Abnormality detection device and abnormality detection method for joints of robot
By using an abnormality detection device in a delta transverse dual swing arm robot, simulating the action track and measuring the motor parameters, the problem of gap detection of spherical joints is solved, and the robot's movement stability and production efficiency are improved.
Patent Information
- Application Number
- CN201910111734.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2018-02-14
- Filing Date
- 2019-02-12
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2039-02-12
AI Technical Summary
The prior art is difficult to effectively detect abnormal gaps in the spherical joints of the delta lateral double swing arm robot, especially when the gap between the joint ball and the shell becomes larger, resulting in deterioration of the positioning accuracy of the robot and increasing vibration, which in turn affects production efficiency and process. normal progress.
The abnormality detection device is adopted, which cooperates with the passive connector through the motor-driven drive drive, simulates the robot to operate along a number of predetermined action tracks, measures the driving torque or current value of the motor, and calculates the change of the index value. To determine whether there is an abnormal gap between the joint elements.
It can accurately determine the abnormal gap of the spherical joint, avoid the problems of reduced positioning accuracy and increased vibration caused by the gap, and improve the stability and reliability of the robot during high-speed operation or collision.
Smart Images

Figure CN110154089B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a device and a method for detecting abnormal clearance of a joint of a robot. Background Art
[0002] As an example of a robot having a connection mechanism, a lateral double-swing arm robot having a delta-type lateral double-swing arm mechanism is known, and the delta-type lateral double-swing arm mechanism has a function of three-dimensionally positioning a movable part having an end actuator. The delta-type lateral double-swing arm robot has a base, a movable part, and a driving connector and a passive connector that connect the base and the movable part. In most cases, three pairs of driving connectors and passive connectors are provided, and by controlling the actions of each pair separately, the movable part can be moved in three degrees of freedom (X, Y, Z).
[0003] Generally, a 3-DOF ball joint is used to connect the passive connector and the driving connector, and the passive connector and the movable part. For example, as a structure of a ball joint, the following structure is known, that is, the joint ball and the housing are separated in a geometric manner, and the joint ball is pulled toward the housing by an elastic body such as a spring (for example, refer to Japanese Patent Publication No. 2002-529258). In addition, a lateral double-swing arm robot with a connecting ball structure that integrates the joint ball and the housing so that the housing and the joint ball are not easily separated is known (for example, refer to Japanese Patent Publication No. 2014-046406).
[0004] On the other hand, a technique for detecting abnormality in the movement of a robot as early as possible and suppressing the decrease in the operation rate of the robot is also known. For example, the following abnormality judgment method is known: by using a low-pass filter with the inverse of the position loop gain as a time constant, the estimated current position is calculated based on the command position, the absolute value of the difference between the actual position detected by the encoder installed on the servo motor as the actual current position and the estimated current position is calculated, the total observed feedback amount is calculated based on the torque command value for driving the servo motor, the differential value of the actual position, and the known disturbance torque, the abnormality detection threshold is calculated based on the estimated speed obtained by differentiating the total observed feedback amount and the estimated current position, and when the absolute value of the difference between the actual position and the estimated current position is greater than the abnormality detection threshold, it is judged to be an abnormal state (for example, refer to Japanese Patent Publication No. 2005-186183).
[0005] In addition, for the joints of the robot body, the following abnormality detection method is known: the natural vibration frequency of the robot arm is measured, and the constant speed movement of the robot arm at the maximum resonance speed due to the vibration generated by the drive motor and the reducer is used as the action condition for abnormality detection of the robot body; when the torque variation value calculated based on the motor torque value obtained by temperature correction using a temperature sensor set on the drive motor or the reducer exceeds a preset threshold value, it is determined that an abnormality has occurred in the robot body (for example, refer to Japanese Patent Gazette No. 2006-281421).
[0006] On the other hand, a lateral double-swing arm robot is known, which is provided with a sensor for detecting the final output of the lateral double-swing arm robot, namely the inclination of the end plate, and based on the output value of the sensor, it is detected that the connection between the connecting parts is released at at least one of the multiple connection positions of the connecting parts of the spherical joint (for example, refer to Japanese Patent Gazette No. 2017-056507).
[0007] Furthermore, a detection device is known that forms an internal passage opening on the surface of the ball head of a spherical joint, and determines whether the connection of the spherical joint is disengaged based on the detection value of the pressure in the internal passage (for example, refer to Japanese Patent Publication No. 2017-013160).
[0008] In a structure in which the joint ball and the housing are separated geometrically, in the event of an unexpectedly high-speed movement or a collision, the restraining force at the joint portion of the passive link for pulling the joint ball toward the housing may be insufficient, and the joint portion may disintegrate.
[0009] On the other hand, in the connecting ball structure in which the joint ball and the shell are integrated, it can be seen that the shell and the joint ball are not easily separated due to mechanical connection even if a collision occurs. However, in the case of using a connecting ball structure, when the joint ball or the shell is worn due to use, a gap is generated between the joint ball and the shell, which may cause a deterioration in the positioning accuracy of the movable part of the robot, an increase in vibration, etc. Sometimes, due to the deterioration of positioning accuracy and the increase in vibration, it is difficult for the robot to perform processing work, assembly work, etc. normally, and even major problems such as reduced production efficiency and stopped production processes may occur. Therefore, in the case where there is an abnormality in the gap of the connecting ball part, it is hoped to know the abnormal state as soon as possible.
