Information processing device, information processing method, and program

Abstract

A configuration is provided that enables a robot to autonomously perform an optimal behavior in accordance with a surrounding environment of the robot or a state of the robot itself. The present invention includes: an environment and body feature extraction unit that receives an input of a detection value of a sensor mounted on a robot, and extracts at least any one of an environment feature amount relating to an environment around the robot or a physical feature amount relating to the robot itself; and a behavior control unit that determines a behavior to be performed by the robot based on the extracted feature amount, and performs the behavior. According to the present invention, an optimal behavior is determined and executed in response to a characteristic quantity, for example, a floor slip degree acquired as an environmental characteristic quantity, and in the case of slip, a walking speed is reduced and walking or the like is executed.

Description

Technical Field

[0001] The present disclosure relates to an information processing device, an information processing method, and a program. More specifically, the present invention relates to an information processing device, an information processing method, and a program in which a robot determines a behavior to be performed by the robot according to a surrounding environment and performs the determined behavior. Background Art

[0002] In recent years, a number of so-called entertainment robots have been used, such as dog-type robots that respond to users.

[0003] Most of such entertainment robots each have a sensor and a learning function, and use, for example, a camera to recognize a user's face and respond to the user.

[0004] Note that a robot having a sensor and a learning function is described in, for example, PTL1 (WO 2019 / 08478) or the like.

[0005] However, the conventional technology described in the above patent documents and the like only realizes a process of determining one behavior according to, for example, the user's voice or the surrounding environment and executing the determined behavior.

[0006] For example, the robot can perform an action of running in the direction of the user in response to a call from the user, but cannot perform detailed control of grasping the condition of the floor and adjusting the running speed.

[0007] [Citation list]

[0008] [Patent Document]

[0009] [PTL1]

[0010] WO 2019 / 08478 Summary of the invention

[0011] [Technical issues]

[0012] For example, in view of the above problems, the purpose of the present disclosure is to provide an information processing device, an information processing method, and a program, which enable the robot itself to control and execute the robot's behavior by taking into account various environmental information around the robot and physical information of the robot's own accessories, etc.

[0013] [Solution to the problem]

[0014] A first aspect of the present disclosure is an information processing device, the information processing device comprising:

[0015] an environment and body feature extraction unit that receives an input of a detection value of a sensor attached to the robot and extracts at least one of an environment feature amount that is a feature amount related to an environment around the robot and a body feature amount that is a feature amount related to the robot itself; and

[0016] A behavior control unit that determines and executes a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0017] In addition, a second aspect of the present disclosure is

[0018] An information processing method executed by an information processing device, comprising:

[0019] an environment and body feature extraction step in which an environment and body feature extraction unit receives an input of a detection value of a sensor attached to the robot and extracts at least one of an environment feature quantity as a feature quantity related to an environment around the robot and a body feature quantity as a feature quantity related to the robot itself; and

[0020] A behavior control step in which the behavior control unit determines and executes a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0021] In addition, a third aspect of the present disclosure is

[0022] A program for causing an information processing device to execute information processing, the program causing execution of:

[0023] an environment and body feature extraction step, in which an environment and body feature extraction unit receives an input of a detection value of a sensor attached to the robot, and extracts at least one of an environment feature quantity as a feature quantity related to the environment around the robot and a body feature quantity as a feature quantity related to the robot itself; and

[0024] A behavior control step in which the behavior control unit is caused to determine and execute the behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0025] Note that the program according to the present disclosure is a program that can be provided to an information processing device or a computer system that can execute various program codes, for example, through a storage medium or a communication medium provided in a computer-readable form. By providing such a program in a computer-readable form, processing according to the program can be implemented on an information processing device or a computer system.

[0026] Other objects, features and advantages of the present disclosure will become apparent through a more detailed description based on the embodiments of the present disclosure described below and the accompanying drawings. Note that in this description, a system is a logical collection of configurations of multiple devices, and devices having various configurations are not limited to devices in the same housing.

[0027] The configuration according to one embodiment of the present disclosure realizes a configuration that enables a robot to autonomously execute an optimal behavior that matches the surrounding environment of the robot or the state of the robot itself.

[0028] More specifically, for example, the configuration includes an environment and body feature extraction unit that receives an input of a detection value of a sensor attached to a robot and extracts at least one of an environment feature quantity as a feature quantity related to the environment around the robot and a body feature quantity as a feature quantity related to the robot itself; and a behavior control unit that determines and executes a behavior to be executed by the robot based on the extracted feature quantity. For example, the configuration acquires the slipperiness of a floor as an environment feature quantity, and in the case where the floor is slippery, for example, performs walking at a reduced walking speed, that is, determines and executes an optimal behavior that matches the feature quantity.

[0029] This configuration realizes a configuration that enables the robot to autonomously perform optimal behavior that matches the robot's surrounding environment or the robot's own state.

[0030] Note that the effects described in this specification are merely exemplary and not restrictive, and there may be additional effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] [ Figure 1 ]

[0032] Figure 1 is a diagram for explaining an example of a dog-type robot as an example of the information processing device according to the present disclosure.

[0033] [ Figure 2 ]

[0034] Figure 2 is a diagram for explaining a sensor attached to a robot according to the present disclosure.

[0035] [ Figure 3 ]

[0036] Figure 3 is a diagram for explaining a sensor attached to a robot according to the present disclosure.

[0037] [ Figure 4 ]

[0038] Figure 4: is a diagram for explaining a configuration example of a robot (information processing device) according to the present disclosure.

[0039] [ Figure 5 ]

[0040] Figure 5 It is a diagram for explaining a specific example of an environmental feature amount and a body feature amount acquired from a sensor detection value by the environmental and body feature amount extraction unit.

[0041] [ Figure 6 ]

[0042] Figure 6 This is a diagram for explaining specific examples of accessories that can be equipped on the robot.

[0043] [ Figure 7 ]

[0044] Figure 7 It is a diagram for explaining a specific example of an environmental feature amount and a body feature amount acquired from a sensor detection value by the environmental and body feature amount extraction unit.

[0045] [ Figure 8 ]

[0046] Figure 8 It is a diagram for explaining a specific example of an environmental feature amount and a body feature amount acquired from a sensor detection value by the environmental and body feature amount extraction unit.

[0047] [ Fig. 9 ]

[0048] Fig. 9 It is a diagram for explaining a specific example of a process in which the robot jumps and detects the reflected sound, hardness, and slipperiness of the floor.

[0049] [ Fig.10 ]

[0050] Fig.10 It is a diagram for explaining a specific example of a process in which the robot jumps and detects the reflected sound, hardness, and slipperiness of the floor.

[0051] [ Fig.11 ]

[0052] Fig.11 It is a diagram for explaining the types of environmental feature quantities and body feature quantities acquired by the environmental and body feature quantity extraction unit by receiving input of sensor detection values.

[0053] [ Fig.12 ]

[0054] Fig.12It is a diagram for explaining the types of environmental feature quantities and body feature quantities that the environmental and body feature quantity extraction unit obtains by receiving input of sensor detection values.

[0055] [ Fig.13 ]

[0056] Fig.13 It is a diagram for explaining the types of environmental feature quantities and body feature quantities that the environmental and body feature quantity extraction unit obtains by receiving input of sensor detection values.

[0057] [ Fig.14 ]

[0058] Fig.14 is a diagram for explaining an example of behavior control of the robot 100 performed by the behavior control unit of the robot.

[0059] [ Fig.15 ]

[0060] Fig.15 is a diagram for explaining an example of behavior control of the robot 100 performed by the behavior control unit of the robot.

[0061] [ Fig.16 ]

[0062] Fig.16 is a diagram for explaining an example of behavior control of the robot 100 performed by the behavior control unit of the robot.

[0063] [ Fig.17 ]

[0064] Fig.17 : is a diagram showing a flowchart for explaining a processing sequence executed by a robot (information processing device) according to the present disclosure.

[0065] [ Fig.18 ]

[0066] Fig.18 : is a diagram showing a flowchart for explaining a processing sequence executed by a robot (information processing device) according to the present disclosure.

[0067] [ Fig.19 ]

[0068] Fig.19 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0069] [ Fig. 20 ]

[0070] Fig. 20 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0071] [ Fig.21 ]

[0072] Fig.21 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0073] [ Fig. 22 ]

[0074] Fig. 22 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0075] [ Fig.23 ]

[0076] Fig.23 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0077] [ Fig.24 ]

[0078] Fig.24 : is a diagram for explaining a configuration example of a robot (information processing device) according to the present disclosure.

[0079] [ Fig.25 ]

[0080] Fig.25 : is a diagram for explaining a configuration example of a robot (information processing device) according to the present disclosure.

[0081] [ Fig.26 ]

[0082] Fig.26 : is a diagram for explaining a configuration example of a robot (information processing device) according to the present disclosure.

[0083] [ Fig. 27 ]

[0084] Fig. 27 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0085] [ Fig.28 ]

[0086] Fig.28 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0087] [ Fig.29 ]

[0088] Fig.29 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0089] [ Fig.30 ]

[0090] Fig.30 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0091] [ Fig.31 ]

[0092] Fig.31 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0093] [ Fig.32 ]

[0094] Fig.32 is a diagram for explaining a specific example of a process performed by a robot according to the present disclosure.

[0095] [ Fig.33 ]

[0096] Fig.33 : is a diagram for explaining a hardware configuration example of a robot (information processing device) according to the present disclosure. DETAILED DESCRIPTION

[0097] Hereinafter, details of the information processing device, information processing method, and program according to the present disclosure will be described with reference to the drawings. Note that the explanation will be given according to the following items.

[0098] 1. Configuration of a robot as an information processing device according to the present disclosure and attached sensors

[0099] 2. Configuration Example of the Robot (Information Processing Device) According to the Present Disclosure

[0100] 2-1. Details and Specific Examples of Processing Performed by Environment and Body Feature Extraction Unit

[0101] 2-2. Details and specific examples of processing performed by the behavior control unit

[0102] 3. Processing sequence executed by the robot (information processing device) according to the present disclosure

[0103] 4. Other configuration examples of the robot (information processing device) according to the present disclosure

[0104] 5. Specific example of processing using a robot (information processing device) according to the present disclosure

[0105] 6. Hardware Configuration Example of the Robot (Information Processing Device) According to the Present Disclosure

[0106] 7. Overview of the Configuration of the Present Disclosure

[0107] [1. Configuration of a robot as an information processing device according to the present disclosure and attached sensors]

[0108] First, the configuration of a robot as an information processing device according to the present disclosure and attached sensors will be described.

[0109] Figure 1 An example of a dog-type robot is shown as an example of the information processing device according to the present disclosure.

[0110] Figure 1 The robot (information processing device) 100 shown in FIG. 1 is a dog-type robot, and is a so-called entertainment-type robot that interacts with a user.

[0111] Figure 1 A robot (information processing device) 100 shown in FIG. 1 is a robot that moves autonomously, for example, that is, performs behavior autonomously.

[0112] The robot 100 performs actions according to the user's instructions, for example, in response to the user's voice, etc. When sensors (camera and microphone) attached to the robot 100 detect the user's call or beckoning, the robot 100 performs an autonomous behavior of running close to the user.

[0113] In addition, the robot can also control facial expressions. For example, when the sensor (camera) detects the user's smile, the robot 100 can also perform a feeling expression showing a smiling expression.

[0114] Note that, for example, the eyes of the robot 100 are configured as a display such as a liquid crystal display, for example, and various eyes associated with various emotions, such as eyes corresponding to a smile, sad eyes, and angry eyes, can be displayed.

[0115] The data processing unit in the robot 10 receives input of detection information of various sensors attached to the robot 100 , and uses the input detection information of the sensors to determine what behavior to perform, that is, what action or feeling expression (presentation) to perform.

[0116] Note that the “behavior” of the robot 100 in this specification includes both the “action” of the robot and the “feeling expression (presentation)” of the robot 100 .

[0117] The data processing unit in the robot 100 uses a machine learning algorithm such as deep learning to determine the optimal behavior pattern that matches the surrounding environment (situation), the robot's physical condition such as the equipment condition of the accessories, or the load (e.g., torque) of the joint components.

[0118] Note that the robot 100 adopts a configuration that enables various accessories such as clothing, shoes, and sole pads to be equipped. Specific examples of these accessories will be described later.

[0119] Will refer to Figure 2 and Figure 3 Sensors attached to the robot 100 are described.

[0120] like Figure 2 As shown in , various sensors are attached to the robot 100 .

[0121] Notice, Figure 3 It is used to illustrate Figure 2 Details of the sensor shown in the figure.

[0122] like Figure 2 and Figure 3 As shown in , for example, the following sensors are attached to the robot 100 .

[0123] S01: Image Sensor

[0124] S02: Voice Sensor

[0125] S03: Acceleration sensor

[0126] S04: Illuminance sensor

[0127] S05: Wind direction / pressure sensor

[0128] S06: Smell Sensor

[0129] S07: Temperature sensor

[0130] S08: Distance Sensor

[0131] S09: Joint Angle Sensor

[0132] S10: Joint torque sensor

[0133] S11: Foot sensor

[0134] S12: Odometer

[0135] Notice, Figure 2 and Figure 3 The sensors shown in are examples of sensors attached to the robot 100 , and a configuration in which other types of sensors are attached may also be employed.

[0136] We will describe Figure 2 and Figure 3 The sensor shown in .

[0137] The image sensor S01 includes, for example, an RGB camera device, and captures an image. Figure 2As shown in , the image sensor S01 is attached to the nose of the robot 100. Basically, the image sensor S01 continuously captures moving images.

[0138] Note that although Figure 2 Only one image sensor S01 is shown, but a configuration including a plurality of camera devices that respectively photograph the front / rear / left / right and up / down directions of the robot 100 may be adopted.

[0139] More specifically, the voice sensor S02 is a microphone for acquiring sound. Figure 2 As shown in , the voice sensor S02 is attached to the ear of the robot 100.

[0140] Note that although Figure 2 Two voice sensors S02 at the left and right ears are shown, but one microphone may also be used, or a configuration including three or more microphones may be employed.

[0141] The acceleration sensor S03 detects the movement of the robot, such as acceleration, angular acceleration, vibration, etc. The acceleration sensor S03 is, for example, Figure 2 Attached to the abdomen of the robot 100 is shown in FIG.

[0142] More specifically, for example, an inertial measurement unit (IMU) is used.

[0143] The illumination sensor S04 detects the brightness around the robot 100. The illumination sensor S04 is, for example, Figure 2 Attached to the eye of the robot 100 is shown in FIG.

[0144] The wind direction / wind pressure sensor S05 detects wind direction and wind force. Figure 2 Attached to the mouth of the robot 10 is shown in FIG.

[0145] The odor sensor S06 detects the odor around the robot 100. The odor sensor S06 is, for example, Figure 2 Attached to the nose of the robot 100 is shown in FIG.

[0146] The odor sensor S06 includes a device for detecting various odors such as gas odor and alcohol odor.

[0147] The temperature sensor S07 detects the temperature around the robot 100. The temperature sensor S07 is, for example, Figure 2 Attached to the head of the robot 100 is shown in FIG.

[0148] The distance sensor S08 detects the distances of various objects around the robot 100. The distance sensor S08 is, for example, Figure 2 Attached to the eye of the robot 10 is shown in FIG.

[0149] The distance sensor S08 is configured as, for example, a millimeter wave radar, an acoustic wave sensor, a light detection and ranging or laser imaging detection and ranging (LiDAR), a time of flight (ToF) sensor, a stereo camera device, etc.

[0150] In the case of using a sound wave sensor, by outputting the barking of a dog from a speaker of the robot 100, receiving input of reflected waves of the barking through a microphone, and analyzing the microphone input signal, the distance and direction to an object that has reflected the output sound can be analyzed.

[0151] The joint angle sensor S09 is attached to the body connection part, knee and ankle of the leg part, neck on the head side, etc. of the robot 100, and detects the angles of these joint parts. The joint angle sensor S09 is configured as, for example, an optical encoder, a magnetic encoder, etc.

[0152] The joint torque sensor S10 is attached to the body connection part, knee and ankle of the leg of the robot 100, the neck on the head side, etc., and detects the torque (torque (load)) of these joint parts. For example, a measuring instrument of the joint motor current value is used.

