JP6248694B2 - Robot, robot system, and control device - Google Patents

Description

  The present invention relates to a robot, a robot system, and a control device.

A robot that controls the position and orientation of an object to be gripped by a gripping part (hand) of a robot arm that constitutes the robot by visual feedback control (also called visual servo) based on image data captured by a camera Has been developed (see, for example, Patent Document 1).
On the other hand, the position and force for setting the mechanical impedance (inertia, damping coefficient, rigidity) generated when external force is applied to the hand of the gripping part to values convenient for the intended work. A robot that performs impedance control, which is control of the above, is known.

JP 2003-211382 A

  The visual servo described above can correct a wide range of deviations, but the control speed is relatively low. On the other hand, the impedance control can correct a local shift near the control point at high speed. However, for example, in an operation in which both visual servo and impedance control can be used, it has been difficult to appropriately determine which one of the two types of control should be used.

  The present invention has been made in view of the above points. A robot, a robot system, and a control device that can perform more appropriate control in an operation in which either of two types of control can be used. The purpose is to provide.

One aspect of the present invention is made to solve the above-described problem, and includes a force sensor, an arm, and a control unit, and the control unit uses a captured image captured by the imaging unit. The robot operates the arm by statistically selected one of the first control and the second control using the output of the force sensor.
With this configuration, the control unit causes the arm to operate according to any one control statistically selected from the first control and the second control. Thereby, the robot can perform more appropriate control in the work in which either of the two types of control can be used.

According to another aspect of the present invention, in the above robot, the control unit is selected based on information indicating a failure number in the past operation result from the first control and the second control. The arm may be operated by any one control.
With this configuration, the robot operates the arm by any one of the two types of control selected based on information indicating the number of failures in the past operation results. As a result, the robot can select, for example, a method with a small number of failures out of the two methods of control, and thus can perform more appropriate arm control.

One embodiment of the present invention is the robot described above, wherein the control unit is selected based on a control time in a past operation result from the first control and the second control. The arm may be operated by one control.
With this configuration, the robot operates the arm by any one of the two types of control selected based on the control time in the past operation result. As a result, the robot can select a method having a high control speed, for example, from among the two methods of control, and thus can perform more appropriate arm control.

Further, according to one aspect of the present invention, in the robot described above, the control unit includes a control time in a plurality of operations tried in the past among the first control and the second control, and the plurality of times. The arm is operated by any one of the selected controls based on the number of successes of the operations.
With this configuration, the robot operates the arm by one of the two types of control selected based on the control time and the number of successes. As a result, the robot can select a method with good work efficiency that has a short cycle time obtained by dividing the total control time by the number of successes, and can perform more appropriate arm control.

Further, according to one aspect of the present invention, in the robot described above, the control unit selects a control with a small number of failures in a past operation result tried a plurality of times from the first control and the second control. When the first control and the second control have the same number of failures in the past operation result, the control in the past operation result of the first control and the second control. A control with a short time may be selected, and the arm may be operated by the selected control.
With this configuration, the robot selects one of the two types of control, which has a small number of failures or a short control time. Accordingly, the robot can select a method having a small number of failures or a short control time among the two methods of control, and thus can perform more appropriate arm control.

In one embodiment of the present invention, in the robot described above, the first control may be a visual servo, and the second control may be impedance control.
With this configuration, the robot operates the arm by any one control statistically selected from visual servo and impedance control. Thereby, the robot can perform more appropriate control in the work in which either of the two types of control of visual servo and impedance control can be used.

One embodiment of the present invention includes an imaging unit, a robot including a force sensor, and a control unit that operates the robot. The control unit uses a captured image captured by the imaging unit. In the robot system, the robot is operated by any one control statistically selected from the first control and the second control using the output of the force sensor.
With this configuration, the robot system operates the robot according to any one of the two types of control selected based on the control time in the past operation result. As a result, the robot system can perform more appropriate robot control in work in which either of the two types of control can be used.

One embodiment of the present invention is a control device that operates a robot including a force sensor, and includes a first control using a captured image captured by an imaging unit and a second control using an output of the force sensor. The control device operates the robot according to any one of statistically selected controls.
With this configuration, the control device causes the robot to operate according to any one of the two types of control selected based on the control time in the past operation result. Thus, more appropriate robot control can be performed in an operation in which either of the two methods of control can be used.

It is a figure which shows the schematic structural example of the robot by 1st Embodiment. It is a block diagram which shows the schematic structural example of the robot control apparatus by 1st Embodiment. It is a flowchart which shows an example of the trial operation | movement of the robot by 1st Embodiment. It is a flowchart which shows an example of selection operation | movement of the control system in 1st Embodiment. It is a figure which shows the schematic structural example of the robot system by 1st Embodiment. It is a flowchart which shows an example of the selection operation | movement of the control system in 2nd Embodiment. It is a figure which shows the schematic structural example of the robot system by 3rd Embodiment. It is a figure which shows the selection example of the control system of the robot system by 3rd Embodiment.

