TW202423638A - Robot system, and control device - Google Patents

Abstract

This robot system comprises: a robot; and an operation mode control unit that, on the basis of information pertaining to the continuation of work performed by the robot, switches the operation mode of the robot or another apparatus forming a robot system which includes the robot.

Description

Robotic systems and control devices

Invention Field

The present disclosure relates to a robot system and a control device.

Invention Background

It is known that a robot is set to a power saving mode to reduce the power consumption of the robot. Patent document 1 states: Regarding an interaction robot, it detects that a person has left and switches to a power saving mode when the person has left. In addition, patent document 2 states that the robot has a power saving mode. Prior art document Patent document

Patent document 1: Japanese Patent Publication No. 2016-95824 Patent document 2: Japanese Patent Publication No. 2019-18272

Invention Summary Problem to be solved by the invention

When constructing a system that can switch a robot to a power-saving mode, a general approach is to switch the robot to a power-saving mode when the robot does not perform an action for a fixed period of time. However, this approach has the following disadvantages: Since the time until entering the power-saving mode is fixed, power is still supplied to the robot even when the robot is not required to perform an action, which may cause unnecessary power consumption. It is desired to provide a technology that can suppress the occurrence of such unnecessary power consumption in a robot system that can switch to a power-saving mode. Means for solving the problem

One aspect of the present disclosure is a robot system comprising: a robot; and an action mode control unit, which switches the action mode of the robot or other machines constituting the robot system including the robot based on information related to the continuation of the robot's operation.

These objects, features and advantages of the present invention, as well as other objects, features and advantages, will become more apparent from the detailed description of typical embodiments of the present invention as shown in the accompanying drawings.

The form used to implement the invention

Next, the embodiments of the present disclosure are described with reference to the drawings. In the drawings, the same components or functional parts are given the same reference symbols. For ease of understanding, the proportions of these drawings are appropriately changed. In addition, the embodiments shown in the drawings are examples for implementing the present invention, and the present invention is not limited to the embodiments shown in the drawings.

First embodiment FIG. 1 is a diagram showing the machine structure of a robot system 100 of an embodiment. As shown in FIG. 1 , the robot system 100 includes: a robot 10, a robot control device 50 for controlling the robot 10, a conveying device 90 for conveying a workpiece, and a visual sensor 70 for photographing and inspecting the workpiece. The robot system 100 may further include a teaching operation panel 40 connected to the robot control device 50. The visual sensor 70 is fixed in the work space in a manner such that it can photograph the upstream side area of the conveying device 90, for example. Furthermore, the visual sensor 70 may also be mounted on the front end (terminator) of the arm of the robot 10. The visual sensor 70 is connected to the robot control device 50. The robot control device 50 is configured to be connected to the conveying device 90 and to obtain information (speed information, etc.) from various sensors (not shown) on the conveying device 90. With this configuration, the robot system 100 is configured to detect the workpiece W conveyed on the conveying device 90 by the visual sensor 70, and the robot 10 can follow the workpiece W conveyed on the conveying device 90 and perform handling such as taking out the workpiece W. The function of the robot following the workpiece and performing actions based on the detection of the visual sensor is also called visual tracking.

As described in detail below, the robot system 100 is configured to control the motion modes of the machines constituting the robot system including the robot 10 and the visual sensor 70 to reduce the unnecessary power consumption of the robot system.

The robot 10 is, for example, a vertical multi-joint robot. Furthermore, as the robot 10, other types of robots such as a horizontal multi-joint robot, a parallel link robot, a two-arm robot, etc. may also be used depending on the operation object.

The robot 10 can perform the required operation by means of a terminal mounted on the wrist. In this embodiment, a hand 11 is used as the terminal. As the hand 11, various types of hands such as a multi-finger gripping hand and a suction hand can be used.

The visual sensor 70 may be, for example, a camera that takes grayscale images or color images, or a stereo camera or three-dimensional sensor that can obtain distance images or three-dimensional point clouds. Assume that the robot control device 50 is equipped with: a function of controlling the visual sensor 70; and a function of performing image processing such as detection processing on the photographed image of the visual sensor 70. The robot control device 50 maintains a model pattern of the workpiece, and can perform image processing to detect the workpiece by pattern matching the image of the workpiece in the photographed image with the model pattern. In the present embodiment, the visual sensor 70 has been calibrated, and the robot control device 50 retains calibration data that defines the relative position relationship between the visual sensor 70 and the robot 10. Thereby, the robot control device 50 can convert the position on the image captured by the visual sensor 70 into a position on a coordinate system (robot coordinate system, etc.) fixed in the work space.

