

Abstract

To improve the accuracy of life prediction for robots equipped with gas springs. [Solution] A life prediction device that predicts the lifespan of a robot equipped with a gas spring based on information about the drive motor includes: an input unit that receives a signal indicating that gas replenishment has been performed; a replenishment timing storage unit that stores the replenishment timing based on the signal input; a current data acquisition unit that acquires current time series data; a feature data acquisition unit that acquires and stores feature time series data; a time series data correction unit that corrects a series of feature time series data so that the progression of a series of feature time series data before the most recent replenishment timing matches the initial feature value after replenishment, which is obtained from one or more initial features after the most recent replenishment timing; and a life prediction unit that creates a first trend line based on the corrected series of feature time series data and determines the timing at which the first trend line reaches a life threshold as the first predicted life timing.

Description

Technical Field

[0001] The present disclosure relates to a life prediction device, a method for using the life prediction device, a life prediction method, a life prediction program, an abnormality determination device, an abnormality determination method, and an abnormality determination program.

Background Art

[0002] Conventionally, a robot maintenance support device for predicting the life of a drive system of a robot and assisting maintenance has been known. Patent Document 1 discloses this type of robot maintenance support device.

[0003] The robot maintenance support device of Patent Document 1 includes a life determination means for determining a period until the current command value reaches a preset value based on a future change trend of the current command value of a servo motor (or a drive motor) constituting the drive system of the robot.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Incidentally, some robots are equipped with gas springs to reduce the load on the drive motor. If the robot maintenance support device described in Patent Document 1 is applied directly to a robot equipped with such a gas spring, the following problems may arise. That is, the gas filled inside the gas spring leaks out with use, and the force it generates decreases over time. When the force generated by the gas spring, i.e., the force assisting the drive motor, decreases, it becomes necessary to increase the output of the drive motor. Increasing the output of the drive motor is achieved by increasing the current command value (more directly, the current value). Therefore, in a robot equipped with a gas spring, the current command value of the drive motor may fluctuate due to factors that are not directly related to the lifespan of the drive system, such as gas leakage in the gas spring. For this reason, if the robot maintenance support device described in Patent Document 1 is applied directly to a robot equipped with a gas spring, the accuracy of lifespan prediction may decrease.

[0006] This disclosure is made in view of the above circumstances, and its purpose is to improve the accuracy of life prediction in robots equipped with gas springs. [Means for solving the problem]

[0007] The problems that this disclosure aims to solve are as described above, and next we will explain the means and effects of solving these problems.

[0008] According to a first aspect of this disclosure, a life prediction device is provided having the following configuration: The life prediction device predicts the life of a robot comprising an arm, a drive motor for driving the arm, and a gas spring for supporting the load acting on the arm and reducing the load on the drive motor, based on information about the drive motor. The life prediction device comprises an input unit, a replenishment timing storage unit, a current data acquisition unit, a feature data acquisition unit, a time series data correction unit, and a life prediction unit. The input unit receives a signal indicating that gas has been replenished to the gas spring. The replenishment timing storage unit stores the replenishment timing, which is the timing at which the gas was replenished, based on the input of the signal to the input unit. The current data acquisition unit acquires current time series data, which is time series data of information about the drive current of the drive motor. The feature data acquisition unit acquires and stores feature time series data, which is time series data of the features of the current time series data. The time-series data correction unit corrects the series of feature time-series data so that the progression of the series of feature time-series data prior to the most recent replenishment timing matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing. The lifetime prediction unit creates a first trend line based on the corrected series of feature time-series data and determines the timing at which the first trend line reaches a predetermined lifetime threshold as the first predicted lifetime timing.

[0009] A second aspect of this disclosure provides a life prediction device having the following configuration: The life prediction device predicts the life of a robot comprising an arm, a drive motor that drives the arm, and a gas spring that supports the load acting on the arm and reduces the load on the drive motor, based on information about the drive motor. The life prediction device comprises a current data acquisition unit, a current data correction unit, a corrected current data storage unit, a corrected feature data acquisition unit, and a life prediction unit. The current data acquisition unit acquires current data, which is data relating to the drive current of the drive motor. The current data correction unit corrects the current data based on internal pressure information, which is information relating to the internal pressure in the gas spring, and position information, which is information relating to the position of the arm. The corrected current data storage unit stores corrected current time series data, which is time series data of the corrected current data. The corrected feature data acquisition unit acquires corrected feature time series data, which is time series data of the features of the corrected current time series data. The lifetime prediction unit creates a trend line based on the corrected feature time series data and determines the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing.

[0010] According to a third aspect of this disclosure, a method of using a life prediction device is provided as follows: That is, this method of using a life prediction device is a method of using a life prediction device according to the first or second aspect described above. In the above method of use, if the gas spring is refilled with gas multiple times, the multiple gas refills are performed with the arms aligned.

[0011] A fourth aspect of this disclosure provides a method for using a life prediction device, which is the same as the method for using a life prediction device according to the first or second aspect described above. In this method, if the gas spring is replenished multiple times, each of the multiple replenishments is performed after determining a target internal pressure value, which is a target value for the internal pressure of the gas spring, based on the surface temperature of the gas spring at the time of the replenishment. In this method, the target internal pressure values ​​are set so that the total amount of gas in the gas spring is the same or approximately the same after each of the multiple replenishments has been performed.

[0012] According to a fifth aspect of this disclosure, the following lifetime prediction method is provided. That is, this lifetime prediction method is a lifetime prediction method that predicts the life of a robot comprising an arm, a drive motor that drives the arm, and a gas spring that supports the load acting on the arm and reduces the load on the drive motor, based on information about the drive motor. A signal is input to indicate that gas has been replenished to the gas spring. Based on the input of the signal, the replenishment timing, which is the timing at which the gas replenishment was performed, is stored. Current time series data, which is time series data of information about the drive current of the drive motor, is acquired. Feature time series data, which is time series data of the feature quantities of the current time series data, is acquired and stored. The series of feature time series data is corrected so that the progression of the series of feature time series data before the most recent replenishment timing matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing. A first trend line is created based on the corrected series of feature time series data, and the timing at which the first trend line reaches a predetermined lifetime threshold is determined as the first predicted lifetime timing.

[0013] According to a sixth aspect of this disclosure, the following life prediction method is provided. That is, this life prediction method is a life prediction method that predicts the life of a robot comprising an arm, a drive motor that drives the arm, and a gas spring that supports the load acting on the arm and reduces the load on the drive motor, based on information about the drive motor. Current data, which is data about the drive current of the drive motor, is acquired. The current data is corrected based on internal pressure information, which is information about the internal pressure in the gas spring, and position information, which is information about the position of the arm. Corrected current time series data, which is time series data of the corrected current data, is stored. Corrected feature time series data, which is time series data of the feature quantities of the corrected current time series data, is acquired. A trend line is created based on the corrected feature time series data, and the timing at which the trend line reaches a predetermined life threshold is determined as the predicted life timing.

[0014] According to a seventh aspect of this disclosure, a life prediction program having the following configuration is provided. That is, the life prediction program predicts the life of a robot comprising an arm, a drive motor that drives the arm, and a gas spring that supports the load acting on the arm and reduces the load on the drive motor, based on information about the drive motor. The life prediction program causes a computer to perform a signal input step, a replenishment timing storage step, a current data acquisition step, a feature data acquisition step, a time series data correction step, and a life prediction step. In the signal input step, a signal is input to indicate that gas has been replenished to the gas spring. In the replenishment timing storage step, the replenishment timing, which is the timing at which the gas was replenished, is stored based on the input of the signal. In the current data acquisition step, current time series data, which is time series data of information about the drive current of the drive motor, is acquired. In the feature data acquisition step, feature time series data, which is time series data of the features of the current time series data, is acquired and stored. In the time series data correction step, the series of feature time series data is corrected so that the progression of the series of feature time series data prior to the most recent replenishment timing matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing. In the lifetime prediction step, a first trend line is created based on the corrected series of feature time series data, and the timing at which the first trend line reaches a predetermined lifetime threshold is determined as the first predicted lifetime timing.

[0015] According to the eighth aspect of this disclosure, a lifetime prediction program having the following configuration is provided. That is, the lifetime prediction program predicts the lifetime of a robot comprising an arm, a drive motor that drives the arm, and a gas spring that supports the load acting on the arm and reduces the load on the drive motor, based on information about the drive motor. The lifetime prediction program causes a computer to perform a current data acquisition step, a current data correction step, a corrected current data storage step, a corrected feature data acquisition step, and a lifetime prediction step. In the current data acquisition step, current data, which is data relating to the drive current of the drive motor, is acquired. In the current data correction step, the current data is corrected based on internal pressure information, which is information relating to the internal pressure in the gas spring, and position information, which is information relating to the position of the arm. In the corrected current data storage step, corrected current time series data, which is time series data of the corrected current data, is stored. In the corrected feature data acquisition step, corrected feature time series data, which is time series data of the features of the corrected current time series data, is acquired. In the lifetime prediction step, a trend line is created based on the corrected feature time series data, and the timing at which the trend line reaches a predetermined lifetime threshold is determined as the predicted lifetime timing.

[0016] As a result, even if gas leaks from the gas spring during use, causing fluctuations in the drive motor's drive current, the effects of these fluctuations can be substantially eliminated through a predetermined correction process, allowing for accurate life prediction.

[0017] According to a ninth aspect of this disclosure, a gas spring life prediction device is provided having the following configuration. That is, the gas spring life prediction device predicts the life of a gas spring in a robot comprising an arm, a drive motor, a gas spring, and a pressure detector. The drive motor drives the arm. The gas spring supports the load acting on the arm and reduces the load on the drive motor. The pressure detector detects the internal pressure of the gas spring. The gas spring life prediction device comprises an internal pressure information acquisition unit, an internal pressure information correction unit, a corrected internal pressure information storage unit, a corrected internal pressure feature data acquisition unit, and a gas spring life prediction unit. The internal pressure information acquisition unit acquires the internal pressure detected by the pressure detector. The internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. The corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit. The Corrected Internal Pressure Feature Data Acquisition Unit acquires Corrected Internal Pressure Feature Time Series Data, which is time series data of the feature quantities of the Corrected Internal Pressure Time Series Data stored in the Corrected Internal Pressure Information Storage Unit. The Gas Spring Life Prediction Unit creates a trend line based on the Corrected Internal Pressure Feature Time Series Data and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing.

[0018] This allows for the correction of the impact on internal pressure caused by changes in the arm's position, thus enabling a better maintenance plan for the gas spring.

[0019] According to a tenth aspect of this disclosure, a gas spring abnormality detection device is provided having the following configuration. That is, this gas spring abnormality detection device determines whether or not there is an abnormality in the gas spring of a robot comprising an arm, a drive motor, a gas spring, and a pressure detector. The drive motor drives the arm. The gas spring supports the load acting on the arm and reduces the load on the drive motor. The pressure detector detects the internal pressure of the gas spring. The gas spring abnormality detection device comprises an internal pressure information acquisition unit, an internal pressure information correction unit, a corrected internal pressure information storage unit, a corrected internal pressure feature acquisition unit, and a gas spring abnormality detection unit. The internal pressure information acquisition unit acquires the internal pressure detected by the pressure detector. The internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. The corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit. The corrected internal pressure feature acquisition unit acquires the corrected internal pressure feature stored in the corrected internal pressure information storage unit. The gas spring abnormality determination unit compares the corrected internal pressure feature acquired by the corrected internal pressure feature acquisition unit with a preset threshold and determines whether or not there is an abnormality in the gas spring based on the comparison result.

[0020] This allows for the correction of the effect on internal pressure caused by changes in the arm's position, enabling accurate determination of whether or not there is a problem with the gas spring.

[0021] According to an eleventh aspect of this disclosure, a gas spring life prediction device is provided having the following configuration: The gas spring life prediction device predicts the life of a gas spring in a robot comprising an arm, a drive motor, a gas spring, a control unit, and a pressure detector. The drive motor drives the arm. The gas spring supports the load acting on the arm and reduces the load on the drive motor. The control unit controls the drive motor. The pressure detector detects the internal pressure of the gas spring each time a detection instruction signal is generated inside the control unit, or each time a control signal is input from the control unit. The gas spring life prediction device comprises an internal pressure information acquisition unit, an internal pressure information correction unit, a corrected internal pressure information storage unit, and a gas spring life prediction unit. The internal pressure information acquisition unit acquires the internal pressure detected by the pressure detector. The internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. The corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit. The gas spring life prediction unit creates a trend line based on the time-series data of the corrected internal pressure stored in the corrected internal pressure information storage unit, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing.

[0022] This allows for the correction of the impact on internal pressure caused by changes in the arm's position, thus enabling a better maintenance plan for the gas spring.

[0023] According to a twelfth aspect of the present disclosure, a gas spring abnormality determination device having the following configuration is provided. That is, this gas spring abnormality determination device determines the presence or absence of an abnormality in the gas spring of a robot including an arm, a drive motor, a gas spring, a control unit, and a pressure detector. The drive motor drives the arm. The gas spring supports the load acting on the arm to reduce the load on the drive motor. The control unit controls the drive motor. The pressure detector detects the internal pressure of the gas spring every time a detection instruction signal is generated inside the control unit or every time a control signal is input from the control unit. The gas spring abnormality determination device includes an internal pressure information acquisition unit, an internal pressure information correction unit, and a gas spring abnormality determination unit. The internal pressure information acquisition unit acquires the internal pressure detected by the pressure detector. The internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. The gas spring abnormality determination unit compares the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, with a preset threshold value, and determines the presence or absence of an abnormality in the gas spring based on the comparison result.

[0024] Thereby, since the influence on the internal pressure due to the change in the position of the arm can be removed by correction, the presence or absence of an abnormality in the gas spring can be determined favorably.

[0025] According to the 13th aspect of the present disclosure, a gas spring life prediction device having the following configuration is provided. That is, this gas spring life prediction device predicts the life of the gas spring of a robot including an arm, a drive motor, and a gas spring. The drive motor drives the arm. The gas spring supports the load acting on the arm to reduce the load on the drive motor. The gas spring life prediction device includes an estimated internal pressure information acquisition unit, an estimated internal pressure information storage unit, and a gas spring life prediction unit. The estimated internal pressure information acquisition unit acquires an estimated internal pressure which is the estimated internal pressure of the gas spring. The estimated internal pressure can be estimated from position information which is information regarding the position of the arm and current data which is data of information regarding the drive current of the drive motor. The estimated internal pressure can also be estimated from the operation information of the gas spring obtained from the position information. The estimated internal pressure information storage unit stores the time series data of the estimated internal pressure acquired by the estimated internal pressure information acquisition unit. The gas spring life prediction unit creates a trend line based on the time series data of the estimated internal pressure stored by the estimated internal pressure information storage unit, and obtains the timing when the trend line reaches a predetermined life threshold as the predicted life timing.

[0026] Thereby, since the life of the gas spring can be predicted without providing a pressure detector, the maintenance plan of the gas spring can be determined at low cost.

[0027] Each of the devices provided in the above-described 9th to 13th aspects can also be grasped as a technical idea of a method and a program.

Effects of the Invention

[0028] According to the present disclosure, the accuracy of life prediction in a robot including a gas spring can be improved.

Brief Description of the Drawings

[0029] [Figure 1]A front view showing the configuration of a robot system according to one embodiment of this disclosure. [Figure 2] A block diagram schematically showing the electrical configuration of the life prediction device of the first embodiment. [Figure 3] A graph showing an example of current time series data. [Figure 4] This graph shows an example of feature time series data, with the horizontal axis representing time and the vertical axis representing features. [Figure 5] A graph illustrating the correction of a series of feature time-series data prior to the most recent replenishment timing in the first embodiment. [Figure 6] A graph illustrating the correction of a series of feature time-series data after the most recent interpolation timing in the first embodiment. [Figure 7] A graph showing the display configuration of the trend line in the first embodiment. [Figure 8] This graph shows the relationship between the surface temperature of a gas spring and the specified pressure, with the horizontal axis representing surface temperature and the vertical axis representing the specified pressure. [Figure 9] A graph illustrating the correction of a series of feature time-series data prior to the most recent replenishment timing in the second embodiment. [Figure 10] A graph illustrating the correction of a series of feature time-series data after the most recent replenishment timing in the second embodiment. [Figure 11] A graph showing the display configuration of the trend line in the second embodiment. [Figure 12] A block diagram schematically showing the electrical configuration of the life prediction device of the third embodiment. [Figure 13] A front view illustrating the dimensions of each part of the robot. [Figure 14] This graph shows the display method of the trend line in the third embodiment, and also shows the feature time series data without correcting the current data for reference. [Figure 15] A block diagram schematically showing the electrical configuration of the gas spring life prediction device of the fourth embodiment. [Figure 16]A graph showing the display configuration of the trend line in the fourth embodiment. [Figure 17] A block diagram schematically showing the electrical configuration of the gas spring abnormality detection device of the fifth embodiment. [Figure 18] A block diagram schematically showing the electrical configuration of the gas spring life prediction device of the sixth embodiment. [Figure 19] A block diagram schematically showing the electrical configuration of the gas spring abnormality detection device of the seventh embodiment. [Figure 20] A front view showing the configuration of the robot system according to the eighth embodiment. [Figure 21] A block diagram schematically showing the electrical configuration of the gas spring life prediction device of the eighth embodiment. [Modes for carrying out the invention]

[0030] Next, embodiments of the present disclosure will be described with reference to the drawings. Figure 1 is a front view showing the configuration of a robot system 10 according to one embodiment of the present disclosure. Figure 2 is a schematic block diagram showing the electrical configuration of the life prediction device 40 of the first embodiment. Figure 3 is a graph showing an example of current time series data.

[0031] As shown in Figures 1 and 2, the robot system 10 according to the first embodiment includes a robot 20, a controller 30, and a life prediction device 40.

