Repair device, repair method, and robot system

Abstract

To improve accuracy of calibration by discriminating and identifying between joint deflection and a backlash as an angle error included in a mechanism error parameter.SOLUTION: A correction device comprises: a measurement point generation unit which generates a plurality of candidate points corresponding to the attitude of the robot within a movable range of the robot on the basis of robot mechanism model information expressing a mechanism of the robot and end effector information about an end effector provided in the robot, and decides the candidate point in which a joint torque τ of each joint of the robot in the attitude corresponding to each candidate point becomes equal to or greater than a prescribed threshold in the plurality of generated candidate points as a measurement point; and a parameter identification unit which calculates a positional error between the calculation position and the actually measured position of the end effector in the attitude corresponding to the measurement point and identifies a mechanism error parameter such that the calculated positional error becomes minimum.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a repair device, a repair method, and a robot system. [Background technology]

[0002] When controlling a robot, calibration is required to correct a position error between a target position commanded to the robot and an actual position measured when the robot operates in accordance with the command.

[0003] As a technology related to robot calibration, for example, Patent Document 1 describes the following: setting a first measurement area and a second measurement area within the movable range of the robot; performing calibration in each of the first measurement area and the second measurement area to identify mechanism error parameters; and when operating the robot according to an operation program, correcting the operation program by applying to each teaching point included in the operation program the mechanism error parameters identified by calibration in the measurement area closest to the teaching point. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6568172 Summary of the Invention [Problem to be solved by the invention]

[0005] The causes of position errors between a robot's target position and its actual measured position include the robot's mechanical error, the end effector's machining error and assembly error, the installation error of the jig that fixes the workpiece, and the workpiece's machining error and assembly error. Of these, the robot's mechanical error has the greatest effect. The mechanical error parameters that cause robot mechanical errors include link length error, link assembly error, and angular errors such as joint deflection, angle offset error, and backlash.

[0006] Among the angular errors, joint deflection, angular offset error, and backlash, joint deflection and backlash depend on the direction of joint torque. For this reason, it is difficult to distinguish and identify joint deflection and backlash in conventional techniques such as the technique described in Patent Document 1. The inability to distinguish and identify joint deflection and backlash is one of the causes of reduced calibration accuracy.

[0007] The present invention has been made in consideration of the above points, and has an object to improve the accuracy of calibration by making it possible to distinguish and identify joint deflection and backlash as angle errors included in mechanism error parameters. [Means for solving the problem]

[0008] The present application includes a number of means for solving at least part of the above problems, examples of which are as follows.

[0009] In order to solve the above-described problems, a correction device according to one aspect of the present invention includes: a measurement point generation unit that generates a plurality of candidate points corresponding to postures of the robot within a movable range of the robot based on robot mechanism model information that represents a mechanism of the robot and end effector information related to an end effector provided on the robot, and determines, as measurement points, those of the generated candidate points at which a joint torque τ of each joint of the robot in the posture corresponding to each of the candidate points is equal to or greater than a predetermined threshold; and a parameter identification unit that calculates a position error between a calculated position and an actual measured position of the end effector in the posture corresponding to the measurement point, and identifies mechanism error parameters so as to minimize the calculated position error. [Effects of the Invention]

[0010] According to the present invention, joint deflection and backlash as angle errors included in mechanism error parameters can be discriminated and identified, thereby improving the accuracy of calibration.

[0011] Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of a robot system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram illustrating an example of information stored in the storage unit. [Figure 3] FIG. 3 is a diagram illustrating an example of the operation program correction process. [Figure 4] FIG. 4 is a flowchart illustrating an example of the measurement point generation process. [Figure 5] FIG. 5 is a diagram for explaining the measurement point generation process. [Figure 6] FIG. 6 is a flowchart illustrating an example of the mechanism error parameter identification process. [Figure 7]FIG. 7 is a diagram for explaining a method for estimating the joint deflection coefficient, the angle offset error, and the backlash. [Figure 8] FIG. 8 is a diagram showing a display example of the input / output screen. DETAILED DESCRIPTION OF THE INVENTION

[0013] An embodiment of the present invention will be described below with reference to the drawings. In all drawings used to describe the embodiment, the same components are generally designated by the same reference numerals, and repeated description thereof will be omitted. Furthermore, in the following embodiments, the components (including element steps, etc.) are not necessarily essential unless otherwise specified or considered to be clearly essential in principle. Furthermore, when the terms "consisting of A," "composed of A," "having A," and "including A" are used, other elements are not excluded unless otherwise specified to include only the relevant element. Similarly, in the following embodiments, when referring to the shape, positional relationship, etc. of components, etc., this includes those that are substantially similar or similar to the shape, etc., unless otherwise specified or considered to be clearly essential in principle.