A robot and a robot system according to an embodiment of the present invention will be described with reference to the drawings.
[First Embodiment]
FIG. 1 is a diagram illustrating a schematic configuration example of a robot 10 according to the first embodiment.
As shown in FIG. 1, the robot 10 is a single-arm robot including one manipulator, and includes a gripping unit 11, a force sensor 12, an arm unit 13, and a support base 15, and is connected to the imaging unit 30. ing. Here, the gripping part 11 and the arm part 13 correspond to the arm 14 of the robot 10. The robot 10 also includes a robot control device 20.
In the present embodiment, a case of a single-arm robot will be described as an example.

The grip part 11 (hand) is provided at the tip of the arm part 13 via the force sensor 12. The gripping unit 11 has, for example, two fingers and can grip an object (target object WK) to be gripped.
The force sensor 12 (force sensor) is provided between the grip portion 11 and the arm portion 13. The force sensor 12 detects triaxial force and torque. Here, the triaxial force acts between the grip portion 11 and the arm portion 13. The torque is about the axis in the attachment direction of the grip portion 11.
The arm portion 13 is connected to the support base 15 so as to be capable of turning and bending.
The support base 15 is connected to the arm portion 13 and is fixed to, for example, a floor surface.

  The imaging unit 30 is configured using a camera and captures an image. In the present embodiment, the imaging unit 30 is disposed at a position where an image including the object WK can be captured. Note that the imaging unit 30 may be provided fixed to a floor surface, a ceiling, a wall surface, or the like. In addition, the imaging unit 30 may be configured to be able to change the imaging direction, the imaging angle, and the like manually by a person or automatically by an apparatus.

  The robot control device 20 controls the arm 14 (the gripping unit 11 and the arm unit 13). For example, the robot control device 20 is connected to the robot 10 and the imaging unit 30 so as to be able to transmit control signals and the like. This connection may be wired or wireless.

  FIG. 2 is a block diagram illustrating a schematic configuration example of the robot control apparatus 20 according to the present embodiment. The robot control device 20 (control device) includes an input unit 21, an output unit 22, a storage unit 23, and a control unit 24.

  The input unit 21 receives input from the outside. The input unit 21 includes, for example, a keyboard and a mouse that receive an operation input operated by a user. Further, the input unit 21 may be configured to include a function of receiving an input from an external device, for example.

  The output unit 22 performs output to the outside. The output unit 22 includes, for example, a screen (for example, a liquid crystal display) that displays various types of information as images to the user, a speaker that outputs audio information to the user, and the like. The output unit 22 may be configured to include a function of outputting information to an external device, for example.

  The storage unit 23 includes a storage medium that stores various types of information. The storage unit 23 stores, for example, information on programs used in the control unit 24 and information on parameters (for example, numerical values) used in various processes. The storage unit 23 stores, for example, target image information indicating a target image (goal image) to be described later, a work sequence that is information indicating a target in impedance control, history information at the time of trial, and the like.

The control unit 24 is a processor including, for example, a CPU (Central Processing Unit), and controls the robot control device 20 in an integrated manner. For example, the control unit 24 performs various processes for controlling the robot 10 (the arm 14 in the present embodiment).
The control unit 24 includes, for example, visual servo (first control) using a captured image captured by the imaging unit 30 and impedance control (second control) using the output of the force sensor 12. The arm 14 is operated by any one control statistically selected based on the history information indicating the operation results tried in the past.

Here, the visual servo is a control method for performing visual feedback control based on image data captured by the imaging unit 30. That is, the visual servo is feedback control based on a captured image captured by the imaging unit 30.
Impedance control is to set the mechanical impedance (inertia, damping coefficient, rigidity) generated when a force is applied from the outside to the hand of the gripping part 11 to a value convenient for the intended work. This is a control method for controlling position and force. That is, the impedance control is feedback control based on the output of the force sensor 12.

Specifically, for example, the control unit 24 selects one of the two control methods described above based on the number of failures and the control time in the past operation result, and the robot 10 (this In the embodiment, the arm 14) is controlled. Here, the control time is, for example, an average control time that is an average of the control times. Note that the control unit 24 calculates the number of failures and the control time based on the history information stored in the storage unit 23.
In addition, the control unit 24 performs control to try both visual servo and impedance control for the work performed by the robot 10 in order to accumulate history information used for selecting a control method. For example, the control unit 24 causes the work to be performed a predetermined number of times (for example, 10 times) by visual servo and impedance control, and causes each storage unit 23 to store each trial result as history information. The control unit 24 associates, for example, identification information (work identification information) for identifying the work, a control time taken in the trial, and success / failure information indicating whether the work has succeeded (or failed), and records history information. Is stored in the storage unit 23. Here, “failure” indicates that the operation of the robot 10 cannot reach the target (goal) in the work, and “success” indicates that the operation of the robot 10 has reached the target (goal). .
The control unit 24 includes a target image acquisition unit 241, a visual servo control unit 242, a target impedance setting unit 243, an impedance control unit 244, and a control selection unit 245.

