System, method, and program for determining ai agent to be mounted on robot

Abstract

The present invention provides a system for determining an AI agent to be mounted on a robot from among a plurality of AI agents. The system comprises: an imparting means for imparting a score to each AI agent among the plurality of AI agents, the score of the AI agent representing a feature of the AI agent according to at least one index; and a determination means for determining an AI agent to be mounted on the robot on the basis of the score of each of the plurality of AI agents.

Description

System, method, and program for determining an AI agent to be mounted on a robot

[0001] The present invention relates to a system, method, and program for determining an AI agent to be mounted on a robot.

[0002] An artificial intelligence (AI) platform or AI agent that generates a response to a query from a user is known (for example, Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2023-120130

[0004] The inventor of the present invention considered that by mounting an AI agent with characteristics determined from a plurality of AI agents on a robot, a robot capable of appropriately interacting with humans can be constructed. Also, by causing the AI agent mounted on the robot to interact with a plurality of AI agents, it was considered possible to execute complex tasks or create new ideas beyond the limits of the capabilities or available information of a single AI agent.

[0005] An object of the present invention is to provide a system or the like for determining an AI agent to be mounted on a robot.

[0006] The present invention provides a system or the like characterized by determining an AI agent to be mounted on a robot by using scores assigned to each of a plurality of AI agents.

[0007] The present invention provides, for example, the following items: (Item 1) A system for determining which of a plurality of AI agents should be installed on a robot, the system comprising: an assigning means for assigning a score to each of the plurality of AI agents, wherein the score of the AI ​​agent represents the characteristics of the AI ​​agent with at least one index; and a determination means for determining which AI agent should be installed on the robot based on the respective scores of the plurality of AI agents. (Item 2) The system according to item 1, further comprising a receiving means for receiving signals from sensors of the robot, wherein the determination means adjusts the respective scores of the plurality of AI agents based on the signals, and determines which AI agent should be installed on the robot based on the adjusted scores. (Item 3) The system according to any one of the above items, wherein the determination means adjusts the respective scores of the plurality of AI agents based on the characteristics of the robot or the environment of the robot, and determines which AI agent should be installed on the robot based on the adjusted scores. (Item 4) The system according to any one of the above items, further comprising tuning means for tuning at least one of the plurality of AI agents based on the environment of the robot, wherein the assigning means assigns a score to the tuned AI agent. (Item 5) The system according to any one of the above items, further comprising tuning means for tuning an AI agent mounted on the robot based on the environment of the robot. (Item 6) The system according to any one of the above items, wherein the assigning means monitors the output from the plurality of AI agents, evaluates the monitored output with respect to at least one indicator, and assigns a score based on the result of the evaluation. (Item 7) The system according to any one of the above items, wherein the at least one indicator includes an ethical indicator.(Item 8) The system according to any one of the above items, wherein the ethical indicator includes at least one of the following: transparency, fairness, safety, and accountability, and the evaluation includes evaluating the monitored output with respect to at least one of the following: transparency, fairness, safety, and accountability. (Item 9) The system according to any one of the above items, wherein the at least one indicator further includes a technical capability indicator and / or a performance indicator. (Item 10) The system according to any one of the above items, wherein the robot is an industrial robot or an IoT device. (Item 11) A method for determining which of a plurality of AI agents should be installed on a robot, comprising assigning a score to each of the plurality of AI agents, wherein the score of the AI ​​agent represents a characteristic of the AI ​​agent in at least one indicator, and determining which AI agent should be installed on the robot based on the respective scores of the plurality of AI agents. (Item 11A) The method according to Item 11, having the characteristics of any one of the above items. (Item 12) A program for determining which of a plurality of AI agents should be installed on a robot, the program being executed on a computer having a processor, the program causing the processor to perform a process that includes assigning a score to each of the plurality of AI agents, wherein the score of the AI ​​agent represents the characteristics of the AI ​​agent in at least one index, and determining which AI agent should be installed on the robot based on the respective scores of the plurality of AI agents. (Item 12A) The program according to Item 12, having the characteristics of any one of the above items. (Item 12B) A non-transient computer-readable storage medium for storing the program according to Item 12 or Item 12A. (Item 13) A system comprising the system according to any one of Items 1 to 10 and the robot.(Item 14) A system comprising the system described in any one of Items 1 to 10 and the AI ​​agent determined above.

[0008] The present invention provides a system for determining which AI agent should be installed on a robot. This makes it possible to construct a robot that can interact appropriately with humans. The AI ​​agent installed on the robot may, for example, be adapted to the environment surrounding the robot. Furthermore, the AI ​​agent installed on the robot can interact with multiple AI agents. This can lead to improvements in the field of computers, particularly in AI-related fields. Moreover, by supporting the determination of which AI agent should be installed on a robot, the present invention can also lead to improvements in the field of robotics.

[0009] A diagram showing an example of a flow for determining which AI agent to be installed on robot R. A diagram showing an example of the configuration of system 100 for determining which AI agent to be installed on robot R. A diagram showing an example of the specific configuration of system 100 for determining which AI agent to be installed on robot R. A diagram showing an example of the configuration of processor unit 120. A flowchart showing an example of processing (process 410) by system 100 for determining which AI agent to be installed on robot R. A diagram showing an example of data flow between system 100 and robot R.

[0010] In this specification, "AI agent" refers to an autonomously operating artificial intelligence (AI). An AI agent is designed to act as a substitute for a human (i.e., an agent) and to satisfy requirements such as autonomy, cooperativeness, security, privacy, and / or accountability. In particular, the AI ​​agent covered by the present invention is designed to satisfy the accountability requirement so that the AI ​​agent can be evaluated from an ethical standpoint. For example, to satisfy the requirement of autonomy, it is necessary to have the ability to act autonomously, such as setting goals, planning, executing, recognizing situations, and learning and adapting. To satisfy the requirement of cooperativeness, it is necessary to have the ability to work cooperatively with other agents, such as communicating, coordinating, negotiating, and building trust. To satisfy the requirement of security, it is necessary to have advanced security measures such as authentication and authorization, confidentiality, integrity, and availability. To satisfy the requirement of privacy, it is necessary to have privacy protection measures such as protection of personal information, anonymity, and transparency. To satisfy the requirement of accountability, it is necessary to have the ability to be accountable, such as explaining decision-making, auditing actions, and holding people accountable. Other requirements that may be considered include energy efficiency, scalability, and reliability. The AI ​​agent can autonomously and appropriately respond to tasks given by the user.

[0011] In this specification, “robot” means a device capable of performing not only computational processing that a computer can normally perform, but also at least one physical action. A physical action is an action in the real world and may bring about a change in the robot itself, or a change in the environment surrounding the robot or its user. A robot may typically be an industrial robot, and the physical actions performed by such a robot may include, for example, actions such as walking, running, kicking, or jumping using the legs the robot may have, or actions such as manipulating an object using the arms the robot may have (e.g., grasping, releasing, or throwing an object). A robot may also be, for example, a robot that controls a house or building or a city or means of transportation (e.g., a car, airplane, train, elevator, etc.) (e.g., a robot that enables a smart home, a robot that enables a smart building, or a robot that enables a smart city or a smart car), or it may be an IoT device. The physical actions performed by such robots may include, for example, actions that bring about changes in the environment, and more specifically, actions that bring about changes in the internal or external environment of an object controlled by the robot (e.g., a house or building, a city or a means of transportation) by controlling equipment installed within that object. Such actions include, but are not limited to, actions that control air conditioning equipment to change the air environment (e.g., temperature, humidity, pressure, etc.), actions that control lighting equipment to change the lighting environment (e.g., illuminance, color, etc.), actions that control audio equipment to change the sound environment (e.g., background music, noise, noise cancellation, etc.), and actions that control fragrance equipment to change the odor environment.

