Robotic system

Abstract

A robot system includes a robot having one or more joint sections (J1), and a determination section connected to the robot. The joint section (J1) includes a motor (9), a reduction mechanism (10) that reduces the rotation of the motor (9), and a torque sensor (11) that can measure the output torque of the reduction mechanism (10). The reduction mechanism (10) includes a plurality of reduction elements that reduce the rotation of the motor (9) at predetermined reduction ratios. The determination section calculates time-series data of the input torque to the reduction mechanism (10), and determines a reduction element in which an abnormality occurs based on time-series data of the rotation speed of the motor (9), the calculated time-series data of the input torque, time-series data of the output torque measured by the torque sensor (11), and the reduction ratios of the reduction elements.

Description

Technical Field

[0001] This invention relates to robot systems. Background Technology

[0002] A known method involves detecting abnormalities in the motor and reducer based on the torque generated by the motor's drive current (see, for example, Patent Document 1).

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent No. 5927440 Summary of the Invention

[0006] The problem the invention aims to solve

[0007] In diagnosing faults in speed reduction mechanisms, pinpointing the location of the fault is crucial. This is because a fault diagnosis necessitates a more detailed investigation of the speed reduction mechanism, or, if there isn't enough time for such an investigation, replacement of the entire mechanism may be necessary. Since detailed investigations are time-consuming and replacing the entire mechanism is costly, it is desirable to be able to easily identify the primary cause of the internal fault within the speed reduction mechanism.

[0008] Solution for solving the problem

[0009] One aspect of the present invention is a robot system comprising: a robot having one or more joints; and a determination unit connected to the robot, wherein each joint includes: a motor; a reduction mechanism for reducing the rotation of the motor; and a torque sensor capable of measuring the output torque of the reduction mechanism, the reduction mechanism having multiple reduction elements that reduce the rotation of the motor at a predetermined reduction ratio, the determination unit calculating timing data of the input torque to the reduction mechanism, and determining, based on timing data of the motor's rotational speed, the calculated timing data of the input torque, the timing data of the output torque measured by the torque sensor, and the reduction ratio of each reduction element, a reduction element that has malfunctioned. Attached Figure Description

[0010] Figure 1 This is an overall structural diagram of the robot system according to the first embodiment of the present invention.

[0011] Figure 2 It means Figure 1 The diagram shows the first rotary joint of the robot in the robot system and the control device.

[0012] Figure 3 It means Figure 2 A schematic diagram of an example of a deceleration mechanism in the first rotary joint.

[0013] Figure 4 This is an explanation Figure 1 A block diagram of the control device for a robot system.

[0014] Figure 5 This is an explanation Figure 4 The determination section stores a graph of the reduction ratio.

[0015] Figure 6 This is an explanation Figure 1 A flowchart of diagnostic methods for robot systems.

[0016] Figure 7 Is following Figure 6 The flowchart following the flowchart.

[0017] Figure 8 This means that in Figure 6 The diagnostic method uses a time-series diagram of motor speed and torque data.

[0018] Figure 9 This is a block diagram illustrating the control device of the robot system according to the second embodiment of the present invention.

[0019] Figure 10 This is an explanation Figure 9 A flowchart of diagnostic methods for robot systems.

[0020] Figure 11 This means that in Figure 10 The timing diagram of torque timing data used in the diagnostic method.

[0021] Figure 12 This is a flowchart illustrating the diagnostic method for a robot system according to the third embodiment of the present invention.

[0022] Figure 13 This is an explanation Figure 12 A schematic diagram of the robot system.

[0023] Figure 14 yes Figure 12 An example of a deceleration mechanism in a robot system is shown in a diagram illustrating an example of a mechanism component requiring a load and the location of the load.

[0024] Figure 15 This is an explanation Figure 1 A schematic diagram of a variation of the position of the torque sensor in a robot system.

[0025] Figure 16 This is an explanation Figure 1A schematic diagram of another variation of the location of the torque sensor in the robot system. Detailed Implementation

[0026] Hereinafter, the robot system 1 of the first embodiment of the present invention will be described with reference to the accompanying drawings.

[0027] like Figure 1 As shown, the robot system 1 of this embodiment includes: a robot 2; and a control device (determination unit) 3, which is connected to the robot 2.

[0028] Robot 2 is, for example, a vertical six-axis multi-joint robot with six rotary joints (joints) J1, J2, J3, J4, J5, and J6. Robot 2 includes: a base 4, which is disposed on the ground; and a rotating body 5, which is supported on the base 4 in a manner capable of rotating about a vertical first axis A. Additionally, Robot 2 includes: a first arm 6, which is supported on the rotating body 5 in a manner capable of rotating about a horizontal second axis B; and a second arm 7, which is supported on the first arm 6 in a manner capable of rotating about a third axis C parallel to the second axis B. Furthermore, Robot 2 includes a three-axis wrist unit 8 mounted on the front end of the second arm 7.

[0029] The wrist unit 8 includes: a first wrist element 30, which is supported on the second arm 7 in a manner that allows it to rotate about a fourth axis D orthogonal to the third axis C; a second wrist element 31, which is supported on the first wrist element 30 in a manner that allows it to rotate about a fifth axis E orthogonal to the fourth axis D; and a third wrist element 32, which is supported on the second wrist element 31 in a manner that allows it to rotate about a sixth axis F orthogonal to both the fourth axis D and the fifth axis E.