[0153] The sole sensor S11 is configured as a hardness sensor or a friction sensor, and detects an object contacted by the robot 100 , for example, the hardness of a floor on which the robot 100 stands or the friction of the floor.

[0154] The foot sensor S11 is, for example, Figure 2 . More specifically, for example, the sole sensor S11 detects whether the floor is made of stone such as marble or a wooden material such as a wooden floor, whether it is carpeted, or whether the robot 100 is on a sofa or a futon. In addition, the sole sensor S11 detects, for example, whether the floor is slippery.

[0155] The odometer S12 is attached to the body connection part, knee, ankle, etc. of the leg of the robot 100. The odometer S12 is, for example, an encoder, and detects the motion of each joint part, and estimates the moving distance of the robot 10 or the robot's own position.

[0156] Note that, as described above, Figure 2 and Figure 3 The sensors shown in are examples of sensors attached to the robot 100 , and a configuration in which other types of sensors are attached may be employed.

[0157] [2. Configuration Example of Robot (Information Processing Device) According to the Present Disclosure]

[0158] Next, a configuration example of a robot (information processing device) according to the present disclosure will be described.

[0159] Figure 4 is a diagram showing a configuration example of a robot (information processing device) 100 according to the present disclosure.

[0160] like Figure 4 As shown in , a robot (information processing device) 100 according to the present disclosure includes a sensing unit 110 and a data processing unit 120 .

[0161] The sensing unit 110 includes a reference Figure 2 and Figure 3 Describes the various sensors that collect information about the surrounding environment.

[0162] like Figure 4 As shown in , the data processing unit 120 includes an environment and body feature extraction unit 121 and a behavior control unit 122.

[0163] The behavior control unit 122 includes a behavior determination unit 123 , a behavior switching control unit 124 , a behavior execution unit 125 , an action control unit 126 , and an expression control unit 127 .

[0164] The environment and body feature extraction unit 121 receives input of various sensor detection data from the sensing unit 110, and extracts feature quantities of the environment around the robot and body feature quantities of the robot itself using a machine learning model based on the input data.

[0165] The environmental feature quantities include various environmental information, such as information about objects around the robot, the material and hardness of the robot's running surface (floor surface), time, temperature, the user's state (smiling, angry, etc.), and noisy state. Specific examples of these environmental feature quantities will be described later.

[0166] The body feature quantities of the robot itself are feature quantities indicating whether accessories such as clothing, shoes and sole pads that the robot can be equipped with are equipped, the driving unit of the robot, and the conditions of the load (torque) of the joints of the legs, the battery status, etc. Specific examples of these body feature quantities will also be described later.

[0167] Next, the behavior control unit 122 will be described.

[0168] The behavior control unit 122 controls the behavior to be performed by the robot 100 .

[0169] As described above, the “behavior” of the robot 100 in the present specification includes both the “action” of the robot 100 and the “feeling expression (presentation)” of the robot 100 .

[0170] The behavior control unit 122 controls at least one of “action” to be performed by the robot 100 and “feeling expression (presentation)” to be performed by the robot 100 according to the feature amount of the environment around the robot or the body feature amount of the robot itself.

[0171] like Figure 4 As shown in , the behavior control unit 122 includes a behavior determination unit 123 , a behavior switching control unit 124 , a behavior execution unit 125 , an action control unit 126 , and an expression control unit 127 .

[0172] Hereinafter, processing performed by each of these components will be described.

[0173] The behavior determination unit 123 determines a behavior to be performed in response to a sensor input of the sensing unit 110 , a user's motion input via, for example, an image sensor (camera), or a user's voice input via a voice sensor (microphone).

[0174] Furthermore, the behavior determination unit 123 also performs processing for determining a behavior to be performed by the robot 100 based on the feature amount of the environment around the robot extracted by the environment and body feature amount extraction unit 121 and the body feature amount of the robot itself.

[0175] The behavior determination unit 123 receives inputs of the surrounding environment (situation), the physical condition of the robot such as the equipment condition of accessories, and the load of each joint (e.g., torque), applies a machine learning algorithm such as deep learning, and determines the optimal behavior pattern that matches the current environment or physical condition.

[0176] Note that, as described above, the “behavior” of the robot 100 in the present specification includes both the “action” of the robot and the “feeling expression (presentation)” of the robot 100 .

[0177] That is, the behavior determination unit 123 determines at least one of an “action” to be performed by the robot 100 and a “feeling expression (presentation)” to be performed by the robot 100 according to the feature quantity of the environment around the robot or the body feature quantity of the robot itself.

[0178] Note that the behavior determination unit 123 performs processing of determining whether there is a behavior that needs to be performed to acquire a valid or necessary sensor detection value from the sensing unit 110 , and determining the behavior as a behavior that needs to be performed if there is such a behavior.

[0179] As described above, the process of executing an action for acquiring a sensor detection value and acquiring the sensor detection value is referred to as an active sensor detection value acquisition process.

[0180] More specifically, for example, when the user's face is on the left side and the robot needs to check the user's facial expression, the behavior determination unit 123 determines the behavior of pointing the direction of the robot 10's face, that is, the direction of the camera device serving as a sensor, in the direction of the user's face as a behavior that needs to be performed, and causes the behavior execution unit 125 to perform the behavior.

[0181] This processing makes it possible to acquire an image of the user's face using an image sensor (camera).

[0182] In addition, for example, when the illumination is low (such as at night) and the captured image of the camera device is not clear, the behavior determination unit 123 determines the behavior of causing the lamp attached to the robot 100 to emit light as a behavior that needs to be performed, and causes the behavior execution unit 125 to perform the behavior.

[0183] This processing makes it possible to acquire a bright face image using an image sensor (camera).

[0184] In addition, for example, in the case where the robot 100 attempts to run on the floor, the behavior determination unit 123 will determine the behavior of stepping on the floor by swinging the robot's legs up and down to check whether the floor is hard or slippery, for example, as a behavior that needs to be performed, and causes the behavior execution unit 125 to execute the behavior.

[0185] The processing makes it possible to use the foot sole sensor to obtain information such as whether the floor is hard or slippery.

[0186] As described above, the behavior determination unit 123 also performs a process of determining a behavior for acquiring a valid or necessary sensor detection value from the sensing unit 110 as a behavior that needs to be performed.

[0187] The behavior switching control unit 124 determines whether the behavior determined by the behavior determination unit 123 is different from the currently executed behavior, and if the behavior is different, outputs behavior pattern information requiring switching and a behavior switching request to the behavior execution unit 125 .

[0188] The behavior execution unit 125 controls the execution of various behaviors of the robot 100. More specifically, the behavior execution unit 125 controls the driving unit 126 and the display control unit 127 to control the motion or expression of the robot 100.

[0189] Note that, upon receiving input of behavior pattern information that needs to be switched and a behavior switching request from the behavior switching control unit 124 , the behavior execution unit 125 executes processing for switching the behavior of the robot 100 according to the input information.

[0190] The driving unit 126 includes driving mechanisms such as motors and actuators for moving various component parts such as the legs, neck, body, and tail of the robot 100, and performs driving processing for realizing the movement of the robot according to instructions of the behavior execution unit.

[0191] The display control unit 127 controls, for example, an image of the eyes of a display at the eye portion of the face of the robot. By performing this control, the display control unit 127 can change the expression of the robot into various states. The display control unit 127 performs display control processing to embody the expression of the robot matching the instruction of the behavior execution unit, such as a happy expression, a smiling expression, a sad expression, a confused expression, and an angry expression.

[0192] Next, we will describe Figure 4 Specific processing examples of the components of the data processing unit 120 shown in FIG. 1 , namely, these environment and body feature extraction unit 121 and behavior control unit 122.

[0193] (2-1. Details and Specific Examples of Processing Performed by Environment and Body Feature Extraction Unit)

[0194] First, we will describe Figure 4 Details and specific examples of the processing performed by the environment and body feature extraction unit 121 of the data processing unit 120 of the robot (information processing device) 100 shown in FIG.

[0195] As described above, the environment and body feature extraction unit 121 receives input of various sensor detection data from the sensing unit 110, and uses a machine learning model based on the input data to extract feature quantities of the environment around the robot and body feature quantities of the robot itself.

[0196] Note that the environmental and physical feature extraction unit 121 not only performs processing of passively receiving input of sensor detection values ​​detected by the sensors of the sensing unit 110 , but also performs active sensor detection value input processing as processing of receiving input of sensor detection values ​​from the sensing unit 110 .

[0197] The active sensor detection value input process is a process in which the robot 100 performs a specific behavior to obtain a specific sensor detection value and acquires the sensor detection value that becomes necessary as a result of the behavior.

[0198] As described previously, the behavior determination unit 123 determines a behavior for acquiring a valid or necessary sensor detection value from the sensing unit 110 as a behavior that needs to be executed, and causes the behavior execution unit 125 to execute the behavior.

[0199] The environment and body feature extraction unit 121 also performs processing of acquiring active sensor detection values ​​as a result of such behavior.

[0200] Will refer to Figure 5 The drawings thereafter describe specific examples of environmental feature quantities and body feature quantities acquired by the environmental and body feature quantity extraction unit 121 from input sensor detection values ​​by receiving input of sensor detection values ​​from the sensing unit 110 .

[0201] Figure 5 An example of passive detection information is shown.

[0202] Figure 5 Shown are tables associated with:

[0203] (a) sensor type (sensor to be used),

[0204] (b) Examples of environmental feature quantities to be extracted, and

[0205] (c) Examples of body feature quantities to be extracted.

[0206] Figure 5 Item (1) in indicates an example following passive detection information.

[0207] (a) Sensor type (sensors to be used) = voice sensor, IMU, joint torque sensor, joint angle sensor and odometer,

[0208] (b) Examples of environmental feature quantities to be extracted = people around the robot, vibrations caused by the robot, voices and footsteps, and sounds caused when handling objects (e.g., the sound of glass colliding), and

[0209] (c) Example of body feature quantity to be extracted = voice of a robot (e.g., footsteps)

[0210] The passive detection information example (1) indicates that upon receiving the input of detection information from at least one of a voice sensor, an IMU, a joint torque sensor, a joint angle sensor, and an odometer,

[0211] The environmental and physical feature extraction unit 121 extracts at least one of the environmental feature quantities of people around the robot, vibrations, sounds and footsteps caused by the robot, and sounds caused when processing objects (for example, the sound of glass colliding) as environmental feature quantities, and extracts at least one of the sounds (for example, footsteps) and vibrations of the robot as physical feature quantities.

[0212] The next passive detection information example (2) indicates that upon receiving the input of detection information of at least one of the image sensor, the voice sensor, and the sole sensor,

[0213] The environment and body feature amount extraction unit 121 extracts the type of floor surface (for example, stone, wood, or carpet) or the slipperiness of the floor, and

[0214] Whether the robot is equipped with accessories and the types of accessories equipped are extracted as body feature quantities.

[0215] Note that, as described above, the robot 100 adopts a configuration that enables various accessories such as clothing, shoes, and sole pads to be equipped.

[0216] Will refer to Figure 6 Specific examples of accessories that the robot 100 can be equipped with are described.

[0217] like Figure 6 As shown in , the robot 100 may be equipped with various accessories, such as clothing 151 , shoes 152 , and sole pads 153 .

[0218] When the robot 100 walks or runs, the shoes 152 and the sole pads 153 provide anti-slip or anti-noise effects.

[0219] Figure 5 The data (2) shown in indicates that the environment and body feature quantity extraction unit 121 receives input of detection information of one of the image sensor, voice sensor, and sole sensor, and determines whether the robot 100 is equipped with accessories such as shoes or sole pads.

[0220] For example, the environmental and body feature extraction unit 121 performs processing to determine whether the robot 100 is equipped with accessories such as shoes or insoles based on an image of the robot 100's legs, or performs processing to determine whether the robot 100 is equipped with accessories such as shoes or insoles based on the footsteps caused when the robot 100 walks.

[0221] Figure 5 The data of the passive detection information examples (3) to (9) shown in are as follows.

[0222] The passive detection information example (3) indicates that the environment and body feature extraction unit 121 receives input of detection information from one of the joint torque sensor and the joint angle sensor, and extracts the load (e.g., torque) of the joint of the robot as a body feature.

[0223] The passive detection information example (4) indicates that the environment and body feature amount extraction unit 121 receives input of detection information of the image sensor and extracts the position of an object (obstacle) around the robot as a body feature amount.

[0224] The passive detection information example (5) indicates that the environment and body feature quantity extraction unit 121 receives the input of the detection information of the distance sensor, and extracts the distance to the object (e.g., object) around the robot and the shape of the object as the environment feature quantity.

[0225] The passive detection information example (6) indicates that the environment and body feature extraction unit 121 receives the input of the detection information of the temperature sensor, extracts the surrounding temperature as the environment feature, and extracts the temperature (fever state) of the robot as the body feature.

[0226] The passive detection information example (7) indicates that the environment and body feature extraction unit 121 receives the input of detection information from the wind direction / wind pressure sensor, extracts the wind direction and wind pressure around the robot as environmental feature quantities, and extracts the running speed and running direction of the robot as body feature quantities.

[0227] The passive detection information example (8) indicates that the environment and body feature quantity extraction unit 121 receives the input of the detection information of the odor sensor and extracts the odor around the robot as the environment feature quantity.

[0228] The passive detection information example (9) indicates that the environment and body feature quantity extraction unit 121 receives the input of the detection information of the illuminance sensor and extracts the brightness around the robot as the environment feature quantity.

[0229] As described above, the environment and body feature amount extraction unit 121 receives input of various sensor detection values ​​from the sensing unit 110 , and acquires various environment feature amounts and body feature amounts from the input sensor detection values.

[0230] Reference Figure 5 The example of feature quantity extraction processing by the described environment and body feature quantity extraction unit 121 is a processing example in which the environment and body feature quantity extraction unit 121 passively receives input of various sensor detection values ​​from the sensing unit 110 and obtains environmental feature quantities and body feature quantities.

[0231] As described above, the environment and body feature extraction unit 121 not only performs processing for passively receiving input of sensor detection values ​​detected by the sensors of the sensing unit 110, but also performs active sensor detection value input processing, in which the robot 100 performs specific behavior to obtain specific sensor detection values ​​and obtains sensor detection values ​​that become necessary as a result of the behavior.

[0232] For example, the behavior determination unit 123 determines a behavior for acquiring a valid or necessary sensor detection value from the sensing unit 110 as a behavior that needs to be performed, and causes the behavior execution unit 125 to execute the behavior. The environment and body feature extraction unit 121 also performs processing for acquiring an active sensor detection value as a result of such behavior.

[0233] Will refer to Figure 7 and Figure 8 A specific example of a process for performing active sensor detection value input processing (i.e., active sensor detection value input processing in which the robot 100 performs a specific behavior to obtain a specific sensor detection value, and obtains sensor detection values ​​that become necessary as a result of the behavior, and obtains environmental feature quantities and body feature quantities) is described.

[0234] Figures 7 and 8 An example of active detection information is shown.

[0235] Figure 7 and Figure 8 Shown are tables associated with:

[0236] (a) sensor type (sensor to be used),

[0237] (b) the active action to be performed,

[0238] (c) Examples of environmental feature quantities to be extracted, and

[0239] (d) Examples of body feature quantities to be extracted.

[0240] Figure 7 Item (1) in indicates an example of the following active detection information.

[0241] (a) Sensor type (sensor to be used) = image sensor,

[0242] (b) Active action to be performed = emit light and acquire bright captured image,

[0243] (c) Examples of environmental feature quantities to be extracted = whether there is an object (obstacle) around the robot and the position of the object, and

[0244] (d) Example of body feature quantity to be extracted = whether the robot is equipped with accessories and the type of accessories

[0245] The active detection information example (1) indicates that when using the image sensor, the environment and body feature quantity extraction unit 121 emits light, acquires a captured image,

[0246] Extract whether there are objects (obstacles) around the robot and the location of the objects as environmental features, and

[0247] Whether the robot is equipped with accessories and the types of accessories are extracted as body feature quantities.

[0248] Note that the light is a light attached to the robot 100 , and emits light under the control of the behavior execution unit 125 .

