Device, action determination method and calculator system

Abstract

To realize cooperative autonomous control by considering emotions of people or other devices.SOLUTION: A device holds an internal model that predicts a state of a future environment, acquires observation data indicating the state of the environment observed by the device, uses the observation data to generate first state information, uses the first state information to generate second state information of an intelligent agent existing in a periphery, determines a self-emotion of the device on the basis of the first state information, determines the self-emotion of the intelligent agent on the basis of the second state information, determines a social emotion on the basis of the self-emotions of the device and intelligent agent, generates future first state information and future second state information by using the internal model, and determines an action on the basis of the future first state information, future second state information, and social emotion.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a behavior control technique for an autonomously controlled device. [Background technology]

[0002] Autonomous control technology is being used in various industries, such as transportation, manufacturing, and nursing care. Devices equipped with autonomous control technology act to achieve a given purpose according to the environment.

[0003] Devices that interact with people are required to be controlled in a way that takes into account the mental state of people in the environment.

[0004] For example, Patent Document 1 discloses a robot device having a function of switching between a behavior selection criterion that takes into account the state of the robot itself and a behavior selection criterion that takes into account the state of others, depending on the situation. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2005-199402 A Summary of the Invention [Problem to be solved by the invention]

[0006] In the technology described in Patent Document 1, each action selection criterion is independent, and therefore it is not possible to realize cooperative autonomous control that takes into account the feelings of a person or other devices.

[0007] The present invention aims to realize cooperative autonomous control that takes into account the emotions of people or other devices. [Means for solving the problem]

[0008] A representative example of the invention disclosed in the present application is as follows: That is, an autonomously controlled device that exists in a space where an intelligent agent that acts based on a mental state exists, the device includes a processor, a storage device connected to the processor, and an interface connected to the processor, holds an internal model that predicts a future state of the environment based on the state of the environment, social emotions that an individual has toward other individuals, and an action, and acquires observation data indicating the state of the environment observed by the device via the interface, generates first state information using the observation data, the first state information including information about the intelligent agent that exists around the device, as the state of the environment grasped by the device, and generates second state information using the first state information, the second state information including information about the device and other intelligent agents that exist around the intelligent agent, as the state of the environment grasped by the intelligent agent. and executing a behavior decision process including: a process of determining a self-emotion of the device based on the first state information; a process of determining a self-emotion of the intelligent agent based on the second state information; a process of determining a social emotion of the device toward the intelligent agent and a social emotion of the intelligent agent toward the device or another intelligent agent based on the self-emotion of the device and the self-emotion of the intelligent agent; a process of generating first predicted state information, which is a prediction of the first state information in the future, and second predicted state information, which is a prediction of the second state information in the future, using the internal model; and a process of determining an action to be taken by the device based on the first predicted state information, the second predicted state information, the social emotion of the device toward the intelligent agent, and the social emotion of the intelligent agent toward the device or another intelligent agent. Effect of the Invention

[0009] According to the present invention, it is possible to realize cooperative autonomous control that takes into account the emotions of a person or other devices. Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0010] [Figure 1] FIG. 2 is a diagram illustrating a hardware configuration of an autonomous device according to a first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a space in which the autonomous device of the first embodiment exists. [Diagram 3] FIG. 2 is a diagram illustrating an example of a functional configuration of an autonomous device according to a first embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of an internal model held by the autonomous device of the first embodiment. [Figure 5A] FIG. 4 is a diagram illustrating an example of an observation data database according to the first embodiment. [Figure 5B] FIG. 4 is a diagram illustrating an example of an observation data database according to the first embodiment. [Figure 6A] FIG. 4 is a diagram illustrating an example of an emotion data database according to the first embodiment. [Figure 6B] FIG. 4 is a diagram illustrating an example of an emotion data database according to the first embodiment. [Figure 7] 1 is a flowchart illustrating an example of a process executed by an autonomous device according to a first embodiment. [Figure 8A] FIG. 4 is a diagram illustrating an example of an emotion determination rule held by the autonomous device of the first embodiment. [Figure 8B] FIG. 4 is a diagram illustrating an example of an emotion determination rule held by the autonomous device of the first embodiment. [Figure 9] FIG. 4 is a diagram illustrating an example of a weight update rule held by the autonomous device of the first embodiment. [Figure 10A] FIG. 4 is a diagram showing an example of a display of a processing result output by the autonomous device of the first embodiment. [Figure 10B] FIG. 4 is a diagram showing an example of a display of a processing result output by the autonomous device of the first embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of a functional configuration of an autonomous device according to a second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. However, the present invention is not to be interpreted as being limited to the description of the embodiment shown below. It will be easily understood by those skilled in the art that the specific configuration can be changed without departing from the concept or purpose of the present invention.