  The target image acquisition unit 241 acquires target image information used for visual servo, for example, from the storage unit 23, and outputs the acquired target image information to the visual servo control unit 242. Here, the target image information is information of an image that becomes a target (goal) when controlling the arm 14 (the gripping unit 11 and the arm unit 13). As the target image (also referred to as a reference image or the like), for example, an image including the target object WK (target object) in the target position and orientation in the same frame as the frame of the image captured by the imaging unit 30 is used. be able to. Such target image information is prepared in advance and stored in the storage unit 23, for example. In the target image, as background information other than the object WK, for example, background information that matches or is similar to the background information in the image captured by the imaging unit 30 may be used. Information (for example, information with no background) may be used.

  The visual servo control unit 242 performs visual servoing on the robot 10 (the gripping unit 11 and the arm) based on information on an image (captured image) captured by the imaging unit 30 and target image information acquired by the target image acquisition unit 241. Part 13). Specifically, the visual servo control unit 242 controls the robot such that the object WK gripped by the robot 10 approaches the target. That is, the visual servo control unit 242 controls the arm 14 so that the error between the target WK gripped by the arm 14 and the target becomes small.

  Here, as the target image, for example, a plurality of different target images may be used, and these images are switched in order and used. As a specific example, the visual servo control unit 242 performs control so that the object WK in the captured image approaches the first target state from the initial state by visual servoing based on the first target image. After that, the visual servo control unit 242 performs control so that the object WK in the captured image approaches the second target state from the first target state by visual servoing based on the second target image. Thereafter, the visual servo control unit 242 uses the visual servo to change the object in the captured image from the second target state to the third target state (for example, the final target state) based on the third target image. It is possible to perform control such that control is performed so as to be close to. Although the case where three different target images are used is shown here, the number of target images may be arbitrary.

  The target impedance setting unit 243 acquires, for example, a work sequence stored in the storage unit 23, generates an instruction (target impedance) of an operation (target operation) at each timing, and uses the generated target impedance as an impedance control unit. To 244.

  The impedance control unit 244 performs impedance control so that the robot 10 performs an operation according to the target impedance output from the target impedance setting unit 243, and a plurality of servo motors (not shown) that drive each joint of the robot 10. To control. When performing this impedance control, the impedance control unit 244 uses the output (detection result) of the force sensor 12 as feedback. Specifically, the impedance control unit 244 controls the robot so that the output of the force sensor 12 approaches the target impedance.

  In order to perform impedance control, it is necessary to detect the force and moment applied to the end effector of the gripper 11, but the force and moment applied to the end effector of the robot 10 (the point that contacts the object WK) are detected. The method of doing is not limited to using the force sensor 12. For example, the force applied to the end effector may be estimated from the axial torque values of the grip portion 11 and the arm portion 13.

  The control selection unit 245 statistically selects any one of the visual servo and impedance control operations performed by the robot 10 based on history information that has been tried a plurality of times in the past. Here, “to select statistically” means to select using a statistic based on past operation results. This selection can be made, for example, by using a statistic (for example, the number of failures, the number of successes, the average control time, etc.) calculated based on history information that is a past operation result of a plurality of trials. One of these is selected. For example, the control selection unit 245, based on the history information stored in the storage unit 23, information indicating the number of failures of the visual servo and the impedance control (for example, the number of failures, the failure rate, the number of successes, the success rate, etc.) ) And the average control time are calculated as statistics. The control selection unit 245 selects one of the visual servo and the impedance control based on information indicating the calculated number of failures and the average control time.

  Specifically, the control unit 24 selects a control with a small number of failures (number of failures) from among visual servo and impedance control. In addition, when the number of failures between the visual servo and the impedance control is equal, the control unit 24 selects a control with a short average control time from the visual servo and the impedance control. Then, the control selection unit 245 operates the robot 10 (arm 14) by the selected control. That is, when the visual servo is selected, the control selection unit 245 operates the robot 10 (arm 14) with respect to the visual servo control unit 242, and when the impedance control is selected, the control selection unit 245 selects the visual control unit 244. Then, the robot 10 (arm 14) is operated.

Next, the operation of the robot 10 according to the present embodiment will be described with reference to the drawings.
FIG. 3 is a flowchart illustrating an example of the trial operation of the robot 10 according to the present embodiment.
Note that the processing shown in this figure is performed, for example, during a period before learning the work by the robot 10 and before actually performing the work by the robot 10. In the example shown in this figure, an example in which the number of trials is 10 for each of the visual servo and the impedance control will be described.

  In FIG. 3, the control unit 24 of the robot control device 20 tries a predetermined operation ten times by visual servo (step S101). That is, the controller 24 causes the visual servo controller 242 to cause the robot 10 (arm 14) to perform an operation of trying a predetermined work ten times. In this case, the visual servo control unit 242 applies a predetermined servo to the robot 10 by visual servoing based on the target image information acquired by the target image acquisition unit 241 from the storage unit 23 and information on the captured image captured by the imaging unit 30. Let the work run.