[0032] Robot 20 performs tasks such as painting, cleaning, welding, and transporting on the workpiece. Robot 20 is, for example, a vertical articulated robot. Robot 20 comprises a base unit 21, a swivel unit 22, a swivel motor 23, an arm 24, a drive motor 25, and a gas spring 26.

[0033] The base portion 21 is a part that constitutes the base of the robot 20 and is fixed to the ground (for example, the floor of a factory).

[0034] The swivel section 22 is rotatably mounted to the base section 21. The swivel section 22 is rotatable around a vertical axis. The swivel section 22 is driven by a swivel motor 23.

[0035] The slewing motor 23 rotates the slewing section 22 relative to the base section 21. As this rotation occurs, the arm 24 attached to the slewing section 22 also rotates relative to the base section 21. The slewing motor 23 in this embodiment has a rotation axis extending vertically. The rotor (not shown) of the slewing motor 23 rotates, causing the slewing section 22 and the arm 24 to rotate around the vertical axis relative to the base section 21 via a reduction gear (not shown). The slewing motor 23 may be a servo motor. The slewing motor 23 is electrically connected to a controller 30, and its operation is controlled by the controller 30.

[0036] Arm 24 is attached to a swivel section 22, and via this swivel section 22, it is attached to a base section 21 so as to be able to rotatably about a vertical axis. In Figure 1, only the base end of arm 24 and its vicinity are shown, but one or more arms (not shown) are swivelably attached to the tip of arm 24. Arm 24 and the arms (not shown) constitute a multi-joint arm, and an end effector (not shown) that performs work on the workpiece is attached to its tip via a wrist section (not shown).

[0037] The drive motor 25 is attached to the joint between the swivel section 22 and the arm 24 via a reduction gear (not shown). The drive motor 25 in this embodiment has a rotation axis that extends horizontally. As the rotor (not shown) of the drive motor 25 rotates, the arm 24 swivels relative to the swivel section 22 about the horizontal axis. The configuration of the drive motor 25 is arbitrary, but in this embodiment it is a servo motor. The drive motor 25 is electrically connected to the controller 30, and its operation is controlled by the controller 30. The drive motor 25 is equipped with a current sensor (not shown) for detecting the drive current. The drive motor 25 is also equipped with a position sensor (not shown) for detecting the angular position of the rotor. The detected value from the current sensor is sent to the controller 30 and the life prediction device 40. The detected value from the position sensor is sent to the controller 30 and used to determine the angular position of the arm 24.

[0038] The gas spring 26 has a cylinder 26a and a piston rod 26b. One end of the gas spring 26 (or the end of the cylinder 26a) is rotatably attached to the base 21. The other end of the gas spring 26 (or the end of the piston rod 26b) is rotatably attached to the arm 24. The cylinder 26a of the gas spring 26 is filled with compressed gas (for example, nitrogen gas), and the pressure of the compressed gas (hereinafter also referred to as the internal pressure of the gas spring 26 or simply internal pressure) acts on the piston rod 26b, generating a force that supports the load (or gravity) acting on the arm 24. This supporting force of the gas spring 26 reduces the load on the drive motor 25. The gas spring 26 is provided with a pressure sensor (not shown) for detecting the internal pressure. The detected value of the pressure sensor (i.e., information on the internal pressure) is sent to the controller 30 and the life prediction device 40. However, in this embodiment, the pressure sensor may be omitted.

[0039] The support force of the gas spring 26 is smallest, for example, when the arm 24 is in the upright position shown in Figure 1, and increases as the rotation angle of the arm 24 around the horizontal axis from the upright position increases. Also, the support force of the gas spring 26 decreases over time due to leakage of internal gas during use. Therefore, the gas spring 26 needs to be replenished with gas periodically or as needed. The support force of the gas spring 26 changes relatively significantly before and after gas replenishment. That is, the support force immediately after gas replenishment is greater than the support force immediately before gas replenishment. The output of the drive motor 25 also fluctuates in accordance with such changes in support force. That is, the output of the drive motor 25 immediately after gas replenishment is less than the output of the drive motor 25 immediately before gas replenishment. The life prediction device 40 of this embodiment predicts the life based on information regarding the drive current of the drive motor 25, as described later, and can suppress the influence of such fluctuations in the output of the drive motor 25 (i.e., fluctuations in the drive current) on the life prediction.

[0040] The controller 30 is configured as a known computer, for example, equipped with a CPU, ROM, RAM, auxiliary storage device, etc. The auxiliary storage device is, for example, an HDD or SSD. The auxiliary storage device stores programs for operating the robot 20, etc.

[0041] As shown in Figure 1, the life prediction device 40 is connected to the controller 30. The life prediction device 40 acquires information about the drive motor 25 of the robot 20 via the controller 30 and predicts the life (for example, the life of the drive system including the drive motor 25 and the reduction gear) based on the acquired information. As shown in Figure 2, the life prediction device 40 includes an input unit 41, a replenishment timing storage unit 42, a current data acquisition unit 43, a feature data acquisition unit 44, a time series data correction unit 45, a life prediction unit 46, and a display unit 47.

[0042] The life prediction device 40 is configured as a known computer equipped with, for example, a CPU, ROM, RAM, auxiliary storage device, etc. The auxiliary storage device is, for example, an HDD, SSD, etc. The auxiliary storage device stores a program for life prediction (i.e., the life prediction program according to this embodiment), etc. Through the cooperation of this hardware and software, the computer can be operated as an input unit 41, a replenishment timing storage unit 42, a current data acquisition unit 43, a feature quantity data acquisition unit 44, a time series data correction unit 45, a life prediction unit 46, a display unit 47, etc. The life prediction program realizes the life prediction method of this embodiment.

[0043] The input unit 41 receives a signal indicating that gas has been replenished to the gas spring 26. Such a signal may be input to the input unit 41 via the controller 30 by an operator operating a teaching pendant, for example, or it may be input to the input unit 41 by operating a separately provided switch.

[0044] The replenishment timing storage unit 42 stores the replenishment timing, which is the timing at which gas replenishment is performed, based on the input of the above signal to the input unit 41. The replenishment timing storage unit 42 may store the timing at which the signal was input as the replenishment timing, or it may store any timing after the timing at which the signal was input as the replenishment timing. The replenishment timing may be stored in date format, in date and time format, or in the format of elapsed time starting from the timing at which the robot 20 is put into use. It is preferable that the replenishment timing storage unit 42 can store multiple replenishment timings.

[0045] The current data acquisition unit 43 stores current time series data, which is time series data related to the drive current of the drive motor 25. In a simple example, a special program is predetermined to operate the robot 20 for life prediction, and by executing this program, the robot 20 reproduces a certain operation pattern for life prediction (hereinafter sometimes referred to as the evaluation operation pattern). The current time series data when the robot 20 operates according to the evaluation operation pattern is stored in the current data acquisition unit 43. Alternatively, the evaluation operation pattern may be automatically detected from the actual operations performed by the robot 20, and the portion corresponding to the evaluation operation pattern may be extracted from the current time series data during the actual operation and stored in the current data acquisition unit 43.

[0046] Figure 3 shows an example of current time-series data. As shown in this graph, in this embodiment, the information regarding the drive current is the detected value of the current sensor described above (i.e., the measured value of the drive current). In this example, the time-series data is a collection of numerous current values ​​obtained by repeatedly detecting the drive current at short, constant time intervals, arranged in chronological order. However, the information regarding the drive current may also be a current command value or a torque command value for the drive motor 25, or a torque monitor value based on the measured current value of the drive motor 25. Furthermore, the information regarding the drive current may also be the position deviation, which is the difference between the measured position of the drive motor 25 (i.e., the detected value of the position sensor described above) and the position command value. The servo driver provides the servo motor with a current command value obtained by multiplying this deviation by a gain, so the trend of the rotational position deviation shows a similar trend to the trend of the current command value. Therefore, the rotational position deviation can also be considered as a type of information regarding the drive current.

[0047] These values, which are available as information regarding the drive current, may be used after any processing has been applied to them (for example, by multiplying them by a predetermined coefficient, or by filtering to remove noise signals).

[0048] The feature data acquisition unit 44 acquires and stores feature time series data, which is time series data of the features of the current time series data. In this embodiment, one feature is obtained from one current time series data as illustrated in Figure 3. Feature time series data can be obtained by calculating a feature for each current time series data that is acquired periodically or irregularly (for example, once a day).

[0049] The features can be any features extracted from the current time series data. For example, the features of the current time series data can be the root mean square (hereinafter also referred to as I2), peak value, peak-to-peak value, or frequency analysis integrated value obtained from the current time series data.

[0050] The peak value is the maximum value of the current waveform. The peak-to-peak value is the value obtained by subtracting the current value of the lower peak from the current value of the higher peak of the current waveform. The maximum value or peak may be the maximum value on the positive side of the current waveform, or the maximum absolute value of the current. The frequency analysis integrated value is proposed as one method of frequency analysis. The frequency analysis integrated value is, for example, the sum of the amplitude spectrum, power spectrum, or power spectral density from 0 or a few Hz to tens of Hz obtained by frequency analysis. The average value may be used instead of the sum.

[0051] The feature of the current time series data may be the dissimilarity between the current time series data and the current time series data acquired by the drive motor 25 immediately after the start of use. The dissimilarity may be determined, for example, based on the DTW distance or Euclidean distance between the two sets of current time series data. For example, the dissimilarity may be determined based on the sum or average of the DTW distance or Euclidean distance.

[0052] The Euclidean distance is a common method for determining the dissimilarity of waveforms, so its explanation will be omitted. DTW is an abbreviation for Dynamic Time Warping. The DTW method is well known, so it will be briefly explained, but in the DTW method, the DTW distance is obtained, which indicates the degree to which two time series data are dissimilar, while substantially allowing for nonlinear stretching and compression in the time axis direction.

[0053] The following explanation will use the case where the feature in question is I2 as an example.

[0054] The time-series data correction unit 45 corrects the series of feature time-series data so that the progression of the series of feature time-series data prior to the most recent replenishment timing (in this example, the timing shown as tt1 in the graph in Figure 4, etc.) matches the post-replenishment initial feature value Va, which is obtained from one or more features in the early stages after the most recent replenishment timing t1. Here, the post-replenishment initial feature value Va may be, for example, the value of one feature immediately after the replenishment timing t1, or it may be the mean or median of a predetermined number of features (for example, three or more or five or fewer) in the early stages after the replenishment timing t1. In this embodiment, the post-replenishment initial feature value Va is the value of one feature immediately after the replenishment timing t1.

[0055] The correction methods described in the previous paragraph will be explained in detail below with reference to Figures 4 to 6. Figure 4 is a graph showing an example of feature time series data, where the horizontal axis is time and the vertical axis is the feature (I2). Figure 5 is a graph to explain the correction of a series of feature time series data prior to the most recent replenishment timing t1 in this embodiment. Figure 6 is a graph to explain the correction of a series of feature time series data after the most recent replenishment timing t1 in this embodiment.

[0056] As shown in Figure 4, the feature quantity (I2) increases over time. This is due to the aging deterioration of the drive system, including the drive motor 25, as well as the decrease in support capacity due to gas leakage from the gas spring 26. The feature quantity decreases significantly before and after the replenishment timings t1 and t2. This large change occurs because the support capacity of the gas spring 26 is restored by gas replenishment. The feature quantity immediately after replenishment timing t1 is slightly larger than the feature quantity located furthest to the left in this graph, and the feature quantity immediately after replenishment timing t2 is slightly larger than the feature quantity immediately after replenishment timing t1. This small difference in feature quantities is thought to be due to the aging deterioration of the drive system, including the drive reducer and the drive motor 25. Examples of aging deterioration of the drive system include a decrease in the efficiency of the reducer and a decrease in the torque constant due to demagnetization of the drive motor 25. The correction of the series of feature quantity time-series data by the life prediction device 40 is performed in order to separate and consider the aging deterioration of the drive system, including the drive motor 25, from the decrease in support capacity due to gas leakage from the gas spring 26.

[0057] Specifically, as shown in Figure 5, the time-series data correction unit 45 corrects the series of feature time-series data before the supplementation timing t1 by multiplying the difference between each feature and the initial feature value Ve before supplementation by the slope ratio k1, and then adding the initial feature value Ve before supplementation to the resulting value. In Figure 5, the series of feature time-series data before correction are shown with white circles, and the series of feature time-series data after correction are shown with black circles. Also in Figure 5, arrows are added to some of the series of features to show the image of the correction.

[0058] Here, the pre-refill initial feature value Ve, as mentioned in the previous paragraph, is obtained from one or more features at the beginning of a series of feature time-series data prior to the refill timing t1, as shown in the graphs of Figures 4 and 5. For example, the pre-refill initial feature value Ve may be the value of the first feature in the series of feature time-series data, or it may be the mean or median of a predetermined number (e.g., 3 or more, and 5 or less) of features at the beginning of the series of feature time-series data. In this embodiment, the pre-refill initial feature value Ve is the value of the first feature in the series of feature time-series data. Note that in graphs other than Figure 4, the number of feature time-series data is smaller than in Figure 4 for easier viewing.

[0059] Furthermore, in the graphs of Figures 4 and 5, the pre-refill final feature value Vb is defined as a value obtained from one or more features at the end of a series of feature time-series data prior to the refill timing t1. For example, the pre-refill final feature value Vb may be the value of the last feature in the series of feature time-series data, or it may be the mean or median of a predetermined number of features (e.g., 3 or more, and 5 or less) at the end of the series of feature time-series data. Alternatively, a trend line approximating the series of feature time-series data prior to the refill timing t1 can be created, and the value indicated by this trend line at refill timing t1 can be defined as the pre-refill final feature value Vb. In this embodiment, the pre-refill final feature value Vb is the value of the last feature in the series of feature time-series data.

[0060] Furthermore, the above slope ratio k1 is the ratio of the post-refill slope Sb to the pre-refill slope Sa (k1 = Sb / Sa), and usually takes a value less than 1. Here, the pre-refill slope Sa is the slope of the first line L1 that passes through the point of the pre-refill early feature value Ve and the point of the pre-refill late feature value Vb in the graphs of Figures 4 and 5. The post-refill slope Sb is the slope of the second line L2 that passes through the point of the pre-refill early feature value Ve and the point of the post-refill early feature value Va in the graphs of Figures 4 and 5.

[0061] To explain the correction in Figure 5 from a different perspective, first, the series of feature time series data prior to the replenishment timing t1 is moved downward parallel to the vertical axis so that the initial feature value Ve (in this example, the first feature in the series of feature time series data) is placed on the horizontal axis. In other words, the series of feature time series data is moved downward in the graph by the initial feature value Ve. Then, by multiplying all the features included in the series of feature time series data by the slope ratio k1, the series of feature time series data is corrected so that the slope of its approximation line becomes smaller. The slope of the approximation line before this correction corresponds to the slope of the first line L1, and the slope of the approximation line after the correction corresponds to the slope of the second line L2. Finally, by moving the series of feature time series data corrected in this way upward to reverse the aforementioned downward movement, the series of time series data shown by the black circles in Figure 5 can be obtained. The corrections described in the previous four paragraphs are substantially equivalent to the correction in this paragraph.

[0062] Furthermore, as shown in Figure 6, the time-series data correction unit 45 corrects the series of feature time-series data after the replenishment timing t1 by multiplying the difference between each of the features included in the series of feature time-series data after the replenishment timing t1 and the initial feature value Va after replenishment by the slope ratio k1, and adding the initial feature value Va after replenishment to the multiplied value. In Figure 6, the series of feature time-series data before correction are shown with white circles, and the series of feature time-series data after correction are shown with black circles. Also, as shown in the same figure, the initial feature value Va after replenishment is the first feature after the replenishment timing t1. The explanation corresponding to the correction graph in Figure 6 is the same as in the previous paragraph and will be omitted. The correction explained in Figure 6 is based on the idea that the gas leakage tendency of the gas spring 26 after the replenishment timing t1 is substantially the same as before the replenishment timing t1.

[0063] As shown in the upper graph of Figure 7, the lifetime prediction unit 46 creates a first trend line TL1 based on a series of feature time series data prior to the corrected most recent replenishment timing t1, and determines the timing at which the first trend line TL1 reaches a predetermined lifetime threshold th as the first predicted lifetime timing LT1. The lifetime prediction unit 46 also creates a second trend line TL2 based on the feature time series data after the corrected most recent replenishment timing t1, and determines the timing at which the second trend line TL2 reaches a predetermined lifetime threshold th as the second predicted lifetime timing LT2. The first trend line TL1 and the second trend line TL2 can be determined by known methods, such as the least squares method. The lifetime prediction unit 46 may, for example, predict the lifetime of the robot's drive system based on which of the first predicted lifetime timing LT1 and the second predicted lifetime timing LT2 is closer to the present time. However, the usage of the first predicted lifetime timing LT1 and the second predicted lifetime timing LT2 is not limited to this.

[0064] The display unit 47 displays information related to lifetime prediction. The display unit 47 may be any type of display. Figure 7 shows how the display content changes over time. When the number of features acquired after the most recent replenishment timing t1 is below a predetermined number, the display unit 47 displays the first trend line TL1 and the first predicted lifetime timing LT1, as shown in the upper graph of Figure 7, and simultaneously displays the second trend line TL2 and the second predicted lifetime timing LT2. As time passes and the number of features acquired after replenishment timing t1 exceeds a predetermined number (five in the illustrated example), the display unit 47 no longer displays the first trend line TL1 and the first predicted lifetime timing LT1, and instead displays the second trend line TL2 and the second predicted lifetime timing LT2, as shown in the lower graph of Figure 7.