[0014] <Configuration Example of Robot System 10 According to an Embodiment of the Present Invention> FIG. 1 shows an example of the configuration of a robot system 10 according to an embodiment of the present invention.

[0015] The robot system 10 includes a correction device 20, a robot controller 30, a robot 40, and an imaging device 50.

[0016] The correction device 20 identifies the mechanism error parameters of the robot 40. Furthermore, the correction device 20 corrects the robot operation program 223 using the identified mechanism error parameters.

[0017] The robot controller 30 is connected to the correction device 20 via a network 31. The robot controller 30 controls the operation of the robot 40 in accordance with the robot operation program 223 corrected by the correction device 20.

[0018] The imaging device 50 is connected to the repair device 20 via the network 31. The imaging device 50 captures an image of the entire robot 40 including the end effector 42 under the control of the repair device 20.

[0019] The correction device 20 consists of a general computer such as a personal computer equipped with a processor such as a CPU (Central Processing Unit), memory such as SDRAM (Static Random Access Memory), storage such as an HDD (Hard Disc Drive) or SSD (Solid State Drive), a communication module, an input device, and a display device.

[0020] The correction device 20 includes a calculation unit 21 , a storage unit 22 , an input unit 23 , a display unit 24 , and a communication unit 25 .

[0021] The calculation unit 21 is realized by a computer processor. The calculation unit 21 has the following functional blocks: an information acquisition unit 211, a measurement point generation unit 212, a measurement point measurement unit 213, a parameter identification unit 214, a position error estimation unit 215, a display control unit 216, and a robot control unit 217. These functional blocks are realized by the computer processor executing a predetermined program.

[0022] The information acquisition unit 211 acquires robot mechanism model information 221 , end effector information 222 , and robot operation program 223 input by the user using the input unit 23 , and stores them in the storage unit 22 .

[0023] The measurement point generation unit 212 generates a plurality of measurement points that represent the posture within the movable range of the robot 40, which are necessary for calculating the mechanism error parameters of the robot 40, based on the robot mechanism model information 221 and the end effector information 222. The method of generating the measurement points will be described later.

[0024] The measurement point measurement unit 213 measures the actual position of the end effector 42 when the robot 40 is made to assume a posture corresponding to each measurement point (hereinafter referred to as the measured position) based on an image captured by the imaging device 50. Note that the measurement point measurement unit 213 may be omitted and the measured position may be measured manually.

[0025] The parameter identification unit 214 calculates the calculated position (hereinafter referred to as the calculated position) of the end effector 42 when the robot 40 is caused to assume a posture corresponding to each measurement point. The parameter identification unit 214 also calculates the position error between the calculated position corresponding to each measurement point and the actual measured position. Furthermore, the parameter identification unit 214 uses, for example, a nonlinear least squares method to identify mechanism error parameters that minimize the position error corresponding to each measurement point.

[0026] The position error estimation unit 215 analyzes the trajectory information included in the robot operation program 223 to identify the target position of the end effector 42. Furthermore, the position error estimation unit 215 estimates the target position of the end effector 42 and the position error when the end effector 42 is actually operated, based on the mechanism error parameter information 225. Furthermore, the position error estimation unit 215 corrects the trajectory information included in the robot operation program 223 by correcting the target position based on the estimated position error.

[0027] The display control unit 216 displays the input / output screen 500 (FIG. 8) on the display unit 24. The robot control unit 217 reads out the corrected robot operation program 223 from the storage unit 22. The robot control unit 217 also outputs the read out corrected robot operation program 223 to the robot controller 30 via the communication unit 25 and the network 31.

[0028] The storage unit 22 is realized by a computer memory and storage. The storage unit 22 stores robot mechanism model information 221, end effector information 222, and a robot operation program 223 input by the user. The storage unit 22 also stores measurement point information 224 and mechanism error parameter information 225 input from the calculation unit 21.

[0029] The input unit 23 is realized by an input device of a computer, and accepts various input operations from the user.

[0030] The display unit 24 is realized by a display device of a computer. The display unit 24 displays, for example, an input / output screen 500 on which the user can input various information and on which identified mechanism error parameters are displayed.

