JP2019166592A - Robot control system and robot control method - Google Patents

Abstract

To provide a service robot that determines a next action of a robot which operates based on a scenario for complicated work, on the basis of collected log information.SOLUTION: A robot control system, which controls an action of a robot that performs autonomous operation to achieve communication with a user, comprises: a log collection section that collects robot log information from the robot, at a service base where the robot provides service on the basis of an action scenario; an action scenario determination section that determines an action scenario for the robot on the basis of the robot log information collected by the log collection section; and an action scenario updating section that updates the action scenario determined by the action scenario determination section and transmits the updated action scenario to the robot.SELECTED DRAWING: Figure 9

Description

  The present invention relates to a service robot that provides various services to human beings, and more particularly, to a behavior control system and a behavior control method for a robot that performs autonomous operation and realizes communication with a user.

  Service robots are expected as a means to solve labor shortages associated with advances in robot technology and declining birthrate and aging population. In order to control the operation of the robot, it is necessary to control it using a scenario including a robot operation command. Therefore, a control system has been proposed that selects the next action by comprehensively judging the situation where the robot is placed from the recognition results of the external environment such as vision and hearing, and the internal state of the robot (for example, patents). Reference 1). Also, a robot control method has been proposed in which a scenario with the highest priority is selected from a plurality of scenarios and executed (for example, Patent Document 2).

JP 2003-334785 A JP2013-158869A

  The technique described in Patent Document 1 evaluates an action performed based on recognition information of many external environments such as vision and hearing, and controls the next action. Therefore, since comprehensive judgment becomes very complicated, it is difficult to take action control, and there is a problem that it cannot be applied to a robot that performs complicated operations.

  The technology described in Patent Document 2 uses industrial robots, and the priority order of scenarios for realizing operations is determined by the operation efficiency derived from the time required to execute the scenario, the average speed, and the success rate. is doing. Therefore, there is a problem that it can be applied to a robot that operates in a simple work scenario, but cannot be applied to a robot that operates in a complicated work scenario such as a counter service or customer service.

  SUMMARY OF THE INVENTION An object of the present invention relates to a service robot that provides various services to humans, and is to provide a robot control system and a robot control method applicable to a robot that operates in a complicated work scenario.

  A robot control system according to the present invention is a robot control system that controls a behavior of a robot that performs autonomous operation and realizes communication with a user, and is provided at a service base where the robot provides a service based on a behavior scenario. A log collecting unit that collects robot log information from the robot, an action scenario determining unit that determines an action scenario of the robot based on the robot log information collected by the log collecting unit, and the action scenario determining unit And a behavior scenario update unit that updates the behavior scenario and transmits the updated behavior scenario to the robot.

  The present invention is also grasped as a robot control method performed by the robot control system.

  The present invention relates to a service robot that provides various services to humans, and can be applied to a robot that operates in a complicated work scenario.

It is a block diagram which shows one Example of a robot control system. It is a figure which shows the functional block of a log collection server. It is a figure which shows the functional block of an operation management server. It is a figure of the operation | movement sequence which shows operation | movement of a log collection program. It is a figure of the operation | movement sequence which shows operation | movement of an action scenario update program. It is a figure of the operation | movement sequence which shows operation | movement of a log information collection program. It is a figure of the operation | movement sequence which shows operation | movement of a field model production | generation program. It is a figure of the operation | movement sequence which shows operation | movement of a field model update program. It is a figure of the operation | movement sequence which shows operation | movement of an action scenario determination program. It is a figure which shows the example of the list of log information. It is a figure which shows the example of the identifier added to log information. It is a figure which shows the structure of log data. It is a figure which shows an example of operation | movement of an action scenario determination program. It is a figure which shows an example of target objective information DB. It is a figure which shows an example of robot information DB. It is a figure which shows an example of external apparatus information DB. It is a figure which shows an example of core system information DB. It is a figure which shows an example of external system information DB. It is a figure which shows an example of field model information DB. It is a figure which shows an example of action scenario information DB. It is a figure which shows an example of question item information. It is a figure which shows an example of a persona. It is a figure which shows an example of an on-site model. It is a figure which shows an example of persona information, a field model, etc. It is the figure which showed an example of the information after last time and the information about persona information and a field model. It is a figure which shows an example of correlation with temperature, store sales (number) of an air conditioner, and store sales (number) of an electric fan.

DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the invention will be described in detail with reference to the accompanying drawings.
In the following embodiments, a behavior scenario is a generic term for applications that control actions including dialogues that a robot utters, user interaction methods, movements and movements of operating parts, and limits one action. is not.

  Further, in the following embodiment, the log is a past history of a robot operating according to an action scenario, a dialogue history with a user, a past audio recording or video recording of an environment camera or the like including the robot, and a past operation of a program. It is a general term for recording and the like, and is not limited to one recording.