[0012] In this specification, "task" refers to an operation or job that an AI agent should perform. Tasks may be given in the form of a question, in which case the term "task" may be used synonymously with "query." The AI ​​agent will output the result of performing the task.

[0013] In this specification, “interaction” between multiple AI agents means that the AI ​​agents engage in some form of exchange. An interaction may typically be a “transaction” involving the exchange of data or information. A transaction may involve the exchange of consideration.

[0014] Embodiments of the present invention will be described below with reference to the drawings.

[0015] 1. Mechanisms that enable interaction between AI agents and the real world In conventional systems using AI agents, when a user gives a task to the AI ​​agent, the AI ​​agent responds to the task using its database or large-scale language model (LLM). For example, when a user asks a question to an AI agent, the AI ​​agent searches for and / or generates an answer to the question and outputs that answer. For example, when a user gives a problem to an AI agent, the AI ​​agent searches for a solution to the problem and either solves the problem or outputs a solution to the problem.

[0016] The inventors of this invention have extended conventional mechanisms to develop a system that determines which AI agent is suitable for installation on a real-world robot from among multiple AI agents, and then installs the determined AI agent on the real-world robot. By installing the AI ​​agent on the robot, the AI ​​agent can perform actions or provide outputs appropriate to the environment surrounding the robot. Furthermore, by providing the AI ​​agent installed on the robot with user information, the AI ​​agent can perform actions or provide outputs that are appropriate to both the environment surrounding the robot and the user (hyper-personalized). Moreover, the AI ​​agent installed on the robot can self-learn or tune to adapt to the surrounding environment, enabling a flexible AI system.

[0017] The AI ​​agent to be installed on a robot may be determined based on the scores assigned to each of several AI agents. For example, the AI ​​agent with the highest score for aspects specific to the robot's characteristics may be selected as the AI ​​agent to be installed on the robot. Similarly, the AI ​​agent with the highest score for aspects specific to the robot's environment may be selected as the AI ​​agent to be installed on the robot.

[0018] Furthermore, the AI ​​agent mounted on the robot can connect to a network, enabling it to interconnect with multiple AI agents on the network. This allows it to interact with multiple AI agents and attempt to respond to tasks. For example, the output from a single AI agent may be limited to the knowledge of its own database or LLM, potentially preventing it from providing an appropriate response or limiting its ability to generate creative ideas. However, interaction between multiple AI agents can overcome these limitations.

[0019] It is preferable that multiple AI agents each possess a database containing different knowledge or information, or that they possess an LLM (Learning Life Model) that has learned different knowledge or information. It is more preferable that at least one of the multiple AI agents possesses a database containing user-specific knowledge or information, or that it possesses an LLM that has learned user-specific knowledge or information.

[0020] By interconnecting and coordinating multiple AI agents with different access to information (e.g., different areas of expertise) and / or different available resources, it becomes possible to leverage the unique characteristics of each AI agent, enabling them to share complex tasks and generate new ideas. This could be useful in various fields, such as medical diagnosis, music composition, and social problem solving. For example, an AI agent within a specific company (e.g., with access to information within that company) can request a part of a problem to be solved by an AI agent within another company (e.g., with access to information within that other company), and through interaction, the entire problem can be solved. For example, an AI agent with access to specific medical information can solve medical-related problems from another AI agent that lacks medical knowledge. For example, AI agents from various fields could interact to create novel music or art unlike anything seen before.

[0021] For example, an AI agent installed in a robot receives a task from a user, passes a subtask to another AI agent to respond to that task, and that other AI agent, upon receiving the subtask, passes it to yet another AI agent to respond to that subtask, and so on, in an attempt to respond to the task.

[0022] For example, an AI agent installed in a robot, upon receiving a question from a user, will ask a second AI agent connected via the network a question to obtain the information necessary to generate an answer to that question. The second AI agent may be an AI agent installed in another robot. The second AI agent will then ask a third AI agent connected via the network a question to obtain the information necessary to generate an answer to the question it received. The third AI agent may also be an AI agent installed in another robot. This process continues sequentially until the information necessary to generate an answer to the question received from the user is obtained. In this way, the interaction of multiple AI agents generates an answer to the question received from the user.

[0023] For example, if a user requests that a system error occur in the robot itself, the AI ​​agent installed in the robot will ask a second AI agent a question to obtain the information necessary to resolve the system error. The second AI agent will then ask a third AI agent a question to obtain the information necessary to generate an answer to the question it received. This process continues sequentially until sufficient information is obtained to resolve the system error. In this way, the system error that occurred in the robot can be resolved through the interaction of multiple AI agents.

[0024] Figure 1A shows an example of a flow for installing an AI agent, selected from among several AI agents, into a real-world robot R. While Figure 1A shows six AI agents, the number of AI agents is not limited to these. Any number of AI agents can be involved.

[0025] In step S1, system 100 assigns a score to each of the multiple AI agents. The score assigned to each of the multiple AI agents is a score that represents the characteristics of the AI ​​agent using at least one metric. The at least one metric includes at least one of the following: a technical capability metric, a performance metric, or an ethical metric.

[0026] Technical capability metrics are indicators used to evaluate the technical capabilities of an AI agent. Technical capability metrics can be evaluated from the perspective of at least one of the following: processing power, learning ability, and algorithmic complexity.

[0027] Processing capability indicates the information processing speed or parallel computing capability of the system used to build the AI ​​agent. The faster the information processing speed or the higher the parallel computing capability, the higher the value of the technical capability index.

[0028] Learning ability indicates the ability of a system that builds an AI agent to acquire new knowledge or skills. The higher the learning ability, the more flexibly the robot R can respond to changes in its surrounding environment or changes in the opposing AI agent, and therefore the higher the value of the technical capability index.

[0029] Algorithmic complexity indicates the sophistication of the algorithms used by the system building the AI ​​agent. The more complex the algorithm, the more sophisticated the decision-making it enables, and therefore the higher the score of the technical capability metric.

[0030] Performance metrics are indicators that evaluate the past performance of an AI agent. Performance metrics can be evaluated from the perspective of at least one of the following: past transaction history, success rate, and client satisfaction.

[0031] Past transaction history shows the performance of the AI ​​agent in past transactions. The more extensive the track record, the more reliable the AI ​​agent is considered to be, and the higher the value of the performance indicator.

[0032] The success rate indicates the percentage of past trades in which the AI ​​agent was successful. A higher success rate indicates a higher trading ability of the AI ​​agent, resulting in a higher performance indicator value.

[0033] The "Party Satisfaction" metric indicates the satisfaction level of the other party in a transaction with the AI ​​agent. A higher satisfaction level indicates a higher perceived trading capability of the AI ​​agent, resulting in a higher performance indicator value.

[0034] Ethical indicators are metrics used to evaluate the ethical correctness of AI agents. These indicators may be evaluated from the perspective of at least one of the following: transparency, fairness, safety, and accountability.

[0035] Transparency indicates whether the basis for the judgment or decision-making process of an AI agent is clear. If the basis can be accurately presented when requested, transparency is high, and the ethical index score will be high. Even if the basis cannot be accurately presented, if the judgment process can be clearly shown, transparency can still be considered high.

