AUTONOMOUSLY ACTING ROBOT SEEKS COOLNESS

Abstract

Autonomously acting robot (100), comprising: a storage device (124, 148); a processor (122, 136) connected to the storage device (124, 148), the processor (122, 136) configured to execute the instructions for: determining a cool point (C) as a movement target point, wherein the cool point (C) is a location having a lower ambient temperature than an ambient temperature at a current location of the autonomously acting robot (100); selecting a route (R1) from a plurality of routes (R1, R2) from the current location to the movement destination point, wherein selecting the route (R1) comprises selecting a route (R1) that is assumed to have a lowest average temperature of the plurality of routes (R1, R2); specifying a direction of movement along the selected route (R1); and a drive mechanism (120) configured to perform the specified movement.

Description

[0001] The present disclosure relates to a robot that autonomously selects an action according to an internal state or an external environment.

[0002] A person perceives various elements of information from an external environment through their sense organs and selects an action. There are times when an action is chosen consciously and times when an action is chosen unconsciously. A repeated action becomes an unconscious action over time, and a new action remains in a zone of consciousness.

[0003] A person believes that he or she has the will to freely choose an action—that is, has free will. The reason a person experiences feelings of affection or hostility toward another person is because he or she believes that the other person also has free will. A person who has free will, or at least an existence that can be assumed to have free will, is also an existence that alleviates a person's sadness.

[0004] One reason a person keeps a pet is because it provides comfort, not because the pet is useful to the person. Precisely because a pet is an existence that more or less gives the impression of having free will, the pet can become a good companion for a person.

[0005] Meanwhile, for various reasons, such as not being able to devote enough time to caring for a pet, not having a home environment suitable for keeping a pet, suffering from allergies, or hating the thought of being separated by death, many people choose not to keep a pet. A robot that acts as a pet can provide the comfort that a pet provides to people who cannot keep a pet (see Patent Document 1).

[0006] JP 2000-323219 A discloses a connecting device and a robot system. Further prior art is represented by JP 2007-061962 A, which discloses a mobile robot, a temperature control device, and a corresponding method.

[0007] Although robot technology has advanced rapidly in recent years, the technology hasn't yet produced a pet-like companion presence. The inventors believe this is because humans don't assume a robot has free will. By observing an action that can only be assumed to be a pet's, a human senses the existence of free will in the pet, feels compassion for the pet, and receives comfort from the pet.

[0008] The inventors believe that if there were a robot that could imitate a human- or animal-like action, in other words, a robot that could autonomously select a human- or animal-like action, compassion towards the robot could be greatly increased.

[0009] The invention, which was made on the basis of a finding of the facts described so far, has as its main object the provision of a behavior control technology that expresses animal-like behavioral characteristics of a robot, and in particular, behavioral characteristics of searching for a cool place.

[0010] An autonomously acting robot in one aspect of the invention includes an operation determining unit that specifies a movement direction and a drive mechanism that executes the movement specified by the operation determining unit.

[0011] The operation determination unit specifies a cool point, which is a point with a temperature lower than that of a current point, as a movement target point.

[0012] An autonomously acting robot in another aspect of the invention includes an operation determination unit that specifies a movement direction, a drive mechanism that executes the movement specified by the operation determination unit, and a detection unit that detects a cooling device or a controller of the cooling device.

[0013] The operation determination unit sets a direction as a movement direction in which the cooling device or the controller is located when an internal temperature of the robot reaches or exceeds a predetermined value or when the internal temperature is predicted to reach or exceed the predetermined value.

[0014] An autonomously acting robot in another aspect of the invention includes an operation determining unit that specifies a movement direction, a drive mechanism that executes the movement specified by the operation determining unit, and a cooling mechanism that causes a speed change of a fan based on an internal temperature.

[0015] The operation determination unit sets the direction of movement in a direction away from a user when the rotational speed of the fan reaches or exceeds a predetermined value, or when the rotational speed of the fan is predicted to reach or exceed the predetermined value.

[0016] An autonomously acting robot in another aspect of the invention includes an operation determining unit that specifies a movement direction, a drive mechanism that executes the movement specified by the operation determining unit, and a cooling mechanism that causes a speed change of a fan based on an internal temperature.

[0017] The cooling mechanism sets an upper limit of the fan speed lower when a user is within a predetermined range than when the user is not there.

[0018] An autonomously acting robot in another aspect of the invention includes an operation determination unit that specifies a movement direction, a drive mechanism that executes the movement specified by the operation determination unit, and a body temperature detection unit that detects a body temperature of a user.

[0019] The operation determining unit instructs the drive mechanism to perform a predetermined movement to guide the user to a cool point when the user's body temperature is a predetermined value or higher.

[0020] According to embodiments of the invention, compassion towards a robot can be easily increased. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1(a) is a front external view of a robot. Fig. 1(b) is a side external view of the robot. Fig. 2 is a configuration diagram of a robot system. Fig. Figure 3 is a schematic view of an emotion map. Fig. 4 is a hardware configuration diagram of the robot. Fig. Figure 5 is a functional block diagram of the robot system. Fig. 6 is a schematic view for describing a function of the coolness-seeking robot. Fig. Figure 7 is a schematic view describing the temporal change of a temperature map. Fig. Figure 8 is a flowchart showing a process in which the robot moves toward a cool spot. Fig. 9 is a functional block diagram of the robot system in a modified example. Description of the embodiments

[0021] Fig. 1A is a front exterior view of a robot 100. Fig. Figure 1B is a side exterior view of robot 100.

[0022] The robot 100 in at least one embodiment is an autonomously acting robot that determines an action or gesture based on an external environment and an internal state. The external environment is detected by various types of sensors, such as a camera or a thermal sensor. The internal state is quantified as various parameters that express emotions of the robot 100. These are described below.

[0023] Assuming an indoor action, the robot 100 of at least one embodiment has, for example, an interior of a private home as its action area. Hereinafter, a person engaging with the robot 100 is referred to as a "user," and a user who is a member of a household to which the robot 100 belongs is referred to as an "owner."

[0024] A body 104 of the robot 100 has a rounded shape throughout and is formed of a soft material with elasticity, such as urethane, rubber, or resin. The robot 100 can be clothed. By adopting the rounded, soft, and comfortable-to-touch body 104, the robot 100 provides the user with a sense of security and a pleasant tactile sensation.

[0025] A total weight of the robot 100 is 15 kilograms or less. In at least one embodiment, the total weight of the robot 100 is 10 kilograms or less. In at least one embodiment, the total weight of the robot 100 is 5 kilograms or less. Most babies begin walking independently 13 months after birth. An average weight of a baby 13 months after birth is slightly more than 9 kilograms for boys and slightly less than 9 kilograms for girls. Therefore, when the total weight of the robot 100 is 10 kilograms or less, a user can hold the robot 100 with an effort virtually equivalent to holding a baby that cannot walk unaided.

[0026] The average weight of a baby less than 2 months after birth is less than 5 kilograms for both boys and girls. As a result, if the total weight of the robot 100 is 5 kilograms or less, a user can hold the robot 100 with an effort practically equivalent to holding a very small baby.

[0027] The user's benefits of being able to easily hold and desire to hold the robot 100 are realized through attributes such as reasonable weight and roundness, softness, and pleasantness to the touch. For the same reasons, the height of the robot 100 is 1.2 meters or less. In at least one embodiment, the height of the robot 100 is 0.7 meters or less.

[0028] The ability to be held is a concept of robot 100 in at least one embodiment.

[0029] The robot 100 moves using a wheel 102. A rotational speed and a rotational direction of two of the wheels 102 can be individually controlled. Furthermore, the wheel 102 can also be pushed upward into an interior of the body 104 of the robot 100 and completely stored within the body 104. A larger part of the wheel 102 is hidden by the body 104 even during locomotion, but when the wheel 102 is completely stored within the body 104, the robot 100 is in a state of immobility (hereinafter referred to as a "sitting state"). In the sitting state, a flat seat surface 108 is in contact with a floor surface.

[0030] The robot 100 has two arms 106. The arms 106 do not have an object-grasping function. The arms 106 can perform simple actions such as lifting, waving, and swinging. The two arms 106 can also be controlled individually.

[0031] A camera is integrated into an eye 110. The eye 110 is also capable of image display using a liquid crystal element or an organic EL element. In addition to the camera integrated into the eye 110, various sensors, such as a highly directivity microphone or an ultrasonic sensor, are mounted in the robot 100. A speaker is also integrated, and the robot 100 is also capable of simple speech.

[0032] A horn 112 is attached to a head portion of the robot 100. Since the robot 100 is lightweight as described above, a user can also lift the robot 100 by grasping the horn 112.

[0033] Fig. 2 is a configuration diagram of a robot system 300.

[0034] The robot system 300 includes the robot 100, a server 200, and a plurality of external sensors 114. The plurality of external sensors 114 (external sensors 114a, 114b, and so on through 114n) are installed in advance in a house. The external sensor 114 can be attached to a wall surface of the house or can be placed on a floor. Position coordinates of the external sensor 114 are registered in the server 200. The position coordinates are defined as x- and y-coordinates in the house, which is designated as the operating range of the robot 100.

[0035] The server 200 is installed in the house. The server 200 and the robot 100, in at least one embodiment, correspond one-to-one. The server 200 determines a basic action of the robot 100 based on information obtained from the sensors integrated into the robot 100 and the multiple external sensors 114.

[0036] The external sensor 114 serves to amplify the sensory components of the robot 100, and the server 200 serves to amplify the computing power of the robot 100.

[0037] The external sensor 114 periodically transmits a wireless signal (hereinafter referred to as a "robot search signal") including the ID (hereinafter referred to as a "beacon identifier") of the external sensor 114. Upon receiving the robot search signal, the robot 100 returns a wireless signal (hereinafter referred to as a "robot response signal") including the beacon identifier. The server 200 measures a time from the transmission of the robot search signal by the external sensor 114 to the receipt of the robot response signal and measures a distance from the external sensor 114 to the robot 100. By measuring the distance between each of the plurality of external sensors 114 and the robot 100, the server 200 obtains the position coordinates of the robot 100.