[0249] The next active detection information example (2) indicates that, when using an image sensor, the environment and body feature extraction unit 121 moves the head, controls the direction of the camera, obtains captured images in various directions, extracts whether there are objects (obstacles) around the robot and the position of the object as environmental features, and extracts whether the robot is equipped with accessories and the type of accessories as body features.

[0250] For example, by pointing a head to which an image sensor (camera) has been attached to capture images of the legs toward the legs and capturing images of the legs, it is possible to check whether the legs of the robot are equipped with accessories such as shoes and sole pads.

[0251] Figure 7 The data of the active detection information examples (3) to (7) shown in are as follows.

[0252] The active detection information example (3) indicates that, when using a voice sensor, the environment and body feature extraction unit 121 outputs the barking action (i.e., barking voice) through a speaker, obtains the reflected sound of the barking voice, and obtains the distance to the object (obstacle) around the robot and the direction of the object (obstacle) as environmental feature quantities.

[0253] The active detection information example (4) indicates that, using a voice sensor or a sole sensor, the environment and body feature extraction unit 121 makes the robot jump, detects the reflected sound, hardness, and slipperiness of the floor, identifies the type of floor surface (for example, stone, wood, or carpet) as environmental features, and extracts whether the robot's legs are equipped with accessories and the type of accessories equipped as body features.

[0254] Will refer to Fig. 9 A specific example of this processing is described.

[0255] Fig. 9 A processing example in the case where the robot 100 jumps is shown.

[0256] The robot 100 stands on the marble floor at time (t1).

[0257] The robot 100 switches to the jumping system at the next time (t2), jumps at (t3), and lands at (t4).

[0258] When landing at time (t4), the legs of the robot 100 collide with the marble floor and cause a loud collision sound (collision sound).

[0259] The voice sensor (microphone) of the robot 100 acquires the loud collision sound, and inputs the collision sound to the environment and body feature quantity extraction unit 121 .

[0260] The environment and body feature extraction unit 121 obtains a feature indicating that the type of the floor surface is stone as an environment feature based on the large collision sound input from the voice sensor (microphone). In addition, the environment and body feature extraction unit 121 obtains a feature indicating that the robot's legs are not equipped with shoes or sole pads as a body feature.

[0261] Fig.10 An example of the process is shown where the floor is a soft "carpet".

[0262] The robot 100 stands on the carpet floor at time (t1).

[0263] The robot 100 switches to the jumping system at the next time (t2), jumps at (t3), and lands at (t4).

[0264] When landing at time (t4), the legs of the robot 100 collide with the carpet, and cause a light collision sound (banging sound).

[0265] The voice sensor (microphone) of the robot 100 acquires the light collision sound, and inputs the collision sound to the environment and body feature quantity extraction unit 121 .

[0266] The environment and body feature extraction unit 121 obtains a feature indicating that the type of the floor surface is a carpet as an environment feature based on the light collision sound input from the voice sensor (microphone). In addition, the environment and body feature extraction unit 121 obtains a feature indicating that the legs of the robot are not equipped with shoes or sole pads as a body feature.

[0267] Back to Figure 7 , other active detection information examples will be described.

[0268] The active detection information example (5) indicates that, using a voice sensor or a sole sensor, the environment and body feature extraction unit 121 makes the robot step without making the robot jump, detects the reflected sound and hardness of the floor, identifies the type of floor surface (such as stone, wood, or carpet) as environmental feature, and extracts whether the robot's legs are equipped with accessories and the type of accessories equipped as body feature.

[0269] The active detection information example (6) indicates that, in the case of using an image sensor, the environment and body feature quantity extraction unit 121 captures an image by throwing a ball, and obtains the concave-convex information of the floor surface around the robot as a processing example of the environment feature quantity.

[0270] When the ball disappears from the image, the environment and body feature quantity extraction unit 121 determines that the floor has steps or a "cliff".

[0271] The active detection information example (7) indicates that, when using a voice sensor, the environmental and physical feature extraction unit 121 obtains the bouncing sound by throwing a ball, and obtains information on the type of floor surface (such as stone, wood, or carpet) around the robot as an example of processing environmental feature quantities.

[0272] For example, the environment and body feature amount extraction unit 121 determines the floor type as “stone such as marble” in the case of a hard sound, and determines the floor type as “carpet” in the case of a soft sound.

[0273] exist Figure 8 In the active detection information example (8) shown in , in the case of using a voice sensor, the environment and body feature quantity extraction unit 121 performs processing in which the robot touches or hits an object and obtains the sound of the touch or hit. The active detection information example (8) indicates that the environment and body feature quantity extraction unit 121 obtains material information about the type of object (e.g., stone, wood, or cloth) around the robot from the sound obtained by the processing as an environmental feature quantity, and extracts whether the robot is equipped with an accessory, the type of accessory, the load of the joint (e.g., torque), etc. as a physical feature quantity.

[0274] In the active detection information example (9), in the case of using a joint torque sensor or a joint angle sensor, the environment and body feature quantity extraction unit 121 causes the robot to add an object thereto. The environment and body feature quantity extraction unit 121 obtains the weight and slipperiness of the added object as environmental feature quantities based on the information about the joint torque and joint angle of, for example, the robot's neck obtained through the processing.

[0275] In the active detection information example (10), when using an image sensor, the environment and body feature extraction unit 121 captures a facial image of a person by facing the direction of the person. Based on the facial image of the person obtained by this process, it is determined whether the person is someone the robot has encountered before.

[0276] Note that when executing this process, the robot 100 acquires a past captured image from a database that has stored previously captured facial images, and executes a process of cross-checking the past captured image and the captured image.

[0277] In the active detection information example (11), in the case of using the voice sensor, the environment and body feature extraction unit 121 performs a process of calling a person. That is, the environment and body feature extraction unit 121 outputs a calling sound via a speaker. The environment and body feature extraction unit 121 determines whether there is a person around the robot and whether the person is a person that the robot has met before based on the response voice of the person obtained through the process.

[0278] Note that, while executing this process, the robot 100 also acquires the past spoken voice from the database which has stored the previously acquired spoken voice of a person, and executes a process of cross-checking the past spoken voice and the voice.

[0279] In the active detection information example (12), in the case of using an image sensor or a voice sensor, the environment and body feature extraction unit 121 takes an action of showing a sad expression or looking away. The environment and body feature extraction unit 121 determines the personality of the person, that is, whether the person is, for example, friendly or indifferent, based on the person's reaction, that is, the facial image or response voice obtained by the processing.

[0280] In the active detection information example (13), in the case of using an image sensor, a voice sensor, and a sole sensor, the environment and body feature extraction unit 121 takes the action of touching and looking at a person. The environment and body feature extraction unit 121 determines the personality of the person, that is, whether the person is, for example, friendly or indifferent, based on the person's reaction, that is, the facial image, response voice, or touch method obtained by the processing.

[0281] As described above, the environment and body feature extraction unit 121 performs the following processing: receiving and analyzing the input of sensor detection values ​​input from the sensor of the sensing unit 110 as a result obtained by performing various behaviors, and acquiring various environment feature quantities and body feature quantities.

[0282] Next, refer to Figures 11 to 13 The types of environmental feature amounts and body feature amounts acquired by the environmental and body feature amount extraction unit 121 by receiving input of detection values ​​of the sensors of the sensing unit 110 are described.

[0283] First, refer to Fig.11 and Fig.12 The types of environmental feature quantities acquired by the environmental and body feature quantity extraction unit 121 by receiving input of detection values ​​of the sensors of the sensing unit 110 are described.

[0284] like Figure 11 to Figure 12As shown in , examples of the environmental feature amounts acquired by the environmental and body feature amount extraction unit 121 by receiving input of detection values ​​of the sensors of the sensing unit 110 include the following feature amounts.

[0285] (a) Feature quantities acquired through terrain and interaction with objects

[0286] (b) Feature quantities obtained through interaction with people

[0287] (c) Feature quantities acquired through interaction with organisms

[0288] (d) Features acquired through interaction with another autonomous system

[0289] “(a) Feature quantities acquired through terrain and interaction with an object” are environmental feature quantities acquired based on sensor detection values ​​associated with terrain or an object (e.g., terrain in a space such as a room where the robot 100 exists or an image associated with an object in the space). More specifically, examples of “(a) Feature quantities acquired through terrain and interaction with an object” include the following environmental feature quantities.

[0290] (a1) Environmental feature quantities related to the structure and properties of the room in which the robot is currently located

[0291] (a2) Environmental characteristics related to the material and structure of the current interactive object

[0292] (a3) Weather-related environmental characteristics

[0293] More specifically, examples of “(a1) environmental feature quantities related to the structure and attributes of the room in which the robot currently exists” include the following environmental feature quantities.

[0294] The material of the floor surface of the room the robot is currently in = hardness, friction, roughness, and concavity / convexity

[0295] · Known room similarity = the degree of change from past observations, even in the case of the same room

[0296] Room usage = kitchen similarity, living room similarity, office similarity, etc.

[0297] Room temperature

[0298] The smell of the room

[0299] For example, “(a1) environmental feature quantities related to the structure and attributes of the room in which the robot currently exists” includes these environmental feature quantities.

[0300] Furthermore, more specifically, examples of “(a2) environmental feature quantities related to the material and structure of the current interactive object” include the following environmental feature quantities.

[0301] Ease of moving objects = e.g., whether toys, obstacles, and doors are movable

[0302] The softness, weight, and slipperiness of the object

[0303] Furthermore, more specifically, examples of “(a3) environmental characteristic amounts related to weather” include the following environmental characteristic amounts.

[0304] Is the weather good or bad = Is it sunny, cloudy, rainy, snowy or hail?

[0305] “(b) Feature amounts acquired through interaction with people” include environmental feature amounts acquired based on sensor detection values ​​associated with people detected by the robot 100. More specifically, examples of “(b) Feature amounts acquired through interaction with people” include the following environmental feature amounts.

[0306] (b1) Frequency of interaction

[0307] For example, whether the person is someone the robot sees every day, someone the robot has not seen for a long time, or someone the robot sees for the first time.

[0308] (b2) Related to changes in people

[0309] For example, the person cuts their hair, changes their clothing, becomes taller, or gains weight.

[0310] (b3) The feeling of the person in front

[0311] For example, facial expressions, tone of voice, and laughter

[0312] (b4) How people interact

[0313] For example, whether the person hits the robot hard, hits the robot lightly, touches the robot, or ignores the robot even though the robot is in the person's field of vision.

[0314] (b5) Are there many people around?

[0315] Furthermore, “(c) feature quantities obtained through interaction with a living being” includes environmental feature quantities obtained based on sensor detection values ​​associated with a living being other than a human being, such as a dog or a cat, detected by the robot 100. More specifically, examples of “(c) feature quantities obtained through interaction with a living being” include the following environmental feature quantities.

[0316] (c1) Is the type of the organism a known type?

[0317] (c2) Interactions from Biological Sources

[0318] For example, whether the creature barks or avoids

[0319] (c3) Whether the animal is a pet owned by the owner

[0320] (c4) Is the size of the organism roughly the same as its own size?

[0321] Furthermore, “(d) feature quantities acquired through interaction with other autonomous systems” include environmental feature quantities acquired based on sensor detection values ​​associated with another autonomous system of another robot, for example, detected by the robot 100. More specifically, examples of “(d) feature quantities acquired through interaction with another autonomous system” include the following environmental feature quantities.

[0322] (d1) Feature quantities obtained through interaction with another entertainment robot, such as conversation

[0323] Next, we will refer to Fig.13 The types of body feature quantities acquired by the environment and body feature quantity extraction unit 121 by receiving input of detection values ​​of the sensors of the sensing unit 110 are described.

[0324] like Fig.13 As shown in , examples of the body feature amount acquired by the environment and body feature amount extraction unit 121 by receiving input of the detection value of the sensor of the sensing unit 110 include the following feature amounts.

[0325] (p) Physical characteristics related to hardware (HW) failure

[0326] (q) Body feature quantity related to body deformation

[0327] (r) Physical characteristics related to accessory equipment

[0328] (s) Physical characteristics such as the occurrence of being tripped by an external object

[0329] “(p) Physical feature amounts related to a failure of own hardware (HW)” are physical feature amounts related to a failure of the hardware of the robot 100 , and include, for example, the following physical feature amounts.

[0330] (p1) Possibility of failure probability of the drive unit

[0331] The body characteristic amount is, for example, a body characteristic amount indicating that, for example, a target torque given as an instruction value cannot be achieved.

[0332] (p2) Possibility of failure probability of sensor unit

[0333] The body feature amount is, for example, a feature amount indicating that a sensor detection value cannot be acquired.

[0334] “(q) Body feature quantities related to deformation of the body” are body feature quantities related to deformation of component parts of the robot 100 , and include, for example, the following body feature quantities.

[0335] (q1) Accessories and equipment

[0336] The body feature amount is, for example, a feature amount indicating that, for example, a leg or an arm has been replaced, or a leg has been replaced with a wheel.

[0337] (q2) Changes to attachments

[0338] The body feature amount is, for example, a feature amount indicating whether an equipped accessory has been changed or whether an accessory has deteriorated over time.

[0339] (q3) Integration with another machine

[0340] “(r) Physical feature quantities related to the condition of equipment of accessories” are physical feature quantities related to the condition of equipment of accessories such as clothing or shoes that the robot 100 can be equipped with, and include, for example, the following physical feature quantities.

[0341] (r1) Likelihood of wearing foot pads (cushioning pads applied to the soles of the feet to reduce footsteps)

[0342] (r2) Whether the person is wearing clothes, shoes, etc.

[0343] “(s) Physical feature amounts such as occurrence of being entangled by an external object” are physical feature amounts related to a situation where the robot 100 is entangled by another object and cannot move, and include, for example, the following physical feature amounts.

[0344] (s1) Tripping over a rope

[0345] As mentioned above Figures 11 to 13 As described, the environmental feature quantities and body feature quantities obtained by the environmental and body feature quantity extraction unit 121 by receiving the input of the detection value of the sensor of the sensing unit 110 include various feature quantities related to the environment around the robot and the robot's own body.

[0346] The environmental feature amount and the body feature amount acquired by the environmental and body feature amount extraction unit 121 by receiving input of detection values ​​of the sensors of the sensing unit 110 are input to the behavior determination unit 123 .

[0347] (2-2. Details and Specific Examples of Processing Executed by Behavior Control Unit)

[0348] Next, we will describe Figure 4 Details and specific examples of the processing performed by the behavior control unit 122 of the data processing unit 120 of the robot (information processing device) 100 shown in FIG.

[0349] As previously referenced Figure 4 Described, Figure 4 The behavior control unit 122 of the data processing unit 120 of the robot (information processing device) 100 shown in FIG. 1 controls the behavior to be performed by the robot 100 .

[0350] Note that, as described above, the behavior to be performed by the robot 100 includes “feeling expression (presentation)” to be performed by the robot 100 in addition to “actions” such as movement of the robot 100, and the behavior control unit 122 controls the robot's actions and feeling expressions.

[0351] As previously referenced Figure 4 As described, the behavior control unit 122 includes a behavior determination unit 123 , a behavior switching control unit 124 and a behavior execution unit 125 .

[0352] The behavior determination unit 123 determines a behavior to be performed by the robot 100 based on the feature amount of the environment around the robot extracted by the environment and body feature amount extraction unit 121 or the body feature amount of the robot itself.

[0353] The behavior switching control unit 124 determines whether the behavior determined by the behavior determination unit 123 is different from the currently executed behavior, and if the behavior is different, outputs behavior pattern information requiring switching and a behavior switching request to the behavior execution unit 125 .

[0354] Furthermore, the behavior execution unit 125 controls execution of various behaviors of the robot 100. More specifically, the behavior execution unit 125 controls the driving unit 126 and the display control unit 127 to control the motion or expression of the robot 100.

[0355] As described above, Figure 4 The behavior control unit 122 of the data processing unit 120 of the robot (information processing device) 100 shown in the figure determines the behavior to be performed by the robot 100 based on the feature values ​​of the environment around the robot extracted by the environment and body feature value extraction unit 121 or the body feature values ​​of the robot itself, switches the behavior when the determined behavior is different from the currently performed behavior, and controls the behavior to be performed by the robot 100.