[0031] The communication unit 25 is realized by a communication module of a computer. The communication unit 25 connects to the robot controller 30 via a network 31 such as the Internet or a mobile phone network, and communicates various types of data.

[0032] The robot 40 is an articulated robot. The robot 40 has an end effector 42 at the tip of an arm 41. The robot 40 performs predetermined operations and tasks under the control of the robot controller 30.

[0033] <Mechanism error parameters> Next, we will explain the mechanical error parameters of the robot 40. The posture of the robot 40 can be expressed by the following equation (1) which indicates the position of each joint.

[0034]

number

[0035] where P k,i+1indicates the position of the (i+1)th joint in a stationary state when the robot 40 is operated so as to assume a posture corresponding to the measurement point k. R is a known rotation matrix acting on the i-th link length L (not shown) connecting the i-th joint and the i+1-th joint. The link length L is a link length (known design value) L * Link length error ΔL i The assembly angle (known design value) α when assembling the i-th link to the i-th joint i The assembly angle α * Assembly error Δα i The joint angle θ of the ith joint corresponding to the measurement point k k,i The joint angle (known command value) θ * joint deflection c i τ k,i , angular offset error Δθ i , and backlash ω i may include:

[0036] Below, link length error ΔL i , assembly error Δα i , joint deflection c i τ k,i , angular offset error Δθ i , and backlash ω i is called the mechanism error parameter. Also, the joint deflection c i τ k,i , angular offset error Δθ i , and backlash ω i are collectively referred to as the angle error.

[0037] <Various information> Next, FIG. 2 shows an example of robot mechanism model information 221, end effector information 222, measurement point information 224, and mechanism error parameter information 225 stored in the storage unit 22.

[0038] The robot 40 is expressed as a Denavit-Hartenberg (DH) model in the robot mechanism model information 221. The robot mechanism model information 221 records an assembly angle α, a joint angle θ, a and d representing the link length L, and a mass in association with each joint axis.

[0039] In the end effector information 222, the mass and center of gravity coordinates of the end effector 42 are recorded.

[0040] In the measurement point information 224, the joint angle θ, joint torque τ, joint deflection cτ, and backlash ω of each of the first to nth joints are recorded in association with the ID of the measurement point.

[0041] The mechanism error parameter information 225 records, in association with each joint, Δa and Δd representing the link length error ΔL, an assembly error Δα, an angle offset error Δθ, a coefficient c of the joint deflection cτ, and a backlash ω.

[0042] <Operation Program Correction Process by Correction Device 20> Next, FIG. 3 is a flowchart illustrating an example of an operation program modification process performed by the modification device 20. In FIG.

[0043] The operation program correction process is started in response to, for example, a user operation on the calculation execution button 505 on the input / output screen 500 (FIG. 8).

[0044] First, the information acquisition unit 211 acquires the robot mechanism model information 221, the end effector information 222, and the robot operation program 223 input by the user using the input unit 23, and stores them in the storage unit 22 (step S1).

[0045] Next, the measurement point generating unit 212 executes a measurement point generating process (step S2).

[0046] 4 is a flowchart illustrating an example of the measurement point generation process. FIG. 5 is a diagram illustrating the measurement point generation process, and shows the joint angles θ1 to θ2 of the joints constituting the robot 40.n It represents a theoretical n-dimensional space with coordinate axes.

[0047] First, as a first constraint, the measurement point generating unit 212 identifies the movable area of ​​the robot 40 based on the robot mechanism model information 221, and randomly generates a plurality of candidate points within the movable area (step S11).

[0048] Next, the measurement point generation unit 212 identifies the joint angle θ of each joint in a state in which the robot 40 is made to assume a posture corresponding to each candidate point. Furthermore, the measurement point generation unit 212 refers to the robot mechanism model information 221 and the end effector information 222, and calculates, for each joint, the joint torque τ when stationary in that state, using the masses of the arm 41 and the end effector 42 on the end effector 42 side from that joint and the distances from that joint to their centers of gravity (step S12).

[0049] Next, as a second constraint, the measurement point generation unit 212 determines, from among the multiple candidate points, candidate points at which the joint torque τ at each joint of the robot 40 is equal to or greater than a threshold value as measurement points. Note that the threshold value may be set to an appropriate value for each joint. Then, the measurement point generation unit 212 records the joint angle θ and the joint torque τ at each joint in the measurement point information 224 of the storage unit 22 in association with an ID that identifies each measurement point (step S13). This completes the measurement point generation process.