[0038] Of course, a method can also be used in which the robot 100 regularly sends position coordinates to the server 200.

[0039] Fig. 3 is a schematic view of an emotion map 116.

[0040] The emotion map 116 is a data table stored in the server 200. The robot 100 selects an action according to the emotion map 116. Fig. The emotion map 116 shown in Figure 3 shows a magnitude of emotional attraction or dislike toward a location of the robot 100. An x-axis and a y-axis of the emotion map 116 indicate two-dimensional spatial coordinates. A z-axis indicates a magnitude of emotional attraction or dislike. When a z-value is a positive value, attraction to a location is high, and when the z-value is a negative value, the robot 100 is disliked toward the location.

[0041] On the emotion map 116 of Fig. 3, a coordinate P1 is a point in an indoor space managed by the server 200 as the action area of ​​the robot 100 where an emotion of attraction is high (hereinafter referred to as a preferred point). The preferred point may be a "safe place," such as behind a sofa or under a table, or may be a place where people gather or a lively place, such as a living room. The safe place may also be a place where the robot 100 has been gently stroked or touched in the past.

[0042] A definition of what type of location the robot 100 prefers is arbitrary, but the preferred location is usually a location preferred by small children or small animals such as dogs or cats and is set as the preferred point.

[0043] A coordinate P2 is a point where an emotion of dislike is high (hereinafter referred to as an "undesirable point"). The undesirable point may be a location where there is a loud noise, such as near a TV, a location likely to leak, such as a bathroom or washroom, a closed space or a dark place, a location where the robot 100 was roughly handled by a user and which evokes an unpleasant memory, or the like.

[0044] A definition of what kind of place the robot 100 does not like is also arbitrary, but the undesirable place is usually a place feared by small children or by small animals such as dogs or cats, and is set as an undesirable point.

[0045] A coordinate Q indicates a current position of the robot 100. The server 200 identifies position coordinates of the robot 100 using the robot search signal regularly transmitted from the multiple external sensors 114 and the robot response signal that responds to the robot search signal. For example, when the external sensor 114 with beacon ID = 1 and the external sensor 114 with beacon ID = 2 each detect the robot 100, the server 200 obtains the distances of the robot 100 from the two external sensors 114 and obtains the position coordinates of the robot 100 from the distances.

[0046] Alternatively, the external sensor 114 with the beacon ID = 1 transmits the robot search signal in multiple directions, and the robot 100 returns the robot response signal upon receiving the robot search signal. In this way, the server 200 can determine the direction and distance of the robot 100 from each external sensor 114. Furthermore, in another embodiment, the server 200 can calculate a distance traveled by the robot 100 from the rotational speed of the wheel 102, thereby determining the current position, or can determine the current position from an image obtained from the camera.

[0047] If the Fig. 3 is provided, the robot 100 moves in a direction toward the preferred point (coordinate P1) or in a direction away from the undesired point (coordinate P2).

[0048] The emotion map 116 changes dynamically. When the robot 100 arrives at coordinate P1, the z-value (attraction emotion) at coordinate P1 decreases over time. Therefore, the robot 100 can mimic animal-like behavior of "being emotionally satisfied" when it arrives at the preferred point (coordinate P1) and "becoming bored" of the location over time. Similarly, the emotion of aversion at coordinate P2 diminishes over time. As time passes, a new preferred point or undesirable point emerges, causing the robot 100 to make a new action selection. The robot 100 is "interested" in a new preferred point and continuously performs action selection.

[0049] The emotion map 116 expresses emotional fluctuations as an internal state of the robot 100. The robot 100 moves toward a preferred point, avoids an undesirable point, stays at the preferred point for a while, and performs the next action over time. With this type of control, the action selection of the robot 100 can be a human-like or animal-like action selection.

[0050] Cards that influence an action of the robot 100 (hereinafter collectively referred to as “action cards”) are not limited to the Fig. 3. For example, various action maps such as curiosity, a desire to avoid fear, a desire to seek safety, and a desire to find physical comfort such as quiet, dim light, coolness, or warmth may be defined. Furthermore, a target point of the robot 100 may be determined by calculating a weighted average of the z-values ​​of each of the multiple action maps.

[0051] In addition to an action map, the robot 100 may also have parameters that indicate the magnitude of various emotions or senses. For example, as a value of a sadness emotion parameter increases, a weighting coefficient of an action map that evaluates locations where the robot 100 feels comfortable may be increased, and the value of this emotion parameter may be decreased when the robot 100 reaches a target point. Similarly, as a value of a parameter indicating a feeling of boredom increases, a weighting coefficient of an action map that evaluates locations where curiosity is satisfied may be increased.

[0052] Fig. 4 is a hardware configuration diagram of the robot 100.

[0053] The robot 100 includes an internal sensor 128, a communicator 126, a storage device 124, a processor 122, a drive mechanism 120, and a battery 118. The units are connected to each other by a power line 130 and a signal line 132. The battery 118 supplies power to each unit via the power line 130. Each unit sends and receives a control signal via the signal line 132. The battery 118 is a rechargeable battery, such as a rechargeable lithium-ion battery, and is a power source for the robot 100.

[0054] The internal sensor 128 is a collection of various types of sensors integrated into the robot 100. Specifically, the internal sensor 128 includes a camera, a high-directivity microphone, an infrared sensor, a thermal sensor, a touch sensor, an acceleration sensor, an odor sensor, and the like. The odor sensor is a conventional sensor that uses a principle in which electrical resistance changes according to the adsorption of a molecule that constitutes an odor source. The odor sensor classifies various odors into several types of categories (hereinafter referred to as "odor categories").

[0055] The communicator 126 is a communication module that performs wireless communication with the server 200 and various types of external devices as targets, such as the external sensor 114 and a mobile device owned by the user. The storage device 124 is configured from a non-volatile memory and a volatile memory and stores a computer program and various types of setting information. The processor 122 is a means for executing a computer program. The drive mechanism 120 is an actuator that controls various mechanisms, such as the wheels 102 and the arms 106.

[0056] In addition, a display device, a speaker, and the like are also mounted in the robot 100.

[0057] The processor 122 selects an action of the robot 100 while communicating with the server 200 or the external sensor 114 via the communicator 126. Various types of external information obtained from the internal sensor 128 also influence the action selection. The drive mechanism 120 mainly controls the wheel 102 and the arm 106. The drive mechanism 120 changes a moving direction and a moving speed of the robot 100 by changing the rotational speed and direction of the two wheels 102. In addition, the drive mechanism 120 can also raise and lower the wheel 102. When the wheel 102 ascends, the wheel 102 is fully supported in the body 104, and the robot 100 comes into contact with a floor surface via the seat surface 108 and assumes a sitting state.

[0058] The arm 106 can be raised by the drive mechanism 120, which pulls the arm 106 along a wire 134. A gesture such as waving the arm can also be performed by swinging the arm 106. A more complex gesture can also be performed by using a large amount of wire 134. That is, with the number of wires 134 in the arm 106, the complexity of the possible gestures by the arm 106 increases.

[0059] Fig. 5 is a functional block diagram of a robot system 300.

[0060] As described above, the robot system 300 includes the robot 100, the server 200, and a plurality of external sensors 114. Each component of the robot 100 and the server 200 is realized by hardware including a computer composed of a CPU (central processing unit), various types of coprocessors, and the like, a storage device that is a memory or storage, and a wired or wireless communication line that connects the computer and the storage device, as well as software stored in the storage device that provides a processing instruction to the computer. A computer program may be configured from a device driver, an operating system, various types of application programs arranged in an upper layer thereof, and a library that provides a common function to the programs.Each block described below refers to a functional unit block rather than a hardware unit configuration.

[0061] Part of the functions of the robot 100 can be realized by the server 200, and part or all of the functions of the server 200 can be realized by the robot 100. Server 200

[0062] The server 200 includes a communication unit 204, a data processing unit 202, and a data storage unit 206. The communication unit 204 manages a process of communication with the external sensor 114 and the robot 100. The data storage unit 206 stores various types of data. The data processing unit 202 executes various types of processes based on the data acquired by the communication unit 204 and the data stored in the data storage unit 206. The data processing unit 202 also functions as an interface between the communication unit 204 and the data storage unit 206.

[0063] The data storage unit 206 includes an operation pattern storage unit 232, a map storage unit 216, and an individual data storage unit 218. The operation pattern storage unit 232 correlates the identifier of operation patterns (hereinafter referred to as "operation identifier") expressing each type of gesture of the robot 100 and selection conditions thereof. The map storage unit 216 stores a plurality of action maps. The individual data storage unit 218 stores information about a user, particularly about an owner. Specifically, the individual data storage unit 218 stores various types of parameters, such as familiarity with a user, as well as physical characteristics and behavioral characteristics of a user. The individual data storage unit 218 may also store attribute information such as age and gender.

[0064] The robot 100 identifies a user based on the user's physical and behavioral characteristics. The robot 100 continuously captures an external environment with the integrated camera. Furthermore, the robot 100 extracts the physical and behavioral characteristics of a person appearing in an image. The physical characteristics may be body-inherent visual characteristics such as body size, selected clothing, presence or absence of glasses, skin gloss, hair color, or ear size, or may also include other characteristics such as average body temperature, odor, or voice quality. The behavioral characteristics are, in particular, behavior-accompanying characteristics such as a user's preferred location, vigorousness of movement, and presence or absence of smoking.For example, the robot 100 extracts behavioral characteristics that an owner identified as a father is often outside the house and at home is often lying motionless on a sofa, but a mother is often in a kitchen and an activity area is extensive.

[0065] The robot 100 clusters users who appear with high frequency as “owners” based on physical characteristics and behavioral characteristics obtained from a large amount of image information and other measurement information.