FIG. 1 is a diagram illustrating an example of the configuration of the robot control system according to the first embodiment.
The robot control system 1000 according to the first embodiment includes a service base 100 that provides a service, an operation management center 110 that performs operation management of the entire system, a customer base base 120 that manages a customer core system, and an external And an external system 130 such as a Web site.
The service base 100, the operation management center 110, the customer core base 120, and the external system 130 are connected by a network 140.

  The service base 100 which is a place to provide a service cooperates with the robot 20 that provides the service, the external device 30 such as an environmental camera that executes the service in cooperation with the robot 20, and the robot 20 and the robot 20 that provide the service. And a log collection server 10 that collects logs transmitted from the external device 30 such as an environmental camera that executes the service.

The operation management center 110 includes an operation management server 40 that is a control device that manages the operation of the robot 20.
The customer core base 120 has a core server 50 that manages customer information, store information, and the like.
The external system 130 includes an external server 60 that manages external search sites, weather information, price information, and the like.

  FIG. 2 is a diagram illustrating an example of the configuration of the log collection server 10. In the log collection server 10, a program (software) is stored in an auxiliary storage device 104 of a general computer in order to execute processing, and a program read by the CPU (Central Processing Unit) 102 is read from the auxiliary storage device 104. It is expanded on the memory 101 and executed. The log collection server 10 communicates with other servers and devices via the network I / F 105.

  An I / O (input / output interface) 103 is a user interface for the user to input an instruction to the log collection server 10 and present the execution result of the program to the user. Input / output devices (for example, a keyboard, a mouse, a touch panel, a display, a printer, etc.) are connected to the I / O 103. The I / O 103 may be connected to a user interface provided by a management terminal connected via a network.

  The CPU 102 is a processor that executes a program stored in the memory 101. The memory 101 includes a ROM (Read Only Memory) that is a nonvolatile storage element and a RAM (Random Access Memory) that is a volatile storage element. The ROM stores an invariant program (for example, BIOS: Basic Input Output System). The RAM is a high-speed and volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores a program stored in the auxiliary storage device 104 and data used when the program is executed.

  The memory 101 stores a log collection program 111 and an action scenario update program 112. The log collection program 111 is a program for executing a log collection process (see FIG. 4) for collecting logs transmitted from the robot 20 and the external device 30. The behavior scenario update program 112 is a program for executing behavior scenario execution processing (see FIG. 5).

  The memory 101 also includes a target purpose information DB (DataBase) 121 (see FIG. 14A), an identifier information DB 122 (see FIG. 11), a robot information DB 123 (see FIG. 14B), an external device information DB 124 (see FIG. 14C), and a backbone. A system information DB 125 (see FIG. 15A), an external system information DB 126 (see FIG. 15B), an on-site model information DB 127 (see FIG. 16), and an action scenario information DB 128 (see FIG. 17) are stored.

  The auxiliary storage device 104 is a large-capacity and nonvolatile storage device such as a magnetic storage device (HDD: Hard Disk Drive) or a flash memory (SSD: Solid State Drive). The auxiliary storage device 104 stores a program executed by the CPU 102 and data used when the program is executed. That is, the program is read from the auxiliary storage device 104, loaded into the memory 101, and executed by the CPU 102.

  The log collection server 10 is a computer system that is physically configured on one computer or a plurality of logical or physical computers, and the program stored in the memory 101 is stored on the same computer. It may operate on a separate thread, or may operate on a virtual machine constructed on a plurality of physical computer resources. Further, the log collection server 10 and other devices may be accommodated in one physical or logical computer. Note that all or a part of the processing realized by executing the program may be realized by hardware (for example, Field-Programmable Gate Array).

  FIG. 3 is a diagram illustrating an example of the configuration of the operation management server 40. The processing contents of the operation management server 40 are stored in the form of a program (software) in the auxiliary storage device 404 of a general computer, and the CPU 402 expands the program read from the auxiliary storage device 404 on the memory 401 and executes it. To do. The operation management server 40 communicates with other servers and devices via the network I / F 405.

  The I / O 403 is a user interface for the user to input an instruction to the operation management server 40 and present the execution result of the program to the user. Input / output devices (for example, a keyboard, a mouse, a touch panel, a display, a printer, etc.) are connected to the I / O 403. The I / O 403 may be connected to a user interface provided by a management terminal connected via a network.

  The CPU 402 is a processor that executes a program stored in the memory 401. The memory 401 includes a ROM that is a nonvolatile storage element and a RAM that is a volatile storage element. The ROM stores an invariant program (for example, BIOS). The RAM is a high-speed and volatile storage element such as a DRAM, and temporarily stores a program stored in the auxiliary storage device 402 and data used when the program is executed.

  The memory 401 stores a goal / purpose setting program 411, a log data collection program 412, a site model generation program 413, a site model update program 414, and an action scenario determination program 415.

  The target / purpose setting program 411 is a program for setting a target or purpose of a service provided by the robot 20. In the goal / purpose setting program 411, for example, a goal or purpose such as a sales amount, the number of customers, and the number of answers to a question are set.