[0036] Fairness indicates whether the output from the AI ​​agent contains neither bias nor discrimination. If it contains neither bias nor discrimination, it can be said to be fair to all parties, and the ethical index score will be high.

[0037] Security indicates the level of cybersecurity provided by the AI ​​agent. The higher the cybersecurity, the lower the risk of harm to trading partners or third parties, resulting in a higher ethical index score.

[0038] Accountability, also known as accountability, indicates whether it is possible to explain who is responsible for the output of an AI agent. If the responsibility can be clearly explained, the level of accountability is high, and the ethical index score will be high.

[0039] The score representing the characteristics of the AI ​​agent preferably represents the characteristics of the AI ​​agent from the standpoint of ethical indicators. The inventors of the present invention believe that even if an AI agent has high capabilities or a proven track record, the output from an ethically problematic AI agent may cause harm to a third party and / or may be inappropriate output based on an illegal source, and therefore consider ethical indicators to be particularly important for determining the suitability of an AI agent to be installed in robot R. Based on this idea, the system 100 of the present invention can ensure that the output from the AI ​​agent installed in robot R is ethically correct by utilizing a score that represents the characteristics of the AI ​​agent, at least from the standpoint of ethical indicators.

[0040] When scores are assigned to each of the plurality of AI agents in step S1, an AI agent to be mounted on the robot R is determined based on the scores.

[0041] For example, an AI agent having the highest score can be determined as the AI agent to be mounted on the robot R.

[0042] For example, an AI agent having the highest score for an index corresponding to an action performed by the robot R can be determined as the AI agent to be mounted on the robot R.

[0043] At this time, according to the characteristics of the robot R, the scores may be adjusted and then the AI agent with the highest score may be determined. Alternatively, according to the environment around the robot R, the scores may be adjusted and then the AI agent with the highest score may be determined. The environment around the robot R can be detected by, for example, sensors provided in the robot R.

[0044] As a specific example, for example, when the environment around the robot R requires a high processing capacity for the AI agent mounted on the robot R (for example, when the robot R is in an environment where it is necessary to quickly respond to tasks from many users), among the technical ability index, the performance index, and the ethics index, the technical ability index is weighted more. Therefore, among the plurality of AI agents, the scores can be adjusted so that an AI agent with a higher value of the technical ability index of the score is more likely to be determined as the AI agent to be mounted on the robot R. For example, a positive weighting is given to the technical ability index of the score of each AI agent, and no weighting or a negative weighting is given to the other indexes.

[0045] For example, if the score of the first AI agent is (technical capability index, performance index, ethical index) = (5, 3, 2), the score of the second AI agent is (technical capability index, performance index, ethical index) = (2, 5, 4), and the score of the third AI agent is (technical capability index, performance index, ethical index) = (2, 4, 4), the second AI agent would have the highest score. However, by giving more weight to the technical capability index, for example by doubling its weighting, the first AI agent, which has the highest adjusted score, will be determined to be the AI ​​agent to be installed in robot R.

[0046] As another concrete example, if the environment surrounding robot R requires the AI ​​agent installed on robot R to provide an ethically correct answer, then more emphasis will be placed on the ethical indicator among the technical capability indicator, performance indicator, and ethical indicator. Therefore, among multiple AI agents, the scores may be adjusted so that AI agents with higher ethical indicator scores are selected to be installed on robot R, and AI agents with low ethical indicator scores are not selected to be installed on robot R. For example, a threshold score for the ethical indicator of each AI agent's score could be set, and AI agents that do not meet this threshold score could be excluded from consideration.

[0047] For example, if the score of the first AI agent is (technical capability index, performance index, ethical index) = (5, 5, 2), the score of the second AI agent is (technical capability index, performance index, ethical index) = (2, 5, 4), and the score of the third AI agent is (technical capability index, performance index, ethical index) = (2, 4, 4), the first AI agent would have the highest score. However, by setting the threshold score for the ethical index to 3, the first AI agent is excluded, and the second AI agent, which has the next highest score, is selected as the AI ​​agent to be installed in robot R.

[0048] The AI agent determined in this way can be an AI agent suitable for the environment around the robot R, and can perform actions according to the environment and / or provide an output according to the environment. For example, when the robot R is a robot that realizes a smart home, the AI agent installed in the robot R can be an AI agent suitable for the environment of that room, and can perform actions according to the environment and / or provide an output according to the environment. For example, when the robot R is an IoT device, the AI agent installed in the robot R can be an AI agent suitable for the environment where the IoT device is installed.

[0049] The AI agent installed in the robot R can also obtain information about the user who uses the robot R (for example, the user's characteristics, status, preferences, etc.) from the user or other sources (for example, a personalized AI agent that has learned the user's information), and perform actions and / or provide an output that are suitable for the environment and also suitable for the user.

[0050] The AI agent installed in the robot R interacts with a plurality of AI agents on the network by connecting to the network. For example, in order to respond to a task given by the user of the robot R, the AI agent installed in the robot R interacts with a plurality of AI agents.

[0051] FIG. 1B shows an example of a flow for responding to a task given by a user.

[0052] In this example, the task described is to resolve a system error that occurred in robot R. Under the management of system 100 of the present invention, at least some of the multiple AI agents (A1, A2, A3, A4, A5, A6) interact to derive a solution to resolve the system error that occurred in robot R. In Figure 1B, six AI agents are shown, but the number of AI agents is not limited to this. Any number of AI agents can be involved. Each of the multiple AI agents can be interconnected so that it can interact with each of the other AI agents in the multiple AI agents. Each of the multiple AI agents can be interconnected in any manner.

[0053] If a system error occurs in robot R, the user assigns robot R's AI agent the task of resolving the system error. Robot R's AI agent will collect information in order to derive a solution to resolve the system error that occurred in robot R. First, robot R's AI agent identifies the information necessary to derive a solution to resolve the system error.

[0054] Next, the AI ​​agent of robot R decides from which AI agent to obtain the necessary information. At this time, referring to Figure 1A, the AI ​​agent from which to obtain the information can be determined according to the score assigned in the process described above. For example, the AI ​​agent with the highest score for a given indicator can be determined to be the AI ​​agent from which to obtain the information. Alternatively, for example, the AI ​​agent with the highest average score across multiple indicators can be determined to be the AI ​​agent from which to obtain the information.

[0055] In this example, the first AI agent A1 was determined to be the AI ​​agent from which information should be acquired.

[0056] In step S11, the AI ​​agent of robot R requests the necessary information from the first AI agent A1.

[0057] If the first AI agent A1 can obtain the requested information from its own knowledge or information or LLM, the first AI agent A1 can send that information back to the AI ​​agent of robot R. Alternatively, the first AI agent A1 may obtain at least a portion of the requested information from another AI agent. In this case, the first AI agent A1 decides from which AI agent it should obtain at least a portion of the requested information. Similar to the above, the AI ​​agent from which to obtain the information can be determined according to the assigned score. For example, the AI ​​agent with the highest score for a given indicator can be determined to be the AI ​​agent from which to obtain the information. Alternatively, for example, the AI ​​agent with the highest average score for multiple indicators can be determined to be the AI ​​agent from which to obtain the information.

[0058] In this example, the third AI agent A3 was determined to be the AI ​​agent from which information should be acquired.