  Next, the control unit 24 stores the trial history by the visual servo in the storage unit 23 (step S102). In other words, the visual servo control unit 242 of the control unit 24 associates, for example, identification information for identifying the work, control time, and success / failure information of the work in each trial, and stores them in the storage unit 23 as history information.

  Next, the control unit 24 tries a predetermined operation ten times by impedance control (step S103). That is, the controller 24 causes the impedance controller 244 to cause the robot 10 (arm 14) to perform a predetermined operation for 10 times. In this case, the impedance control unit 244 causes the robot 10 to perform impedance control based on the target impedance information generated from the work sequence stored in the storage unit 23 by the target impedance setting unit 243 and the output of the force sensor 12. A predetermined work is executed.

  Next, the control unit 24 stores the trial history by the impedance control in the storage unit 23 (step S104). That is, the impedance control unit 244 of the control unit 24 associates, for example, identification information for identifying the work, control time, and success / failure information of the work in each trial, and stores them in the storage unit 23 as history information. The control unit 24 ends the trial operation process after the process of step S104.

Next, referring to FIG. 4, the control method selection operation by the robot 10 will be described.
FIG. 4 is a flowchart showing an example of a control method selection operation in the present embodiment.
4, first, the control unit 24 of the robot control device 20 calculates the number of failures NFV and the average control time TV based on the trial history by visual servo (step S201). That is, the control selection unit 245 of the control unit 24 uses the visual servo trial history (history information for 10 times) stored in the storage unit 23 and the number of failures NFV in 10 trials and the control of 10 times. An average control time TV, which is an average value of time, is calculated.

  Next, the control selection unit 245 calculates the number of failures NFI and the average control time TI based on the trial history by impedance control (step S202). That is, the control selection unit 245, based on the trial history (history information for 10 times) by impedance control stored in the storage unit 23, the number of failures NFI of 10 trials and the average value of 10 control times An average control time TI is calculated.

Next, the control selection unit 245 compares the number of failures NFV by visual servo with the number of failures NFI by impedance control, and branches the process according to the comparison result (step S203). When the number of failures NFI due to impedance control is smaller than the number of failures NFV due to visual servo (step S203: NFV> NFI), the control selection unit 245 advances the process to step S205 and operates the robot 10 by impedance control.
In addition, when the number of failure times NFV by visual servo is smaller than the number of failures NFI by impedance control (step S203: NFV <NFI), the control selection unit 245 advances the process to step S206 and operates the robot 10 by visual servo.
Further, when the visual servo failure frequency NFV is equal to the impedance control failure frequency NFI (step S203: NFV = NFI), the control selection unit 245 advances the process to step S204 and performs comparison based on the average control time.

  In step S204, the control selection unit 245 determines whether or not the average control time TI by impedance control is equal to or less than the average control time TV by visual servo. In other words, the control selection unit 245 compares the average control time TV by visual servo with the average control time TI by impedance control, and the average control time TI by impedance control is equal to or less than the average control time TV by visual servo ( In step S204: YES), the process proceeds to step S205. Further, when the average control time TI by impedance control is longer than the average control time TV by visual servo (step S204: NO), the control selection unit 245 advances the process to step S206.

In step S205, the control selection unit 245 operates the robot 10 by impedance control. That is, the control selection unit 245 causes the impedance control unit 244 to operate the robot 10.
In step S206, the control selection unit 245 operates the robot 10 by visual servo. That is, the control selection unit 245 causes the visual servo control unit 242 to operate the robot 10.

In the example described above, the control selection unit 245 selects any one of the visual servo and the impedance control based on the number of failures and the average control time. The control method may be selected based on one of the average control time. For example, in step S203 of FIG. 4 described above, the control selection unit 245 executes the process of step S204 when the number of failures NFV by visual servo is equal to the number of failures NFI by impedance control (step S203: NFV = NFI). Instead, the process may proceed to step S205. For example, the control selection unit 245 may not execute the process of step S203 of FIG.
Further, the control selection unit 245 may exchange the determination based on the number of failures with the determination based on the average control time. For example, the control selection unit 245 selects a control with a short average control time in the past operation result from visual servo) and impedance control, and selects a control with a small number of failures when the average control time is equal. Also good.

As described above, the robot 10 according to the present embodiment includes the force sensor 12 (an example of a force sensor), the arm 14, and the control unit 24. The control unit 24 performs first control (for example, visual servo) using the captured image captured by the imaging unit 30 and second control (for example, impedance control) using the output of the force sensor 12. The arm 14 is operated by any one of the statistically selected controls.
As a result, the robot 10 according to the present embodiment statistically determines which control should be used in an operation that can use either of the two types of control, and based on the determination result. Since control can be selected, more appropriate control can be performed. For example, the robot 10 according to the present embodiment can correct a wide range of deviation, but corrects a visual servo whose control speed is relatively low and a local deviation near the control point at high speed. It is possible to statistically determine which one of the impedance controls that can be used is better. That is, since the robot 10 according to the present embodiment switches the optimum control method for each work and operates the arm 14, more appropriate control can be performed.