[0065] When using the robot system 10 (or life prediction device 40) described above, if the gas spring 26 is refilled with gas multiple times, it is preferable that the angular positions of the arm 24 (more specifically, the arm 24 supported by the gas spring 26 that is the target of the gas refill) be aligned during these multiple refills. The display unit 47 may also display a prompt to the operator to fix the arm 24 at a predetermined angular position when gas refilling is performed. Alignment of the arm 24 can be achieved, for example, by manually changing the angular position of the arm 24 to match a predetermined angle using a function that displays the angle of the joint corresponding to the drive motor 25 in real time on the teaching pendant. Alignment of the arm 24 may also be performed by the operator measuring the angular position of the arm 24 with a spirit level or angle gauge. Alignment of the arm 24 may also be performed so that the marks (not shown) attached to the arm 24 and the swivel unit 22 align. Instead of manual positioning, the controller 30 may control the drive motor 25 to maintain the arm 24 at a predetermined angular position when gas replenishment is performed.

[0066] Since the pressure P of a gas follows the equation of state PV=nRT, the pressure P changes depending on the temperature T, even if the number of moles n and the volume V are the same. When the gas spring 26 is replenished multiple times during the use of the robot system 10 (or life prediction device 40), it is preferable that the gas replenishment is carried out according to the target internal pressure value, which is the target value of the internal pressure of the gas spring 26, determined based on the surface temperature of the gas spring 26 at the time of gas replenishment. This target internal pressure value is set so that the total amount of gas in the gas spring 26 (the number of moles n mentioned above) is the same or approximately the same after each of the multiple gas replenishments. Here, "value A and value B are approximately the same" means that the difference between value A and value B is 1% or less of the larger of value A and value B. Figure 8 is a graph showing the relationship between the surface temperature of the gas spring 26 and the specified pressure (i.e., the target internal pressure value), with the horizontal axis being the surface temperature and the vertical axis being the specified pressure. As shown in the figure, the higher the surface temperature of the gas spring 26, the higher the target internal pressure of the gas spring 26. The operator can, for example, use this graph to determine the target internal pressure and then replenish the gas so that the internal pressure of the gas spring 26 reaches that target value. The target internal pressure may also be determined by the controller 30 by substituting the surface temperature of the gas spring 26 into a calculation formula, or by using a table that shows the relationship between the surface temperature and the target internal pressure. Furthermore, when gas replenishment is performed, the display unit 47 may display the target internal pressure determined by the controller 30 and / or the current internal pressure of the gas spring 26.

[0067] The internal pressure target value is determined based on the surface temperature of the gas spring 26 before replenishment, but the temperature may change before and after gas replenishment. If the surface temperature of the gas spring 26 changes after replenishment, the internal pressure target value may be recalculated and the gas replenishment may be repeated based on the new internal pressure target value. If the new internal pressure target value is lower than the original target value, a small amount of gas can be released from the gas spring 26 and the gas replenishment can be repeated.

[0068] As described above, the life prediction device 40 of this embodiment predicts the life of a robot 20, which comprises an arm 24, a drive motor 25 that drives the arm 24, and a gas spring 26 that supports the load acting on the arm 24 and reduces the load on the drive motor 25, based on information about the drive motor 25 of the robot 20. The life prediction device 40 comprises an input unit 41, a replenishment timing storage unit 42, a current data acquisition unit 43, a feature data acquisition unit 44, a time series data correction unit 45, and a life prediction unit 46. The input unit 41 receives a signal indicating that gas has been replenished to the gas spring 26. The replenishment timing storage unit 42 stores the replenishment timings t1 and t2, which are the timings when gas has been replenished, based on the signal input to the input unit 41. The current data acquisition unit 43 acquires current time series data, which is time series data of information about the drive current of the drive motor 25. The feature data acquisition unit 44 acquires and stores feature time series data, which is time series data of the features of the current time series data. The time series data correction unit 45 corrects a series of feature time series data so that the progression of a series of feature time series data prior to the most recent replenishment timing t1 matches the post-replenishment initial feature value Va obtained from one or more initial features after the most recent replenishment timing t1. The lifetime prediction unit 46 creates a first trend line TL1 based on the corrected series of feature time series data and determines the timing at which the first trend line TL1 reaches a predetermined lifetime threshold th as the first predicted lifetime timing LT1.

[0069] As a result, even if gas leaks from the gas spring 26 during use, causing fluctuations in the drive current of the drive motor 25, the effects of these fluctuations can be substantially eliminated by a predetermined correction process, allowing for accurate life prediction. In other words, while gas leakage in the gas spring 26 affects the current time series data and, consequently, the feature time series data, the correction process of this embodiment yields feature time series data that has been corrected to substantially eliminate such effects. Since life prediction is performed based on the corrected feature time series data, the prediction accuracy can be improved.

[0070] In the lifetime prediction device 40 of this embodiment, when considering a graph where the horizontal axis is time and the vertical axis is features, the slope of the first line L1 passing through the point of the pre-replenishment initial feature value Ve, which is obtained from one or more initial features of a series of feature time-series data, and the point of the pre-replenishment final feature value Vb, which is obtained from one or more final features before the most recent replenishment timing t1, is defined as the pre-replenishment slope Sa. The slope of the second line L2 passing through the point of the pre-replenishment initial feature value Ve and the point of the post-replenishment initial feature value Va is defined as the post-replenishment slope Sb, and the ratio of the post-replenishment slope Sb to the pre-replenishment slope Sa is defined as the slope ratio k1. The time series data correction unit 45 corrects the series of feature time series data prior to the most recent supplementation timing t1 by multiplying the difference between each of the features included in the series of feature time series data prior to the most recent supplementation timing t1 and the initial feature value Ve before supplementation by the slope ratio k1, and adding the initial feature value Ve before supplementation to the multiplied value.

[0071] This allows for the appropriate correction of the series of feature time-series data prior to the most recent replenishment timing t1 through simple processes such as determining the pre-replenishment slope Sa, post-replenishment slope Sb, and slope ratio k1.

[0072] In the lifetime prediction device 40 of this embodiment, the time series data correction unit 45 corrects the series of feature time series data after the most recent replenishment timing t1 by multiplying the difference between each of the features included in the series of feature time series data after the most recent replenishment timing t1 and the initial feature value Va after replenishment by a slope ratio k1, and adding the initial feature value Va after replenishment to the multiplied value. The lifetime prediction unit 46 creates a second trend line TL2 based on the corrected series of feature time series data after the most recent replenishment timing t1, and determines the timing at which the second trend line TL2 reaches a predetermined lifetime threshold th as the second predicted lifetime timing LT2.

[0073] This allows for even simpler and more appropriate correction of the series of feature time series data after the most recent replenishment timing t1 by using the already determined slope ratio k1. However, it is also possible to determine a new slope ratio, different from the aforementioned slope ratio k1, for the series of feature time series data after the most recent replenishment timing t1, and use this new slope ratio to correct the series of feature time series data.

[0074] The life prediction device 40 of this embodiment further includes a display unit 47 that displays both the first trend line TL1 and the first predicted life timing LT1, and the second trend line TL2 and the second predicted life timing LT2.

[0075] This allows the operator to consider the maintenance timing of the robot 20 based on both the first predicted life timing LT1 and the second predicted life timing LT2. However, the display unit 47 may display only one of the following: the first trend line TL1 and the first predicted life timing LT1, or the second trend line TL2 and the second predicted life timing LT2.

[0076] In the lifetime prediction device 40 of this embodiment, if the number of features included in the series of feature time-series data after the most recent replenishment timing t1 is greater than or equal to a predetermined number, the display unit 47 does not display the first trend line TL1 and the first predicted lifetime timing LT1, but displays the second trend line TL2 and the second predicted lifetime timing LT2.

[0077] This ensures that the reliability of the second trendline TL2 and the second predicted lifetime timing LT2, which are based on relatively newer information, is sufficiently high. By displaying only the second trendline TL2 and the second predicted lifetime timing LT2, the reliability of the displayed information is ensured, while reducing the burden on the operator considering the maintenance timing of the robot 20. However, the display unit 47 may also display the first trendline TL1 and the first predicted lifetime timing LT1 if the number of features included in the series of feature time-series data after the most recent replenishment timing t1 is greater than or equal to a predetermined number.

[0078] In the life prediction device 40 of this embodiment, the feature quantities of the current time series data are the root mean square, peak value, peak-to-peak value, or frequency analysis integrated value obtained from the current time series data, or the dissimilarity between the current time series data and the current time series data acquired by the drive motor 25 immediately after the start of use.

[0079] This allows for accurate lifetime prediction using appropriate features.

[0080] In the life prediction device 40 of this embodiment, the information regarding the drive current is the measured value of the drive current, the current command value for the drive motor 25, or the position deviation which is the difference between the measured position and the position command value in the drive motor 25.

[0081] This allows for accurate lifespan prediction using information on appropriate drive currents.

[0082] In the method of using the life prediction device of this embodiment, when the gas spring 26 is refilled with gas multiple times, the multiple gas refills are performed with the arms 24 in the same position.

[0083] This ensures that the conditions for each gas replenishment are consistent, thereby preventing variations in those conditions from affecting the feature time series data. Consequently, the reliability of the correction performed by the time series data correction unit 45, and consequently the reliability of the lifespan prediction performed by the lifespan prediction unit 46, can be increased.

[0084] In the method of using the life prediction device of this embodiment, when the gas spring 26 is replenished with gas multiple times, each of the multiple gas replenishments is performed after determining the target internal pressure value, which is the target value of the internal pressure of the gas spring 26, based on the surface temperature of the gas spring 26 at the time of replenishment. The target internal pressure value is set so that the total amount of gas (total molar amount of gas) in the gas spring 26 is the same or approximately the same after each of the multiple gas replenishments has been performed.

[0085] This ensures that the total amount of gas present in the gas spring 26 is substantially equal after each gas replenishment, thus avoiding the impact of variations in the total gas amount on the feature time series data. Consequently, the reliability of the correction by the time series data correction unit 45, and consequently the reliability of the lifetime prediction by the lifetime prediction unit 46, can be improved.

[0086] The life prediction method of this embodiment is a life prediction method that predicts the life of a robot 20, which comprises an arm 24, a drive motor 25 that drives the arm 24, and a gas spring 26 that supports the load acting on the arm 24 and reduces the load on the drive motor 25, based on information about the drive motor 25. A signal is input to the gas spring 26 indicating that gas has been replenished. Based on the input of the signal, the replenishment timing, which is the timing when gas replenishment was performed, is stored. Current time series data, which is time series data of information about the drive current of the drive motor 25, is acquired. Feature time series data, which is time series data of the features of the current time series data, is acquired and stored. The series of feature time series data is corrected so that the progression of the series of feature time series data before the most recent replenishment timing t1 matches the initial post-replenishment feature value Va obtained from one or more initial features after the most recent replenishment timing t1. A first trend line TL1 is created based on the corrected series of feature time series data, and the timing at which the first trend line TL1 reaches a predetermined life threshold th is determined as the first predicted life timing LT1.

[0087] As a result, even if gas leaks from the gas spring 26 during use, causing fluctuations in the drive current of the drive motor 25, the effects of these fluctuations can be substantially eliminated by a predetermined correction process, allowing for accurate life prediction.

[0088] The life prediction program of this embodiment is a life prediction program that predicts the life of a robot 20, which comprises an arm 24, a drive motor 25 that drives the arm 24, and a gas spring 26 that supports the load acting on the arm 24 and reduces the load on the drive motor 25, based on information about the drive motor 25. The life prediction program causes a computer to execute a signal input step, a replenishment timing storage step, a current data acquisition step, a feature data acquisition step, a time series data correction step, and a life prediction step. In the signal input step, a signal is input to the gas spring 26 indicating that gas has been replenished. In the replenishment timing storage step, the replenishment timing, which is the timing at which gas replenishment was performed, is stored based on the signal input. In the current data acquisition step, current time series data, which is time series data of information about the drive current of the drive motor 25, is acquired. In the feature data acquisition step, feature time series data, which is time series data of the features of the current time series data, is acquired and stored. In the time series data correction step, the feature time series data is corrected so that the progression of the series of feature time series data prior to the most recent replenishment timing t1 matches the post-replenishment initial feature value Va, which is obtained from one or more initial features after the most recent replenishment timing t1. In the lifetime prediction step, a first trend line TL1 is created based on the corrected series of feature time series data, and the timing at which the first trend line TL1 reaches a predetermined lifetime threshold th is determined as the first predicted lifetime timing LT1.

[0089] As a result, even if gas leaks from the gas spring 26 during use, causing fluctuations in the drive current of the drive motor 25, the effects of these fluctuations can be substantially eliminated by a predetermined correction process, allowing for accurate life prediction.

[0090] Next, a second embodiment will be described. Figure 9 is a graph illustrating the correction of a series of feature time series data prior to the most recent replenishment timing t1 in this embodiment. Figure 10 is a graph illustrating the correction of a series of feature time series data after the most recent replenishment timing t1 in this embodiment. In the description of this embodiment, components that are the same as or similar to those in the previously described embodiment will be denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0091] As shown in Figure 9, the time series data correction unit 45 of this embodiment corrects the series of feature time series data before the most recent supplementation timing t1 by rotating all features included in the series of feature time series data before the most recent supplementation timing t1 with the point of the initial feature value Ve before supplementation (in this example, the first feature in the series of feature time series data before the most recent supplementation timing t1) as the rotation center, and applying a rotation transformation corresponding to the difference angle θ. Known methods can be used for this rotation transformation. In Figure 9, the series of feature time series data before correction are shown as white circles, and the series of feature time series data after correction are shown as black circles.

[0092] Here, in the graph of Figure 9, the difference angle θ is the angle between the pre-refill line Lb and the post-refill line La. The pre-refill line Lb is a straight line passing through the point of the pre-refill initial feature value Ve and the point of the pre-refill final feature value Vb (in this example, the value of the last feature in the series of feature time series data before the most recent refill timing t1). The post-refill line La is a straight line passing through the point of the pre-refill initial feature value Ve and the point of the post-refill initial feature value Va (in this example, the value of the first feature in the series of feature time series data after the most recent refill timing t1).

[0093] Furthermore, as shown in Figure 10, the time-series data correction unit 45 of this embodiment corrects the series of feature time-series data after the most recent replenishment timing t1 by performing a rotation transformation corresponding to the difference angle θ on all features included in the series of feature time-series data after the most recent replenishment timing t1, using the point of the initial feature value Va after replenishment as the rotation center.

[0094] As shown in Figure 11, the lifetime prediction unit 46 creates a first trend line TL1 (upper graph in Figure 11) based on a series of corrected feature time series data prior to the most recent corrected replenishment timing t1, and determines the timing at which the first trend line TL1 reaches a predetermined lifetime threshold th as the first predicted lifetime timing LT1. The lifetime prediction unit 46 also creates a second trend line TL2 (lower graph in Figure 11) based on a series of feature time series data after the most recent corrected replenishment timing t1, and determines the timing at which the second trend line TL2 reaches a predetermined lifetime threshold th as the second predicted lifetime timing LT2.

[0095] As shown in the upper graph of Figure 11, the display unit 47 of this embodiment displays the first trend line TL1 and the first predicted lifetime timing LT1, but does not display the second trend line TL2 and the second predicted lifetime timing LT2, if the number of features included in the series of feature time series data after the most recent replenishment timing t1 is less than a predetermined number (5 in the illustrated example). On the other hand, as shown in the lower graph of Figure 11, the display unit 47 does not display the first trend line TL1 and the first predicted lifetime timing LT1, but displays the second trend line TL2 and the second predicted lifetime timing LT2, if the number of features included in the series of feature time series data after the most recent replenishment timing t1 is greater than or equal to a predetermined number.

[0096] As described above, in the lifetime prediction device 40 of this embodiment, when considering a graph where the horizontal axis is time and the vertical axis is features, the pre-refill line Lb is defined as the line passing through the point of the pre-refill initial feature value Ve, which is obtained from one or more initial features of the series of feature time series data, and the point of the pre-refill final feature value Vb, which is obtained from one or more final features before the most recent refill timing t1, and the post-refill line La is defined as the line passing through the point of the pre-refill initial feature value Ve and the point of the post-refill initial feature value Va, and the difference angle θ is defined as the angle between the pre-refill line Lb and the post-refill line La. The time series data correction unit 45 corrects the series of feature time series data before the most recent refill timing t1 by performing a rotation transformation corresponding to the difference angle θ on all features included in the series of feature time series data before the most recent refill timing t1, with the point of the pre-refill initial feature value Ve as the rotation center.

[0097] This allows for the appropriate correction of a series of feature time-series data prior to the most recent interpolation timing t1 through simple processes such as calculating the difference angle θ, which is the angle between the pre-interpolation line Lb and the post-interpolation line La.

[0098] In the lifetime prediction device of this embodiment, the time series data correction unit 45 corrects the series of feature time series data after the most recent replenishment timing t1 by performing a rotation transformation corresponding to the difference angle θ on all features included in the series of feature time series data after the most recent replenishment timing t1, using the point of the initial feature value Va after replenishment as the rotation center. The lifetime prediction unit 46 creates a second trend line TL2 based on the corrected series of feature time series data after the most recent replenishment timing t1, and determines the timing at which the second trend line TL2 reaches a predetermined lifetime threshold th as the second predicted lifetime timing LT2.

[0099] This allows for even simpler and more appropriate correction of the feature time series data after the most recent replenishment timing t1 by using the already calculated difference angle θ. However, it is also possible to calculate a new difference angle, different from the difference angle θ, for the feature time series data after the most recent replenishment timing t1, and use this new difference angle to correct the feature time series data.

[0100] In the lifetime prediction device 40 of this embodiment, the display unit 47 displays the first trend line TL1 and the first predicted lifetime timing LT1, but does not display the second trend line TL2 and the second predicted lifetime timing LT2, if the number of features included in the series of feature time series data after the most recent replenishment timing t1 is less than a predetermined number. On the other hand, the display unit 47 does not display the first trend line TL1 and the first predicted lifetime timing LT1, but displays the second trend line TL2 and the second predicted lifetime timing LT2, if the number of features included in the series of feature time series data after the most recent replenishment timing t1 is greater than or equal to a predetermined number.

