Control device and control method, and robot hand

Abstract

The present disclosure relates to a control device and a control method capable of further improving a measurement function, and to a robot hand. The robot hand is provided with substantially columnar sensor fingers configured to be rotatable about a rotation axis along the longitudinal direction and provided with a plurality of sensors side by side on a side surface along the circumferential direction. Then, it is determined whether or not to switch a sensor to be used for controlling the operation of the robot hand among the plurality of sensors, and a sensor finger on which the sensor that has been determined to be switched to is provided is controlled to rotate about the rotation axis. The present technology can be applied to, for example, a robot hand provided with a plurality of sensor fingers.

Description

Control device, control method, and robot hand

[0001] The present disclosure relates to a control device, a control method, and a robot hand, and more particularly to a control device, a control method, and a robot hand that are capable of further improving measurement functions.

[0002] Conventionally, in order for a robotic hand to perform highly accurate operations, it is necessary to improve the measurement function for measuring the surrounding environment, objects, etc. For example, when a robotic hand performs an operation of grasping an object with its fingertips, various sensors can be mounted on the fingertips of the robotic hand to provide feedback on the distance and contact with the object, enabling appropriate control.

[0003] For example, Patent Document 1 discloses a robot hand that controls the rotation of a link member around the finger longitudinal axis based on the output of a pressure sensor provided in the finger mechanism that grasps the object to be grasped, so that the finger mechanism is in an optimal position for the object to be grasped.

[0004] Japanese Patent Application Laid-Open No. 2006-43843

[0005] However, it is difficult to properly mount multiple sensors on the fingertips of a robot hand, which has limited mounting space, but there has been a demand for robot hands to be equipped with the measurement functions necessary for high-precision operation.

[0006] The present disclosure has been made in view of such circumstances, and aims to make it possible to further improve measurement functions.

[0007] A control device according to one aspect of the present disclosure includes a sensor switching determination unit that determines whether to switch a sensor to be used as a sensor for controlling the operation of a robot equipped with a sensor unit, out of a plurality of sensors arranged side by side along the circumferential direction of a roughly columnar sensor unit that is configured to be rotatable around a rotation axis along the longitudinal direction, and a rotation control unit that controls the rotation of the sensor unit, around the rotation axis, in which the sensor determined by the sensor switching determination unit to be switched is installed.

[0008] A control method according to one aspect of the present disclosure includes a control device determining whether to switch a sensor to be used as a sensor for controlling the operation of a robot equipped with a sensor unit from among a plurality of sensors arranged side by side along the circumferential direction of an approximately cylindrical sensor unit configured to be rotatable around a rotation axis along the longitudinal direction, and controlling the rotation of the sensor unit in which the sensor determined to be switched is installed, around the rotation axis.

[0009] A robot hand according to one aspect of the present disclosure comprises a sensor finger having a generally cylindrical shape that is rotatable around a rotation axis along a longitudinal direction and that has a plurality of sensors arranged along a circumferential direction on its side; a sensor switching determination unit that determines whether to switch one of the plurality of sensors to be used as a sensor to be used for controlling the operation of the robot hand that includes the sensor finger; and a control device that controls the rotation of the sensor finger, around the rotation axis, that is provided with the sensor that has been determined by the sensor switching determination unit to be switched.

[0010] In one aspect of the present disclosure, a determination is made as to whether or not to switch a sensor to be used as a sensor to be used for controlling the operation of a robot equipped with a sensor unit among a plurality of sensors arranged side by side along the circumferential direction of an approximately columnar sensor unit configured to be rotatable around a rotation axis along the longitudinal direction, and rotation around the rotation axis of the sensor unit equipped with the sensor determined to be switched is controlled.

[0011] 13 is a diagram illustrating a configuration example of a first embodiment of a robot hand to which the present technology is applied. FIG. 14 is a diagram illustrating an operation of grasping a target object with a robot hand. FIG. 15 is a block diagram illustrating a first configuration example of a control device. FIG. 16 is a diagram illustrating a determination method for determining sensor switching based on a threshold value. FIG. 17 is a diagram illustrating an example of a correspondence relationship between a usage scene and sensors in use. FIG. 18 is a diagram illustrating an example of input information and output information of a neural network. FIG. 19 is a diagram illustrating input information. FIG. 20 is a diagram illustrating output information. FIG. 21 is a flowchart illustrating a control process for controlling the operation of a robot hand. FIG. 19 is a diagram illustrating a configuration example of a second embodiment of a robot hand to which the present technology is applied. FIG. 22 is a block diagram illustrating a second configuration example of a control device. FIG. 23 is a diagram illustrating an operation of grasping a target object with two fingers. FIG. 24 is a diagram illustrating definitions of sensors in use in a task of grasping a target object with three fingers. FIG. 25 is a diagram illustrating sensors in use defined in the task shown in FIG. 13. FIG. 26 is a diagram illustrating definitions of sensors in use in a task of pressing a button with the fingertip of one finger. FIG. 27 is a diagram illustrating sensors in use defined in the task shown in FIG. 15. FIG. 28 is a diagram illustrating a first application example in which multiple sensors are used simultaneously as sensors in use. FIG. 29 is a diagram illustrating a second application example in which multiple sensors are used by periodically switching between sensors in use. FIG. 29 is a diagram illustrating in-hand manipulation. 1 is a block diagram illustrating an example of the configuration of an embodiment of a computer to which the present technology is applied.

[0012] Hereinafter, specific embodiments to which the present technology is applied will be described in detail with reference to the drawings.

[0013] <First Configuration Example of Robot Hand> FIG. 1 is a diagram showing a configuration example of a first embodiment of a robot hand to which the present technology is applied.

[0014] 1 is configured such that a robot body 21 is equipped with a movement device 22 and a robot arm 23, and a robot hand 24 is provided at the tip of the robot arm 23. For example, the robot 11 can move using the movement device 22, adjust the position of the robot hand 24 using the robot arm 23, and perform an operation of grasping an arbitrary target object 12 with the robot hand 24.

[0015] The robot hand 24 is configured by providing a hand base 31 attached to the robot arm 23 with a plurality of fingers 32 (two fingers 32-1 and 32-2 in the example shown in FIG. 1), as well as a recognition sensor 33.

[0016] The fingers 32 are configured to enable the robot hand 24 to perform various operations, such as grasping the target object 12 .

[0017] The recognition sensor 33 is a sensor used to recognize the target object 12 to be grasped by the robot hand 24, and outputs detection data indicating the result of detecting the target object 12. For example, the recognition sensor 33 can be an RGB(-D) sensor capable of acquiring color images and distance information, a ToF (Time of Flight) sensor capable of acquiring distance images based on the time of flight of light, or a Lidar (Light Detection and Ranging) sensor capable of measuring distance and shape by emitting laser light. Note that, in addition to the configuration in which the recognition sensor 33 is provided in the hand base 31 as shown in FIG. 1 , the recognition sensor 33 may also be provided in, for example, the head of the robot main body 21.

[0018] As shown in the enlarged view, the finger portion 32 is configured such that a sensor finger 41 is connected to the hand base 31 via a link finger 42. The link finger 42 is connected to the hand base 31 so as to be rotatable around a joint portion 43, and the sensor finger 41 is connected to the link finger 42 so as to be rotatable around a joint portion 44.