[0066] Although a method for identifying a user based on user ID is simple and reliable, it requires that the user have a device capable of providing user ID. Meanwhile, the method for identifying a user based on physical or behavioral characteristics requires an image recognition method that is complex, but it has the advantage of being able to identify even a user who does not have a mobile device. Either of the two methods can be used alone, or user identification can be performed using the two methods together in a complementary manner.

[0067] In at least one embodiment, users are clustered based on physical and behavioral characteristics, and a user is identified using deep learning (a multi-layer neural network). Details are described below.

[0068] The robot 100 has an internal familiarity parameter for each user. When the robot 100 detects an action that indicates affection toward the robot 100, such as picking up the robot 100 or talking to the robot 100, the familiarity with that user increases. Familiarity decreases with a user who is not engaged with the robot 100, a user who behaves rudely, or a user who is rarely encountered.

[0069] The data processing unit 202 includes a position management unit 208, a map management unit 210, a recognition unit 212, an operation determination unit 222, and a familiarity management unit 220.

[0070] The position management unit 208 identifies the position coordinates of the robot 100 using the Fig. 2. The position management unit 208 can also track a user's position coordinates in real time.

[0071] The map management unit 210 changes the parameters of each coordinate in the plurality of action maps using the Fig. 3. A temperature map management unit 226, which is part of the functions of the map management unit 210, manages a temperature map, which is a type of action map. The temperature map is described below.

[0072] The map management unit 210 may select one of the multiple action maps or may calculate a weighted average of the z-values ​​of the multiple action maps. For example, the z-values ​​at a coordinate R1 and a coordinate R2 on an action map A are 4 and 3, and the z-values ​​at the coordinate R1 and the coordinate R2 on an action map B are -1 and 3. When calculating a simple average, the total z-value at the coordinate R1 is 4 - 1 = 3, and the total z-value at the coordinate R2 is 3 + 3 = 6, which is why the robot 100 steers toward the coordinate R2 instead of the coordinate R1.

[0073] If action card A is weighted 5 times with respect to action card B, the total z-value at coordinate R1 is 4 × 5 - 1 = 19 and the total z-value at coordinate R2 is 3 × 5 + 3 = 18, which is why robot 100 steers in the direction of coordinate R1.

[0074] The recognition unit 212 detects an external environment. Detection of the external environment includes various types of detection, such as detecting weather or season based on temperature and humidity, and detecting shelter (a safe area) based on light and temperature levels. The recognition unit 212 further includes a person detection unit 214 and a response detection unit 228. The person detection unit 214 detects a person from an image captured by the camera integrated in the robot 100 and extracts the person's physical and behavioral characteristics.Further, based on the physical feature information and behavioral feature information registered in the individual data storage unit 218, the person recognition unit 214 determines which person, such as a father, mother, or eldest son, the filmed user, that is, the user viewed by the robot 100, corresponds to. The person recognition unit 214 includes an expression recognition unit 230. The person recognition unit 214 includes an expression recognition unit 230. The expression recognition unit 230 derives an emotion of a user using image recognition of an expression of the user.

[0075] In addition to a person, the person recognition unit 214 also extracts features of, for example, a cat or a dog that is a pet. The following description assumes that not only a person but also a pet is included as a user or owner.

[0076] The response detection unit 228 detects various response actions performed with respect to the robot 100 and classifies the actions as pleasant or unpleasant. The response detection unit 228 also detects a response action of an owner with respect to an action of the robot 100, thereby classifying the response action as a positive or negative reaction.

[0077] Pleasant and unpleasant actions are distinguished depending on whether a user's response action is pleasant or unpleasant for an animal. For example, being hugged is a pleasant action for the robot 100, and being kicked is an unpleasant action for the robot 100. Positive and negative reactions are distinguished depending on whether a user's response action indicates a pleasant emotion or an unpleasant emotion of the user. For example, being hugged is a positive reaction indicating a pleasant emotion of the user, and being kicked is a negative reaction indicating an unpleasant emotion of the user.

[0078] The operation determination unit 222 of the server 200 determines an operation (motion and gesture) of the robot 100 in cooperation with an operation determination unit 150 of the robot 100. The operation determination unit 222 includes a motion determination unit 234 and an action determination unit 236. The motion determination unit 234 compiles a movement target point of the robot 100 and a movement route for the movement target point based on an action map selection by the map management unit 210. The motion determination unit 234 compiles multiple movement routes and can select any one of the movement routes after this operation. The action determination unit 236 selects a gesture of the robot 100 from the multiple operation patterns of the operation pattern storage unit 232.

[0079] The familiarity management unit 220 manages the familiarity for each user. As described above, the familiarity is registered as a piece of individual data in the individual data storage unit 218. Details of the familiarity management will be described below. Robot 100

[0080] The robot 100 includes a communication unit 142, a data processing unit 136, a data storage unit 148, a drive mechanism 120, an internal sensor 128, and a cooling mechanism 162.

[0081] The communication unit 142 corresponds to the communicator 126 (see Fig. 4) and manages a communication process with the external sensor 114 and the server 200. The data storage unit 148 stores various types of data. The data storage unit 148 corresponds to the storage device 124 (see Fig. 4). The data processing unit 136 executes various types of processes based on the data acquired by the communication unit 142 and the data stored in the data storage unit 148. The data processing unit 136 corresponds to the processor 122 and a computer program executed by the processor 122. The data processing unit 136 also functions as an interface between the communication unit 142, the internal sensor 128, the drive mechanism 120, and the data storage unit 148.

[0082] The internal sensor 128 includes a temperature detection unit 152.

[0083] The temperature detection unit 152 measures a user's body temperature and an external surface temperature. The temperature detection unit 152 includes a non-contact temperature sensor, such as a radiation thermometer or thermography, and a contact temperature sensor, such as a thermistor, a resistance temperature detector, a thermocouple, or an IC temperature sensor.

[0084] The cooling mechanism 162 includes a heat sink and a fan. The fan of the cooling mechanism 162 draws in outside air and dissipates the heat trapped in an interior space. An operating level of the cooling mechanism 162 can be controlled by the speed of the fan, which is controlled according to a temperature inside the robot 100.

[0085] The data storage unit 148 includes an operation pattern storage unit 160 that defines various operation types of the robot 100.

[0086] The action identifier and an action selection condition are correlated in the action pattern storage unit 160 of the server 200. For example, a selection probability of an action pattern A upon detection of an unpleasant action is recorded in correlation with the action identifier. The action determination unit 236 of the server 200 selects an action pattern based on this type of selection condition.

[0087] The operation identifier and a method for controlling various types of actuators to realize an operation thereof are defined in the operation pattern storage unit 160 of the robot 100. Specifically, an operation timing, an operation time, an operation direction, and the like of the various types of actuators (the drive mechanism 120) are chronologically defined for each operation pattern to express various gestures, such as retracting the wheel 102 and sitting down, raising the arm 106, causing the robot 100 to perform a rotating action by causing the two wheels 102 to rotate in opposite directions or causing only one wheel 102 to rotate, shaking by causing the wheel 102 to rotate in a state where the wheel 102 is in a retracted state, and stopping once and looking back when moving away from a user.

[0088] The data processing unit 136 includes a recognition unit 156 and the operation determination unit 150. The operation determination unit 150 of the robot 100 determines an operation of the robot 100 in cooperation with the operation determination unit 222 of the server 200. The operation determination unit 150 includes a movement determination unit 138 and an action determination unit 140.

[0089] The drive mechanism 120 includes a motion drive unit 144 and an action drive unit 146. The motion determination unit 138, together with the motion determination unit 234 of the server 200, determines a movement direction of the robot 100. A movement based on an action map can be determined by the server 200, and an instantaneous movement, such as avoiding an obstacle, can be determined by the motion determination unit 138 of the robot 100. The motion drive unit 144 causes the robot 100 to move toward a target point by driving the wheel 102 according to an instruction from the motion determination unit 138.

[0090] Although an action map determines the main element of the movement direction of the robot 100, the robot 100 can also perform an action that corresponds to familiarity.

[0091] The operation identifier selected by the action determination unit 236 of the server 200 is forwarded to the robot 100, and the action determination unit 140 of the robot 100 instructs the action driving unit 146 to execute an operation pattern corresponding to the operation identifier.

[0092] Part of the complex operation patterns may be determined by the server 200, and other operation patterns may be determined by the robot 100. Alternatively, a basic operation pattern may be determined by the server 200, and an additional operation pattern may be determined by the robot 100. It is sufficient that the manner in which operation pattern determination processes are divided between the server 200 and the robot 100 is designed according to the specifications of the robot system 300.

[0093] The action determination unit 140 can also perform a gesture of holding up both arms 106 as a gesture of requesting "a hug" when a user with a high degree of familiarity is nearby, and can also perform a gesture of not wanting to be hugged any longer by causing the wheel 102 to rotate in the opposite direction in a retracted state when the "hug" becomes boring. The action drive unit 146 causes the robot 100 to perform various gestures by driving the wheel 102 and the arm 106 according to an instruction from the action determination unit 140.

[0094] The detection unit 156 analyzes external information received from the internal sensor 128. The detection unit 156 is capable of visual detection (a visual unit), olfactory detection (an olfactory unit), sound detection (an auditory unit), and tactile detection (a tactile unit).

[0095] The recognition unit 156 periodically films an external viewing angle using the integrated camera (internal sensor 128) and recognizes a user who is a moving object such as a person or a pet. These characteristics are transmitted to the server 200, and the person recognition unit 214 of the server 200 extracts the physical characteristics of the moving object. Furthermore, the recognition unit 156 also detects a user's smell and voice. The smell and sound (voice) are classified into several types using a previously known method. Furthermore, the temperature detection unit 152 can also detect a touch temperature.