[0050] According to the measurement point generation process, it is possible to generate measurement points corresponding to an attitude in which the joint deflection in the angle error is dominant and backlash is negligible.

[0051] Returning to Fig. 3, next, the measurement point measurement unit 213 measures the actual measurement positions corresponding to the generated measurement points (step S3).

[0052] Next, the parameter identifying unit 214 executes a mechanism error parameter identifying process (step S4).

[0053] 6 is a flowchart illustrating an example of the mechanism error parameter identification process. First, the parameter identification unit 214 initializes the mechanism error parameter information 225 stored in the storage unit 22 (step S21). Next, the parameter identification unit 214 calculates the calculated position of the end effector 42 when the robot 40 is caused to assume a posture corresponding to each measurement point, according to the following equation (2) (step S22).

[0054]

number

[0055] As mentioned above, at the measurement point, the joint deflection is dominant in the angular error. Therefore, compared with equation (1), equation (2) is more accurate in determining the joint angle θ of the ith joint corresponding to measurement point k. k,i From the equation, the angle offset error Δθ i , and backlash ω i is omitted.

[0056] Next, the parameter identification unit 214 calculates the position error between the calculated position corresponding to each measurement point and the actual position. In addition, the parameter identification unit 214 uses, for example, a nonlinear least squares method to find the link length error ΔL that minimizes the position error corresponding to each measurement point. i , assembly error Δα i , and identify the angular error (joint deflection cτ) (step S23). However, since the joint deflection is dominant in the angular error at each measurement point, the identified angular error can be regarded as the joint deflection cτ.

[0057] Next, the parameter identification unit 214 estimates the coefficient c of the joint deflection cτ, the angle offset error Δθ, and the backlash ω at each joint (step S24).

[0058] 7 is a diagram for explaining a method for estimating the coefficient c, the angle offset error Δθ, and the backlash ω, in which the X-axis represents the joint torque and the Y-axis represents the angle error.

[0059] First, the parameter identification unit 214 plots the joint torque τ on the X-axis and the angular error (joint deflection cτ) on the Y-axis for each measurement point for each joint, estimates two straight lines representing the angular error (joint deflection cτ) that changes depending on the joint torque τ, and calculates the coefficient c. Next, the parameter identification unit 214 calculates the first Y-intercept of the straight line in the region X>0, Y>0 (angle offset error Δθ+backlash ω). The parameter identification unit 214 also calculates the second Y-intercept (-backlash ω) of the straight line in the region X<0, Y<0. Furthermore, the parameter identification unit 214 subtracts the absolute value of the second Y-intercept from the value of the first Y-intercept to calculate the angle offset error Δθ, and subtracts the angle offset error Δθ from the value of the first Y-intercept to calculate the backlash ω. This completes the mechanism error parameter identification process.

[0060] Returning to Fig. 3, the parameter identifying unit 214 then stores the identified mechanism error parameters as mechanism error parameter information 225 in the storage unit 22. Furthermore, the parameter identifying unit 214 records the joint deflection cτ and the backlash ω, among the identified mechanism error parameters, in the measurement point information 224 in the storage unit 22 (step S5).

[0061] Next, the position error estimation unit 215 corrects the current robot operation program 223 stored in the storage unit 22 based on the mechanism error parameter information 225 (step S6). This completes the operation program correction process by the correction device 20.

[0062] According to the operation program correction process described above, the joint deflection cτ and the backlash ω, which are angular errors included in the mechanism error parameters, can be discriminated and identified, thereby improving the accuracy of the calibration.

[0063] <Display example of input / output screen 500> 8 shows an example of an input / output screen 500 that is displayed on the display unit 24 and presented to the user. The input / output screen 500 has a mechanism error parameter display area 501, a measurement point information display area 502, a robot mechanism model input button 503, an end effector information input button 504, and a calculation execution button 505.

[0064] The mechanism error parameter display area 501 displays the identified mechanism error parameters based on the mechanism error parameter information 225 stored in the storage unit 22. The mechanism error parameter display area 501 also displays a robot mechanism model representing the mechanism of the robot based on the robot mechanism model information 221 stored in the storage unit 22. The measurement point information display area 502 displays, for each joint, the joint angle θ, joint torque τ, joint deflection cτ, and backlash ω corresponding to each measurement point based on the measurement point information 224 stored in the storage unit 22.