[0019] The sensor finger 41 is a generally columnar sensor unit configured to be rotatable around a rotation axis along the longitudinal direction relative to the connection portion with the link finger 42. The sensor finger 41 is configured with a plurality of different types of sensors arranged side by side along the circumferential direction of the sensor finger 41, and the orientation of these sensors can be changed by rotating around the rotation axis. For example, when grasping the target object 12, the sensor finger 41 rotates around the rotation axis to change the sensor facing the target object 12, thereby switching the sensor (hereinafter referred to as the "sensor in use") used to sense the target object 12, the surrounding environment, and the like in order to control the operation of the robot 11.

[0020] 1 , a tactile sensor 51 and a distance measurement sensor 52 are provided on circumferential side surfaces of the sensor finger 41 so as to face opposite each other. The sensor finger 41 is configured so that the sensor in use can be switched between the tactile sensor 51 and the distance measurement sensor 52 by rotating about a rotation axis shown by a dashed line. For example, the rotation of the sensor finger 41 is controlled so that when the tactile sensor 51 is used as the sensor in use, the tactile sensor 51 faces the target object 12, and when the distance measurement sensor 52 is used as the sensor in use, the distance measurement sensor 52 faces the target object 12.

[0021] Therefore, when the finger 32 approaches the target object 12, the robot hand 24 uses the distance sensor 52 as the active sensor to measure the distance to the target object 12. Then, when the finger 32 comes into contact with the target object 12, the robot hand 24 rotates the sensor finger 41 around the rotation axis to switch the active sensor, and uses the tactile sensor 51 as the active sensor to detect the tactile sensation (e.g., reaction force, pressure, etc.) generated by contact with the target object 12. In this way, the robot hand 24 can switch the active sensor depending on the distance to the target object 12.

[0022] The operation of gripping the target object 12 with the robot hand 24 will be described with reference to FIG.

[0023] 2A, the robot hand 24 is moved by the moving device 22 and the robot arm 23 toward the target object 12 recognized by the recognition sensor 33. At this time, since the robot hand 24 is away from the target object 12, the distance measurement sensor 52 is used as the sensor to be used.

[0024] 2B, the robot hand 24 moves until the target object 12 is placed between the fingers 32-1 and 32-2, and then the fingers 32-1 and 32-2 operate to sandwich the target object 12. At this time, the distance measurement sensors 52-1 and 52-2 used as sensors face the target object 12 and measure the distance to the target object 12 until the fingers 32-1 and 32-2 approach the target object 12.

[0025] 2C, the finger portions 32-1 and 32-2 approach the target object 12 until the distance therebetween is less than a predetermined distance, and just before coming into contact with the target object 12, the sensor fingers 41-1 and 41-2 rotate about the rotation axis so that the tactile sensors 51-1 and 51-2 directly face the target object 12. This switches the sensors in use from the distance measurement sensors 52-1 and 52-2 to the tactile sensors 51-1 and 51-2. The tactile sensors 51-1 and 51-2 then come into contact with the target object 12 to detect touch, and the target object 12 can be grasped by the finger portions 32-1 and 32-2.

[0026] In this way, the robot hand 24 can improve its measurement function by switching between the tactile sensor 51 and the distance measurement sensor 52 as the sensor to be used, and can perform operations with higher accuracy.

[0027] The sensor finger 41 may be configured to use a combination of the tactile sensor 51 and the distance sensor 52, or may be configured to use a combination of various types of sensors, such as a force sensor, a temperature sensor, a vibration sensor, Lidar, or an RGB(-D) sensor. Of course, a configuration using a combination of multiple types of sensors other than these may also be used. For example, a temperature sensor is used to measure the temperature of the target object 12 or the surrounding environment, and can detect contact with the target object 12 based on changes in temperature. A vibration sensor is used to estimate the surface material of the target object 12 or the surrounding environment, and can sensitively detect contact with the target object 12 based on changes in vibration.

[0028] The sensor finger 41 may be configured to use a combination of a long-distance measurement sensor 52 and a short-distance measurement sensor 52. In this configuration, the sensor finger 41 can switch the sensor in use so that the long-distance measurement sensor 52 is used when the sensor finger 41 is far from the target object 12, and the short-distance measurement sensor 52 is used when the sensor finger 41 approaches the target object 12.

[0029] The sensor finger 41 may be configured to use a combination of a high-load tactile sensor 51 and a low-load tactile sensor 51. In this configuration, the sensor finger 41 can switch the sensor to be used depending on the hardness of the target object 12 when it comes into contact with the target object 12, so that the high-load tactile sensor 51 is used when the hardness is high, and the low-load tactile sensor 51 is used when the hardness is low.

[0030] The sensor finger 41 can be used in a variety of ways, such as by switching between a plurality of sensors, or by constantly rotating around a rotation axis when not in contact with the target object 12, and continuously updating the sensor values ​​obtained from the plurality of sensors (for example, like a radar).

[0031] FIG. 3 is a block diagram showing an example of the configuration of a control device that controls the operation of the robot hand 24.

[0032] As shown in FIG. 3, the control device 61 is configured to include a recognition unit 71, sensor value acquisition units 72-1 and 72-2, sensor value processing units 73-1 and 73-2, a sensor switching determination unit 74, a rotation axis control unit 75, a sensor value switching unit 76, and a robot control unit 77.

[0033] The recognition unit 71 recognizes the environment around the robot 11 and the robot hand 24, the state of the robot 11 and the target object 12, and various other estimable situations based on the detection data detected by the recognition sensor 33 and the detection data detected by other sensors (not shown).The recognition unit 71 then supplies recognition information indicating the results of the recognition (e.g., RGB-D data, object recognition information, scene recognition information, robot state recognition information, etc.) to the sensor switching determination unit 74.

[0034] The sensor value acquisition unit 72-1 acquires raw sensor values ​​(e.g., current values ​​and voltage values) output from the tactile sensor 51 and supplies them to the sensor value processing unit 73-1. Similarly, the sensor value acquisition unit 72-2 acquires raw sensor values ​​(e.g., current values ​​and voltage values) output from the distance measurement sensor 52 and supplies them to the sensor value processing unit 73-2.

[0035] The sensor value processing unit 73-1 performs signal processing on the raw sensor values ​​supplied from the sensor value acquisition unit 72-1 to acquire sensor values ​​indicative of the tactile data detected by the tactile sensor 51, and supplies these to the sensor switching determination unit 74 and the sensor value switching unit 76. Similarly, the sensor value processing unit 73-2 performs signal processing on the raw sensor values ​​supplied from the sensor value acquisition unit 72-2 to acquire sensor values ​​indicative of the distance measurement data detected by the distance measurement sensor 52, and supplies these to the sensor switching determination unit 74 and the sensor value switching unit 76.