[0096] When a strong force is applied to the robot 100, the detection unit 156 detects this using a built-in acceleration sensor, and the response detection unit 228 of the server 200 detects that a "violent action" has been performed by a nearby user. If a user lifts the robot 100 by grasping the horn 112, this can also be detected as a violent action. If a user speaks in a state of facing the robot 100 within a certain volume range and frequency band, the response detection unit 228 of the server 200 detects that a speech action has been performed with respect to the robot 100.Furthermore, when a temperature within the range of body temperature is detected, the response detection unit 228 of the server 200 detects that a "touch action" has been performed by a user, and when upward acceleration is detected in a state where a touch is detected, the response detection unit 228 of the server 200 detects that a "hug" has been performed. Body contact when a user lifts the body 104 can also be detected, and a hug can also be detected by a load acting on the wheel 102 decreasing.

[0097] In this way, the reaction recognition unit 228 of the server 200 recognizes various types of reactions of a user toward the robot 100. "Pleasant" or "unpleasant" and "positive" or "negative" are correlated with a portion of typical reaction actions among these various types of reaction actions. In general, almost all reaction actions that are pleasant actions are positive reactions, and almost all reaction actions that are unpleasant actions are negative reactions. Pleasant and unpleasant actions relate to familiarity, and positive and negative reactions influence the action selection of the robot 100.

[0098] A series of recognition processes, including detecting, analyzing, and determining, may be performed by the recognition unit 212 of the server 200 alone, or may be performed by the recognition unit 156 of the robot 100 alone, or the two may perform the recognition processes while sharing roles.

[0099] The familiarity management unit 220 of the server 200 changes the familiarity toward a user according to a response action detected by the detection unit 156. Essentially, the familiarity toward a user who performs a pleasant action increases, while the familiarity toward a user who performs an unpleasant action decreases.

[0100] The recognition unit 212 of the server 200 can determine whether a response is pleasant or unpleasant, and the map management unit 210 can change the z-value of the point where the pleasant or unpleasant action was performed on an action map representing the "preference for a location." For example, if a pleasant action is performed in a living room, the map management unit 210 can determine a preferred point with a high probability in the living room. In this case, a positive feedback benefit is realized by the robot 100 favoring the living room and further favoring the living room because it is the recipient of a pleasant action in the living room.

[0101] The human recognition unit 214 of the server 200 recognizes a moving object from various types of data obtained from the external sensor 114 or the internal sensor 128 and extracts features (physical and behavioral characteristics) therefrom. Furthermore, the human recognition unit 214 performs cluster analysis on multiple moving objects based on these features. Not only a human, but also a pet such as a dog or a cat can be a target of analysis as a moving object.

[0102] The robot 100 periodically performs image acquisition, and the human recognition unit 214 detects a moving object from the images and extracts features of the moving object. When a moving object is detected, physical and behavioral features are also extracted from the odor sensor, the built-in high-directivity microphone, the temperature sensor, and the like. For example, when a moving object appears in an image, various features are extracted, such as having a beard, being active early in the morning, wearing red clothes, smelling perfume, having a loud voice, wearing glasses, wearing a skirt, having white hair, being tall, being plump, having a tan, or being on a sofa.

[0103] If a moving object (user) with a beard is often active in the early morning (wakes up early) and rarely wears red clothes, a first profile will be created, which is a cluster (user) who wakes up early, has a beard, and rarely wears red clothes. Meanwhile, if a moving object with glasses often wears a skirt, but the moving object has no beard, a second profile will be created, which is a cluster (user) who wears glasses and a skirt, but definitely does not have a beard.

[0104] Although the above example is a simple example, the first profile corresponding to a father and the second profile corresponding to a mother are formed according to the method described above, and the robot 100 recognizes that there are at least two users (owners) in this house.

[0105] Note that robot 100 does not need to recognize that the first profile is the "father." In all cases, it is sufficient that robot 100 can recognize a figure that is "a cluster with a beard, often wakes up early, and rarely wears red clothes."

[0106] It is assumed that the robot 100 newly recognizes a moving object (user) in a state where this type of cluster analysis is completed.

[0107] At this time, the human recognition unit 214 of the server 200 extracts features from measurement information of an image or the like obtained from the robot 100, and determines which cluster a moving object near the robot 100 corresponds to using deep learning (a multi-layer neural network). For example, if a moving object with a beard is detected, the probability that the moving object is the father is high. If the moving object is active early in the morning, it is even more certain that the moving object corresponds to the father. Meanwhile, if a moving object wearing glasses is detected, there is a possibility that the moving object is the mother.If the moving object has a beard, the moving object is neither the mother nor the father, so the person recognition unit 214 determines that the moving object is a new person who has not been cluster analyzed.

[0108] The formation of a cluster by feature extraction (cluster analysis) and the application to cluster-accompanying feature extraction (deep learning) can be carried out simultaneously.

[0109] The familiarity with a moving object (user) changes according to the way the user handles the robot 100.

[0110] The familiarity management unit 220 increases or decreases the familiarity with each clustered user. The familiarity changes mainly according to (1) sensing (visual recognition), (2) physical contact, and (3) speech. 1. Recording

[0111] When a toddler is detected in an image filmed by robot 100, the toddler is "visually recognized" by robot 100. More specifically, it is determined that visual recognition occurs when, using deep learning based on the characteristic information obtained from the filmed image and other characteristic information obtained from the odor sensor and the like during filming, it is determined that the features of the detected moving object match the cluster (profile) of the toddler. When it is determined that visual recognition occurs, familiarity management unit 220 increases the familiarity of the toddler. The more often a user is recognized, the more familiarity tends to increase.

[0112] According to this type of control method, the robot 100 emulates animal-like behavior to develop a sense of proximity to a frequently encountered person.

[0113] Familiarity is not limited to simple recognition, but can also increase with "eye contact." The recognition unit 156 of the robot 100 can recognize a facial image of an opposing user, detect a gaze direction from the facial image, and recognize that "eye contact has been established" when the gaze direction is directed toward the robot 100 for a predetermined time or longer. 2. Physical contact

[0114] When the robot 100 visually recognizes a user and detects a touch (body contact) from the user, it is determined that the user has shown interest in the robot 100, and familiarity increases. For example, when the robot 100 is touched by the mother, the familiarity management unit 220 increases the mother's familiarity. The robot 100 can detect a touch of the robot 100 by covering an outer shell with a piezoelectric fabric. A touch can also be detected by the user's body temperature, which is detected by the temperature sensor. When the robot 100 detects a hug, familiarity can be significantly increased based on the strong affection shown toward the robot 100.

[0115] However, if the robot 100 detects a violent action such as being kicked, hit, or grabbing the horn 112, the familiarity management unit 220 reduces the familiarity. For example, if the robot 100 is thrown by the toddler, the familiarity management unit 220 significantly reduces the familiarity with the toddler.

[0116] According to this kind of control method, the robot 100 emulates an animal-like behavior to develop a sense of closeness to a person who gently touches the robot 100, but dislikes a violent person. 3. Speaking

[0117] Familiarity also changes when robot 100 detects speech directed at robot 100. For example, familiarity increases when robot 100 detects the name of robot 100 or an affectionate term within a predetermined volume range. Typical terminological patterns such as "you are cute," "you are funny," or "come here" can be registered in advance as affectionate terms, and whether a term is an affectionate term or not can be determined using speech recognition. Meanwhile, familiarity may be reduced if robot 100 is spoken to at a high volume that exceeds a normal volume range. For example, familiarity is reduced if robot 100 is scolded in a loud voice or if it is surprised.

[0118] Furthermore, familiarity may be reduced when an offensive term is directed at the robot 100. Typical terminological patterns such as "stop," "stay away," "go away," or "idiot" can be registered in advance as offensive terms, and whether a term is an offensive term or not can be determined using speech recognition.

[0119] The name of the robot 100 may be registered in advance by a user. Alternatively, the robot 100 may recognize a term that is used with a certain frequency among various terms addressed to the robot 100 as the name of the robot 100. In this case, commonly used terms such as "Hey" and "come here" may be removed from the candidates for name recognition.

[0120] According to the control method described above, the robot 100 sets a high familiarity level for a frequently encountered person, a person who frequently touches the robot 100, and a person who frequently talks to the robot 100. Meanwhile, the familiarity level decreases for a rarely encountered person, a person who infrequently touches the robot 100, a violent person, and a person who scolds in a loud voice. The robot 100 changes the familiarity level of each user based on various elements of the external environment information acquired by the sensors (visual, tactile, and acoustic).

[0121] The familiarity management unit 220 decreases the familiarity over time. For example, the familiarity management unit 220 may decrease the familiarity of all users by 1 every 10 minutes. If a user no longer engages with the robot 100, or in other words, if a user no longer treats the robot 100 in a friendly manner, the user cannot maintain a close relationship with the robot 100.

[0122] The actual robot 100 autonomously performs a complex action selection according to an action card. The robot 100 acts while influenced by multiple action cards based on various parameters, such as loneliness, boredom, and curiosity. When the effect of the action cards is removed, or in an internal state where the effect of the action cards is low, the robot 100 essentially attempts to approach a person with high familiarity and attempts to move away from a person with low familiarity.

[0123] The actions of the robot 100 are classified below according to familiarity. 1. A cluster with extremely high familiarity

[0124] The robot 100 strongly expresses a feeling of affection by approaching a user (hereinafter referred to as an approach action) and performing an affectionate gesture, which is defined in advance as a gesture indicating goodwill towards a person. 2. A cluster with comparatively high familiarity

[0125] The robot 100 performs only one approach action. 3. A cluster with comparatively low familiarity

[0126] The robot 100 does not perform any special action. 4. A cluster with particularly low familiarity

[0127] The robot 100 performs a retreat action.

[0128] According to the control method described so far, the robot 100 approaches the user when it finds a user with high familiarity, and conversely, moves away from the user when it finds a user with low familiarity. According to this type of control method, the robot 100 can express so-called "shyness" through behavior. Furthermore, when a visitor (a user A with low familiarity) appears, the robot 100 can move away from the visitor and toward a family member (a user B with high familiarity). In this case, user B can perceive that the robot 100 is shy and uncomfortable, and relies on user B. Through this type of behavioral expression, user B feels the joy of being chosen and relied upon, as well as a related feeling of affection.