[0010] It can be considered that the existing method of using motor torque to determine abnormality is not suitable for detecting abnormality in the ball joint of a delta-type lateral double-swing arm robot. The reasons for this are that the ball joint of a delta-type lateral double-swing arm robot is usually a joint that moves passively following the robot's posture, is arranged at a position relatively far from the drive motor, and it is difficult to determine which joint among multiple joints has an abnormality.
[0011] Furthermore, in the conventional method of detecting the connection disengagement of a ball joint, it is difficult to detect that the gap between the joint ball and the housing (socket) becomes larger in a ball joint having a structure in which the connection is easily disengaged. Summary of the invention
[0012] One embodiment of the present invention is an abnormality detection device that detects abnormal gaps between dual elements of a pair connected to a passive connector in a robot, the robot comprising a driving connector driven by a motor, a plurality of passive connectors driven by the movement of the driving connector, and a plurality of pairs respectively connected to the plurality of passive connectors, wherein the abnormality detection device comprises: a score calculation unit that simulates the movement of the robot along a plurality of predetermined motion trajectories, and calculates scores for the plurality of motion trajectories, the scores being calculated when an abnormal gap occurs in a pair of objects that should be determined from among the plurality of pairs. The present invention relates to a method for generating a robot motion for moving the robot along a motion track having the score being greater than or equal to a predetermined threshold value among the plurality of motion tracks; a measuring unit for measuring a driving torque or a current value of the motor when the robot moves in accordance with the robot motion; an index value calculating unit for calculating an index value based on a magnitude of a change in a value related to the driving torque or the current value measured by the measuring unit; and a determining unit for determining whether or not there is an abnormal gap between the dual elements of the object pair based on the magnitude of the index value.
[0013] Another embodiment of the present invention is an abnormality detection device that detects an abnormal gap between dual elements of a pair connected to a passive connector in a robot, the robot comprising a driving connector driven by a motor, a plurality of passive connectors driven by the movement of the driving connector, and a plurality of pairs respectively connected to the plurality of passive connectors, wherein the abnormality detection device comprises: a simulation execution unit that simulates the movement of the robot along a plurality of predetermined movement trajectories, assuming that there is a gap between the dual elements of the target pair specified from the plurality of pairs for which the presence or absence of the abnormal gap should be determined. , selecting an action trajectory in which a collision between the dual elements caused by the gap occurs only in the object pair from the plurality of action trajectories; an action generating unit, which generates a robot action for moving the robot along the action trajectory selected by the simulation executing unit; a measuring unit, which measures the driving torque or current value of the motor when the robot moves according to the robot action; an index value calculating unit, which calculates an index value based on the magnitude of a change in a value related to the driving torque or current value measured by the measuring unit; and a determining unit, which determines whether there is an abnormal gap between the dual elements of the object pair according to the magnitude of the index value.
[0014] Another embodiment of the present invention is an abnormality detection method for detecting an abnormal gap between dual elements of a pair connected to a passive connector in a robot, the robot comprising a driving connector driven by a motor, a plurality of passive connectors driven by the movement of the driving connector, and a plurality of pairs connected to the plurality of passive connectors, wherein the abnormality detection method comprises: performing a simulation in which the robot is caused to move along a plurality of predetermined motion trajectories, and calculating scores for the plurality of motion trajectories, the scores becoming high when a collision between dual elements occurs in a target pair specified from the plurality of pairs for which the presence or absence of an abnormal gap is to be determined, and becoming low when a collision between dual elements occurs in a pair other than the target pair; generating a robot motion for causing the robot to move along a motion trajectory in which the score is greater than a predetermined threshold value among the plurality of motion trajectories; measuring a driving torque or a current value of the motor when the robot moves according to the robot motion; calculating an index value based on a magnitude of a change in a value related to the measured driving torque or current value; and determining whether or not there is an abnormal gap between the dual elements of the target pair based on the magnitude of the index value.
[0015] Still another aspect of the present invention is an abnormality detection method for detecting an abnormal gap between dual elements of a pair connected to a passive connector in a robot, the robot comprising a driving connector driven by a motor, a plurality of passive connectors driven by the movement of the driving connector, and a plurality of pairs connected to the plurality of passive connectors, wherein the abnormality detection method comprises: performing a simulation for causing the robot to move along a plurality of predetermined motion trajectories, respectively, and selecting, from the plurality of motion trajectories, a motion trajectory in which a collision between the dual elements due to the gap occurs only in the target pair, assuming that a gap exists between the dual elements of a target pair specified from the plurality of pairs for which the presence or absence of the abnormal gap is to be determined; generating a robot motion for causing the robot to move along the motion trajectory selected by the simulation; measuring a driving torque or a current value of the motor when the robot moves according to the robot motion; calculating an index value based on a magnitude of a change in a value related to the measured driving torque or current value; and determining, based on the magnitude of the index value, whether or not there is an abnormal gap between the dual elements of the target pair. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The objects, features, and advantages of the present invention will become more apparent from the following description of the embodiments in conjunction with the accompanying drawings.
[0017] Figure 1 The diagram shows an abnormality detection device according to a suitable embodiment and a delta-type lateral double-swing arm robot as an example of an application target of the device.
[0018] Figure 2 Yes means Figure 1 A partially enlarged view of the structure of each spherical joint of the lateral double-swing arm robot.
[0019] Figure 3 It is a diagram showing a structural model of a lateral double-swing arm robot.
[0020] Figure 4 This is a graph showing an example of the time change of the driving torque of the motor.
[0021] Figure 5 This is a diagram showing an example of a time change of a differential value of a drive torque of a motor.