Furthermore, although the function of controlling the visual sensor 70 to perform image processing is installed in the robot control device 50, an image processing device having the function of controlling the visual sensor 70 to perform image processing may be configured in the robot system as a device independent of the robot control device 50.

The robot control device 50 controls the actions of the robot 10 according to the action program or the instructions from the teaching operation panel 40. The robot control device 50 may also have a hardware configuration as a general computer, and the general computer has a processor 51 (FIG. 2), a memory (ROM, RAM, non-volatile memory, etc.), a storage device, an operation unit, an input/output interface, a network interface, etc.

The teaching operation panel 40 is used as an operation terminal for teaching or various settings of the robot 10. As the teaching operation panel 40, a teaching device constituted by a tablet terminal or the like may also be used. The teaching operation panel 40 may also have a hardware configuration as a general computer, wherein the general computer has a processor, a memory (ROM, RAM, non-volatile memory, etc.), a storage device, an operation unit, a display unit 41 (FIG. 2), an input/output interface, a network interface, etc.

FIG2 is a functional block diagram of the robot control device 50. As shown in FIG2, the robot control device 50 has an action control unit 151, a detection unit 152, and an action mode control unit 153. The robot control device 50 may further include a learning unit 154. These functional blocks may also be implemented by the processor 51 of the robot control device 50 executing software.

The robot control device 50 has a memory unit 155. The memory unit 155 is a memory device composed of, for example, a non-volatile memory or a hard disk device. The memory unit 155 stores: a robot program for controlling the robot 10, a program (visual program) for image processing such as workpiece detection based on an image captured by a camera, and other various setting information. The memory unit 155 may further store data required for image processing such as calibration data and workpiece model data.

The motion control unit 151 controls the motion of the robot according to the robot program or the command from the teaching operation panel 40. The robot control device 50 includes a servo control unit (not shown) which performs servo control on the servo motor of each axis according to the command for each axis generated by the motion control unit 151.

The detection unit 152 provides a function of detecting the workpiece W based on the image captured by the visual sensor 70. The detection unit 152 can detect the workpiece W conveyed on the conveying device 90 and provide its detection position to the motion control unit 151. In addition, the motion control unit 151 can obtain information such as the conveying speed of the conveying device 90 from various sensors (pulse encoder, etc.) on the conveying device 90. With this function, the motion control unit 151 can perform an action of taking out the workpiece W while following the movement of the workpiece W based on the detection result of the workpiece W by the visual sensor 70.

The action mode control unit 153 can control the action mode of the robot 10 or the visual sensor 70 based on the detection result of the detection unit 152. The control of the action mode performed by the action mode control unit 153 includes: (A) control of switching the action mode of the robot 10; and (B) control of switching the action mode of the visual sensor 70. In this way, in addition to controlling the action mode of the robot 10, the action mode control unit 153 can also control the action mode of the visual sensor 70, thereby reducing the redundant power consumption of the entire robot system 100.

The function (A) of controlling the action mode of the robot 10 by the action mode control unit 153 is described. The control of the action mode of the robot 10 by the action mode control unit 153 includes the following. (A1) The function of switching the robot from the normal mode to the energy saving mode. (A2) The function of switching the robot from the energy saving mode to the normal action mode.

The above function (A1) is explained with reference to FIG3 . When the robot 10 is in the general action mode, the detection unit 152 performs imaging at a certain time interval (the first imaging interval T ₁ ) to detect the workpiece W. Here, it is assumed that the workpiece W is a group of workpieces G as shown in FIG3 and arrives at irregular time intervals. That is, the workpiece group G is sometimes transported at approximately certain time intervals, and sometimes at random time intervals. In this case, as an example, if it is assumed that the minimum time interval for the arrival of the workpiece group G is INT, the action mode control unit 153 can also set the first imaging interval T ₁ for the visual sensor 70 to be shorter than INT (to be T ₁ <INT). This time interval INT can also be set based on the actual measurement value. The time interval INT can also be set in the robot control device 50 by the user operating through the teaching operation panel 40.