  For example, the goal / purpose setting program 411 obtains a behavior scenario for operating the robot 20 from the behavior scenario DB 128 by selecting a goal (H1), place (H2), purpose (H3), time (H4), and behavior (H5). The combined action scenario is confirmed (S101 in FIG. 4). In this case, the target / purpose setting program 411 has a target (H1) for sales amount (h11), a place (H2) for air conditioner department (h22), a purpose (H3) for sale (h33), and a time (H4) for humans. When it comes (h43), the behavior scenario A when the behavior (H5) is PR (h55) is defined and set as h11 × h22 × h33 × h43 × h55. The target or purpose set by the target / purpose setting program 411 is stored in the target / purpose information DB 421.

  The log data collection program 412 is a program for executing log collection processing (see FIG. 6) for collecting logs transmitted from the log collection server 10. The field model generation program 413 is a program for executing a field model creation process (see FIG. 7) for creating a field model from the log data transmitted from the log collection server 10. The on-site model update program 414 is a program for executing on-site model update processing (see FIG. 8) for updating the on-site model. The action scenario determination program 415 is a program for executing an action scenario determination process (see FIG. 9) for determining an action scenario for determining the operation of the robot.

  Further, the memory 401 includes a target purpose information DB (DataBase) 421 (see FIG. 14A), an identifier information DB 422 (see FIG. 11), a robot information DB 423 (see FIG. 14B), an external device information DB 424 (see FIG. 14C), and a backbone system. An information DB 425 (see FIG. 15A), an external system information DB 426 (see FIG. 15B), an on-site model information DB 427 (see FIG. 16), and an action scenario information DB 428 (see FIG. 17) are stored.

  The auxiliary storage device 404 is a large-capacity non-volatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD). The auxiliary storage device 404 stores a program executed by the CPU 402 and data used when the program is executed. That is, the program is read from the auxiliary storage device 404, loaded into the memory 401, and executed by the CPU 402.

  The operation management server 40 is a computer system that is physically configured on one computer or a plurality of logical or physical computers, and the program stored in the memory 401 is stored on the same computer. It may operate on a separate thread, or may operate on a virtual machine constructed on a plurality of physical computer resources. In addition, the operation management server 40 and other devices may be accommodated in one physical or logical computer. Note that all or a part of the processing realized by executing the program may be realized by hardware (for example, Field-Programmable Gate Array).

  FIG. 4 is a flowchart illustrating an example of log collection processing. The log collection process is a process performed when the log collection program 111 is executed by the CPU 102 of the log collection server 10. The log collection program 111 is a process performed when the log collection server 10 acquires information from the operation management server 40 that the robot 20 and the external device 30 start a service.

  When the log collection program 111 acquires an action scenario for the operation of the robot 20 set by the target / purpose setting program 411 from the operation management server 40 (S101), the target / purpose information DB 121, the robot information DB 123, the external By referring to the device information DB 124, the core system information DB 125, and the external system information DB 126, a list 200 (see FIG. 10) of devices and log information used in the acquired action scenario is created, and a device for collecting log information is determined ( S102). The types of devices used in the action scenario and the types of log information to be acquired from the devices are associated in advance, and a log information list 200 (see FIG. 10) is created according to the association. Is done.

  When the log collection program 111 acquires log information from the robot 20 and the external device 30 among the devices determined in S102 (S103), the identifier is added to the header portion of the acquired log information according to the identifier information DB 122 (see FIG. 11). Is added (S104), and log data 220 (see FIG. 12) is created. The log collection program 111 calculates the correlation between the target purpose identifier of the log data 220 and the target purpose information of the target purpose information DB 121 (S105), and if the correlation is weak, stores the log data 220 in the auxiliary storage device 104. (S106), the process ends. When the correlation is strong, the created log information list 200 is transmitted together with the log data 220 to the operation management server 40 (S107), and the process is terminated.

  An example of the operation of the log collection program 111 is as follows. The log collection program 111 uses the behavior scenario DB 128 (FIG. 17) as a behavior scenario for the robot 20 to operate, from the goal (H1), the place (H2), the purpose (H3), the time (H4), and the behavior (H5). An action scenario combining the above is acquired (S101). Here, the target (H1) is the sales amount (h11), the place (H2) is the air conditioner section (h22), the purpose (H3) is sold (h33), and the time (H4) is when a person comes (h43). As an action scenario A when (H5) is PR (h55), it is defined as h11 × h22 × h33 × h43 × h55.

  When the log collection program 111 acquires the action scenario A from the operation management server 40 (S101), the log collection program 111 has an ID that matches the sales amount (h11) that is the target of the action scenario A from the target purpose information DB 121 (FIG. 14A). 0001 is selected, the machine 1 (a11) is selected from the robot information DB 123 (FIG. 14B), the temperature sensor (b21) is selected from the external device information DB 124 (FIG. 14C), and the device for collecting log information is determined ( S102).