[0101] As a result, when the reliability of the second trend line TL2 and the second predicted life timing LT2 may be low, only the first trend line TL1 and the first predicted life timing LT1 are displayed, while when their reliability is sufficiently high, only the second trend line TL2 and the second predicted life timing LT2 are displayed. This ensures the reliability of the displayed information while reducing the burden on workers considering maintenance timing for the robot 20.

[0102] Next, a third embodiment will be described. Figure 12 is a block diagram schematically showing the electrical configuration of the life prediction device 50 of this embodiment. In the description of this embodiment, the same or similar components as those in the previously described embodiments will be denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0103] As shown in Figure 12, the life prediction device 50 of this embodiment includes a current data acquisition unit 51, a current data correction unit 52, a corrected current data storage unit 53, a corrected feature data acquisition unit 54, a life prediction unit 55, and a display unit 56.

[0104] The current data acquisition unit 51 acquires current data, which is data relating to the drive current of the drive motor 25. The information relating to the drive current may be the same as that in the first embodiment described above.

[0105] The current data correction unit 52 corrects the current data based on internal pressure information, which is information regarding the internal pressure in the gas spring 26, and position information, which is information regarding the position of the arm 24.

[0106] The correction of the current data described above will be explained in detail. In the following explanation, the internal pressure information is the internal pressure of the gas spring 26 detected by the pressure detector 27, and the position information is the angular position of the arm 24 determined based on the angular position of the drive motor 25, etc.

[0107] The moment M around the joint axis of the arm 24 that is canceled by the gas spring 26 can be expressed by the formula M = L × F × sinθ1. Hereafter, this moment may be referred to as the cancellation moment. Here, as shown in Figure 13, L is the distance between the center O1 of the joint axis and the center O2 of the tip of the gas spring 26. F is the magnitude of the support force in the direction along the center line of the gas spring 26, and is expressed as the product of the internal pressure of the gas spring 26 and the pressure-receiving area of ​​the piston rod 26b.

[0108] θ1 is the angle formed by a straight line passing through the center O1 of the joint axis and the center O2 of the tip of the gas spring 26 with respect to the center line of the gas spring 26. θ1 can be determined by geometric calculation from the angular position (i.e., position information) of the arm 24, based on information such as the position of the center O1 of the joint axis and the mounting positions of both ends of the gas spring 26.

[0109] Let V0 be the maximum cylinder volume of the gas spring 26 when it is attached to the swivel section 22 and the arm 24 (i.e., the maximum volume of the space between the bottom surface of the cylinder 26a and the pressure-receiving surface of the piston rod 26b in the attached state of the gas spring 26). Let P0 be the specified internal pressure of the gas spring 26 corresponding to the maximum cylinder volume V0 (i.e., the internal pressure of the gas spring 26 when the gas spring 26 is at its maximum cylinder volume V0 immediately after gas replenishment). Let Vn be the cylinder volume when the arm 24 is at an arbitrary angular position. Based on these, the initial internal pressure Pn0 of the gas spring 26 when the arm 24 is at an arbitrary angular position can be calculated as Pn0 = P0 × V0 / Vn. P0 and V0 are known. Vn can also be calculated geometrically from the angular position of the arm 24, similar to θ1 described above. Therefore, the initial internal pressure Pn0 can also be calculated corresponding to any angular position of the arm 24.

[0110] The current data correction unit 52 calculates the support force F by multiplying the initial internal pressure Pn0, obtained from the angular position of the arm 24, by the pressure-receiving area of ​​the piston rod 26b. Subsequently, the current data correction unit 52 calculates the cancellation moment M by substituting the angle θ1 obtained from the angular position of the arm 24 and the support force F into the above-mentioned cancellation moment formula. Since this cancellation moment corresponds to the initial internal pressure Pn0, it may hereafter be called the initial cancellation moment. The current data correction unit 52 multiplies this initial cancellation moment M by a predetermined coefficient K (i.e., a coefficient K that takes into account the reduction ratio of the speed reducer and the torque constant of the drive motor 25, etc.). This makes it possible to obtain the drive current Im that corresponds to the initial cancellation moment M when the arm 24 is at an arbitrary angular position (Im = K × M). Hereafter, this drive current may hereafter be called the initial cancellation equivalent current.

[0111] As described above, the gas in the gas spring 26 leaks over time. At the present time, after a suitable period has elapsed since gas replenishment, the internal pressure of the gas spring 26 (i.e., internal pressure information) when the arm 24 is at an arbitrary angular position is denoted as Pnm. This Pnm is detected by a pressure detector 27 provided on the gas spring 26. The pressure detector 27 is connected to the life prediction device 50 via the controller 30. The correction current data storage unit 53 calculates the pressure reduction rate μn when the arm 24 is at an arbitrary angular position according to the formula μn = Pnm / Pn0. This pressure reduction rate μn is the ratio of the current internal pressure Pnm to the initial internal pressure Pn0, and represents how much the internal pressure of the gas spring 26 has decreased. Note that the smaller the value of the pressure reduction rate μn, the greater the internal pressure of the gas spring 26 has decreased. The pressure reduction rate can also be called the internal pressure retention rate. The pressure detector 27 may be connected directly to the life prediction device 50 without going through the controller 30.

[0112] Alternatively, instead of the pressure detector 27, a force sensor (not shown) for measuring the current support force F of the gas spring 26 may be provided on the cylinder 26a or piston rod 26b of the gas spring 26. Or, such a force sensor may be provided at a part where the gas spring 26 is rotatably supported. Alternatively, the current support force F of the gas spring 26 may be indirectly determined using a strain gauge (not shown) or the like to measure the amount of deformation of the end of the gas spring 26. These support forces F are measured or determined as the support force of the gas spring 26 when the arm 24 is at an arbitrary angular position. Here, the support force Fm of the gas spring 26 immediately after gas replenishment is determined in advance by multiplying the initial internal pressure Pn0 when the arm 24 is at the same angular position by the pressure-receiving area of ​​the piston rod 26b. In this case, the pressure reduction rate μn can be determined according to the formula μn = F / Fm. It is desirable that the measurements described in this paragraph be performed with the arm 24 stopped.

[0113] The moment resulting from the decrease in the support force F of the gas spring 26 due to gas leakage must be compensated for by the torque generated by the drive motor 25. The current data correction unit 52 calculates the increase in drive current ΔIm at the current time when the arm 24 is at any angular position by dividing the initial cancellation equivalent current Im by the pressure reduction rate μn (ΔIm = Im / μn). This equation means that the smaller the pressure reduction rate μn, that is, the greater the decrease in the internal pressure of the gas spring 26, the larger the increase in drive current ΔIm becomes. Furthermore, this increase ΔIm is the increase in drive current due to gas leakage from the gas spring 26. By subtracting the above increase ΔIm, which changes according to the angular position of the arm 24, from the measured value of the drive current shown in Figure 3, the current data can be corrected so that the effect of gas leakage is substantially eliminated.

[0114] The correction of the current data described above may be performed, for example, by calculation based on the above-mentioned multiple formulas, or by simplifying at least part of the calculation using a table that defines the relationship between the angular position of arm 24 and the initial internal pressure Pn0 and / or the initial cancellation equivalent current Im.

[0115] The corrected current data storage unit 53 stores corrected current time series data, which is time series data of the corrected current data.

[0116] The correction feature data acquisition unit 54 acquires correction feature time series data, which is time series data of the feature quantities of the correction current time series data. The feature quantities of the correction current time series data can be arbitrarily selected, as in the first embodiment described above. The feature quantity in this embodiment is I2, as in the first embodiment described above.

[0117] The lifetime prediction unit 55 creates a trend line TL based on the corrected feature time series data and determines the timing at which the trend line TL reaches a predetermined lifetime threshold th as the predicted lifetime timing LT. Here, the trend line TL is created based on the current data corrected as described above, that is, current data from which the effect of gas leakage from the gas spring 26 has been substantially eliminated. Therefore, by performing lifetime prediction using such a trend line TL, the prediction accuracy can be improved.

[0118] As shown in Figure 14, the display unit 56 displays the trend line TL and the predicted lifetime timing LT. In Figure 14, the corrected feature time series data is shown as black circles, while the feature time series data without current data correction (i.e., the feature time series data in which the effect of gas leakage from the gas spring 26 has not been eliminated) is shown as white circles. In this embodiment, the latter white circle time series data is not actually displayed on the display unit 56, but it may be displayed.

[0119] As described above, the life prediction device 50 of this embodiment predicts the life of a robot 20, which comprises an arm 24, a drive motor 25 that drives the arm 24, and a gas spring 26 that supports the load acting on the arm 24 and reduces the load on the drive motor 25, based on information about the drive motor 25. The life prediction device 50 comprises a current data acquisition unit 51, a current data correction unit 52, a corrected current data storage unit 53, a corrected feature data acquisition unit 54, and a life prediction unit 55. The current data acquisition unit 51 acquires current data, which is data relating to the drive current of the drive motor 25. The current data correction unit 52 corrects the current data based on internal pressure information, which is information relating to the internal pressure in the gas spring 26, and position information, which is information relating to the position of the arm 24. The corrected current data storage unit 53 stores corrected current time series data, which is time series data of the corrected current data. The corrected feature data acquisition unit 54 acquires corrected feature time series data, which is time series data of the features of the corrected current time series data. The lifetime prediction unit 55 creates a trend line TL based on the corrected feature time series data and determines the timing at which the trend line TL reaches a predetermined lifetime threshold th as the predicted lifetime timing.

[0120] As a result, even if gas leaks from the gas spring 26 during use, causing fluctuations in the drive current of the drive motor 25, the effects of these fluctuations can be substantially eliminated by a predetermined correction process, allowing for accurate life prediction. In other words, while gas leakage in the gas spring 26 affects the current data, the correction process of this embodiment provides corrected current time series data that substantially eliminates such effects. Since life prediction is performed based on corrected feature time series data obtained from the corrected current time series data, the prediction accuracy can be improved.

[0121] In the life prediction device 50 of this embodiment, the internal pressure information is the internal pressure of the gas spring 26 detected by the pressure detector 27.

[0122] This allows for effective use of the pressure sensor 27's detection values ​​to appropriately correct the current data.

[0123] The life prediction method of this embodiment is a life prediction method that predicts the life of a robot 20, which comprises an arm 24, a drive motor 25 that drives the arm 24, and a gas spring 26 that supports the load acting on the arm 24 and reduces the load on the drive motor 25, based on information about the drive motor 25. Current data, which is data about the drive current of the drive motor 25, is acquired. The current data is corrected based on internal pressure information, which is information about the internal pressure in the gas spring 26, and position information, which is information about the position of the arm 24. Corrected current time series data, which is time series data of the corrected current data, is stored. Corrected feature time series data, which is time series data of the feature quantities of the corrected current time series data, is acquired. A trend line TL is created based on the corrected feature time series data, and the timing at which the trend line TL reaches a predetermined life threshold th is determined as the predicted life timing LT.

[0124] As a result, even if gas leaks from the gas spring 26 during use, causing fluctuations in the drive current of the drive motor 25, the effects of these fluctuations can be substantially eliminated by a predetermined correction process, allowing for accurate life prediction.

[0125] The life prediction program of this embodiment is a life prediction program that predicts the life of a robot 20, which comprises an arm 24, a drive motor 25 that drives the arm 24, and a gas spring 26 that supports the load acting on the arm 24 and reduces the load on the drive motor 25, based on information about the drive motor 25. The life prediction program causes a computer to execute a current data acquisition step, a current data correction step, a corrected current data storage step, a corrected feature data acquisition step, and a life prediction step. In the current data acquisition step, current data, which is data relating to the drive current of the drive motor 25, is acquired. In the current data correction step, the current data is corrected based on internal pressure information, which is information relating to the internal pressure in the gas spring 26, and position information, which is information relating to the position of the arm 24. In the corrected current data storage step, corrected current time series data, which is time series data of the corrected current data, is stored. In the corrected feature data acquisition step, corrected feature time series data, which is time series data of the features of the corrected current time series data, is acquired. In the lifetime prediction step, a trend line TL is created based on the corrected feature time series data, and the timing at which the trend line TL reaches a predetermined lifetime threshold th is determined as the predicted lifetime timing LT.

[0126] As a result, even if gas leaks from the gas spring 26 during use, causing fluctuations in the drive current of the drive motor 25, the effects of these fluctuations can be substantially eliminated by a predetermined correction process, allowing for accurate life prediction.

[0127] Next, a modified example of the third embodiment described above will be explained. In this explanation of the modified example, the same or similar components as those in the previously described embodiment will be denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0128] In this modified example, instead of using the detected value from the pressure detector 27 installed on the gas spring 26 as the internal pressure information used to correct the current data, the estimated internal pressure of the gas spring 26, estimated from the position information and current data, is used. As an estimation method, for example, the following method (i) to (iv) can be considered. That is, (i) with the arm 24 stopped at a predetermined stopping position, the measured value Ia of the drive current of the drive motor 25 is obtained. (ii) The drive current Ic of the drive motor 25 at the stopping position is calculated assuming there is no gas leakage from the gas spring 26. Hereinafter, this drive current may be called the initial drive current. Details of this calculation will be described later. (iii) The difference between the measured drive current Ia and the calculated initial drive current Ic is calculated. (iv) Based on this difference, the estimated internal pressure Pnme of the gas spring 26 at the current time is calculated. Details of this calculation will be described later. Furthermore, by replacing the internal pressure Pnm in the third embodiment with the estimated internal pressure Pnme, the increase in drive current ΔIm when the arm 24 is at any angular position can be determined.

[0129] The calculation of the initial drive current Ic in (ii) is described below. The moment M1 based on the weight of the arm 24 when the arm 24 is at an arbitrary angular position can be obtained by calculation based on the weight of the arm 24 and the position of its center of gravity. Also, as described in the third embodiment, the initial cancellation moment M when the arm 24 is at an arbitrary angular position can be obtained by calculation. Therefore, assuming that there is no gas leakage from the gas spring 26, the torque Mm (=M1-M) that the drive motor 25 should bear is uniquely determined according to the angular position of the arm 24. By multiplying this torque Mm by the coefficient K described above, the initial drive current Ic can be obtained.

[0130] The estimation of the internal pressure described in (iv) can be performed, for example, as follows: The difference between the measured drive current Ia and the calculated initial drive current Ic is divided by the initial cancellation equivalent current Im to obtain the current ratio R (R = (Ia - Ic) / Im). Subtracting the obtained ratio R from 1 and multiplying the resulting value by the initial internal pressure Pn0 gives the estimated internal pressure Pnme (Pnme = (1 - R) × Pn0).

[0131] In this modified example, the current data can be appropriately corrected without using the detected value from the pressure detector 27. Alternatively, the internal pressure information may be determined based on a combination of the estimated internal pressure and the detected value from the pressure detector. For example, the average of the two could be used as the internal pressure information. Alternatively, the value representing the larger reduction in pressure could be used as the internal pressure information.

[0132] This modified example describes a method for determining the gas pressure reduction rate μn from the movement of the robot 20 and the drive current Ic, but the estimation method is not limited to the above. For example, it is conceivable to obtain the operation information of the gas spring 26 (e.g., the stroke amount of the piston rod 26b relative to the cylinder 26a) from the movement of the robot 20 (more specifically, the position information of the arm 24), and without using the drive current Ic information, calculate the amount of gas leakage from an empirical formula based on the operation information of the gas spring 26 (e.g., an empirical formula that defines the relationship between the integrated value of the stroke amount of the piston rod 26b and the amount of gas leakage), and estimate the pressure reduction rate μn. Based on the estimated pressure reduction rate μn, the internal pressure of the gas spring 26 can be estimated. This method is based on the existence of a positive correlation between the amount of gas leakage (or reduction in internal pressure) in the gas spring 26 and the stroke amount of the piston rod 26b.

[0133] Next, a fourth embodiment will be described. Figure 15 is a schematic block diagram showing the electrical configuration of the gas spring life prediction device 60 of this embodiment. Figure 16 is a graph showing the display mode of the trend line TLg in the fourth embodiment. In the description of this embodiment, the same or similar components as in the previously described embodiments are denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0134] The gas spring life prediction device 60 of this embodiment is specifically designed to predict the life of the gas spring 26. The predicted life of the gas spring 26 may be either the life due to deterioration of the mechanical drive unit or the life due to gas leakage. As mentioned above, the life due to gas leakage can be extended by replenishing the gas.

[0135] As shown in Figure 15, the gas spring life prediction device 60 includes an internal pressure data acquisition unit (internal pressure information acquisition unit) 61, an internal pressure data correction unit (internal pressure information correction unit) 62, a corrected internal pressure data storage unit (corrected internal pressure information storage unit) 63, a corrected internal pressure feature data acquisition unit 64, a gas spring life prediction unit 65, and a display unit 66.

[0136] The internal pressure data acquisition unit 61 acquires internal pressure data, which is data relating to the internal pressure of the gas spring 26 when the robot 20 operates according to the evaluation operation pattern. In this embodiment, the internal pressure data is the internal pressure of the gas spring 26 actually detected by the pressure detector 27. In the process of the robot 20 performing an operation corresponding to one evaluation operation pattern, the internal pressure data acquisition unit 61 acquires the internal pressure data in the form of time-series data by repeatedly acquiring detected values ​​from the pressure detector 27 at short time intervals. In the first embodiment described above, current time-series data as shown in Figure 3 is obtained each time the robot 20 performs an operation corresponding to one evaluation operation pattern, but in this embodiment, internal pressure time-series data is obtained instead of current time-series data.

[0137] The internal pressure data correction unit 62 corrects the internal pressure time series data based on position information, which is information regarding the position of the arm 24.