[0356] Next, we will refer to Figures 14 to 16 Description by Figure 4An example of behavior control of the robot 100 performed by the behavior control unit 122 of the data processing unit 120 of the robot (information processing device) 100 shown in FIG.

[0357] Examples of the behavior control of the robot 100 performed by the behavior control unit 122 include: Figures 14 to 16 The following behavior controls of various types are shown in .

[0358] (A) Mobile control

[0359] (B) Grip control

[0360] (C) Environmental Adaptive Behavior Control

[0361] (D) Feeling expression and behavior control

[0362] The following will refer to Figures 14 to 16 Describe specific examples of these behavioral controls.

[0363] (A) Mobile control

[0364] First, refer to Fig.14 A specific example of "(A) movement control" is described.

[0365] The "(A) movement control" of the robot 100 executed by the behavior control unit 122 is movement processing of the robot 100, that is, control processing related to movement such as walking and running. Specific examples of this "(A) movement control" include, for example, the following control.

[0366] (a1) Control for the purpose of improving accuracy

[0367] (a2) Control for the purpose of improving mobility

[0368] (a3) Control for the purpose of switching to an operation that does not use a damaged part

[0369] Specific examples of control to be performed as “(a1) control for the purpose of improving accuracy” include control processing such as the following:

[0370] (a11) Odometer Calibration Control

[0371] (a12) Gait change control.

[0372] "(a11) Correction control of the odometer" is an odometer correction process that matches the friction of the floor surface on which the robot 100 runs, and is a running distance measurement of the robot using the odometer or a correction process of the robot's own position estimation algorithm.

[0373] Upon receiving input of information related to the friction of the floor surface on which the robot 100 runs as an environmental feature quantity from the environment and body feature detection unit 121, the behavior control unit 122 performs odometer correction control that matches the input friction of the floor surface.

[0374] In the case where the friction of the floor surface on which the robot 100 runs is different, and even in the case where the robot 100 walks or runs in a hurry with the same number of steps, the movement amount will change. Therefore, the robot's running distance measurement or self-position estimation algorithm can be corrected using the odometer according to the friction of the floor surface, and the correct movement amount or self-position can be estimated.

[0375] “(a12) Gait change control” is a process of changing and controlling the walking pattern of the robot 100 in matching with the frictional force of the floor surface on which the robot 100 runs.

[0376] Upon receiving input of information related to the friction of the floor surface on which the robot 100 runs as an environmental feature amount from the environment and body feature detection unit 121, the behavior control unit 122 performs gait change control matching the input friction of the floor surface.

[0377] For example, if the friction of the floor surface on which the robot 100 runs is low, that is, if the floor surface is slippery, if the stride is increased, the robot 100 is likely to slip and fall. In addition, when the robot 100 performs walking at a high speed, such as running in a hurry, the robot 100 is likely to slip and fall.

[0378] In the case where the friction of the floor surface on which the robot 100 runs is low, that is, in the case where the floor surface is slippery, the walking mode is set so that the robot 100 walks with a reduced stride and a low walking speed to prevent such an accident from occurring.

[0379] Alternatively, the robot 100 may be configured to move by utilizing the slipperiness of the floor, ie, to move like skating instead of walking, for example.

[0380] On the other hand, for example, in the case where the friction of the floor surface on which the robot 100 runs is large, that is, the floor surface is hardly slippery, the robot 100 is unlikely to slip and fall even if the stride is increased. In such a case, control is performed so that the robot 100 walks with an increased stride.

[0381] Note that when the friction of the floor surface is large, that is, when the floor surface is almost not slippery, if the robot 100 performs high-speed walking such as running in a hurry, the robot 100 may fall, and therefore speed control matching the friction of the floor surface is performed.

[0382] Specific examples of the control to be performed as "(a2) control for the purpose of improving mobility" include control processing, for example,

[0383] (a21) Switching control to leaving the trapped place

[0384] (a22) Control switches to a behavior of acting on the object to enable movement.

[0385] “(a21) Switching control to the behavior of leaving the trapped area” is a control for causing the robot 100 to leave the object that has caught the leg when the leg of the robot 100 is caught by the object.

[0386] Upon receiving input of information indicating that the robot 100 has been tripped by an object as an environmental feature amount from the environment and body feature detection unit 121 , the behavior control unit 122 performs control to start a leaving behavior based on the information.

[0387] In such a case, the behavior control unit 122 analyzes the stumbling pattern, and executes an exiting behavior that matches the analysis result.

[0388] “(a22) Control switching to a behavior that acts on an object to enable movement” is control performed when the robot 100 continues walking by moving the specific object, for example, when the robot 100 is walking or approaching a specific object.

[0389] Upon receiving input of information indicating that the robot 100 has approached an object as an environmental feature amount from the environment and body feature detection unit 121 , the behavior control unit 122 performs control switching to a behavior acting on the object to enable movement based on the information.

[0390] More specifically, for example, in a case where there is a door in front of the robot 100 , the robot 100 is caused to open the door and continue moving.

[0391] Furthermore, for example, when there is a movable obstacle in front of the robot 100, the robot 100 is caused to move the obstacle. Furthermore, when the obstacle is immovable, the robot 100 moves while avoiding the obstacle.

[0392] Specific examples of control to be executed as "(a3) control for the purpose of switching to an action that does not use the damaged part" include, for example:

[0393] (a31) Walking control while protecting the injured leg.

[0394] For example, upon receiving input of a physical characteristic quantity indicating that the torque of a specific leg is abnormally high from the environment and physical characteristic detection unit 121, the behavior control unit 122 determines that the leg has a malfunction such as some kind of fault, and controls walking using legs other than that leg.

[0395] (B) Grip control

[0396] Next, we will refer to Fig.15 A specific example of “(B) Grip control” is described.

[0397] The "(B) gripping control" of the robot 100 performed by the behavior control unit 122 is a gripping process of the robot 100, that is, a control process when the robot 100 grips an object using a mouth, hands, or legs. Specific examples of this "(B) gripping control" include, for example, the following control.

[0398] (b1) Grip action switching control

[0399] (b2) Grasping possibility control

[0400] “(b1) Gripping action switching control” is a control process related to the action mode of the gripping action for the object that the robot 100 has encountered, and specific examples of the control include:

[0401] (b11) Action switching control for grasping according to body shape (e.g., arm length and arm end shape).

[0402] Upon receiving input of information related to an object that the robot 100 has encountered as an environmental feature amount from the environment and body feature detection unit 121 , the behavior control unit 122 performs gripping motion switching control that matches the input information on the object.

[0403] More specifically, for example, in the case where the robot holds a large ball, the behavior control unit 122 causes the robot to open its legs and hold the ball. In the case of a small ball, the behavior control unit 122 performs control so that the robot holds the ball in its mouth.

[0404] “(b2) Grasping possibility control” is a control process related to determination classification as to whether the robot 100 takes action to grasp an encountered object.

[0405] Upon receiving input of information related to an object that the robot 100 has encountered as an environmental feature amount from the environment and body feature detection unit 121 , the behavior control unit 122 performs grasping possibility control matching the input information of the object.

[0406] Specific examples of “(b2) grasping possibility control” include control processing such as:

[0407] (b21) Grasp only movable objects

[0408] (b22) Grasp only soft objects

[0409] (b23) Grasp only light objects.

[0410] (C) Environmental Adaptive Behavior Control

[0411] Next, we will refer to Fig.15 Describe specific examples of “(C) environmental adaptation behavior control”.

[0412] “(C) Environment-adaptive behavior control” of the robot 100 executed by the behavior control unit 122 is a control process for causing the robot 100 to execute a behavior that is adapted to the environment in which the robot 100 exists.

[0413] Specific examples of this "(C) environment-adaptive behavior control" include, for example, the following control.

[0414] (c1) Control of avoiding dangerous places

[0415] (c2) Control of obtaining unknown environmental information

[0416] “(c1) Control for avoiding a dangerous place” is a control process for causing the robot 100 to perform a behavior of avoiding a dangerous place, and specific examples of the control include, for example, control processes such as:

[0417] (c11) Avoid going to places where the robot may get wet, such as kitchens and sink areas

[0418] (c12) Do not go to places where there are many static obstacles or many dynamic obstacles.

[0419] Upon receiving input of information indicating that the robot 100 is about to approach a dangerous place as an environmental feature amount from the environment and body feature detection unit 121 , the behavior control unit 122 performs control of avoiding the dangerous place to cause the robot 100 to perform a behavior of avoiding the dangerous place.

[0420] More specifically, as described above, the behavior control unit 122 performs control of not causing the robot to go to a place where the robot may get wet such as a kitchen, a sink area, a place where there are many static obstacles, or a place where there are many dynamic obstacles.

[0421] “(c2) Control of acquiring unknown environmental information” is control processing that causes the robot 100 to perform information collection behavior related to an unknown place when the robot 100 encounters an unknown and unexperienced environment, that is, for example, when the robot 100 enters an unknown room.

[0422] Upon receiving input of information indicating that the robot 100 is in an unknown environment that has never been seen before as an environmental feature amount from the environment and body feature detection unit 121 , the behavior control unit 122 performs control of acquiring the unknown environmental information according to the input information.

[0423] Specific examples of “(c2) control for acquiring unknown environmental information” include, for example, causing the robot to perform the following control:

[0424] (c21) Move around the area the robot first arrives at to plan subsequent efficient movements and collect terrain information.

[0425] (D) Feeling expression and behavior control

[0426] Next, we will refer to Fig.16 Describe specific examples of “(D) feeling expression and behavior control”.

[0427] The "(D) Feeling expression and behavior control" of the robot 100 executed by the behavior control unit 122 is a process of controlling the feeling expression and behavior of the robot 100. Specific examples of this "(D) Feeling expression and behavior control" include, for example, the following control.

[0428] (d1) Switching the control of sensory expression

[0429] (d2) Switching behavior control

[0430] “(d1) Control of switching the feeling expression” is a control process for causing the robot 100 to perform the feeling expression, and specific examples of the control include, for example, control processes such as:

[0431] (d11) Changes in sensation when the robot collides with a hard or soft object

[0432] (d12) Being surprised by changes in physical characteristics or environmental characteristics.

[0433] When receiving input of information related to the occurrence of a certain action for the robot 100 as an environmental characteristic quantity, or input of information related to a change in a physical characteristic quantity or an environmental characteristic quantity from the environment and body characteristic detection unit 121, the behavior control unit 122 performs control to cause the robot 100 to perform a sensory expression matching this information.

[0434] More specifically, as described above,

[0435] Since (d11) the feeling changes when the robot collides with a hard object or a soft object, the behavior control unit 122 causes the robot 100 to perform a feeling expression such as sadness, happiness, or anger.

[0436] In addition, due to (d12) being surprised by a change in a physical characteristic quantity or a change in an environmental characteristic quantity, the behavior control unit 122 causes the robot 100 to perform a feeling expression showing a sad and confused face when a malfunction occurs in the robot 100, and to perform a feeling expression showing a hot expression when the temperature of the surrounding environment becomes high and becomes a hot environment.

[0437] “(d2) Control of switching behavior” is a control process for causing the robot 100 to perform various behaviors, and specific examples of the control include control processes such as:

[0438] (d21) Toggle behavior to silent / sound

[0439] (d22) Playing games using sliding surfaces

[0440] (d23) Jump only on soft floor surfaces

[0441] (d24) Rest only in a cool or warm place

[0442] (d25) Move quietly in a quiet environment (e.g., an office) and around the home.

[0443] Upon receiving input of environmental feature quantities as information related to various environments or physical feature quantities related to the robot's own body from the environment and body feature detection unit 121 , the behavior control unit 122 controls the behavior of the robot 100 according to these feature quantities.

[0444] For example, upon receiving input of information indicating that the robot 100 stands on a floor that easily makes a loud sound as an environmental feature amount, the behavior control unit 122 performs control of switching to a behavior of not making a sound as indicated in (d21) above.

[0445] Furthermore, upon receiving input of information indicating that the robot 100 stands on a slippery floor as an environmental feature amount, the behavior control unit 122 performs control switching to a behavior of playing a game utilizing sliding on the floor surface as indicated above (d22).

[0446] Furthermore, upon receiving input of information indicating that the robot 100 stands on a soft floor as an environmental feature amount, the behavior control unit 122 performs control to switch to a behavior of jumping only on the soft floor surface as indicated in (d23) above.

[0447] Furthermore, upon receiving input of information indicating that the robot 100 is in a cool place, a warm place, or the like as an environmental feature amount, the behavior control unit 122 performs control to switch to a behavior of taking a rest as indicated in (d24) above.

[0448] Furthermore, upon receiving input of information indicating that the robot 100 is in a stationary environment (eg, an office) as an environmental feature amount, the behavior control unit 122 performs control of switching to the above (d25) behavior of acting quietly.

[0449] Refer to above Figures 14 to 16 It has been described by Figure 4 An example of behavior control of the robot 100 performed by the behavior control unit 122 of the data processing unit 120 of the robot (information processing device) 100 shown in FIG.

[0450] As can be understood from the above description, the behavior control unit 122 determines the behavior to be performed by the robot 100 based on the feature quantities of the environment around the robot extracted by the environment and body feature quantity extraction unit 121 or the body feature quantities of the robot itself, and performs control to make the robot perform the determined behavior.

[0451] [3. Processing sequence executed by a robot (information processing device) according to the present disclosure]

[0452] Next, a processing sequence performed by the robot (information processing device) according to the present disclosure.

[0453] Fig.17 is a diagram showing a flowchart for explaining a processing sequence executed by the robot (information processing device) 100 according to the present disclosure.

[0454] Note that according to Fig.17 and the flowcharts in the subsequent figures are processed by using Figure 4 The above-described configuration shown in FIG. 1 is performed by the robot (information processing device) 100 .

[0455] Despite Figure 4 Although not shown in the figure, the processing can be executed according to a program stored in a storage unit of the robot 100, for example, and executed under the control of a data processing unit (control unit) including a CPU or the like having a program execution function.

[0456] In the following, we will describe Fig.17 The processing in each step of the flowchart shown in .

[0457] (Step S101)

[0458] First, in step S101 , the data processing unit 120 of the robot 100 receives input of sensor detection information.

[0459] The processing is Figure 4The environment and body feature amount extraction unit 121 of the data processing unit 120 of the robot (information processing device) 100 shown in FIG. 1 inputs the processing of sensor detection values ​​from various sensors constituting the sensing unit 110 .

[0460] As previously referenced Figures 2 to 4 As described, for example, various sensors are attached to the robot 100, such as an image sensor, a voice sensor, an acceleration sensor (IMU), an illumination sensor, a wind direction / wind pressure sensor, an odor sensor, a temperature sensor, a distance sensor, a joint angle sensor, a joint torque sensor, a foot sensor, and an odometer.

[0461] Figure 4 The environment and body feature quantity extraction unit 121 of the data processing unit 120 of the robot (information processing device) 100 shown in FIG. 1 receives input of sensor detection values ​​from these sensors.

[0462] Note that the process of receiving input of these sensor detection values ​​includes two types of processes: a passive input process of passively receiving input of sensor detection values ​​detected by the sensors of the sensing unit 110 , and an active input process.

[0463] The active input process is a process in which the robot 100 performs a specific behavior to obtain a specific sensor detection value, and receives an input of a sensor detection value that becomes necessary as a result of the behavior.

[0464] Will refer to Fig.18 The detailed processing of the sensor detection value input processing in step S101 is described.

[0465] Fig.18 The process shown in is for illustration purposes only. Fig.17 Flowchart of detailed processing of sensor detection value input processing in step S101 of the process shown in FIG. Fig.18 The processing in each step of the process shown in .

[0466] (Step S121)

[0467] First, in step S121, the data processing unit 120 of the robot 100 determines whether the behavior determination unit 123 has determined the behavior that needs to be performed to acquire the environmental feature quantity and the body feature quantity.

[0468] More specifically, the determination unit 123 determines whether to perform the reference Fig. 9 and Fig.10 The processing of the described behavior corresponds to that of the described behavior.