[0022] Figure 6 is a flowchart illustrating an example of an abnormality detection method according to a suitable embodiment.
[0023] Figure 7 a~ Figure 7 c is a diagram showing a model of a spherical joint, Figure 7a represents the case where there is no gap between the dual elements, Figure 7 b represents the case where there is a gap between the dual elements and sliding motion occurs between the dual elements. Figure 7 c represents the case where there is a gap between the dual elements and separation / collision motion occurs between the dual elements.
[0024] Figure 8 This is a diagram showing an example of parameters used for score calculation in a tabular format.
[0025] Figure 9a and Figure 9b is a diagram showing an example of the initial conditions of the robot's motion trajectory. Figure 9a Indicates 4 tracks starting from the predetermined position. Figure 9b Indicates the status of each track being rated.
[0026] Fig.10 It is a graph comparing the detection results of abnormal gaps of specific pairs in a suitable embodiment with the actual measured values.
[0027] Fig.11 It is a diagram schematically showing another configuration example to which the present embodiment can be applied.
[0028] Fig.12 This is a diagram schematically showing another structural example to which the present embodiment can be applied. DETAILED DESCRIPTION
[0029] Figure 1 This is a diagram showing the schematic structure of an abnormality detection device involved in a suitable embodiment of the present disclosure, an example of a structure to which the abnormality detection device can be applied, namely, a delta-type lateral double-swing arm robot. The lateral double-swing arm robot (hereinafter also referred to as the robot) 10 includes a base portion 12, a movable portion 14 arranged separately from the base portion 12 (usually below the base portion 12), two or more (three in the illustrated example) connecting portions 16a to 16c that connect the base portion 12 and the movable portion 14 and each have one degree of freedom relative to the base portion 12, and multiple (usually the same number as the connecting portions, three in the illustrated example) servo motors and other motors 18a to 18c that drive the connecting portions 16a to 16c, respectively, and an end effector 19 (in the illustrated example) such as a robot hand can be mounted on the movable portion 14. Figure 3 (see Figure 2).
[0030] The connecting portion 16a is composed of a driving connecting member 20a connected to the base portion 12, and a pair (2) of passive connecting members 22a that connect the driving connecting member 20a and the movable portion 14 and extend parallel to each other. The driving connecting member 20a and the passive connecting member 22a are connected by a pair (2) of first joints 24a. In addition, the movable portion 14 and the passive connecting member 22a are connected by a pair (2) of second joints 26a. In addition, in the present embodiment, the first joint and the second joint (dual) are both configured as spherical joints (spherical bearings).
[0031] Figure 2 1 is a partially enlarged view showing the structure (connector spherical structure) of each spherical joint (here, the spherical joint 24a or 26a) of the robot 10. The spherical joint 24a includes a joint ball (convex portion) 28, a housing (concave portion) 30 for accommodating the joint ball 28, and a gasket 32 disposed between the joint ball 28 and the housing 30. Figure 1 As shown, the robot 10 has a restriction plate 34a connected to the housing of the first spherical joint 24a on the driving link side (upper side) of the passive link to restrict the rotation of the two parallel passive links 22a around the respective axes.
[0032] The other connecting parts 16b and 16c can also have the same structure as the connecting part 16a, so the corresponding structural elements are added with only the reference symbols at the end changed (for example, the elements corresponding to the passive connecting part 22a are added with reference symbols 22b or 22c), and detailed descriptions are omitted.
[0033] like Figure 1 As schematically shown in the figure, the double-swing arm robot 10 is connected to a control device 36 that controls the motion of the robot 10 . In addition, the abnormality detection device 38 for detecting abnormal clearance of a spherical joint comprises: a score calculation unit 40, which simulates causing the robot 10 to move along a plurality of predetermined motion trajectories, respectively, and calculates scores (described later) for each of the plurality of motion trajectories, the score becoming high when a collision between dual elements (joint elements) occurs in a target dual (target joint) specified from a plurality of duals (joints) for determining whether or not there is an abnormal clearance, and becoming low when a collision between dual elements occurs in a dual other than the target dual; an action generation unit 42, which generates a robot action for causing the robot 10 to move along an action trajectory having a score greater than a predetermined first threshold value among the plurality of motion trajectories; a measuring unit 44, which measures a driving torque or current value of a motor when the robot 10 moves according to the generated robot action; an index value calculation unit 46, which calculates an index value (described later) based on the magnitude of a change in the driving torque or current value measured by the measuring unit 44; and a determination unit 48, which determines whether or not there is an abnormal clearance between the dual elements of the target dual based on the magnitude of the calculated index value.
[0034] The robot control device 36 is configured to generate an action command for causing the robot 10 to move based on the robot action generated by the action generation unit 42, and to control each axis (motor) of the robot 10 based on the action command. In addition, the abnormality detection device 38 may also include a storage unit 50 such as a memory for storing the above-mentioned multiple action trajectories, the calculated scores and index values, and data related to the first threshold value and the second threshold value, and a display unit 52 such as a display for presenting the above-mentioned simulation and determination results to the operator.
[0035] The abnormality detection device 38 can be realized as a computing device such as a personal computer (PC) connected to the robot control device 36 and equipped with a processor and a memory. Figure 1 In the figure, the abnormality detection device 38 is shown as a device independent of the robot control device 36, but it can also be combined into the control device 36 in the form of a processor and a memory. Furthermore, a device such as a PC can be responsible for part of the abnormality detection function, and the robot control device 36 can be responsible for other functions.