[0012] In the configurations of the invention described below, the same or similar configurations or functions are given the same reference numerals, and duplicated explanations are omitted.

[0013] In this specification, the terms "first," "second," "third," and the like are used to identify components and do not necessarily limit the number or order.

[0014] In order to facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in the drawings, etc. may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not limited to the position, size, shape, range, etc. disclosed in the drawings, etc. EXAMPLES

[0015] Fig. 1 is a diagram illustrating a hardware configuration of the autonomous device of the first embodiment. Fig. 2 is a diagram illustrating an example of a space in which the autonomous device of the first embodiment exists.

[0016] The autonomous device 100 is a robot, a car, or the like, and behaves based on inputs obtained from the environment and the emotions of people and other autonomous devices present in the environment. The autonomous device 100 moves within a space 200, for example, and transports luggage. In the space 200, autonomous devices 201 and 204 and people 202 and 203 exist. The autonomous device 100 behaves in consideration of the emotions of the autonomous devices 201 and 204 and the people 202 and 203.

[0017] In this specification, an individual (a human or an autonomous device 100) that performs an action based on a mental state is referred to as an intelligent agent. An intelligent agent that the autonomous device 100 is aware of is referred to as an aware intelligent agent.

[0018] The emotions of an intelligent agent include self-emotions within itself and social emotions that an intelligent agent (an individual with emotions) has toward other intelligent agents. Self-emotions are, for example, fear, joy, etc. Social emotions are, for example, sympathy, trust, resentment, etc.

[0019] The autonomous device 100 includes a processor 101, a memory 102, an auxiliary storage device 103, a network interface 104, an observation device 105, a control device group 106, an input device 107, and an output device 108. Note that the hardware configuration of the autonomous device 100 is merely an example and is not limited thereto. For example, the autonomous device 100 does not need to include the input device 107 and the output device 108.

[0020] The processor 101 executes a program stored in the memory 102. The processor 101 executes processing according to the program, thereby operating as a functional unit (module) that realizes a specific function. In the following explanation, when a process is explained with a functional unit as the subject, it indicates that the processor 101 is executing a program that realizes the functional unit.

[0021] The memory 102 stores the programs executed by the processor 101 and information executed by the programs. The memory 102 is also used as a work area. The auxiliary storage device 103 is a large-capacity storage device that permanently stores data. The auxiliary storage device 103 is, for example, a hard disk drive (HDD) and a solid state drive (SSD). The programs and information stored in the memory 102 may be stored in the auxiliary storage device 103. In this case, the processor 101 reads the programs and information from the auxiliary storage device 103 and loads them into the memory 102.

[0022] The network interface 104 communicates with external devices via a network. The observation device 105 is a camera, a LiDAR, a microphone, a millimeter wave radar, an acceleration sensor, etc., and acquires observation data 350 (see FIG. 3) including information for grasping the state of the environment.

[0023] The control device group 106 includes motors, speech devices, etc. The input device 107 includes a keyboard, mouse, touch panel, etc., and receives input from the user. The output device 108 includes a display, speaker, etc., and outputs information to the user.

[0024] Fig. 3 is a diagram illustrating an example of a functional configuration of the autonomous device 100 according to the first embodiment. Fig. 4 is a diagram illustrating an example of an internal model held by the autonomous device 100 according to the first embodiment.

[0025] The autonomous device 100 includes an intelligent agent detection / tracking unit 301, a state information generation unit 302, a free energy calculation unit 303, an emotion estimation unit 304, a behavior decision unit 305, and a learning unit 306. The autonomous device 100 also holds an observation data database 311, an internal model database 312, and an emotion data database 313.

[0026] The autonomous device 100 acquires observation data 350 via the observation device 105. The observation data 350 is input to an intelligent agent detection / tracking unit 301 and a free energy calculation unit 303. In addition, the autonomous device 100 accepts behavior goal information 351 via the network interface 104 or the input device 107.

[0027] The observation data 350 is, for example, image and point cloud data, etc. The action goal information 351 is information for controlling the selection of an action.