[0469] That is, the data processing unit 120 determines whether the behavior determination unit 123 has determined to perform a jumping behavior for performing processing of acquiring active sensor detection values ​​for detecting the hardness and slipperiness of the floor on which the robot stands when the robot 100 is about to run.

[0470] Note that the situation in which the behavior determination unit 123 determines the behavior required to be performed to obtain the environmental feature quantity and the body feature quantity is set in advance. For example, only when the robot 100 starts walking, the behavior determination unit 123 performs the process of determining the behavior required to be performed to obtain the environmental feature quantity and the body feature quantity. Alternatively, in the case where the robot 100 performs all other new behaviors including walking, the behavior determination unit 123 may be set to perform the process of determining the preliminary behavior required to be performed to obtain the environmental feature quantity and the body feature quantity before starting all other new behaviors.

[0471] When it is determined in step S121 that the behavior determination unit 123 has determined the behavior that needs to be performed to acquire the environmental feature amount and the body feature amount, the flow moves to step S123.

[0472] On the other hand, in the case where it is determined that the behavior determination unit 123 has not yet determined the behavior that needs to be performed to acquire the environmental feature amount and the body feature amount, the flow moves to step S122.

[0473] (Step S122)

[0474] The processing in step S122 is performed in the event that it is determined in step S121 that the behavior determination unit 123 has not determined the behavior required to be performed to acquire the environmental feature amount and the body feature amount.

[0475] In this case, the environment and body feature quantity extraction unit 121 performs the processing in step S122. That is, in step S122, the environment and body feature quantity extraction unit 121 receives the input of the passive sensor detection information.

[0476] The environment and body feature amount extraction unit 121 performs a process of passively receiving input of sensor detection values ​​from various sensors constituting the sensing unit 110 .

[0477] (Step S123)

[0478] On the other hand, in the event that it is determined in step S121 that the behavior determination unit 123 has determined the behavior that needs to be performed to acquire the environmental feature amount and the body feature amount, the processing in step S123 is performed.

[0479] In this case, in step S123 , the behavior execution unit 125 causes the robot 100 to execute the behavior determined by the behavior determination unit 123 .

[0480] For example, as previously referenced Fig. 9 and Fig.10 As described, the behavior execution unit 125 causes the robot 100 to execute the jumping process to execute the process of acquiring active sensor detection values ​​for detecting the hardness and slipperiness of the floor on which the robot 100 stands.

[0481] (Step S124)

[0482] After the predetermined behavior is performed in step S123 , the environment and body feature amount extraction unit 121 performs processing in step S124 .

[0483] In step S124 , the environment and body feature quantity extraction unit 121 receives input of active sensor detection information.

[0484] That is, in step S124, the environment and body feature amount extraction unit 121 receives input of active sensor detection information input from the sensor of the sensing unit 110 as a result of performing the action in step S123.

[0485] As described above, in Fig.17 In step S101 of the flow shown in FIG. 1 , at least one of two types of processing of passive input processing and active input processing for passively receiving input of a sensor detection value detected by the sensor of the sensing unit 110 is performed.

[0486] After performing one of these processes, the flow moves to Fig.17 Step S102 of the process shown in .

[0487] (Step S102)

[0488] Next, we will describe Fig.17 The processing in step S102 of the flow shown in .

[0489] In step S102 , the environment and body feature amount extraction unit 121 performs a process of extracting an environment feature amount and a body feature amount using the sensor detection value input in step S101 .

[0490] Note that the environmental feature quantities are, for example, feature quantities related to the surrounding environment of the robot 100 as described above, and are feature quantities related to various environments around the robot 100, such as information related to surrounding objects, the material and hardness of the robot's running surface (floor surface), time, temperature, the user's state (e.g., smiling, angry), and noisy conditions.

[0491] In addition, the body feature quantities are the body feature quantities of the robot itself, and are the following feature quantities: whether accessories that the robot can be equipped with (for example, clothing, shoes, sole pads) are attached, the driving unit of the robot, and the load (torque) condition of the joints of the legs, for example, the battery status, etc.

[0492] Note that the above Figure 5 Specific examples of environmental feature quantities and body feature quantities acquired from passive sensor detection values ​​are shown.

[0493] also, Figure 7 and Figure 8 Specific examples of environmental feature quantities and body feature quantities acquired from active sensor detection values ​​are shown.

[0494] also, Fig.11 and Fig.12 A specific example of an environmental feature quantity is shown, and Fig.13 Specific examples of body feature quantities are shown.

[0495] In step S102, the environment and body feature amount extraction unit 121 performs extraction of, for example, Figure 5 , Figure 7 and Figure 8 And in addition Figures 11 to 13 Processing of environmental feature quantities and body feature quantities shown in .

[0496] The environmental feature amount and the body feature amount acquired by the environmental and body feature amount extraction unit 121 are input to the behavior determination unit 123 .

[0497] (Step S103)

[0498] Next, in step S103 , the robot 100 determines a behavior to be performed.

[0499] This processing is done by Figure 4 The process performed by the behavior determination unit 123 of the behavior control unit 122 of the data processing unit 120 shown in FIG.

[0500] The behavior determination unit 123 determines a behavior to be performed by the robot 100 based on the feature amount of the environment around the robot extracted by the environment and body feature amount extraction unit 121 or the body feature amount of the robot itself.

[0501] Specific examples of the behavior to be determined are accompanied by previous references such as Figures 14 to 16 The behavior of the control described is one of the following:

[0502] (A) Mobile control

[0503] (B) Grip control

[0504] (C) Environmental Adaptive Behavior Control

[0505] (D) Sensory expression and behavioral control.

[0506] (Step S104)

[0507] Next, in step S104 , the robot 100 determines whether a behavior transition (switching of behaviors) is necessary.

[0508] This processing is done by Figure 4 The process performed by the behavior switching control unit 124 of the behavior control unit 122 of the data processing unit 120 shown in FIG.

[0509] The behavior switching control unit 124 determines whether the behavior determined by the behavior determination unit 123 is different from the currently executed behavior, and if the behavior is different, outputs behavior pattern information requiring switching and a behavior switching request to the behavior execution unit 125 .

[0510] In step S104 , when the behavior switching control unit 124 determines that the behavior determined by the behavior determination unit 123 is different from the currently executed behavior, that is, when it is determined that behavior transition (switching of behavior) is necessary, the flow moves to step S111 .

[0511] On the other hand, in the case where the behavior switching control unit 124 determines that the behavior determined by the behavior determination unit 123 is the same as the currently executed behavior, that is, determines that behavior transition (switching of behavior) is unnecessary, the flow moves to step S105 .

[0512] (Step S105)

[0513] The processing in step S105 is processing executed if the behavior switching control unit 124 determines in step S104 that the behavior determined by the behavior determination unit 123 is the same as the currently executed behavior, that is, determines that behavior transition (switching of behavior) is unnecessary.

[0514] In this case, in step S105, the current behavior continues to be executed.

[0515] This processing is done by Figure 4 The process is performed by the behavior execution unit 125 of the behavior control unit 122 of the data processing unit 120 shown in FIG.

[0516] The behavior execution unit 125 controls the execution of various behaviors of the robot 100. More specifically, the behavior execution unit 125 controls the driving unit 126 and the display control unit 127 to control the motion or expression of the robot 100.

[0517] The behavior executed by the behavior execution unit 125 is accompanied by a previous reference such as Figures 14 to 16 Describes the behavior of controls, i.e., those that:

[0518] (A) Mobile control

[0519] (B) Grip control

[0520] (C) Environmental Adaptive Behavior Control

[0521] (D) Sensory expression and behavioral control.

[0522] (Step S111)

[0523] The processing in step S111 is processing performed when the behavior switching control unit 124 determines in step S104 that the behavior determined by the behavior determination unit 123 is different from the currently executed behavior, that is, determines that behavior transition (switching of behavior) is necessary.

[0524] In this case, in step S111 , the robot 100 performs a behavior switching process that matches the behavior transition.

[0525] That is, the behavior switching control unit 124 outputs the behavior pattern information that needs to be switched and the behavior switching request to the behavior execution unit 125 .

[0526] (Step S112)

[0527] In this case, in step S112 , the robot 100 performs the behavior after the switching process.

[0528] That is, the behavior execution unit 125 executes the behavior after the switching specified by the behavior switching control unit 124. More specifically, the behavior execution unit 125 controls the driving unit 126 and the display control unit 127 to control the motion or expression of the robot 10.

[0529] The behavior executed by the behavior execution unit 125 is accompanied by the above reference, for example Figures 14 to 16 Describes the behavior of controls, i.e., those that:

[0530] (A) Mobile control

[0531] (B) Grip control

[0532] (C) Environmental Adaptive Behavior Control

[0533] (D) Sensory expression and behavioral control.

[0534] The execution control of the behavior in step S105 or step S112 returns to step S101 to execute the input processing of the sensor continuous presence information, and the processing in step S101 and subsequent steps is repeatedly executed.

[0535] These processes enable the robot 100 to execute optimal behavior determined based on various environmental feature quantities and body feature quantities obtained from sensor detection values.

[0536] Next, we will refer to Fig.19 The following drawings describe the Fig.17 A specific example of the processing performed by the flowchart in FIG.

[0537] Fig.19 A robot 100 and a user 20 of the robot 100 are shown.

[0538] The robot 100 is located at a position (P1) far from the user 20. The user 20 calls the robot 100 while waving at the robot 100 to call the robot 100 to approach the user.

[0539] Note that the robot 100 is on a carpet, and the user 20 is on a marble floor. The carpet has the characteristic that the carpet is almost non-slip and causes a small footstep sound. On the other hand, the marble has the characteristic that the marble is slippery and causes a large footstep sound.

[0540] Robot 100 is based on Figure 4 The robot shown in the configuration recognizes the call of the user 20 based on the detection information of the sensor of the sensing unit 110 .

[0541] That is, the behavior determination unit 123 in the behavior control unit 122 of the data processing unit 120 of the robot 100 detects the waving action of the user 20 from the captured image of the image sensor (camera), and detects the user's call from the sound acquired by the voice sensor (microphone).

[0542] The behavior determination unit 123 determines that a behavior of running close to the user 20 is performed as a behavior in response to the detected call from the user.

[0543] This behavior determination process is performed by applying, for example, a machine learning algorithm.

[0544] Next, the behavior determination unit 123 of the robot 100 performs processing of the behavior pattern of running close to the user 20. For example, the behavior determination unit 123 determines a behavior pattern (walking pattern) such as whether the robot 100 is running hurriedly at high speed or walking at low speed or increasing or decreasing the stride.

[0545] The behavior determination unit 123 uses the environmental feature amount and the body feature amount extracted by the environmental and body feature extraction unit 121 based on the sensor detection information for a process of determining the behavior pattern (walking pattern).

[0546] The behavior determination unit 123 determines whether there is an active behavior required for obtaining sensor detection values ​​that is input from the environment and body feature extraction unit 121 and can be used to process and determine the behavior pattern (walking pattern), and if such a behavior exists, determines the behavior as a behavior that needs to be performed.

[0547] Here, the behavior determination unit 123 determines the behavior of actively acquiring sensor detection values ​​for obtaining environmental feature quantities as environmental information required to determine the walking mode of the robot 100 for approaching the user 20, and the environmental feature quantities are, for example, floor information such as material, whether the floor is slippery, and whether the floor causes large footsteps or small footsteps.

[0548] For example, Fig. 20 As shown in , the behavior determination unit 123 determines the action of the robot 100's legs hitting the floor, i.e., stepping on the floor and obtaining the detection value of the sole sensor, the detection sound of the voice sensor, etc., as the action that needs to be executed, and enables the behavior execution unit 125 to execute the action.

[0549] Furthermore, the behavior determination unit 123 may point the image sensor (camera) in the direction of the floor, and cause the behavior execution unit 125 to execute a behavior of acquiring an image of the floor.

[0550] Fig. 20 Step S201 shown in FIG. 1 is described as a process of executing behavior for this active sensor detection value acquisition process.

[0551] The processing in step S201 is similar to that in the previous reference. Fig.18 The processing in step S123 of the described flow corresponds.

[0552] pass Fig. 20 The active sensor detection value obtained by the process in step S201 shown in , that is, the stepping process as the behavior for the active sensor detection value acquisition process, is input to the environment and body feature extraction unit 121.

[0553] This treatment is similar to the previous reference Fig.18 The processing in step S124 of the described flow corresponds.

[0554] For example, detection values ​​of a sole sensor, detection sounds of a voice sensor, and further sensor detection values ​​of an image of a floor or the like obtained by an image sensor (camera) are input to the environment and body feature extraction unit 121 .

[0555] The environment and body feature extraction unit 121 analyzes these input sensor detection values ​​and extracts an environment feature amount and a body feature amount.

[0556] More specifically, the environment and body feature extraction unit 121 extracts environmental feature quantities such as the material of the floor, the slipperiness of the floor, and whether the floor causes large or small footsteps when walking, and body feature quantities such as whether accessories such as shoes or insoles are equipped.

[0557] Fig. 20 The example shown in Fig. 20 Step S202 in the example indicates obtaining the following information as an example of the environmental feature amount.

[0558] Floor material = "carpet"

[0559] Slipperiness = Almost not slippery

[0560] Is the footsteps loud or soft = soft

[0561] By this Fig. 20 The processing indicated by step S202 in Fig.17 The processing corresponds to step S102 of the process in .

[0562] The environment and body feature extraction unit 121 inputs these extracted environment feature quantities to the behavior determination unit 123 .

[0563] The behavior determination unit 123 determines the optimal behavior, that is, the walking pattern, based on these feature amounts.

[0564] More specifically, as e.g. Fig. 20 As indicated in step S203 in , the behavior determination unit 123 determines that the person is running hurriedly with an increased stride.

[0565] This is the optimal behavior mode (walking mode) based on environmental characteristic quantities indicating that the material of the floor is "carpet", the floor is almost non-slip and the floor also causes small footsteps, and the behavior determination unit 123 determines that running hurriedly with an increased stride as the optimal walking mode based on these environmental characteristic quantities.

[0566] By this Fig. 20 The processing indicated by step S203 in Fig.17 The processing corresponds to step S103 of the process in .

[0567] Then, the robot 100 executes a behavior according to the determined behavior pattern. That is, the behavior execution unit 125 causes the robot 100 to start running hurriedly with an increased stride.

[0568] Fig.21It is a diagram for explaining the state after the robot 100 starts to run hurriedly.

[0569] The robot 100 starts running hurriedly with an increasing stride from position (P1).

[0570] However, at the point in time when the robot 100 reaches the position (P2), the floor surface subsequently changes from "carpet" to "marble".

[0571] The sensors of the robot 100 also continuously perform sensing processing while the robot 100 runs.

[0572] For example, the sole sensor continuously detects the hardness and friction of the floor surface and inputs the detection values ​​to the environment and body feature extraction unit 121.

[0573] The environment and body feature extraction unit 121 recognizes the change in the sensor detection value input from the sole sensor at the time point when the robot 100 reaches the position (P2). That is, the environment and body feature extraction unit 121 recognizes that the hardness and friction of the floor surface have changed, and performs a process of updating the environment feature amount according to the new detection value.

[0574] If Fig.21 As indicated in step S221, the following environmental feature quantities are extracted based on the passive detection information of the sensor.

[0575] Floor material = "marble"

[0576] Slipperiness = Slippery

[0577] Is the footsteps loud or soft = loud

[0578] As described above, the environment and body feature extraction unit 121 detects that the floor has been changed into marble at the time point when the position of the robot 100 reaches (P2).

[0579] By this Fig.21 The processing indicated by step S221 in the above is the input of the passive sensor detection value and the feature extraction processing. Fig.18 The processing in step S122 of the process in Fig.17 The processing corresponds to step S102 of the process in .

[0580] The environment and body feature extraction unit 121 inputs these extracted environment feature quantities to the behavior determination unit 123 .

[0581] The behavior determination unit 123 determines the optimal behavior, that is, a new walking pattern, based on these feature amounts.

[0582] More specifically, the behavior determination unit 123 is composed of, for example, Fig.21As indicated in step S222 in the flowchart, it is determined to walk at a low speed with a reduced stride.