[0036] The sensor switching determination unit 74 determines whether to switch the sensor in use based on the recognition information supplied from the recognition unit 71, the sensor values ​​(tactile data) supplied from the sensor value processing unit 73-1, and the sensor values ​​(distance measurement data) supplied from the sensor value processing unit 73-2. If the sensor switching determination unit 74 determines to switch the sensor in use, it supplies sensor information indicative of the sensor to be used after the switch to the sensor value switching unit 76. For example, if the sensor switching determination unit 74 determines to switch from the tactile sensor 51 to the distance measurement sensor 52, it supplies sensor information indicative of the distance measurement sensor 52 to the sensor value switching unit 76, and if the sensor switching determination unit 74 determines to switch from the distance measurement sensor 52 to the tactile sensor 51, it supplies sensor information indicative of the tactile sensor 51 to the sensor value switching unit 76.

[0037] The sensor switching determination unit 74 also calculates the rotation angle and rotation speed for rotating the sensor finger 41 around the rotation axis so that the sensor to be used after switching (one of the tactile sensor 51 and the distance measurement sensor 52) faces the target object 12. The sensor switching determination unit 74 then supplies the rotation axis control unit 75 with rotation command information for commanding the sensor finger 41 to rotate at the calculated rotation angle and rotation speed.

[0038] The rotation axis control unit 75 controls the rotation of the sensor finger 41 in accordance with the rotation command information supplied from the sensor switching determination unit 74, and positions one of the tactile sensor 51 and the distance measurement sensor 52 to be used as the sensor facing the target object 12.

[0039] In response to being supplied with the sensor usage information from the sensor switching determination unit 74, the sensor value switching unit 76 switches the output of the sensor value so that the sensor value output from the sensor in use indicated by the sensor usage information is supplied to the robot control unit 77. For example, when the sensor usage information indicates the tactile sensor 51, the sensor value switching unit 76 switches the output so that the sensor value (tactile data) supplied from the sensor value processing unit 73-1 is supplied to the robot control unit 77. On the other hand, when the sensor usage information indicates the distance measurement sensor 52, the sensor value switching unit 76 switches the output so that the sensor value (distance measurement data) supplied from the sensor value processing unit 73-2 is supplied to the robot control unit 77.

[0040] The robot control unit 77 outputs control commands to control the operation of each part of the robot hand 24, such as the rotation of the link fingers 42 relative to the hand base 31 or the rotation of the sensor fingers 41 relative to the link fingers 42, based on the sensor value (i.e., one of the tactile data and the distance measurement data) supplied from the sensor value switching unit 76, and controls the operation of the robot hand 24. Furthermore, the robot control unit 77 can control the entire robot 11, such as controlling the movement by the movement device 22 and controlling the adjustment of the position of the robot hand 24 by the robot arm 23, as necessary.

[0041] The robot hand 24 configured as described above can mount the tactile sensor 51 and the distance measurement sensor 52 on the sensor finger 41 of the robot hand 24, which has limited mounting space, so that they face opposite each other. This allows the robot hand 24 to switch between the tactile sensor 51 and the distance measurement sensor 52 as the sensor to be used by rotating the sensor finger 41 about a rotation axis along the longitudinal direction of the sensor finger 41, depending on, for example, the target object 12, the surrounding environment, the manipulation or task being performed, etc. The robot hand 24 can then appropriately switch between tactile data and distance measurement data to use for control, thereby improving measurement functionality.

[0042] When the robot hand 24 is configured by providing three or more types of sensors on the sensor fingers 41, the control device 61 can be configured to include three or more sensor value acquisition units 72 and three or more sensor value processing units 73. The sensor switching determination unit 74 determines which of the three or more types of sensors to use, and the sensor value switching unit 76 can switch which of the three or more types of sensor values ​​to supply to the robot control unit 77.

[0043] Furthermore, the robot hand 24 may employ a configuration in which sensors with different dynamic ranges, accuracies, etc. are provided on the sides of the sensor finger 41 so that they face opposite each other. In this configuration, by rotating the sensor finger 41 around the rotation axis, sensor values ​​with different dynamic ranges, accuracies, etc. can be acquired, making it possible to accommodate a wider dynamic range, accuracies, etc.

[0044] <Sensor Switching Determination Method> First to third determination methods by which the sensor switching determination unit 74 determines whether to switch the sensor in use will be described with reference to FIGS.

[0045] First, the first determination method is a method of determining whether to switch the sensor in use according to a preset rule.

[0046] 4 as a reference, the sensor switching determination unit 74 can determine whether to switch the sensor to be used in accordance with a rule that uses the distance measurement sensor 52 when the sensor value is equal to or greater than the threshold, and uses the tactile sensor 51 when the sensor value is less than the threshold. Here, distance measurement data indicating the distance to the target object 12 measured by the distance measurement sensor 52 can be used as the sensor value, and when the distance to the target object 12 becomes closer than a certain distance, the sensor switching determination unit 74 can determine to switch the sensor to be used to the tactile sensor 51. Note that in a configuration in which multiple distance measurement sensors 52 are used, the minimum value or average value of the multiple distance measurement data may be used.

[0047] 5, a correspondence between the usage scene (task) of the robot 11 recognized by the recognition unit 71 and the sensor to be used is set in advance, and the sensor switching determination unit 74 can determine whether to switch the sensor to be used according to the correspondence (rule). For example, the sensor switching determination unit 74 can determine whether to switch the sensor to be used so that a temperature sensor is used when the usage scene is cooking, a tactile sensor is used when the usage scene is plating, a distance sensor is used when the usage scene is operating appliances, and a vibration sensor is used when the usage scene is cleaning.

[0048] Furthermore, when the sensor finger 41 is configured to use a combination of force sensors with different sensitivities, the sensor switching determination unit 74 can determine whether to switch the sensor in use, as will be described below.

[0049] For example, the sensor switching determination unit 74 can predict the mass of the target object 12 based on the recognition result of the target object 12, and determine to switch the sensor to be used so as to use a force sensor with sensitivity appropriate for that mass. Note that the sensor switching determination unit 74 can predict the mass of the target object 12 based on the recognition result of the size of the target object 12, for example. Furthermore, the sensor switching determination unit 74 can predict the mass of the target object 12 based on the recognition result of the object name of the target object 12 by referring to a general mass stored in association with the object name.

[0050] Furthermore, the sensor switching determination unit 74 can estimate the hardness of the target object 12 by contacting the target object 12 once and determine whether to switch the sensor to be used so as to use a force sensor with sensitivity corresponding to the hardness. For example, the sensor switching determination unit 74 can determine whether to switch sensors so as to use a high-sensitivity force sensor when the hardness of the target object 12 is low, and a low-sensitivity force sensor when the hardness of the target object 12 is high. Note that the sensor switching determination unit 74 can estimate the hardness of the target object 12 based on, for example, the contact force when contacting the target object 12, the pressing distance after contacting the target object 12, etc.

[0051] Next, the second judgment method is a method in which a predetermined processing is performed using a function or the like on the recognition information output from the recognition unit 71, and based on the processing result indicating the result of that processing, a judgment is made as to whether to switch the sensor to be used so that the most appropriate sensor is used.

[0052] Next, as shown in FIG. 6, the third judgment method is a method for determining whether to switch the sensor to be used by using a neural network 81 obtained by learning in which the recognition information (object recognition information, scene recognition information, robot state recognition information, etc.) output from the recognition unit 71 and / or the sensor value output from the sensor value processing unit 73 are used as input information, and sensor use information (sensor number, sensor evaluation value, etc.) indicating the optimal sensor to be used is used as output information.