  The log collection program 111 collects log information from the device determined above (S103), and creates log data 220 (see FIG. 12) (S104). For example, the log collection program 111 analyzes the collected log information, and determines which target / purpose indicated in the identifier information DB 122 (FIG. 11) corresponds to the content of the log. As a result, for example, the log collection program 111 adds 0001 as an identifier corresponding to the target / purpose / sales amount to the header portion of the collected log information. When the log collection program 111 collects log information from the robot 20, the log collection program 111 adds a11 to the header portion as an identifier An corresponding to the device / robot (ID of the robot 20), and logs information from the external device 30. Is collected, the log data 220 is created by adding b21 to the header portion as an identifier Bn corresponding to the device / external device (ID of the external device 30). In this way, the log collection program 111 associates, with collected log information, an identifier indicating the target / purpose corresponding to the log information and an identifier indicating the device that executes the target / purpose. 220 is generated.

  Further, the log collection program 111 associates the collected log information with the generated log data 220 and an identifier indicating the type of the log information. For example, the log collection program 111 analyzes the collected log information, and if the type of log information is log information in the image data format of the robot 20, refer to the identifier information DB 122 (FIG. 11), The identifier 01 corresponding to the image data is read and added to the header portion of the collected log information.

  Further, if the log information content type is log information including the conversation of the robot 20, the log collection program 111 reads the identifier 001 corresponding to the face-to-face log with reference to the identifier information DB 122 (FIG. 11). And add it to the header of the collected log information.

  The log collection program 111 uses a target / purpose identifier (for example, 0001) of the created log data 220 and a target / target value ID (for example, the target purpose information DB 121 corresponding to the target (for example, sales amount: h11) defined in the action scenario A). 0001) and whether the two match each other (S105; Yes), it is determined that the correlation is strong, and is created together with the log data 220 to the operation management server 40 The log information list 200 is transmitted (S107), and the process is terminated.

  On the other hand, the log collection program 111 determines whether or not the target / purpose identifier (for example, 0002) of the created log data 220 matches the target purpose value ID (for example, 0001) of the target purpose information DB 121. Are determined to be inconsistent (S105; No), it is determined that the correlation is weak, the log data 220 is stored and stored in the auxiliary storage device 104 (S106), and the process ends.

  FIG. 5 is a flowchart illustrating an example of behavior scenario update processing. The behavior scenario update process is a process performed when the behavior scenario update program 112 is executed by the CPU 102 of the log collection server 10. The behavior scenario update program 112 is a process performed when the log collection server 10 acquires information for starting a service from the operation management server 40.

  When the action scenario update program 112 acquires an action scenario from the operation management server 40 (S201), the action scenario update program 112 reads the acquired action scenario and confirms its contents (S202), and is stored in the action scenario DB 128 (FIG. 17). The difference with the action scenario used when operating is calculated, and it is determined whether there is a difference between them (S203). If the behavior scenario update program 112 determines that there is no difference between the two (S203; No), the behavior scenario update program 112 discards the acquired behavior scenario (S204) and ends the processing.

  On the other hand, if the behavior scenario update program 112 determines that there is a difference between the two (S203; Yes), the behavior scenario used at the previous operation is updated to the acquired behavior scenario (S205). The updated action scenario is stored in the DB 128 and transmitted to the robot 20 (S206), and the process ends.

  An example of the operation of the behavior scenario update program 112 is as follows. The action scenario update program 112 combines a goal (H1), a place (H2), a purpose (H3), a time (H4), and an action (H5) from the action scenario DB 128 as an action scenario for the robot 20 to operate. An action scenario is acquired and the content is confirmed (S201, S202). Here, as an action scenario of this time, the target (H1) is the sales amount (h11), the place (H2) is the air conditioner section (h22), the purpose (H3) is sold (h33), and the time (H4) is visited. The behavior scenario A when the time (h43) and the behavior (H5) are PR (h55) is defined as h11 × h22 × h33 × h43 × h55. As the previous action scenario, the goal (H1) is the sales amount (h11), the place (H2) is received (h21), the purpose (H3) is sold (h33), the time (H4) is in the morning (h41), The behavior scenario B when the behavior (H5) is PR (h55) is defined as h11 × h21 × h33 × h41 × h55.

  After executing S201 and S202, the behavior scenario update program 112 calculates the difference between the behavior scenario B used when the previous operation was stored and stored in the behavior scenario DB 128 and the behavior scenario A acquired this time (S203). . For example, the behavior scenario update program 112 updates the location (H2) from the reception (h21) to the air conditioner department (h22), and updates the time (H4) when the person comes from the morning (h41) (h43). Then, the updated action scenario is transmitted to the robot 20 (S205, S206).

  FIG. 6 is a flowchart illustrating an example of log data collection processing. The log data collection process is a process performed when the log data collection program 412 is executed by the CPU 402 of the operation management server 40. The log data collection program 412 is a process performed when the operation management server 40 receives log data from the log collection server 10.