[0138] The volume of the gas spring 26 changes according to the position of the arm 24, and the internal pressure fluctuates accordingly. The internal pressure data correction unit 62 corrects each internal pressure value that makes up the internal pressure time series data, based on the position information of the arm 24 at the time of internal pressure detection, so that it corresponds to the internal pressure when the arm 24 is in a predetermined basic position. This eliminates the influence of volume changes on the internal pressure. The correction can be performed using the equation of state PV=nRT. The basic position of the arm 24 can be arbitrarily determined, but for example, it can be the position shown in Figure 13.

[0139] The corrected internal pressure data storage unit 63 stores corrected internal pressure time-series data, which is time-series data of internal pressure data corrected by the internal pressure data correction unit 62.

[0140] The Corrected Internal Pressure Feature Data Acquisition Unit 64 acquires Corrected Internal Pressure Feature Time Series Data, which is time series data of the features of the Corrected Internal Pressure Time Series Data. In this embodiment, one feature is obtained from one Corrected Internal Pressure Time Series Data. Time series data of features can be obtained by calculating the feature for each of the Corrected Internal Pressure Time Series Data that are acquired periodically or irregularly.

[0141] Based on the same principles as in the first embodiment, the root mean square of the internal pressure obtained from the internal pressure time series data can be used as the feature quantity. In this case, the corrected internal pressure feature quantity data acquisition unit 64 acquires time series data of the root mean square value. However, the feature quantity is not limited to the root mean square. The corrected internal pressure time series data is affected by acceleration and deceleration in the stroke direction of the gas spring 26, or external vibrations applied to the gas spring 26. For this reason, it is preferable to use the mean or median value instead of the root mean square value as the feature quantity of the corrected internal pressure, as this can better reduce such effects.

[0142] The gas spring lifetime prediction unit 65 creates a trend line TLg based on the corrected internal pressure feature time series data, and determines the timing at which the trend line TLg reaches a predetermined lifetime threshold thg as the predicted lifetime timing LTg of the gas spring 26. The trend line TLg can be determined by a known method, such as the least squares method.

[0143] As shown in Figure 16, the display unit 66 displays the trend line TLg and the predicted life timing LTg. This allows the operator to appropriately consider the maintenance timing of the gas spring 26.

[0144] As described above, the gas spring life prediction device 60 of this embodiment predicts the life of the gas spring 26 provided in the robot 20. The robot 20 includes an arm 24, a drive motor 25, a gas spring 26, and a pressure detector 27. The drive motor 25 drives the arm 24. The gas spring 26 supports the load acting on the arm 24 and reduces the load on the drive motor 25. The pressure detector 27 detects the internal pressure of the gas spring 26. The gas spring life prediction device 60 includes an internal pressure data acquisition unit 61, an internal pressure data correction unit 62, a corrected internal pressure data storage unit 63, a corrected internal pressure feature data acquisition unit 64, and a gas spring life prediction unit 65. The internal pressure data acquisition unit 61 acquires the internal pressure detected by the pressure detector 27. The internal pressure data correction unit 62 corrects the internal pressure acquired by the internal pressure data acquisition unit 61 based on position information, which is information regarding the position of the arm 24. The corrected internal pressure data storage unit 63 stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure data correction unit 62. The corrected internal pressure feature data acquisition unit 64 acquires corrected internal pressure feature time-series data, which is the time-series data of the features of the corrected internal pressure time-series data stored in the corrected internal pressure data storage unit 63. The gas spring lifetime prediction unit 65 creates a trend line TLg based on the corrected internal pressure feature time-series data and determines the timing at which the trend line TLg reaches a predetermined lifetime threshold thg as the predicted lifetime timing LTg.

[0145] This allows the effect of changes in the position of the arm 24 on the internal pressure to be eliminated by correction, thus enabling a proper maintenance plan for the gas spring 26.

[0146] In the gas spring life prediction device 60 of this embodiment, the feature quantities of the time-series data of the corrected internal pressure are the median, mean, or root mean square obtained from the time-series data of the corrected internal pressure.

[0147] This makes it possible to appropriately suppress the effects of the correction internal pressure characteristics on acceleration and deceleration in the stroke direction of the gas spring 26, or on external vibrations applied to the gas spring 26.

[0148] Next, a fifth embodiment will be described. Figure 17 is a schematic block diagram showing the electrical configuration of the gas spring abnormality detection device 60x of this embodiment. In the description of this embodiment, the same or similar components as those in the previously described embodiments are denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0149] Unlike the fourth embodiment described above, which determines the predicted lifetime timing LTg of the gas spring 26, this embodiment determines whether or not an abnormality has occurred in the gas spring 26 based on whether or not the feature quantity is above a threshold.

[0150] The gas spring abnormality detection device 60x comprises an internal pressure data acquisition unit 61, an internal pressure data correction unit 62, a corrected internal pressure data storage unit 63, a corrected internal pressure feature data acquisition unit (corrected internal pressure feature acquisition unit) 64, a gas spring abnormality detection unit 67, and a display unit 66.

[0151] In this embodiment, the gas spring abnormality determination device 60x includes a gas spring abnormality determination unit 67 instead of the gas spring life prediction unit 65 of the fourth embodiment. The gas spring abnormality determination unit 67 compares the feature quantities acquired by the corrected internal pressure feature quantity data acquisition unit 64 with a threshold value and determines whether or not an abnormality has occurred in the gas spring 26 based on this comparison result. For example, if the value of the root mean square of the feature quantity is greater than or equal to the threshold value, it can be determined to be abnormal, and if it is less than the threshold value, it can be determined to be normal. The feature quantities are arbitrary, as in the fourth embodiment, and for example, the median or mean value can be used. The threshold value may be equal to or different from the life threshold thg of the fourth embodiment described in Figure 16.

[0152] In the gas spring abnormality detection device 60x of this embodiment, a trend line TLg is not used for abnormality detection. Therefore, unlike the fourth embodiment in which the corrected internal pressure feature data acquisition unit 64 acquires a time series of feature quantities, it is sufficient for the corrected internal pressure feature data acquisition unit 64 to acquire only the most recent feature quantity.

[0153] As described above, the gas spring abnormality detection device 60x of this embodiment determines whether or not there is an abnormality in the gas spring 26 provided by the robot 20. The gas spring abnormality detection device 60x comprises an internal pressure data acquisition unit 61, an internal pressure data correction unit 62, a corrected internal pressure data storage unit 63, a corrected internal pressure feature data acquisition unit 64, and a gas spring abnormality detection unit 67. The internal pressure data acquisition unit 61 acquires the internal pressure detected by the pressure detector 27. The internal pressure data correction unit 62 corrects the internal pressure acquired by the internal pressure data acquisition unit 61 based on position information, which is information regarding the position of the arm 24. The corrected internal pressure data storage unit 63 stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure data correction unit 62. The corrected internal pressure feature data acquisition unit 64 acquires the feature quantities of the time-series data of the corrected internal pressure stored by the corrected internal pressure data storage unit 63. The gas spring abnormality determination unit 67 compares the corrected internal pressure feature quantities acquired by the corrected internal pressure feature quantity data acquisition unit 64 with a preset threshold and determines whether or not there is an abnormality in the gas spring 26 based on the comparison result.

[0154] This allows the influence of changes in the position of the arm 24 on the internal pressure to be eliminated by correction, thus enabling accurate determination of whether or not there is a problem with the gas spring 26.

[0155] In the gas spring abnormality detection device 60x of this embodiment, the corrected internal pressure feature quantity acquired by the corrected internal pressure feature quantity data acquisition unit 64 is the median, mean, or root mean square obtained from the time-series data of the corrected internal pressure.

[0156] This makes it possible to appropriately suppress the effects of the correction internal pressure characteristics on acceleration and deceleration in the stroke direction of the gas spring 26, or on external vibrations applied to the gas spring 26.

[0157] Next, a sixth embodiment will be described. Figure 18 is a schematic block diagram showing the electrical configuration of the gas spring life prediction device 70 of this embodiment. In the description of this embodiment, the same or similar components as those in the previously described embodiments are denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0158] The gas spring life prediction device 70 of this embodiment differs from the gas spring life prediction device 60 of the fourth embodiment in that it does not acquire internal pressure data in the form of time-series data and does not acquire feature quantities. The gas spring life prediction device 70 comprises an internal pressure data acquisition unit (internal pressure information acquisition unit) 71, an internal pressure data correction unit (internal pressure information correction unit) 72, a corrected internal pressure data storage unit (corrected internal pressure information storage unit) 73, a gas spring life prediction unit 74, and a display unit 75.

[0159] The internal pressure data acquisition unit 71 acquires internal pressure data, which is data relating to the internal pressure of the gas spring 26, at appropriate timings. In this embodiment, the internal pressure data is the internal pressure of the gas spring 26 actually detected by the pressure detector 27, similar to the fourth embodiment. In this embodiment, the robot 20 performs an operation corresponding to the evaluation operation pattern. Each time this operation is performed, the internal pressure data acquisition unit 71 acquires internal pressure data only once. Unless the internal pressure data waveform is sampled during the process of the evaluation operation pattern, the number of times internal pressure data is acquired is not limited to one, but may be multiple times.

[0160] In this embodiment, the lifespan prediction of the robot 20 and the lifespan prediction of the gas spring 26 are performed in parallel. The lifespan prediction of the robot 20 can be performed, for example, by the lifespan prediction device 40 of the first embodiment, the lifespan prediction device 50 of the third embodiment, etc.

[0161] The acquisition of internal pressure data will be described in detail below. For the purpose of predicting the lifespan of the robot 20, the robot 20 performs an operation corresponding to an evaluation operation pattern, which is based on the execution of a program in the robot 20's controller 30. In this embodiment, the program contains an instruction for the robot 20 to detect the internal pressure of the gas spring 26 during or before / after the execution of an operation corresponding to one evaluation operation pattern. In the controller 30, a detection instruction signal is generated internally in response to this instruction in the program. Based on this detection instruction signal, the controller 30 outputs a control signal to the pressure detector 27, controlling the pressure detector 27 to detect the internal pressure of the gas spring 26. Thus, the detection instruction signal generated internally in the controller 30 corresponds to the control signal input from the controller 30 to the pressure detector 27. The detection result of the pressure detector 27 may be output to the gas spring lifespan prediction device 70 via the controller 30, or it may be output directly to the gas spring lifespan prediction device 70.

[0162] The timing of the detection instruction signal can be arbitrarily determined. However, while the arm 24 is in operation, vibrations of the gas inside the cylinder 26a of the gas spring 26 may affect the detected value of the pressure detector 27. Also, when the arm 24 is in operation, vibrations resulting from its operation may affect the detected value of the pressure detector 27. Therefore, it is preferable that the program described above be written so that the internal pressure is detected by the pressure detector 27 when the arm 24 is stopped.

[0163] The following describes in detail an example of a program for when the robot 20 performs a regeneration operation. Before the regeneration operation program is executed in the controller 30, the current value of the drive motor 25 (servo motor) is zero. At this time, the posture of the arm 24 is maintained because an electromagnetic brake (not shown) is operating at the joint.

[0164] When the regeneration program is started in the controller 30, the brake is released, and almost simultaneously, current begins to flow to the servo motor. At this point, the output shaft of the servo motor is controlled to stop. After a certain amount of time has elapsed, which is necessary for the angle of the output shaft of the servo motor to stabilize, the servo motor starts to rotate. This effectively starts the operation of the robot 20. Subsequently, the robot 20 performs the operations described in the program.

[0165] Once the programmed sequence of actions is complete, the servo motor is controlled to stop rotating. After the servo motor stops rotating, the electromagnetic brake is controlled to operate again. Then, the program execution ends.

[0166] In the controller 30, a program start signal is generated internally when program execution begins. Control to start the rotation of the servo motor is performed based on the servo motor rotation signal generated internally in the controller 30. Control to release the electromagnetic brake is performed based on the brake release signal generated internally in the controller 30.

[0167] The program for causing the robot 20 to perform actions corresponding to the evaluation action pattern is substantially the same as the program for the playback action described above, except that the actions to be performed by the robot 20 are different.

[0168] When instructing the robot 20 to perform an action corresponding to the evaluation action pattern, the program start signal, servo motor rotation signal, or brake release signal can be used as the detection instruction signal. In this case, the effort of specifically writing an instruction to detect the internal pressure of the gas spring 26 into the program for instructing the robot 20 to perform an action corresponding to the evaluation action pattern can be omitted. Normally, the arm 24 is stopped when the program starts executing, when the servo motor starts rotating, or when the brake is released. Therefore, by using the program start signal, servo motor rotation signal, or brake release signal as the detection instruction signal, the internal pressure of the gas spring 26 can be detected while the arm 24 is stopped. As a result, the internal pressure of the gas spring 26 can be accurately detected while preventing the effects of gas vibrations and the like.

[0169] The internal pressure data correction unit 72 corrects the internal pressure data based on position information, which is information regarding the position of the arm 24. The internal pressure data correction unit 72 eliminates the influence on the internal pressure caused by changes in the position of the arm 24 through correction.

[0170] The position of arm 24 at the time of acquiring internal pressure data is usually constant in simple cases, but it may change for some reason. For example, when there are multiple types of evaluation operation patterns. Another example is when detection instruction signals are generated at multiple timings within a single evaluation operation pattern, and internal pressure data is acquired for each of them. Furthermore, when internal pressure data is acquired triggered by the automatic detection of an evaluation operation pattern from the actual movements performed by robot 20, it is possible that multiple movements determined to match the evaluation operation pattern may not strictly coincide with each other.

[0171] The corrected internal pressure data storage unit 73 stores corrected internal pressure time-series data, which is time-series data of internal pressure data corrected by the internal pressure data correction unit 72. The corrected internal pressure time-series data can be obtained by correcting the internal pressure data acquired repeatedly by the internal pressure data acquisition unit 71 periodically or irregularly (for example, once a day) and arranging them in chronological order.

[0172] The gas spring life prediction unit 74 creates a trend line based on the corrected internal pressure time series data stored in the corrected internal pressure data storage unit 73, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing of the gas spring 26. This process is the same as that of the gas spring life prediction unit 65 in the fourth embodiment, except for creating a trend line using the corrected internal pressure time series data, so a detailed explanation is omitted.

[0173] The display unit 75 displays a trend line and predicted life timing based on the prediction processing results performed by the gas spring life prediction unit 74. An example of the display screen is similar to that of the fourth embodiment (Figure 16), so its explanation is omitted. This allows the operator to appropriately consider the maintenance timing of the gas spring 26.

[0174] As described above, the robot 20 of this embodiment includes an arm 24, a drive motor 25, a gas spring 26, a controller 30, and a pressure detector 27. The drive motor 25 drives the arm 24. The gas spring 26 supports the load acting on the arm 24 and reduces the load on the drive motor 25. The controller 30 controls the drive motor 25. The pressure detector 27 detects the internal pressure of the gas spring 26 each time a detection instruction signal is generated inside the controller 30. The gas spring life prediction device 70 includes an internal pressure data acquisition unit 71, an internal pressure data correction unit 72, a corrected internal pressure data storage unit 73, and a gas spring life prediction unit 74. The internal pressure data acquisition unit 71 acquires the internal pressure detected by the pressure detector 27. The internal pressure data correction unit 72 corrects the internal pressure acquired by the internal pressure data acquisition unit 71 based on position information, which is information regarding the position of the arm 24. The corrected internal pressure data storage unit 73 stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure data correction unit 72. The gas spring life prediction unit 74 creates a trend line based on the time-series data of the corrected internal pressure stored in the corrected internal pressure data storage unit 73, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing.

[0175] This allows the effect of changes in the position of the arm 24 on the internal pressure to be eliminated by correction, thus enabling a proper maintenance plan for the gas spring 26.

[0176] Next, a seventh embodiment will be described. Figure 19 is a schematic block diagram showing the electrical configuration of the gas spring abnormality detection device 70x of this embodiment. In the description of this embodiment, the same or similar components as those in the previously described embodiments are denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0177] Unlike the sixth embodiment described above, which determines the predicted life timing of the gas spring 26, this embodiment determines whether or not an abnormality has occurred in the gas spring 26 based on whether or not the feature quantity is above a threshold. In this respect, the gas spring abnormality determination device 70x of this embodiment is similar to the gas spring abnormality determination device 60x of the fifth embodiment.

[0178] The gas spring abnormality detection device 70x comprises an internal pressure data acquisition unit 71, an internal pressure data correction unit 72, a corrected internal pressure data storage unit 73, a gas spring abnormality detection unit 76, and a display unit 75.

[0179] In the gas spring abnormality determination device 70x of this embodiment, a gas spring abnormality determination unit 76 is provided instead of the gas spring life prediction unit 74 of the sixth embodiment. The gas spring abnormality determination unit 76 compares the corrected internal pressure stored in the corrected internal pressure data storage unit 73 with a threshold value and determines whether or not an abnormality has occurred in the gas spring 26 based on this comparison result.

[0180] As described above, the gas spring abnormality determination device 70x of this embodiment determines whether or not there is an abnormality in the gas spring 26 provided by the robot 20. The gas spring abnormality determination device 70x comprises an internal pressure data acquisition unit 71, an internal pressure data correction unit 72, and a gas spring abnormality determination unit 76. The internal pressure data acquisition unit 71 acquires the internal pressure detected by the pressure detector 27. The internal pressure data correction unit 72 corrects the internal pressure acquired by the internal pressure data acquisition unit 71 based on position information, which is information regarding the position of the arm 24. The gas spring abnormality determination unit 76 compares the corrected internal pressure stored in the corrected internal pressure data storage unit 73 with a preset threshold and determines whether or not there is an abnormality in the gas spring 26 based on the comparison result.

[0181] This allows the influence of changes in the position of the arm 24 on the internal pressure to be eliminated by correction, thus enabling accurate determination of whether or not there is a problem with the gas spring 26.

[0182] Next, we will describe a modified example in which temperature compensation is applied to the fourth to seventh embodiments described above.