[0053] The object recognition information is information for recognizing an object captured in an image, and may be, for example, an object class name, a bounding box, a mask, or the like.

[0054] The object class name can be a predefined object name (e.g., apple, orange, banana, etc.). Note that instead of using a single object name, the object class name can also be using likelihoods calculated for multiple possible candidate object names (e.g., likelihoods of apple, orange, and banana, as in the example shown in A of FIG. 7).

[0055] The bounding box can be the coordinates of a rectangular area that represents the position of the target object in the image. For example, Fig. 7B shows an example of a bounding box for an apple. Note that the bounding box may be three-dimensional data.

[0056] The mask can be a region of pixels that make up the target object in the image. For example, an example of a banana mask is shown in FIG. 7C. The mask may be three-dimensional data.

[0057] The scene recognition information is information for recognizing a usage scene of the robot 11, and may include, for example, location information, task information, relationship information, and the like.

[0058] The location information is information indicating the location where the robot 11 is located (for example, the kitchen, living room, office, etc.).

[0059] The task information is information indicating the task that the robot 11 is to execute (for example, grabbing, cutting, pushing, etc.).

[0060] The relationship information is information indicating the relationship between objects, and for example, graph information (on, near, behind, etc.) as shown in D of FIG. 7 can be used.

[0061] The robot state recognition information is information for recognizing the state of the robot 11, and may use, for example, the joint angles of each part of the robot 11, the speed, the force, the posture of the end effector, and the like.

[0062] The input information input to the neural network 81 may be, for example, image data, time series data of the output of the tactile sensor 51, time series data of the output of the distance measurement sensor 52, or trajectory data of the robot 11.

[0063] The sensor number is a number that directly specifies the sensor to be used. As shown in an example in A of Fig. 8, the sensor number can be set so that sensor number 0 specifies a force sensor, sensor number 1 specifies a distance measurement sensor (long distance), sensor number 2 specifies a distance measurement sensor (short distance), sensor number 3 specifies a tactile sensor, and sensor number 4 specifies a temperature sensor.

[0064] The sensor evaluation value is an evaluation value that evaluates which sensor would increase the success rate of the task at that time. For example, the likelihood calculated for each sensor may be used. Note that the sensor evaluation value changes over time. For example, B in FIG. 8 shows an example of the likelihoods of a force sensor, a distance sensor (long distance), a distance sensor (short distance), a tactile sensor, and a temperature sensor. In the illustrated example, it is shown that using the tactile sensor increases the success rate of the task.

[0065] The robot hand 24 configured as described above has multiple sensors arranged side by side along the circumferential direction of the sensor finger 41, and can switch the sensor being used depending on the application of each sensor by rotating the sensor finger 41 around a rotation axis along the longitudinal direction. This allows for a simpler structure than, for example, a configuration in which multiple sensors are mounted on the same plane, which can reduce manufacturing costs and the risk of interaction (interference) between sensors.

[0066] Furthermore, the robot hand 24 can cover a wide dynamic range using multiple sensors, and the specifications of each sensor can be lowered, thereby reducing costs. The robot hand 24 is also designed to easily replace sensors in advance as needed, allowing the sensors to be changed depending on the task, thereby enabling a wide range of tasks to be performed. Furthermore, the robot hand 24 can easily secure space for each sensor, allowing the specifications of each sensor (e.g., spatial resolution) to be improved.

[0067] <Example of Control Processing> An example of control processing in which the control device 61 controls the operation of the robot hand 24 will be described with reference to the flowchart shown in FIG.

[0068] In step S11 , the recognition unit 71 recognizes the target object 12 and the surrounding environment based on the detection data supplied from the recognition sensor 33 , and supplies recognition information indicating the recognition result to the sensor switching determination unit 74 .

[0069] In step S12, the sensor switching determination unit 74 determines whether or not to switch the sensor in use based on the recognition information supplied from the recognition unit 71 in step S11 and the sensor value supplied from the sensor value processing unit 73.

[0070] If the sensor switching determination unit 74 determines in step S12 that the sensor in use is to be switched, the process proceeds to step S13. On the other hand, if the sensor switching determination unit 74 determines in step S12 that the sensor in use is not to be switched, steps S13 and S14 are skipped and the process proceeds to step S15.

[0071] In step S13, the sensor switching determination unit 74 calculates the rotation angle and rotation speed for rotating the sensor finger 41 around the rotation axis so that the sensor to be used after switching faces the target object 12, and supplies rotation command information instructing the rotation of the sensor finger 41 to the rotation axis control unit 75. The rotation axis control unit 75 controls the rotation of the sensor finger 41 at the rotation angle and rotation speed in accordance with the rotation command information.

[0072] In step S14, the sensor switching determination unit 74 supplies used sensor information indicating the used sensor to be used after the switching to the sensor value switching unit 76. The sensor value switching unit 76 switches the output of the sensor value so as to supply the sensor value output from the used sensor indicated by the used sensor information to the robot control unit 77.

[0073] In step S15, the robot control unit 77 outputs control commands to control the operation of each part of the robot hand 24 based on the sensor values ​​supplied from the sensor value switching unit 76, and controls the operation of the robot hand 24. Thereafter, the process returns to step S11, and the same processes are repeated thereafter.

[0074] By having the control device 61 execute the control processing described above, the sensor values ​​used to control the operation of the robot hand 24 can be appropriately switched, thereby further improving the measurement function.

[0075] <Second Configuration Example of Robot Hand> Fig. 10 is a diagram showing a configuration example of a second embodiment of a robot hand to which the present technology is applied. Note that in a robot hand 24A shown in Fig. 10, components common to the robot hand 24 shown in Fig. 1 are denoted by the same reference numerals, and detailed descriptions thereof will be omitted.

[0076] 1 has two fingers 32-1 and 32-2 attached to a hand base 31, whereas the robot hand 24A has four fingers 32A-1 to 32A-4 attached to a hand base 31A, as shown in FIG. 10. Although not shown, the robot hand 24A is also provided with a recognition sensor 33 (see FIG. 1). The fingers 32A-1 to 32A-4 are configured in the same manner, and when there is no need to distinguish between them, they will be referred to as finger 32A.

[0077] The finger unit 32A is configured by three sensor fingers 41a to 41c connected via joints (not shown). The sensor finger 41a is rotatably connected to the hand base 31A by the joint, the sensor finger 41b is rotatably connected to the sensor finger 41a by the joint, and the sensor finger 41c is rotatably connected to the sensor finger 41b by the joint.

[0078] 1, the sensor fingers 41a to 41c are configured to be rotatable around a rotation axis along the longitudinal direction, and are configured such that a plurality of sensors of different types are provided side by side along the circumferential direction of the sensor fingers 41a to 41c. In Fig. 10, the sensor fingers 41a to 41c are illustrated so that the sides on which the tactile sensors 51a to 51c are provided face each other, and distance measuring sensors 52a to 52c (not shown) are provided on the opposite sides.

[0079] Fig. 11 is a block diagram showing an example of the configuration of a control device that controls the operation of the robot hand 24A. In the control device 61A shown in Fig. 11, components common to the control device 61 shown in Fig. 3 are denoted by the same reference numerals, and detailed description thereof will be omitted.

