Robot system and control method of robot system

Abstract

To provide a robot system and a control method of the robot system capable of suppressing the failure of work and the repeated operation that damages a workpiece.SOLUTION: A robot system includes: a sensor that identifies a workpiece; a robot that suctions by an end-effector and transfers the workpiece; and a robot control device that controls driving of the robot. When the workpiece is identified as a first workpiece by the sensor, the robot control device operates the robot by a first drive command based on a first suction condition. and when the transfer of the workpiece according to the first drive command fails, the robot control device updates the first suction condition. Further, when the workpiece is identified as a second workpiece different from the first workpiece by the sensor, the robot control device operates the robot according to a second drive command based on a second suction condition, and when the conveyance of the workpiece according to the second drive command fails, the robot control device updates the second suction condition.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a robot system and a control method for a robot system. [Background technology]

[0002] For example, the robot system described in Patent Document 1 operates according to an operation command determined according to data of an object acquired by a sensor. Specifically, the robot system includes a robot that grasps and transports an object, a sensor that measures the object, a data analysis unit that analyzes first data of the object acquired by measuring the object by the sensor, a database that stores second data of the object in association with a status of the object determined corresponding to an operation mode of the robot when the first data is acquired, and an operation instruction unit that gives an operation instruction to the robot to process the object according to the operation command determined according to the second data of the object in the operation mode determined according to the status of the object stored in the database. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-011913 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, when a robot picks up an object and performs a specified task, if the operation commands given by the operation instruction unit are inappropriate for the object, there is a risk that the robot will fail to perform the task on the object or will repeatedly perform operations that damage the object. [Means for solving the problem]

[0005] The robot system of the present invention includes a sensor for identifying a workpiece; a robot having a robot arm and an end effector disposed on the robot arm, the robot adsorbing the workpiece with the end effector and performing a predetermined operation; A robot control device that controls the driving of the robot, The robot control device includes: When the sensor identifies the workpiece as a first workpiece, the robot is operated according to a first drive command based on a first pickup condition, and when the operation according to the first drive command fails, the first pickup condition is updated; If the sensor identifies the workpiece as a second workpiece different from the first workpiece, the robot is operated according to a second drive command based on a second pickup condition, and if the operation with the second drive command fails, the second pickup condition is updated.

[0006] A control method for a robot system of the present invention is a control method for a robot system including a sensor for identifying a workpiece, a robot arm, and an end effector disposed on the robot arm, the end effector adsorbing the workpiece and performing a predetermined operation, the control method comprising the steps of: When the sensor identifies the workpiece as a first workpiece, the robot is operated according to a first drive command based on a first pickup condition, and when the operation according to the first drive command fails, the first pickup condition is updated; If the sensor identifies the workpiece as a second workpiece different from the first workpiece, the robot is operated according to a second drive command based on a second pickup condition, and if the operation with the second drive command fails, the second pickup condition is updated. [Brief description of the drawings]

[0007] [Figure 1] 1 is a configuration diagram of a robot system according to a first embodiment. [Diagram 2] FIG. 13 is a diagram showing a state in which the end effector adsorbs a first workpiece. [Diagram 3]FIG. 13 is a diagram showing a state in which the end effector adsorbs a second workpiece. [Figure 4] 11 is a flowchart illustrating a control method for the robot system. [Diagram 5] FIG. 2 is a diagram illustrating an example of a database. [Figure 6] 10 is a flowchart illustrating a control method for the robot system according to the second embodiment. [Figure 7] 13 is a flowchart illustrating a control method for a robot system according to a third embodiment. [Figure 8] 13 is a flowchart illustrating a control method of a robot system according to a fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A robot system and a control method for a robot system according to the present invention will be described in detail below with reference to the embodiments shown in the accompanying drawings.

[0009] First Embodiment Fig. 1 is a configuration diagram of a robot system according to a first embodiment. Fig. 2 is a diagram showing a state in which an end effector picks up a first workpiece. Fig. 3 is a diagram showing a state in which an end effector picks up a second workpiece. Fig. 4 is a flowchart for explaining a control method of the robot system. Fig. 5 is a diagram showing an example of a database.

[0010] 1 is applied to a transport and assembly system that transports and assembles a workpiece W. Such a robot system 1 includes a transport device 5 that transports the workpiece W, a sensor 4 that identifies the workpiece W on the transport device 5, a robot 2 that grasps the workpiece W on the transport device 5 and performs a predetermined task, and a robot control device 6 that controls the driving of each of these components.

[0011] <Robot 2> As shown in FIG. 1, the robot 2 has a robot body 20 and an end effector 3 attached to the robot body 20.

[0012] The robot body 20 is a six-axis vertical articulated robot having six drive axes, and includes a base 21 fixed to the floor and a robot arm 22 rotatably connected to the base 21. The robot arm 22 includes six arms 221, 222, 223, 224, 225, and 226 rotatably connected in this order from the base 21 side, and includes six joints J1, J2, J3, J4, J5, and J6. Among these joints J1 to J6, the joints J2, J3, and J5 are bending joints, and the joints J1, J4, and J6 are torsion joints. Although not shown, a drive mechanism including a motor for moving the joint and an encoder for detecting the amount of rotation of the joint is installed in each of the joints J1, J2, J3, J4, J5, and J6, and the motor is driven by servo control that feeds back the output of the encoder.