[0065] The robot mechanism model input button 503 is an operation button that allows the user to instruct the selection and reading of the robot mechanism model information 221. The end effector information input button 504 is an operation button that allows the user to instruct the selection and reading of the end effector information 222. The calculation execution button 505 is an operation button that allows the user to instruct the start of the mechanism error parameter identification process.

[0066] On the input / output screen 500, the user can select and instruct to read the robot mechanism model information 221 and the end effector information 222. The user can also check the identified mechanism error parameters and the generated measurement point information.

[0067] <Modification> Regarding the generation of measurement points by the measurement point generation unit 212, a third constraint may be added to the above-described first and second constraints. That is, as the third constraint, the measurement point generation unit 212 may analyze the robot operation program 223 and determine, as the measurement point, a candidate point that is close to a teaching point that the robot 40 passes through before completing its operation (the distance is equal to or less than a predetermined threshold). By adding the third constraint, it is possible to identify mechanism error parameters with higher accuracy.

[0068] The present invention is not limited to the above-described embodiments, and various modifications are possible. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with or add to the configuration of another embodiment.

[0069] Furthermore, some or all of the above-described configurations, functions, processing units, processing means, etc. may be implemented in hardware, for example, by designing them as integrated circuits. Furthermore, the above-described configurations, functions, etc. may be implemented in software by a processor interpreting and executing a program that implements each function. Information such as programs, tables, and files that implement each function may be stored in memory, a storage device such as a hard disk or SSD, or a storage medium such as an IC card, SD card, or DVD. Furthermore, the control lines and information lines shown are those considered necessary for explanation, and do not necessarily represent all control lines and information lines in the product. In reality, it can be assumed that almost all components are interconnected. [Explanation of symbols]

[0070] 10...Robot system, 20...Correction device, 21...Calculation unit, 211...Information acquisition unit, 212...Measurement point generation unit, 213...Measurement point actual measurement unit, 214...Parameter identification unit, 215...Position error estimation unit, 216...Display control unit, 217...Robot control unit, 22...Memory unit, 221...Robot mechanism model information, 222...End effector information, 223...Robot operation program, 224...Measurement point information, 225...Mechanism error parameter information, 23...Input unit, 24...Display unit, 25...Communication unit, 30...Robot controller, 31...Network, 40...Robot, 41...Arm, 42...End effector, 50...Imaging device, 500...Input / output screen

Claims

1. A measurement point generation unit that generates a plurality of candidate points corresponding to the posture of the robot within the movable range of the robot based on robot mechanism model information representing the mechanism of the robot and end effector information regarding the end effector provided on the robot, and determines, from among the plurality of generated candidate points, the candidate points at which the joint torque τ of each joint of the robot in the posture corresponding to each of the candidate points is equal to or greater than a predetermined threshold, as measurement points; a parameter identification unit that calculates a position error between a calculated position and an actually measured position of the end effector in the posture corresponding to the measurement point, and identifies a mechanism error parameter so as to minimize the calculated position error; The parameter identification unit Identifying the link length error ΔL, the assembly error Δα, and the angle error as the mechanism error parameters so as to minimize the calculated position error; Based on the identified angle error and joint torque τ at each joint, a coefficient c of joint deflection cτ, an angle offset error Δθ, and a backlash ω are estimated. Correction device.

2. A measurement point generation unit that generates a plurality of candidate points corresponding to the posture of the robot within the movable range of the robot based on robot mechanism model information representing the mechanism of the robot and end effector information regarding the end effector provided on the robot, and determines, from among the plurality of generated candidate points, the candidate points at which the joint torque τ of each joint of the robot in the posture corresponding to each of the candidate points is equal to or greater than a predetermined threshold, as measurement points; a parameter identification unit that calculates a position error between a calculated position and an actually measured position of the end effector in the posture corresponding to the measurement point, and identifies a mechanism error parameter so as to minimize the calculated position error; The measurement point generation unit analyzes a robot operation program for controlling the operation of the robot, and determines the candidate point within a predetermined distance from a teaching point corresponding to the posture of the robot until the operation corresponding to the robot operation program is completed as the measurement point. Correction device.

3. 3. The correction device according to claim 1 or 2, a measurement point measurement unit for measuring the actual measurement position, Correction device.

4. 3. The correction device according to claim 1 or 2, a position error estimating unit that estimates a position error of the end effector when the robot is operated in accordance with a robot operation program for controlling the operation of the robot, based on the mechanism error parameters. Correction device.