[0036] Figure 3 Yes means Figure 1 FIG. 1 is a diagram showing a structural model of a lateral double swing arm robot 10. The lateral double swing arm robot 10 has a closed-loop connection structure including three rotation drive units (motors) and 12 passive pairs (here, spherical joints). In addition, the 12 spherical joints are represented as S i,j,k , one of the 12 spherical joints is designated in advance (by the operator, etc.) for which an abnormal gap (described later) should be detected.
[0037] Refer again Figure 2 As the robot 10 moves, the joint ball 28 slides relative to the pad 32 in each joint, but in order to minimize the friction resistance at this time, in many cases, the pad 32 is made of a low-friction material such as resin. However, due to the repeated movement of the robot, the pad 32 wears, so a gap (air gap) is generated between the joint ball 28 and the pad 32 (housing 30). If the gap is larger than a certain value, the positioning accuracy of the robot 10 may be deteriorated, and the vibration accompanying the movement of the robot may increase. Therefore, in this embodiment, it is detected whether the gap is larger than a certain value (whether an abnormal gap has occurred).
[0038] Figure 4This is a graph showing an example of the time variation of the measured value of the driving torque of the motor driving the connecting member in a spherical joint (ball pair). Here, graph 54 shows the case where the gap between the dual elements (here, the joint ball and the housing) can be ignored (normal), and graph 56 shows the case where there is a gap of a certain value or more (abnormal).
[0039] In the case where there is a gap in a certain spherical joint (ball pair), if the lateral double-swing arm robot (representative points such as the end effector) is moved along a predetermined motion trajectory, in the ball pair with the gap, either a sliding motion between the dual elements (here, the joint ball and the housing) or a collision motion in which the dual elements separate and collide will occur. In the latter case, an impact force is generated, which is transmitted to the actuator (motor) of the drive connection via the intermediate connection and other bearings (joint, pair). The inventors of the present application have discovered that Figure 4 As shown in Figure 56, the driving torque of the actuator (motor) to which the impact force is transmitted changes greatly when there is a pair with an abnormal gap (above a certain value) compared to the case where a normal pair is used, and the abnormality can be determined by evaluating the size of the change.
[0040] Specifically, first, in order to quantify the magnitude of the variation in the driving torque value caused by the collision between the coupled elements, an index value (hereinafter also referred to as abnormality) AS represented by the following formula (1) is introduced.
[0041]
[0042] In formula (1), τ i represents the driving torque obtained from the monitoring current of the i-th actuator, RMS represents the second-order mean square root of the value in the time interval (t0, t1), and c represents a vector consisting of the sizes of the gaps of all spherical pairs. In addition, "c = 0" represents the case where the gaps of all pairs are ideal (zero). Here, as the time interval (t0, t1), for example, Figure 4 The time before and after the acceleration starts as shown in the formula (1) indicates that the greater the change (variation) in the driving torque value when the paired elements collide with respect to the driving torque value measured when there is no gap in all the paired elements, the greater the abnormality.
[0043] Here, it is preferable to set the time interval (t0, t1) to include the change of the driving torque caused by the initial collision between the coupled elements (illustrated by reference symbol 58). For example, the time t0 is set before the initial collision, and the time t1 is set to an appropriate value that includes the initial change, such as 0.1 seconds after the time t0. In addition, the time interval can be determined based on experience, but it can also be calculated (predicted) based on the size of the gap and the acceleration when the robot starts to move.
[0044] In addition, the driving torque (τ i ), but the time differential value of the driving torque can also be used instead. Figure 5 The relationship between the time and the time differential value of the driving torque in this case is shown in . Figure 5 The graph shows the difference between normal and abnormal conditions. In any case, the change in the driving torque can be detected, so a suitable index value for determining whether there is an abnormal gap can be obtained. However, if the time differential value of the driving torque is used, there is a tendency to easily detect collisions between the dual elements.
[0045] In addition, instead of using the value related to the driving torque (here, the driving torque or its time differential value), a value related to the current value of the motor (for example, the current value or its time differential value) may be used. Generally, since the driving torque is proportional to the current value, when using a value related to the current value, it is also possible to Figure 4 or Figure 5 The same applies.
[0046] Below, refer to Figure 6 The flowchart of FIG. 1 illustrates an example of the processing (abnormality detection method) of the abnormality detection device involved in this embodiment. First, in step S1, the operator or the like selects or specifies (preferably one) of the 12 pairs (spherical joints) for which it is desired to determine whether an abnormal gap has occurred as the target pair. Here, the specified spherical joint (spherical pair) is represented as S i’,j’,k’ .
[0047] Next, in steps S2 to S10, the robot motion trajectory suitable for abnormality detection is selected, and the robot motion corresponding to the motion trajectory is generated. In this embodiment, in order to determine the pair with an abnormal gap, under the assumption that there is a gap between the dual elements of the target pair (here, the joint ball and the housing), a trajectory (the motion trajectory of the representative point of the robot) that produces collision or proximity of the dual elements only in the target pair is obtained through simulation. In other words, a robot motion is generated that performs the sliding motion described later even if there is a gap in all or more than half of the pairs other than the target pair. Such a robot motion can be obtained through the following steps.