[0013] However, the configuration of the robot 2 is not particularly limited. For example, the number of arms provided in the robot arm 22 is not limited to six. The robot 2 may also be a dual-arm robot, a horizontally articulated robot (SCARA robot), or the like. The robot 2 may also be a robot that is not fixed to the floor and is capable of self-propelling.

[0014] The end effector 3 is a suction head 30. The suction head 30 has a number of suction nozzles 31 arranged in a row. As shown in Figs. 2 and 3, the suction head 30 sucks and grips the workpiece W by reducing the pressure inside the suction nozzle 31 while pressing the suction surface of at least one of the suction nozzles 31 against the surface of the workpiece W. The suction force of the workpiece W (hereinafter, simply referred to as "suction force") can be adjusted by changing the pressure inside the suction nozzle 31. Specifically, the lower the pressure inside the suction nozzle 31, the stronger the suction force.

[0015] Moreover, each suction nozzle 31 has elasticity and is configured to deform along the surface of the workpiece W. With such a configuration, one suction head 30 can accommodate workpieces W of various shapes, making the suction head 30 suitable for this embodiment in which various types of workpieces W are used.

[0016] The configuration of the suction head 30 is not particularly limited. For example, each suction nozzle 31 may be hard and not substantially elastically deformable. The arrangement of the plurality of suction nozzles 31 is also not particularly limited, and for example, the plurality of suction nozzles 31 may be arranged in two rows. The number of suction nozzles 31 is also not particularly limited, and may be one, for example.

[0017] <Conveyor device 5> As shown in FIG. 1, the transport device 5 is a belt conveyor 50, which transports the workpiece W to the working area of ​​the robot 2. Such a belt conveyor 50 has a belt 51, a transport roller that sends the belt 51, a motor that drives the transport roller, and a transport amount sensor that outputs a signal according to the rotation amount of the transport roller to the robot control device 6. Note that in FIG. 1, each part other than the belt 51 is omitted. During operation of the robot system 1, the robot control device 6 executes feedback control to match the transport speed of the workpiece W indicated by the output of the transport amount sensor with a target transport speed that is a control target. This allows the workpiece W to be transported stably at a desired speed.

[0018] However, the configuration of the transport device 5 is not particularly limited. Also, the transport device 5 may be omitted. In this case, for example, the workpieces W may be stored randomly in a tray, and the robot 2 may take out the workpieces W from there.

[0019] <Sensor 4> 1, the sensor 4 is disposed above the transport device 5 on the upstream side of the working area of ​​the robot 2. The sensor 4 identifies the workpiece W transported by the transport device 5 to the working area of ​​the robot 2. The sensor 4 includes a camera 40 and an image recognition unit 41.

[0020] The camera 40 captures an image of the workpiece W on the transport device 5 from above the transport device 5. The image recognition unit 41 identifies the workpiece W based on the image captured by the camera 40, and detects the position and posture of the workpiece W. For ease of explanation, the following will be described taking as an example a case where the workpiece W is identified as either a first workpiece W1 or a second workpiece W2 different from the first workpiece W1. However, the present invention is not limited to this, and the workpiece W may be identified into three or more types, for example.

[0021] In this embodiment, the workpiece W is identified and its posture is detected by template matching. In this method, 3D models of the first workpiece W1 and the second workpiece W2 are created in advance using CAD (Computer Aided Design). Next, the created 3D models are placed in various postures through 360°, the postures of the first workpiece W1 and the second workpiece W2 in the contour shapes at that time are estimated, and a learning model is generated in which the contour shapes and postures are linked. Then, the workpiece W is extracted from the image acquired by the camera 40, and the extracted workpiece W is compared with the learning model to identify the workpiece W and calculate the posture of the workpiece W. According to this method, the workpiece W can be identified and its posture detected with high accuracy.

[0022] Furthermore, each pixel in the image captured by the camera 40 is associated with a robot coordinate system set in the robot 2, and the position in the robot coordinate system is known. Therefore, the image recognition unit 41 can identify the coordinates of the workpiece W based on the position of the workpiece W in the image. The image recognition unit 41 can estimate the trajectory of the workpiece W (the position at each time) based on the time the image was captured, the position of the workpiece W at that time, and the transport speed of the transport device 5.

[0023] Although the method for identifying the workpiece W and the method for detecting the position and posture of the workpiece W have been described above, these methods are not particularly limited. Furthermore, the sensor 4 may be disposed on the robot 2, for example. Furthermore, as long as the workpiece W can be identified, the configuration of the sensor 4 is not particularly limited, and for example, an infrared sensor, an ultrasonic sensor, or the like may be used.

[0024] <Robot control device 6> The robot control device 6 controls the driving of the robot 2, the sensor 4, and the transport device 5. Such a robot control device 6 is, for example, composed of a computer, and has a processor (CPU) for processing information, a memory communicatively connected to the processor, and an external interface for connecting to an external device. Various programs executable by the processor are stored in the memory, and the processor can read and execute the various programs stored in the memory. The robot control device 6 may be composed of multiple processors. Also, the robot control device 6 may have a part or all of the image recognition unit 41 that the sensor 4 has.