  When the log data collection program 412 receives the log data 220 from the log collection server 10 transmitted in S107 in FIG. 4 (S301), the log data collection program 412 confirms the identifier of the header portion of the log data 220, and the target purpose information DB 421 and the log It is determined whether or not all the identifiers in the header portion of the data 220 match (S302). When the log data collection program 412 determines that the two do not match (S302; No), the log data collection program 412 stores the received log data 220 in the auxiliary storage device 404 (S303). On the other hand, if the log data collection program 412 determines that they match (S302; Yes), the log data collection program 412 calculates a difference from the previously acquired log data 220 stored in the auxiliary storage device 404, and whether there is a difference. It is determined whether or not (S304). When the log data collection program 412 determines that there is no difference between them (S304; No), the received log data 220 is stored in the auxiliary storage device 404 (S305), and the process is terminated. On the other hand, if the log data collection program 412 determines that there is a difference between them (S304; Yes), it transmits the received log data 220 to the on-site model generation program 413 (S306), and ends the process.

  An example of the operation of the log data collection program 412 is as follows. The log data collection program 412 combines a goal (H1), a place (H2), a purpose (H3), a time (H4), and a behavior (H5) from the behavior scenario DB 428 as behavior scenarios for the robot 20 to operate. Check the action scenario. Here, the target (H1) is the sales amount (h11), the place (H2) is the air conditioner section (h22), the purpose (H3) is sold (h33), and the time (H4) is when a person comes (h43). As behavior scenario A when (H5) is PR (h55), it is defined as h11 × h22 × h33 × h43 × h55.

  When the log data collection program 412 receives the log data 220 from the log collection server 10 (S301), the log data collection program 412 confirms the ID (eg, 0001) of the target / target identifier in the header portion of the log data 220. The log data collection program 412 stores the target / purpose identifier ID (for example, 0002) of the header portion of the previous log data 220 stored in the target purpose information DB 421 and the target purpose of the header portion of the log data 220 received in S301. It is determined whether or not the identifier ID (for example, 0001) matches (S302). If it is determined that they do not match (S302; No), the received log data 220 is stored in the auxiliary storage device 404. (S303). On the other hand, if the log data collection program 412 determines that they match (S302; Yes), it further compares the device identifiers in S302.

  The log data collection program 412 determines whether the device identifier in the header portion of the received log data 220 is an identifier An indicating a robot, and determines that the device identifier in the header portion is an identifier An indicating a robot. In this case, the device identifier (for example, a12) in the header portion of the previous log data 220 is compared with the device identifier (for example, a11) in the header portion of the log data 220 received this time, and it is determined whether or not they match. If the log data collection program 412 determines that they do not match, the log data 220 received is stored in the auxiliary storage device 404 (S303), and the process ends.

  On the other hand, if the log data collection program 412 determines that the two match, the log data collection program 412 further compares the log 1 identifiers. The case where the device identifier in the header portion of the received log data 220 is an identifier Bn indicating an external device is determined in the same manner as in the case of the identifier An indicating a robot.

  In S302, the log data collection program 412 stores the log 1 identifier (for example, 01 indicating image data) in the header portion of the previous log data 220 stored in the auxiliary storage device 404 and the log data 220 received this time. A log 1 identifier (for example, 00 indicating voice data) in the header portion is compared to determine whether or not they match. If the log data collection program 412 determines that they do not match, the log data 220 received is stored in the auxiliary storage device 404 (S303), and the process ends.

  On the other hand, if the log data collection program 412 determines that the two match, the log data collection program 412 further compares the log 2 identifiers.

  The log data collection program 412 stores the log 2 identifier (for example, 002 indicating the detection log) of the header portion of the previous log data 220 stored in the auxiliary storage device 404 and the log 2 of the header portion of the log data 220 received this time. An identifier (for example, 000 indicating a conversation log) is compared to determine whether or not the two match. If the log data collection program 412 determines that they do not match, the log data 220 received is stored in the auxiliary storage device 404 (S303), and the process ends.

  On the other hand, the log data collection program 412 stores the log 2 identifier (for example, 000 indicating a conversation log) of the header portion of the previous log data 220 stored in the auxiliary storage device 404 and the header portion of the log data 220 received this time. The log 2 identifier (for example, 000 indicating the conversation log) is compared, and if it is determined that they match, the presence / absence of a difference at the time of the previous collection is further determined (S304).

  The log data collection program 412 compares the log information of the previous log data 220 stored in the auxiliary storage device 404 with the log information of the log data 220 received this time, and determines whether there is a difference between the two. . If the log data collection program 412 determines that there is no difference between the two (S304; No), it stores the received log data 220 in the auxiliary storage device 404 (S303) and ends. On the other hand, if it is determined that there is a difference between the two (S304; Yes), the log data collection program 412 transmits the received log data 220 to the on-site model generation program 413 (S306), and ends the process.

  In FIG. 6, the log data collection program 412 has described the processing for the log data 220 received from the log collection server 10, but in addition to this, the log information is received from the core server 50 and the external server 60 and received. Log information, the action scenario set by the target / purpose setting program 411, the target / purpose information DB 421, the basic system information DB 425, the external system information DB 426, and the log information list 200, and the basic server Generate log data 220 for the external server. For example, as with the log collection program 111, the following may be performed.