When the visual sensor 70 takes images at a certain time interval (the first imaging interval T ₁ ), for example, when the workpiece is not detected for a certain number of times, the action mode control unit 153 causes the robot 10 to shift to an action mode that reduces power consumption. In this specification, an action mode that reduces power consumption of the robot 10 is referred to as an energy-saving mode of the robot. For example, in the energy-saving mode of the robot 10, power is stopped from being supplied to the servo motors of the joints of the robot, and the joints of the robot are braked to maintain the posture of the robot. Furthermore, the energy-saving mode of the robot may be supplemented with a control that reduces power consumption of machines related to controlling the robot (teaching operation panel or robot control device). For example, a control may be added to turn off the backlight of the display of the teaching operation panel 40 in the power saving mode of the robot.

Description of function (A2) (function for shifting the robot from the energy saving mode to the general action mode). Even after shifting the robot 10 to the energy saving mode, the robot control device 50 still causes the visual sensor 70 to perform imaging and detection at a certain time interval (the second imaging interval T ₂ ). Furthermore, the setting of the second imaging interval T ₂ of the robot 10 in the energy saving mode will be described later. Assuming that the robot 10 is in the energy saving mode, the workpiece group G arrives as shown in FIG. 4 , and the workpiece W is detected by the visual sensor 70 (detection unit 152). In this case, the action mode control unit 153 returns the robot 10 to the general action mode and starts the removal operation of the workpiece W.

The function (B) of controlling the action mode of the visual sensor 70 by the action mode control unit 153 is described. When the state in which the workpiece W is not detected appears and the robot 10 enters the energy saving mode as described above, it is assumed that the state in which the workpiece W is not detected will continue for a certain period of time. The action mode control unit 153 has the following function: after the robot 10 is transferred to the energy saving mode, the imaging interval of the visual sensor 70 is changed to the second imaging interval T ₂ . In this way, after the robot 10 enters the energy saving mode, the imaging interval is made longer than in the normal action mode to reduce the power consumption of the visual sensor 70, so that the redundant power consumption of the robot system 100 is further reduced. In this way, the action mode control unit 153 not only has the function of shifting the robot 10 to the energy saving mode, but also has the function of shifting the visual sensor 70 to an action mode that reduces power consumption.

The action mode control unit 153 has the following functions as a function for setting the second imaging interval _T2 after the robot 10 shifts to the power saving mode: (C1) accepting input from the outside (user input or input from an external device); and (C2) obtaining the second imaging interval _T2 by learning.

The function of accepting the user input of the second imaging interval _T2 can also be realized in the form of accepting user operation through the operation unit of the teaching operation panel 40.

The function of learning the second imaging interval _T2 is performed by the learning unit 154. In the case where the workpiece is not detected and the mode is switched to the energy saving mode, the state of not detecting the workpiece should continue for a certain period of time. In addition, it is assumed that the duration of the state of not detecting the workpiece will vary depending on, for example, the time period. For example, in a day (24 hours), the time interval for the workpiece to be transported may become longer or shorter in a certain time period. For example, it is also assumed that the following situation may occur: around noon, the time interval for the workpiece to be transported becomes longer, and at a certain time period in the morning or afternoon, the time interval for the workpiece to be transported becomes shorter.

Thus, based on the possible correlation between the start time of not detecting the workpiece and its duration, the learning unit 154 acquires the second imaging interval T ₂ after the robot shifts to the energy saving mode through learning. Specifically, the learning unit 154 performs mechanical learning (supervisory learning) using learning data 154b, which uses the start time of not detecting the workpiece as input data and the duration as answer data (teacher data). The learning unit 154 collects and stores the learning data 154b of the start time of not detecting the workpiece and its duration, for example, for several days. Then, the inference engine 154a is made to learn the learning data 154b to construct a learning model 154c. The inference engine 154a can be constructed by using a nonlinear regression model or a neural network (NN), for example.

When the learning model 154c is constructed, as shown in FIG5, the inference engine 154a can output an estimated value of the duration by inputting the start time of the undetected workpiece. The action mode control unit 153 can also use the duration estimated by the inference engine 154a as the setting value of the second imaging interval _T2 . Alternatively, the action mode control unit 153 can also adjust the duration based on the duration estimated by the inference engine 154a and set it as the second imaging time _T2 . For example, the time obtained by subtracting a predetermined time from the estimated duration can be set as the second imaging interval _T2 .

In this way, by adopting a learning method and estimating the imaging interval of the vision sensor after the robot switches to the power saving mode, the imaging interval of the vision sensor can be lengthened according to the actual conditions of the production equipment, thereby reducing the power consumption of the vision sensor.

Furthermore, when collecting learning data, the start time and the end time of the undetected workpiece can be saved in advance, and when learning is performed, the duration can be calculated as the difference between the start time and the end time. The learning method can also be as follows: while the robot system 100 is running, learning data is obtained and learning is performed to gradually construct the learning model 154c.