[0583] The walking is an optimal behavior pattern (walking pattern) based on environmental characteristic quantities indicating that the material of the floor is "marble", the floor is slippery, and the floor causes loud footsteps, and the behavior determination unit 123 determines walking at a reduced stride and at a low speed as the optimal walking pattern based on these environmental characteristic quantities.

[0584] By this Fig.21 The processing indicated by step S222 in Fig.17 The processing corresponds to step S103 of the process in .

[0585] Then, the robot 100 performs a behavior according to the determined behavior pattern. That is, the behavior execution unit 125 causes the robot 100 to walk at a reduced pace and at a low speed, and to run in a row close to the user 20 .

[0586] As described above, the robot 100 according to the present disclosure adopts a configuration that determines and executes an optimal behavior according to various environmental feature quantities and body feature quantities obtained from sensor detection values.

[0587] Note, refer to Figures 19 to 21 The described example is a specific example in the case where the robot 100 is not equipped with accessories such as shoes and sole pads.

[0588] In this case, the environment and body feature extraction unit 121 detects that no accessories are equipped as the body feature amount of the robot 100, and the behavior determination unit 123 determines the optimal behavior based on this detection information.

[0589] In contrast, in the case where the robot 100 is equipped with accessories such as shoes or insoles, the environment and body feature extraction unit 121 detects, for example, "shoes" as a body feature of the robot 100, and the behavior determination unit 123 determines the optimal behavior based on the detection information.

[0590] Will refer to Fig. 22 and subsequent figures describe specific examples.

[0591] like Fig. 22 As shown in , it is assumed that the robot 100 is equipped with a “shoe” 152 as an accessory that the robot 100 can be equipped with.

[0592] By being equipped with "shoes", the robot 100 hardly slips or causes small footsteps.

[0593] Fig.23 A processing example is shown in the case where the robot 100 is equipped with “shoes” 152 as an accessory.

[0594] The environment and body feature extraction unit 121 of the robot 100 extracts an environment feature amount and a body feature amount using active sensor detection values ​​when the robot 100 is at the position (P1).

[0595] That is, the environment and body feature extraction unit 121 performs an action of hitting the floor with the legs, for example, and extracts the following environment feature amount and body feature amount.

[0596] (a) Environmental characteristics

[0597] Floor material = "carpet"

[0598] Slipperiness = Almost not slippery

[0599] Is the footsteps loud or soft = soft

[0600] (b) Physical characteristics

[0601] Equipment Accessories = Shoes

[0602] Slipperiness = Almost not slippery

[0603] Is the footsteps loud or soft = soft

[0604] The environment and body feature extraction unit 121 inputs these extracted environment feature quantities to the behavior determination unit 123 .

[0605] The behavior determination unit 123 determines the optimal behavior, that is, the walking pattern, based on these feature amounts.

[0606] More specifically, the behavior determination unit 123 determines as determined by, for example Fig.23 The robot 100 at position (P1) in FIG. 1 is indicated to be running hurriedly with an increased stride.

[0607] The running is an optimal behavior pattern (walking pattern) based on environmental feature quantities indicating that the material of the floor is "carpet", the floor is almost not slippery and the floor also causes small footsteps, and physical feature quantities indicating that the equipment is equipped with shoes as accessories, the floor is almost not slippery and the floor also causes small footsteps, and the behavior determination unit 123 determines that hurried running with an increased stride is the optimal walking pattern based on these environmental feature quantities.

[0608] The robot 100 starts running hurriedly with an increased stride length from the position (P1) according to the determined behavior.

[0609] However, at the point in time when the robot 100 reaches the position (P2), the floor surface subsequently changes from "carpet" to "marble".

[0610] The sensors of the robot 100 continuously perform sensing processing even when the robot 100 runs.

[0611] For example, the sole sensor continuously detects the hardness and friction of the floor surface and inputs the detection values ​​to the environment and body feature extraction unit 121.

[0612] The environment and body feature extraction unit 121 performs a process of updating the environment feature quantity and the body feature quantity according to the new sensor detection value at the time point when the robot 100 reaches the position (P2). For example, the environment and body feature extraction unit 121 extracts the following new environment feature quantity and body feature quantity.

[0613] (a) Environmental characteristics

[0614] Floor material = "marble"

[0615] Slipperiness = Slippery

[0616] Is the footsteps loud or soft = loud

[0617] (b) Physical characteristics

[0618] Equipment Accessories = Shoes

[0619] Slipperiness = Almost not slippery

[0620] Is the footsteps loud or soft = soft

[0621] As described above, the environment and body feature extraction unit 121 detects that the floor has been changed into marble at the time point when the position of the robot 100 reaches (P2).

[0622] The environment and body feature extraction unit 121 inputs these extracted environment feature quantities and body feature quantities to the behavior determination unit 123 .

[0623] The behavior determination unit 123 determines the optimal behavior, that is, a new walking pattern, based on these feature amounts.

[0624] The behavior determination unit 123 determines the optimal behavior pattern (walking pattern) by considering two feature quantities: an environmental feature quantity indicating that the material of the floor is "marble", the floor is slippery, and the floor causes large footsteps, and a physical feature quantity indicating that the floor is almost not slippery and the floor also causes small footsteps due to the "shoes" equipped as an accessory.

[0625] In this example, the behavior determination unit 123 determines, based on these environmental feature amounts, hurried running with an increased stride as the optimal walking pattern.

[0626] This running is the optimal behavior mode (walking mode) according to the fact that feature quantities include environmental feature quantities indicating that the material of the floor is "marble", the floor is slippery and the floor causes loud footsteps, and physical feature quantities indicating that the floor is almost not slippery and the floor also causes quiet footsteps because the robot 100 is equipped with "shoes" as an accessory, and the behavior determination unit 123 determines the behavior of running hurriedly with an increased stride as the optimal walking mode based on these environmental feature quantities.

[0627] Thus, the robot 100 continuously scurries from the position ( P1 ) to ( P2 ) and then to ( P3 ) with increasing strides, reaching the position ( P3 ) of the user 20 .

[0628] As described above, the robot 100 according to the present disclosure determines and executes an optimal behavior according to various environmental feature quantities and body feature quantities obtained from sensor detection values.

[0629] [4. Other configuration examples of the robot (information processing device) according to the present disclosure]

[0630] Next, other configuration examples of the robot (information processing device) of the present disclosure will be described.

[0631] Previously referenced Figure 4 A configuration example of the robot (information processing device) 100 according to the present disclosure has been described.

[0632] Figure 4 The configuration of the robot (information processing device) 100 shown in is a configuration example of the robot (information processing device) 100 that performs the processing according to the present disclosure.

[0633] Next, we will refer to Fig.24 and the accompanying drawings Figure 4 A configuration example of a robot (information processing device) 100 having a different configuration as shown in FIG.

[0634] The following configuration examples will be described below in sequence.

[0635] (a) Example of configuration including sensor database

[0636] (b) Configuration example including environment / feature quantity database

[0637] (c) Configuration Example Including Sensing Database and Environment / Feature Quantity Database

[0638] (a) Example of configuration including sensing database

[0639] First, a configuration example of a robot (information processing device) 100B including a sensing database will be described.

[0640] Fig.24 is a diagram showing a configuration example of a robot (information processing device) 100B including a sensing database.

[0641] Fig.24 The robot (information processing device) 100B shown in includes a sensing unit 110 and a data processing unit 120 .

[0642] The sensing unit 110 includes a reference Figure 2 and Figure 3 Describes the various sensors that collect information about the surrounding environment.

[0643] like Fig.24 As shown in , the data processing unit 120 includes an environment and body feature extraction unit 121 , a behavior control unit 122 , and a sensing database (DB) 131 .

[0644] The behavior control unit 122 includes a behavior determination unit 123 , a behavior switching control unit 124 , a behavior execution unit 125 , an action control unit 126 , and an expression control unit 127 .

[0645] The components other than the sensor database (DB) 131 are the same as those previously referred to. Figure 4 The parts described are the same parts.

[0646] The sensing database (DB) 131 receives input of various sensor detection data from the sensing unit 110 and associates various sensor detection values ​​with detection time information to record them. That is, the sensing database (DB) 131 stores historical data of sensor detection values.

[0647] The environment and body feature amount extraction unit 121 refers to the history data of sensor detection values ​​stored in the sensing database (DB) 131 as needed.

[0648] The environment and body feature extraction unit 121 receives input of various sensor detection data from the sensing unit 110, and extracts feature quantities of the environment around the robot and body feature quantities of the robot itself using a machine learning model based on the input data.

[0649] The environment and body feature amount extraction unit 121 also performs a process of cross-checking the current sensor detection value input from the sensing unit 110 and the history data of the sensor detection value stored in the sensing database (DB) 131 .

[0650] For example, the environmental and physical feature extraction unit 121 compares a facial image of a person acquired by the image sensor (camera) with a facial image stored in the sensing database (DB) 131 and previously detected by the sensor, and determines whether the person acquired by the image sensor (camera) has been previously photographed, that is, whether the person is a person that the robot 100 has previously encountered.

[0651] The environment and body feature extraction unit 121 also performs the following processing: for example, cross-checking the current image of not only a person but also a room, scenery, etc. with past images stored in the sensing database (DB) 131, and determining whether the room or scenery is a room or scenery that the robot 100 has previously seen.

[0652] For example, the environmental and physical feature extraction unit 121 compares the voice of a person acquired by the voice sensor (microphone) with the audio (voice) stored in the sensing database (DB) 131 and previously detected by the sensor, and determines whether the voice of the person acquired by the voice sensor (microphone) has also been previously captured, that is, whether the person is a person the robot 100 has previously encountered.

[0653] As described above, by using past sensor detection values ​​stored in the sensing database (DB) 131, the environmental and physical feature extraction unit 121 can perform a process of cross-checking the current sensor detection value input from the sensing unit 110 and the historical data of the sensor detection value stored in the sensing database (DB) 131, and determine whether the current sensor value matches the previously acquired sensor detection value.

[0654] Therefore, the environment and body feature amount extraction unit 121 can extract, for example, an environment feature amount indicating that the person in front of the robot 100 is a person that the robot 100 sees for the first time or a person that the robot 100 has encountered many times.

[0655] Furthermore, for example, the environment and body feature amount extraction unit 121 may extract environment feature amounts indicating, for example, that the room in which the robot 100 exists is a room that the robot 100 enters for the first time or a room that the robot 100 has entered multiple times.

[0656] (b) Configuration example including environment / feature quantity database

[0657] Next, a configuration example of the robot (information processing device) 100C including an environment / feature amount database will be described.

[0658] Fig.25 is a diagram showing a configuration example of a robot (information processing device) 100C including an environment / feature amount database.

[0659] Fig.25 A robot (information processing device) 100C shown in FIG. 1 includes a sensing unit 110 and a data processing unit 120 .

[0660] The sensing unit 110 includes a reference Figure 2 and Figure 3 Describes the various sensors that collect information about the surrounding environment.

[0661] like Fig.25 As shown in , the data processing unit 120 includes an environment and body feature extraction unit 121 , a behavior control unit 122 , and an environment / feature amount database (DB) 132 .

[0662] The behavior control unit 122 includes a behavior determination unit 123 , a behavior switching control unit 124 , a behavior execution unit 125 , an action control unit 126 , and an expression control unit 127 .

[0663] The components other than the environment / feature amount database (DB) 132 are the same as those previously referred to. Figure 4 The parts described are the same parts.

[0664] The environment / feature amount database (DB) 132 records the environment feature amount and the body feature amount extracted by the environment and body feature amount extraction unit 121 in association with the extracted time information. That is, the environment / feature amount database (DB) 132 stores the history data of the extracted feature amounts.

[0665] More specifically, the environment / feature amount database (DB) 132 records environment feature amounts such as the color of a person's clothing, the shape of a room, the furniture of the room, and the arrangement of the furniture in association with time information.

[0666] The historical data of the feature amount stored in the environment / feature amount database (DB) 132 is referred to by the behavior determination unit 123 as needed.

[0667] For example, the behavior determination unit 123 compares the current environmental feature quantity or physical feature quantity input from the environmental and physical feature quantity extraction unit 121 with the past environmental feature quantities or physical feature quantities stored in the environmental / feature quantity database (DB) 132, detects changes in the environment around the robot or changes in the robot itself, and determines the optimal behavior that matches the difference in changes.

[0668] More specifically, for example, upon receiving an input of a feature quantity indicating that the floor of a room is a "marble floor with carpet" as a current environmental feature quantity from the environment and body feature quantity extraction unit 121, and obtaining a feature quantity indicating that the floor of a room is a "marble floor without carpet" as a past environmental feature quantity from the environment / feature quantity database (DB) 132, the behavior determination unit 123 performs processing for recognizing that the user has laid a carpet and determining that happily running around is a behavior that needs to be performed.

[0669] In addition, for example, when a feature quantity indicating "the user is a user with short hair" is received as a current environmental feature quantity from the environment and body feature quantity extraction unit 121, and a feature quantity indicating "the user is a user with long hair" is obtained as a past environmental feature quantity from the environment / feature quantity database (DB) 132, the behavior determination unit 123 performs processing of recognizing that "the user has cut his hair" and determining showing a surprised expression as a behavior that needs to be performed.

[0670] (c) Configuration Example Including Sensing Database and Environment / Feature Quantity Database

[0671] Next, a configuration example of a robot (information processing device) 100D including a sensing database and an environment / feature amount database will be described.

[0672] Fig.26 is a diagram showing a configuration example of a robot (information processing device) 100D including a sensing database and an environment / feature amount database.

[0673] Fig.26 A robot (information processing device) 100D shown in FIG. 1 includes a sensing unit 110 and a data processing unit 120 .

[0674] The sensing unit 110 includes a reference Figure 2 and Figure 3 Describes the various sensors that collect information about the surrounding environment.

[0675] like Fig.24 As shown in , the data processing unit 120 includes an environment and body feature extraction unit 121 , a behavior control unit 122 , a sensing database (DB) 131 , and an environment / feature amount database (DB) 132 .

[0676] The behavior control unit 122 includes a behavior determination unit 123 , a behavior switching control unit 124 , a behavior execution unit 125 , an action control unit 126 , and an expression control unit 127 .

[0677] The components other than the sensing database (DB) 131 and the environment / feature amount database (DB) 132 are the same as those previously referred to. Figure 4 The parts described are the same parts.

[0678] Should Fig.26 The configuration shown in Fig.24 The configuration shown in Fig.25 The configuration obtained by combining the configurations shown in .

[0679] That is, the sensing database (DB) 131 receives input of various sensor detection data from the sensing unit 110 and associates various sensor detection values ​​with detection time information to record. That is, the sensing database (DB) 131 stores historical data of sensor detection values.

[0680] The history data of sensor detection values ​​stored in the sensing database (DB) 131 is referred to by the environment and body feature amount extraction unit 121 as needed.

[0681] Furthermore, the environment / feature amount database (DB) 132 associates the environment feature amount and the body feature amount extracted by the environment and body feature amount extraction unit 121 with the extracted time information to record them. That is, the environment / feature amount database (DB) 132 stores the history data of the extracted feature amounts.

[0682] More specifically, the environment / feature amount database (DB) 132 associates environment feature amounts, such as the color of a person's clothing, the shape of a room, the furniture of the room, and the arrangement of the furniture.

[0683] The historical data of the feature amount stored in the environment / feature amount database (DB) 132 is referred to by the behavior determination unit 123 as needed.

[0684] Should Fig.26 The configuration shown in the previous reference can be executed using Fig.24 The processing and use of the data stored in the sensing database (DB) 131 described herein refer to Fig.25 The configuration of any one of the processes of the data stored in the environment / feature amount database (DB) 132 is described.

[0685] [5. Specific example of processing using a robot (information processing device) according to the present disclosure]

[0686] Next, a specific example of a process using a robot (information processing apparatus) according to the present disclosure will be described.

[0687] Will refer to Fig. 27 and the subsequent drawings describe specific examples of processing using a robot (information processing device) according to the present disclosure.

[0688] The following specific examples will be described in sequence.