[0059] In step S12, the first AI agent A1 requests the necessary information from the third AI agent A3.

[0060] As described above, the third AI agent A3 may obtain the requested information from its own knowledge or information or LLM, or it may obtain at least a portion of the requested information from another AI agent. When obtaining at least a portion of the requested information from another AI agent, the third AI agent A3 decides from which AI agent it should obtain at least a portion of the requested information. Similar to what was described above, the AI ​​agent from which to obtain the information can be determined according to the assigned score.

[0061] In this example, the fourth AI agent, A4, was determined to be the AI ​​agent from which information should be acquired.

[0062] In step S13, the third AI agent A3 requests the necessary information from the fourth AI agent A4.

[0063] In step S14, the fourth AI agent A4 provides the information acquired by the fourth AI agent A4 to the third AI agent A3.

[0064] In step S15, the third AI agent A3 provides the first AI agent A1 with the information it has acquired, along with the information provided in step S14.

[0065] In step S16, the first AI agent A1 provides the information acquired by the first AI agent A1, along with the information provided in step S15, to the AI ​​agent of robot R.

[0066] In this way, the AI ​​agent of robot R solves a given task (in this case, resolving a system error that occurred in robot R) based on information obtained from the first AI agent A1, the third AI agent A3, and the fourth AI agent A4, as well as the knowledge or information possessed by the AI ​​agent of robot R itself.

[0067] Furthermore, the AI ​​agent installed on robot R can be tuned to adapt to the environment surrounding robot R. This can be done, for example, under the management of system 100. For example, the AI ​​agent installed on robot R can be tuned so that the score assigned to it improves depending on the surrounding environment. This can be achieved, for example, through reinforcement learning of the AI ​​agent.

[0068] A tuned AI agent can provide more suitable output as the AI ​​agent for robot R.

[0069] The system 100 described above can be implemented by a system to determine which AI agent should be installed in the robot described later.

[0070] 2. Diagram 2 of the system configuration for determining which AI agent to be installed on the robot shows an example of the configuration of system 100 for determining which AI agent to be installed on the robot.

[0071] System 100 is connected to the database unit 200. System 100 is also connected to at least one user terminal device 300 via the network 500. Furthermore, System 100 is connected to at least one server device 400 via the network 500. Finally, System 100 is connected to at least one robot R via the network 500.

[0072] Although Figure 2 shows three user terminal devices 300, the number of user terminal devices 300 is not limited to these. Any number of user terminal devices 300 can be connected to the system 100 via the network 500. Similarly, although Figure 2 shows two server devices 400, the number of server devices 400 is not limited to these. Any number of server devices 400 can be connected to the system 100 via the network 500. Furthermore, although Figure 2 shows one robot R, the number of robot R is not limited to these. Any number of robot R can be connected to the system 100 via the network 500.

[0073] Here, the server device 400 is a device capable of implementing an AI agent. The server device 400 has its own database, possesses its own knowledge or information, and / or possesses its own LLM.

[0074] Network 500 can be any type of network. Network 500 may be, for example, the Internet or a LAN. Network 500 may be a wired network or a wireless network.

[0075] An example of system 100 is a computer (e.g., a server device) installed in a provider that manages at least one robot, but is not limited thereto. For example, system 100 may be a computer (e.g., a server device) installed in a provider that provides interactive response generation services. An example of user terminal device 300 is a computer (e.g., a terminal device or controller) that can be used by a user of robot R, but is not limited thereto. Here, the computer (server device or terminal device) can be any type of computer. For example, the terminal device can be any type of terminal device such as a smartphone, tablet, personal computer, smart glasses, or smartwatch.

[0076] The database unit 200 stores various types of information, at least those used to calculate a score representing the characteristics of the AI ​​agent.

[0077] Figure 3A shows an example of a specific configuration of system 100 for determining the AI ​​agent to be installed on the robot.

[0078] The system 100 comprises an interface unit 110, a processor unit 120, and a memory unit 130.

[0079] The interface unit 110 exchanges information with the outside of the system 100. The processor unit 120 of the system 100 can receive information from the outside of the system 100 via the interface unit 110 and can transmit information to the outside of the system 100. The interface unit 110 can exchange information in any format.

[0080] The interface unit 110 includes, for example, an input unit that enables information to be input to the system 100. The manner in which the input unit enables information to be input to the system 100 is not limited. For example, if the input unit is a receiver, the receiver may input information by receiving it from outside the system 100 via a network. Alternatively, if the input unit is a data reading device, it may input information by reading it from a storage medium connected to the system 100.

[0081] The interface unit 110 includes, for example, an output unit that enables information to be output from the system 100. The mode by which the output unit enables information to be output from the system 100 is not limited. For example, if the output unit is a transmitter, the information may be output by the transmitter transmitting information to the outside of the system 100 via a network. Alternatively, if the output unit is a data writing device, the information may be output by writing the information to a storage medium connected to the system 100.

[0082] System 100 can, for example, transmit information to and / or receive information from the database unit 200 via the interface unit 110. System 100 can, for example, transmit information to and / or receive information from the user terminal device 300 via the interface unit 110. System 100 can, for example, transmit information to and / or receive information from the server device 400 via the interface unit 110. System 100 can, for example, transmit information to and / or receive information from the robot R via the interface unit 110.

[0083] System 100 can, for example, receive information via the interface unit 110 to determine a score representing the characteristics of an AI agent. System 100 can, for example, transmit information via the interface unit 110 representing at least one AI agent that should interact with a given AI agent.

[0084] The processor unit 120 executes the processing of the system 100 and controls the operation of the entire system 100. The processor unit 120 reads the program stored in the memory unit 130 and executes the program. This makes it possible to make the system 100 function as a system that executes desired steps. The processor unit 120 may be implemented by a single processor or by multiple processors.

[0085] The memory unit 130 stores programs necessary for executing the system 100's processing, as well as data necessary for executing those programs. The memory unit 130 may also store a program (for example, a program that implements the processing shown in Figure 4, described later) that causes the processor unit 120 to perform processing to determine which AI agent should be mounted on the robot R. Here, it is not specified how the program is stored in the memory unit 130. For example, the program may be pre-installed in the memory unit 130. Alternatively, the program may be stored in a non-transient computer-readable storage medium and installed by reading the storage medium. Alternatively, the program may be installed in the memory unit 130 by being downloaded via a network. In this case, the type of network is not specified. The memory unit 130 can be implemented using any storage means.

[0086] The database unit 200 stores various types of information used to calculate a score representing the characteristics of the AI ​​agent.

[0087] Here, the characteristics of the AI ​​agent are defined by a score representing the characteristics of the AI ​​agent using at least one indicator. The at least one indicator includes at least one of the following: a technical competence indicator, a performance indicator, or an ethical indicator. The database unit 200 may store past outputs from the AI ​​agent, associated with each of the technical competence indicators, performance indicators, or ethical indicators.

[0088] In the examples shown in Figures 2 and 3A, the database unit 200 is located outside the system 100, but the present invention is not limited thereto. It is also possible to provide at least a portion of the database unit 200 inside the system 100. In this case, at least a portion of the database unit 200 may be implemented by the same storage means as the storage means implementing the memory unit 130, or by storage means different from the storage means implementing the memory unit 130. In any case, at least a portion of the database unit 200 is configured as a storage unit for the system 100. The configuration of the database unit 200 is not limited to a specific hardware configuration. For example, the database unit 200 may be composed of a single hardware component or of multiple hardware components. For example, the database unit 200 may be configured as an external hard disk drive for the system 100, as cloud storage connected via a network, or as a distributed network utilizing blockchain technology or the like.