[0030] like Figure 2 As shown, the first rotary joint J1 includes a pair of joint components, which are composed of a base 4 supported in a manner that allows rotation about a first axis A and a rotating body 5. Furthermore, the first rotary joint J1 includes a servo motor (motor) 9 fixed to one joint component, i.e., the base 4; and a reduction mechanism 10 disposed between the pair of joint components. Additionally, the first rotary joint J1 includes a torque sensor 11 disposed between the reduction mechanism 10 and the other joint component, i.e., the rotating body 5, and is capable of detecting the torque applied between them.

[0031] The servo motor 9 has a motor shaft 12 that is driven to rotate, and an encoder 13 that detects the rotation angle of the motor shaft 12.

[0032] The reduction mechanism 10 is a mechanism that reduces the rotation of the motor shaft 12 and transmits it to another joint component. It has multiple reduction elements 14, 15, and 16 connected in series or parallel. The reduction mechanism 10 can have any configuration, for example, Figure 3 As shown, a reduction mechanism with three reduction elements 14, 15, and 16 connected in series will be described.

[0033] The first reduction element (reduction element) 14 includes: a first gear 17 fixed to the motor shaft 12 of the servo motor 9; a second gear 18 meshing with the first gear 17; and a first bearing (bearing) 20 supporting the first shaft 19 to which the second gear 18 is fixed, in a manner that allows it to rotate about its long axis.

[0034] The second reduction element (reduction element) 15 includes: a third gear 21 fixed to the first shaft 19; a fourth gear 22 meshing with the third gear 21; and a second bearing (bearing) 24 that supports the second shaft 23, on which the fourth gear 22 is fixed, in a manner that allows it to rotate about its long axis.

[0035] The third reduction element (reduction element) 16 includes: a fifth gear 25 fixed to the second shaft 23; a sixth gear 26 meshing with the fifth gear 25; and a third bearing (bearing) 28 that supports the third shaft 27, to which the sixth gear 26 is fixed, in a manner that allows it to rotate about its long axis. The third shaft 27 is fixed to the output flange 29 of the reduction mechanism 10, and a torque sensor 11 is fixed to the flange surface 29a of the output flange 29.

[0036] The reduction ratio between the first gear 17 and the second gear 18 is R1, the reduction ratio between the third gear 21 and the fourth gear 22 is R2, and the reduction ratio between the fifth gear 25 and the sixth gear 26 is R3.

[0037] The rotational speed N of the motor shaft 12 of the servo motor 9 is reduced to N / R1 by the first reduction element 14, to N / (R1·R2) by the second reduction element 15, and to N / (R1·R2·R3) by the third reduction element 16. That is, the reduction ratio observed from the servo motor 9 side of the first reduction element 14 is R1, the reduction ratio observed from the servo motor 9 side of the second reduction element 15 is R1·R2, and the reduction ratio observed from the servo motor 9 side of the third reduction element 16 is R1·R2·R3.

[0038] The second rotary joint J2 has the same structure as the first rotary joint J1, except that the pair of joint components are the rotating body 5 and the first arm 6.

[0039] The third rotary joint J3 has the same structure as the first rotary joint J1, except that the pair of joint components are the first arm 6 and the second arm 7.

[0040] The fourth rotary joint J4 has the same structure as the first rotary joint J1, except that the pair of joint components are the second arm 7 and the first wrist element 30.

[0041] The fifth rotary joint J5 has the same structure as the first rotary joint J1, except that the pair of joint components are the first wrist element 30 and the second wrist element 31.

[0042] The sixth rotary joint J6 has the same structure as the first rotary joint J1, except that the pair of joint components are the second wrist element 31 and the third wrist element 32.

[0043] The control device 3 includes at least one processor and at least one memory. For example... Figure 4 As shown, the control device 3 includes: a robot control unit 33, which supplies command current to the servo motors 9 by feeding back the rotation angle values ​​from the encoders 13 of the servo motors 9 of each of the rotary joints J1, J2, J3, J4, J5, and J6; a determination unit 34; and a notification unit 35.

[0044] The following parameters are input to the determination unit 34: the command current value output from the robot control unit 33, the rotation angle value output from the encoder 13 of the servo motor 9, and the output torque, i.e., the measured torque Ts, detected by the torque sensor 11. Figure 5 As shown, the determination unit 34 stores the reduction ratios R1, R2, and R3 of each reduction element 14, 15, and 16. Based on the command current value input from the robot control unit 33, the determination unit 34 calculates the input torque input from the servo motor 9 to the reduction mechanism 10, and calculates the ideal output torque, i.e., the ideal torque Ti, of the reduction mechanism 10 according to the input torque and the reduction ratios R1, R2, and R3.

[0045] Furthermore, the determination unit 34 calculates the differential torque TD1 by subtracting the measured torque Ts from the calculated ideal torque Ti. Moreover, the determination unit 34 performs frequency analysis on the differential torque TD1 using a known method such as FFT to calculate the frequency f of the periodic component contained in the differential torque TD1.