  When the log data collection program 412 collects log information from the core server 50 and the external server 60, the log data collection program 412 analyzes the collected log information and corresponds to any target / purpose indicated in the identifier information DB 122 (FIG. 11). It is determined whether it is. As a result, for example, the log collection program 111 adds 0001 as an identifier corresponding to the target / purpose / sales amount to the header portion of the collected log information. When the log data collection program 412 collects log information from the backbone server 50 (customer management server), the log data collection program 412 uses C1 as the identifier Cn corresponding to the device / main server (the ID of the backbone server 50) in the header portion. In addition, when log information is collected from the external server 60, E1 is added to the header portion as an identifier En corresponding to the device / external server (external server 60 ID), and log data 220 is created. As described above, the log data collection program 412 associates, with the collected log information, an identifier indicating the target / purpose corresponding to the log information and an identifier indicating the server that executes the target / purpose. Data 220 is generated.

  Further, the log data collection program 412 associates the collected log information with an identifier indicating the type of the log information with the generated log data 220. For example, the log data collection program 412 analyzes the collected log information, and if the type of log information is the log information of the core server 50 (customer management server), refer to the identifier information DB 122 (FIG. 11). Then, the identifier 20 corresponding to the customer management server is read and added to the header portion of the collected log information.

  Further, if the type of log information is log information including the number of points, the log data collection program 412 reads the identifier 200 corresponding to the number of points with reference to the identifier information DB 122 (FIG. 11). And add it to the header of the collected log information.

  And the log data collection program 412 performs the process similar to each process of S301-S306 of FIG. 6 with respect to the log data 220 about the produced | generated core server and external server.

  FIG. 7 is a flowchart illustrating an example of the on-site model generation process. The on-site model generation process is a process performed when the on-site model generation program 413 is executed by the CPU 402 of the operation management server 40. The on-site model generation program 413 is processing performed when the operation management server 40 receives the log data 220 collected and generated by the log data collection program 412.

  The site model generation program 413 receives the log data 220 from the log data collection program 412 (S401), confirms the identifier of the header portion of the log data 220, and, similarly to S302, the target purpose information DB 421 and the header of the log data 220 It is determined whether or not all the identifiers of the part match (S402). When the site model generation program 413 determines that the identifiers of the header portions of the target purpose information DB 421 and the log data 220 do not match (S402; No), the received log data 220 is stored in the auxiliary storage device 404 ( S403).

  On the other hand, if the site model generation program 413 determines that the two match (S402; Yes), it determines whether the device identifier of the log data 220 is an identifier indicating the robot 20 (S404). When the on-site model generation program 413 determines that the device identifier is not an identifier indicating the robot 20 (S404; No), it reads the log information included in the received log data 220, and the status of the environment in which the service is provided For example, site model information (FIG. 16) indicating the state of the robot or the state of the external device is created (S405), the created site model information is transmitted to the site model update program 414 (S408), and processing is performed. finish.

  On the other hand, when the on-site model generation program 413 determines that the device identifier is an identifier indicating the robot 20 (S404; Yes), the stored on-site model information DB 427 is read and the question item (FIG. 18) is read. On the other hand, an answer is created from the log information included in the received log data 220 (S406), and persona information (FIG. 18) indicating a virtual person image providing the service is created (S407). The on-site model generation program 413 transmits the created persona information to the on-site model update program 414 (S408), and ends the process.

  An example of the operation of the on-site model generation program 413 is as follows. When the site model generation program 413 receives the log data 220 from the log data collection program 412 (S401), the target / purpose identifier ID (for example, customer collection) of the header portion of the previous log data 220 stored in the target / purpose information DB 421 is received. 0002) indicating the number of persons and the target / purpose identifier ID (for example, 0001 indicating the sales amount) in the header portion of the log data 220 received this time are compared to determine whether or not they match (S402). When it is determined that the two do not match (S402; No), the on-site model generation program 413 stores the received log data 220 in the auxiliary storage device 404 (S403). On the other hand, if the site model generation program 413 determines that the two match (S402; Yes), it further compares the device identifiers (S404).

  When the on-site model generation program 413 determines that the device identifier is An indicating a robot (S404; Yes), the voice data included in the log information (corresponding to the log 1 identifier 00) or the conversation included in the voice data Log (corresponding to log 2 identifier 000), image data included in log information (corresponding to log 1 identifier 01), face-to-face log (corresponding to log 2 identifier 001) and detection log (log) 2) (corresponding to two identifiers 002) is analyzed, answers to the questions included in the question item information 1001 (FIG. 18A) are determined by majority (S406), and persona information 1002 is created (S407). The question item information 1001 and the persona information 1002 will be described later with reference to FIGS. 18A and 18B. The interval for collecting the answers may be one week or one hour. For example, the site model generation program 413 analyzes the gender obtained by analyzing the face-to-face log (log 2 identifier 001) and the detection log (log 2 identifier 002) included in the image data identified by the log 1 identifier for one week. If the ratio is 60% male, 35% female, and 5% unknown, the gender recorded as persona information is male. The on-site model generation program 413 transmits the persona information 1002 created in this way to the on-site model update program 414 (S408), and ends the process.