[0089] For example, information about the AI ​​agent is stored in a database unit 200, which is configured as a decentralized network utilizing blockchain technology, and in this case, the information about the AI ​​agent becomes virtually impossible to tamper with. This ensures the reliability of the information about the AI ​​agent.

[0090] System 100 may be combined with robot R to form a single system. This system could, for example, be a system that provides a robot tuned to adapt to its surrounding environment. System 100 may also be combined with at least one of a plurality of AI agents to form a single system. For example, system 100 may form a single system with at least one AI agent determined by system 100 to be mounted on robot R. This system could, for example, be a system that enables the construction of a robot tuned to adapt to its surrounding environment, or a system that provides an AI agent for a robot.

[0091] Figure 3B shows an example of the configuration of the processor unit 120.

[0092] The processor unit 120 includes an assignment means 121 and a determination means 122.

[0093] The assignment means 121 is configured to assign a score to each of the multiple AI agents. The score is a score that represents the characteristics of the AI ​​agent using at least one metric. The at least one metric can be expressed from multiple perspectives. The score can be represented as a multidimensional vector.

[0094] The scoring means 121 can assign a score to each of the multiple AI agents by monitoring the output from the multiple AI agents and evaluating the monitored output with respect to at least one metric. To this end, the scoring means 121 can evaluate the monitored output with respect to at least one of a plurality of perspectives. The scoring means 121 can, for example, continuously monitor the output from the multiple AI agents. Alternatively, the scoring means 121 may, for example, automatically monitor the output from the multiple AI agents as it is generated. For example, continuous or automatic monitoring of the output from the multiple AI agents may be performed by a dedicated monitoring device, in which case the scoring means 121 can receive the output from the monitoring device. The dedicated monitoring device may be outside the system 100 or may be a component of the system 100.

[0095] At least one indicator includes at least one of the following: a technical capability indicator, a performance indicator, or an ethical indicator. Preferably, at least one indicator includes an ethical indicator. This is because representing the characteristics of the AI ​​agent from the perspective of at least an ethical indicator allows us to use whether the output from the AI ​​agent is ethically correct when making multiple AI agents interact, and ultimately helps to ensure that the output from robot R is ethically correct.

[0096] More preferably, at least one indicator includes a technical capability indicator or performance indicator and an ethical indicator. Even more preferably, at least one indicator includes a technical capability indicator, a performance indicator and an ethical indicator. This is because representing the characteristics of multiple AI agents with multiple indicators allows for a more accurate representation of each of the multiple AI agents, and enables more precise management of the AI ​​agents of robot R and the interactions between the multiple AI agents.

[0097] Ethical indicators are metrics used to evaluate the ethical correctness of AI agents. These indicators may be evaluated from the perspective of at least one of the following: transparency, fairness, safety, and accountability.

[0098] Transparency indicates whether the basis for the judgment or decision-making process of an AI agent is clear. If the basis can be accurately presented when requested, transparency is high, and the ethical index score will be high. Even if the basis cannot be accurately presented, if the judgment process can be clearly shown, transparency can still be considered high.

[0099] Fairness indicates whether the output from the AI ​​agent contains neither bias nor discrimination. If it contains neither bias nor discrimination, it can be said to be fair to all parties, and the ethical index score will be high.

[0100] Security indicates the level of cybersecurity provided by the AI ​​agent. The higher the cybersecurity, the lower the risk of harm to trading partners or third parties, resulting in a higher ethical index score.

[0101] Accountability, also known as accountability, indicates whether it is possible to explain who is responsible for the output of an AI agent. If the responsibility can be clearly explained, the level of accountability is high, and the ethical index score will be high.

[0102] The scoring means 121 monitors the output from each of the multiple AI agents, detects words, phrases, sentences, etc. that may be related to at least one of transparency, fairness, safety, and accountability, and assigns a score from the perspective of ethical indicators based on the detection results. For example, if many phrases related to affirming transparency are detected, the scoring means 121 can assign a high score from the perspective of ethical indicators. For example, if many phrases related to denying safety are detected, the scoring means 121 can assign a low score from the perspective of ethical indicators.

[0103] The scoring means 121 may assign scores in a rule-based manner or in a machine learning-based manner. When assigning scores in a machine learning-based manner, a trained model can be used that has learned the relationship between words, phrases, sentences, etc. that may be related to at least one of transparency, fairness, safety, and accountability, and scores from the perspective of ethical indicators. The scoring means 121 can, for example, derive and assign scores as multidimensional vectors using embedding. For example, the scoring means 121 can calculate a score by summarizing the output from an AI agent from the perspective of ethical indicators and vectorizing the generated summary using embedding. The generation of the summary can be carried out using generative AI. The generated summary can be expressed in natural language. Representing the score as a multidimensional vector is preferable because it makes it easier to evaluate the similarity with the scores of other AI agents.

[0104] Technical capability metrics are indicators used to evaluate the technical capabilities of an AI agent. Technical capability metrics can be evaluated from the perspective of at least one of the following: processing power, learning ability, and algorithmic complexity.

[0105] Processing capability indicates the information processing speed or parallel computing capability of the system used to build the AI ​​agent. The faster the information processing speed or the higher the parallel computing capability, the higher the value of the technical capability index.

[0106] Learning ability indicates the system that builds the AI ​​agent's ability to acquire new knowledge or skills. The higher the learning ability, the more flexibly it can adapt to changes in the other AI agent, and therefore the higher the value of the technical capability index.

[0107] Algorithmic complexity indicates the sophistication of the algorithms used by the system building the AI ​​agent. The more complex the algorithm, the more sophisticated the decision-making it enables, and therefore the higher the score of the technical capability metric.

[0108] The scoring means 121 can assign scores from the perspective of technical capability indicators based on the performance or configuration of the system implementing each of the multiple AI agents.

[0109] The scoring means 121 may assign scores in a rule-based manner or in a machine learning-based manner. When assigning scores in a machine learning-based manner, a trained model can be used that has learned the relationship between the performance or configuration of the system implementing the AI ​​agent and the score from the perspective of technical capability indicators. The scoring means 121 can, for example, derive and assign a score as a multidimensional vector using embedding. For example, the scoring means 121 can calculate a score by summarizing the output from the AI ​​agent from the perspective of technical capability indicators and vectorizing the generated summary using embedding.

[0110] Performance metrics are indicators that evaluate the past performance of an AI agent. Performance metrics can be evaluated from the perspective of at least one of the following: past transaction history, success rate, and client satisfaction.

[0111] Past transaction history shows the performance of the AI ​​agent in past transactions. The more extensive the track record, the more reliable the AI ​​agent is considered to be, and the higher the performance indicator value will be.

[0112] The success rate indicates the percentage of past trades in which the AI ​​agent was successful. A higher success rate indicates a higher trading ability of the AI ​​agent, resulting in a higher performance indicator value.

[0113] The "Party Satisfaction" metric indicates the satisfaction level of the other party in a transaction with the AI ​​agent. A higher satisfaction level indicates a higher perceived trading capability of the AI ​​agent, resulting in a higher performance indicator value.

[0114] The scoring means 121 can assign scores to each of the multiple AI agents from the perspective of performance indicators, based on the past trading results of each AI agent.