[0025] The configuration of the robot system 1 has been described above. Next, a control method for the robot system 1 will be described. For ease of explanation, the following will be described taking as an example a case where the first and second workpieces W1 and W2 are randomly transported as the workpieces W by the transport device 5, and the robot 2 sucks and holds the first workpiece W1 and inserts it into the part Q1, and sucks and holds the second workpiece W2 and inserts it into the part Q2, as an assembly operation. However, the content of the operation is not particularly limited.

[0026] 4, the control method of the robot system 1 includes an identification step S1 for identifying a workpiece W, a pickup condition determination step S2 for determining pickup conditions according to the type of workpiece W identified in the identification step S1, a work step S3 for suction-holding the workpiece W under the pickup conditions determined in the pickup condition determination step S2 and performing a predetermined work, a determination step S4 for determining whether the work step S3 has succeeded or failed, and a pickup condition update step S5 for updating the pickup conditions based on the determination result in the determination step S4. Each of these steps S1 to S5 will be described in order below.

[0027] <Identification step S1> In the identification step S1, the camera 40 captures an image of the transport device 5 at a set period. The image recognition unit 41 checks the image captured by the camera 40 and determines whether a workpiece W is present in the image. If a workpiece W is present in the image, the image recognition unit 41 extracts the workpiece W from the image and identifies whether the workpiece W is the first workpiece W1 or the second workpiece W2 by the template matching described above, and detects the posture of the workpiece W. The image recognition unit 41 also detects the position (coordinates) of the workpiece W at the time the image was captured from the position of the workpiece W in the image, and estimates the trajectory of the workpiece W. The image recognition unit 41 then transmits each piece of detected information to the robot control device 6.

[0028] In addition, if a workpiece other than the first and second workpieces W1, W2 is transported by the transport device 5, it may be identified as the one having the greater similarity between the first and second workpieces W1, W2, and then, if necessary, information about that workpiece may be registered in the database DB.

[0029] <<Adsorption condition determination step S2>> The robot control device 6 stores a database DB that predetermines suction conditions for suctioning the workpiece W, specifically, a first suction condition for suctioning the first workpiece W1 and a second suction condition for suctioning the second workpiece W2. As shown in FIG. 5, the database DB predetermines the suction force (pressure in the suction nozzle 31) of the workpiece W, the settable maximum suction force and minimum suction force, and the suction nozzle 31 used for suction (hereinafter also referred to as the "used suction nozzle 310") for each of the first and second suction conditions. In this way, by setting the suction conditions for each type of workpiece W, the first and second workpieces W1 and W2 can be sucked and worked under appropriate conditions. Therefore, it is possible to effectively suppress work failures and stably perform appropriate work. However, the items to be determined in the first and second suction conditions are not particularly limited.

[0030] The maximum suction force and the minimum suction force can be appropriately set for each of the first and second workpieces W1, W2 depending on the work content, the configuration (weight, fragility) of the first and second workpieces W1, W2, and the like.

[0031] When the robot control device 6 acquires each piece of information obtained in the identification step S1 from the image recognition unit 41, it checks the database DB and determines the suction conditions according to the workpiece W. In other words, if the target workpiece W is the first workpiece W1, it selects the first suction conditions, and if the target workpiece W is the second workpiece W2, it selects the second suction conditions.

[0032] Although the adsorption condition determination step S2 has been described above, the method of determining the adsorption conditions is not particularly limited. For example, the adsorption conditions may be determined using a trained machine learning model. In other words, a machine learning model that has been trained using a large amount of sample data is prepared, and when the identification result of the workpiece W is input to the machine learning model, the machine learning model analyzes the input data, performs evaluation, and outputs the result (adsorption conditions). Then, the robot control device 6 determines the result output from the machine learning model as the adsorption condition. Even with such a method, the first and second workpieces W1 and W2 can be adsorbed and worked under appropriate conditions. Therefore, it is possible to effectively suppress work failures and stably perform appropriate work.

[0033] The database DB may be stored in, for example, a server and may be accessible via a communication line such as the Internet. With this configuration, the database DB can be shared by a plurality of robot systems 1.

[0034] <Work step S3> In the operation step S3, the robot control device 6 controls the driving of the robot 2 according to the suction conditions determined in the suction condition determination step S2 and according to the operation command determined according to the posture and trajectory of the workpiece W detected in the identification step S1. Specifically, when the target workpiece W is the first workpiece W1, the suction conditions are first set to the first suction conditions. Then, the robot control device 6 controls the driving of the robot 2 according to the first operation command determined according to the posture and trajectory of the first workpiece W1, including the posture of the suction head 30 when suctioning the first workpiece W1, the position and timing of suctioning the first workpiece W1, and the transport trajectory of the first workpiece W1 by the robot 2, and performs the operation of suction-holding the first workpiece W1 and inserting it into the part Q1. Similarly, when the target workpiece W is the second workpiece W2, the suction conditions are first set to the second suction conditions. Furthermore, the drive of the robot 2 is controlled according to a second operation command which is determined according to the posture and trajectory of the second work W2 and includes the posture of the suction head 30 when suctioning the second work W2, the position and timing for suctioning the second work W2, and the transport trajectory of the second work W2 by the robot 2, and the like, and the second work W2 is suction-held and inserted into part Q2.