[0028] The observation data database 311 is a database for managing the observation data 350 and state information that represents the state of the environment. The internal model database 312 is a database for managing the internal model in the free energy principle. The emotion data database 313 is a database for managing the self-emotions and social emotions of the intelligent agent.

[0029] The intelligent agent detection / tracking unit 301 uses the observation data 350 to detect an intelligent agent present around the autonomous device 100 and also tracks the intelligent agent. The intelligent agent detection / tracking unit 301 generates the coordinates and speed of the detected intelligent agent as self-state information that represents the state of the environment grasped by the autonomous device 100. The intelligent agent detection / tracking unit 301 stores the observation data 350 and state information in the observation data database 311, and outputs the observation data 350 and processing results to the state information generation unit 302.

[0030] The state information generating unit 302 uses the observation data 350 and the state information to generate state information (other state information) that represents the state of the environment grasped by the cognitive intelligent agent. For example, when coordinates and speed are obtained as state information, the other state information can be generated by subjecting the state information to coordinate conversion. When the observation data 350 is an image, an image (other observation data) acquired by the intelligent agent can be generated using existing technology such as Novel View Synthesis, and other state information can be generated from the other observation data. The present invention is not limited to the algorithm for generating other state information.

[0031] In the following, when there is no need to distinguish between self status information and other status information, they will be referred to as status information.

[0032] The free energy calculation unit 303 calculates the free energy for determining the self-emotion and the social-emotion of the intelligent agent (the autonomous device 100 and the cognitive intelligent agent). The free energy calculation unit 303 also calculates the expected free energy for determining the behavior of the intelligent agent (the autonomous device 100 and the cognitive intelligent agent).

[0033] The free energy is determined by the internal model Q(s,o) that is determined by the generative model P(s,o) of the latent state s of the environment and the observed data o, and the parameters θ that approximate the generative model of the latent variables s of the environment. θ It is defined by the formula (1) using (s) and the observation data o. D KL[ ] is the Kullback-Leibler distance and E[ ] is the expectation value.

[0034]

number

[0035] Here, the internal model of the first embodiment will be described with reference to FIG. 4. FIG. 4 is an example of the internal model, and is an example of configuring an encoder and a decoder of a latent state by a neural network. The internal model receives the self state, the intelligent agent state, the social emotion, and the behavior as inputs, and outputs a predicted self state and a predicted intelligent agent state. The social emotion is input as binary data, a one-hot code, or a combination of a value calculated from the free energy of the self intelligent agent and a value calculated from the free energy of the other intelligent agent. The self state is the state of the self intelligent agent included in the state information, and the intelligent agent state is the state of the other intelligent agent included in the state information. The encoder has a probability sampling method (Reparameterization Trick) used in a variational autoencoder (VAE) and samples the latent state probabilistically. That is, the internal model is a model that probabilistically outputs future state information from current state information. The internal model may have a parameter that controls learning, such as β-VAE.

[0036] The expected free energy is defined by Equation (2). π represents an action sequence, and s and o represent the latent state transition sequence and observed data sequence when π is executed. C is a probability distribution that represents the agent's preference for the observed data.

[0037]

number

[0038] In active inference, the expected free energy for each possible behavioral hypothesis is calculated, and the probability distribution of behavior is determined, for example, using the softmax function of equation (3) so that the smaller the expected free energy, the higher the probability of selecting that behavioral hypothesis.

[0039]

number

[0040] The free energy calculation unit 303 calculates the free energy of the autonomous device 100 using the self state information and the internal model. The free energy calculation unit 303 calculates the free energy of the intelligent agent using the other state information and the internal model. The internal model of the intelligent agent is required to calculate the free energy of the intelligent agent. In the first embodiment, the internal model of the autonomous device 100 is substituted for the internal model of the intelligent agent based on the simulation theory that people empathize with others by simulating the sensations of others using their own bodies.

[0041] The free energy calculation unit 303 generates future own state information by inputting the own state information to the internal model, and uses the future own state information to calculate the expected free energy of the autonomous device 100. The free energy calculation unit 303 generates future other state information by inputting other state information to the internal model, and calculates the expected free energy of the cognitive intelligent agent using the future other state information.

[0042] The emotion estimation unit 304 estimates the autonomous device 100's own emotion, the autonomous device 100's social emotion toward the intelligent agent, the intelligent agent's own emotion, and the intelligent agent's social emotion toward the autonomous device 100 or other intelligent agents based on the free energies of the autonomous device 100 and the cognitive intelligent agents. The estimation method will be described later.