Furthermore, although the example of the start time of using the undetected workpiece is described here as the input data, since the workpieces of the assumed operation objects are various items ranging from food, medicine, daily necessities to mechanical parts, the input data that may be related to the duration of the energy saving mode (the duration of the state of undetected workpiece) may include information such as the date, day of the week, month, season, and meteorological conditions (including weather, temperature, and humidity) when the workpiece is not detected. Furthermore, the robot program or the internal state of the robot may also include information that may be related to the duration of the energy saving mode, such as the type or content of the robot's operation, the progress of the robot program, the status of various data or signals related to the program execution, the movement status of the robot or the work tool, etc. Therefore, one or more of these data may be learned together with the start time as input data, or may replace the start time as input data. Furthermore, it is also possible to imagine an event in which an external signal is used to force the robot to start up in the energy saving mode to perform an operation (for example, someone presses a button to force the robot to start up). This external signal may also become the main factor affecting the duration of the energy saving mode. Therefore, information related to such external signals can also be taken into learning as input data. In this way, the learning unit 154 can also use data related to the duration of the state in which the workpiece is not detected as input data for learning. Furthermore, when using input data as exemplified above, in order to efficiently learn useful data, the input data can also be appropriately pre-processed. The learning of the learning unit 154 can also introduce deep learning techniques.

Furthermore, in order to obtain the above input data, the robot control device 50 may also be connected to a sensor 62 or an external signal 63 as shown in FIG2 . In this case, the sensor 62 is, for example, a temperature sensor or a humidity sensor for obtaining weather conditions. The robot control device 50 may also obtain information about the start time, date, day of the week, month, and season from an internal timer.

Fig. 6 shows the above-mentioned action mode control processing of the robot system 100 as a flow chart. The control of Fig. 6 is performed under the control of the processor 51 of the robot control device 50. In the flow chart of Fig. 6, each process on the left side corresponds to the process in the general action mode, and each process on the right side corresponds to the process of the robot in the energy saving mode.

When the processing in the general operation mode is started, the processing of repeating the imaging and detection at the first imaging interval _T1 as described above is performed until the workpiece W is detected. Specifically, the visual sensor 70 performs imaging (step S1), and the detection unit 152 performs processing to detect the workpiece W based on the photographed image (step S2). Then, in step S3, as the detection result of the detection unit 152, it is determined whether the workpiece W is detected. When the workpiece W is detected (S3: Yes), the action control unit 151 performs the workpiece removal action (step S5).

When the workpiece W is not detected (S3: No), the action mode control unit 153 determines whether the number of times the workpiece is not detected is greater than a certain number (step S4). When the number of times the workpiece is not detected is less than a certain number (S4: No), the processing of imaging and detection (steps S1, S2) is repeated. That is, imaging and detection by the visual sensor 70 are repeated until the workpiece W is detected. The processor 51 controls to loop steps S1 to S4 (S4: No judgment) with the first imaging cycle _T1 (repetition of imaging and detection).

When the number of times the workpiece W is not detected exceeds a certain number (S4: Yes), the action mode control unit 153 causes the robot 10 to shift to the energy saving mode (step S6). Next, the action mode control unit 153 changes the imaging interval of the visual sensor 70 to the second imaging interval _T2 (step S7). In the case of setting the second imaging interval _T2 by user input (non-learning mode), the action mode control unit 153 sets the second imaging interval _T2 to the value set in advance by user input (step S8).

In the case of setting the second imaging interval _T2 by learning (learning mode), the action mode control unit 153 obtains the duration time by inputting the start time of not detecting the workpiece W into the inference engine 154a. Here, it is assumed that the learning is completed in advance and the learning model 154c is constructed. Then, the action mode control unit 153 sets the second imaging interval _T2 based on the estimated duration time (step S9).

Then, the imaging and detection (steps S10, S11, S12: No judgment loop) performed by the visual sensor 70 and the detection unit 152 are performed at the second imaging interval _T2 (steps S10, S11, S12: No). During the imaging and detection at the second imaging interval _T2 , when the workpiece W is detected (S12: Yes), the action mode control unit 153 returns the robot 10 to the normal action mode (step S13). Then, the robot 10 performs the removal action (step S5).

In the above-mentioned motion mode control processing, the following structure is adopted: in any mode of the robot's general motion mode and energy-saving mode, the workpiece can be photographed and detected by the visual sensor at a predetermined time interval, and the motion mode can be switched according to the detection result. In this way, the change of the workpiece condition can be quickly grasped, and the motion mode of the robot system can be transferred in a timely manner.