[0183] In this modified example, a temperature sensor (not shown) is provided on the gas spring 26. The detection unit of the temperature sensor can be attached, for example, to the outer wall of the cylinder 26a. The detection unit of the temperature sensor may also be attached to an appropriate location on the robot 20 (preferably near the cylinder 26a). The temperature sensor may be provided on the robot 20, on the controller 30, or on the gas spring life prediction device 60, 70 or the gas spring abnormality determination device 60x, 70x. The temperature sensor may be a room temperature sensor if the temperature of the cylinder 26a can be considered to be the same as the room temperature. The detected value of the temperature sensor is input to the gas spring life prediction device 60, 70 or the gas spring abnormality determination device 60x, 70x of the fourth to seventh embodiments. The gas spring life prediction device 60, 70 or the gas spring abnormality determination device 60x, 70x includes a temperature acquisition unit (not shown) that acquires the temperature detected by the temperature sensor.

[0184] As mentioned above, the pressure P of a gas follows the equation of state PV=nRT, so even if the number of moles n and the volume V are the same, the pressure P changes depending on the temperature T. On the other hand, it is difficult to maintain a constant temperature in the cylinder 26a of the gas spring 26. In order to eliminate the effects of temperature changes, in this modified example, the internal pressure detected by the pressure detector 27 is corrected based on the temperature detected by the temperature sensor so that it becomes a pressure corresponding to a predetermined reference temperature. The correction can be performed using the above-mentioned equation of state. The reference temperature can be arbitrarily determined, but for example, it can be 20°C. In the fourth and fifth embodiments, the internal pressure data correction unit 62 can perform the temperature correction, and in the sixth and seventh embodiments, the internal pressure data correction unit 72 can perform it. The correction based on the position information mentioned above may be performed first, or the correction based on the temperature sensor's detected value may be performed first.

[0185] As described above, in this modified example, the gas spring life prediction device 60, 70 or the gas spring abnormality determination device 60x, 70x includes a temperature acquisition unit (not shown) that acquires the temperature detected by the temperature sensor. The internal pressure data correction units 62, 72 correct the internal pressure of the gas spring 26 based on the temperature detected by the temperature sensor so that it corresponds to an internal pressure equivalent to a predetermined reference temperature.

[0186] This allows the effect of temperature changes on internal pressure to be corrected and eliminated. Therefore, the gas spring life prediction devices 60 and 70 can accurately determine the maintenance plan for the gas spring 26. The gas spring abnormality detection devices 60x and 70x can accurately determine whether or not there is an abnormality in the gas spring 26.

[0187] Next, an eighth embodiment will be described. Figure 20 is a front view showing the configuration of the robot system 10 according to the eighth embodiment. Figure 21 is a block diagram schematically showing the electrical configuration of the gas spring life prediction device 80 of this embodiment. In the description of this embodiment, the same or similar components as those in the previously described embodiments will be denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0188] In this embodiment, as shown in Figure 20, the gas spring 26 is not equipped with a pressure sensor 27. Instead, the controller 30 includes an internal pressure data estimation unit 31 that estimates the pressure of the gas spring 26.

[0189] As explained in the modified example of the third embodiment, the internal pressure of the gas spring 26 can be estimated from position information and current data. Specifically, when a series of operations corresponding to the evaluation operation pattern is completed, the arm 24 of the robot 20 is usually stopped at a predetermined position. In this state, the internal pressure data estimation unit 31 performs the following (i) to (iv): (i) obtain the measured value Ia of the drive current of the drive motor 25. (ii) calculate the initial drive current Ic mentioned above. (iii) find the difference between the measured drive current Ia and the calculated initial drive current Ic. (iv) calculate the estimated internal pressure of the gas spring 26 at the current time based on this difference. Since this method has already been explained in the modified example of the third embodiment, a detailed explanation of (ii) and (iv) will be omitted.

[0190] Once the series of processes (i) to (iv) are completed, one estimated internal pressure for the latest gas spring 26 is obtained. The internal pressure data estimation unit 31 outputs the obtained estimated internal pressure to the gas spring life prediction device 80.

[0191] As shown in Figure 21, the gas spring life prediction device 80 includes an estimated internal pressure data acquisition unit (estimated internal pressure information acquisition unit) 81, an estimated internal pressure data storage unit (estimated internal pressure information storage unit) 82, a gas spring life prediction unit 83, and a display unit 84.

[0192] The estimated internal pressure data acquisition unit 81 acquires the pressure of the gas spring 26 estimated by the internal pressure data estimation unit 31.

[0193] The estimated internal pressure data storage unit 82 stores estimated internal pressure time-series data, which is time-series data of estimated internal pressure acquired by the estimated internal pressure data acquisition unit 81. The estimated internal pressure time-series data can be obtained by arranging the estimated internal pressure data acquired repeatedly by the estimated internal pressure data acquisition unit 81 periodically or irregularly (for example, once a day) in chronological order.

[0194] The gas spring life prediction unit 83 creates a trend line based on the estimated internal pressure time series data stored in the estimated internal pressure data storage unit 82, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing of the gas spring 26.

[0195] The display unit 84 displays a trend line and predicted life timing based on the prediction processing results performed by the gas spring life prediction unit 83.

[0196] In this embodiment, it is assumed that the estimation of the internal pressure of the gas spring 26 is performed each time with the arm 24 stopped at a predetermined position. Therefore, in this embodiment, the internal pressure data correction units 62 and 72 that were provided in the fourth and sixth embodiments are omitted. However, if the position of the arm 24 when estimating the internal pressure of the gas spring 26 is uncertain, an internal pressure data correction unit that corrects the estimated internal pressure data based on position information may be provided.

[0197] Each time a program is executed that causes the robot 20 to perform a series of actions corresponding to an evaluation action pattern, the internal pressure data estimation unit 31 estimates the internal pressure of the gas spring 26 once. The internal pressure data estimation unit 31 can store the estimated internal pressure. Each time the internal pressure data estimation unit 31 estimates the internal pressure of the gas spring 26, it updates the stored contents with the latest estimated internal pressure. Once the update is complete, an estimated internal pressure update signal is generated inside the controller 30. Based on the estimated internal pressure update signal, the controller 30 outputs the latest estimated internal pressure to the gas spring life prediction device 80. The estimated internal pressure data acquisition unit 81 acquires the estimated internal pressure input to the gas spring life prediction device 80. In this way, the estimated internal pressure data acquisition unit 81 can acquire the latest estimated internal pressure each time the estimated internal pressure of the gas spring 26 is updated in the internal pressure data estimation unit 31 of the controller 30.

[0198] In a modified version of the third embodiment, an empirical formula defining the relationship between the cumulative stroke amount of the piston rod 26b and the amount of gas leakage is described. The internal pressure data estimation unit 31 can also estimate the internal pressure of the gas spring 26 based on this empirical formula. That is, the internal pressure data estimation unit 31 determines the stroke amount of the piston rod 26b relative to the cylinder 26a from the position information, calculates the amount of gas leakage from the cumulative value of this stroke amount based on the empirical formula described above, and estimates the pressure reduction rate. The internal pressure data estimation unit 31 estimates the internal pressure of the gas spring 26 based on the estimated pressure reduction rate. When estimating the internal pressure based on the pressure reduction rate, the estimated internal pressure data can also be obtained in the form of time-series data by performing this estimation iteratively at short time intervals. However, considering the computational load, it is advantageous to configure the system so that the internal pressure is estimated once each time the program is executed.

[0199] Instead of providing an internal pressure data estimation unit 31 in the controller 30, the estimated internal pressure data acquisition unit 81 in the gas spring life prediction device 80 may estimate the internal pressure of the gas spring 26 by calculation.

[0200] As described above, the gas spring life prediction device 80 of this embodiment comprises an estimated internal pressure data acquisition unit 81, an estimated internal pressure data storage unit 82, and a gas spring life prediction unit 83. The estimated internal pressure data acquisition unit 81 acquires the estimated internal pressure. This estimated internal pressure is the internal pressure of the gas spring 26 estimated from position information and current data, or the internal pressure of the gas spring 26 estimated from the operation information of the gas spring 26 obtained from the position information. The estimated internal pressure data storage unit 82 stores the time-series data of the estimated internal pressure acquired by the estimated internal pressure data acquisition unit 81. The gas spring life prediction unit 83 creates a trend line based on the time-series data of the estimated internal pressure acquired by the estimated internal pressure information acquisition unit, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing.

[0201] This makes it possible to predict the lifespan of the gas spring 26 without installing a pressure sensor 27, thus enabling the creation of a low-cost maintenance plan for the gas spring 26.

[0202] Next, a modified example in which temperature compensation is applied to the eighth embodiment described above will be explained.

[0203] In the eighth embodiment described above, an example is described in which the drive current Ia of the servo motor is actually measured and the internal pressure of the gas spring 26 is estimated using this measurement. When using this estimation method, applying temperature compensation is preferable because it eliminates the influence of temperature changes on the estimated internal pressure.

[0204] In this modified example, similar to the modifications in the fourth to seventh embodiments where temperature compensation is applied, a temperature sensor (not shown) is provided on the gas spring 26. The value detected by the temperature sensor is input to the controller 30. The internal pressure data estimation unit 31 of the controller 30 corrects the estimated internal pressure of the gas spring 26 based on the temperature detected by the temperature sensor so that it becomes a pressure corresponding to the aforementioned reference temperature. The correction can be performed using the above-described equation of state.

[0205] As described above, in this modified gas spring life prediction device, the estimated internal pressure acquired by the estimated internal pressure data acquisition unit 81 is corrected to become the internal pressure of the gas spring 26 corresponding to a predetermined reference temperature, based on the temperature acquired by a temperature sensor capable of detecting temperature.

[0206] This allows the influence of temperature changes on the estimated internal pressure to be eliminated through correction. Therefore, a maintenance plan for the gas spring 26 can be properly established.

[0207] Preferred embodiments and modifications of the present disclosure have been described above, but the above configuration can be modified as follows, for example. Modifications may be made individually or in any combination of multiple modifications.

[0208] The second trend line TL2 shown in Figure 7 may be created based only on feature time series data from the present to a predetermined number of days later. The same applies to the trend line TL in Figure 14.

[0209] In the example in Figure 7, a single comprehensive trend line may be created based on both the series of feature time series data prior to the corrected most recent supplementation timing t1 and the feature time series data after the corrected most recent supplementation timing t1. This trend line may also be created based only on feature time series data from the present within a predetermined number of days. The same applies to the example in Figure 11.

[0210] An arm (not shown) may be provided as a second arm, which is pivotably attached to the tip of arm 24, and a second gas spring (not shown) may be provided between arm 24 and the second arm. In this case, a second drive motor for driving the second arm is provided at the joint between arm 24 and the second arm. The support force of the second gas spring can cancel at least a portion of the load acting on the second arm (or the gravitational moment of the second arm), thereby reducing the load on the second drive motor. The technology of this disclosure is also applicable to the second arm, the second drive motor, and the second gas spring.

[0211] For example, the pressure reduction ratio μ* in the second gas spring (i.e., the ratio of the current internal pressure to the initial internal pressure of the second gas spring) can be determined, and the current data of the second drive motor can be corrected using the determined pressure reduction ratio μ* in the same manner as in the third embodiment described above. The current internal pressure of the second gas spring may be measured using a second pressure detector (not shown) or estimated using a predetermined method. As an example of the latter, the operation information of the second gas spring (for example, the stroke amount of the piston rod relative to its cylinder) can be determined from the movement of the robot 20 (more specifically, the position information of the arm 24 and the second arm), and the amount of gas leakage in the second gas spring can be calculated from an empirical formula that defines the relationship between the integrated value of the stroke amount of the piston rod and the amount of gas leakage, thereby estimating the pressure reduction ratio μ* in the second gas spring. By dividing the initial cancellation equivalent current of the second drive motor by the estimated pressure reduction ratio μ*, the increase in the drive current of the second motor when the arm 24 and the second arm are at any angular position at the present time can be determined. By subtracting this increase from the measured value of the second motor's drive current, the current data for the second drive motor can be corrected so that the effects of gas leakage are substantially eliminated.

[0212] The gas spring 26 in the above embodiments and modified examples has a configuration in which its overall length increases as the internal pressure increases, but instead, a gas spring having a configuration in which its overall length decreases as the internal pressure increases may be used. In this case as well, the technology of this disclosure can be applied. As an example, similar to the third embodiment described above, a pressure detector, a force sensor (e.g., a load cell), or a strain gauge can be provided on the gas spring to detect the internal pressure or support force of the gas spring, and the current data of the drive motor 25 can be corrected after determining the pressure reduction rate. As another example, the estimated internal pressure of the gas spring can be estimated from the position information of the arm 24 and the current data of the drive motor 25, or from the operation information of the gas spring, and the current data of the drive motor 25 can be corrected after determining the pressure reduction rate.

[0213] The life prediction devices 40 and 50 do not need to be provided separately from the controller 30, but may be built into the controller 30. Alternatively, the life prediction devices 40 and 50 may not have a separate computer functioning as a CPU, RAM, ROM, auxiliary storage device, etc., and may be implemented using the computer of the robot 20's controller 30. In this case, the display units 47 and 56 can be configured, for example, as part of the robot 20's teaching pendant. The same applies to the gas spring life prediction devices 60, 70, and 80, and the gas spring abnormality detection devices 60x and 70x.

[0214] The functions of the elements disclosed herein can be performed using circuits or processing circuits, including general-purpose processors, dedicated processors, integrated circuits, ASICs (Application Specific Integrated Circuits), conventional circuits, and / or combinations thereof, configured or programmed to perform the disclosed functions. A processor is considered a processing circuit or circuit because it includes transistors and other circuits. In this disclosure, a circuit, unit, or means is hardware that performs the enumerated functions, or hardware programmed to perform the enumerated functions. The hardware may be hardware disclosed herein, or other known hardware that is programmed or configured to perform the enumerated functions. If the hardware is a processor, which is considered a type of circuit, then the circuit, means, or unit is a combination of hardware and software, and the software is used to configure the hardware and / or the processor.