In the present embodiment, the control unit 24 includes information indicating the number of failures in the past operation result (for example, the visual control) and the second control (for example, impedance control) (for example, the visual control). Based on the number of failures, the arm 14 is operated by any one of the selected controls.
Thereby, since the robot 10 according to the present embodiment can select, for example, a method with a small number of failures out of two methods of control, the robot 10 can perform more appropriate control of the arm 14.

In the present embodiment, the control unit 24 selects the first control (for example, visual servo) or the second control (for example, impedance control) based on the average control time in the past operation result. The arm 14 is operated by any one of the controls.
Thereby, since the robot 10 by this embodiment can select the method of short average control time among control of two methods, for example, it can perform more suitable control of the arm 14. FIG.

In the present embodiment, the control unit 24 has a small number of failures in the past operation results that have been tried a plurality of times between the first control (for example, visual servo) and the second control (for example, impedance control). Select control. When the number of failures in the past operation results of the first control and the second control is equal, the control unit 24 determines the average control time in the past operation results of the first control and the second control. Choose a short control. Then, the control unit 24 operates the arm 14 by the selected control.
As a result, the robot 10 according to the present embodiment can select a method having a small number of failures or a short average control time among the two methods of control, and thus can perform more appropriate control of the arm 14. .

Moreover, in this embodiment, the control part 24 operates the arm 14 by any one control statistically selected based on the historical information tried several times in the past among two control.
Thereby, since the robot 10 according to the present embodiment statistically selects appropriate control based on history information that has been tried a plurality of times in the past, it is possible to perform more appropriate control of the arm 14.

Also, the robot control device 20 according to the present embodiment is a control device that operates the robot 10 including the force sensor 12, and performs first control (for example, visual) using a captured image captured by the imaging unit 30. Of the second control (for example, impedance control) using the output of the servo) and the force sensor 12, the robot 10 is operated by any one control statistically selected.
Thereby, similarly to the robot system 1, the robot control apparatus 20 according to the present embodiment can more appropriately control the robot 10.

In addition, the robot 10 is good also as the robot system 1 provided with the robot control apparatus 20 outside as shown in FIG. In this case, the robot system 1 includes, for example, a robot 10, a robot control device 20, and an imaging unit 30. That is, the robot system 1 according to the present embodiment includes the imaging unit 30, the robot 10 including the force sensor 12, and the control unit 24 (robot control device 20) that operates the robot 10. The control unit 24 includes first control (for example, visual servo) using the captured image captured by the imaging unit 30 and second control (for example, impedance control) using the output of the force sensor 12. Of these, the robot 10 is operated by any one of the statistically selected controls.
Thereby, the robot system 1 by this embodiment can control the robot 10 more appropriately.

[Second Embodiment]
Next, a robot 10 according to a second embodiment will be described with reference to the drawings.
In the first embodiment described above, an example in which the control unit 24 selects a control method based on the number of failures and the average control time has been described. However, in the present embodiment, the control unit 24 is an index of work efficiency. An example of selecting a control method based on the average cycle time will be described.
Note that the configuration of the robot 10 according to the present embodiment is the same as that of the first embodiment shown in FIGS.

  In the present embodiment, the control selection unit 245 of the control unit 24 selects any one of the visual servo and the impedance control based on the average cycle time. The control selection unit 245 calculates the average cycle time (CT) based on the history information stored in the storage unit 23 as shown in the following equation (1).

Here, the total control time indicates the total control time in a plurality of attempted operations (for example, 10 times) in the history information. The total control time includes the control time taken when the failure occurs. The success count indicates the number of successes in a plurality of trials. That is, the average cycle time (CT) is obtained by dividing the total control time in a plurality of operations tried in the past by the number of successes in the plurality of operations. Note that (success rate × number of trials) may be used instead of the number of successes.
Thus, the average cycle time becomes shorter as the control time is shorter, and becomes shorter as the number of successes increases. The control selection unit 245 calculates an average cycle time CTV based on visual servo trial history (history information) and an average cycle time CTI based on impedance control trial history (history information), respectively, and the calculated average cycle time is shorter. Select the control.

Next, the operation of the robot 10 according to the present embodiment will be described with reference to the drawings.
The trial operation of the robot 10 according to this embodiment is the same as that of the first embodiment shown in FIG.

Here, the control method selection operation according to the present embodiment will be described with reference to FIG.
FIG. 6 is a flowchart illustrating an example of a control method selection operation in the present embodiment.
In FIG. 6, first, the control unit 24 of the robot control device 20 calculates the average cycle time CTV based on the trial history by the visual servo (step S301). That is, the control selection unit 245 of the control unit 24 calculates the average cycle time CTV using the above-described equation (1) based on the trial history (the history information for 10 times) by the visual servo stored in the storage unit 23. To do.