In the above-mentioned motion mode control processing, the following configuration is adopted: in the general motion mode of the robot, the visual sensor performs imaging and detection at predetermined time intervals, and when the workpiece is not detected for more than a predetermined number of times, the robot is switched to the energy saving mode. According to this configuration, the switch to the energy saving mode can be performed in a timely manner, and compared with the configuration in which the time to enter the energy saving mode is set to a fixed time, the time that the robot consumes excess power can be reduced.

Furthermore, if the operation mode is switched to the energy saving mode according to the above operation mode control process, the load of the visual sensor for imaging and detection can be reduced, and the power consumption can be reduced. In other words, the redundant power consumption can be reduced as much as possible according to the status of the robot system.

FIG. 7 is a flow chart showing a learning process in which the learning unit 154 collects learning data while performing learning. This process is executed under the control of the learning unit 154 (processor 51). This process is executed in parallel with the action mode control process of FIG. 6. Furthermore, in the stage before the learning is completed, the second imaging interval _T2 of the action mode control process of FIG. 6 can also be set based on the user input.

When this process is started, the learning unit 154 checks whether the robot 10 has shifted from the normal operation mode to the energy saving mode (step S101). The learning unit 154 waits until the robot 10 shifts from the normal operation mode to the energy saving mode (S101: No). When the robot 10 shifts to the energy saving mode (S101: Yes), the learning unit 154 records the start time t1 when the workpiece W is not detected (step S102).

Then, the learning unit 154 checks whether the transition of the robot 10 from the energy saving mode to the general action mode has occurred (step S103). The learning unit 154 waits until the transition of the robot 10 from the energy saving mode to the general action mode occurs (S103: No). When the transition of the robot 10 from the energy saving mode to the general action mode is detected (S103: Yes), the learning unit 154 records the time t2 at that time (that is, the end time of the state where the workpiece is not detected) (step S104). Then, the learning unit 154 calculates the undetected continuation time by t2-t1, and causes the inference engine 154a to learn the undetected start time t1 and its continuation time (t2-t1) (step S105). Then, the process returns to step S101.

The learning unit 154 may also be configured to terminate the learning process of FIG. 7 after a predetermined period (e.g., several days) has passed after the transfer of the learning process of FIG. 7 or when a sufficient amount of learning data has been learned.

Furthermore, although FIG. 7 shows an operation example in which learning is performed step by step while learning data is collected, the learning of the inference engine 154a may be performed after a sufficient amount of learning data is collected.

Second embodiment The following describes a robot system 100A of the second embodiment. FIG. 8 is a diagram showing the machine configuration of the robot system 100A and the functional blocks of the robot control device 50 of the second embodiment. In this embodiment, the robot 10A is provided with an external force detector 61 for detecting an external force applied to the robot 10A. In this configuration, the robot 10A can be configured as, for example, a cooperative robot. Furthermore, in the second embodiment, the conveying device 90 or the visual sensor 70 can also be omitted.

The external force detector 61 is, for example, a force sensor mounted on the wrist flange of the robot 10A. In this case, the force sensor may also be, for example, a 6-axis force sensor, which detects forces in the directions of the X, Y, and Z axes, and torques around the axes. Alternatively, the external force detector 61 may also be constituted by torque sensors disposed at the joints of the robot 10A.

Furthermore, the robot control device 50 has a contact detection unit 156, which detects whether a person is in contact with the robot 10A based on the external force detected by the external force detector 61. The contact detection unit 156 can determine that the person and the robot 10A are in contact (i.e., they are working in a collaborative manner) when the magnitude of the external force detected by the external force detector 61 is greater than a predetermined threshold, for example.

In the present embodiment, the action mode control unit 153A provides a function of switching the action mode of the robot 10A by using the output from the external force detector 61 as information related to the continuation of the operation of the robot 10A.

As a specific example of operation, assume that the robot control device 50 is executing a program for performing collaborative work between the robot 10A and a human. In this case, the operation mode control unit 153A is: (k1) based on the detection result of the contact detection unit 156, when the time when the human does not contact the robot 10A becomes longer than a predetermined time, the robot 10A is shifted to the energy saving mode.

In the collaborative work between humans and robots, the state where humans are not in contact with the robot can be regarded as a state where no work is requested from the robot. If the above procedure (k1) is followed, it is possible to accurately grasp the occurrence of such a state that no work is requested from the robot, and the robot can be quickly transferred to the energy saving mode. Therefore, it is possible to prevent the robot from consuming power unnecessarily.