[0080] That is, the control device 61A has a common configuration with the control device 61 of FIG. 3 in that it includes a recognition unit 71, sensor value acquisition units 72-1 and 72-2, sensor value processing units 73-1 and 73-2, and a robot control unit 77.

[0081] The control device 61A differs from the control device 61 of FIG. 3 in that it includes a sensor switching determination unit 74A, a rotation axis control unit 75A, and a sensor value switching unit 76A.

[0082] The sensor switching determination unit 74A determines whether to switch the sensor in use for each of the sensor fingers 41a to 41c of the finger unit 32A for each of the four fingers 32A-1 to 32A-4 based on the recognition information, sensor values ​​(tactile data), and sensor values ​​(distance measurement data). If the sensor switching determination unit 74 determines to switch the sensor in use, it supplies sensor information indicative of the sensor to be used after the switch to the sensor value switching unit 76A. The sensor switching determination unit 74A also supplies rotation command information instructing rotation for each of the sensor fingers 41a to 41c of the finger unit 32A for each of the four fingers 32A-1 to 32A-4 to the rotation axis control unit 75A.

[0083] The rotation axis control unit 75A controls the rotation of the sensor fingers 41 for each of the sensor fingers 41a to 41c of the finger unit 32A for each of the four finger units 32A-1 to 32A-4 in accordance with the rotation command information supplied from the sensor switching determination unit 74A. This allows each of the four finger units 32A-1 to 32A-4 to orient the sensor in use directly toward the target object 12 for each of the sensor fingers 41a to 41c of the finger unit 32A.

[0084] In response to the supply of the sensor usage information from the sensor switching determination unit 74A, the sensor value switching unit 76A switches the output of the sensor values ​​for each of the four finger units 32A-1 to 32A-4, so as to supply the sensor values ​​output from the sensor in use indicated by the sensor usage information to the robot control unit 77 for each of the sensor fingers 41a to 41c of the finger unit 32A.

[0085] The robot hand 24A configured as described above can switch the sensor used for each of the sensor fingers 41a to 41c of the finger portion 32A for each of the four finger portions 32A-1 to 32A-4, and supply the sensor values ​​output from each sensor used to the robot control unit 77.

[0086] This allows the robot hand 24A to switch the sensor in use for each of the fingers 32A-1 to 32A-4, for example. Therefore, the robot hand 24A can use the tactile sensor 51 as the sensor in use for the fingers 32A used to grasp the target object 12, and can use the distance measurement sensor 52 as the sensor in use for the fingers 32A not used to grasp the target object 12.

[0087] Furthermore, the robot hand 24A can switch the sensor to be used for each of the sensor fingers 41a to 41c of the finger portion 32A, for example. As a result, the robot hand 24A can use the tactile sensor 51 as the sensor to be used for the sensor finger 41 used to contact the target object 12, and can use the distance measurement sensor 52 as the sensor to be used for the sensor finger 41 not used to contact the target object 12.

[0088] This allows the robot hand 24A to simultaneously perform high-precision contact control based on contact data obtained from the tactile sensor 51, and optimize the grasping posture while measuring the distance to the target object 12 (the distance between the finger 32A used to grasp the target object 12 and the target object 12, or the distance between the sensor finger 41 used to contact the target object 12 and the target object 12) based on distance measurement data obtained from the distance measurement sensor 52.

[0089] 12 shows an example of an operation in which the robot hand 24A grasps the target object 12 by using the fingers 32A-3 and 32A-4 to grasp the target object 12 and the sensor fingers 41c-3 and 41c-4 to contact the target object 12. When performing such an operation, the tactile sensor 51c-3 is used as the active sensor in the sensor finger 41c-3, the tactile sensor 51c-4 is used as the active sensor in the sensor finger 41c-4, and the distance measurement sensor 52 is used as the active sensor in the other sensor fingers 41. This makes it possible to optimize the grasping posture while measuring the distance to the target object 12 based on the distance measurement data acquired from the distance measurement sensor 52, thereby realizing control of the grasping posture in accordance with the shape of the target object 12.

[0090] Furthermore, the robot hand 24A can define in advance the sensors to be used for various tasks for each of the sensor fingers 41a to 41c of the four fingers 32A-1 to 32A-4.

[0091] For example, as shown in FIG. 13, in a task such as grasping a target object 12 with three fingers 32A-2 to 32A-4, the distance measurement sensor 52a-1 of the sensor finger 41a-1, the distance measurement sensor 52b-1 of the sensor finger 41b-1, the distance measurement sensor 52c-1 of the sensor finger 41c-1, the distance measurement sensor 52a-2 (not shown) of the sensor finger 41a-2, the distance measurement sensor 52a-3 (not shown) of the sensor finger 41a-3, the distance measurement sensor 52a-4 (not shown) of the sensor finger 41a-4, and the distance measurement sensor 52a-4 (not shown) of the sensor finger 41b-4 are defined to be used as the sensors to be used. 14, the distance measurement sensor 52 is defined to be used as the sensor finger 41a-1, the sensor finger 41b-1, the sensor finger 41c-1, the sensor finger 41a-2, the sensor finger 41a-3, the sensor finger 41a-4, and the sensor finger 41b-4, and the tactile sensor 51 is defined to be used as the sensor finger 41b-2, the sensor finger 41c-2, the sensor finger 41c-3, the sensor finger 41c-3, and the sensor finger 41c-4, respectively.

[0092] Also, for example, in a task such as pressing a button with the fingertip of one finger portion 32A-1, the tactile sensor 51c-1 of sensor finger 41c-1 is defined to be used as the sensor to be used, and the distance measurement sensor 52 is defined to be used as the sensor to be used for the other fingers, as shown in Fig. 15. That is, as shown in Fig. 16, the tactile sensor 51 is used as the sensor to be used for sensor finger 41c-1, and the distance measurement sensor 52 is defined to be used as the sensor to be used for sensor fingers 41a-1, 41b-1, 41a-2, 41b-2, 41c-2, 41a-3, 41b-3, 41c-3, 41a-4, 41b-4, and 41c-4.

[0093] In addition to defining the sensors to be used for each sensor finger 41 according to the task, the ratio of sensors to be used according to the task (for example, the ratio between the tactile sensors 51 and the distance measurement sensors 52) may be defined in advance. This makes it possible to switch the sensors to be used for each sensor finger 41 according to the task, and to appropriately change the ratio between the tactile sensors 51 and the distance measurement sensors 52 used as sensors (how many of which sensors are used for which task).

[0094] For example, as shown in Figure 13, in a task such as grasping a target object 12, the contact area with the target object 12 becomes large, a certain grasping force is required, and control of the grasping posture in accordance with the surface shape of the target object 12 is required. Therefore, as shown in Figure 14, the proportion of tactile sensor 51 used as the sensor is defined to be high (compared to a task such as that shown in Figure 15).

[0095] Furthermore, as shown in FIG. 15, in a task such as pressing a button, no contact occurs other than with the sensor finger 41 that presses the button, and the distance between the other sensor fingers 41 and the button needs to be adjusted. Therefore, as shown in FIG. 16, the ratio of using the distance measurement sensor 52 as the sensor to be used is defined to be higher (compared to a task such as that shown in FIG. 13).