[0046] The determination unit 34 compares the value N / f with the reduction ratios 1, R1, R1·R2, and R1·R2·R3. The value N / f is obtained by dividing the calculated frequency f by the rotational speed N of the motor shaft 12, which is based on the change in the rotational angle value detected by the encoder 13. Then, if the difference between the value N / f and any one of the reduction ratios 1, R1, R1·R2, and R1·R2·R3 is within a predetermined threshold, the determination unit 34 determines that the corresponding reduction element 14, 15, or 16 may be malfunctioning.

[0047] When the value N / f is approximately equal to the reduction ratio of 1, the servo motor 9 or the first gear 17 of the first reduction element 14 fixed to the motor shaft 12 may malfunction.

[0048] When the value N / f is approximately equal to the reduction ratio R1, the second gear 18 or the third gear 21 fixed to the first shaft 19 of the first reduction element 14 may malfunction.

[0049] When the value N / f is approximately the reduction ratio R1·R2, the fourth gear 22 or the fifth gear 25 fixed to the second shaft 23 of the second reduction element 15 may malfunction.

[0050] When the value N / f is approximately the reduction ratio R1·R2·R3, the sixth gear 26 of the third shaft 27 fixed to the third reduction element 16 may have a defect.

[0051] Then, the notification unit 35 notifies the outside of the determination result determined by the determination unit 34. The notification unit 35 can make the notification in any way, such as by displaying on the monitor of the control device 3, by displaying with a light, or by displaying with a buzzer, thereby notifying the outside of the part where there may be a malfunction.

[0052] Next, the method for diagnosing whether the deceleration mechanism 10 is malfunctioning in the robot system 1 of this embodiment will be described.

[0053] The operator operates the control device 3, thereby causing the robot 2 to move. The robot 2 can perform any action, such as causing the six rotary joints J1, J2, J3, J4, J5, and J6 to move simultaneously or in multiples. However, preferably, a diagnostic procedure is executed to cause the six rotary joints J1, J2, J3, J4, J5, and J6 to move one by one. Since the diagnostic methods for each rotary joint J1, J2, J3, J4, J5, and J6 are the same, the diagnostic method for the first rotary joint J1 will be described here as an example.

[0054] like Figure 6As shown, the operator executes a diagnostic procedure to initiate only the constant-speed movement of the first rotary joint J1 of robot 2 (step S1). This causes the servo motor 9 to operate using the command current value output from the robot control unit 33, driving the rotary body 5 to rotate relative to the base 4 around the first axis A. In this state, the rotation angle value detected by the encoder 13 of the servo motor 9 is input to the determination unit 34, where the rotational speed N of the servo motor 9 is obtained based on the time change of the rotation angle value (step S2).

[0055] Additionally, the command current value output from the robot control unit 33 is input to the determination unit 34, where the ideal torque Ti based on the input torque of the servo motor 9 is calculated (step S3). The ideal torque Ti is calculated by multiplying the input torque by the reduction ratio R1·R2·R3 of the reduction mechanism 10 as a whole and the transmission efficiency.

[0056] Simultaneously, the measured torque Ts detected by the torque sensor 11 is input to the determination unit 34 (step S4). The acquired rotational speed N, the calculated ideal torque Ti, and the detected measured torque Ts are stored in correspondence with the rotation angle value, the command current value, and the time when the measured torque Ts is input to the determination unit 34 (step S5).

[0057] Then, it is determined whether the action of robot 2 has ended (step S6). If it has not ended, the process starting from step S2 is repeated. In step S6, if it is determined that the action of robot 2 used for diagnosis has ended, as follows... Figure 8 As shown, timing data of the servo motor 9's rotational speed N, ideal torque Ti, and measured torque Ts are acquired within a predetermined time range. Additionally, as... Figure 7 As shown, the measured torque Ts is obtained by subtracting the time series data of the actual torque Ts from the time series data of the ideal torque Ti. Figure 8 The timing data of the differential torque (first differential torque) TD1 shown (step S7).

[0058] Determine whether a differential torque TD1 exceeding a predetermined threshold Th1 is generated in the acquired timing data of the differential torque TD1 (step S8). If there is a moment when a differential torque TD1 exceeding the threshold Th1 is generated, such as... Figure 7 As shown, the frequency f of the periodic component contained in the time series data of the differential torque TD1 is extracted (step S9).

[0059] Then, in the determination unit 34, the value N / f, obtained by dividing the extracted frequency f by the rotational speed N of the servo motor 9, is calculated (step S10). The calculated value N / f is compared with the reduction ratios 1, R1, R1·R2, and R1·R2·R3 in sequence (steps S11 to S14).

[0060] In step S11, if it is determined that the value N / f is equal to 1 (the difference between the value N / f and 1 is less than a predetermined threshold Th2), the servo motor 9 rotating at speed N and the first gear 17 may be malfunctioning (step S15). If the rotor of the servo motor 9 malfunctions or a tooth of the first gear 17 is missing, the output torque fluctuates at the same frequency as the speed N. If it is determined that a malfunction may exist, the notification unit 35 notifies the outside of this situation (step S19).