Then, the action mode control unit 153A may return the robot 10A to the normal action mode when the contact detection unit 156 detects again that a person has touched the robot 10A.

Furthermore, in the above-mentioned second embodiment, although the output of the external force detector 61 is used to detect whether a person and a robot are in contact, there may also be an example of a structure in which a mechanical switch installed on the arm of the robot 10A is used to detect whether a person and a robot are in contact.

As described above, when the robot 10A is in the energy saving mode, the robot 10A will start when the contact detection unit 156 detects that a person is in contact with the robot 10A. In this case, the event that the contact detection unit 156 detects that the person is in contact with the robot 10A may become the main factor that affects the duration of the energy saving mode. Therefore, the output signal of the contact detection unit 156 can also be taken into the learning of the learning unit 154 as input data that affects the duration of the energy saving mode.

Furthermore, the above describes an example of changing the action mode of the robot 10A in response to the fact that the robot 10A has not detected contact with a person for a certain period of time when the robot 10A is operated as a cooperative robot. The robot 10A can also be operated as a robot that performs force control (precision fitting operations, etc.) based on the output of the external force detector 61. In addition, when the robot 10A is configured to carry a container-type hand device as a terminator, and the aforementioned container-type hand device has the function of serving as a container for transporting workpieces, the external force detector 61 can be used to detect the introduction of the workpiece into the container-type hand device by the change in weight (that is, the change in the force acting on the robot 10A). In such cases, if the change of the external force on the robot 10A is not detected for a certain period of time, it can be considered that the robot 10A is not required to perform any operation. Therefore, the action mode control unit 153A can also perform the following control: when the change of the external force on the robot 10A is not detected for a certain period of time, the action mode of the robot 10A is transferred to the energy saving mode. By performing such control, the same effect as the above situation can be obtained.

The function allocation of the robot control device 50 shown in Fig. 2 and Fig. 8 is an example, and various modifications can be made to the function allocation. For example, there may also be a configuration example in which a part of the function block arranged in the robot control device 50 is arranged on the teaching operation panel 40 side.

The functions of the teaching operation panel 40 and the functions of the robot control device 50 may be collectively positioned as a robot control device.

The configuration of the above-mentioned embodiment can be applied to controlling the motion mode of industrial machinery, which is various types of industrial machinery that performs operations on workpieces by detecting workpieces with visual sensors.

The functional blocks of the robot control device shown in FIG. 2 and FIG. 8 may also be implemented by the processor of the robot control device executing various software stored in a memory device, or may also be implemented by a configuration mainly based on hardware such as an ASIC (Application Specific Integrated Circuit).

The programs for executing various processing such as the action mode control processing (FIG. 6) and the learning processing (FIG. 7) of the above-mentioned implementation form can be recorded in various computer-readable recording media (e.g. semiconductor memories such as ROM, EEPROM, flash memory, magnetic recording media, CD-ROM, DVD-ROM and other optical discs).

As described above, according to each embodiment, the change of the status related to the operation of the robot can be accurately grasped, and the action mode of the robot system can be transferred in a timely manner. That is, the action mode of the robot or other machines constituting the robot system can be accurately switched based on the information related to the continuation of the robot's operation. In addition, in this way, the occurrence of unnecessary power consumption in the robot system can be suppressed.

Although the present disclosure is described in detail, the present disclosure is not limited to the above-mentioned embodiments. These embodiments may be subject to various additions, replacements, changes, partial deletions, etc., without departing from the gist of the present disclosure or the intent of the present disclosure derived from the contents recorded in the scope of the patent application and its equivalents. Furthermore, these embodiments may also be combined and implemented. For example, in the above-mentioned embodiments, the order of each action or the order of each processing is represented as an example and is not limited to such order. Furthermore, the same applies to the use of numerical values or formulas in the description of the above-mentioned embodiments.