[0043] The behavior decision unit 305 decides the behavior to be performed by the autonomous device 100 based on the expected free energies of the autonomous device 100 and the cognitive intelligent agent, equation (3), and the behavior goal information 351. The learning unit 306 learns the internal model.

[0044] Regarding each functional unit of the autonomous device 100, multiple functional units may be integrated into one functional unit, or one functional unit may be divided into multiple functional units.

[0045] 5A and 5B are diagrams illustrating an example of the observation data database 311 according to the embodiment 1. In the observation data database 311, a table 500 and a table 510 are stored.

[0046] Table 500 is a table that stores self-state information generated by the intelligent agent detection / tracking unit 301. Table 500 includes entries each consisting of an intelligent agent ID 501, a timestamp 502, a position 503, and a speed 504. One entry exists for a pair of an intelligent agent and a timestamp. One entry represents the state of an intelligent agent recognized by the autonomous device 100 at a certain time.

[0047] The intelligent agent ID 501 is a field for storing identification information of a detected intelligent agent. The state of the autonomous device 100 itself is also managed in the table 500. The timestamp 502 is a field for storing a timestamp included in the observation data 350. The position 503 and the speed 504 are a group of fields for storing values ​​calculated from the observation data 350 and representing the state grasped by the autonomous device 100.

[0048] Table 510 is a table for storing other state information generated by the state information generating unit 302. Table 510 includes entries each consisting of an intelligent agent ID 511, a timestamp 512, a position 513, and a speed 514. One entry exists for a pair of an intelligent agent and a timestamp. One entry represents other state information of one cognitive intelligent agent.

[0049] The intelligent agent ID 511 is a field for storing identification information of the detected intelligent agent. Note that there is no entry for the autonomous device 100 in the table 510. The timestamp 512 is a field for storing a timestamp. The timestamp 512 stores the timestamp of the observation data 350 of the state information used to generate the other state information. The position 513 and the speed 514 are a group of fields for storing values ​​representing the state grasped by the cognitive intelligent agent.

[0050] 6A and 6B are diagrams showing an example of emotion data database 313 in Example 1. In emotion data database 313, a table 600 and a table 610 are stored.

[0051] Table 600 is a table for managing own emotions. Table 600 includes entries each consisting of an intelligent agent ID 601, a timestamp 602, and own emotions 603. One entry exists for each pair of an intelligent agent and a timestamp.

[0052] The intelligent agent ID 601 is a field for storing identification information of an intelligent agent. The time stamp 602 is a field for storing a time stamp of state information used to calculate the own emotion. The own emotion 603 is a field for storing the own emotion of the intelligent agent.

[0053] The table 610 is a table for managing social emotions. There is a table 610 for each intelligent agent. The table 610 shown in FIG. 6 is a table for managing social emotions for the cognitive intelligent agent of the autonomous device 100.

[0054] Table 610 includes entries consisting of a timestamp 611, an intelligent agent ID 612, a location 613, a speed 614, and a social sentiment 615. There is one entry for each pair of an intelligent agent and a timestamp.

[0055] The timestamp 611 is a field for storing the timestamp of the state information used to calculate the expected free energy. When multiple observation data 350 are input as time-series data, the timestamp of the first or last observation data 350 of the time-series data is stored in the timestamp 611. The intelligent agent ID 612 is a field for storing the identification information of the intelligent agent. The position 613 and the speed 614 are a group of fields for storing predicted state information calculated using an internal model. The social emotion 615 is a field for storing the social emotion of the intelligent agent toward other intelligent agents.

[0056] Fig. 7 is a flowchart illustrating an example of a process executed by the autonomous device 100 according to the first embodiment. Fig. 8A and Fig. 8B are diagrams illustrating an example of an emotion determination rule held by the autonomous device 100 according to the first embodiment. Fig. 9 is a diagram illustrating an example of a weight update rule held by the autonomous device 100 according to the first embodiment.

[0057] The autonomous device 100 executes the process described below using the received observation data 350. The execution timing can be set arbitrarily. For example, the process may be executed when the observation data 350 is received, or when a predetermined number of observation data 350 is accumulated.

[0058] The intelligent agent detection / tracking unit 301 uses the observation data 350 to track and trace the intelligent agent (step S101).

[0059] The state information generating unit 302 generates other state information of the cognitive intelligent agent using the observation data 350 (step S102).