[0215] [Aspect] The embodiments described above are specific examples of the following embodiments. (Aspect 1) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction device that predicts the lifespan of a robot equipped with the drive motor based on information relating to the drive motor, An input unit to which a signal indicating that gas has been replenished to the gas spring is input, A replenishment timing storage unit stores the replenishment timing, which is the timing at which the gas replenishment was performed, based on the input of the signal to the input unit. A current data acquisition unit acquires current time-series data, which is time-series data of information regarding the drive current of the aforementioned drive motor. A feature data acquisition unit acquires and stores feature time series data, which is feature time series data of the feature quantities of the current time series data. A time-series data correction unit corrects the series of feature time-series data so that the progression of the series of feature time-series data prior to the most recent replenishment timing matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing, A lifetime prediction unit creates a first trend line based on the corrected series of feature time series data and determines the timing at which the first trend line reaches a predetermined lifetime threshold as the first predicted lifetime timing. A life prediction device equipped with the following features. (Aspect 2) A life prediction device according to Embodiment 1, When considering a graph where the horizontal axis is time and the vertical axis is the feature, the slope of the first line passing through the point of the pre-refill early feature value obtained from one or more of the initial features in the series of feature time-series data and the point of the pre-refill late feature value obtained from one or more of the final features before the most recent refill timing is defined as the pre-refill slope, the slope of the second line passing through the point of the pre-refill early feature value and the point of the post-refill early feature value is defined as the post-refill slope, and the ratio of the post-refill slope to the pre-refill slope is defined as the slope ratio. The time-series data correction unit corrects the series of feature time-series data prior to the most recent replenishment timing by multiplying the difference between each of the feature quantities included in the series of feature time-series data prior to the most recent replenishment timing and the initial feature quantity value before replenishment by the slope ratio, and adding the initial feature quantity value before replenishment to the multiplied value, thereby correcting the series of feature time-series data prior to the most recent replenishment timing. (Aspect 3) A life prediction device according to embodiment 2, The time-series data correction unit corrects the series of feature time-series data after the most recent replenishment timing by multiplying the difference between each of the features included in the series of feature time-series data after the most recent replenishment timing and the initial feature value after replenishment by the slope ratio, and adding the initial feature value after replenishment to the multiplied value. The lifetime prediction unit creates a second trend line based on a series of feature time series data after the corrected most recent supplementation timing, and determines the timing at which the second trend line reaches the predetermined lifetime threshold as the second predicted lifetime timing, in this lifetime prediction device. (Aspect 4) A life prediction device according to Embodiment 1, When considering a graph where the horizontal axis is time and the vertical axis is the feature, the line passing through the point of the pre-refill early feature value obtained from one or more of the initial features in the series of feature time-series data and the point of the pre-refill late feature value obtained from one or more of the final features before the most recent refill timing is defined as the pre-refill line, the line passing through the point of the pre-refill early feature value and the point of the post-refill early feature value is defined as the post-refill line, and the angle between the pre-refill line and the post-refill line is defined as the difference angle. The time-series data correction unit corrects the series of feature time-series data preceding the most recent replenishment timing by performing a rotation transformation corresponding to the difference angle on all features included in the series of feature time-series data preceding the most recent replenishment timing, using the point of the initial feature value before replenishment as the rotation center. (Appendix 5) A life prediction device according to aspect 4, The time-series data correction unit corrects the series of feature time-series data after the most recent replenishment timing by using the point of the initial feature value after replenishment as the rotation center and performing a rotation transformation corresponding to the difference angle for all features included in the series of feature time-series data after the most recent replenishment timing. The lifetime prediction unit creates a second trend line based on a series of feature time series data after the corrected most recent supplementation timing, and determines the timing at which the second trend line reaches the predetermined lifetime threshold as the second predicted lifetime timing, in this lifetime prediction device. (Aspect 6) A life prediction device according to embodiment 3 or embodiment 5, A life prediction device further comprising a display unit that displays both the first trend line and the first predicted life timing, and the second trend line and the second predicted life timing. (Aspect 7) A life prediction device according to aspect 6, Life prediction device wherein, when the number of features included in the series of feature time series data after the most recent replenishment timing is greater than or equal to a predetermined number, the display unit does not display the first trend line and the first predicted life timing, but displays the second trend line and the second predicted life timing. (Pattern 8) A life prediction device according to embodiment 3 or embodiment 5, It further includes a display unit that shows information related to lifespan prediction, The aforementioned display unit is If the number of features included in the series of feature time-series data after the most recent replenishment timing is less than a predetermined number, the first trend line and the first predicted lifetime timing are displayed, while the second trend line and the second predicted lifetime timing are not displayed. A lifetime prediction device that, if the number of features included in the series of feature time-series data after the most recent replenishment timing is greater than or equal to a predetermined number, does not display the first trend line and the first predicted lifetime timing, but displays the second trend line and the second predicted lifetime timing. (Aspect 9) A life prediction device according to any one of embodiments 1 to 8, Life prediction device, wherein the feature quantities of the current time series data are the root mean square, peak value, peak-to-peak value, or frequency analysis integrated value obtained from the current time series data, or the dissimilarity between the current time series data and current time series data acquired in a drive motor immediately after the start of use. (Aspect 10) A life prediction device according to aspect 9, Life prediction device, wherein the feature quantity of the current time series data is the dissimilarity, which is determined based on the sum or average value of the Euclidean distance or DTW distance between the current time series data and the current time series data acquired in the drive motor immediately after the start of use. (Aspect 11) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction device that predicts the lifespan of a robot equipped with the drive motor based on information relating to the drive motor, A current data acquisition unit acquires current data, which is data relating to the drive current of the aforementioned drive motor. A current data correction unit corrects the current data based on internal pressure information, which is information regarding the internal pressure within the gas spring, and position information, which is information regarding the position of the arm. A corrected current data storage unit stores corrected current time-series data, which is the time-series data of the corrected current data. A correction feature data acquisition unit acquires correction feature time series data, which is time series data of the feature quantities of the correction current time series data. A lifetime prediction unit creates a trend line based on the corrected feature time series data and determines the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing. A life prediction device equipped with the following features. (Aspect 12) A life prediction device according to embodiment 11, The life prediction device wherein the internal pressure information is the internal pressure of the gas spring detected by a pressure detector. (Aspect 13) A life prediction device according to embodiment 11, Life prediction device, wherein the internal pressure information is the estimated internal pressure of the gas spring estimated from the position information and the current data, or the estimated internal pressure of the gas spring estimated from the operation information of the gas spring obtained from the position information. (Aspect 14) A life prediction device according to any one of the embodiments 11 to 13, Life prediction device, wherein the feature quantities of the corrected current time series data are the root mean square, peak value, peak-to-peak value, or frequency analysis integrated value obtained from the corrected current time series data, or the dissimilarity between the corrected current time series data and the corrected current time series data obtained in the drive motor immediately after the start of use. (Aspect 15) A life prediction device according to embodiment 14, Life prediction device, wherein the feature quantity of the corrected current time series data is the dissimilarity determined based on the sum or average value of the Euclidean distance or DTW distance between the corrected current time series data and the corrected current time series data acquired in the drive motor immediately after the start of use. (Aspect 16) A life prediction device according to any one of embodiments 1 to 15, Life prediction device, wherein the information relating to the drive current is a measured value of the drive current, a current command value for the drive motor, or a position deviation which is the difference between the measured position and the position command value in the drive motor. (Aspect 17) A method for using a life prediction device described in any one of embodiments 1 to 16, A method for using a life prediction device, wherein, when the gas spring is refilled multiple times, the multiple refills are performed with the arms aligned. (Aspect 18) A method for using a life prediction device described in any one of embodiments 1 to 17, When the gas spring is refilled multiple times, each of the multiple gas refills is performed after determining the target internal pressure value, which is the target internal pressure of the gas spring, based on the surface temperature of the gas spring at the time of the gas refill. A method for using a life prediction device, wherein the internal pressure target value is set so that the total amount of gas in the gas spring is the same or approximately the same after each of the multiple gas replenishments has been performed. (Aspect 19) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction method for predicting the life of a robot equipped with the drive motor, based on information about the drive motor, A signal is input to indicate that gas has been replenished to the gas spring. Based on the input of the aforementioned signal, the replenishment timing, which is the timing at which the gas replenishment was performed, is stored. The current time series data, which is time series data of information regarding the drive current of the aforementioned drive motor, is acquired. The feature time series data, which is the feature time series data of the aforementioned current time series data, is acquired and stored. The series of feature time series data prior to the most recent replenishment timing is corrected so that it matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing. A lifetime prediction method comprising creating a first trend line based on the corrected series of feature time series data, and determining the timing at which the first trend line reaches a predetermined lifetime threshold as a first predicted lifetime timing. (Aspect 20) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction method for predicting the life of a robot equipped with the drive motor, based on information about the drive motor, The current data, which is data relating to the drive current of the aforementioned drive motor, is acquired. Based on the internal pressure information, which is information regarding the internal pressure within the gas spring, and the position information, which is information regarding the position of the arm, the current data is corrected. The corrected current time series data, which is the time series data of the corrected current data, is stored. The corrected feature time series data, which is the time series data of the feature quantities of the corrected current time series data, is obtained. A lifetime prediction method that creates a trend line based on the corrected feature time series data and determines the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing. (Aspect 21) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction program that predicts the lifespan of a robot equipped with the drive motor, based on information about the drive motor, A signal input step in which a signal indicating that gas has been replenished to the gas spring is input, A replenishment timing storage step that stores the replenishment timing, which is the timing at which the gas replenishment was performed, based on the input of the signal, A current data acquisition step is to acquire current time series data, which is time series data of information regarding the drive current of the drive motor, A feature data acquisition step involves acquiring and storing feature time series data, which is feature time series data of the feature quantities of the current time series data. A time-series data correction step in which the series of feature time-series data prior to the most recent replenishment timing is corrected so that it matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing, A lifetime prediction step involves creating a first trend line based on the corrected series of feature time series data, and determining the timing at which the first trend line reaches a predetermined lifetime threshold as the first predicted lifetime timing. A lifespan prediction program that is executed by a computer. (Aspect 22) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction program that predicts the lifespan of a robot equipped with the drive motor, based on information about the drive motor, A current data acquisition step is to acquire current data, which is data relating to the drive current of the drive motor, A current data correction step in which the current data is corrected based on internal pressure information, which is information regarding the internal pressure in the gas spring, and position information, which is information regarding the position of the arm. A corrected current data storage step, which stores corrected current time-series data, which is the time-series data of the corrected current data, A correction feature data acquisition step involves acquiring correction feature time series data, which is time series data of the feature quantities of the correction current time series data; A lifetime prediction step involves creating a trend line based on the corrected feature time series data and determining the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing. A lifespan prediction program that is executed by a computer. (Aspect 23) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A gas spring life prediction device for predicting the life of the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, A corrected internal pressure feature data acquisition unit acquires corrected internal pressure feature time series data, which is time series data of the feature quantities of the time series data of the corrected internal pressure stored in the corrected internal pressure information storage unit. A gas spring life prediction unit creates a trend line based on the corrected internal pressure feature time series data and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction device equipped with the following features. (Aspect 24) A gas spring life prediction device according to embodiment 23, A gas spring life prediction device in which the feature quantities of the time-series data of the corrected internal pressure are the median, mean, or root mean square obtained from the time-series data of the corrected internal pressure. (Aspect 25) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A gas spring abnormality determination device for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, A corrected internal pressure feature acquisition unit acquires the corrected internal pressure feature quantity stored in the corrected internal pressure information storage unit, A gas spring abnormality determination unit compares the corrected internal pressure feature quantity acquired by the corrected internal pressure feature quantity acquisition unit with a preset threshold and determines whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection device equipped with the following features. (Aspect 26) A gas spring abnormality detection device according to embodiment 25, A gas spring abnormality determination device wherein the characteristic quantity of the corrected internal pressure acquired by the corrected internal pressure characteristic quantity acquisition unit is the median, mean, or root mean square obtained from the time series data of the corrected internal pressure. (Aspect 27) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a detection instruction signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring life prediction device for predicting the life of the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, A gas spring life prediction unit creates a trend line based on the time-series data of the corrected internal pressure stored in the corrected internal pressure information storage unit, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction device equipped with the following features. (Aspect 28) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a detection instruction signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring abnormality determination device for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A gas spring abnormality determination unit compares the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, with a preset threshold and determines whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection device equipped with the following features. (Aspect 29) A gas spring life prediction device according to embodiment 23, 24, or 27, It includes a temperature acquisition unit that acquires the temperature detected by the temperature sensor. The internal pressure information correction unit corrects the internal pressure of the gas spring based on the temperature acquired by the temperature acquisition unit so that it corresponds to an internal pressure equivalent to a predetermined reference temperature, and is a gas spring life prediction device. (Aspect 30) A gas spring abnormality detection device according to embodiment 25, 26, or 28, It includes a temperature acquisition unit that acquires the temperature detected by the temperature sensor. The internal pressure information correction unit corrects the internal pressure of the gas spring based on the temperature acquired by the temperature acquisition unit so that it corresponds to an internal pressure equivalent to a predetermined reference temperature, and is a gas spring abnormality determination device. (Aspect 31) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A gas spring life prediction device for predicting the life of the gas spring of a robot having the same, An estimated internal pressure information acquisition unit that acquires an estimated internal pressure, which is the internal pressure of the gas spring estimated from position information that is information regarding the position of the arm and current data that is data regarding the drive current of the drive motor, or the internal pressure of the gas spring estimated from the operation information of the gas spring obtained from the position information, An estimated internal pressure information storage unit that stores time-series data of the estimated internal pressure acquired by the estimated internal pressure information acquisition unit, A gas spring life prediction unit that creates a trend line based on the time-series data of the estimated internal pressure stored by the estimated internal pressure information storage unit and obtains the timing at which the trend line reaches a predetermined life threshold as the predicted life timing, A gas spring life prediction device comprising the same. (Aspect 32) The gas spring life prediction device according to Aspect 31, The estimated internal pressure acquired by the estimated internal pressure information acquisition unit is corrected based on the temperature detected by a temperature sensor so as to be the internal pressure of the gas spring corresponding to a predetermined reference temperature. A gas spring life prediction device. (Aspect 33) An arm, A drive motor that drives the arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector that detects the internal pressure of the gas spring, A gas spring life prediction method for predicting the life of the gas spring of a robot having the same, Acquire the internal pressure detected by the pressure detector, Correct the acquired internal pressure based on position information that is information regarding the position of the arm, Store time-series data of the corrected internal pressure that is the corrected internal pressure, Acquire corrected internal pressure feature quantity time-series data that is time-series data of feature quantities of the time-series data of the corrected internal pressure stored, A gas spring lifetime prediction method comprising creating a trend line based on the corrected internal pressure feature time series data, and determining the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing. (Aspect 34) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A method for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: The internal pressure detected by the pressure detector is acquired, The acquired internal pressure is corrected based on position information, which is information regarding the position of the arm. The time-series data of the corrected internal pressure, which is the corrected internal pressure, is stored. The feature quantities of the time-series data of the stored corrected internal pressure are obtained, A method for determining whether a gas spring is abnormal, comprising comparing the aforementioned feature quantity with a preset threshold and determining whether or not there is an abnormality in the gas spring based on the comparison result. (Aspect 35) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a detection instruction signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A method for predicting the lifespan of a gas spring in a robot equipped with the following features: The internal pressure detected by the pressure detector is acquired, The acquired internal pressure is corrected based on position information, which is information regarding the position of the arm. The time-series data of the corrected internal pressure, which is the corrected internal pressure, is stored. A gas spring life prediction method comprising creating a trend line based on the stored time-series data of the corrected internal pressure, and determining the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. (Aspect 36) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a detection instruction signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A method for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: The internal pressure detected by the pressure detector is acquired, The acquired internal pressure is corrected based on position information, which is information regarding the position of the arm. A method for determining whether a gas spring is abnormal, comprising comparing the corrected internal pressure, which is the corrected internal pressure, with a preset threshold, and determining whether or not there is an abnormality in the gas spring based on the comparison result. (Aspect 37) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A method for predicting the lifespan of a gas spring in a robot equipped with the following features: The internal pressure of the gas spring is estimated from position information, which is information relating to the position of the arm, and current data, which is data relating to the drive current of the drive motor, or the estimated internal pressure is obtained from the operation information of the gas spring, which is obtained from the position information. The time-series data of the acquired estimated internal pressure is stored, A gas spring life prediction method for creating a trend line based on the time series data of the estimated internal pressure stored and obtaining the timing when the trend line reaches a predetermined life threshold as the predicted life timing. (Aspect 38) An arm, A drive motor that drives the arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector that detects the internal pressure of the gas spring, A gas spring life prediction program for predicting the life of the gas spring of a robot including: An internal pressure information acquisition step of acquiring the internal pressure detected by the pressure detector; An internal pressure information correction step of correcting the acquired internal pressure based on position information which is information regarding the position of the arm; A corrected internal pressure information storage step of storing time series data of the corrected internal pressure which is the corrected internal pressure; A corrected internal pressure feature amount data acquisition unit that acquires corrected internal pressure feature amount time series data which is time series data of the feature amount of the time series data of the corrected internal pressure stored; A gas spring life prediction step of creating a trend line based on the corrected internal pressure feature amount time series data and obtaining the timing when the trend line reaches a predetermined life threshold as the predicted life timing; A gas spring life prediction program that causes a computer to execute. (Aspect 39) An arm, A drive motor that drives the arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector that detects the internal pressure of the gas spring, A gas spring abnormality determination program for determining the presence or absence of an abnormality in the gas spring of a robot including: An internal pressure acquisition step of acquiring the internal pressure detected by the pressure detector; An internal pressure information correction step is performed to correct the acquired internal pressure based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage step that stores time-series data of the corrected internal pressure, which is the corrected internal pressure, A corrected internal pressure feature acquisition step involves acquiring the feature quantities of the time-series data of the stored corrected internal pressure, A gas spring abnormality determination step involves comparing the aforementioned feature quantity with a preset threshold and determining whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection program that is executed by a computer. (Approach 40) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a detection instruction signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring life prediction program for predicting the life of the gas spring of a robot equipped with the following: An internal pressure information acquisition step is to acquire the internal pressure detected by the pressure detector, An internal pressure information correction step is performed to correct the acquired internal pressure based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the corrected internal pressure. A gas spring life prediction step involves creating a trend line based on the stored time-series data of the corrected internal pressure, and determining the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction program that is run on a computer. (Aspect 41) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a detection instruction signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring abnormality determination program for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: An internal pressure information acquisition step is to acquire the internal pressure detected by the pressure detector, An internal pressure information correction step is performed to correct the acquired internal pressure based on position information, which is information regarding the position of the arm. A gas spring abnormality determination step involves comparing the corrected internal pressure, which is the corrected internal pressure, with a preset threshold and determining whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection program that is executed by a computer. (Aspect 42) Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A gas spring life prediction program for predicting the life of the gas spring of a robot equipped with the following: An estimated internal pressure information acquisition step, which acquires an estimated internal pressure, which is the internal pressure of the gas spring estimated from position information, which is information relating to the position of the arm, and current data, which is data relating to the drive current of the drive motor, or the internal pressure of the gas spring estimated from the operation information of the gas spring obtained from the position information, An estimated internal pressure information storage step, which stores the time-series data of the acquired estimated internal pressure, A gas spring life prediction step involves creating a trend line based on the stored time-series data of the estimated internal pressure, and determining the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction program that is run on a computer. [Explanation of Symbols]

[0216] 20 Robots 24 Arms 25 Drive motor 26 Gas springs 40 Life prediction device 41 Input section 42 Replenishment timing memory unit 43 Current data acquisition unit 44 Feature Data Acquisition Unit 45 Time-series data correction unit 46 Life Prediction Unit 47 Display section 50 Life prediction device 51 Current data acquisition unit 52 Current Data Correction Unit 53 Correction current data storage unit 54 Corrected Feature Data Acquisition Unit 55 Life Prediction Unit 56 Display section

Claims

1. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction device that predicts the lifespan of a robot equipped with the drive motor based on information relating to the drive motor, An input unit to which a signal indicating that gas has been replenished to the gas spring is input, A replenishment timing storage unit stores the replenishment timing, which is the timing at which the gas replenishment was performed, based on the input of the signal to the input unit. A current data acquisition unit acquires current time-series data, which is time-series data of information regarding the drive current of the aforementioned drive motor. A feature data acquisition unit acquires and stores feature time series data, which is feature time series data of the feature quantities of the current time series data. A time-series data correction unit corrects the series of feature time-series data so that the progression of the series of feature time-series data prior to the most recent replenishment timing matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing, A lifetime prediction unit creates a first trend line based on the corrected series of feature time series data and determines the timing at which the first trend line reaches a predetermined lifetime threshold as a first predicted lifetime timing. A life prediction device equipped with the following features.

2. A life prediction device according to claim 1, When considering a graph where the horizontal axis is time and the vertical axis is the feature, the slope of the first line passing through the point of the pre-refill early feature value obtained from one or more of the initial features in the series of feature time-series data and the point of the pre-refill late feature value obtained from one or more of the final features before the most recent refill timing is defined as the pre-refill slope, the slope of the second line passing through the point of the pre-refill early feature value and the point of the post-refill early feature value is defined as the post-refill slope, and the ratio of the post-refill slope to the pre-refill slope is defined as the slope ratio. The time-series data correction unit corrects the series of feature time-series data prior to the most recent replenishment timing by multiplying the difference between each of the feature quantities included in the series of feature time-series data prior to the most recent replenishment timing and the initial feature quantity value before replenishment by the slope ratio, and adding the initial feature quantity value before replenishment to the multiplied value, thereby correcting the series of feature time-series data prior to the most recent replenishment timing.