The following notes are further described with respect to the above-mentioned embodiments and variations. (Note 1) A robot system (100) comprises: a robot (10); and an action mode control unit (153) for switching the action mode of the robot (10) or other machines constituting the robot system (100) including the robot (10) based on information related to the continuation of the operation of the robot (10). (Note 2) The robot system (100) as described in Note 1 is provided with a sensor, wherein the sensor is used to detect a condition related to the continuation of the operation of the robot, and the action mode control unit (153) switches the action mode of the robot (10) or other machines constituting the robot system (100) based on the detection result of the sensor. (Note 3) The robot system (100) as described in Note 2, wherein the sensor is a visual sensor (70) capable of photographing a workpiece at a predetermined photographing interval, the robot system further comprises a detection unit (152), the detection unit (152) detects the workpiece based on the photographed image of the visual sensor (70), and the action mode control unit (153) switches the action mode of the robot (10) or the visual sensor (70) based on the detection result of the detection unit (152). (Note 4) The robot system (100) as described in Note 3, wherein the action mode control unit (153) causes the robot (10) to shift to a power saving mode that can reduce power consumption in response to the detection result that the workpiece is not detected. (Note 5) The robot system (100) as described in Note 4, wherein the action mode control unit (153) causes the robot (10) to shift to the power saving mode in response to the detection result that the workpiece is not detected a predetermined number of times in the detection performed by the detection unit (152) on the photographic image obtained at the predetermined photographic interval. (Note 6) The robot system (100) as described in Note 4 or 5, wherein the action mode control unit (153) causes the robot (10) to shift from the energy saving mode to the normal action mode in response to the detection result of the workpiece being detected. (Note 7) The robot system (100) as described in any one of Notes 3 to 6, wherein the action mode control unit (153) changes the imaging interval of the visual sensor (70) in response to the detection result of the workpiece not being detected. (Note 8) The robot system (100) as described in Note 7, wherein the action mode control unit (153) changes the imaging interval of the visual sensor (70) to the second imaging interval in response to the detection unit (152) not detecting the workpiece a predetermined number of times in the detection of the image of the workpiece taken by the visual sensor (70) at the first imaging interval. (Note 9) The robot system (100) as described in Note 8, wherein the action mode control unit (153) accepts user input for setting the second imaging interval. (Note 10) The robot system (100) as described in Note 8 further comprises a learning unit (154), wherein the learning unit (154) learns based on input data related to the duration of the state in which the workpiece is not detected, thereby setting the second imaging interval. (Note 11) As described in Note 10, the robot system (100), wherein the learning unit (154) constructs a learning model by learning learning data, wherein the learning data is at least one of the start time, date, day of the week, month, season, weather condition, sensor input, external signal, robot program, and internal state of the robot when the workpiece is not detected as input data, and the continuation time after the workpiece is not detected as teacher data. (Note 12) A robot system (100) as described in Note 11, wherein the learning unit (154) inputs at least one of the start time, date, day of the week, month, season, weather conditions, sensor input, external signal, robot program, and internal state of the robot when the workpiece is not detected to the learning model that has been learned, thereby obtaining an estimated value of the duration after the workpiece is not detected, and setting the second imaging interval based on the estimated value. (Note 13) The robot system (100) as described in Note 11, wherein the learning unit (154) records at least one of the start time, date, day of the week, month, season, weather condition, sensor input, external signal, robot program, and internal state of the robot when the workpiece is not detected during operation of the robot system (100) for a predetermined period, and the end time or duration of the state in which the workpiece is not detected, thereby obtaining the learning data. (Note 14) The robot system (100) as described in any one of Notes 8 to 13, wherein the second imaging interval is longer than the first imaging interval. (Note 15) The robot system (100A) as described in Note 2, wherein the sensor is a sensor (61) for detecting a force acting on the robot, and the action mode control unit (153A) switches the robot (10A) to a power saving mode when no change in the force acting on the robot (10A) is detected for a certain period of time. (Note 16) The robot system (100A) as described in Note 15, wherein the action mode control unit (153A) switches the robot (10A) to a power saving mode when no contact of an object with the robot (10A) is detected for a certain period of time based on the output of the sensor (61). (Note 17) A control device (50) is a control device (50) for controlling a robot (10), wherein the control device (50) has an action mode control unit (153), and the action mode control unit (153) switches the action mode of the robot (10) or other machines constituting a robot system (100) including the robot (10) based on information related to the continuation of the operation of the robot (10).

10,10A: robot 11: hand 40: teaching operation panel 41: display unit 50: robot control device 51: processor 61: external force detector 62: sensor 63: external signal 70: visual sensor 90: conveying device 100,100A: robot system 151: motion control unit 152: detection unit 153,153A: motion mode control unit 154: learning unit 154a: inference engine 154b: learning data 154c: learning model 155: memory unit 156: contact detection unit G: workpiece group S1~S13,S101~S105: step t1: start time t2: time T ₁ : first imaging interval T ₂ :2nd imaging interval W:Workpiece

FIG. 1 is a diagram showing the machine configuration of a robot system of one embodiment. FIG. 2 is a functional block diagram of a robot control device. FIG. 3 is a diagram for explaining the action status of a robot. FIG. 4 is a diagram for explaining the action status of a robot. FIG. 5 is a diagram showing the configuration of an inference engine. FIG. 6 is a flowchart showing the action mode control process. FIG. 7 is a flowchart showing the learning process of a learning unit. FIG. 8 is a diagram showing the machine configuration of a robot system of a second embodiment.