[0129] Meanwhile, when user A, who is a visitor, frequently visits and talks to and touches the robot 100, the familiarity of the robot 100 with the user A gradually increases, and the robot 100 stops performing a shyness action (a withdrawal action) toward the user A. The user A may also feel affection toward the robot 100 by perceiving that the robot 100 has become accustomed to the user A.

[0130] The action selection described so far does not necessarily have to be executed continuously. For example, if an internal parameter indicating the curiosity of robot 100 is high, weight is given to an action map from which a location is determined where curiosity is satisfied, which is why there is a possibility that robot 100 will not select an action influenced by familiarity. If the outdoor sensor 114 installed in the hallway detects a user's return home, robot 100 can also perform an action to greet the user with maximum priority. Cooling search function

[0131] The robot 100 is configured to cool a heat-generating element such as the CPU by introducing outside air. The lower the outside air temperature, the easier it is to cool a heat-generating element, and the operating level of a device such as a fan that forcibly circulates air can be limited. Generally, as the operating level of a cooling device increases, the fan speed and the noise level increase. Cooling noise makes the robot 100, which emulates animal-like behavior characteristics, feel the presence of a "machine" and is preferably suppressed as much as possible. Therefore, the robot 100 in this embodiment includes a function of moving to a cool location to reduce the mechanical noise generated during cooling as much as possible.

[0132] Fig. 6 is a schematic view for describing a function of the coolness-seeking robot.

[0133] The area in which the robot 100 can move in this embodiment is the entire interior of a Fig. 6. The external sensor 114 is installed in various areas within the house. The robot 100 regularly measures the room temperature (the peripheral temperature) as it moves within the house. The temperature map management unit 226 compiles a temperature map showing a peripheral temperature distribution in the area where the robot 100 can move by correlating the room temperature and the position coordinates of the measurement point. The temperature map is information in which the room temperature and position coordinates are correlated, and a high-temperature area 170 and a low-temperature area 172 are identified by the temperature map. The high-temperature area 170 is an area where the room temperature is equal to or greater than a threshold value M1, and the high-temperature area 170 is an area where the room temperature is less than a threshold value M2 (M2 ≤ M1).

[0134] In this embodiment, the high temperature region 170 is a region of 25 degrees Celsius or higher, and the low temperature region 172 is a region of less than 15 degrees Celsius. In Fig. 6, several high temperature ranges 170a to 170d and several low temperature ranges 172a to 172g are recorded.

[0135] The high-temperature area 170 may be, for example, next to a window or near a heat-generating body, such as an electrical appliance. The low-temperature area 172 may be a location exposed to air from an air conditioning system, located behind an object, or the like.

[0136] The robot 100 exhibits behavioral characteristics of preferring the low-temperature region 172 and avoiding the high-temperature region 170. When the indoor temperature rises, the robot 100 searches for a point where the room temperature is lower than the current point in order to lower the temperature. Specifically, based on the temperature map, the robot 100 (server 200) specifies a cool point C, where the room temperature is presumably lower than the current point, as the movement target point.

[0137] The cool spot C is desirably a point located in the low-temperature area 172a where the room temperature is particularly low. The movement determination unit 234 of the server 200 may specify any point in the nearest low-temperature area 172 as the cool spot C, or may specify the point closest to the current point that has a predetermined room temperature or lower as the cool spot C. Alternatively, the movement determination unit 234 may specify the point on the temperature map with the lowest room temperature as the cool spot C. It is sufficient that the cool spot C is at least a point where the room temperature is lower or where the room temperature is assumed to be lower than the current point.

[0138] The movement determination unit 234 of the server 200 specifies several movement routes moving toward the cool point C. Two movement routes, which are a route R1 and a route R2, are in Fig. 6. Route R2 passes through the high-temperature area 170a, but route R1 does not pass through the high-temperature area 170. In this case, the movement determination unit 234 selects route R1, which has the lower temperature. This is done to limit an increase in the internal temperature as much as possible, even when the robot 100 is moving.

[0139] The movement determination unit 234 may select a movement route in which an average value of the room temperatures is low at each predetermined interval. If the room temperatures of each one meter in a three-meter movement route are 25 degrees, 22 degrees, and 27 degrees, the average value thereof, which is 25 degrees, may be specified as the room temperature (hereinafter referred to as "route temperature") of the movement route. Alternatively, the movement determination unit 234 may specify a maximum room temperature or a minimum room temperature in a movement route as the route temperature. If the route temperatures of multiple movement routes are the same, or if a difference between the route temperatures is equal to or smaller than a predetermined threshold, the movement determination unit 234 may select the shorter movement route.

[0140] The server 200 notifies the robot 100 of the movement target point (cool point C) and the route R1 for the movement target point, and the robot 100 moves toward the cool point C along the route R1.

[0141] Since the robot 100 includes a heat-generating body such as the processor 122, the temperature inside a frame of the robot 100 (the internal temperature) may rise above the room temperature. Although the robot 100 can dissipate heat by causing the built-in fan to rotate, in the same way as a conventional computer, the robot 100 can also lower the internal temperature by autonomously moving toward the cool point C. This is an advantage characteristic of an autonomously acting robot. Furthermore, according to this type of control method, animal-like behavioral characteristics of "disliking heat" can be expressed.

[0142] Fig. Figure 7 is a schematic view describing the temporal change of a temperature map.

[0143] The room temperature changes depending on time. Fig. 7 shows a temperature map 180 in three time ranges, which are a temperature map 180a from 6:00 a.m. to 9:00 a.m., a temperature map 180b from 9:00 a.m. to 12:00 p.m. and a temperature map 180c from 12:00 p.m. to 3:00 p.m.

[0144] The robot 100 regularly measures the room temperature. The temperature map management unit 226 records a measurement time and date that correlate with the room temperature and the measurement point. For example, if the room temperature is measured between 6:00 a.m. and 9:00 a.m., the temperature map management unit 226 updates the temperature map 180a. If the room temperature is measured multiple times at the same point within the same time range, the most recent room temperature may be recorded in the temperature map 180, or the average room temperature of the multiple measured room temperature may be recorded.

[0145] According to Fig. 7, in the temperature map 180a, the low temperature area 172 is larger than the high temperature area 170, but the high temperature area 170 increases as time progresses from the temperature map 180b to the temperature map 180c. According to the multiple temperature maps 180, information is obtained that the room temperature next to a south-facing window rises during the day, and it is always cool behind a sofa.

[0146] The robot system 300 can appropriately specify the cool point C by referring to the temperature map 180 according to the time range of a motion start point.

[0147] Fig. 8 is a flowchart showing a process in which the robot 100 moves toward the cool point C.

[0148] The internal sensor 128 regularly measures the internal temperature of the robot 100. The Fig. The process shown in Figure 8 is repeatedly executed at a time of measuring the indoor temperature. When an indoor temperature t is greater than a predetermined threshold T1 (Y in S10), the movement determination unit 234 searches for the cool spot C by referring to the temperature map 180 (S14).

[0149] Even if the indoor temperature t is equal to or less than the threshold T1 (N in S10), the movement determination unit 234 also searches for the cool point C (S14) if a change rate Δt of the indoor temperature is greater than a predetermined threshold T2 (Y in S12). The change rate of the indoor temperature can be defined as a rate of temperature rise in a predetermined period of time, for example, every five seconds. This is because at a high rate of temperature rise, it is predicted that cooling will soon be required, even if the indoor temperature t is low at that time.

[0150] The motion determination unit 234 sets the cool point C as the motion target point and determines a motion route to reach the cool point C (S16). Once the motion target point and the motion route are determined, the motion determination unit 138 of the robot 100 outputs a motion instruction to the motion drive unit 144 (S18).

[0151] When the internal temperature t is equal to or lower than the threshold T1 (N in S10) and the change rate Δt of the internal temperature is equal to or lower than the threshold T2 (N in S12), the robot 100 does not move to the cool point C.

[0152] The actual behavioral characteristics of the robot 100 are somewhat more complex. If the preferred point P1 with the large z-value is in the Fig. 4 exists, the robot 100 can select the favorite point P1 instead of the cool point C as a movement target point, even when the internal temperature t is high. When the internal temperature t is particularly high, the robot 100 can move toward the cool point C even if the favorite point P1 with the large z-value exists.

[0153] When a cool, dark place exists and the robot 100 has a trait of disliking dark places, a behavioral characteristic of avoiding dark places and a behavioral characteristic of seeking a cool place conflict. When the internal temperature is extremely high, an incentive to seek a cool place is stronger than an incentive to avoid dark places. A movement target point of the robot 100 is determined based on multiple action maps, including the temperature map 180, and various parameters such as internal temperature and familiarity.

[0154] Fig. 8 is a flowchart schematically showing the motion characteristics based only on the temperature map 180 of the multiple action maps.

[0155] So far, the robot 100 and the robot system 300 including the robot 100 have been described based on an embodiment.

[0156] The robot 100 performs an action selection that cannot be replicated using one or more action cards, is difficult to predict, and is animal-like. Using this type of method, the robot 100 emulates animal-like action selection.

[0157] An animal's actions are influenced by various factors, but room temperature is one of them. When the body temperature rises, an animal attempts to limit the rise in body temperature by moving to the shade or the like. When the internal temperature rises, especially the temperature near a heat-generating source such as the battery 118 or the processor 122, the robot 100 in this embodiment also moves autonomously in search of the cool point C. For this reason, animal-like behavioral characteristics can be expressed by the robot 100.

[0158] In addition, a failure such as a failure of the processor 122 or corruption of data in the memory (storage device 124) in the computer may occur when the temperature is high. When lowering the internal temperature using a cooling device such as a fan, the noise level increases as the operating level of the cooling function increases. Noise during cooling makes the robot 100, which emulates animal-like behavior characteristics, feel the presence of a "machine," and therefore the robot 100 may be in a bad mood. Meanwhile, the robot 100 has the characteristic of autonomous action. By moving to the cool point C of its own free will, the robot 100 can perform cooling without overly depending on a cooling device. This type of control method also contributes to energy saving by the robot 100.