[0060] The free energy calculation unit 303 starts loop processing of the intelligent agent (step S103). Specifically, the free energy calculation unit 303 selects one intelligent agent from among the autonomous device 100 and the cognitive intelligent agent.

[0061] The free energy calculation unit 303 calculates the free energy using the state information of the intelligent agent (step S104).

[0062] The emotion estimation unit 304 determines the user's own emotion based on the free energy (step S105).

[0063] Here, we will explain the emotion determination rules managed by the autonomous device 100. The autonomous device 100 holds a table 800 and a table 810 as emotion determination rules.

[0064] The table 800 is a table for managing emotion determination rules for determining one's own emotion. The table 800 stores entries each consisting of an F'801 and a own emotion802.

[0065] F'801 is a field for storing conditions related to the value of the first derivative of the free energy. Own emotion 802 is a field for storing a value indicating the self emotion. In this embodiment, when the value of the first derivative of the free energy is positive, the self emotion is determined to be negative, and when the value of the first derivative of the free energy is negative, the self emotion is determined to be negative.

[0066] The judgment rules shown in the table 800 are merely examples and are not intended to be limiting. For example, the self-emotion may be defined for a combination of a condition related to the value of the first derivative of the free energy and a condition related to the value of the second derivative. This allows a variety of self-emotions to be set.

[0067] The table 810 is a table for managing emotion determination rules for determining social emotions. The table 810 stores entries each consisting of a self emotion (self) 811, a self emotion (other) 812, and a social emotion 813.

[0068] Self emotion (self) 811 is a field for storing the self emotion of the self intelligent agent. Self emotion (other) 812 is a field for storing the self emotion of another intelligent agent. Social emotion 813 is a field for storing a value indicating a social emotion.

[0069] The emotion estimation unit 304 uses the free energy of the intelligent agent and the table 800 to determine the intelligent agent's own emotion.

[0070] The free energy calculation unit 303 judges whether the processing has been completed for all intelligent agents (step S106).

[0071] If the process has not been completed for all intelligent agents, the free energy calculation unit 303 returns to step S103 and selects a new intelligent agent.

[0072] When the processing for all intelligent agents is completed, the feeling deduction unit 304 starts loop processing of the intelligent agents (step S107). Specifically, the feeling deduction unit 304 selects one intelligent agent from the autonomous device 100 and the cognitive intelligent agents. The selected intelligent agent is described as a target intelligent agent.

[0073] The emotion estimation unit 304 determines the social emotion of the target intelligent agent toward other intelligent agents (step S108). Specifically, the following process is executed.

[0074] (S108-1) The emotion estimation unit 304 selects one intelligent agent from among the intelligent agents excluding the target intelligent agent. The selected intelligent agent is referred to as a sub-target intelligent agent.

[0075] (S108-2) The emotion estimation unit 304 determines the social emotion of the target intelligent agent toward the sub-target intelligent agent based on the self-emotion of the target intelligent agent, the self-emotion of the sub-target intelligent agent, and the table 810.

[0076] (S108-3) The feeling deduction unit 304 judges whether or not the processing is completed for all intelligent agents except the target intelligent agent. If the processing is not completed for all intelligent agents except the target intelligent agent, the feeling deduction unit 304 returns to S108-1 and selects a new sub-target intelligent agent. If the processing is completed for all intelligent agents except the target intelligent agent, the feeling deduction unit 304 ends the processing of step S108.

[0077] The free energy calculation unit 303 generates future state information of the intelligent agent (step S109).

[0078] Specifically, the free energy calculation unit 303 acquires the latest social emotions from the table 610 of the intelligent agent stored in the emotion data database 313. The free energy calculation unit 303 acquires the action of the intelligent agent as a hypothesis from the action decision unit 305. The free energy calculation unit 303 calculates future state information by inputting the state information, social emotions, and actions of the intelligent agent into the internal model.

[0079] The free energy calculation unit 303 uses the future state information to calculate the expected free energy of the intelligent agent for a hypothetical action (step S110).

[0080] The feeling estimation unit 304 determines whether the processing has been completed for all intelligent agents (step S111).

[0081] If the processing has not been completed for all intelligent agents, the feeling estimation unit 304 returns to step S107 and selects a new intelligent agent.

[0082] When the processing is completed for all the intelligent agents, the behavior decision unit 305 decides the behavior of the autonomous device 100 based on the expected free energy and the social sentiment (step S112).