10:Robots

11: Hands

40: Teaching operation panel

41: Display unit

50:Robot control device

51: Processor

62:Sensor

63: External signal

70: Visual sensor

90: Transport device

100:Robotic system

151: Motion control unit

152: Testing Department

153: Action mode control unit

154: Study Department

154a: Inference Engine

154b: Study materials

155: Memory Department

Claims (17)

A robot system comprises: a robot; and an action mode control unit for switching the action mode of the robot or other machines constituting the robot system including the robot based on information related to the continuation of the operation of the robot. The robot system of claim 1 is provided with a sensor, wherein the sensor is used to detect a condition related to the continuation of the operation of the robot, and the action mode control unit switches the action mode of the robot or other machines constituting the robot system based on the detection result of the sensor. The robot system of claim 2, wherein the sensor is a visual sensor capable of photographing a workpiece at a predetermined photographing interval, the robot system further comprises a detection unit, the detection unit detects the workpiece based on the photographic image of the visual sensor, and the action mode control unit switches the action mode of the robot or the visual sensor based on the detection result of the detection unit. A robot system as claimed in claim 3, wherein the action mode control unit causes the robot to shift to a power saving mode that can reduce power consumption in response to the detection result that the workpiece is not detected. A robot system as claimed in claim 4, wherein the action mode control unit causes the robot to shift to the energy saving mode in response to the detection result that the workpiece is not detected a predetermined number of times during the detection by the detection unit of the photographic images obtained at the predetermined photographic interval. A robot system as claimed in claim 4, wherein the action mode control unit causes the robot to shift from the energy saving mode to the general action mode in response to the detection result of the workpiece. A robot system as claimed in any one of claims 3 to 6, wherein the action mode control unit changes the imaging interval of the visual sensor in response to the detection result that the workpiece is not detected. A robot system as claimed in claim 7, wherein the action mode control unit changes the imaging interval of the visual sensor to the second imaging interval in response to the detection unit failing to detect the workpiece a predetermined number of times in the detection of the photographic image of the workpiece taken by the visual sensor at the first imaging interval. A robot system as claimed in claim 8, wherein the action mode control unit accepts user input for setting the second imaging interval. The robot system of claim 8 further comprises a learning unit, which learns based on input data related to the duration of the state in which the workpiece is not detected, thereby setting the second imaging interval. A robot system as claimed in claim 10, wherein the learning unit constructs a learning model by learning learning data, wherein the learning data uses at least one of the start time, date, day of the week, month, season, weather conditions, sensor input, external signal, robot program, and internal state of the robot when the workpiece is not detected as input data, and uses the continuation time after the workpiece is not detected as teacher data. A robot system as claimed in claim 11, wherein the learning unit inputs at least one of the start time, date, day of the week, month, season, weather conditions, sensor input, external signal, robot program, and internal state of the robot when the workpiece was not detected into the learning model that has completed learning, thereby obtaining an estimated value of the duration after the workpiece was not detected, and setting the second imaging interval based on the estimated value. A robot system as claimed in claim 11, wherein the learning unit records, during the operation of the robot system, at least one of the start time, date, day of the week, month, season, weather conditions, sensor input, external signal, robot program, and internal state of the robot when the workpiece is not detected, and the end time or continuation time when the workpiece is not detected, for a predetermined period of time, thereby obtaining the learning data. A robot system as claimed in claim 8, wherein the second imaging interval is longer than the first imaging interval. As in the robot system of claim 2, the sensor is a sensor for detecting the force acting on the robot, and the action mode control unit switches the robot to a power saving mode when no change in the force acting on the robot is detected for a certain period of time. A robot system as claimed in claim 15, wherein the action mode control unit switches the robot to a power saving mode when no object contact with the robot is detected for a certain period of time based on the output of the sensor. A control device is a control device for controlling a robot, the control device comprising: An action mode control unit switches the action mode of the robot or other machines constituting a robot system including the robot based on information related to the continuation of the operation of the robot.

Images