[0035] <Determination step S4> In the judgment step S4, the robot control device 6 judges whether the work step S3 has been successful or unsuccessful. "The work step S3 has been successful" means that the first and second workpieces W1 and W2 can be inserted into the parts Q1 and Q2 without causing damage to the first and second workpieces W1 and W2 due to excessive suction force. On the other hand, "the work step S3 has failed" means, for example, that the first and second workpieces W1 and W2 cannot be inserted into the parts Q1 and Q2 because the first and second workpieces W1 and W2 fall during the work due to insufficient suction force, or that the first and second workpieces W1 and W2 can be inserted into the parts Q1 and Q2, but damage to the first and second workpieces W1 and W2 occurs due to excessive suction force. However, the distinction between success and failure is not particularly limited and can be appropriately set depending on the work content, the required quality, etc.

[0036] The method for determining whether the work step S3 is successful or unsuccessful is not particularly limited. In this embodiment, the parts Q1 and Q2 after the work is completed are photographed by a camera (not shown), and the obtained image is analyzed using image recognition technology to determine whether the work step S3 is successful or unsuccessful. This makes it possible to easily and accurately determine whether the work step S3 is successful or unsuccessful. In addition, this method also makes it easy to determine the cause of the failure, that is, whether the work failed due to the workpiece W falling due to insufficient suction force, or whether the work failed due to damage to the workpiece W due to excessive suction force.

[0037] <<Adsorption condition update step S5>> In the suction condition update step S5, the robot control device 6 changes the suction conditions stored in the database DB based on the result of the determination in the determination step S4.

[0038] If the determination result in the determination step S4 is failure, the robot control device 6 first determines in step S51 whether the cause of the failure is damage to the workpiece W due to an excessive suction force.

[0039] If the cause of the failure is damage to the workpiece W due to excessive suction force, the robot control device 6 determines in step S52 whether the suction force currently set in the database DB is the minimum suction force that can be set, that is, whether the suction force cannot be reduced any further. Specifically, if the target workpiece W is the first workpiece W1, it determines whether the suction force A set in the first suction condition in the database DB is the minimum suction force Amin. Similarly, if the target workpiece W is the second workpiece W2, it determines whether the suction force B set in the second suction condition in the database DB is the minimum suction force Bmin.

[0040] If the suction force set in the database DB is not the minimum suction force, the robot control device 6 reduces the suction force stored in the database DB in step S53 and updates the result. Specifically, if the target workpiece W is the first workpiece W1, the suction force A set in the first suction condition in the database DB is reduced, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the suction force B set in the second suction condition in the database DB is reduced, and the result is updated as a new second suction condition. As a result, the next work is performed under the updated suction condition, and the suction force is smaller than that of the current work. Therefore, the possibility of damage to the workpiece W due to excessive suction force is reduced.

[0041] When the suction force set in the database DB is the minimum suction force, the robot controller 6 skips step S53.

[0042] On the other hand, if the cause of the failure is not damage to the workpiece W due to excessive suction force, that is, if the workpiece W falls during the operation due to insufficient suction force, the robot control device 6 determines in step S54 whether the suction force currently set in the database DB is the maximum suction force that can be set, that is, whether the suction force cannot be increased any further. Specifically, if the target workpiece W is the first workpiece W1, it is determined whether the suction force A set in the first suction condition of the database DB is the maximum suction force Amax. Similarly, if the target workpiece W is the second workpiece W2, it is determined whether the suction force B set in the second suction condition of the database DB is the maximum suction force Bmax.

[0043] If the suction force set in the database DB is not the maximum suction force, the robot control device 6 increases the suction force stored in the database DB and updates the result in step S55. Specifically, if the target workpiece W is the first workpiece W1, the suction force A set in the first suction condition in the database DB is increased, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the suction force B set in the second suction condition in the database DB is increased, and the result is updated as a new second suction condition. As a result, the next work is performed under the updated suction condition, and the suction force is greater than that of the current work. Therefore, the possibility of the workpiece W falling due to insufficient suction force is reduced.

[0044] If the suction force set in the database DB is the maximum suction force, the robot controller 6 skips step S55.

[0045] By performing the above steps S51 to S55 every time the work step S3 fails, the adsorption conditions gradually converge to the appropriate values, and the number of failures in the work step S3 decreases. Therefore, it is possible to effectively prevent the same work failures from being repeated many times, and to perform the work efficiently.

[0046] Here, the amount of reduction in the suction force in step S53 is fixed to a predetermined value ΔM, and the suction force is reduced by the predetermined value ΔM each time step S53 is performed. According to such a method, the amount of reduction in the suction force becomes constant, which simplifies the control and makes it easier to converge the suction force to an appropriate value. However, this is not limited to this, and the predetermined value ΔM may be changed randomly. Also, the predetermined value ΔM may be different between the first workpiece W1 and the second workpiece W2.

[0047] Similarly, the amount of increase in the suction force in step S55 is fixed to a predetermined value ΔP, and the suction force is increased by the predetermined value ΔP each time step S55 is performed. According to such a method, the amount of increase in the suction force becomes constant, which simplifies control and makes it easier to converge the suction force to an appropriate value. However, this is not limited to this, and the predetermined value ΔP may be changed randomly. Also, the predetermined value ΔP may be different between the first workpiece W1 and the second workpiece W2.