[0096] The robot hand 24A configured as described above can switch the sensors used for each of the multiple sensor fingers 41 constituting each of the multiple finger portions 32A. This allows the robot hand 24A to adjust the physical quantities measured for each finger portion 32A and each sensor finger 41, thereby improving the control performance of the robot hand 24A. Furthermore, the robot hand 24A can adjust the combination of sensors provided on the sensor fingers 41 in accordance with the task purpose and the required movement quality, thereby improving task performance.

[0097] <Application Examples> Application examples of the robot hand 24 will be described with reference to FIGS.

[0098] Referring to FIG. 17, a first application example in which a plurality of sensors are simultaneously used as active sensors will be described.

[0099] FIG. 17 shows a side view of the sensor finger 41 facing the target object 12 .

[0100] For example, when only the tactile sensor 51 is used as the sensor in use, the angle of the sensor finger 41 is controlled so that the tactile sensor 51 faces the target object 12, as shown in A of Fig. 17. Similarly, when only the distance measuring sensor 52 is used as the sensor in use, the angle of the sensor finger 41 is controlled so that the distance measuring sensor 52 faces the target object 12, as shown in B of Fig. 17.

[0101] 17C, when the tactile sensor 51 and the distance measurement sensor 52 are used simultaneously as the active sensors, the angle of the sensor finger 41 is controlled so that both a part of the tactile sensor 51 and a part of the distance measurement sensor 52 face the target object 12, i.e., so that the tactile sensor 51 and the distance measurement sensor 52 can simultaneously sense the target object 12. For example, the angle of the sensor finger 41 is controlled so that an intermediate region between the region where the tactile sensor 51 is provided and the region where the distance measurement sensor 52 is provided faces the target object 12 directly.

[0102] According to this first application example, the robot hand 24 can simultaneously use the tactile sensor 51 and the distance measurement sensor 52 as sensors in use.

[0103] In this way, by using multiple sensors simultaneously, the robot hand 24 can adjust the type of physical quantity fed back to the control, thereby improving control performance. Furthermore, the robot hand 24 can adjust the importance (weight) of the physical quantity measured by each sensor simply by controlling the rotation angle, making it possible to perform control that matches the required operation quality, specifications, etc.

[0104] Referring to FIG. 18, a second application example in which a plurality of sensors are periodically switched and used as the sensor in use will be described.

[0105] FIG. 18 shows a side view of the sensor finger 41 facing the target object 12 .

[0106] For example, by constantly rotating the sensor finger 41 at a predetermined rotation speed, the tactile sensor 51 and the distance measurement sensor 52 can be used alternately as the sensors to be used, with periodic switching occurring depending on the rotation speed of the sensor finger 41.

[0107] A use example of applying the second application example to a configuration that uses a combination of a long-distance measurement sensor 52 and a short-distance measurement sensor 52 will be described. For example, in a manipulation task in which the distance to the target object 12 varies within a range from 0 m (contact) to 2 m or more, a sensor with a wider dynamic range is required. Therefore, by constantly rotating the sensor finger 41 at a predetermined rotation speed and alternately using the long-distance measurement sensor 52 and the short-distance measurement sensor 52 as the active sensor, the distance to the target object 12 can be measured over a wider dynamic range.

[0108] A use example of applying the second application example to a configuration using a combination of an RGB camera and a tactile sensor 51 will be described. For example, the robot hand 24 can perform in-hand manipulation, in which the target object 12 is moved within the robot hand 24 by rotating the sensor fingers 41 using them as rollers. By alternately using the RGB camera and the tactile sensor 51 as the sensors in accordance with the rotation of the sensor fingers 41 when performing such in-hand manipulation, it is possible to recognize the texture of the target object 12 with the RGB camera while acquiring tactile data with the tactile sensor 51.

[0109] Referring to FIG. 19, in-hand manipulation using a combination of a tactile sensor 51 and a distance measuring sensor 52 will be described.

[0110] 19 shows an example of in-hand manipulation in which the target object 12 is moved within the robot hand 24A using the sensor fingers 41c-1 to 41c-4 of the robot hand 24A. In the example shown, the sensor fingers 41c-1 to 41c-3 are placed below the target object 12, and the sensor finger 41c-4 is placed above the target object 12, and the target object 12 moves from left to right.

[0111] 19A, the robot hand 24A grasps the target object 12 with the sensor fingers 41c-1 and 41c-4. At this time, the tactile sensor 51c-1 of the sensor finger 41c-1 faces the target object 12, and the tactile sensor 51c-4 of the sensor finger 41c-4 faces the target object 12, and the tactile sensor 51c-1 and the tactile sensor 51c-4 are used as the sensors in use. Furthermore, the sensor fingers 41c-2 and 41c-3 that are not used to grasp the target object 12 have their distance measurement sensors 52c-2 and 52c-3 facing upward, and the distance measurement sensors 52c-2 and 52c-3 are used as the sensors in use.

[0112] 19B, the sensor finger 41c-4 moves itself so as to move the target object 12 toward the right, and rotates with this movement so as not to cause any displacement in the contact surface with the target object 12. At this time, the sensor finger 41c-1 rotates with the movement of the target object 12 so as not to cause any displacement in the contact surface with the target object 12. The sensor finger 41c-2 then rotates so that the tactile sensor 51c-2 faces the target object 12 directly. As a result, the target object 12 is grasped by the sensor fingers 41c-1, 41c-2, and 41c-4, and the tactile sensors 51c-1 and 51c-2 are used as the sensors in use.

[0113] Next, as shown in C of FIG. 19 , the sensor finger 41c-4 moves itself so as to move the target object 12 toward the right, and rotates with this movement so as not to cause any misalignment in the contact surface with the target object 12. At this time, the sensor finger 41c-2 rotates with the movement of the target object 12 so as not to cause any misalignment in the contact surface with the target object 12. The sensor finger 41c-3 then rotates so that the tactile sensor 51c-3 faces the target object 12 directly. As a result, the target object 12 is grasped by the sensor fingers 41c-2, 41c-3, and 41c-4, and the tactile sensors 51c-2 and 51c-3 are used as the sensors in use. Note that the sensor finger 41c-1 that is no longer being used to grasp the target object 12 rotates so that the distance measurement sensor 52c-1 faces upward, and the distance measurement sensor 52c-1 is used as the sensor in use.

[0114] 19D, the sensor finger 41c-4 moves itself so as to move the target object 12 toward the right, and rotates as it moves so as not to cause any misalignment in its contact surface with the target object 12. After the target object 12 moves, the sensor finger 41c-4 can adjust its rotation so that the tactile sensor 51c-4 faces the target object 12 directly. The sensor finger 41c-3 rotates as the target object 12 moves so as not to cause any misalignment in its contact surface with the target object 12. As a result, the target object 12 is grasped by the sensor fingers 41c-3 and 41c-4, and the tactile sensors 51c-3 and 51c-4 are used as the active sensors. The sensor finger 41c-2, which is no longer being used to grasp the target object 12, rotates so that the distance measurement sensor 52c-2 faces upward, and the distance measurement sensor 52c-2 is used as the active sensor.