5. 5. The correction device of claim 4, The position error estimating unit corrects the robot operation program based on the estimated position error. A correction device comprising:

6. A repair method using a repair device, a measurement point generating step of generating a plurality of candidate points corresponding to the posture of the robot within a movable range of the robot based on robot mechanism model information representing a mechanism of the robot and end effector information related to an end effector provided on the robot, and determining, as measurement points, candidate points among the generated plurality of candidate points at which the joint torque τ of each joint of the robot in the posture corresponding to each of the candidate points is equal to or greater than a predetermined threshold value; a parameter identification step of calculating a position error between a calculated position and an actually measured position of the end effector in the posture corresponding to the measurement point, and identifying a mechanism error parameter so as to minimize the calculated position error; The parameter identification step includes: Identifying the link length error ΔL, the assembly error Δα, and the angle error as the mechanism error parameters so as to minimize the calculated position error; Based on the identified angle error and joint torque τ at each joint, a coefficient c of joint deflection cτ, an angle offset error Δθ, and a backlash ω are estimated. How to fix it.

7. A correction method using a correction device, comprising: a measurement point generating step of generating a plurality of candidate points corresponding to the posture of the robot within a movable range of the robot based on robot mechanism model information representing a mechanism of the robot and end effector information related to an end effector provided on the robot, and determining, as measurement points, candidate points among the generated plurality of candidate points at which the joint torque τ of each joint of the robot in the posture corresponding to each of the candidate points is equal to or greater than a predetermined threshold value; a parameter identification step of calculating a position error between a calculated position and an actually measured position of the end effector in the posture corresponding to the measurement point, and identifying a mechanism error parameter so as to minimize the calculated position error; The measurement point generation step analyzes a robot operation program for controlling the operation of the robot, and determines the candidate point within a predetermined distance from a teaching point corresponding to the posture of the robot until the operation corresponding to the robot operation program is completed as the measurement point. How to fix it.

8. Robots and a correction device that corrects a robot operation program for controlling the operation of the robot, The correction device is A plurality of candidate points corresponding to the posture of the robot are generated within a movable range of the robot based on robot mechanism model information representing the mechanism of the robot and end effector information relating to an end effector provided to the robot, and among the plurality of candidate points generated, a measurement point generation unit that determines, as measurement points, the candidate points at which the joint torque τ of each joint of the robot in the posture corresponding to each of the candidate points is equal to or greater than a predetermined threshold; a parameter identification unit that calculates a position error between a calculated position and an actually measured position of the end effector in the posture corresponding to the measurement point, and identifies a mechanism error parameter so as to minimize the calculated position error; a position error estimating unit that estimates a position error of the end effector when the robot is operated in accordance with the robot operation program based on the mechanism error parameters, and corrects the robot operation program based on the estimated position error; a robot control unit that controls the robot based on the modified robot operation program, The parameter identification unit Identifying the link length error ΔL, the assembly error Δα, and the angle error as the mechanism error parameters so as to minimize the calculated position error; Based on the identified angle error and joint torque τ at each joint, a coefficient c of joint deflection cτ, an angle offset error Δθ, and a backlash ω are estimated. Robot system.

9. A robot, a correction device that corrects a robot operation program for controlling the operation of the robot, The correction device is A plurality of candidate points corresponding to the posture of the robot are generated within a movable range of the robot based on robot mechanism model information representing the mechanism of the robot and end effector information relating to an end effector provided to the robot, and among the plurality of candidate points generated, a measurement point generation unit that determines, as measurement points, the candidate points at which the joint torque τ of each joint of the robot in the posture corresponding to each of the candidate points is equal to or greater than a predetermined threshold; a parameter identification unit that calculates a position error between a calculated position and an actually measured position of the end effector in the posture corresponding to the measurement point, and identifies a mechanism error parameter so as to minimize the calculated position error; a position error estimating unit that estimates a position error of the end effector when the robot is operated in accordance with the robot operation program based on the mechanism error parameters, and corrects the robot operation program based on the estimated position error; a robot control unit that controls the robot based on the modified robot operation program, The measurement point generation unit analyzes a robot operation program for controlling the operation of the robot, and determines the candidate point within a predetermined distance from a teaching point corresponding to the posture of the robot until the operation corresponding to the robot operation program is completed as the measurement point. Robot system.

Images