[0061] In step S11, if it is determined that the value N / f is different from 1, it is determined whether the value N / f is equal to R1 (the difference between the value N / f and R1 is less than a predetermined threshold Th2) (step S12). In step S12, if it is determined that the value N / f is equal to R1, the second gear 18 and the third gear 21 fixed to the first shaft 19 rotating at a speed of N / R1 may be malfunctioning (step S16). In this case, the notification unit 35 is also used to notify the outside of this situation (step S19).

[0062] Furthermore, in step S12, if it is determined that the value N / f is different from R1, it is determined whether the value N / f is equal to R1·R2 (the difference between the value N / f and R1·R2 is less than a predetermined threshold Th2) (step S13). In step S13, if it is determined that the value N / f is equal to R1·R2, the fourth gear 22 and the fifth gear 25 fixed to the second shaft 23 rotating at a speed of N / (R1·R2) may be malfunctioning (step S17). In this case, the notification unit 35 is also used to notify the outside of this situation (step S19).

[0063] Furthermore, in step S13, if it is determined that the value N / f is different from R1·R2, it is determined whether the value N / f is equal to R1·R2·R3 (the difference between the value N / f and R1·R2·R3 is less than a predetermined threshold Th2) (step S14). In step S14, if it is determined that the value N / f is equal to R1·R2·R3, the sixth gear 26 fixed to the third shaft 27 rotating at a speed of N / (R1·R2·R3) may be malfunctioning (step S18). In this case, the notification unit 35 is also used to notify the outside of this situation (step S19).

[0064] The threshold Th2 in steps S11 to S14 can be the same or different.

[0065] After the diagnosis of the first rotational joint J1 is completed, the diagnoses of the second rotational joint J2 to the sixth rotational joint J6 can be performed sequentially.

[0066] Thus, the robot system 1 according to this embodiment has the following advantages: instead of determining the malfunction of the entire reduction mechanism 10, it is possible to determine the main factors causing the malfunction within the reduction mechanism 10, namely, the malfunction of each reduction element 14, 15, and 16 within the reduction mechanism 10. Therefore, without disassembling the entire reduction mechanism 10 for re-investigation or replacement, it is possible to replace the identified malfunctioning reduction elements 14, 15, and 16, minimizing the number of replacement parts and significantly reducing maintenance time and costs.

[0067] Furthermore, in this embodiment, it is configured to determine whether there is a malfunction in the first to sixth gears 17, 18, 21, 22, 25, and 26 of the servo motor 9 or each reduction element 14, 15, and 16. However, the determination can also be made in the same manner if there is a malfunction in other mechanical components. For example, considering that if there is a malfunction in the first bearing 20 supporting the first shaft 19, the output torque vibrates at the same or similar frequency as the rotational speed N / R1, the malfunction can also be determined in the same manner.

[0068] Furthermore, considering that if there is a defect in the second bearing 24 supporting the second shaft 23, the output torque will vibrate at the same or similar frequency as the rotational speed N / R1·R2, the defect can also be determined in the same way.

[0069] Furthermore, considering that if there is a defect in the third bearing 28 supporting the third shaft 27, the output torque will vibrate at the same or similar frequency as the rotational speed N / R1·R2·R3, the defect can also be determined in the same way.

[0070] Next, the robot system of the second embodiment of the present invention will be described with reference to the accompanying drawings.

[0071] In the description of this embodiment, the parts with the same general structure as the robot system 1 of the first embodiment described above are marked with the same reference numerals, and the description is omitted.

[0072] like Figure 9 As shown, in the robot system of this embodiment, the control device 3 includes a storage unit 36 ​​for storing timing data of the reference torque TR. The reference torque TR is the differential torque between the ideal torque Ti and the measured torque Ts of the robot 2 in its normal state, obtained at the time of manufacture, etc. For example, at the time of manufacture, etc., the timing data of the reference torque TR is obtained by executing the control unit 36 ​​for storing the reference torque TR. Figure 6 The timing data of the differential torque TD1 obtained up to step S7 can be pre-stored in the storage unit 36.

[0073] In the robot system of this embodiment, the determination unit 34 subtracts the timing data of the reference torque TR read from the storage unit 36 ​​from the timing data of the calculated differential torque TD1. Thus, as... Figure 10 As shown, the timing data of the differential value (second differential torque) TD2 is calculated (step S21). The relationship between the differential torque TD1, the reference torque TR, and the differential value TD2 is as follows: Figure 11 As shown.

[0074] Then, the determination unit 34 determines, for example, whether the absolute value of the average value of the difference value TD2 exceeds a predetermined threshold Th3 (step S22). If the absolute value of the average value of the difference value TD2 exceeds the threshold Th3, it determines that there is a problem with any of the bearings 20, 24, 28 or the lubricant inside the reduction mechanism 10, and notifies the notification unit 35 of the situation (step S23).

[0075] The differential torque TD1 is the resistance component inside the reduction mechanism 10 obtained by subtracting the measured torque Ts from the ideal torque Ti. It varies depending on the preload of bearings 20, 24, and 28 and the state of the lubricant.