[0115] When performing such in-hand manipulation, the robot hand 24A uses the tactile sensor 51c as the active sensor for the sensor finger 41c that is in contact with the target object 12, and uses the distance measurement sensor 52c as the active sensor for the sensor finger 41c that is not in contact with the target object 12. This allows the robot hand 24A to simultaneously acquire tactile data and distance measurement data that are required to move the target object 12 within the robot hand 24A, thereby enabling more accurate control of the operation.

[0116] According to this second application example, the robot hand 24 can periodically switch between the tactile sensor 51 and the distance measurement sensor 52 as the sensor in use.

[0117] In this way, by periodically switching between multiple sensors as the sensors in use, the robot hand 24 can widen the dynamic range of the sensors with a simple structure. Furthermore, in operations that require rotation of the sensor fingers 41, such as in-hand manipulation, the robot hand 24 can frequently acquire multiple physical quantities, thereby enabling more delicate operations.

[0118] The sensor fingers 41 may be provided, for example, on the fingers, nails, or palm of the robot hand 24. Alternatively, the sensor fingers 41 may be provided at locations that come into contact with other objects, such as the feet of a drone or a legged robot.

[0119] <Example of Computer Configuration> Next, the above-described series of processes (control method) can be performed by hardware or software. When the series of processes is performed by software, a program constituting the software is installed in a general-purpose computer or the like.

[0120] FIG. 20 is a block diagram showing an example of the configuration of an embodiment of a computer in which a program for executing the above-described series of processes is installed.

[0121] The program can be recorded in advance on the hard disk 105 or ROM 103 as a recording medium built into the computer.

[0122] Alternatively, the program can be stored (recorded) on a removable recording medium 111 driven by the drive 109. Such a removable recording medium 111 can be provided as a so-called package software. Here, examples of the removable recording medium 111 include a flexible disk, a CD-ROM (Compact Disc Read Only Memory), an MO (Magneto Optical) disk, a DVD (Digital Versatile Disc), a magnetic disk, and a semiconductor memory.

[0123] The program can be installed into the computer from the removable recording medium 111 as described above, or can be downloaded to the computer via a communication network or a broadcasting network and installed on the built-in hard disk 105. That is, the program can be transferred to the computer wirelessly from a download site via an artificial satellite for digital satellite broadcasting, or transferred to the computer by wire via a network such as a LAN (Local Area Network) or the Internet.

[0124] The computer includes a CPU (Central Processing Unit) 102 , to which an input / output interface 110 is connected via a bus 101 .

[0125] When a user inputs a command via the input / output interface 110 by operating the input unit 107, the CPU 102 executes a program stored in a read-only memory (ROM) 103 in accordance with the command. Alternatively, the CPU 102 loads a program stored on a hard disk 105 into a random access memory (RAM) 104 and executes the program.

[0126] As a result, the CPU 102 performs processing according to the flowchart described above or processing performed by the configuration of the block diagram described above. Then, the CPU 102 outputs the processing results from the output unit 106 via the input / output interface 110, transmits them from the communication unit 108, or records them on the hard disk 105, as necessary.

[0127] The input unit 107 is made up of a keyboard, a mouse, a microphone, etc. The output unit 106 is made up of an LCD (Liquid Crystal Display), a speaker, etc.

[0128] In this specification, the processing performed by a computer according to a program does not necessarily have to be performed in chronological order according to the order described in the flowchart. In other words, the processing performed by a computer according to a program also includes processing that is executed in parallel or individually (for example, parallel processing or object-based processing).

[0129] The program may be processed by a single computer (processor), or may be distributed among multiple computers. Furthermore, the program may be transferred to and executed on a remote computer.

[0130] Furthermore, in this specification, a system refers to a collection of multiple components (devices, modules (components), etc.), regardless of whether all of the components are contained in the same housing. Therefore, multiple devices housed in separate housings and connected via a network, and a single device housed in a single housing with multiple modules, are both systems.

[0131] Also, for example, a configuration described as one device (or processing unit) may be divided and configured as multiple devices (or processing units). Conversely, configurations described above as multiple devices (or processing units) may be combined and configured as one device (or processing unit). Of course, configurations other than those described above may be added to the configuration of each device (or each processing unit). Furthermore, as long as the configuration and operation of the entire system are substantially the same, part of the configuration of one device (or processing unit) may be included in the configuration of another device (or other processing unit).

[0132] Furthermore, for example, the present technology can be configured as a cloud computing system in which a single function is shared and processed collaboratively by a plurality of devices via a network.

[0133] Furthermore, for example, the above-described program can be executed in any device, as long as the device has the necessary functions (functional blocks, etc.) and can obtain the necessary information.

[0134] Also, for example, each step described in the above flowchart can be executed by one device or can be shared and executed by multiple devices. Furthermore, if one step includes multiple processes, the multiple processes included in that one step can be executed by one device or can be shared and executed by multiple devices. In other words, multiple processes included in one step can be executed as multiple step processes. Conversely, processes described as multiple steps can be executed collectively as a single step.

[0135] In addition, the processing of the steps of a program executed by a computer may be executed in chronological order according to the order described in this specification, or may be executed in parallel or individually at the required timing, such as when a call is made. In other words, as long as no contradiction occurs, the processing of each step may be executed in an order different from the order described above. Furthermore, the processing of the steps of this program may be executed in parallel with the processing of another program, or may be executed in combination with the processing of another program.

[0136] It should be noted that the present technologies described in this specification can be implemented independently and singly, unless a contradiction arises. Of course, any two or more of the present technologies can also be implemented in combination. For example, part or all of the present technologies described in any embodiment can be implemented in combination with part or all of the present technologies described in other embodiments. Furthermore, part or all of any of the present technologies described above can also be implemented in combination with other technologies not described above.