[0048] Next, in step S56, the robot control device 6 determines whether the suction nozzle 31 (hereinafter also referred to as the "used suction nozzle 310") used to pick up the workpiece W is appropriate. If the used suction nozzle 310 is not appropriate, the robot control device 6 changes the used suction nozzle 310 stored in the database DB and updates the result in step S57. Specifically, if the target workpiece W is the first workpiece W1, the used suction nozzle 310 set in the first suction condition in the database DB is changed, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the used suction nozzle 310 set in the second suction condition in the database DB is changed, and the result is updated as a new second suction condition. As a result, the next work is performed under the updated suction conditions, and is performed using a suction nozzle 31 different from the current work. Although it is unknown whether the next work will be successful due to such an update, by repeatedly performing step S56, it is possible to find the optimal used suction nozzle 310. This effectively prevents the same mistakes from being made over and over again, allowing for efficient work.

[0049] The method of determining whether the suction nozzle 310 in use is appropriate is not particularly limited, but in this embodiment, when the suction force set in the database DB in step S52 is determined to be the minimum suction force, or when the suction force set in the database DB in step S54 is determined to be the maximum suction force, the suction nozzle 310 in use is determined to be inappropriate. In other words, the suction force is first tried to be changed, and if no improvement in the work is seen even after trying the suction force change, the suction nozzle 310 in use is tried to be changed. The selection of the suction nozzle 310 in use can be performed relatively easily from the configuration of the workpiece W and the suction head 30. Therefore, it is considered that the possibility that the failure of the work is caused by the use of an inappropriate suction nozzle 310 in use is lower than the possibility that the failure of the work is caused by the use of an inappropriate suction force. Therefore, as described above, by first trying to change the suction force and then trying to change the suction nozzle 310 in use, the suction conditions can be converged to the appropriate value with fewer operations.

[0050] The control method of the robot system 1 has been described above. In the above-mentioned method, the suction force is set between the minimum suction force and the maximum suction force, but the present invention is not limited to this, and the suction force may be set to less than the minimum suction force or more than the maximum suction force. In this case, it is preferable to notify the user of this by an alarm or the like. In addition, in the present embodiment, the method of updating the database DB has been described, but when the suction condition is determined using a machine learning model, the failure data may be provided to the machine learning model as sample data and re-learned.

[0051] The robot system 1 and the control method of the robot system 1 have been described above. As described above, the robot system 1 includes a sensor 4 for identifying the workpiece W, a robot arm 22 and an end effector 3 arranged on the robot arm 22, a robot 2 for adsorbing the workpiece W with the end effector 3 and performing a predetermined task, and a robot control device 6 for controlling the driving of the robot 2. In addition, when the workpiece W is identified by the sensor 4 as the first workpiece W1, the robot control device 6 operates the robot 2 by a first drive command based on the first adsorption condition, and when the task by the first drive command fails, updates the first adsorption condition. Similarly, when the workpiece W is identified by the sensor 4 as the second workpiece W2 different from the first workpiece W1, the robot control device 6 operates the robot 2 by a second drive command based on the second adsorption condition, and when the task by the second drive command fails, updates the second adsorption condition. With this configuration, the first and second adsorption conditions gradually converge to appropriate values, and the number of failures in the task decreases. Therefore, it is possible to effectively prevent the same task failures from being repeated many times, and efficient work can be performed.

[0052] As described above, the end effector 3 is a suction head 30 having a plurality of suction nozzles 31. The update of the first suction condition includes changing the suction nozzle 31 used to pick up the first workpiece W1, i.e., the suction nozzle in use 310. Similarly, the update of the second suction condition includes changing the suction nozzle 31 used to pick up the second workpiece W2, i.e., the suction nozzle in use 310. With this configuration, it is possible to find the optimal suction nozzle in use 310 while repeating the work. Therefore, it is possible to effectively prevent the same work failure from being repeated many times, and to perform the work efficiently.

[0053] As described above, the robot control device 6 determines that the work has failed if the workpiece W is damaged after the work. The update of the first suction condition when the work with the first drive command fails includes reducing the suction force of the first workpiece W1. Similarly, the update of the second suction condition when the work with the second drive command fails includes reducing the suction force of the second workpiece W2. This configuration reduces the possibility of the workpiece W being damaged due to excessive suction force in the subsequent work. By updating the first and second suction conditions in this way every time the work fails, the suction conditions gradually converge to appropriate values, and the number of work failures decreases. Therefore, it is possible to effectively prevent the same work failures from being repeated many times, and to perform efficient work.

[0054] As described above, the attractive force is decreased by a predetermined value ΔM in each of the updates of the first and second attractive conditions. According to this method, the amount of decrease in the attractive force becomes constant, which simplifies the control and makes it easier to converge the attractive force to an appropriate value.

[0055] As described above, the robot control device 6 determines that the work has failed if the workpiece W falls during the work. The update of the first suction condition when the work with the first drive command fails includes increasing the suction force of the first workpiece W1. Similarly, the update of the second suction condition when the work with the second drive command fails includes increasing the suction force of the second workpiece W2. This configuration reduces the possibility of the workpiece W falling due to insufficient suction force in the subsequent work. By updating the first and second suction conditions in this way every time the work fails, the suction conditions gradually converge to appropriate values, and the number of work failures decreases. Therefore, it is possible to effectively prevent the same work failures from being repeated many times, and to perform efficient work.

[0056] As described above, in updating the first and second adsorption conditions, the adsorptive force is increased by a predetermined value ΔP. According to this method, the amount of increase in the adsorptive force becomes constant, which simplifies the control and makes it easier to converge the adsorptive force to an appropriate value.