  The on-site model generation program 413 has a device identifier other than An indicating a robot, for example, Bn indicating an external device, Cn indicating a basic server (customer management system), Dn indicating a basic server (customer management server), When it is determined that it is En indicating an external server (S404; No), the log information included in the received log data 220 is analyzed to create a site model 1003 (FIG. 19) (S405). As a creation method of the on-site model 1003 (FIG. 19), for example, the following processing may be executed. The on-site model generation program 413 refers to the log information of the external server (weather forecast) identified by the device identifier E1 included in the log data 220, and changes the temperature included in the log information (for example, one week) Change). If the analyzed temperature change is a value indicating an increase, the on-site model generation program 413 sets the value (for example, e11) in the increase item of the external system information DB 426 (FIG. 15B). Similarly, the on-site model generation program 413 sets the value (for example, e13) in the decrease item of the external system information DB 426 if the analyzed temperature change is a value indicating a decrease, and the analyzed temperature change changes. If not, the value (for example, e12) is set in a certain item of the external system information DB 426.

  Further, for example, the on-site model generation program 413 refers to the log information of the basic server (customer management) identified by the device identifier C1 included in the log data 220, and determines the number of points included in the log information as the basic system The value (for example, c12) is set in the point number item of the information DB 125 (FIG. 15A). Other items included in the log information (for example, presence / absence of point card, purchase history, number of visits) are set similarly.

  Then, the on-site model generation program 413 collects the basic system information DB 125 and the external system information DB 426 on which each item obtained from the log information is set, and records them in the on-site model information DB 427 (see FIG. 16). At this time, in order to make the layout the same, the on-site model generation program 413 changes daily values of the number of store visits and sales items stored in the core system information DB 125 (FIG. 15A) (for example, for one week) Change), as in the case of the external system information DB 426 (FIG. 15B), the change in the number of visits to the store and the amount of sales is increased, constant, or decreased, and the resulting value is calculated as the on-site model information. It records in the increase item, fixed item, and decrease item of DB427.

  The on-site model generation program 413 can identify the on-site model ID between the main server side and the external server side when recording the main system information DB 125 and the external system information DB 426 in the on-site model information DB 427 as described above. As described above, each ID is assigned in a different ID system. In FIG. 16, the site model generation program 413 assigns IDs identified by F1 to Fn as the site model IDs on the backbone server side, and IDs identified by G1 to Gn as the site model IDs on the external server side. Give.

  Further, the site model generation program 413 creates a site model 1003 (FIG. 19) from the site model information DB 427 shown in FIG. For example, when the value of each item in the on-site model information DB 427 is an increase item, the on-site model generation program 413 converts it into a character string of an expression corresponding to the increase item value such as “sales increase” or “temperature rise” The character string is recorded in the field model item of the field model 1003, and the field model ID and type shown in FIG. 16 are set in the field model ID and field model log information item corresponding to the field model item, respectively. In this way, the on-site model generation program 413 finally creates the on-site model 1003 from the log data 220.

  FIG. 8 is a flowchart illustrating an example of the on-site model update process. The on-site model update process is a process performed when the on-site model update process program 414 is executed by the CPU 402 of the operation management server 40. The on-site model update program 414 is a process performed by receiving persona information 1002 or on-site model 1003 from the on-site model generation program 413.

  Upon receiving the persona information 1002 or the field model 1003 from the field model generation program 413 (S501), the field model update processing program 414 confirms whether it is the field model 1003 or the persona information 1002 (S502), and the field model information. Store in the DB 427 (S503). The on-site model update processing program 414 calculates the difference between the stored on-site model and the previously stored on-site model, and determines whether there is a difference between them (S504). When it is determined that there is no difference between the two (S504; No), the on-site model update processing program 414 transmits the on-site model stored last time to the action scenario determination program 415 (S506), and ends the processing.

  On the other hand, if the site model update processing program 414 determines that there is a difference between them (S504; Yes), the site model is updated using the correlation function and majority function with the previously stored site model (S505). ) The updated site model is stored in the site model information DB 427 and transmitted to the action scenario determination program 415 (S506), and the process is terminated. The determination (S504), update (S505), and transmission (S506) of the difference are similarly executed for the persona information 1002.

  An example of the operation of the on-site model update processing program 414 is as follows. The on-site model update processing program 414 confirms the persona information or the on-site model from the on-site model generation program 413 (S502), and stores it in the on-site model information DB 427 (S503).