[0137] <Examples of Combinations of Configurations> The present technology can also be configured as follows. (1) A control device comprising: a sensor switching determination unit that determines whether to switch a sensor to be used as a sensor to be used for controlling an operation of a robot including a substantially columnar sensor unit, among a plurality of sensors arranged side by side along a circumferential direction of the sensor unit that is configured to be rotatable about a rotation axis along a longitudinal direction; and a rotation control unit that controls rotation about the rotation axis of the sensor unit in which the sensor determined to be switched by the sensor switching determination unit is provided. (2) The control device described in (1) above, further comprising: a recognition unit that recognizes a target object that is a target of an operation of the robot; and a sensor value processing unit that acquires sensor values ​​indicating detection data detected by the sensor, wherein the sensor switching determination unit determines whether to switch the sensor to be used based on recognition information indicating a result of the recognition unit recognizing the target object and the plurality of sensor values. (3) The control device described in (2) above, further comprising: a robot control unit that controls an operation of the robot; and a sensor value switching unit that switches output so that the sensor value acquired from the sensor determined to be switched by the sensor switching determination unit is supplied to the robot control unit. (4) The control device according to (3), wherein the sensor switching determination unit determines whether to switch the sensor in use according to a preset rule. (5) The control device according to (4), wherein the sensor unit is provided with a distance measurement sensor that measures the distance to the target object and a tactile sensor that detects tactile sensations generated by contact with the target object, and the sensor switching determination unit determines whether to switch the sensor in use according to the rule that the distance measurement sensor is used when the distance to the target object is equal to or greater than a threshold, and the tactile sensor is used when the distance to the target object is less than the threshold. (6) The control device according to (3), wherein the sensor switching determination unit performs predetermined processing on recognition information obtained by the recognition unit recognizing the target object, and determines whether to switch the sensor in use based on a processing result indicating a result of the processing.(7) The control device according to (3), wherein the sensor switching determination unit determines whether to switch the sensor to be used by using a neural network obtained by learning using at least one of recognition information obtained by the recognition unit recognizing the target object and the sensor value as input information, and sensor information to be used that indicates the optimal sensor to be used as output information. (8) The control device according to any of (3) to (7), wherein the sensor unit is a sensor finger that constitutes a finger unit of a robot hand, and the robot control unit controls the operation of the robot hand. (9) The control device according to (8), wherein the finger unit is constituted by a plurality of the sensor fingers connected via joints, and the robot hand is constituted by a plurality of the finger units, and the sensor switching determination unit determines whether to switch the sensor to be used for each of the plurality of sensor units that constitute the finger unit. (10) The control device according to any of (1) to (9), wherein the rotation control unit controls angles of the plurality of sensor units with respect to the target object so that the plurality of sensors provided in one sensor unit can sense the target object simultaneously, and the plurality of sensors are simultaneously used as the active sensor. (11) The control device according to any of (1) to (10), wherein the rotation control unit constantly rotates the sensor unit at a predetermined rotation speed, and the plurality of sensors are alternately used as the active sensor so that they can be switched according to the rotation speed. (12) A control method including: a control device determining whether to switch the sensor to be used as the active sensor for controlling an operation of a robot including the sensor unit, from among a plurality of sensors provided side by side along a circumferential direction of a substantially columnar sensor unit configured to be rotatable about a rotation axis along a longitudinal direction, and controlling the rotation of the sensor unit, about the rotation axis, where the sensor determined to be switched is provided.(13) A robot hand comprising a control device having: a sensor finger having a generally columnar shape configured to be rotatable around a rotation axis along the longitudinal direction and having a plurality of sensors arranged on a side surface along the circumferential direction; a sensor switching determination unit that determines whether to switch a sensor from among the plurality of sensors to be used as a sensor to be used for controlling the operation of the robot hand having the sensor finger; and a rotation control unit that controls the rotation around the rotation axis of the sensor finger having the sensor determined to be switched by the sensor switching determination unit.

[0138] It should be noted that the present embodiment is not limited to the above-described embodiment, and various modifications are possible within the scope of the gist of the present disclosure. Furthermore, the effects described in this specification are merely examples and are not intended to be limiting, and other effects may also be obtained.

[0139] DESCRIPTION OF SYMBOLS 11 Robot, 12 Target object, 21 Robot body, 22 Moving device, 23 Robot arm, 24 Robot hand, 31 Hand base, 32 Finger portion, 33 Recognition sensor, 41 Sensor finger, 42 Link finger, 43 and 44 Joint portion, 51 Tactile sensor, 52 Distance measurement sensor, 61 Control device, 71 Recognition unit, 72 Sensor value acquisition unit, 73 Sensor value processing unit, 74 Sensor switching determination unit, 75 Rotation axis control unit, 76 Sensor value switching unit, 77 Robot control unit, 81 Neural network

Claims

1. A control device comprising: a sensor switching determination unit that determines whether to switch a sensor to be used as a sensor for controlling the operation of a robot equipped with a sensor unit, out of multiple sensors arranged side by side along the circumferential direction of a substantially columnar sensor unit that is configured to be rotatable around a rotation axis along the longitudinal direction; and a rotation control unit that controls the rotation of the sensor unit, around the rotation axis, in which the sensor determined by the sensor switching determination unit to be switched is installed.

2. The control device according to claim 1, further comprising: a recognition unit that recognizes a target object that is the target of the robot's operation; and a sensor value processing unit that acquires sensor values ​​that indicate detection data detected by the sensor, wherein the sensor switching determination unit determines whether to switch the sensor in use based on recognition information that indicates the result of the recognition unit's recognition of the target object and the plurality of sensor values.

3. The control device according to claim 2, further comprising: a robot control unit that controls the operation of the robot; and a sensor value switching unit that switches the output so that the sensor value obtained from the sensor that is determined to be switched by the sensor switching determination unit is supplied to the robot control unit.

4. The control device according to claim 3, wherein the sensor switching determination unit determines whether to switch the sensor in use according to a preset rule.

5. The control device described in claim 4, wherein the sensor unit is provided with a distance sensor that measures the distance to the target object and a tactile sensor that detects the tactile sensation caused by contact with the target object, and the sensor switching determination unit determines to switch the sensor to be used in accordance with the rule that the distance sensor is used when the distance to the target object is equal to or greater than a threshold, and the tactile sensor is used when the distance to the target object is less than the threshold.

6. The control device according to claim 3, wherein the sensor switching determination unit performs predetermined processing on the recognition information obtained when the recognition unit recognizes the target object, and determines whether to switch the sensor to be used based on the processing result indicating the result of that processing.

7. The control device described in claim 3, wherein the sensor switching determination unit uses at least one of the recognition information obtained by the recognition unit recognizing the target object and the sensor value as input information, and determines whether to switch the sensor to be used using a neural network obtained by learning in which used sensor information indicating the optimal sensor to be used is output information.

8. The control device according to claim 3, wherein the sensor unit is a sensor finger that constitutes a finger portion of a robot hand, and the robot control unit controls the operation of the robot hand.

9. The control device according to claim 8, wherein the finger unit is configured by connecting a plurality of the sensor fingers via joints, the robot hand is configured with a plurality of the finger units, and the sensor switching determination unit determines whether or not to switch the sensor in use for each of the plurality of sensor units constituting the finger unit for each of the plurality of finger units.

10. The control device described in claim 1, wherein the rotation control unit controls the angles of the multiple sensor units relative to the target object so that the multiple sensors provided in one sensor unit can sense the target object simultaneously, and the multiple sensors are used simultaneously as the sensors in use.

11. The control device according to claim 1, wherein the rotation control section constantly rotates the sensor section at a predetermined rotation speed, and the plurality of sensors are alternately used as the sensors in use so as to be switched according to the rotation speed.

12. A control method including: a control device determining whether or not to switch a sensor to be used as a sensor to be used for controlling the operation of a robot equipped with a sensor unit, from among a plurality of sensors arranged side by side along the circumferential direction of a substantially columnar sensor unit configured to be rotatable around a rotation axis along the longitudinal direction; and controlling the rotation around the rotation axis of the sensor unit in which the sensor determined to be switched is installed.

13. A robot hand comprising: a sensor finger having a generally columnar shape, configured to be rotatable around a rotation axis along the longitudinal direction and having a plurality of sensors arranged on a side surface along the circumferential direction; a sensor switching determination unit that determines whether or not to switch a sensor from among the plurality of sensors to be used as a sensor to be used for controlling the operation of the robot hand that is equipped with the sensor finger; and a control device that controls the rotation of the sensor finger, around the rotation axis, that is equipped with the sensor that is determined to be switched by the sensor switching determination unit.

Images