[0057] As described above, the control method of the robot system 1 is a control method of the robot system 1 including a sensor 4 for identifying the workpiece W, a robot arm 22, and a robot 2 having an end effector 3 arranged on the robot arm 22, and adsorbing the workpiece W with the end effector 3 to perform a predetermined task. Then, when the sensor 4 identifies the workpiece W as the first workpiece W1, the robot 2 is operated by a first drive command based on the first adsorption condition, and when the task with the first drive command fails, the first adsorption condition is updated. Similarly, when the sensor 4 identifies the workpiece W as the second workpiece W2 different from the first workpiece W1, the robot 2 is operated by a second drive command based on the second adsorption condition, and when the task with the second drive command fails, the second adsorption condition is updated. With this configuration, the first and second adsorption conditions gradually converge to appropriate values, and the number of failures in the task decreases. Therefore, it is possible to effectively prevent the task from being repeatedly and similarly repeated, and the task can be performed efficiently.

[0058] <Second embodiment> FIG. 6 is a flowchart illustrating a control method for the robot system according to the second embodiment.

[0059] The robot system 1 according to this embodiment is similar to the robot system 1 according to the first embodiment, except that the adsorption condition update step S5 is different. In the following description, the robot system 1 according to this embodiment will be described with a focus on the differences from the first embodiment, and the description of the similarities will be omitted. In the drawings of this embodiment, the same reference numerals are used for the same configurations as those in the above-mentioned embodiment.

[0060] Hereinafter, the adsorption condition updating step S5 of the robot system 1 according to this embodiment will be described with reference to FIG.

[0061] <<Adsorption condition update step S5>> If the determination result in the determination step S4 is failure, the robot control device 6 first determines in step S51 whether the cause of the failure is damage to the workpiece W due to an excessive suction force.

[0062] If the cause of the failure is damage to the workpiece W due to an excessive suction force, the robot control device 6 determines in step S52 whether the suction force currently set in the database DB is the minimum suction force.

[0063] If the attractive force set in the database DB is not the minimum attractive force, the robot controller 6 determines in step S531 whether the difference Δ1 between the current attractive force and the minimum attractive force is equal to or larger than a preset threshold value.

[0064] Then, if the difference Δ1 is equal to or greater than the threshold, the robot control device 6 reduces the suction force by a first predetermined value ΔM1 in step S532. Specifically, if the target workpiece W is the first workpiece W1, the suction force set in the first suction condition in the database DB is reduced by the first predetermined value ΔM1, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the suction force set in the second suction condition in the database DB is reduced by the first predetermined value ΔM1, and the result is updated as a new second suction condition.

[0065] On the other hand, if the difference Δ1 is less than the threshold, the robot control device 6 reduces the suction force by a second predetermined value ΔM2 smaller than the first predetermined value ΔM1 in step S533. Specifically, if the target workpiece W is the first workpiece W1, the suction force set in the first suction condition in the database DB is reduced by the second predetermined value ΔM2, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the suction force set in the second suction condition in the database DB is reduced by the second predetermined value ΔM2, and the result is updated as a new second suction condition.

[0066] According to this method, the attractive force can be changed up to the minimum attractive force, which increases the possibility of finding an appropriate attractive force. Note that, in this embodiment, the predetermined value ΔM is switched between a first predetermined value ΔM1 and a second predetermined value ΔM2 based on a threshold value, but this is not limited thereto, and for example, the predetermined value ΔM may be changed linearly or logarithmically based on the magnitude of the difference Δ1.

[0067] The second predetermined value ΔM2 is not particularly limited, but may be, for example, about 1 / 10 or more and 1 / 2 or less of the first predetermined value ΔM1.

[0068] When the suction force set in the database DB is the minimum suction force, the robot controller 6 skips steps S531, S532, and S533.

[0069] On the other hand, if the cause of the failure is not damage to the workpiece W due to excessive suction force, that is, if the workpiece W falls during transportation due to insufficient suction force, the robot control device 6 determines in step S54 whether the suction force currently set in the database DB is the maximum suction force.

[0070] If the attractive force set in the database DB is not the maximum attractive force, the robot controller 6 determines in step S551 whether the difference Δ2 between the current attractive force and the maximum attractive force is equal to or larger than a preset threshold value.

[0071] Then, if the difference Δ2 is equal to or greater than the threshold, the robot control device 6 increases the suction force by a first predetermined value ΔP1 in step S552. Specifically, if the target workpiece W is the first workpiece W1, the suction force set in the first suction condition in the database DB is increased by the first predetermined value ΔP1, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the suction force set in the second suction condition in the database DB is increased by the first predetermined value ΔP1, and the result is updated as a new second suction condition.

[0072] On the other hand, if the difference Δ2 is less than the threshold, the robot control device 6 increases the suction force by a second predetermined value ΔP2 smaller than the first predetermined value ΔP1 in step S553. Specifically, if the target workpiece W is the first workpiece W1, the suction force set in the first suction condition in the database DB is increased by the second predetermined value ΔP2, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the suction force set in the second suction condition in the database DB is increased by the second predetermined value ΔP2, and the result is updated as a new second suction condition.