[0048] In step S2, the score calculation unit 40 or the like randomly (arbitrarily) selects a motion trajectory for scoring described later from a plurality of motion trajectories that are predetermined (stored in the storage unit 50 or the like). Here, for example, the position (coordinate) x and the velocity x of the end effector of the robot relative to the stationary coordinate system are used. And acceleration The initial conditions of any trajectory are defined as shown in the following formula (2). In formula (2), W represents the robot's working area, R 3 represents a real three-dimensional space. However, there is an upper limit to the acceleration of the robot, so in the following explanation, a max Indicates the upper limit of the magnitude of acceleration. In addition, T represents the initial value of the motion trajectory of the robot classified according to the initial conditions, but in the following description, this initial value is simply referred to as the trajectory.
[0049]
[0050] In steps S3 to S6, the passive pair is assigned a combination of gaps or gaps (for example, in this embodiment, there are 12 joints, so there are 2 12 The initial motion of the pair with a gap when the robot moves along the selected action trajectory is obtained by simulation. Here, as a model of the ball pair, consider Figure 7 a to 7c show three types of models (a) to (c). i,j,k Represents the dual S i,j,k In addition, in these models, the friction between the dual elements is not considered.
[0051] In step S3, model (a) is applied to the pair without gap setting, and model (b) is temporarily applied to the pair with gap setting. Here, model (a) is an ideal constraint condition (c) that there is no gap between the pair elements (joint ball 28 and housing 30). i,j,k= 0), while model (b) is a model that imposes a constraint condition that there is a gap between the dual elements (c i,j,k >0) and the dual elements slide relative to each other.
[0052] In model (b), it can be considered that a pair of balls 64 and 66 of negligible size are hypothetically arranged at the center of each joint element (joint ball 28 and housing 30), and the pair of balls 64 and 66 are connected by a hypothetical connection 68 with a mass of zero (equivalent to c i,j,k ) connection. The direction of the imaginary connecting member 68 at rest is such that the ball pairs 64 and 66 at its two ends are located on the line of action of the dual force F. The result of simulating the activity of the dual according to the initial conditions of the selected orbit is that when the traction force acts on the imaginary connecting member in a resting posture, that is, when the magnitude of the dual force F when the traction direction is positive is greater than 0, the center of the joint ball 28 (ball pair 64) moves on the circumference of the imaginary circle 70 relative to the center of the housing 30 (ball pair 66), that is, the joint ball 28 does not separate / collide with the housing 30, but slides. That is, the condition that the model (b) representing the sliding motion should satisfy is that the above-mentioned dual force F is greater than 0.
[0053] On the other hand, model (c) imposes the constraint that there is a gap between the dual elements (c i,j,k >0), and the model when the dual elements are separated / colliding with each other. The condition of applying model (c) is that the result of simulating the dual activity according to the initial conditions of the selected orbit is the dual force f when the joint ball 28 is stationary relative to the housing 30 and the relative acceleration of the joint ball 28 relative to the housing 30 The inner product of is negative (that is, the two vectors form an obtuse angle). In this case, as in model (b), the center of the joint ball 28 (ball pair 64) does not move on the circumference of the imaginary circle 70 relative to the center of the housing 30 (ball pair 66), but temporarily leaves from a point 72 on the circumference of the imaginary circle 70 and moves inside the imaginary circle 70, and then reaches another point 74 on the circumference of the imaginary circle 70 again. As a result, the joint ball 28 performs a collision motion relative to the housing 30.
[0054] Next, in order to simulate the movement of the robot (end effector, etc.) when it moves along the motion trajectory selected in step S2, the motion equation related to the position and posture of each connection member is solved in consideration of input torque and gravity, etc. (step S4), and the conditions that the models (b) and (c) of the duality with a gap should satisfy are derived (step S5). Next, in step S6, it is determined whether each duality with a gap satisfies the derived conditions, and for the duality that does not satisfy the conditions, the model (b) and the model (c) are exchanged (step S7), and the calculation of step S4 is performed again. As described above, in step S3, the model (b) is temporarily applied to all the dualities with gaps, but through the processing of steps S4 to S7, the model (c) is applied to the dualities that do not satisfy the conditions of the model (b), so that models without contradictions can be set for all the dualities in the selected trajectory.
[0055] Next, in step S8, the score (Sc value) of the selected action trajectory is calculated and stored in the above-mentioned storage unit 50, etc. Here, the score (Sc) refers to an index value, which is set to a higher value when the model (c) is applied to the object pair (joint) specified in step S1 than when the model (b) is applied, and a lower value when the model (c) is applied to a pair other than the object pair than when the model (b) is applied, and can be calculated, for example, according to the following formula (3).
[0056]
[0057] Figure 8 A specific example of the u value in formula (3) is expressed in tabular form. The u value is zero when model (c) is applied to the object duality, and when model (b) is applied to a duality other than the object duality that is set to have a gap. In addition, the u value is a negative value when model (b) is applied to the object duality, and when model (c) is applied to a duality other than the object duality that is set to have a gap. However, in this embodiment, it is important to perform collision motion in the object duality, so the absolute value of the u value of the former is increased. However, this is an example, and the value of u can be appropriately set / changed according to the matching with the actual measured value described later, etc.
[0058] In addition, in formula (3) and Figure 8 In the example, the maximum score is 1, and the action trajectory with a score of 1 indicates a trajectory that generates collisions of the dual elements only in the target dual, regardless of how gaps are generated in other duals. Therefore, when the robot is moved in an action trajectory with a score of 1, collision motion is generated only in the target dual, and sliding motion is generated in duals other than the target dual (even if there are gaps).