3. A life prediction device according to claim 2, The time-series data correction unit corrects the series of feature time-series data after the most recent replenishment timing by multiplying the difference between each of the features included in the series of feature time-series data after the most recent replenishment timing and the initial feature value after replenishment by the slope ratio, and adding the initial feature value after replenishment to the multiplied value. The lifetime prediction unit creates a second trend line based on a series of feature time series data after the corrected most recent supplementation timing, and determines the timing at which the second trend line reaches the predetermined lifetime threshold as the second predicted lifetime timing, in this lifetime prediction device.

4. A life prediction device according to claim 1, When considering a graph where the horizontal axis is time and the vertical axis is the feature, the line passing through the point of the pre-refill early feature value obtained from one or more of the initial features in the series of feature time-series data and the point of the pre-refill late feature value obtained from one or more of the final features before the most recent refill timing is defined as the pre-refill line, the line passing through the point of the pre-refill early feature value and the point of the post-refill early feature value is defined as the post-refill line, and the angle between the pre-refill line and the post-refill line is defined as the difference angle. The time-series data correction unit corrects the series of feature time-series data preceding the most recent replenishment timing by performing a rotation transformation corresponding to the difference angle on all features included in the series of feature time-series data preceding the most recent replenishment timing, using the point of the initial feature value before replenishment as the rotation center.

5. A life prediction device according to claim 4, The time-series data correction unit corrects the series of feature time-series data after the most recent replenishment timing by using the point of the initial feature value after replenishment as the rotation center and performing a rotation transformation corresponding to the difference angle for all features included in the series of feature time-series data after the most recent replenishment timing. The lifetime prediction unit creates a second trend line based on a series of feature time series data after the corrected most recent supplementation timing, and determines the timing at which the second trend line reaches the predetermined lifetime threshold as the second predicted lifetime timing, in this lifetime prediction device.

6. A life prediction device according to claim 3 or claim 5, A life prediction device further comprising a display unit that displays both the first trend line and the first predicted life timing, and the second trend line and the second predicted life timing.

7. A life prediction device according to claim 6, Life prediction device wherein, when the number of features included in the series of feature time series data after the most recent replenishment timing is greater than or equal to a predetermined number, the display unit does not display the first trend line and the first predicted life timing, but displays the second trend line and the second predicted life timing.

8. A life prediction device according to claim 3 or claim 5, It further includes a display unit that shows information related to lifespan prediction, The aforementioned display unit is If the number of features included in the series of feature time-series data after the most recent replenishment timing is less than a predetermined number, the first trend line and the first predicted lifetime timing are displayed, while the second trend line and the second predicted lifetime timing are not displayed. A lifetime prediction device that, if the number of features included in the series of feature time-series data after the most recent replenishment timing is greater than or equal to a predetermined number, does not display the first trend line and the first predicted lifetime timing, but displays the second trend line and the second predicted lifetime timing.

9. A life prediction device according to any one of claims 1 to 5, Life prediction device, wherein the feature quantities of the current time series data are the root mean square, peak value, peak-to-peak value, or frequency analysis integrated value obtained from the current time series data, or the dissimilarity between the current time series data and current time series data acquired in a drive motor immediately after the start of use.

10. A life prediction device according to claim 9, Life prediction device, wherein the feature quantity of the current time series data is the dissimilarity determined based on the sum or average value of the Euclidean distance or DTW distance between the current time series data and the current time series data acquired in the drive motor immediately after the start of use.

11. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction device that predicts the lifespan of a robot equipped with the drive motor based on information relating to the drive motor, A current data acquisition unit acquires current data, which is data relating to the drive current of the aforementioned drive motor. A current data correction unit corrects the current data based on internal pressure information, which is information regarding the internal pressure within the gas spring, and position information, which is information regarding the position of the arm. A corrected current data storage unit stores corrected current time-series data, which is the time-series data of the corrected current data. A correction feature data acquisition unit acquires correction feature time series data, which is time series data of the feature quantities of the correction current time series data. A lifetime prediction unit creates a trend line based on the corrected feature time series data and determines the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing. A life prediction device equipped with the following features.

12. A life prediction device according to claim 11, The life prediction device wherein the internal pressure information is the internal pressure of the gas spring detected by a pressure detector.

13. A life prediction device according to claim 11, Life prediction device, wherein the internal pressure information is the estimated internal pressure of the gas spring estimated from the position information and the current data, or the estimated internal pressure of the gas spring estimated from the operation information of the gas spring obtained from the position information.

14. A life prediction device according to any one of claims 11 to 13, Life prediction device, wherein the feature quantities of the corrected current time series data are the root mean square, peak value, peak-to-peak value, or frequency analysis integrated value obtained from the corrected current time series data, or the dissimilarity between the corrected current time series data and the corrected current time series data obtained in the drive motor immediately after the start of use.

15. A life prediction device according to claim 14, Life prediction device, wherein the feature quantity of the corrected current time series data is the dissimilarity determined based on the sum or average value of the Euclidean distance or DTW distance between the corrected current time series data and the corrected current time series data acquired in the drive motor immediately after the start of use.

16. A life prediction device according to any one of claims 1 to 5, or any one of claims 11 to 13, Life prediction device, wherein the information relating to the drive current is a measured value of the drive current, a current command value for the drive motor, or a position deviation which is the difference between the measured position and the position command value in the drive motor.

17. A method for using a life prediction device according to any one of claims 1 to 5, or any one of claims 11 to 13, A method for using a life prediction device, wherein, when the gas spring is refilled multiple times, the multiple refills are performed with the arms aligned.

18. A method for using a life prediction device according to any one of claims 1 to 5, or any one of claims 11 to 13, When the gas spring is refilled multiple times, each of the multiple gas refills is performed after determining the target internal pressure value, which is the target internal pressure of the gas spring, based on the surface temperature of the gas spring at the time of the gas refill. A method for using a life prediction device, wherein the internal pressure target value is set so that the total amount of gas in the gas spring is the same or approximately the same after each of the multiple gas replenishments has been performed.

19. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction method for predicting the life of a robot equipped with the drive motor, based on information about the drive motor, A signal is input to indicate that gas has been replenished to the gas spring. Based on the input of the aforementioned signal, the replenishment timing, which is the timing at which the gas replenishment was performed, is stored. The current time series data, which is time series data of information regarding the drive current of the aforementioned drive motor, is acquired. The feature time series data, which is the feature time series data of the aforementioned current time series data, is acquired and stored. The series of feature time series data prior to the most recent replenishment timing is corrected so that it matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing. A lifetime prediction method comprising creating a first trend line based on the corrected series of feature time-series data, and determining the timing at which the first trend line reaches a predetermined lifetime threshold as a first predicted lifetime timing.

20. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction method for predicting the life of a robot equipped with the drive motor, based on information about the drive motor, The current data, which is data relating to the drive current of the aforementioned drive motor, is acquired. Based on the internal pressure information, which is information regarding the internal pressure within the gas spring, and the position information, which is information regarding the position of the arm, the current data is corrected. The corrected current time series data, which is the time series data of the corrected current data, is stored. The corrected feature time series data, which is the time series data of the feature quantities of the corrected current time series data, is obtained. A lifetime prediction method that creates a trend line based on the corrected feature time series data and determines the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing.

21. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction program that predicts the lifespan of a robot equipped with the drive motor, based on information about the drive motor, A signal input step in which a signal indicating that gas has been replenished to the gas spring is input, A replenishment timing storage step that stores the replenishment timing, which is the timing at which the gas replenishment was performed, based on the input of the signal, A current data acquisition step is to acquire current time series data, which is time series data of information regarding the drive current of the drive motor, A feature data acquisition step involves acquiring and storing feature time series data, which is feature time series data of the feature quantities of the current time series data. A time-series data correction step in which the series of feature time-series data prior to the most recent replenishment timing is corrected so that it matches the initial post-replenishment feature value obtained from one or more of the initial features after the most recent replenishment timing, A lifetime prediction step involves creating a first trend line based on the corrected series of feature time series data, and determining the timing at which the first trend line reaches a predetermined lifetime threshold as the first predicted lifetime timing. A lifespan prediction program that is executed by a computer.

22. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A life prediction program that predicts the lifespan of a robot equipped with the drive motor, based on information about the drive motor, A current data acquisition step is to acquire current data, which is data relating to the drive current of the drive motor, A current data correction step in which the current data is corrected based on internal pressure information, which is information regarding the internal pressure in the gas spring, and position information, which is information regarding the position of the arm. A corrected current data storage step, which stores corrected current time-series data, which is the time-series data of the corrected current data, A correction feature data acquisition step involves acquiring correction feature time series data, which is time series data of the feature quantities of the correction current time series data; A lifetime prediction step involves creating a trend line based on the corrected feature time series data and determining the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing. A lifespan prediction program that is executed by a computer.

23. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A gas spring life prediction device for predicting the life of the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, A corrected internal pressure feature data acquisition unit acquires corrected internal pressure feature time series data, which is time series data of the feature quantities of the time series data of the corrected internal pressure stored in the corrected internal pressure information storage unit. A gas spring life prediction unit creates a trend line based on the corrected internal pressure feature time series data and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction device equipped with the following features.

24. A gas spring life prediction device according to claim 23, A gas spring life prediction device in which the characteristic quantities of the time-series data of the corrected internal pressure are the median, mean, or root mean square obtained from the time-series data of the corrected internal pressure.

25. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A gas spring abnormality determination device for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, A corrected internal pressure feature acquisition unit acquires the corrected internal pressure feature quantities stored in the corrected internal pressure information storage unit, A gas spring abnormality determination unit compares the corrected internal pressure feature quantity acquired by the corrected internal pressure feature quantity acquisition unit with a preset threshold and determines whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection device equipped with the following features.

26. A gas spring abnormality detection device according to claim 25, A gas spring abnormality determination device wherein the characteristic quantity of the corrected internal pressure acquired by the corrected internal pressure characteristic quantity acquisition unit is the median, mean, or root mean square obtained from the time series data of the corrected internal pressure.

27. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a control signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring life prediction device for predicting the life of the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, A gas spring life prediction unit creates a trend line based on the time-series data of the corrected internal pressure stored in the corrected internal pressure information storage unit, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction device equipped with the following features.

28. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a control signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring abnormality determination device for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: An internal pressure information acquisition unit that acquires the internal pressure detected by the pressure detector, An internal pressure information correction unit corrects the internal pressure acquired by the internal pressure information acquisition unit based on position information, which is information regarding the position of the arm. A gas spring abnormality determination unit compares the corrected internal pressure, which is the internal pressure corrected by the internal pressure information correction unit, with a preset threshold and determines whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection device equipped with the following features.

29. A gas spring life prediction device according to claim 23, 24, or 27, It includes a temperature acquisition unit that acquires the temperature detected by the temperature sensor. The internal pressure information correction unit corrects the internal pressure of the gas spring based on the temperature acquired by the temperature acquisition unit so that it corresponds to an internal pressure equivalent to a predetermined reference temperature, and is a gas spring life prediction device.

30. A gas spring abnormality detection device according to claim 25, 26, or 28, It includes a temperature acquisition unit that acquires the temperature detected by the temperature sensor. The internal pressure information correction unit corrects the internal pressure of the gas spring based on the temperature acquired by the temperature acquisition unit so that it corresponds to an internal pressure equivalent to a predetermined reference temperature, and is a gas spring abnormality determination device.

31. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A gas spring life prediction device for predicting the life of the gas spring of a robot equipped with the following: An estimated internal pressure information acquisition unit acquires an estimated internal pressure, which is the internal pressure of the gas spring estimated from position information, which is information relating to the position of the arm, and current data, which is data relating to the drive current of the drive motor, or the internal pressure of the gas spring estimated from the operation information of the gas spring obtained from the position information. An estimated internal pressure information storage unit stores the time-series data of the estimated internal pressure acquired by the estimated internal pressure information acquisition unit, A gas spring life prediction unit creates a trend line based on the time-series data of the estimated internal pressure stored in the estimated internal pressure information storage unit, and determines the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction device equipped with the following features.

32. A gas spring life prediction device according to claim 31, The estimated internal pressure acquired by the estimated internal pressure information acquisition unit is corrected based on the temperature detected by the temperature sensor so that it corresponds to the internal pressure of the gas spring at a predetermined reference temperature, in a gas spring life prediction device.

33. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A method for predicting the lifespan of a gas spring in a robot equipped with the following features: The internal pressure detected by the pressure detector is acquired, The acquired internal pressure is corrected based on position information, which is information regarding the position of the arm. The time-series data of the corrected internal pressure, which is the corrected internal pressure, is stored. The corrected internal pressure feature time series data, which is the time series data of the feature quantities of the stored time series data of the corrected internal pressure, is obtained. A gas spring lifetime prediction method comprising creating a trend line based on the corrected internal pressure feature time series data, and determining the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing.

34. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A method for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: The internal pressure detected by the pressure detector is acquired, The acquired internal pressure is corrected based on position information, which is information regarding the position of the arm. The time-series data of the corrected internal pressure, which is the corrected internal pressure, is stored. The feature quantities of the time-series data of the stored corrected internal pressure are obtained, A method for determining whether a gas spring is abnormal, comprising comparing the aforementioned feature quantity with a preset threshold and determining whether or not there is an abnormality in the gas spring based on the comparison result.

35. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a control signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A method for predicting the lifespan of a gas spring in a robot equipped with the following features: The internal pressure detected by the pressure detector is acquired, The acquired internal pressure is corrected based on position information, which is information regarding the position of the arm. The time-series data of the corrected internal pressure, which is the corrected internal pressure, is stored. A gas spring life prediction method comprising creating a trend line based on the stored time-series data of the corrected internal pressure, and determining the timing at which the trend line reaches a predetermined life threshold as the predicted life timing.

36. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a control signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A method for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: The internal pressure detected by the pressure detector is acquired, The acquired internal pressure is corrected based on position information, which is information regarding the position of the arm. A method for determining whether a gas spring is abnormal, comprising comparing the corrected internal pressure, which is the corrected internal pressure, with a preset threshold, and determining whether or not there is an abnormality in the gas spring based on the comparison result.

37. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A method for predicting the lifespan of a gas spring in a robot equipped with the following features: The internal pressure of the gas spring is estimated from position information, which is information relating to the position of the arm, and current data, which is data relating to the drive current of the drive motor, or the estimated internal pressure is obtained from the operation information of the gas spring, which is obtained from the position information. The time-series data of the acquired estimated internal pressure is stored, A gas spring life prediction method comprising creating a trend line based on the stored time-series data of the estimated internal pressure, and determining the timing at which the trend line reaches a predetermined life threshold as the predicted life timing.

38. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A gas spring life prediction program for predicting the life of the gas spring of a robot equipped with the following: An internal pressure information acquisition step is to acquire the internal pressure detected by the pressure detector, An internal pressure information correction step is performed to correct the acquired internal pressure based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage step that stores time-series data of the corrected internal pressure, which is the corrected internal pressure, A corrected internal pressure feature data acquisition unit acquires corrected internal pressure feature time series data, which is time series data of the feature quantities of the stored corrected internal pressure time series data. A gas spring lifetime prediction step involves creating a trend line based on the corrected internal pressure feature time series data and determining the timing at which the trend line reaches a predetermined lifetime threshold as the predicted lifetime timing. A gas spring life prediction program that is run on a computer.

39. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A pressure detector for detecting the internal pressure of the gas spring, A gas spring abnormality determination program for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: An internal pressure acquisition step is to acquire the internal pressure detected by the pressure detector, An internal pressure information correction step is performed to correct the acquired internal pressure based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage step that stores time-series data of the corrected internal pressure, which is the corrected internal pressure, A corrected internal pressure feature acquisition step involves acquiring the feature quantities of the time-series data of the stored corrected internal pressure, A gas spring abnormality determination step involves comparing the aforementioned feature quantity with a preset threshold and determining whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection program that is executed by a computer.

40. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a control signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring life prediction program for predicting the life of the gas spring of a robot equipped with the following: An internal pressure information acquisition step is to acquire the internal pressure detected by the pressure detector, An internal pressure information correction step is performed to correct the acquired internal pressure based on position information, which is information regarding the position of the arm. A corrected internal pressure information storage unit stores time-series data of the corrected internal pressure, which is the corrected internal pressure. A gas spring life prediction step involves creating a trend line based on the stored time-series data of the corrected internal pressure, and determining the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction program that is run on a computer.

41. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A control unit that controls the drive motor, Each time a detection instruction signal is generated within the control unit, or each time a control signal is input from the control unit, a pressure detector is provided to detect the internal pressure of the gas spring. A gas spring abnormality determination program for determining whether or not there is an abnormality in the gas spring of a robot equipped with the following: An internal pressure information acquisition step is to acquire the internal pressure detected by the pressure detector, An internal pressure information correction step is performed to correct the acquired internal pressure based on position information, which is information regarding the position of the arm. A gas spring abnormality determination step involves comparing the corrected internal pressure, which is the corrected internal pressure, with a preset threshold and determining whether or not there is an abnormality in the gas spring based on the comparison result. A gas spring abnormality detection program that is executed by a computer.

42. Arm and A drive motor that drives the aforementioned arm, A gas spring that supports the load acting on the arm and reduces the load on the drive motor, A gas spring life prediction program for predicting the life of the gas spring of a robot equipped with the following: An estimated internal pressure information acquisition step, which acquires an estimated internal pressure, which is the internal pressure of the gas spring estimated from position information, which is information relating to the position of the arm, and current data, which is data relating to the drive current of the drive motor, or the internal pressure of the gas spring estimated from the operation information of the gas spring obtained from the position information, An estimated internal pressure information storage step, which stores the time-series data of the acquired estimated internal pressure, A gas spring life prediction step involves creating a trend line based on the stored time-series data of the estimated internal pressure, and determining the timing at which the trend line reaches a predetermined life threshold as the predicted life timing. A gas spring life prediction program that is run on a computer.

Images