[0083] The behavior decision unit 305 of the first embodiment decides the behavior a based on the integrated expected free energy shown in formula (4) and the behavior goal information 351. Here, the index i represents a cognitive intelligent agent. G my is the expected free energy of the autonomous device 100, and G oth i is the expected free energy of a cognitive intelligent agent. W my and W oth i is the weight.

[0084]

number

[0085] The behavior determining unit 305 adjusts the weights according to the social sentiment. For example, the weights are adjusted based on a table 900 as shown in Fig. 9. The table 900 includes entries each consisting of a social sentiment 901 and a weight adjustment rule 902. The social sentiment 901 is a field for storing the social sentiment. The weight adjustment rule 902 is a field for storing the weight adjustment rule. It is assumed that the amount of change in the weights is set in advance.

[0086] The autonomous device 100 can output to the outside the results of the processing described in Fig. 7. Fig. 10A and Fig. 10B are diagrams showing examples of displaying the processing results output by the autonomous device 100 of the first embodiment.

[0087] A screen 1000 shown in FIG. 10A includes icon explanation columns 1001 and 1002 , and display columns 1003 , 1004 , and 1005 .

[0088] Icon explanation columns 1001 and 1002 are columns for displaying explanations of icons indicating self-emotions and social emotions.

[0089] The display field 1003 is a field that displays the results of various processes based on the autonomous device 100. The display field 1003 displays the autonomous device 100's own emotions, the autonomous device 100's social emotions toward the cognitive intelligent agent, and predicted results of the actions of the autonomous device 100 and the cognitive intelligent agent. The display field 1004 is a field that displays the results of various processes based on the cognitive intelligent agent specified by the user. The display field 1005 is a field that displays text that indicates the processing results.

[0090] In addition, the free energy, predicted observation data, weights, and the like may be displayed in display fields 1003 and 1004.

[0091] 10B, information about the free energy used to determine the emotion may be displayed on the screen 1000. For example, a graph showing the first and second derivatives of the free energy of each intelligent agent is displayed in a display field 1003. A graph showing the first derivative of the free energy of a specified intelligent agent on the horizontal axis and the first derivative of the free energy of other intelligent agents on the horizontal axis is displayed in a display field 1004.

[0092] The learning unit 306 executes learning processing of the internal model using state information, social emotions, etc. When the internal model is configured by an encoder and a decoder using a neural network, the learning processing is executed by inputting social emotions as conditions to the encoder and the decoder. Also, when the internal model has parameters that control learning, such as β-VAE, the parameters that control learning are set based on social emotions.

[0093] According to the first embodiment, the autonomous device 100 can control its behavior based on its own emotions and social emotions toward a cognitive intelligent agent. That is, it can realize cooperative autonomous control that takes into account the emotions of a person or other devices.

[0094] (Variation) A computer system that centrally manages the autonomous devices 100 may hold a functional configuration as shown in FIG. 3, acquire observation data 350 from the autonomous devices 100, calculate free energy, estimate emotions, and determine actions. EXAMPLES

[0095] The second embodiment is different in that the autonomous device 100 transmits and receives information to and from other autonomous devices via communication. The second embodiment will be described below, focusing on the differences from the first embodiment.

[0096] The hardware configuration of the autonomous device 100 of the second embodiment is the same as that of the first embodiment. The functional configuration of the autonomous device 100 of the second embodiment is partially different from that of the first embodiment. Fig. 11 is a diagram illustrating an example of the functional configuration of the autonomous device 100 of the second embodiment.

[0097] The autonomous device 100 of the second embodiment newly includes a cooperation unit 307. The other functional configurations are the same as those of the first embodiment. The cooperation unit 307 transmits and receives information between the autonomous devices 100. The cooperation unit 307 is connected to the intelligent agent detection / tracking unit 301, the state information generation unit 302, the free energy calculation unit 303, and the emotion estimation unit 304, and outputs information according to the type of information received from another autonomous device 100. In addition, the cooperation unit 307 acquires information from any one of the intelligent agent detection / tracking unit 301, the state information generation unit 302, the free energy calculation unit 303, and the emotion estimation unit 304, and transmits the information to the other autonomous device 100.

[0098] The data structure of each database held by the autonomous device 100 in the second embodiment is the same as that in the first embodiment.

[0099] The process executed by the autonomous device 100 of the second embodiment is the same as that of the first embodiment. However, in each processing step, if there is information received from another autonomous device 100, that information is used.