[0059] Scores are calculated for all pre-randomly selected action trajectories (step S9), and after all action trajectories have been scored, the best action trajectory is selected based on the scores (step S10). By scoring multiple action trajectories, the eligibility of action trajectories as gaps for detecting object pairs can be quantitatively compared / judged.
[0060] In addition, the optimal motion trajectory in step S10 is selected by comparing the calculated score with a predetermined first threshold value (for example, 0.7, 0.8, or 0.9). By selecting at least one motion trajectory with a score above the first threshold value, it is possible to generate a robot action that only produces a collision or approaches the object pair. As described above, it can be said that the motion trajectory with a score of 1 is ideal, so you can simply select a motion trajectory with a score of 1 or the maximum, but sometimes you cannot obtain a motion trajectory with a score of 1 by setting any joint as the object pair, so it is preferable to set a first threshold value less than 1 as described above. In addition, the magnitude of the impact force generated by the collision is not fixed, and depending on the size of the gap, it is sometimes impossible to detect the above-mentioned changes in the drive torque or current value. Therefore, multiple motion trajectories with scores exceeding the first threshold value can be selected as the optimal motion trajectory, and each selected motion trajectory can be processed as described later.
[0061] Alternatively, in step S10, instead of selecting one or more action trajectories with scores greater than the first threshold as the optimal trajectory, an optimization method as described below may be used. Formula (4) shows an example of an optimization method. Here, first, based on the scores and the first threshold Sc th The size relationship between them classifies multiple action tracks into groups A and B. th Among the action tracks belonging to group A with the above scores, select the one with the dissatisfaction threshold Sc th The action track corresponding to the action track that is farthest from the action track belonging to group B is the score. However, in formula (4), the set of tracks A and B are defined by formula (5). In addition, d(T, T') represents the set of all tracks (W, R 3 ) can be calculated according to formula (6). In addition, r in formula (6) is the proportionality factor between acceleration and displacement.
[0062]
[0063]
[0064]
[0065] By using formulas (4) to (6) for optimization, it is possible to select the track farthest from the track with a score lower than the threshold (group B) from the tracks with a score higher than the first threshold (group A). Alternatively, it is also possible to select multiple tracks from group A that are more than a predetermined distance away from group B. Optimization will effectively prevent the combination of colliding ball pairs from changing (resulting in a large change in score) due to a small change in the track, and it can be said to be a robust method compared to simply selecting the action track with the largest score or above the first threshold.
[0066] Figure 9a and Figure 9b This shows an example of using the above optimization method to select the orbit that best fits the gap of the pair to be determined. Figure 1 In the lateral double-swing arm robot shown in FIG. 1 , a randomly selected motion trajectory starting from a certain position x0 is set as a max =30m / s 2 Sc th =0.7, r = 0.003s 2 , investigate the orbits that produce collisions only in certain ball pairs or have a tendency to approach them. Figure 9a As shown in the figure, the robot starts to move from a state where the representative point is stationary at a certain position x0 (speed is 0) with an acceleration of The 4 tracks T1 to T4 start the action.
[0067] In addition, when scoring each track as shown in FIG9 , tracks T2 and T4 belong to the above group A, and tracks T1 and T3 belong to the above group B. Here, if the maximum score is simply selected, track T2 is selected as the optimal track, but in the above optimization method, among the tracks belonging to group A, track T4, which is farthest from the tracks included in group B, is selected as the optimal track.
[0068] In addition, Figure 9a and Figure 9b In the example, only four action trajectories starting from a specific point (x0) are shown, but you can further select an action trajectory starting from x0, or you can select multiple action trajectories starting from positions other than x0. As an example, hundreds of thousands to millions of action trajectories can be randomly selected and scored.
[0069] In addition, the processing (simulation) of steps S4 to S7 may be omitted. Figure 6For example, when the processing of steps S4 to S7 results in a result that collision occurs only in the object pair (model (c) is applied) and sliding motion occurs in all pairs other than the object pair (model (b) is applied), the action trajectory can be selected as the optimal action trajectory without performing score calculation. In this case, the score calculation unit 40 functions as a simulation execution unit.
[0070] Refer again Figure 6 The control device 36 generates a robot action (action instruction) for moving the robot along the optimal action track selected in step S10, and according to the robot action, the robot is actually moved (step S11), and the driving torque (or current value) of the motor at this time is measured (step S12). In addition, as the measuring unit 44 for measuring the driving torque or current value of the motor, a torque sensor, an ammeter, or other well-known means can be used, so detailed description is omitted.
[0071] Next, the above-mentioned index value (abnormality degree) calculated based on the measured driving torque (or current value) is compared with a predetermined second threshold value (described later) (step S13). If the index value exceeds the second threshold value, it is determined that there is an abnormal (above a certain value) gap in the object pair (steps S14, S15).
[0072] The abnormality detection device 38 can be suitably executed in any case where it is integrated with the robot control device 36 or provided as an independent device such as a PC. Figure 6 For example, when a processing unit such as a PC is connected to the robot control unit 36, the processing unit may execute the above steps S1 to S10, and the robot control unit 36 may execute the steps after step S11. In this case, the structural elements (processor, memory, etc.) in the processing unit and the robot control unit 36 for executing the steps after step S11 are included in the abnormality detection device involved in the present invention.