[0076] The differential value TD2 represents the extent to which the current differential torque TD1 has changed relative to the normal state of robot 2 by subtracting the reference torque TR from the differential torque TD1.

[0077] By subtracting the reference torque TR from the differential torque TD1, even if the differential torque TD1 contains the inherent offsets and vibrations of each rotary joint J1, J2, J3, J4, J5, and J6, these can be removed under normal conditions. This provides the advantages of high-precision extraction of changes from the normal state and high-precision determination of the possibility of adverse conditions.

[0078] When the average value of the differential TD2 is negative, the following adverse conditions may occur: reduction in the preload of bearings 20, 24, and 28 due to wear; softening due to deterioration of the grease used as a lubricant; or, in the case where the reduction mechanism 10 includes a belt, a decrease in belt tension, etc.

[0079] On the other hand, if the average value of the difference TD2 is positive, the following possibilities can be considered. That is, the following adverse situations may occur: wear powder from bearings 20, 24, 28 and other mechanism components caused by the deterioration of the lubricating grease may clog bearings 20, 24, 28; debris from gears 17, 18, 21, 22, 25, 26 and other mechanism components may clog bearings 20, 24, 28; or the lubricating grease may harden due to wear powder.

[0080] According to this embodiment, the following advantages are available: In addition to the malfunctions of gears 17, 18, 21, 22, 25, and 26 in each of the reduction elements 14, 15, and 16 in the first embodiment, it is also possible to notify any malfunction of bearings 20, 24, and 28 or lubricant within the reduction mechanism 10. The following advantages are also available: By adjusting the preload of bearings 20, 24, and 28, the rigidity and rotational accuracy of the rotary joints J1, J2, J3, J4, J5, and J6 can be improved. Furthermore, in the event of suspected lubricant malfunction, replacing the lubricant can repair the malfunction without replacing other mechanical components, significantly reducing maintenance time and costs.

[0081] Furthermore, in this embodiment, a potential defect in bearings 20, 24, 28, or the lubricant is indicated based on whether the absolute value of the average value of the difference value TD2 is greater than a predetermined threshold Th3. Alternatively, the details of the defect can be further categorized and notified using the reference numerals of the average value of the difference value TD2.

[0082] In addition, in this embodiment, the determination unit 34 determines the likelihood of a defect based on the absolute value of the average value of the difference value TD2. Alternatively, the timing data of the difference value TD2 can be extracted based on the timing data of the difference value TD2 calculated in step S21 and the timing data of the stored rotation angle values, for the time periods when the rotary joints J1, J2, J3, J4, J5, and J6 begin to move and the time periods when the rotary joints J1, J2, J3, J4, J5, and J6 move at a constant speed.

[0083] Furthermore, the likelihood of a malfunction can be determined based on the extracted difference value TD2 within the time period. Specifically, the maximum value of the difference value TD2 during the time periods when the rotary joints J1, J2, J3, J4, J5, and J6 begin to operate is a non-periodic component, corresponding to the resistance component generated by the static friction of the mechanism components. It can be seen that if the absolute value of the maximum value of the difference value TD2 during the time periods when the rotary joints J1, J2, J3, J4, J5, and J6 begin to operate exceeds the predetermined threshold Th3, the preload of bearings 20, 24, and 28 is too high.

[0084] Furthermore, the average value of the difference TD2 during the time period when the rotary joints J1, J2, J3, J4, J5, and J6 move at a constant speed is also a non-periodic component, and corresponds to the resistance component generated by the dynamic friction of the mechanism components. It can be seen that when the absolute value of the average value of the difference TD2 during the time period when the rotary joints J1, J2, J3, J4, J5, and J6 move at a constant speed exceeds a predetermined threshold Th3, softening or hardening of the lubricant occurs due to deterioration.

[0085] Alternatively, the timing data of the differential value TD2 during the time period when the rotary joints J1, J2, J3, J4, J5, and J6 move at a constant speed can be used for frequency extraction in step S9. Therefore, compared to using the timing data of the differential torque TD1 over the entire time period during the movement of the rotary joints J1, J2, J3, J4, J5, and J6, the frequency of the periodic component contained in the differential torque TD1 can be determined with higher accuracy.

[0086] Furthermore, in this embodiment, the differential value TD2 is calculated by subtracting the timing data of the reference torque TR from the timing data of the differential torque TD1. Since the reference torque TR is the differential torque TD1, which is measured, for example, by performing a diagnostic procedure under predetermined conditions when the robot 2 is manufactured, it is preferable that the on-site diagnostics are also performed under the same conditions as at the time of manufacture.

[0087] However, for example, it is difficult to ensure that the temperature of the reduction gear 10 during diagnosis is consistent with the temperature of the reduction gear 10 at the time of manufacture. Therefore, it is preferable to provide a temperature sensor capable of detecting the temperature of the reduction gear 10, and the determination unit 34 is provided with a correction unit that corrects the timing data of the reference torque TR based on the temperature detected by the temperature sensor. Alternatively, instead of a temperature sensor that directly detects the temperature of the reduction gear 10, the temperature detected by, for example, the temperature sensor provided with the torque sensor 11 can be used.