  Next, the control selection unit 245 calculates the average cycle time CTI based on the trial history by impedance control (step S302). That is, the control selection unit 245 calculates the average cycle time CTI using the above-described equation (1) based on the trial history (history information for 10 times) by the impedance control stored in the storage unit 23.

  Next, the control selection unit 245 determines whether or not the average cycle time CTI by the impedance control is equal to or less than the average cycle time CTV by the visual servo (step S303). That is, the control selection unit 245 compares the average cycle time CTV by visual servo with the average cycle time CTI by impedance control, and the average cycle time CTI by impedance control is equal to or less than the average cycle time CTV by visual servo ( In step S303: YES), the process proceeds to step S304. Further, when the average cycle time CTI by impedance control is longer than the average cycle time CTV by visual servo (step S303: NO), the control selection unit 245 advances the process to step S305.

In step S304, the control selection unit 245 operates the robot 10 by impedance control. That is, the control selection unit 245 causes the impedance control unit 244 to operate the robot 10.
In step S305, the control selection unit 245 operates the robot 10 by visual servo. That is, the control selection unit 245 causes the visual servo control unit 242 to operate the robot 10.

As described above, in the present embodiment, the control unit 24 performs a plurality of operations tried in the past among the first control (for example, visual servo) and the second control (for example, impedance control). Based on the control time and the number of successes among the plurality of operations, the arm 14 is operated by any one of the selected controls. That is, the control unit 24 calculates an average cycle in which the total control time (for example, total control time) in a plurality of operations tried in the past is divided by the number of successes in the plurality of operations among visual servo and impedance control. Based on the time, the arm 14 is operated by any one of the selected controls.
As a result, the robot 10 according to the present embodiment can select a method with good work efficiency that has a short average cycle time, for example, among the two types of control, and thus can perform more appropriate arm control. Since the average cycle time is an index (statistic) that takes into consideration both the work time and the number of failures (or the number of successes), the robot 10 according to the present embodiment can be simplified by using the average cycle time. With this technique, it is possible to select an appropriate method considering both the work time and the number of failures (or the number of successes).

[Third Embodiment]
Next, a robot 10a and a robot system 1a according to a third embodiment will be described with reference to the drawings.
In the first and second embodiments described above, an example in the case of a single-arm robot provided with one manipulator has been described, but the present invention can also be applied to a robot provided with two or more manipulators. In this embodiment, an example in the case of applying to a robot system 1a including a dual-arm robot will be described.

FIG. 7 is a diagram illustrating a schematic configuration example of the robot system 1a according to the present embodiment.
FIG. 7 is a front view showing the appearance of the robot system 1a. However, the control device 20 is shown for convenience, and is provided inside the main body of the robot 10a.
The robot system 1a shown in this figure includes a storage unit 16, body members 17 to 19, a left manipulator 14A (first manipulator), a right manipulator 14B (second manipulator), and a left wheel 41. The right wheel 42, the imaging unit 30a, and the robot control device 20a are provided. The manipulator 14A has a force sensor 12A and a gripping part 11A, and the manipulator 14B has a force sensor 12B and a gripping part 11B.

Here, the robot system 1a according to the present embodiment includes, as an example, an imaging unit 30a, a robot control device 20a, and a robot 10a (consisting of other components). Note that the imaging unit 30a can be regarded as being integrally included in the robot 10a.
In the present embodiment, the robot 10a is a double-arm robot provided with two manipulators (14A, 14B) each constituting an arm (arm). The robot 10a is configured integrally with the robot control device 20a, and its operation is controlled by a control signal input from the robot control device 20a.

On the upper surface of the storage unit 16, a body member 17, a body member 18, and a body member 19 are sequentially attached to the upper side, and a manipulator 14A that constitutes a left arm is attached to the left side of the uppermost body member 19, A manipulator 14B that constitutes a right arm is attached to the right side of the body member 19, a wheel 41 is attached to the left side of the bottom surface of the storage unit 16, and a wheel 42 is attached to the right side of the bottom surface of the storage unit 16, An imaging unit 30 a is attached to the upper surface of the storage unit 16.
Note that the robot 10a according to the present embodiment can be moved by rotating the left wheel 41 and the right wheel 42 by applying an external force manually or automatically by the apparatus.
The storage unit 16 stores a robot control device 20a.

Each of the manipulators (14A, 14B) is a kind of vertical articulated robot and functions as a robot arm. Each manipulator (14A, 14B) has a gripping portion (11A, 11B) at its tip.
The degrees of freedom of the respective manipulators (14A, 14B) may be arbitrary degrees of freedom, for example, may be 3-axis degrees of freedom, 6-axis degrees of freedom, 7 It may be a degree of freedom of the axis, or another degree of freedom.