[0073] According to this method, since the attractive force can be changed up to the limit of the maximum attractive force, the possibility of finding an appropriate attractive force is increased. Note that, in this embodiment, the predetermined value ΔP is switched between a first predetermined value ΔP1 and a second predetermined value ΔP2 based on a threshold value, but this is not limited thereto, and for example, the predetermined value ΔP may be changed linearly or logarithmically based on the magnitude of the difference Δ2.

[0074] The second predetermined value ΔP2 is not particularly limited, but may be, for example, about 1 / 10 or more and 1 / 2 or less of the first predetermined value ΔP1.

[0075] When the suction force set in the database DB is the maximum suction force, the robot controller 6 skips steps S551, S552, and S553.

[0076] The rest of the process is the same as in the first embodiment, and so a description thereof will be omitted.

[0077] As described above, in updating the first and second adsorption conditions, if the difference Δ1 between the current adsorption force and the minimum adsorption force is equal to or greater than the threshold, the adsorption force is reduced by a first predetermined value ΔM1, and if the difference Δ1 is less than the threshold, the adsorption force is reduced by a second predetermined value ΔM2 smaller than the first predetermined value ΔM1. According to this method, the adsorption force can be changed up to the minimum adsorption force, which increases the possibility of finding an appropriate adsorption force.

[0078] In updating the first and second adsorption conditions, if the difference Δ2 between the current adsorption force and the maximum adsorption force is equal to or greater than a threshold value, the adsorption force is increased by a first predetermined value ΔP1, and if the difference Δ2 is less than the threshold value, the adsorption force is increased by a second predetermined value ΔP2 smaller than the first predetermined value ΔP1. According to this method, the adsorption force can be changed up to the maximum adsorption force, which increases the possibility of finding an appropriate adsorption force.

[0079] The second embodiment as described above can also achieve the same effects as the first embodiment described above.

[0080] <Third embodiment> FIG. 7 is a flowchart illustrating a control method for the robot system according to the third embodiment.

[0081] The robot system 1 according to this embodiment is similar to the robot system 1 according to the first embodiment, except that the adsorption condition update step S5 is different. In the following description, the robot system 1 according to this embodiment will be described with a focus on the differences from the first embodiment, and the description of the similarities will be omitted. In the drawings of this embodiment, the same reference numerals are used for the same configurations as those in the above-mentioned embodiment.

[0082] Hereinafter, the adsorption condition updating step S5 of the robot system 1 according to this embodiment will be described with reference to FIG.

[0083] <<Adsorption condition update step S5>> In the suction condition update step S5 of this embodiment, even if the judgment result in the judgment step S4 is success, the suction conditions are updated and the suction force is reduced. For example, even if the work itself is successful, the work may be successful with a smaller suction force. Therefore, as in this embodiment, even if the judgment result in the judgment step S4 is success, by updating the suction conditions and reducing the suction force, there is a possibility that the work may be successful with a smaller suction force. Therefore, power saving operation is possible, and damage to the workpiece W can be further reduced.

[0084] Specifically, if the determination result in the determination step S4 is successful, the robot control device 6 determines whether the suction force currently set in the database DB is the minimum suction force in step S581. If the suction force set in the database DB is not the minimum suction force, the robot control device 6 reduces the suction force stored in the database DB in step S582 and updates the result. Specifically, if the target workpiece W is the first workpiece W1, the suction force set in the first suction condition in the database DB is reduced, and the result is updated as a new first suction condition. Similarly, if the target workpiece W is the second workpiece W2, the suction force set in the second suction condition in the database DB is reduced, and the result is updated as a new second suction condition.

[0085] In this way, by updating the suction conditions even when the operation is successful, the operation can be completed with a suction force closer to the minimum suction force required for the operation to be successful. This enables power saving operation and further reduces damage to the workpiece W caused by suction.

[0086] As described above, in the robot system 1 of this embodiment, the robot control device 6 updates the first suction condition even when the work on the first workpiece W1 is successful, and updates the second suction condition even when the work on the second workpiece W2 is successful. Furthermore, the update of the first suction condition and the update of the second suction condition each include reducing the suction force. In this way, by updating the suction condition even when the work is successful, the work can be performed with a suction force that is closer to the minimum suction force required for the work to be successful. This makes it possible to drive with low power consumption, and further reduces damage to the workpiece W caused by suction.

[0087] The third embodiment as described above can also achieve the same effects as the first embodiment described above.

[0088] <Fourth embodiment> FIG. 8 is a flowchart illustrating a control method for the robot system according to the fourth embodiment.

[0089] The robot system 1 according to this embodiment is similar to the robot system 1 according to the first embodiment, except that the adsorption condition update step S5 is different. In the following description, the robot system 1 according to this embodiment will be described with a focus on the differences from the first embodiment, and the description of the similarities will be omitted. In the drawings of this embodiment, the same reference numerals are used for the same configurations as those in the above-mentioned embodiment.

[0090] Hereinafter, the adsorption condition updating step S5 of the robot system 1 according to this embodiment will be described with reference to FIG.

[0091] <<Adsorption condition update step S5>> In the adsorption condition update step S5 of this embodiment, if the judgment result in the judgment step S4 is failure, the robot control device 6 calculates the failure probability of the work performed under the same adsorption conditions in step S50, and judges whether the calculated failure probability is equal to or greater than a preset reference value. If the failure probability is equal to or greater than the reference value, the robot control device 6 proceeds to step S51. On the other hand, if the failure probability is less than the reference value, the robot control device 6 ends the adsorption condition update step S5 without updating the adsorption conditions. According to this method, it is possible to effectively prevent the adsorption conditions from being deteriorated due to the work accidentally failing due to other factors that occur accidentally or unexpectedly, even though the current adsorption conditions are appropriate.