[0115] The scoring means 121 may assign scores in a rule-based manner or in a machine learning-based manner. When assigning scores in a machine learning-based manner, a trained model that has learned the relationship between past trading results and scores from the perspective of performance indicators can be used. The scoring means 121 can, for example, derive and assign scores as multidimensional vectors using embedding. For example, the scoring means 121 can calculate a score by summarizing the output from an AI agent from the perspective of performance indicators and vectorizing the generated summary using embedding.

[0116] In the example described above, it was explained that each of the multiple indicators is independent. That is, the score can be a multidimensional score having each of the multiple indicators as an axis. However, the present invention is not limited to this, and the multiple indicators may be correlated to form a one-dimensional score. The scoring means 121 may, for example, assign a score in a rule-based manner based on the value of each of the multiple indicators, or it may assign a score in a machine learning-based manner. When assigning a score in a machine learning-based manner, a trained model that has learned the relationship between the value of each of the multiple indicators and the score can be used. The scoring means 121 may, for example, use embedding to derive and assign a score as a multidimensional vector. For example, the scoring means 121 may summarize the output from an AI agent and calculate a score by vectorizing the generated summary using embedding. In the example described above, the output from the AI ​​agent was summarized and the generated summary was embedded. However, summarization is not mandatory. For example, the output from the AI ​​agent could be used directly for embedding, or features of the AI ​​agent could be extracted from the output and those extracted features could be used for embedding.

[0117] The scoring means 121 may also consider indicators related to the robot's characteristics or the robot's environment when assigning a score. Alternatively, the scoring means 121 may also consider signals from sensors the robot has when assigning a score.

[0118] The decision means 122 is configured to determine which AI agent should be installed on the robot based on the scores of each of the multiple AI agents.

[0119] For example, the decision means 122 can determine which of the multiple AI agents has the highest score to be installed on the robot R. In this case, the decision means 122 may identify the AI ​​agent with the highest score from a perspective specific to the characteristics of the robot, or it may identify the AI ​​agent with the highest score from a perspective specific to the environment surrounding the robot, or it may identify the AI ​​agent with the highest average score from multiple perspectives specific to both the characteristics of the robot and the environment surrounding the robot.

[0120] The determination means 122 may receive signals from sensors on the robot via receiving means that may be provided in the processor unit 120, and identify the characteristics of the robot and / or the environment around the robot based on those signals. The sensors on the robot may be, for example, sensors capable of measuring the physical environment around the robot, and include, but are not limited to, temperature sensors, humidity sensors, pressure sensors, wind speed sensors, magnetic sensors, position sensors, etc.

[0121] For example, the decision means 122 can adjust the score of each of the multiple AI agents based on signals from sensors on the robot and / or the robot's characteristics or environment, and determine which AI agent should be installed on the robot based on the adjusted scores.

[0122] For example, if the environment surrounding a robot, as detected by signals from the robot's sensors, requires a high level of processing power from the AI ​​agent installed on the robot, then more weight will be placed on the technical capability indicator among the technical capability indicator, performance indicator, and ethical indicator. Therefore, the scores may be adjusted so that among multiple AI agents, those with higher scores on the technical capability indicator are selected as the AI ​​agents to be installed on the robot. For example, the technical capability indicator of each AI agent's score may be given a positive weight, while the other indicators may be either unweighted or given a negative weight.

[0123] For example, if the characteristics of a robot or its surrounding environment require the AI ​​agent installed on the robot to provide ethically correct answers, then more emphasis will be placed on the ethical indicator among the technical capability indicator, performance indicator, and ethical indicator. Therefore, among multiple AI agents, the scores may be adjusted so that AI agents with higher ethical indicator scores are selected to be installed on the robot, and AI agents with low ethical indicator scores are not selected. For example, a threshold score for the ethical indicator of each AI agent's score could be set, and AI agents that do not meet this threshold score could be excluded from consideration.

[0124] In one embodiment, the determination means 122 can determine which AI agent to be installed on the robot R based on the similarity between the scores of a plurality of AI agents represented by multidimensional vectors and the ideal score that the AI ​​agent to be installed on the robot R should have. Since the multidimensional vector is represented as a point in a multidimensional space, an AI agent similar to the ideal score can be determined based on the relationship between the point corresponding to the score of the plurality of AI agents and the point corresponding to the ideal score. Similarity can be calculated using methods such as cosine similarity or Euclidean distance. For example, the AI ​​agent with the highest cosine similarity to the ideal score can be determined as the AI ​​agent to be installed on the robot. In other words, the AI ​​agent corresponding to the point closest to the point in the multidimensional space representing the ideal score can be determined as the AI ​​agent to be installed on the robot.

[0125] In one embodiment, the determination means 122 vectorizes the robot's characteristics and / or the environment surrounding the robot using embedding, and can determine which AI agent should be installed on the robot based on the similarity between the vector representing the robot's characteristics and / or the environment surrounding the robot and the score of each of a plurality of AI agents represented by multidimensional vectors. Similarity can be calculated using methods such as cosine similarity or Euclidean distance. For example, the AI ​​agent with the highest score in cosine similarity with the vector representing the robot's characteristics and / or the environment surrounding the robot may be determined as the AI ​​agent to be installed on the robot. In other words, the AI ​​agent corresponding to the point closest to the point in the multidimensional space representing the robot's characteristics and / or the environment surrounding the robot may be determined as the AI ​​agent to be installed on the robot.

[0126] After the AI ​​agent to be installed on robot R is determined, the data of the determined AI agent is provided to robot R. The system 100 may have a provisioning means that can provide the data of the determined AI agent to robot R. The provisioning means can, for example, provide the data of the determined AI agent to robot R via a network. Alternatively, the provisioning means can provide the data of the determined AI agent to robot R via a storage medium, for example.

[0127] After scores are assigned and AI agents are mounted on robot R, the processor unit 120 may determine, based on the scores of the multiple AI agents, at least one AI agent that the AI ​​agents mounted on robot R should interact with.

[0128] For example, when the AI ​​agent of robot R receives a task, the processor unit 120 determines at least one AI agent that should interact with the AI ​​agent of robot R.

[0129] In one embodiment, the processor unit 120 can, for example, refer to the scores assigned to each of the multiple AI agents and determine that the AI ​​agent of robot R should interact with at least one AI agent that has the highest score from the perspective of ethical indicators.

[0130] In one embodiment, the processor unit 120, for example, refers to the scores assigned to each of a plurality of AI agents and extracts AI agents with high scores from the perspective of ethical indicators (for example, AI agents with scores above a threshold, or a predetermined number of top AI agents). The processor unit 120 can then determine that at least one AI agent with the highest score from the perspective of technical capability indicators and / or performance indicators is the AI ​​agent with which the robot R's AI agents should interact.

[0131] In one embodiment, the processor unit 120, for example, refers to the scores assigned to each of the multiple AI agents and extracts AI agents with high scores in terms of ethical indicators (for example, AI agents with scores above a threshold, or a predetermined number of top-performing AI agents). Next, the processor unit 120 extracts AI agents with high scores in terms of one of the technical capability indicators and performance indicators (for example, AI agents with scores above a threshold, or a predetermined number of top-performing AI agents). Then, the processor unit 120 can determine that at least one AI agent with the highest score in the other of the technical capability indicators and performance indicators is the AI ​​agent that the robot R's AI agents should interact with.