[0100] By transmitting and receiving information between the autonomous devices 100, it is possible to grasp an intelligent agent that has not been detected by another autonomous device 100 due to occlusion. In addition, it is expected to have the effect of reducing the amount of calculation of the autonomous device 100. In addition, more coordinated control between the autonomous devices 100 can be realized.

[0101] The present invention is not limited to the above-mentioned embodiment, but includes various modified examples. For example, the above-mentioned embodiment describes the configuration in detail to easily explain the present invention, and the present invention is not necessarily limited to the configuration including all the described configurations. Also, it is possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0102] In addition, each of the above configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. The present invention can also be realized by software program code that realizes the functions of the embodiments. In this case, a storage medium on which the program code is recorded is provided to a computer, and a processor included in the computer reads the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-mentioned embodiments, and the program code itself and the storage medium storing it constitute the present invention. Examples of storage media for supplying such program code include flexible disks, CD-ROMs, DVD-ROMs, hard disks, SSDs (Solid State Drives), optical disks, magneto-optical disks, CD-Rs, magnetic tapes, non-volatile memory cards, ROMs, etc.

[0103] Furthermore, the program code for realizing the functions described in this embodiment can be implemented in a wide range of program or script languages, such as assembler, C / C++, perl, Shell, PHP, Python, Java (registered trademark), etc.

[0104] Furthermore, the program code of the software that realizes the functions of the embodiments may be distributed over a network and stored in a storage means such as a computer's hard disk or memory, or in a storage medium such as a CD-RW or CD-R, and the processor of the computer may read out and execute the program code stored in the storage means or storage medium.

[0105] In the above-mentioned embodiment, the control lines and information lines are shown as those considered necessary for the explanation, and not all the control lines and information lines are shown in the product. All the components may be connected to each other. [Explanation of symbols]

[0106] 100 Autonomous Device 101 Processor 102 Memory 103 Auxiliary storage device 104 Network Interface 105 Observation Equipment 106 Control Units 107 Input Devices 108 Output Device 200 space 201 Autonomous Device 202 people 301 Intelligent Agent Detection / Tracking Division 302 Status Information Generation Unit 303 Free Energy Calculation Unit 304 Emotion estimation part 305 Action Decision-Making Department 306 Learning Department 307 Collaboration Department 311 Observation Data Database 312 Internal Model Database 313 Emotion Data Database 350 observation data 351 Action Target Information 1000 screens

Claims

1. A device that exists in a space where an intelligent agent that acts based on a mental state exists and performs autonomous control, A processor, a storage device connected to the processor, and an interface connected to the processor, maintaining an internal model that predicts future environmental conditions based on the state of the environment, the social emotions that the individual has towards other individuals, and the individual's behavior; acquiring, via the interface, observation data indicative of an environmental state observed by the device; a process of generating first state information including information about the intelligent agents present around the device as an environmental state grasped by the device using the observation data; a process of generating second state information, using the first state information, as an environmental state grasped by the intelligent agent, the second state information including information on the devices and other intelligent agents existing around the intelligent agent; determining a self-emotion of the device based on the first state information; determining a self-emotion of the intelligent agent based on the second state information; determining a social emotion of the device towards the intelligent agent and a social emotion of the intelligent agent towards the device or another intelligent agent based on a self-emotion of the device and a self-emotion of the intelligent agent; generating first predicted state information, which is a prediction of the first state information in the future, and second predicted state information, which is a prediction of the second state information in the future, using the internal model; and determining an action to be taken by the device based on the first predicted state information, the second predicted state information, social sentiment of the device toward the intelligent agent, and social sentiment of the intelligent agent toward the device or another intelligent agent.

2. 2. The apparatus of claim 1, Using the first state information, calculate a free energy of the device in a free energy principle, and determine a self-emotion of the device based on the free energy of the device; Using the second state information, calculate the free energy of the intelligent agent, and determine a self-emotion of the intelligent agent based on the free energy of the intelligent agent; An apparatus for determining a social emotion of the apparatus toward the intelligent agent and a social emotion of the intelligent agent toward the apparatus or another intelligent agent based on a combination of a self-emotion of the apparatus and a self-emotion of the intelligent agent.

3. 3. The apparatus of claim 2, Calculating an expected free energy of the device under a free energy principle using the first predicted state information; calculating the expected free energy of the intelligent agent using the second predicted state information; determining a behavior of the device based on a weighted sum of the expected free energy of the device and the expected free energy of the intelligent agent; An apparatus, characterized in that a weight to be multiplied to each of the expected free energy of the device and the expected free energy of the intelligent agent is determined based on a social sentiment of the device toward the intelligent agent and a social sentiment of the intelligent agent toward the device or another intelligent agent.