[0073] Fig.10 In order to verify whether the gap is properly determined by the selected action track, Figure 1 The results of an experiment in which a gap was actually applied to the spherical joints in an actual device of a delta-type lateral double-swing arm robot shown in FIG. Here, as shown in the white bar graph, the four joints (S2, 2,1 , S 2,2,2 , S 3,2,1 , S 3,2,2) applies an excessive gap of about 0.8mm to 1.5mm, and the other 8 joints are in an ideal state with no gap or can be ignored. In this example, the optimal motion trajectory is selected when the 12 joints are respectively used as the object pair, and the robot is actually moved along the selected motion trajectory, and the above-mentioned index value (abnormality) is calculated based on the measured motor current value. In addition, the value of the current sensor (monitoring current value) set inside the robot control device is used as the current value of the actuator (motor) used for calculating the abnormality.
[0074] Fig.10 The black bar graph shows the calculated abnormality (dimensionless quantity). 4 By setting the second threshold value as the comparison object with the abnormality, it is possible to make a judgment that conforms to the actual situation in three of the four joints that actually have excessive gaps. By appropriately setting the second threshold value in this way, it is possible to fairly accurately judge whether the object pair has an abnormal gap.
[0075] In addition, as described in step S10, when determining a specific abnormal gap, the (optimal) motion trajectory selected to make the robot actually move can be either one or multiple. In the case of selecting multiple motion trajectories as the optimal motion trajectory, the determination result (the magnitude relationship between the index value and the second threshold) may be different for each motion trajectory, but in this case, from the perspective of safety, it is preferred that if at least one result in which the index value exceeds the second threshold is obtained, it is also determined that an abnormal gap has occurred in the object pair. In addition, in this embodiment, as a specific example of determining whether there is an abnormal gap between the dual elements based on the size of the index value, the case of using the magnitude relationship between the index value and the second threshold is described, but other methods can also be used. For example, for the same pair, the index value calculated this time and the index value calculated in the past can be compared, and when the difference between the two index values exceeds a predetermined threshold, it is determined that an abnormal gap has occurred. As the past index value in this case, either the most recent index value or the average value of multiple index values calculated in the past can be used.
[0076] As described above, in this embodiment, the action track that only produces collision motion in the object pair or a track close to it can be automatically selected from a plurality of tracks by simulation, so that the robot actually moves along the selected action track. Figure 4 or Figure 5In the case of a change in the driving torque or current value as shown, it can be determined that there is an abnormal gap in the object pair. In addition, if the object pair is changed and the same process is performed, it can be determined whether there is an abnormal gap in all joints included in the robot, thereby efficiently determining the joint that should be replaced or repaired.
[0077] In addition, in this embodiment, a lateral double-swing arm robot is described as a robot to which the abnormality detection device and abnormality detection method of the present invention can be applied, but the application object is not limited to this. As other suitable examples to which the abnormality detection device and abnormality detection method of the present invention can be applied, there can be listed: Fig.11 or Fig.12 A robot is schematically shown having at least partially a closed loop type connection mechanism.
[0078] Fig.11 A robot 80 having a driving joint unit 76 including two driving joints and a planar connection mechanism including three passive joints 78 is shown, and a load is applied to the front end. Fig.12 The robot 86 has a five-joint connection mechanism including two driving joints 82 and three passive joints 84, and can be used for positioning devices, etc. Figure 1 The lateral double-swing arm robot shown in the figure similarly has a driving connector driven by an electric motor, a plurality of passive connectors driven by the movement of the driving connector, and a plurality of pairs respectively connected to the plurality of passive connectors, so that it is possible to determine / detect joints (passive pairs) with abnormal gaps in the same way as described above.
[0079] In addition, in the present embodiment, a spherical joint (ball joint) is described as a dual (joint) to which the abnormality detection device and abnormality detection method involved in the present disclosure can be applied, but the application object is not limited to this. The abnormality detection device and the abnormality detection method can also be applied to a hinge structure (rotational joint) with a degree of freedom of 1, for example. In this case, as a dual element, the rotational joint (hinge structure) has a roughly cylindrical member (convex portion) and a roughly cylindrical member (concave portion) engaged with the cylindrical member. In such a hinge structure, due to time-dependent degradation (wear of the cylindrical member or the cylindrical member, or wear of the gasket between the cylindrical member and the cylindrical member), an abnormal gap in the diameter direction may also be generated between the cylindrical member and the cylindrical member, so the abnormality detection device and the abnormality detection method involved in the present disclosure can also be applied.
[0080] According to the present disclosure, it is possible to easily and accurately estimate whether or not there is an abnormal gap in the pair (joint) of the robot, and to identify the pair in which the abnormal gap has occurred.
Claims
1. An abnormality detection device for detecting an abnormal gap between dual elements of a dual connected to a passive connector in a robot, the robot comprising: A drive connection driven by an electric motor; A plurality of passive connecting members driven by the action of the driving connecting member; A plurality of pairs respectively connected to the plurality of passive connectors, It is characterized in that The abnormality detection device has: a score calculation unit that performs a simulation of causing the robot to move along a plurality of predetermined motion trajectories, respectively, and calculates a score for each of the plurality of motion trajectories, the score becoming higher when a collision between paired elements occurs in a target pair specified from among the plurality of pairs for which the presence or absence of an abnormal gap should be determined, and becoming lower when a collision between paired elements occurs in a pair other than the target pair; a motion generating unit for generating a robot motion for causing the robot to move along the motion track having the score being equal to or greater than a predetermined threshold value among the plurality of motion tracks; a measuring unit that measures a driving torque or a current value of the motor when the robot moves according to the robot motion; an index value calculation unit that calculates an index value based on the magnitude of a change in a value related to the drive torque or the current value measured by the measurement unit; as well as a determination unit for determining whether there is an abnormal gap between the dual elements of the object pair according to the magnitude of the index value; The above-mentioned action generation unit classifies the above-mentioned multiple action trajectories into a first group and a second group according to the size relationship between each score and the above-mentioned threshold, and generates a robot action corresponding to the following action trajectory, namely, an action trajectory belonging to the first group with a score above the above-mentioned threshold that is farthest away from an action track belonging to the second group with a score less than the above-mentioned threshold, or an action track that is separated from an action track belonging to the above-mentioned second group by a predetermined distance or more.