[0159] Since the invention is not limited to the at least one embodiment or modified example described so far, components may be changed or implemented without departing from the scope of the disclosure. Various implementations may be formed by combining several of the components disclosed in the at least one embodiment or modified example described so far. Furthermore, some components may be removed from the entirety shown in the at least one embodiment or modified example described so far.

[0160] Although a description has been given assuming that the robot system 300 is configured from a robot 100, a server 200, and a plurality of external sensors 114, part of the functions of the robot 100 may be realized by the server 200, and part or all of the functions of the server 200 may be allocated to the robot 100. One server 200 may control a plurality of robots 100, or a plurality of servers 200 may control one or more of the robots 100 in cooperation.

[0161] A third device different from the robot 100 and the server 200 can manage part of the functions. A collection of the functions of the robot 100 and the Fig. The functions of the server 200 described in Figure 5 can also be comprehensively understood as a "robot." It is sufficient that a method for distributing the multiple functions required to implement the invention with respect to one or more hardware elements is determined, taking into account the processing capability of each hardware element, the specifications required for the robot system 300, and the like.

[0162] As described above, “the robot in the narrow sense” is the robot 100 excluding the server 200, but “the robot in the broad sense” is the robot system 300. It is believed that there is a possibility that many functions of the server 200 will be integrated into the robot 100 in the future.

[0163] In this embodiment, a description has been given assuming that the temperature map management unit 226 compiles and updates the temperature map 180 as the robot 100 measures the room temperature accordingly, but the temperature map 180 may be set in advance in the server 200 as "innate and unchangeable prior knowledge." A plurality of temperature maps 180 corresponding to a plurality of time ranges may be initially set in the map storage unit 216.

[0164] In the same way, one or more cool spots C may initially be pre-determined in the server 200 as "innate knowledge." For example, the cool spot C may be pre-determined as being behind a sofa or a bedroom. The robot 100 may perform an action assuming that the cool spot C determined in this way is "a point with a lower temperature than the current point."

[0165] The robot 100 can independently search for the cool point C without referring to the temperature map 180. For example, the robot 100 can measure the peripheral temperature using a thermal sensor and identify a point with the lowest peripheral temperature as the cool point C.

[0166] A description has been given assuming that the robot 100 includes a fan for lowering the internal temperature. The robot 100 can measure the operating level of the fan, specifically, the rotational speed and rotation time of the fan, and move to the cool point C when the operating level reaches a predetermined threshold or higher, or when the operating level is assumed to reach the predetermined threshold or higher. The robot 100 can assume that the operating level is increasing when the rate of change of the operating level of the fan is high, or can assume that the operating level is increasing when an operation with a high load is to be performed. When an operation with a high load is to be performed, it can be assumed that the internal temperature will rise.

[0167] Since the cooling mechanism 162 is not limited to a fan, it is sufficient for the mechanism to exhaust air inside the robot 100 and admit outside air.

[0168] When a user turns off the cooler when going out, the room temperature is assumed to be rising. Therefore, the robot 100 can assume that the indoor temperature is rising when a user goes out. If it has innate or experiential knowledge that the temperature of a room is highest at 2:00 PM, the robot 100 can move to the cool point C before 2:00 PM. In this way, the robot 100 can predict a time when the indoor temperature reaches a predetermined temperature or higher based not only on the rate of change of the indoor temperature but also on innate or experiential knowledge.

[0169] When a user's return home is detected by the external sensor 114 on a hot summer day, the robot 100 may approach the user. This is because the robot 100 may expect the user who has returned home to turn on the cooler. Furthermore, when it is hot, the robot 100 may prompt the user to turn on the cooler by performing a back-and-forth gesture between the user and the cooler, or the like.

[0170] A time range of the temperature map 180 can be defined not only based on a time of day, but also based on a date or a season. For example, the temperature map 180 from 9:00 a.m. to 12:00 p.m. on June 7 can be prepared separately from the temperature map 180 from 9:00 a.m. to 12:00 p.m. on June 6. The cool point C can be more easily specified by the temperature map 180 corresponding to different prepared time ranges. Multiple temperature maps 180 can be prepared not only according to time ranges but also according to weather conditions such as sunny, cloudy, or rainy. The server 200 can obtain weather information by connecting to a weather forecast website.

[0171] The robot 100 can move quickly past a high-temperature point and slowly past a low-temperature point on the movement route to the cool point C.

[0172] The server 200 estimates the route temperature of each movement route based on the temperature map 180. Furthermore, the server 200 selects a movement route with a lower route temperature with priority. "Priority selection" may mean selecting the movement route with a lower route temperature with a higher probability.

[0173] If there are multiple robots 100, a temperature map can be shared.

[0174] The external sensor 114 may include a temperature sensor. Furthermore, the temperature map management unit 226 may create and update the temperature map 180 according to the temperature information obtained from the plurality of external sensors 114.

[0175] Fig. 9 is a functional block diagram of the robot system 300 in a modified example.

[0176] In the robot system 300 of a first modified example, the communication unit 142 includes a cooling information unit 154. The cooling information unit 154 executes a "cooling report" to be described below with respect to another robot. The internal sensor 128 includes a body temperature detection unit 164. The body temperature detection unit 164 detects the temperature (body temperature) of a user or another robot. The body temperature detection unit 164 includes a non-contact temperature sensor such as a radiation thermometer or thermography, and a contact temperature sensor such as a thermistor, a resistance temperature detector, a thermocouple, or an IC temperature sensor. The temperature detection unit 152 can also perform the function of the body temperature detection unit 164.

[0177] The robot 100 has several action patterns (movements). Various movements are defined, such as pivoting the arm 106, approaching an owner while turning, and looking closely at an owner with the head tilted to the side.

[0178] The operation pattern storage unit 232 stores a "movement file" that defines the control details of a movement. Each movement is identified by a movement identifier. The movement file is also downloaded to the operation pattern storage unit 232 of the robot 100. Which movement should be executed can be determined by the server 200 or by the robot 100.

[0179] Many movements of the robot 100 are configured as compound movements that include multiple individual movements. For example, when the robot 100 approaches an owner, the movement can be expressed as a combination of a movement of the unit changing direction toward the owner, a movement of the unit approaching while raising an arm, a movement of the unit approaching while shaking the body, and a movement of the unit sitting while raising both arms. By combining these four types of movements, a movement of "approaching an owner, raising an arm along the way, and finally sitting down after shaking the body" is realized. A rotation angle, angular velocity, and the like of an actuator provided in the robot 100 are defined, which are correlated with a time axis in the motion file.Different movements are expressed by controlling each actuator along with the timing according to the motion file (actuator control information).

[0180] The changeover time for switching from a previous unit movement to a subsequent unit movement is called an "interval." It is sufficient to define an interval according to the time required for a change in the unit's movement or details of a movement. The length of an interval can be controlled.

[0181] Hereinafter, the settings involved in controlling an action of robot 100, such as which movement is selected and when, and the output control of each actuator in realizing a movement, are collectively referred to as "action characteristics." The action characteristics of robot 100 are defined by a movement selection algorithm, a movement selection probability, a movement file, and the like.

[0182] In addition to an action map, the robot 100 also has parameters that indicate the magnitude of various emotions or senses. For example, when a value of a loneliness emotion parameter increases, a weighting coefficient of an action map that evaluates places where the robot 100 feels comfortable is increased, and the value of this emotion parameter is decreased when the robot 100 reaches a destination point. Similarly, when a value of a parameter indicating a feeling of boredom increases, it is sufficient for a weighting coefficient of an action map that evaluates places where curiosity is satisfied to be increased.

[0183] When the recognition unit 156 detects a moving object such as a person or a pet, the body temperature detection unit 162 measures the body temperature of the moving object. When the body temperature of the moving object is a predetermined value or higher, for example, 37 degrees Celsius or higher, the motion determination unit 234 of the server 200 (or the motion determination unit 138 of the robot 100) sets a cool spot specified based on the temperature map 180 as the movement target point. Furthermore, the action determination unit 236 of the server 200 (or the action determination unit 140 of the robot 100) selects a movement (hereinafter referred to as a "cooling guide movement") that is pre-correlated with the moving object according to the conditions described above, and causes the action driving unit 146 to execute the movement.It is sufficient that the cool-guiding movement is a specific movement defined to guide the moving object to a cool spot, such as swinging the arm 106, heading toward the cool spot after aligning the body 104 toward the moving object, aligning the arm 106 toward the cool spot, moving toward the cool spot while looking back at the moving object, or staying at the cool spot. The cool-guiding movement is desirably a movement such that the moving object is interested and wants to approach. According to this type of control method, heatstroke is easily prevented by guiding a person or pet with rising body temperature to a cool spot.

[0184] Multiple types of cool-guiding motions can be defined according to user attributes. For example, if the user is a child, the action determination unit 150 can cause the robot 100 to perform a cool-guiding motion to escape to a cool spot after touching the user. The user (child) is guided to the cool spot by running after the robot 100. If the user is an adult, the action determination unit 150 can cause the robot 100 to perform a cool-guiding motion by touching the user with the arm 106, pointing the arm 106 toward a cool spot, or the like. The action determination unit 150 can select a cool-guiding motion based not only on a user's age but also on another characteristic (parameter) such as gender or familiarity.

[0185] The recognition unit 212 may manage a cooling guidance movement correlated with success or failure as history information. For example, when executing a cooling guidance movement M1, the recognition unit 212 determines that the cooling guidance movement M1 was successful when a user P1 moves to a cool spot, and updates a success rate of the user P1 and the cooling guidance movement M1. The operation determination unit 150 may select a cooling guidance movement by referring to the history information. Specifically, it is sufficient for the operation determination unit 150 to select a cooling guidance movement with a high success rate for the user P1 when it wants to guide the user P1 to a cool spot. If the guidance fails at this time, it is sufficient for the operation determination unit 150 to attempt guidance again by selecting the cooling guidance movement with the next higher success rate.