[0088] Furthermore, it is impractical for the robot 2 to perform the same diagnostic actions as it would during factory testing, given that the robot is performing these actions on-site. Therefore, the timing data of the reference torque TR and the timing data of the rotation angle values ​​measured when measuring the reference torque TR can be pre-stored in the storage unit 36. Moreover, during diagnostic testing, the reference torque TR can be corrected at each instant using the speed and acceleration of each rotary joint J1, J2, J3, J4, J5, and J6, calculated from the rotation angle values ​​detected by the encoders 13 on each rotary joint J1, J2, J3, J4, J5, and J6, and timing data of the reference torque TR for the diagnostic operation can be generated.

[0089] Furthermore, in this embodiment, by extracting the timing data of the difference value TD2 between the time periods when the rotary joints J1, J2, J3, J4, J5, and J6 begin to move and the time periods when the rotary joints J1, J2, J3, J4, J5, and J6 move at a constant speed, the resistance components generated by static and dynamic friction are used to determine adverse conditions. Alternatively, it can also be used to measure the smoothness of the rotation of the reduction mechanism 10. Since the change in the internal resistance of the reduction mechanism 10 is known, the lubrication state and the state of shaft vibration caused by the wear of gears 17, 18, 21, 22, 25, 26 and bearings 20, 24, 28 can also be estimated. In addition, when the reduction mechanism 10 has a belt, the belt tension can also be estimated.

[0090] Furthermore, in this embodiment, the time series data of the difference TD2 between the starting time of rotary joints J1, J2, J3, J4, J5, and J6 and the time series of rotary joints J1, J2, J3, J4, J5, and J6 operating at a constant speed are extracted from the time series data of the difference value TD2 calculated using the measured torque Ts based on torque sensor 11. Alternatively, other encoders can be installed at the output position of each reduction mechanism 10, and the time series of the starting time of rotary joints J1, J2, J3, J4, J5, and J6 and the time series of the time series of the rotary joints J1, J2, J3, J4, J5, and J6 operating at a constant speed can be calculated.

[0091] Alternatively, a force sensor can be used instead of the torque sensor 11 installed at the output position of the reduction mechanism 10 at each of the rotating joints J1, J2, J3, J4, J5, and J6.

[0092] Next, the robot system of the third embodiment of the present invention will be described with reference to the accompanying drawings.

[0093] In the description of this embodiment, the parts with the same general structure as the robot system 1 in the first embodiment and the second embodiment described above are marked with the same reference numerals, and the description is omitted.

[0094] In the robot system of this embodiment, in addition to performing the same diagnostics as the robot 2 of the first or second embodiment, notification is also given regarding any malfunctions of any mechanical component of each deceleration element 14, 15, 16 of each deceleration mechanism 10.

[0095] exist Figure 12 In the example, a case is illustrated where a diagnosis is performed by the robot system of this embodiment while the robot system 1 of the first embodiment is performing a diagnosis. However, the diagnoses can be performed in parallel or alternately.

[0096] In the robot system of this embodiment, with Figure 9 Similarly, in the second embodiment shown, the control device 3 includes a storage unit 36 ​​connected to the determination unit 34. The storage unit 36 ​​stores the main dimensions of each link component constituting the robot 2 (rotor 5, first arm 6, second arm 7, and wrist unit 8) and the output positions (coordinates of the center position of the output flange 29) of each reduction mechanism 10. Furthermore, the storage unit 36 ​​stores, for each mechanism component within each reduction mechanism 10, a coefficient for calculating the third load based on the second load described below, a threshold for the force and torque that begins to affect the mechanism component, and a time limit for exceeding the threshold.

[0097] The coefficients, thresholds, and time limits are stored as values ​​corresponding to the forces Fx, Fy, and Fz in the three mutually orthogonal axes x, y, and z, and the torques Mx, My, and Mz around each axis. All or more of the forces and torques in these six axes are collectively referred to as the load. The three axes x, y, and z are defined for each reduction mechanism 10.

[0098] like Figure 12 As shown, after calculating the differential torque TD1 (step S7), the determination unit 34 estimates the first load applied to the front end of the wrist unit 8 of the robot 2 (the center position of the flange surface of the wrist flange) (step S31). At this time, the first load is estimated based on the differential torque TD1 of the multiple axis portions of the robot 2.

[0099] Next, the determination unit 34 estimates the second load (step S32) based on the first load, which is the output position (center position of flange surface 29a of output flange 29) of the reduction mechanism 10 applied to each rotary joint J1, J2, J3, J4, J5, J6.

[0100] Next, the determination unit 34 estimates the third load applied to the mechanism components constituting each reduction mechanism 10 based on the second load for each rotary joint J1, J2, J3, J4, J5, J6 (step S33).

[0101] The first to third loads were estimated as time-series data.

[0102] The determination unit 34 determines which mechanism component may be malfunctioning based on the timing data of the third load of each mechanism component and the timing data of the differential torque TD1 calculated in the first or second embodiment (step S34). Then, if any mechanism component is determined to be malfunctioning, the notification unit 35 notifies the unit of this condition (step S35).

[0103] The following describes the structure of robot 2. Figure 13 The structure shown, and the external force X acting on it. Figure 13 Using the case shown by the arrow as an example, the diagnostic method for the robot system in this embodiment will be explained.