In addition, each grip part (11A, 11B) grips an object (target object WK) to be gripped.
In the present embodiment, by operating the manipulators (14A, 14B), the object WK grasped by the grasping portions (11A, 11B) can be moved, that is, the position and posture of the object WK can be changed. Is possible.

The imaging unit 30a is configured using a camera and captures an image. In the present embodiment, the imaging unit 30a is provided in an arrangement capable of capturing an image including the target object WK gripped by the gripping units (11A, 11B).
In the present embodiment, the imaging unit 30a is provided on the upper surface of the storage unit 16. However, as another configuration example, the imaging unit 30a may be provided in any other location, for example, other in the robot 10a. May be provided, or may be provided fixed to a floor surface, a ceiling, a wall surface, or the like as a place other than the robot 10a.
Further, the imaging unit 30a may be configured to be able to change the imaging direction, the imaging angle, and the like by applying an external force manually or automatically by the apparatus.

  The robot control device 20a controls the left arm manipulator 14A and the right arm manipulator 14B. That is, as in the first embodiment, the robot controller 20a controls the robot 10a (the left arm manipulator 14A and the right arm) by controlling any one of the visual servo and the impedance control statistically selected. The manipulator 14B) is operated.

  The configuration and operation of the robot control device 20a in this embodiment are the same as those of the robot control device 20 in the first and second embodiments, except that the two manipulators (14A, 14B) are controlled. Therefore, the description is omitted here. When controlling the manipulator (14A, 14B) by impedance control, the robot control device 20a in the present embodiment controls the manipulator 14A based on the output of the force sensor 12A, and the manipulator based on the output of the force sensor 12B. 14B is controlled. Moreover, when controlling the manipulator (14A, 14B) by visual servo, the robot controller 20a controls the manipulator (14A, 14B) based on the captured image captured by the imaging unit 30a.

As described above, the robot system 1a in the present embodiment is a double-armed robot 10a including two manipulators (14A and 14B). The robot system 1a includes force sensors (12A, 12B), manipulators (14A, 14B) (an example of an arm), a robot control device 20a, and an imaging unit 30a. The robot control device 20a operates the robot 10a (the left arm manipulator 14A and the right arm manipulator 14B) by any one control selected statistically from visual servo and impedance control.
Thereby, even if the robot system 1a in this embodiment is the two-arm robot 10a, it can control the robot 10a more appropriately similarly to the 1st and 2nd embodiment.

In this embodiment, the robot controller 20a can also control the left arm manipulator 14A and the right arm manipulator 14B separately. That is, the robot control device 20a can be controlled by a combination as shown in FIG. 8, for example, a method of statistically selecting in the first and second embodiments from among the four control combinations shown in FIG. Any one combination may be selected using. Here, the first manipulator shown in FIG. 8 corresponds to, for example, the left arm manipulator 14A, and the second manipulator corresponds to, for example, the right arm manipulator 14B.
In this way, by controlling the manipulators (14A, 14B) separately, the manipulators (14A, 14B) can be controlled by switching the two control methods separately. 10a can be controlled.

The present invention is not limited to the above embodiments, and can be modified without departing from the spirit of the present invention.
For example, in each of the embodiments described above, the embodiment in which each robot system 1 (1a) is implemented when the control methods of visual servo and impedance control are tried a plurality of times has been described. It is not something. For example, when there are a plurality of robot systems 1 (1a) of the same type, the robot system 1 (1a) is statistically controlled based on the history information of a plurality of units tried in each robot system 1 (1a). A method may be selected. In this case, the robot system 1 (1a) can select a control method more appropriately in consideration of variations among the robot systems 1 (1a). Also, for example, when trying with two robot systems 1 (1a), the number of trials with each robot system 1 (1a) should be reduced, for example, ten trials are usually made five times with each system. And the time required for the trial can be shortened.
When there are a plurality of robot systems 1 (1a), a representative robot system 1 (1a) is determined, and each robot system 1 ( 1a) may select the control method statistically.

Further, in each of the above embodiments, the example in which the information indicating the number of failures, the average control time, the average cycle time, and the like are applied as the statistic used for selection of the control method has been described. For example, power consumption may be used, and other statistics may be used. In addition to the number of failures, the information indicating the number of failures may be a failure rate, the number of successes, a success rate (yield), or the like.
Further, in each of the above-described embodiments, the example in which the control unit 24 calculates the above-described statistic when the control method is selected has been described. However, when the history information is acquired by performing a trial, the control unit 24 However, these statistics may be calculated and stored in the storage unit 23 in advance. In this case, the control unit 24 selects a control method based on a statistic stored in advance in the storage unit 23.

  Further, in each of the embodiments described above, the example in which the control unit 24 selects the control method every time when performing the work by the robot 10 (10a) has been described. The identification information and the determination result may be associated with each other and stored in the storage unit 23 in advance. In this case, the control unit 24 selects a control method based on the determination result stored in advance in the storage unit 23.