[0186] The body temperature detection unit 164 can periodically measure the body temperature of each moving object, and the communication unit 142 can notify the server 200 of the measured body temperature. Furthermore, the recognition unit 212 can record the average body temperature of each moving object by registering the measured body temperature in the individual data storage unit 218. When a difference between the measured body temperature of a moving object and the average body temperature of the moving object is a predetermined value or greater, the operation determination unit 222 of the server 200 (or the operation determination unit 150 of the robot 100) can set a cool spot as the movement target point and select a cool-guiding movement.Alternatively, the operation determination unit 150 may select a cooling guide movement when the peripheral temperature of a moving object is a predetermined temperature or higher.

[0187] If the detection unit 156 also detects another robot R1, the body temperature detection unit 164 measures the body temperature (radiation temperature) of the robot R1. The robot R1 in this case may be of the same type as the robot 100 or of a type that does not include a cool spot control function. The detection unit 212 of the server 200 (or the detection unit 156 of the robot 100) may detect a moving object as a "robot" that has a shape other than that of a person or a pet. Alternatively, the detection unit 212 (or the detection unit 156) may detect another robot by detecting an identification signal periodically transmitted from the robot R1 using a sensor (not shown).

[0188] The detection unit 156 is configured such that the cooling information unit 154 transmits cooling information, or more specifically, a signal specifying coordinates of a cool spot (hereinafter referred to as a "cool signal"), to the robot R1 when the body temperature (radiation temperature) of the robot R1, measured from the radiated heat, is a predetermined value or higher. Provided that the robot R1 has a function of receiving a cooling signal from the robot 100 and moving to a cool spot specified by the cooling signal, the robot R1 can move toward a cool spot in the same way as the robot 100 when the temperature is high, even if the robot R1 does not have a function of finding a cool spot by itself.

[0189] The average body temperature of the robot R1 is recorded, and if the difference between the measured body temperature and the average body temperature of the robot R1 is a predetermined threshold difference or greater, the cooling information unit 154 may transmit a cooling signal. Furthermore, the cooling information unit 154 may transmit a cooling signal if the peripheral temperature of the robot R1 is a predetermined temperature or higher.

[0190] When the peripheral temperature regularly measured by the temperature detection unit 152 during movement is a predetermined threshold temperature or higher, for example, 30 degrees Celsius or higher, the action determination unit 140 may set the movement speed to high. Alternatively, when the peripheral temperature of the current point shown in the temperature map 180 during movement is a predetermined threshold temperature or higher, the action determination unit 140 may also set the movement speed of the robot 100 to high. According to this type of control method, the time spent by the robot 100 in a high-temperature area can be reduced as much as possible even during movement, thereby easily limiting the internal temperature of the robot 100.For the same reason, the action determination unit 140 may reduce the movement speed when the peripheral temperature measured by the temperature detection unit 152 or the peripheral temperature of the current point shown in the temperature map 180 is a predetermined threshold temperature or lower, for example, 10 degrees or lower.

[0191] The recognition unit 212 (or the recognition unit 156) can recognize an image of an air conditioner installed indoors or a remote control thereof. The fact that the air conditioner or the remote control is installed can be registered in advance in the robot 100. When the internal temperature of the robot 100 exceeds a threshold value, or when a rate at which the internal temperature rises exceeds a threshold value, the operation determination unit 222 (or the operation determination unit 150) can set a point directly under the air conditioner or a point where the remote control of the air conditioner is present as the movement target point, or the robot 100 can move in the direction where the air conditioner and the remote control are present.If the detection unit 212 (or the detection unit 156) cannot detect a cool spot, the operation determination unit (or the operation determination unit 150) may set a point directly under the air conditioner or a point where the air conditioner's remote control is present as the movement target point. By moving near an air conditioner or the like, the robot 100 may make an unspoken appeal to a user, indicating that the robot 100 desires a lowering of the indoor temperature.

[0192] The detection unit 156 can identify a cooling device, such as an air conditioner, based on the detection result of the temperature detection unit 152. For example, if there is a point on the temperature map 180 with a significantly lower temperature than the surrounding environment, for example, a point that is one degree or more lower than the average indoor temperature, the detection unit 156 can determine that an air conditioner is located there. If the positions of multiple air conditioners are pre-registered in the server 200 or the robot 100, the detection unit 156 can search for an air conditioner near the current point based on the registered information.

[0193] The robot 100 is not limited to an air conditioner and can set a movement target point with another cooling device, such as a fan, as a target. Furthermore, since this type of request is meaningless when no user (person) is present, the robot 100 can move to a cooling device only when the detection unit 156 detects a user. The robot 100 can prompt a user to use a cooling device not only by moving to the cooling device, but also by having the operation determination unit 140 direct the arm 106 toward the cooling device.

[0194] When a fan is operating when the indoor temperature is high, or when a rise rate of the indoor temperature is high, the action determination unit 150 can set a movement target point in front of the fan. The recognition unit 212 (or the recognition unit 156) can determine a position of the fan and an operating state thereof by recognizing an image of the fan and a rotation of a blade thereof. According to this type of control method, not only can the indoor temperature be lowered by using an external cooling device such as a fan, but also an animal-like trait, namely, going near a fan because it is hot, can be expressed through the action. The communication unit 142 of the robot 100 can independently cause the operation of a cooling device by sending a control signal to an air conditioner or a fan.

[0195] When the internal temperature rises or when a rate of increase of the internal temperature is high, the cooling mechanism 162 is configured to increase the rotational speed of the fan. However, noise is generated when the rotational speed of the fan increases. The sound of the rotating fan not only causes noise but also makes a user feel that the robot 100 is a "machine," which is undesirable in view of the concept of animal-like characteristics. When the internal temperature is high or when the rotational speed of the fan is high, the movement determination unit 234 (or the movement determination unit 138) may determine a movement direction in a direction away from a user.When multiple users are present, or more specifically, when multiple users are detected from an image captured by the camera, the motion determination unit 234 (or the motion determination unit 138) may set a point spaced apart from each of the multiple users as the movement target point. According to this control method, the rotation speed of the fan may be increased after the robot 100 has moved sufficiently far away from a user, which may make it difficult for a user to hear the noise of the fan.

[0196] For the same reason, the cooling mechanism 162 is configured such that an upper limit value for the fan speed can be set when the robot 100 is within a predetermined range of a user or when a user is being filmed. Improvement can be made to quiet operation by providing an upper limit value for the fan speed when a user is nearby, and improvement can be made to cooling performance by removing the upper limit value for the fan speed when no user is nearby. A first upper limit value T1 and a second upper limit value T2 (>T1) can be registered in advance for the fan speed. The cooling mechanism 162 is configured such that the first upper limit value T1 can be set when the detection unit 156 detects a user, and the second upper limit value T2 can be set when no user is detected.Furthermore, the robot 100 may measure the distance between a user and the robot 100 using a conventional distance measuring device, and set the upper limit of the rotational speed higher as the distance increases. Furthermore, even when the robot 100 is located next to a user, the movement determination unit 234 may set a point away from the user as the movement target point, and then cancel the upper limit of the fan rotational speed when the internal temperature of the robot 100 rises to a particularly high level.

[0197] An indoor temperature distribution can be measured using a thermal sensor installed indoors, for example, attached to a ceiling. The temperature map management unit 226 can regularly receive temperature distribution information from the thermal sensor and update the temperature map 180.

[0198] When the internal temperature of the robot 100 is a predetermined value or higher, the movement determination unit 234 of the server 200 may set the high-temperature area 170 as a passage-prohibited area.

[0199] When the internal temperature is a predetermined value or higher, or when the internal temperature is a predetermined value or higher, the action determination unit 140 of the robot 100 may appeal to a user to cause a cooling device to operate by pointing the arm 106 toward a cooling device such as an air conditioner or a fan. Alternatively, the movement determination unit 138 of the robot 100 may instruct the movement drive unit 144 to cause the robot 100 to approach a cooling device, or the action determination unit 140 may cause the robot 100 to alternately perform a movement of orienting the body 104 toward a user and a movement of orienting the body 104 toward a cooling device.Since it is not limited to a cooling device, the same type of movement can be performed with a refrigerator as the target, or attention can be drawn by tapping a window with the arm 106.

[0200] As described above, the temperature map 180 can be shared by multiple robots 100. For example, assume that the temperature map 180 is formed based on a temperature measurement function of a robot 100A. At this time, the temperature map management unit 226 of the server 200 can grant access to the temperature map 180 to another robot 100B. Furthermore, the operation determination unit 222 of the server 200 can control both the robot 100A and the robot 100B based on the same temperature map 180. According to this type of control method, the robot 100B can select an action based on the temperature distribution in the same way as the robot 100A, even if the robot 100B does not have a function for forming the temperature map 180.

[0201] When the internal temperature is a predetermined value or higher, or when the indoor temperature is a predetermined value or higher, the robot 100 may perform various movements to indicate that "it is hot." For example, when the robot 100 is wearing clothes, the action determination unit 140 of the robot 100 may cause the robot 100 to perform a movement with the desire to take off the clothes, for example, a movement of hitting a button of the clothes with the arm 106. In addition, the action determination unit 140 may cause the robot 100 to perform a movement of fanning air to the face by moving the arm 106 like a fan, or the movement determination unit 138 may cause the robot 100 to perform an action expressing "lethargy caused by heat" by performing a movement of swaying while moving.

[0202] The cooling mechanism 162 is configured such that the fan speed can be increased when no person (user) is detected in the environment, thereby reducing the internal temperature to a predetermined value, for example, 15 degrees or lower, and the fan speed or operating time can be restricted when a person is detected. Instead of a fan, a heat source such as the processor 122 can also be cooled with a quiet Peltier element. The cooling mechanism 162 is configured such that a heat source can be cooled with a Peltier element instead of a fan or in addition to a fan when the fan speed reaches a predetermined upper limit.Similarly, the cooling mechanism 162 may be configured to cool a heat source using a Peltier element when a user is present in the environment, and to cool a heat source using a fan when there is no user present in the environment, that is, when no person appears in an image captured by the integrated camera. According to this type of control method, cooling performance can be improved and quietness can be increased.