[0104] In this case, the first load is estimated based on the measured torque Ts detected by the three torque sensors 11 in the second rotary joint J2, the third rotary joint J3 and the fifth rotary joint J5 (step S31).

[0105] The determination unit 34 uses the rotation angle value from the encoder 13 and the main dimensions of each linkage component (rotor 5, first arm 6, second arm 7 and wrist unit 8) to calculate the front end position of the wrist unit 8, and estimates the first load based on the calculated wrist front end position, the output position of each reduction mechanism 10 and the measured torque Ts.

[0106] The first load is estimated as (Fx, Fy, Fz, Mx, My, Mz) using the forces in the orthogonal three-axis directions and the torques around each axis. The same logic applies to the other forces and torques acting on robot 2.

[0107] Next, the determination unit 34 calculates the coordinate transformation matrix for transforming from the coordinate system at the front end of the wrist unit 8 to the coordinate system of the output position of each deceleration mechanism 10, and uses the calculated coordinate transformation matrix and the first load to calculate the second load applied to the output position of each deceleration mechanism 10 (step S32).

[0108] Then, the determination unit 34 calculates the third load applied to each mechanism component as timing data by multiplying the second load by the coefficient stored in the storage unit 36 ​​(step S33). For example, as Figure 14 As shown by reference numerals P1 to P10 in the attached figures, the timing data of the third load applied to the center positions P1 to P6 of the first gear 17 to the sixth gear 26 and the center positions P7 to P10 of the first bearing 20 to the third bearing 28 are calculated.

[0109] In step S34, the determination unit 34 compares the timing data of the third load applied to each mechanism component with the timing data of the differential torque TD1 calculated in step S7, aligning them with the time axis. Thus, the determination unit 34 can determine that a malfunction may occur in a mechanism component that varies at the same frequency f as the differential torque TD1.

[0110] Furthermore, in step S34, the determination unit (lifetime estimation unit) 34 accumulates the time during which the estimated third load exceeds the threshold stored in the storage unit 36. Based on the accumulated time and the limit time stored in the storage unit 36, the determination unit 34 estimates the remaining lifetime of the mechanism component. Then, the estimated remaining lifetime of the mechanism component is notified by the notification unit 35. This allows for the external notification of mechanism components with a high probability of failure before a malfunction occurs.

[0111] Furthermore, in this embodiment, the timing data of the third load is compared with the timing data of the differential torque TD1 to determine the likelihood of a malfunction. Alternatively, frequency analysis can be performed on the acquired timing data of the third load for all mechanism components using FFT or similar methods to extract the frequency of the periodic component and determine that a malfunction exists in a mechanism component with a frequency consistent with the frequency of the periodic component of the differential torque TD1.

[0112] Furthermore, for mechanical components that indicate a potential malfunction through the third embodiment, notification can also be provided as supplementary information to the malfunction notification in the first or second embodiment. Alternatively, if it is determined that a mechanical component may be malfunctioning, the determination unit 34 can notify the user by causing the mechanical component to operate at a speed that would make the malfunction easily noticeable, or the determination unit 34 can adjust the movement speed of the robot 2.

[0113] Furthermore, in this embodiment, the determination unit 34 is configured to compare the timing data of the differential torque TD1 with the timing data of the third load in each mechanism component to determine the mechanism component where a malfunction has occurred. Alternatively, the determination unit 34 may also include a learning completion model that has learned the timing data of the third load calculated when the constant speed reduction mechanism 10 is in a normal state at the time of manufacture. Moreover, when the robot 2 is performing an action, the calculated timing data of the third load for each mechanism component can be input into the learning completion model, and the deviation rate from the third load in the normal state can be calculated for each mechanism component, thereby using threshold values ​​or the like to determine the mechanism component where a malfunction has occurred.

[0114] Furthermore, in the above embodiments, examples were given of torque sensors 11 being provided at each rotary joint J1, J2, J3, J4, J5, and J6, but... Figure 15 As shown, a six-axis force sensor 37 can also be configured between the base 4 and the surface to be set. Additionally, as... Figure 16 As shown, a force sensor 37 can also be configured at the front end of the wrist unit 8.

[0115] Furthermore, in the above embodiments, a vertical six-axis multi-joint robot with six rotary joints J1, J2, J3, J4, J5, and J6 was described as an example. However, alternatively, a robot system with a horizontal four-axis multi-joint robot, a robot with a different number of rotary joints (7), or any other robot system of any form 2 can also be used. Additionally, instead of rotary joints J1, J2, J3, J4, J5, and J6, linear joints can be provided as joints.