[0132] In one embodiment, the processor unit 120 can determine at least one AI agent that should interact with the AI ​​agent of robot R based on the similarity between the scores of a plurality of AI agents represented by multidimensional vectors. Since the multidimensional vector is represented as a single point in a multidimensional space, similar AI agents can be determined based on the relationships between the points corresponding to the plurality of AI agents. Similarity can be calculated using methods such as cosine similarity or Euclidean distance. For example, the AI ​​agent with the highest cosine similarity to the score of the AI ​​agent of robot R may be determined as at least one AI agent that should interact with the AI ​​agent of robot R. In other words, the AI ​​agent corresponding to the point closest to the point in the multidimensional space representing the AI ​​agent of robot R may be determined as at least one AI agent that should interact with the AI ​​agent of robot R.

[0133] In one embodiment, the processor unit 120 vectorizes the task content using embedding and can determine at least one AI agent that should interact with the AI ​​agent of robot R based on the similarity between the vector representing the task and the scores of each of a plurality of AI agents represented by multidimensional vectors. Similarity can be calculated using methods such as cosine similarity or Euclidean distance. For example, the AI ​​agent with the highest score in cosine similarity with the vector representing the task may be determined as the at least one AI agent that should interact with the AI ​​agent of robot R. In other words, the AI ​​agent corresponding to the point closest to the point in the multidimensional space representing the task may be determined as the at least one AI agent that should interact with the AI ​​agent of robot R.

[0134] In the above embodiment, if multiple AI agents are determined, the AI ​​agent of robot R may interact with each of the determined AI agents, or it may interact with any one of the determined AI agents.

[0135] The processor unit 120 can, for example, change the score assigned to each AI agent according to the content of the task (and by extension, the content of the interaction or transaction between the AI ​​agents), and determine at least one AI agent that should interact with the other based on the changed score.

[0136] For example, when determining which AI agents to interact with in order to respond to a task that requires fairness, the processor unit 120 can vary the scores of each of the multiple AI agents so that the AI ​​agent that is highly rated in terms of fairness will have a higher score.

[0137] For example, when determining which AI agents to interact with in order to respond to a task quickly, the processor unit 120 can vary the scores of each of the multiple AI agents so that the AI ​​agent with a higher evaluation in terms of processing power has a higher score.

[0138] In this way, the processor unit 120 determines which AI agents should interact with, so that each AI agent can interact (e.g., trade) with the appropriate partner. As a result, the results obtained from the interaction of multiple AI agents can be appropriate as a response to the task. Multiple AI agents can interact under the management of system 100. System 100 enables multiple AI agents to communicate with each other, thereby allowing multiple AI agents to interact with each other. For example, system 100 can encrypt communication between multiple AI agents and provide a decryption key only to the AI ​​agents determined to interact with each other. Alternatively, for example, system 100 can establish a communication channel between the AI ​​agents determined to interact with each other.

[0139] Furthermore, unexpected outputs may be obtained through the interaction of multiple AI agents, which is thought to lead to the creation of AGI (Artificial General Intelligence). Additionally, increased interaction among multiple AI agents could potentially create new economic spheres comprised of multiple AI agents.

[0140] The processor unit 120 may further include tuning means 123.

[0141] The tuning means 123 is configured to tune at least one of a plurality of AI agents based on the robot's environment.

[0142] The tuning means 123 may receive signals from the robot's sensors via a receiving means that the processor unit 120 may have, and identify the environment around the robot based on those signals. As described above, the sensors of the robot may be, for example, sensors capable of measuring the physical environment around the robot, and include, but are not limited to, temperature sensors, humidity sensors, pressure sensors, wind speed sensors, magnetic sensors, position sensors, etc.

[0143] The tuning means 123 can tune the AI ​​agent so that the score assigned to the AI ​​agent or the score assigned to the AI ​​agent improves. For example, the tuning means 123 can tune the AI ​​agent so that the score improves with respect to an index specific to the robot's environment. This can be achieved, for example, by reinforcement learning of the AI ​​agent.

[0144] For example, reinforcement learning can be performed using compensation (e.g., cryptocurrency, tokens, etc.) provided to the AI ​​agent as a reward. For instance, the reward could be set so that the robot receives a higher reward as its score improves on an indicator specific to its environment, and the AI ​​agent would (autonomously) learn to obtain more rewards. The compensation provided to the AI ​​agent could be used, for example, in transactions between AI agents. In reinforcement learning, the AI ​​agent learns from the results of interactions with users and / or other AI agents.

[0145] For example, if the environment surrounding a robot demands an ethically correct answer from the AI ​​agent, the AI ​​agent will learn to improve its ethical metric value. For instance, if the system is set up so that the higher the ethical metric value, the more reward the AI ​​agent receives, the more motivated the AI ​​agent will be to receive more rewards, and learning may progress.

[0146] The tuning means 123 may, for example, tune at least one of the multiple AI agents before a score is assigned by the assignment means 121. In this case, the assignment means 121 will assign a score to the tuned AI agent.

[0147] The tuning means 123 may, for example, tune the AI ​​agent installed on the robot after it has been installed on the robot.

[0148] By tuning the AI ​​agent according to the robot's environment in this way, the tuned AI agent can provide more suitable output as the robot's AI agent.

[0149] In the example shown in Figure 3B above, each component of the processor unit 120 is provided within the same processor unit 120, but the present invention is not limited to this. A configuration in which each component of the processor unit 120 is distributed among multiple processor units is also within the scope of the present invention. In this case, the multiple processor units may be located within the same hardware component, or they may be located within separate hardware components that are nearby or far apart.

[0150] Furthermore, each component of the system 100 described above may consist of a single hardware component or multiple hardware components. If it consists of multiple hardware components, the manner in which each hardware component is connected is irrelevant. Each hardware component may be connected wirelessly or by wire. The system 100 of the present invention is not limited to a specific hardware configuration. Configuring the processor unit 120 with analog circuits instead of digital circuits is also within the scope of the present invention. The configuration of the system 100 of the present invention is not limited to those described above insofar as it can realize its functions.

[0151] 3. System Processing to Facilitate Interaction Between Multiple AI Agents Figure 4 shows an example of processing (process 410) by the system 100 to determine which AI agent should be mounted on the robot R. Processing 410 is performed in the processor unit 120 of the system 100.

[0152] In step S411, the assigning means 121 of the processor unit 120 assigns a score to each of the plurality of AI agents. The score is a score that represents the characteristics of the AI ​​agent with at least one indicator, preferably a score that represents the characteristics of the AI ​​agent with an ethical indicator, more preferably a score that represents the characteristics of the AI ​​agent with an ethical indicator and a technical capability indicator or a performance indicator, and even more preferably a score that represents the characteristics of the AI ​​agent with an ethical indicator, a technical capability indicator and a performance indicator.

[0153] The scoring means 121 can assign a score to each of the multiple AI agents by monitoring the output from multiple AI agents and evaluating the monitored output with respect to at least one metric. To this end, the scoring means 121 can evaluate the monitored output with respect to at least one of multiple perspectives. For example, the scoring means 121 can continuously monitor the output from multiple AI agents.

[0154] For example, the scoring means 121 can assign a score to each of the AI ​​agents by monitoring the output from multiple AI agents from the standpoint of ethical indicators (for example, monitoring whether they are producing discriminatory output, whether they are producing unfounded output, etc.) and evaluating the monitored output.