[0092] The reference value is not particularly limited and can be set appropriately depending on the required work precision, etc., and can be set to about 80%, for example.

[0093] The fourth embodiment as described above can also achieve the same effects as the first embodiment described above.

[0094] The robot system and the control method for the robot system according to the present invention have been described above based on the illustrated embodiment, but the present invention is not limited to this. The configuration of each part can be replaced with any configuration having a similar function. In addition, any other configuration or process may be added to the present invention. [Explanation of symbols]

[0095] 1...robot system, 2...robot, 20...robot body, 21...base, 22...robot arm, 221...arm, 222...arm, 223...arm, 224...arm, 225...arm, 226...arm, 3...end effector, 30...suction head, 31...suction nozzle, 310...suction nozzle used, 4...sensor, 40...camera, 41...image recognition unit, 5...transport device, 50...belt conveyor, 51...belt, 6...robot control device, A...suction force, Amax...maximum suction force, Amin...minimum suction force, B...suction force, Bmax...maximum suction force, Bmin...minimum suction force, DB...database, J1...joint, J2...joint, J3...joint, J4...joint, J5...joint, J6...joint, Q 1...part, Q2...part, S1...identification step, S2...pickup condition determination step, S3...operation step, S4...judgment step, S5...pickup condition update step, S50...step, S51...step, S52...step, S53...step, S531...step, S532...step, S533...step, S54...step, S55...step, S551...step, S552...step, S553...step, S56...step, S57...step, S581...step, S582...step, W...workpiece, W1...first workpiece, W2...second workpiece, ΔM...predetermined value, ΔM1...first predetermined value, ΔM2...second predetermined value, ΔP...predetermined value, ΔP1...first predetermined value, ΔP2...second predetermined value

Claims

1. A sensor for identifying the workpiece; a robot having a robot arm and an end effector disposed on the robot arm, the robot adsorbing the workpiece with the end effector and performing a predetermined operation; A robot control device that controls the driving of the robot, The robot control device includes: When the sensor identifies the workpiece as a first workpiece, the robot is operated by a first drive command based on a first suction condition, and when the operation based on the first drive command fails, the first suction condition is updated; A robot system characterized in that, when the sensor identifies the workpiece as a second workpiece different from the first workpiece, the robot is operated with a second drive command based on a second suction condition, and when the operation with the second drive command fails, the second suction condition is updated.

2. the end effector is a suction head having a plurality of suction nozzles, updating the first suction condition includes changing the suction nozzle used to suction the first workpiece, The robot system according to claim 1 , wherein the update of the second pickup condition includes changing the pickup nozzle used to pick up the second workpiece.

3. The robot control device determines that the task has failed if the workpiece is damaged after the task, updating the first suction condition when the work in response to the first drive command has failed includes reducing an attraction force for the first workpiece; The robot system according to claim 1 , wherein updating the second suction condition when the task in response to the second drive command has failed includes reducing an attraction force for the second workpiece.

4. The robot system according to claim 3 , wherein the suction force is reduced by a predetermined value in each of the updating of the first suction condition and the updating of the second suction condition.

5. 4. The robot system according to claim 3, wherein, in updating the first attraction condition and updating the second attraction condition, if a difference between the current attraction force and a minimum attraction force is equal to or greater than a threshold value, the attraction force is reduced by a first predetermined value, and if the difference is less than the threshold value, the attraction force is reduced by a second predetermined value smaller than the first predetermined value.

6. The robot control device determines that the task has failed if the workpiece falls during the task, updating the first suction condition when the work in response to the first drive command has failed includes increasing an attraction force for the first workpiece; The robot system according to claim 1 , wherein updating the second suction condition when the task in response to the second drive command has failed includes increasing a suction force for the second workpiece.

7. The robot system according to claim 6 , wherein the suction force is increased by a predetermined value in each of the updating of the first suction condition and the updating of the second suction condition.

8. 7. The robot system according to claim 6, wherein, in updating the first attraction condition and updating the second attraction condition, if a difference between the current attraction force and a maximum attraction force is equal to or greater than a threshold value, the attraction force is increased by a first predetermined value, and if the difference is less than the threshold value, the attraction force is increased by a second predetermined value smaller than the first predetermined value.

9. the robot control device updates the first suction condition when the operation on the first workpiece is successful, and updates the second suction condition when the operation on the second workpiece is successful; The robot system according to claim 1 , wherein the updating of the first attraction condition and the updating of the second attraction condition each include decreasing the attraction force.

10. A control method for a robot system including a sensor for identifying a workpiece, a robot arm, and an end effector disposed on the robot arm, the end effector adsorbing the workpiece and performing a predetermined operation, comprising: When the sensor identifies the workpiece as a first workpiece, the robot is operated by a first drive command based on a first suction condition, and when the operation based on the first drive command fails, the first suction condition is updated; A control method for a robot system, characterized in that when the sensor identifies the workpiece as a second workpiece different from the first workpiece, the robot is operated with a second drive command based on a second suction condition, and when the operation with the second drive command fails, the second suction condition is updated.

Images