2. The abnormality detection device according to claim 1, characterized in that: The motion generation unit generates a robot motion corresponding to the motion trajectory with the highest score.
3. An abnormality detection device for detecting an abnormal gap between a pair of elements of a pair connected to a passive connector in a robot, the robot comprising: A drive connection driven by an electric motor; A plurality of passive connecting members driven by the action of the driving connecting member; A plurality of pairs respectively connected to the plurality of passive connectors, It is characterized in that The abnormality detection device has: a simulation execution unit for performing a simulation of causing the robot to perform actions along a plurality of predetermined action trajectories, respectively, and selecting, from the plurality of action trajectories, an action trajectory in which a collision between the pair elements due to the gap occurs only in the pair, assuming that there is a gap between the pair elements of the target pair specified from the plurality of pairs for which the presence or absence of an abnormal gap is to be determined, wherein in the simulation, a first model or a second model is applied to the pair in which a gap is set, the first model being a model in which a constraint condition that a gap exists between the pair elements and the pair elements slide against each other, and the second model being a model in which a constraint condition that a gap exists between the pair elements and the pair elements separate / collide against each other, and the selected action trajectory is an action trajectory in which the second model is applied only to the target pair and the first model is applied to all pairs other than the target pair; an action generation unit for generating a robot action for causing the robot to move along the action trajectory selected by the simulation execution unit; a measuring unit that measures a driving torque or a current value of the motor when the robot moves according to the robot motion; an index value calculation unit that calculates an index value based on a magnitude of a change in a value related to the drive torque or the current value measured by the measurement unit; and A determination unit determines whether or not there is an abnormal gap between the paired elements of the object pair based on the magnitude of the index value.
4. The abnormality detection device according to any one of claims 1 to 3, characterized in that: The index value calculation unit calculates an index value based on a magnitude of a change in a time differential value of the drive torque or a magnitude of a change in a time differential value of the current value.
5. The abnormality detection device according to any one of claims 1 to 3, characterized in that: The driving connection member and the passive connection member constitute at least one closed-loop connection member.
6. An abnormality detection method for detecting an abnormal gap between dual elements of a dual connected to a passive connector in a robot, the robot comprising: A drive connection driven by an electric motor; A plurality of passive connecting members driven by the action of the driving connecting member; A plurality of pairs respectively connected to the plurality of passive connectors, It is characterized in that The anomaly detection method includes: A simulation is performed in which the robot is caused to move along a plurality of predetermined motion trajectories, respectively, and scores are calculated for each of the plurality of motion trajectories, the scores becoming higher when a collision between the pair elements occurs in a target pair specified from the plurality of pairs for which the presence or absence of an abnormal gap should be determined, and becoming lower when a collision between the pair elements occurs in a pair other than the target pair; generating a robot motion for causing the robot to move along a motion trajectory having the score being equal to or greater than a predetermined threshold value among the plurality of motion trajectories; measuring the driving torque or current value of the motor when the robot moves according to the robot movement; calculating an index value based on the magnitude of a change in a value related to the measured drive torque or current value; According to the value of the above index, it is determined whether there is an abnormal gap between the dual elements of the above object pair. The abnormality detection method includes: classifying the above-mentioned multiple action trajectories into a first group and a second group according to the size relationship between each score and the above-mentioned threshold, and generating robot actions corresponding to the following action trajectories, namely, the action trajectory belonging to the first group with a score above the above-mentioned threshold that is farthest away from the action track belonging to the second group with a score less than the above-mentioned threshold, or the action track that is separated from the action track belonging to the above-mentioned second group by a predetermined distance or more.
7. An abnormality detection method for detecting an abnormal gap between dual elements of a dual connected to a passive connector in a robot, the robot comprising: A drive connection driven by an electric motor; A plurality of passive connecting members driven by the action of the driving connecting member; A plurality of pairs respectively connected to the plurality of passive connectors, It is characterized in that The anomaly detection method includes: A simulation is performed in which the robot is caused to move along a plurality of predetermined motion trajectories respectively. Under the assumption that there is a gap between the dual elements of a target pair specified from the plurality of pairs for which the presence or absence of an abnormal gap should be determined, a motion trajectory in which a collision between the dual elements due to the gap occurs only in the target pair is selected from the plurality of motion trajectories. In the simulation, a first model or a second model is applied to a pair in which a gap is set. The first model is a model in which a constraint condition that a gap exists between the dual elements and the dual elements slide against each other. The second model is a model in which a constraint condition that a gap exists between the dual elements and the dual elements separate / collide against each other. The selected motion trajectory is an motion trajectory in which the second model is applied only to the target pair and the first model is applied to all pairs other than the target pair. generating a robot motion for causing the robot to move along the motion trajectory selected through the simulation; measuring the driving torque or current value of the motor when the robot moves according to the robot movement; calculating an index value based on a magnitude of a change in a value related to the measured drive torque or current value; Based on the size of the above index value, determine whether there is an abnormal gap between the dual elements of the above object pair.