  Further, in each of the above embodiments, the case where one work is performed by one control method has been described. However, the present invention is not limited to this, and one work is divided into a plurality of periods and a plurality of work processes. The system 1 (1a) may select the control method for each period or each work process, and switch the control method to operate the robot 10 (10a). In addition, the robot system 1 (1a) may select the control method according to the position of the gripper 11 (11A, 11B), for example, and operate the robot 10 (10a) by switching the control method.

  Further, in each of the above embodiments, the robot system 1 (1a) has been described with an example in which impedance control is selected with priority, but visual servo may be prioritized. Further, the robot system 1 (1a) may change the priority control method according to the production plan or the like. The robot system 1 (1a) gives priority to impedance control when, for example, work delivery is imminent, and has a margin in the production plan, but wants to maintain the operating rate of the robot 10 (10a) For example, visual servo may be prioritized.

  Further, the robot system 1 (1a) may review the selection of the control method by periodically performing a trial, for example. In this case, even when the state of the robot system 1 (1a) is changed due to a change over time, the robot system 1 (1a) can appropriately operate the robot 10 (10a). The timing for reviewing the selection of the control method is not limited to this. For example, the control method may be reviewed when the number of failures increases or when the average control time in the work increases.

  Further, in each of the above embodiments, the case where the robot 10 (10a) is operated by switching between the two control methods of visual servo and impedance control has been described, but the robot system 1 (1a) is limited to this. It is not something. For example, the robot system 1 (1a) may operate the robot 10 (10a) by switching between three or more control methods. Further, the control method is not limited to visual servo and impedance control, and may be another control method such as position control, for example.

  Note that a program for realizing the function of an arbitrary component in the apparatus described above (for example, the robot controller 20 (20a)) is recorded on a computer-readable recording medium, and the program is stored in the computer system. You may make it read and execute. Here, the “computer system” includes hardware such as an OS (Operating System) and peripheral devices. “Computer-readable recording medium” means a portable disk such as a flexible disk, a magneto-optical disk, a ROM (Read Only Memory), a CD (Compact Disk) -ROM, or a hard disk built in a computer system. Refers to the device. Further, the “computer-readable recording medium” means a volatile memory (RAM: Random Access) inside a computer system that becomes a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. Memory that holds a program for a certain period of time, such as Memory).

In addition, the above program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.
Further, the above program may be for realizing a part of the functions described above. Further, the program may be a so-called difference file (difference program) that can realize the above-described functions in combination with a program already recorded in the computer system.

  In addition, some or all of the functions described above may be realized as an integrated circuit such as an LSI (Large Scale Integration). Each function described above may be individually implemented as a processor, or a part or all of the functions may be integrated into a processor. Further, the method of circuit integration is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. In addition, when an integrated circuit technology that replaces LSI appears due to the advancement of semiconductor technology, an integrated circuit based on the technology may be used.

DESCRIPTION OF SYMBOLS 1,1a ... Robot system 10, 10a ... Robot, 11, 11A, 11B ... Grasping part, 12, 12A, 12B ... Force sensor, 13 ... Arm part, 14 ... Arm, 14A, 14B ... Manipulator, 15 ... Support Table, 16 ... Storage unit, 17-19 ... Body part, 20, 20a ... Robot control device, 21 ... Input unit, 22 ... Output unit, 23 ... Storage unit, 24 ... Control unit, 30, 30a ... Imaging unit, 41 , 42 ... wheels, 241 ... target image acquisition unit, 242 ... visual servo control unit, 243 ... target impedance setting unit, 244 ... impedance control unit, 245 ... control selection unit, WK ... object

Claims (8)

A force sensor;
Arm,
A control unit,
The controller is
The arm is operated by statistically selected one of the first control using the captured image captured by the imaging unit and the second control using the output of the force sensor. ,
robot. The controller is
Based on the information indicating the number of failures in the past operation results among the first control and the second control, the arm is operated by any one selected control.
The robot according to claim 1. The controller is
Of the first control and the second control, based on the control time in the past operation result, the arm is operated by any one selected control,
The robot according to claim 1 or 2. The controller is
One of the first control and the second control selected based on the control time in a plurality of operations tried in the past and the number of successes in the plurality of operations. Operating the arm by control,
The robot according to claim 1. The controller is
Select a control with a small number of failures in the past operation results tried a plurality of times from the first control and the second control,
When the number of failures in the past operation results of the first control and the second control is equal, a control time in the past operation results of the first control and the second control. Choose a short control and
The arm is operated by the selected control.
The robot according to claim 1. The robot according to any one of claims 1 to 5, wherein the first control is a visual servo, and the second control is an impedance control. An imaging unit;
A robot with a force sensor;
A control unit for operating the robot,
The controller is
The robot is operated by statistically selected one of the first control using the captured image captured by the imaging unit and the second control using the output of the force sensor. Let
Robot system. A control device for operating a robot equipped with a force sensor,
The robot is operated by any one control statistically selected from the first control using the captured image captured by the imaging unit and the second control using the output of the force sensor.
Control device.

Images