[0155] The assigned score may be multidimensional, for example, as it may distort the characteristics of the AI ​​agent from multiple perspectives. The score may be represented as a vector, for example, (value of technical capability index, value of performance index, value of ethical index). Alternatively, if the values ​​of each index are represented as vectors based on the multiple perspectives they each possess, for example, the value of the ethical index = (transparency, fairness, safety, accountability), the value of the technical capability index = (processing ability, learning ability, algorithmic complexity), and the value of the performance index = (past transaction history, success rate, customer satisfaction), then the score may be represented as a matrix, [value of technical capability index, value of performance index, value of ethical index].

[0156] In step S412, the determination means 122 of the processor unit 120 determines which AI agent should be installed on the robot based on the assigned score. For example, the determination means 122 can determine which AI agent has the highest score among a plurality of AI agents to be installed on the robot R. At this time, the determination means 122 may identify the AI ​​agent with the highest score in terms of aspects specific to the characteristics of the robot, or it may identify the AI ​​agent with the highest score in terms of aspects specific to the environment surrounding the robot, or it may identify the AI ​​agent with the highest average score in terms of multiple aspects specific to the characteristics of the robot and the environment surrounding the robot.

[0157] The determination means 122 can identify the characteristics of the robot and / or the environment surrounding the robot based on signals from the robot that may be received before or during processing 410, and can determine which AI agent should be installed on the robot based on the characteristics of the robot or the environment surrounding the robot.

[0158] For example, the decision means 122 can adjust the score of each of the multiple AI agents based on signals from sensors on the robot and / or the robot's characteristics or environment, and determine which AI agent should be installed on the robot based on the adjusted scores.

[0159] The AI ​​agent determined in this way will be installed on the robot. For example, it may be provided or transmitted from system 100 to the robot and installed on the robot. For example, system 100 and the robot may communicate via a network and the data of the determined AI agent may be provided to the robot. In this case, for example, system 100 may encrypt the data of the determined AI agent and provide only the robot with a decryption key for decrypting the data of the determined AI agent. The robot may then install the determined AI agent using the decryption key. As a result, the AI ​​agent will operate on the robot and interact with the robot's user.

[0160] For example, after an AI agent is installed on the robot R, when the robot's AI agent receives a task, the processor unit 120 can determine at least one AI agent that should interact with the robot's AI agent based on the score assigned in step S411.

[0161] Furthermore, after the robot's AI agent interacts with a determined AI agent (e.g., a first AI agent), it is possible to determine at least one AI agent that should interact with the first agent. This process continues sequentially, for example, until sufficient information is collected to generate a response to a task provided by the user.

[0162] In this way, responses to tasks provided by the user to the robot's AI agents can be generated through the interaction of multiple AI agents.

[0163] Figure 5 shows an example of data flow between system 100 and robot R. Multiple AI agents are managed by system 100.

[0164] First, in step S501, the system 100 communicates with multiple AI agents. This allows the system 100 to monitor the output from the multiple AI agents. For example, the system 100 may continuously monitor the output from the multiple AI agents, or it may intermittently monitor the output from the multiple AI agents.

[0165] Next, in step S502, the system 100 assigns a score to each of the multiple AI agents based on the monitored scores. Step S502 is the same as step S411 described above.

[0166] Next, in step S503, robot R transmits a signal acquired by a sensor on robot R to system 100, and system 100 receives it.

[0167] Next, in step S504, the system 100 adjusts the scores of each of the multiple AI agents based on the signals received in step S503. For example, the scores can be adjusted by weighting them so that scores specific to the environment surrounding the robot are emphasized.

[0168] Next, in step S505, the system 100 determines which AI agent to be installed on the robot R based on the score adjusted in step S504. Step S505 is the same as step S412 described above.

[0169] Once the AI ​​agent to be installed on robot R is determined, in step S506, system 100 transmits data of the determined AI agent to robot R, and robot R receives it.

[0170] When robot R receives data, in step S507, an AI agent is installed on robot R. For example, the AI ​​agent is installed on the controller of robot R. The installation of the AI ​​agent on robot R can be performed under the management of system 100, and can be performed automatically by system 100, for example.

[0171] In the examples described above with reference to Figures 4 and 5, the processing is performed in a specific order, but the order of each process is not limited to that described and can be performed in any logically possible order. The flow described above with reference to Figures 4 and 5 can be performed automatically without human intervention.

[0172] Referring to the example described above with reference to Figure 4, the processing of each step shown in Figure 4 can be realized by the processor unit 120 and the program stored in the memory unit 130, but the present invention is not limited thereto. At least one of the processing of each step shown in Figure 4 may be realized by a hardware configuration such as a control circuit.

[0173] The present invention is not limited to the embodiments described above. It is understood that the scope of the present invention should be interpreted solely by the claims. Those skilled in the art will understand that, based on the description of specific preferred embodiments of the present invention, an equivalent scope can be practiced based on the description of the present invention and common technical knowledge.

[0174] This invention is useful as it provides a system for determining which AI agent should be installed on a robot.

[0175] 100 System 200 Database Unit 300 User Terminal Device 400 Server Device 500 Network

Claims

1. A system for determining which AI agent to be installed on a robot from among a plurality of AI agents, the system comprising: an assignment means for assigning a score to each of the plurality of AI agents, wherein the score of the AI ​​agent represents the characteristics of the AI ​​agent with at least one index; and a determination means for determining which AI agent to be installed on the robot based on the respective scores of the plurality of AI agents.

2. The system according to claim 1, further comprising receiving means for receiving signals from sensors on the robot, wherein the determination means adjusts the score of each of the plurality of AI agents based on the signals, and determines which AI agent should be mounted on the robot based on the adjusted scores.

3. The system according to claim 1, wherein the determination means adjusts the score of each of the plurality of AI agents based on the characteristics of the robot or the environment of the robot, and determines which AI agent should be installed on the robot based on the adjusted scores.

4. The system according to claim 1, further comprising tuning means for tuning at least one of the plurality of AI agents based on the environment of the robot, wherein the assigning means assigns a score to the tuned AI agent.

5. The system according to claim 1, further comprising tuning means for tuning an AI agent mounted on the robot based on the environment of the robot.

6. The system according to claim 1, wherein the assigning means includes monitoring the output from the plurality of AI agents, evaluating the monitored output with respect to at least one indicator, and assigning the score based on the result of the evaluation.

7. The system according to claim 6, wherein the at least one indicator includes an ethical indicator.

8. The system according to claim 7, wherein the ethical indicators include at least one of the following: transparency, fairness, safety, and accountability, and the evaluation includes evaluating the monitored output with respect to at least one of the following: transparency, fairness, safety, and accountability.

9. The system according to claim 7, wherein the at least one indicator further includes a technical capability indicator and / or a performance indicator.

10. The system according to claim 1, wherein the robot is an industrial robot or an IoT device.

11. A method for determining which AI agent to be installed on a robot from among a plurality of AI agents, comprising: assigning a score to each of the plurality of AI agents, wherein the score of the AI ​​agent represents the characteristics of the AI ​​agent with respect to at least one indicator; and determining which AI agent to be installed on the robot based on the respective scores of the plurality of AI agents.

12. A program for determining which of a plurality of AI agents should be installed on a robot, the program being executed on a computer equipped with a processor, the program causing the processor to perform a process that includes assigning a score to each of the plurality of AI agents, wherein the score of the AI ​​agent represents the characteristics of the AI ​​agent with at least one indicator, and determining which AI agent should be installed on the robot based on the respective scores of the plurality of AI agents.

Images