[0689] (Specific Example 1) Processing of switching the robot's lying down action according to the softness of the floor surface (Specific Example 2) Processing of switching the robot's walking action according to the characteristics of the floor surface and the accessories equipped

[0690] (Specific Example 3) The robot's response when placed in an unknown environment

[0691] (Specific Example 4) Processing of switching the robot's behavior according to the user's action on the robot (Specific Example 1) Processing of switching the robot's lying down action according to the softness of the floor surface

[0692] First, a process of switching the lying-down motion of the robot according to the softness of the floor surface will be described as (Specific Example 1).

[0693] When in e.g. Fig. 27 When the robot 100 suddenly performs a lying-down action on a hard floor surface such as marble as shown in the upper figures (a1) and (a2) in FIG.

[0694] However, when Fig. 27 When lying down slowly without any power as shown in the lower figures (b1) to (b3) in FIG.

[0695] Furthermore, although the sound is insignificant in the case of a soft floor surface such as a carpet, when lying down on a soft floor surface such as a carpet without power, the robot 100 cannot roll much, a larger load is applied to the joints of the robot 100, and malfunctions may occur depending on the situation.

[0696] The above-described robot 100 according to the present disclosure does not cause these problems.

[0697] More specifically, for example, the robot 100 performs a behavior of hitting a floor surface to acquire an active sensor detection value.

[0698] The voice sensor picks up the sound caused by the action of hitting the floor surface, and the IMU sensor picks up the vibration.

[0699] The environment and body feature extraction unit 121 receives input of these sensor detection values, and extracts an environment feature amount using a discriminator using machine learning.

[0700] More specifically, the environment and body feature extraction unit 121 determines the hardness of the floor surface.

[0701] The environment and body feature extraction unit 121 inputs information on the hardness of the floor surface as the environment feature amount extracted by the environment and body feature extraction unit 121 to the behavior determination unit 123 .

[0702] The behavior determination unit 123 determines the optimal behavior based on the information of the hardness of the floor surface as the environmental feature amount extracted by the environment and body feature extraction unit 121 .

[0703] For example, in the case where the floor surface is a hard floor surface such as marble, the behavior determination unit 123 performs the following steps: Fig. 27 The slow lying down movement shown in (b1) to (b3) in FIG.

[0704] Furthermore, in the case where the floor surface is a soft floor surface such as carpet, Fig.28 As shown in (a1) to (a3) ​​in FIG. 1 , the behavior determination unit 123 performs a lying down action that does not cause a burden on the legs.

[0705] As described above, by applying the process according to the present disclosure, the lying-down motion of the robot 100 may be performed as an optimal lying-down motion matching the hardness or softness of the floor surface.

[0706] (Specific Example 2) Processing of switching the robot's walking motion according to the characteristics of the floor surface and the equipped accessories

[0707] Next, a process of switching the walking motion of the robot according to the characteristics of the floor surface and the equipped accessories will be described as (Specific Example 2).

[0708] For example, despite the previous reference Figure 6 The described "shoes" or "foot pads" provide an anti-slip effect, but when the robot 100 runs on, for example, a "carpet" having strong friction and long fibers, there is a case where the legs get caught and the robot 100 easily falls.

[0709] The above-described robot 100 according to the present disclosure can solve this problem.

[0710] More specifically, for example, the robot 100 performs a behavior of hitting a floor surface to acquire an active sensor detection value.

[0711] The voice sensor picks up the sound caused by the action of hitting the floor surface, and the IMU sensor picks up the vibration.

[0712] The environment and body feature extraction unit 121 receives input of these sensor detection values, and extracts an environment feature amount using a discriminator using machine learning.

[0713] More specifically, the environment and body feature extraction unit 121 acquires information indicating that the floor surface is “carpet” as the environment feature amount.

[0714] Furthermore, the environment and body feature extraction unit 121 detects that the legs of the robot 100 are equipped with “shoes” or “sole pads” as body feature quantities.

[0715] Information indicating floor surface=“carpet” as environmental feature quantities extracted by the environmental and body feature extraction unit 121 and body feature quantities indicating that the legs of the robot 100 are equipped with “shoes” or “sole pads” are input to the behavior determination unit 123 .

[0716] The behavior determination unit 123 determines the optimal behavior based on the environmental feature amount (floor surface = "carpet") and the body feature amount (the legs of the robot 100 are equipped with "shoes" or "sole pads") extracted by the environmental and body feature extraction unit 121.

[0717] For example, the behavior determination unit 123 determines a behavior of walking slowly on a carpet with a reduced step as a behavior to be performed so that the legs of the robot 100 are not tripped. Alternatively, the behavior determination unit 123 makes a determination so that the robot 100 does not walk on a carpet.

[0718] As described above, by applying the process according to the present disclosure, the walking motion of the robot 100 can be performed as an optimal walking motion that matches the material of the floor surface and the accessories equipped on the robot 100 .

[0719] (Specific Example 3) The robot's response when placed in an unknown environment

[0720] Next, the reaction of the robot when it is placed in an unknown environment will be described as (Specific Example 3).

[0721] For example, in the case where the robot 100 is placed in a new environment, more specifically, for example, when the robot 100 sees a scene that the robot 100 has never seen before, the expression of the robot 100 is set to a happy expression or a surprised expression.

[0722] By making such a change in expression, the robot 100's feeling expression for the environmental change can be performed as close to that of a person or an actual animal.

[0723] The above-described robot 100 according to the present disclosure can perform such feeling expression.

[0724] More specifically, the robot 100 can input, for example, an image acquired by an image sensor (camera) constituting the sensing unit 110 of the robot 100 , distance information acquired by a distance sensor, and the like to the environment and body feature extraction unit 121 .

[0725] The environment and body feature extraction unit 121 receives input of these sensor detection values, and extracts environmental feature quantities related to the environment surrounding the robot 100 .

[0726] The environment and body feature extraction unit 121 also performs the extraction of the environmental feature quantity and the previously stored reference Fig.25 The environment / feature amount database 132 described herein performs a matching process (cross-check process) with the environment feature amount. The environment and body feature extraction unit 121 determines the degree of matching with the environment feature amount acquired from the sensor detection value.

[0727] In the case where past environmental feature quantities that have a high degree of matching with the environmental feature quantities extracted from the sensor detection values ​​are not extracted from the accumulated data, the environmental and body feature extraction unit 121 determines that the scenery currently viewed by the robot 100 is a new environment.

[0728] In response to the result of the determination, the environment and body feature extraction unit 121 generates an environment feature amount indicating that the current environment is a new environment, and outputs the environment feature amount to the behavior determination unit 123 .

[0729] The behavior determination unit 123 determines to output a happy expression or a surprised expression as a behavior to be performed based on the environmental feature amount (current environment=new environment) extracted by the environmental and physical feature extraction unit 121 .

[0730] For example Fig.29 As shown in , the behavior execution unit 125 changes the expression of the robot 100 to a happy expression or a surprised expression according to the determination information of the behavior determination unit 123.

[0731] Furthermore, the behavior determination unit 125 may determine to perform an exploratory behavior of causing the robot 100 to walk around in the new environment according to, for example, the fact that the current environment is a new environment, and cause the behavior execution unit 125 to execute the exploratory behavior.

[0732] As described above, by applying the processing according to the present disclosure, it is possible to change the expression of the robot 100 in response to the environment surrounding the robot and make the robot 100 react closer to a real human or animal.

[0733] (Specific Example 4) Processing of switching the robot's behavior according to the user's action on the robot

[0734] Next, a process of switching the behavior of the robot according to the user's action on the robot will be described as (Specific Example 4).

[0735] For example, there is a case where the robot 100 is caressed or hit by the user. Even the case where the robot 100 is hit includes a case where the robot 100 is hit gently or is hit to cause pain.

[0736] The robot 100 according to the present disclosure can react accurately in response to these various user's actions.

[0737] The robot 100 inputs a sensor detection value detected by a sensor when a user's action (eg, an action of stroking, light impact, or heavy impact) is performed to the environment and body feature extraction unit 121 .

[0738] Examples of the sensor detection value are detection values ​​of a vibration sensor such as an IMU, a voice sensor, an image sensor, and the like.

[0739] The environment and body feature extraction unit 121 performs analysis processing applying a machine learning algorithm based on these sensor detection values ​​and determines the type of the user's action. That is, the environment and body feature extraction unit 121 extracts the result obtained by determining, for example, whether the user is performing a light touch, a light impact, or a heavy impact as an environmental feature amount.

[0740] For example, the environment and body feature extraction unit 121 extracts one of the following environment feature quantities based on the sensor detection value.

[0741] Environmental feature quantity = user is gently touching

[0742] Environmental feature quantity = user is gently bumping

[0743] Environmental feature quantity = user is hitting hard

[0744] For example, the environment and body feature extraction unit 121 extracts these different environmental feature quantities based on the sensor detection values.

[0745] The environment and body feature extraction unit 121 inputs the extracted environment feature amount to the behavior determination unit 123 .

[0746] The behavior determination unit 123 determines different behaviors accordingly based on the environmental feature quantity extracted by the environment and body feature extraction unit 121, that is, one of the following environmental feature quantities:

[0747] Environmental feature quantity = user is gently stroking

[0748] Environmental feature quantity = user is gently bumping

[0749] Environmental feature quantity = user is hitting hard.

[0750] Upon receiving input such as “environmental feature amount = user is gently stroking” or “environmental feature amount = user is gently bumping”, the behavior determination unit 123 determines the behavior of changing the expression of the robot 100 to a happy expression as the behavior to be performed.

[0751] On the other hand, in the case of receiving the input of "environmental feature amount=user is hitting hard", the behavior determination unit 123 determines the behavior of changing the expression of the robot 100 to a sad expression or an angry expression as the behavior to be performed.

[0752] The behavior execution unit 125 changes the expression of the robot 100 to a happy expression, a sad expression, or a surprised expression according to the determination information of the behavior determination unit 123 .

[0753] As described above, by applying the process according to the present disclosure, it is possible to make the robot 100 react to various user reactions more naturally.

[0754] Next, we will refer to Fig.30 and subsequent drawings describe specific examples of other processing that the robot (information processing device) 100 according to the present disclosure can perform.

[0755] Figure 30 to Figure 32 A table for explaining a specific example of processing matching the type of environmental feature amount extracted by the environmental and body feature extraction unit 121 is shown.

[0756] More specifically, Figure 30 to Figure 32 A specific example of processing matching the following types as the types of environmental feature amounts extracted is shown.

[0757] (1) Type of environmental characteristic quantity = Floor surface material

[0758] (2) Type of environmental characteristic quantity = structure and type of room

[0759] (3) Type of environmental characteristic quantity = Material of the object

[0760] (4) Type of environmental characteristic quantity = state of human

[0761] (5) Type of environmental feature quantity = state of the robot itself

[0762] In the table, each of the following data is represented correspondingly by the type of the extracted environmental feature amount.

[0763] (a) Types of environmental characteristics

[0764] (b) Example of sensing data for extracting feature quantities

[0765] (c) Example of Sensing Processing

[0766] (d) Examples of actions to be identified and performed

[0767] first, Fig.30 The data shown in FIG. 1 , that is, the type of the environmental feature quantity extracted by the environmental and body feature extraction unit 121, is

[0768] (1) Type of environmental characteristic quantity = material of floor surface.

[0769] A processing example in this case will be described.

[0770] In the case where the type of environmental feature quantity extracted by the environmental and body feature extraction unit 121 is (a) type of environmental feature quantity = material of the floor surface, (b) sensor detection values ​​of a vibration sensor, a voice sensor, a joint angle sensor or a joint torque sensor such as an IMU are used as sensing data for extracting the feature quantity.

[0771] Furthermore, as an example of the sensing process when (c) each sensing data indicated in (b) above is acquired, for example, the following sensing process is used.

[0772] (c1) Sensing while walking

[0773] (c2) interacting with and sensing the floor surface (jumping on, hitting, and stroking the floor surface)

[0774] (c3) Sense vibration / sound caused by people / objects / robots and floor surfaces in the surrounding environment

[0775] Furthermore, examples of (d) actions to be determined and performed include the following actions.

[0776] (d1) Calibrate odometer

[0777] (d2) Change of gait

[0778] (d3) Change walking mode

[0779] (d4) Playing games by sliding on the floor surface

[0780] (d5) Playing games using the hardness of the floor surface

[0781] The reason for “(d1) calibrating the odometer” is the change in the amount of movement due to the difference in frictional force of the floor surface.

[0782] The reason for "(d2) changing gait" is that the robot has different appropriate gaits depending on the friction of the floor surface. In addition, the reason also includes that the robot is easy to fall on a floor surface with high friction, and further, walking at a high speed on a hard floor surface is noisy, and therefore the robot needs to walk slowly.

[0783] Specific examples of "(d3) changing the walking mode" include, for example, moving while sliding like skating on a slippery place.

[0784] Specific examples of “(d4) playing a game utilizing sliding of the floor surface” include performing an action of playing a game while causing an object to slide and sliding oneself, for example.

[0785] Specific examples of “(d5) playing a game utilizing the hardness of the floor surface” include, for example, performing tap dancing.

[0786] Next, Fig.31 The data shown in , that is, the types of environmental feature quantities extracted by the environmental and body feature extraction unit 121, are:

[0787] (2) Type of environmental characteristic quantity = structure and type of room

[0788] A processing example in this case will be described.

[0789] When the type of environmental feature quantity extracted by the environmental and body feature extraction unit 121 is (a) environmental feature quantity type = structure and type of the room, (b) sensor detection values ​​of distance sensors, image sensors, voice sensors, etc. are used as sensing data for extracting feature quantities.

[0790] Furthermore, as an example of the sensing process when (c) each sensing data indicated by (b) above is acquired, for example, the following sensing process is used.

[0791] (c1) Sensing while walking

[0792] (c2) Throwing and sensing an object (e.g., detecting whether the object disappears from an image captured by an image sensor (camera))

[0793] (c3) Barking and sensing (acquiring and analyzing reverberation echoes using a voice sensor (microphone), and analyzing the distance to objects in the surrounding environment)

[0794] Furthermore, examples of (d) actions to be determined and performed include the following actions.

[0795] (d1) Exploration of rooms with complex structures

[0796] (d2) Exploration in an unknown room

[0797] (d3) Being surprised in an unknown room

[0798] (d4) Approach and inspect furniture when it has changed, and register the changed furniture on the map

[0799] (d5) Don’t go to dangerous places

[0800] (d6) Behavior according to room characteristics

[0801] Specific examples of “(d1) Exploration in a case where the structure of a room is complicated” include, for example, processing of causing the robot 100 to move around in the room. Note that the robot 100 may be set to only look around without exploring in an open place.

[0802] Specific examples of “(d5) Do not go to dangerous places” include the following behaviors: do not go to places where the robot may get wet such as kitchen or sink areas, do not go to places where there are many obstacles, and further, back off if there are steps / cliffs.

[0803] Specific examples of “(d6) behavior according to characteristics of a room” include, for example, moving around in a noisy room and remaining quiet in a quiet room.

[0804] A specific example is moving to a room with high brightness.

[0805] With regard to smells, specific examples include avoiding rooms with unpleasant odors.

[0806] With regard to temperature, a specific example includes moving from a hot room to a cool room.

[0807] Specific examples include these behaviors.

[0808] Next, Fig.31 The data shown in FIG. 1 , that is, the type of the environmental feature quantity extracted by the environmental and body feature extraction unit 121, is

[0809] (3) Type of environmental characteristic quantity = Material of the object

[0810] A processing example in this case will be described.

[0811] In the case where the type of the environmental feature amount extracted by the environmental and body feature extraction unit 121 is the following (a):

[0812] (a) Type of environmental characteristic quantity = material of the object,

[0813] Sensor detection values ​​of a vibration sensor such as an IMU, a voice sensor, a joint angle sensor, a joint torque sensor, an image sensor, and the like are used as sensing data for extracting feature amounts.

[0814] Furthermore, as an example of the sensing process when (c) each sensing data indicated by (b) above is acquired, for example, the following sensing process is used.

[0815] (c1) Poke and sense the object

[0816] (c2) sensing how a person or other robot interacts with an object (obtaining images captured by, for example, an image sensor (camera))

[0817] Furthermore, examples of (d) actions to be determined and performed include the following actions.