4. 2. The apparatus of claim 1, The intelligent agent includes another device that executes the behavior decision process; The device acquires a calculation result calculated in the action decision process from the other device, The device executes the action decision process using the first state information and the acquired calculation result.

5. 2. The apparatus of claim 1, executing a learning process for learning the internal model; A device, characterized in that the learning process uses social emotions of the device toward the intelligent agent and social emotions of the intelligent agent toward the device or other intelligent agents.

6. A behavior decision method executed by an autonomously controlled device that exists in a space in which an intelligent agent that performs behavior based on a mental state exists, comprising the steps of: The apparatus comprises: A processor, a storage device connected to the processor, and an interface connected to the processor, maintaining an internal model that predicts future environmental conditions based on the state of the environment, the social emotions that the individual has towards other individuals, and the individual's behavior; The behavior determination method includes: A first step of the processor acquiring, via the interface, observation data indicative of an environmental state observed by the device; a second step of the processor using the observation data to generate first state information including information about the intelligent agent present around the device as an environmental state grasped by the device; a third step of the processor generating, using the first state information, second state information including information on the devices and other intelligent agents present around the intelligent agent as an environmental state grasped by the intelligent agent; a fourth step of the processor determining a self-emotion of the device based on the first state information; a fifth step of the processor determining a self-emotion of the intelligent agent based on the second state information; a sixth step of the processor determining a social sentiment of the device towards the intelligent agent and a social sentiment of the intelligent agent towards the device or another intelligent agent based on a self-sentiment of the device and a self-sentiment of the intelligent agent; a seventh step of the processor using the internal model to generate first predicted state information that is a prediction of the first state information in the future and second predicted state information that is a prediction of the second state information in the future; and an eighth step in which the processor determines an action to be taken by the device based on the first predicted state information, the second predicted state information, the social sentiment of the device toward the intelligent agent, and the social sentiment of the intelligent agent toward the device or another intelligent agent.

7. The behavior determination method according to claim 6, The fourth step includes: The processor calculates a free energy of the device in a free energy principle using the first state information; and determining a self-emotion of the device based on the free energy of the device, The fifth step includes: the processor calculating the free energy of the intelligent agent using the second state information; and determining a self-emotion of the intelligent agent based on the free energy of the intelligent agent, The sixth step of the method for determining behavior includes a step in which the processor determines the social emotions of the device toward the intelligent agent and the social emotions of the intelligent agent toward the device or another intelligent agent based on a combination of the self-emotions of the device and the self-emotions of the intelligent agent.

8. The behavior determination method according to claim 7, The eighth step includes: The processor calculates an expected free energy of the device in a free energy principle using the first predicted state information; the processor calculating the expected free energy of the intelligent agent using the second predicted state information; and determining a behavior of the device based on a weighted sum of the expected free energy of the device and the expected free energy of the intelligent agent, A method for determining behavior, characterized in that weights to be multiplied to each of the expected free energy of the device and the expected free energy of the intelligent agent are determined based on the social sentiment of the device toward the intelligent agent, and the social sentiment of the intelligent agent toward the device or another intelligent agent.

9. A computer system that exists in a space where an intelligent agent that acts based on a mental state exists and is connected to a device that performs autonomous control, maintaining an internal model that predicts future environmental conditions based on the state of the environment, the social emotions that the individual has towards other individuals, and the individual's behavior; Obtaining observation data indicative of an environmental state observed by the device; generating first state information including information about the intelligent agent present around the device as an environmental state grasped by the device using the observation data; using the first state information to generate second state information including information on the devices and other intelligent agents present around the intelligent agent as an environmental state grasped by the intelligent agent; determining a self-emotion of the device based on the first state information; determining a self-emotion of the intelligent agent based on the second state information; determining a social sentiment of the device towards the intelligent agent and a social sentiment of the intelligent agent towards the device or another intelligent agent based on the self-sentiment of the device and the self-sentiment of the intelligent agent; generating first predicted state information, the first predicted state information being a prediction of future states of the first state information, and second predicted state information, the second predicted state information being a prediction of future states of the second state information, using the internal model; A computer system that determines an action to be taken by the device based on the first predicted state information, the second predicted state information, social sentiment of the device toward the intelligent agent, and social sentiment of the intelligent agent toward the device or another intelligent agent.

Images