[0203] A cool spot may be set at a location not exposed to direct sunlight, such as in the shade or next to a door. The detection unit 156 of the robot 100 may detect a "dark spot" where the illumination is at a predetermined level or less using a light sensor, and the operation determination unit 150 may set the "dark spot" as a cool spot. Furthermore, the detection unit 156 of the robot 100 may detect a "windy spot" from a cooling device or from outside using a wind direction sensor. Furthermore, the operation determination unit 150 may set the "windy spot" as a cool spot.

[0204] Additionally, when a fan is rotated or a Peltier element is activated, the action determination unit 140 of the robot 100 may cause the robot 100 to perform an action indicating that cooling is in progress. For example, the action determination unit 140 may cause the robot 100 to perform a prespecified characteristic action, such as slowly raising and lowering the arm 106 in a sitting state.

[0205] As described above, the action characteristics of the robot 100 change based on various external and internal factors. For example, when a user with high familiarity exists, a momentum of wanting to approach the user is generated. Meanwhile, when the internal temperature is high, a momentum of wanting to move away from a user and drive a fan to thereby lower the internal temperature is generated, or a momentum of wanting to move toward a cool spot is generated. The action determination unit 222 (or the action determination unit 150 of the robot 100) can compare familiarity and internal temperature, and, for example, give priority to an approaching action based on familiarity when the internal temperature is 30 degrees or less, and give priority to moving toward a cool spot when the internal temperature exceeds 30 degrees.In this way, the action determination unit 222 can change the action characteristics of the robot 100 by selecting various eigendynamics according to a strength thereof. SUMMARY

[0206] A robot 100 sets a cool point C as a movement target point and specifies the route coordinates R1 and R2 for reaching the cool point C. The robot 100 moves to the cool point C via the route R1, which has a lower temperature. The robot 100 searches for the cool point C by referring to a temperature map showing a temperature distribution in an area where the robot 100 can move. Furthermore, the robot 100 creates and updates the temperature map by measuring the peripheral temperature using a thermometer integrated in the robot 100. DRAWINGS

[0207] [ Fig. 2] 200 SERVERS[ Fig. 4] 118 BATTERY 120 DRIVE MECHANISM 122 PROCESSOR 124 STORAGE DEVICE 126 COMMUNICATOR 128 INTERNAL SENSOR[ Fig. 5] 100 ROBOTS 114 EXTERNAL SENSOR 120 DRIVE MECHANISM 128 INTERNAL SENSOR 136 DATA PROCESSING UNIT 138 MOTION DETERMINATION UNIT 140 UNIT OF ACTION 142 COMMUNICATION UNIT 144 MOTION DRIVE UNIT 146 ACTION DRIVE UNIT 148 DATA STORAGE UNIT 150 PROCESS DETERMINATION UNIT 152 TEMPERATURE SENSING UNIT 156 RECOGNITION UNIT 160 OPERATION PATTERN STORAGE UNIT 162 COOLING MECHANISM 200 SERVERS 202 DATA PROCESSING UNIT 204 COMMUNICATION UNIT 206 DATA STORAGE UNIT 208 POSITION MANAGEMENT UNIT 210 CARD MANAGEMENT UNIT 212 DETECTION UNIT 214 Personnel Recognition Unit 216 CARD STORAGE UNIT 218 STORAGE UNIT FOR INDIVIDUAL DATA 220 FAMILIARITY MANAGEMENT UNIT 222 PROCESS DETERMINATION UNIT 226 TEMPERATURE CARD MANAGEMENT UNIT 228 REACTION DETECTION UNIT 230 Expression Recognition Unit 232 OPERATION PATTERN STORAGE UNIT 234 MOTION DETERMINATION UNIT 236 UNIT OF ACTION[ Fig. 8] BEGINNING S14 SEARCH COOL POINT S16 COURSE SELECTION S18 MOVEMENT INSTRUCTION BACK[ Fig. 9] 100 ROBOTS 114 EXTERNAL SENSOR 120 DRIVE MECHANISM 128 INTERNAL SENSOR 136 DATA PROCESSING UNIT 138 MOTION DETERMINATION UNIT 140 UNIT OF ACTION 142 COMMUNICATION UNIT 144 MOTION DRIVE UNIT 146 ACTION DRIVE UNIT 148 DATA STORAGE UNIT 150 PROCESS DETERMINATION UNIT 152 TEMPERATURE SENSING UNIT 154 COOLING INFORMATION UNIT 156 RECOGNITION UNIT 160 OPERATION PATTERN STORAGE UNIT 162 COOLING MECHANISM 164 BODY TEMPERATURE RECORDING UNIT 200 SERVERS 202 DATA PROCESSING UNIT 204 COMMUNICATION UNIT 206 DATA STORAGE UNIT 208 POSITION MANAGEMENT UNIT 210 CARD MANAGEMENT UNIT 212 DETECTION UNIT 214 Personnel Recognition Unit 216 CARD STORAGE UNIT 218 STORAGE UNIT FOR INDIVIDUAL DATA 220 FAMILIARITY MANAGEMENT UNIT 222 PROCESS DETERMINATION UNIT 226 TEMPERATURE CARD MANAGEMENT UNIT 228 REACTION DETECTION UNIT 230 Expression Recognition Unit 232 OPERATION PATTERN STORAGE UNIT 234 MOTION DETERMINATION UNIT 236 UNIT OF ACTION

Claims

[1] Autonomously acting robot (100), comprising: a storage device (124, 148); a processor (122, 136) connected to the storage device (124, 148), the processor (122, 136) configured to execute the instructions for: determining a cool point (C) as a movement target point, wherein the cool point (C) is a location having a lower ambient temperature than an ambient temperature at a current location of the autonomously acting robot (100); selecting a route (R1) from a plurality of routes (R1, R2) from the current location to the movement destination point, wherein selecting the route (R1) comprises selecting a route (R1) that is assumed to have a lowest average temperature of the plurality of routes (R1, R2); specifying a direction of movement along the selected route (R1); and a drive mechanism (120) configured to perform the specified movement. [2] The autonomously acting robot (100) of claim 1, further comprising a thermal sensor (128) configured to sense an ambient temperature around the autonomously acting robot (100), wherein the processor (122) is configured to determine the cool spot (C) based on the sensed ambient temperature. [3] The autonomously acting robot (100) according to claim 1, wherein the processor (122) is configured to determine the cool spot (C) by referring to a temperature map showing an ambient temperature distribution in an area in which the robot (100) can move. [4] Autonomously acting robot (100) according to claim 3, further comprising: a thermal sensor (128) configured to sense an ambient temperature around the autonomously acting robot (100), wherein the processor (122) is configured to determine the temperature map based on the sensed ambient temperature. [5] The autonomously acting robot (100) of claim 4, wherein the processor (122) is configured to generate the temperature map in association with a time of day at which the ambient temperature is sensed. [6] The autonomously acting robot (100) according to claim 1, wherein the processor (122) is configured to determine the cool point (C) as a movement target point in response to an internal temperature of the autonomously acting robot (100) being a predetermined value or higher. [7] The autonomously acting robot (100) of claim 1, wherein the processor (122) is configured to determine the cool point (C) as a movement target point in response to an internal temperature of the autonomously acting robot (100) being predicted to reach or exceed a predetermined value. [8] The autonomously acting robot (100) according to claim 1, wherein the processor (122) is configured to determine the cool point (C) as a movement target point in response to an operation level of a cooling function in an interior of the autonomously acting robot (100) reaching or exceeding a predetermined value. [9] The autonomously acting robot (100) of claim 1, wherein the processor (122) is configured to determine the cool point (C) as a movement target point in response to an operating level of a cooling function in an interior of the autonomously acting robot (100) being predicted to reach or exceed a predetermined value. [10] The autonomously acting robot (100) of claim 8, further comprising a cooling mechanism (162) configured to adjust a speed of a fan based on an internal temperature of the autonomously acting robot (100), wherein the processor (122) is configured to specify the direction of movement in a direction away from a user in response to the speed of the fan reaching or exceeding a predetermined speed. [11] The autonomously acting robot (100) of claim 8, further comprising a cooling mechanism (162) configured to adjust a speed of a fan based on an internal temperature of the autonomously acting robot (100), wherein the processor (122) is configured to specify the direction of movement in a direction away from a user in response to the speed of the fan being predicted to reach or exceed a predetermined speed. [12] The autonomously acting robot (100) of claim 10, wherein the cooling mechanism (162) is configured to set an upper limit of the rotational speed of the fan that is lower in response to the user being within a predetermined distance from the autonomously acting robot. [13] The autonomously acting robot (100) according to claim 1, further comprising a body temperature detection unit (164) configured to detect a body temperature of a moving object, wherein the processor (122) is configured to instruct the drive mechanism (120) to perform a predetermined movement to guide the moving object to the cool point (C) in response to the body temperature of the moving object being a predetermined value or higher. [14] Autonomously acting robot (100) according to claim 1, further comprising: a body temperature detection unit (164) configured to detect a temperature of a second robot (R1); and a cooling information unit (154), wherein the processor (122) is configured to instruct the cooling information unit (154) to send the cool spot (C) to the second robot (R1) in response to the detected temperature of the second robot (R1) being a predetermined value or higher. [15] The autonomously acting robot (100) of claim 1, wherein the processor (122) is configured to increase a movement speed of the drive mechanism (120) in response to the autonomously acting robot (100) traversing an area having a temperature higher than a predetermined threshold temperature. [16] The autonomously acting robot (100) of claim 1, further comprising a transmitter, wherein the processor (122) is configured to instruct the transmitter to send a signal to an external cooling device to adjust a direction of the external cooling device in response to a failure to determine the cool spot (C).

Images