[0818] (d1) Change how you grip an object (grip mode)

[0819] More specifically, for example, the robot only grasps objects that are soft and easily deformable.

[0820] This is a process of selecting and grasping an object that is easy to grasp and hardly causes trouble on the robot side.

[0821] Furthermore, the robot does not grip smooth objects (objects that have high reflectivity and cause overexposure).

[0822] This is because the object may be made of a material such as metal, slippery and heavy.

[0823] Next, Fig.32 The data shown in , that is, the types of environmental feature quantities extracted by the environmental and body feature extraction unit 121, are:

[0824] (4) Type of environmental characteristic quantity = human status

[0825] A processing example in this case will be described.

[0826] In the case where the type of environmental feature quantity extracted by the environmental and body feature extraction unit 121 is (a) type of environmental feature quantity = human state: (b) sensor detection values ​​of image sensors, voice sensors, etc. are used as sensing data for extracting feature quantities.

[0827] Furthermore, as an example of the sensing process when (c) each sensing data indicated by (b) above is acquired, for example, the following sensing process is used.

[0828] (c1) Sensing while walking

[0829] Furthermore, examples of (d) actions to be determined and performed include the following actions.

[0830] (d1) Stay still and don’t move when the person’s tone or expression indicates anger, move closer to the person when the person is depressed, and move more when the person is happy

[0831] (d2) Taking sniffing action when a person's clothing changes

[0832] (d3) Acting in a surprised manner when a person has had their hair cut or beard shaved

[0833] (d4) If the robot has never seen a human before, it will not approach a human to show fear.

[0834] (d5) Take a snuggling action with the person the robot has seen the most times

[0835] (d6) Take a happy action towards a person the robot has not seen for a long time

[0836] Next, Fig.32 The data shown in , that is, the types of environmental feature quantities extracted by the environmental and body feature extraction unit 121, are:

[0837] (5) Type of environmental feature quantity = state of the robot itself

[0838] A processing example in this case will be described.

[0839] In the case where the type of environmental feature quantity extracted by the environmental and body feature extraction unit 121 is the following (a) type of environmental feature quantity = state of itself (robot): (b) sensor detection values ​​of image sensors, vibration sensors such as IMU, voice sensors, joint angle sensors, joint torque sensors, etc. are used as sensing data for extracting feature quantities.

[0840] Furthermore, as an example of the sensing process when (c) each sensing data indicated by (b) above is acquired, for example, the following sensing process is used.

[0841] (c1) Sensing while walking

[0842] (c2) Take fixed actions and perform sensing

[0843] Furthermore, examples of (d) actions to be determined and performed include the following actions.

[0844] (d1) In the event of a failure at a component of the robot's body, the broken component is not used. Alternatively, it is moved so that force is not applied to the component.

[0845] (d2) Changes in how you walk, because the friction with the floor surface changes when you wear foot pads or shoes

[0846] (d3) Lifting the legs to get out of the way when the legs are caught in the rope

[0847] (d4) When the legs are replaced with accessories such as wheels, switching to an action that supports the accessories such as the replacement wheels

[0848] As described above, the robot (information processing device) 100 according to the present disclosure determines and executes various optimal behaviors according to the type of environmental feature amount extracted by the environment and body feature extraction unit 121 .

[0849] [6. Hardware Configuration Example of Robot (Information Processing Device) According to the Present Disclosure]

[0850] Next, we will refer to Fig.33 A hardware configuration example of the robot (information processing device) 100 according to the present disclosure is described.

[0851] Will describe Fig.33 The hardware configuration shown in .

[0852] A central processing unit (CPU) 301 functions as a data processing unit that performs various types of processing according to a program stored in a read-only memory (ROM) 302 or a storage unit 308. For example, processing according to the sequence described in the above embodiment is performed. A random access memory (RAM) 303 stores a program or data executed by the CPU 301. The CPU 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304.

[0853] The CPU 301 is connected to the input / output interface 305 via the bus 304, and the input / output interface 305 is connected to an input unit 306 including various switches, a touch panel and a microphone, and is additionally connected to a user input unit, a sensor detection value acquisition unit of various sensors 321 such as a camera device and a microphone, and an output unit 307 including a display and a speaker.

[0854] In addition, the output unit 307 also outputs driving information for the driving unit 322 .

[0855] For example, the CPU 301 receives input of commands, situation data, and the like input from the input unit 306 , executes various types of processing, and outputs processing results to the output unit 307 .

[0856] The storage unit 308 connected to the input / output interface 305 is formed of, for example, a hard disk or the like, and stores programs executed by the CPU 301 and various data. The communication unit 309 functions as a transmission / reception unit for data communication via a network such as the Internet or a local area network, and communicates with an external device.

[0857] Furthermore, in addition to the CPU, a graphics processing unit (GPU) may be provided as a dedicated processing unit for image information input from the camera device.

[0858] The drive 310 connected to the input / output interface 305 drives a removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory such as a memory card to record or read data.

[0859] [7. Summary of Configuration of the Present Disclosure]

[0860] The embodiments of the present disclosure have been described in detail above with reference to specific embodiments. However, it is obvious to those skilled in the art that the embodiments may be modified and substituted without departing from the gist of the present disclosure. That is, the present invention has been disclosed according to the form shown, but the present disclosure should not be interpreted restrictively. The gist of the present disclosure should be determined in consideration of the claims.

[0861] Note that the technology disclosed in this specification can have the following configurations.

[0862] (1) An information processing device comprising:

[0863] an environment and body feature extraction unit that receives an input of a detection value of a sensor attached to the robot and extracts at least any one of an environment feature quantity that is a feature quantity related to an environment around the robot and a body feature quantity that is a feature quantity related to the robot itself; and

[0864] A behavior control unit that determines and executes a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0865] (2) The information processing device according to (1),

[0866] The information processing device causes the robot to perform a behavior for acquiring a detection value of a sensor, and performs active sensor detection value acquisition processing.

[0867] (3) In the information processing device described in any one of (1) and (2), the environment and body feature extraction unit applies a machine learning model and extracts an environmental feature quantity as a feature quantity related to the environment around the robot or a body feature quantity as a feature quantity related to the robot itself based on the detection value of the sensor.

[0868] (4) In the information processing device described in any one of (1) to (3), the environment and body feature extraction unit extracts a feature quantity of at least any one of the material, hardness, and slipperiness of the floor surface where the robot is located as the environment feature quantity.

[0869] (5) In the information processing device described in any one of (1) to (4), the environment and body feature extraction unit extracts a feature amount indicating whether the robot is equipped with an accessory as the body feature amount.

[0870] (6) In the information processing device described in any one of (1) to (5), the behavior control unit determines at least any one of the robot's movement behavior, grasping behavior, environmental adaptation behavior, sensory expression and manners based on the feature quantity extracted by the environment and body feature extraction unit.

[0871] (7) In the information processing device described in any one of (1) to (6), the behavior control unit determines the walking pattern of the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0872] (8) In the information processing device described in any one of (1) to (7), the behavior control unit determines at least any one of the walking speed and stride of the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0873] (9) In the information processing device described in any one of (1) to (8), the behavior control unit determines an action that the robot should perform based on the feature amount extracted by the environment and body feature extraction unit, and causes the robot to perform the determined action.

[0874] (10) In the information processing device described in any one of (1) to (9), the behavior control unit performs processing to determine the expression of the robot based on the feature quantity extracted by the environment and body feature extraction unit and change the robot to the determined expression.

[0875] (11) In the information processing device described in any one of (1) to (10), the behavior control unit includes:

[0876] a behavior determination unit that determines a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit; and

[0877] A behavior execution unit causes the robot to execute the behavior determined by the behavior determination unit.

[0878] (12) In the information processing device described in (11), the behavior control unit further includes a behavior switching control unit that performs behavior switching control when the behavior determined by the behavior determination unit is different from the behavior currently performed by the robot.

[0879] (13) The information processing device according to any one of (1) to (12), wherein:

[0880] The information processing device further includes a sensing database storing detection values ​​of the sensors,

[0881] The environment and body feature extraction unit compares the sensor detection value at the current time point with the past sensor detection values ​​stored in the sensing database, and determines whether the environment around the robot is a previously experienced environment or a new environment.

[0882] (14) The information processing device according to any one of (1) to (13), wherein:

[0883] The information processing device further includes an environment and body feature amount database storing at least any one of the environment feature amount and the body feature amount extracted by the environment and body feature extraction unit, and

[0884] Wherein, the behavior control unit

[0885] The environmental feature quantity or the physical feature quantity at the current time point is compared with the past environmental feature quantity or the physical feature quantity stored in the environmental and physical feature quantity database, and

[0886] At least any one of a change in the environment around the robot and a change in the robot itself is detected based on the comparison result, and an optimal behavior matching the detected change is determined and executed.

[0887] (15) An information processing method executed in an information processing device, comprising:

[0888] an environment and body feature extraction step in which an environment and body feature extraction unit receives an input of a detection value of a sensor attached to the robot and extracts at least any one of an environment feature quantity as a feature quantity related to an environment around the robot and a body feature quantity as a feature quantity related to the robot itself; and

[0889] A behavior control step in which the behavior control unit determines and executes a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0890] (16) A program for causing an information processing device to execute information processing, the program causing execution of:

[0891] an environment and body feature extraction step in which an environment and body feature extraction unit is caused to receive an input of a detection value of a sensor attached to the robot, and extract at least any one of an environment feature quantity as a feature quantity related to an environment around the robot and a body feature quantity as a feature quantity related to the robot itself; and

[0892] A behavior control step in which the behavior control unit is caused to determine and execute a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

[0893] In addition, the series of processing described in this specification can be performed by hardware, software or a combination of the two. In the case where a series of processing is to be performed by software, the series of processing can be performed by installing a program recording the processing sequence into the memory of a computer embedded in special hardware or by installing the program into a general-purpose computer capable of performing various types of processing. For example, the program can be recorded in advance on a recording medium. The program can be received via a network such as a local area network (LAN) or the Internet and installed into a built-in recording medium such as a hard disk, rather than being installed into a computer from a recording medium.

[0894] Note that the various processes described in this specification may be performed continuously in the order described, or may be performed in parallel or individually, depending on the processing capability of the device that performs the processing or as needed. In addition, in this specification, a system is a logical collection of configurations of multiple devices, and devices having each configuration are not limited to devices in the same housing.

[0895] [Industrial Applicability]

[0896] As described above, the configuration according to one embodiment of the present disclosure realizes a configuration that enables a robot to autonomously perform optimal behavior matching the surrounding environment of the robot or the state of the robot itself.

[0897] More specifically, for example, the configuration includes: an environment and body feature extraction unit that receives an input of a detection value of a sensor attached to the robot, and extracts at least one of an environment feature amount that is a feature amount related to an environment around the robot and a body feature amount that is a feature amount related to the robot itself, and

[0898] A behavior control unit that determines and executes the behavior to be performed by the robot based on the extracted feature quantity. For example, this configuration acquires the slipperiness of the floor as the environmental feature quantity, determines and executes the best behavior that matches the feature quantity in the case of a slippery floor, and executes walking at a reduced walking speed, etc.

[0899] This configuration realizes a configuration that enables the robot to autonomously execute optimal behavior that matches the robot's surrounding environment or the robot's own state.

[0900] [reference numerals list]

[0901] 100 Robots (information processing devices)

[0902] 110 sensing units

[0903] 120 data processing units

[0904] 121 Environment / Feature Extraction Unit

[0905] 122 Behavior Control Unit

[0906] 123 Behavior determination unit

[0907] 124 Behavior switching control unit

[0908] 125 Behavior Execution Unit

[0909] 126 Motion Control Unit

[0910] 127 Expression Control Unit

[0911] 131 Sensor Data Database

[0912] 132Environment / Feature Database

[0913] 151 Clothing (Accessories)

[0914] 152 Shoes (Accessories)

[0915] 153 foot pad (accessory)

[0916] 301CPU

[0917] 302ROM

[0918] 303RAM

[0919] 304 bus

[0920] 305 Input / Output Interface

[0921] 306 Output Unit

[0922] 307 Output Unit

[0923] 308 storage unit

[0924] 309 Communication Unit

[0925] 310 Drive

[0926] 311 Removable Media

[0927] 321 Sensor

[0928] 322 drive part

Claims

1. An information processing device, comprising: an environment and body feature extraction unit that receives an input of a detection value of a sensor attached to the robot and extracts at least any one of an environment feature quantity that is a feature quantity related to an environment around the robot and a body feature quantity that is a feature quantity related to the robot itself; and A behavior control unit determines and executes a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

2. The information processing device according to claim 1, wherein: The information processing device causes the robot to perform a behavior for acquiring a detection value of the sensor, and performs active sensor detection value acquisition processing.

3. The information processing device according to claim 1, wherein: The environment and body feature extraction unit applies a machine learning model and extracts an environment feature quantity as a feature quantity related to the environment around the robot or a body feature quantity as a feature quantity related to the robot itself based on the detection value of the sensor.

4. The information processing device according to claim 1, wherein: The environment and body feature extraction unit extracts at least any one feature quantity of the material, hardness and slipperiness of the floor surface where the robot is located as the environment feature quantity.

5. The information processing device according to claim 1, wherein: The environment and body feature extraction unit extracts a feature amount indicating whether the robot is equipped with an accessory as a body feature amount.

6. The information processing device according to claim 1, wherein: The behavior control unit determines at least any one of the movement behavior, grasping behavior, environment adaptation behavior, feeling expression, and mannerism of the robot based on the feature amount extracted by the environment and body feature extraction unit.

7. The information processing device according to claim 1, wherein: The behavior control unit determines a walking pattern of the robot based on the feature amount extracted by the environment and body feature extraction unit.

8. The information processing device according to claim 1, wherein: The behavior control unit determines at least any one of a walking speed and a stride of the robot based on the feature amount extracted by the environment and body feature extraction unit.

9. The information processing device according to claim 1, wherein: The behavior control unit determines an action that the robot should perform based on the feature amount extracted by the environment and body feature extraction unit, and causes the robot to perform the determined action.

10. The information processing device according to claim 1, wherein: The behavior control unit performs a process of determining an expression of the robot based on the feature amount extracted by the environment and body feature extraction unit and causing the robot to change to the determined expression.

11. The information processing device according to claim 1, wherein: The behavior control unit comprises: a behavior determination unit that determines a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit; and A behavior execution unit causes the robot to execute the behavior determined by the behavior determination unit.

12. The information processing device according to claim 11, wherein: The behavior control unit further includes a behavior switching control unit that performs behavior switching control in a case where the behavior determined by the behavior determination unit is different from a behavior currently performed by the robot.

13. The information processing device according to claim 1, wherein: The information processing device further includes a sensing database storing the detection value of the sensor, The environment and body feature extraction unit compares the sensor detection value at the current time point with the past sensor detection values ​​stored in the sensing database, and determines whether the environment around the robot is a previously experienced environment or a new environment.

14. The information processing device according to claim 1, wherein: The information processing device further includes an environment and body feature amount database storing at least any one of the environment feature amount and the body feature amount extracted by the environment and body feature extraction unit, Wherein, the behavior control unit comparing the environmental feature quantity or the physical feature quantity at the current time point with the past environmental feature quantity or the physical feature quantity stored in the environmental and physical feature quantity database, and At least any one of a change in the environment around the robot and a change in the robot itself is detected based on the comparison result, and an optimal behavior matching the detected change is determined and executed.

15. An information processing method executed in an information processing device, comprising: an environment and body feature extraction step in which an environment and body feature extraction unit receives an input of a detection value of a sensor attached to the robot and extracts at least any one of an environment feature quantity as a feature quantity related to an environment around the robot and a body feature quantity as a feature quantity related to the robot itself; and a behavior control step in which a behavior control unit determines and executes a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

16. A program for causing an information processing device to execute information processing, the program causing execution of: an environment and body feature extraction step in which an environment and body feature extraction unit is caused to receive an input of a detection value of a sensor attached to the robot, and extract at least any one of an environment feature quantity as a feature quantity related to an environment around the robot and a body feature quantity as a feature quantity related to the robot itself; and A behavior control step in which a behavior control unit is caused to determine and execute a behavior to be performed by the robot based on the feature amount extracted by the environment and body feature extraction unit.

Images