[0116] Explanation of reference numerals in the attached figures:

[0117] 1: Robot System

[0118] 2: Robot

[0119] 3: Control device (judgment unit)

[0120] 9: Servo motor (motor)

[0121] 10: Speed ​​reduction mechanism

[0122] 11: Torque sensor

[0123] 13: Encoder

[0124] 14: First reduction gear (reduction element)

[0125] 15: Second reduction gear (reduction element)

[0126] 16: Third reduction gear (reduction element)

[0127] 17: First gear (gear, mechanism component)

[0128] 18: Second gear (gear, mechanism component)

[0129] 20: First bearing (bearing, mechanism component)

[0130] 21: Third Gear (Gear, Mechanism Component)

[0131] 22: Fourth Gear (Gear, Mechanism Component)

[0132] 24: Second bearing (bearing, mechanism component)

[0133] 25: Fifth Gear (Gear, Mechanism Component)

[0134] 26: Sixth Gear (Gear, Mechanism Component)

[0135] 28: Third bearing (bearing, mechanism component)

[0136] 34: Judgment Department (Life Estimation Department)

[0137] 35: Notification Department

[0138] J1: First rotational joint (joint)

[0139] J2: Second Rotational Joint (Joint)

[0140] J3: Third Rotational Joint (Joint)

[0141] J4: Fourth Rotational Joint (Joint)

[0142] J5: Fifth Rotational Joint (Joint)

[0143] J6: Sixth Rotational Joint (Joint)

[0144] f: Frequency

[0145] N: Rotational speed

[0146] Ti: Ideal torque

[0147] Ts: Measured torque

[0148] TD1: Differential torque (first differential torque)

[0149] TD2: Differential value (second differential torque)

[0150] Th1, Th2, Th3, Th4: Thresholds

[0151] R1, R2, R3: Reduction ratio

Claims

1. A robot system, characterized in that, have: A robot having one or more joints; and The decision-making unit, which is connected to the robot, The joint includes: a motor; a reduction mechanism that reduces the rotation of the motor; and a torque sensor that measures the output torque of the reduction mechanism. The reduction mechanism includes multiple reduction elements, which reduce the rotation of the motor at a predetermined reduction ratio. The determination unit calculates the timing data of the input torque to the reduction mechanism, and determines the reduction element that is malfunctioning based on the timing data of the motor speed, the timing data of the calculated input torque, the timing data of the output torque measured by the torque sensor, and the reduction ratio of each reduction element.

2. The robot system according to claim 1, characterized in that, The determination unit calculates the ideal output torque of the deceleration mechanism, i.e., the ideal torque, based on the input torque and the overall deceleration ratio of the deceleration mechanism. It also calculates the difference between the output torque measured by the torque sensor, i.e., the measured torque, and the ideal torque, i.e., the first differential torque. Based on the first differential torque and the deceleration ratio of each deceleration element, it determines the deceleration element that has malfunctioned.

3. The robot system according to claim 2, characterized in that, The determination unit determines that a malfunction has occurred in the reduction element having a reduction ratio approximately as follows: the value is obtained by dividing the frequency of the periodic component contained in the first differential torque by the rotational speed of the motor.

4. The robot system according to claim 2, characterized in that, The determination unit stores the timing data of the differential torque of the deceleration mechanism when it is normal, i.e., the reference data, and calculates the difference between the timing data of the first differential torque and the reference data, i.e., the second differential torque, and determines the deceleration element that has malfunctioned based on the second differential torque.

5. The robot system according to claim 4, characterized in that, The robot system is equipped with a temperature sensor that detects the temperature of the deceleration mechanism. The determination unit includes a correction unit that corrects the reference data based on the temperature detected by the temperature sensor.

6. The robot system according to any one of claims 1 to 5, characterized in that, The robot system includes a notification unit that notifies the deceleration element of any malfunctions.

7. The robot system according to claim 2 or 3, characterized in that, If the magnitude of the non-periodic component contained in the first differential torque exceeds a predetermined threshold, the determination unit determines that there is a malfunction in the lubricant or in the bearing of any of the reduction elements.

8. The robot system according to claim 3, characterized in that, The robot system includes an encoder that detects the amount of movement in each of the joints. Each of the aforementioned speed reduction elements has one or more mechanical components. The determination unit calculates the timing data of the load applied to each of the mechanism components based on the timing data of the movement amount detected by the encoder and the timing data of the output torque measured by the torque sensor, and determines that any mechanism component in which the frequency of the periodic component in the calculated load is consistent with the frequency of the periodic component in the first differential torque has a malfunction.

9. The robot system according to claim 8, characterized in that, The robot system includes a notification unit that notifies the deceleration element and the mechanism components of any adverse conditions determined by the determination unit.

10. The robot system according to claim 8 or 9, characterized in that, The robot system includes a lifespan estimation unit that, based on time-series data of the loads applied to each of the mechanical components, accumulates the time during which the loads above a predetermined threshold are applied, and estimates the lifespan of each of the mechanical components based on the accumulated time.

11. The robot system according to claim 3, characterized in that, The robot system includes an encoder that detects the amount of movement in each of the joints. Each of the aforementioned speed reduction elements has one or more mechanical components. The determination unit has a learning completion model, which is obtained by learning using time-series data of a reference load. The reference load is the load applied to each component of the mechanism, calculated when the deceleration mechanism is in normal operation. The determination unit is used for, Based on the timing data of the movement detected by the encoder and the timing data of the output torque measured by the torque sensor, the timing data of the load applied to each of the mechanism components is calculated. By inputting the calculated time-series data of the load into the learning-completed model, the mechanism component that has malfunctioned is determined based on the deviation rate from the benchmark load.

Images