  When the field model update processing program 414 receives the persona information 1002, the field model update processing program 414 compares the persona information 1002 with the previous persona information stored in the field model DB 247, and compares the presence / absence of the difference with the previous one. If there is no difference between the two (S504; No), the on-site model update processing program 414 transmits the persona information 1002 to the behavior scenario determination program 415 (S506), and ends the processing. On the other hand, when there is a difference between the two (S504; Yes), the on-site model update processing program 414 determines the received persona information 1002 and the previous persona information stored in the on-site model DB 247 (for example, in the upper part of FIG. 20). Persona information 1004) shown in FIG. 21 is compared with each other and majority processing is performed for each item. Persona information after the majority processing (for example, persona information 1006 shown in FIG. 21) is stored in the on-site model information DB 427 (S503), and an action scenario is stored. The persona information 1006 is transmitted to the determination program 415 (S506), and the process is terminated.

  On the other hand, when the on-site model update processing program 414 receives the on-site model 1003, the on-site model update program 414 compares with the previous on-site model stored in the on-site model information DB 247, and compares the presence or absence of the difference with the previous time. If there is no difference between the two (S504; No), the on-site model update processing program 414 transmits the on-site model 1003 to the action scenario determination program 415 (S506), and ends the processing. On the other hand, when there is a difference between the two (S504; Yes), the site model update processing program 414 compares the difference between the received site model 1003 and the previous site model 1005 stored in the site model information DB 247. When the correlation between temperature and sales is strong (for example, when the value calculated by the above correlation function or majority function is equal to or greater than a certain threshold), the previous field model 1005 (for example, The site model 1007 shown in the lower part of FIG. 21 is updated, the site model 1007 is transmitted to the action scenario determination program 415 (S506), and the process is terminated.

  In FIG. 5, the behavior scenario update program 112 determines whether or not the correlation between the site model determined last time and the site model determined this time is strong as described above, and the correlation between the two is not less than a certain level. When it is determined that there is, an action scenario corresponding to the on-site model determined this time may be updated.

  FIG. 9 is a flowchart illustrating an example of action scenario determination processing. The behavior scenario determination process is a process performed when the behavior scenario determination program 415 is executed by the CPU 402 of the operation management server 40. The action scenario determination program 415 is a process performed by receiving a field model or persona information from the field model update program 414.

  The behavior scenario determination program 415 receives the site model or persona information from the site model update program 414 (S601), and confirms that it is a site model (S602). The behavior scenario determination program 415 confirms the behavior scenario stored in the behavior scenario information DB 428 when the received site model is created (S603), and the goal / purpose information stored in the goal / purpose information DB 421. Is confirmed (S604). The behavior scenario determination program 415 compares the received site model with the target / purpose information, and determines whether the target / purpose is achieved by the site model (S605).

  If the action scenario determination program 415 determines that the goal / purpose has been achieved by the field model (S605; Yes), the action scenario is determined without change (S608), and the action is determined to the log collection server 10. The scenario is transmitted (S609), and the process ends. On the other hand, when the action scenario determination program 415 determines that the goal / purpose has not been achieved by the site model (S605; No), the correlation between the site models is compared (S606), and the action scenario is updated (S606). S607), an action scenario is determined (S608), the action scenario is transmitted to the log collection server 10 (S609), and the process is terminated.

  An example of the operation of the action scenario determination program 415 is as follows. The action scenario determination program 415 receives from the on-site model update program 414 the persona information 1006 after the majority process and the on-site model 1007 that has been determined to have a strong correlation and updated (S601). The action scenario is confirmed (S602).

  The action scenario is a combination of the goal (H1), the place (H2), the purpose (H3), the time (H4), and the action (H5). The previous action scenario is converted from the target (H1) to the sales amount (h11). ), Place (H2) is air conditioner (h22), purpose (H3) is sold (h33), time (H4) is when people come (h43), action (H5) is PR (h55) The behavior scenario A is defined as h11 × h22 × h33 × h43 × h55.

  The action scenario determination program 415 confirms the previous target purpose information from the target purpose information DB 421 (S604). If the set target / target value name is 0001 indicating the sales amount, the field model ID is received from the received field model 1007. The numerical value of F1 indicating customer management (store) is confirmed, and it is confirmed whether or not the target has been achieved (S605). If the action scenario determination program 415 achieves the goal (S605; Yes), the action scenario is set to h00 in all action IDs as an action scenario, or determined to be the same action scenario A as the previous time (S608), and the log collection server 10 The action scenario data is transmitted to (S609), and the process is terminated.

  If the goal cannot be achieved (S605; No), the behavior scenario determination program 415 derives a correlation function from the received on-site model 1007, and the result of the correlation function between the temperature and the fan sales (for example, in FIG. 22). A behavior scenario is created using the correlation graph 1008) and the processed persona information 1006 (S605).

  Specifically, the behavior scenario determination program 415 sets the goal (H1) of the behavior scenario to the sales amount (h11), the location (H2) to the air conditioner department (h22), the purpose (H3) to sale (h33), and the time (H4). ) Is defined as h11 × h22 × h33 × h43 × h55 as the action scenario A when the person (h43) comes to PR (h55) for the action (H5), the persona information 1006 after processing is Refer to the information on the increase in remarks from “air conditioner” to “fan” and the on-site model 1007 after processing, and read the information on “rising temperature” and “increased fan sales”, and the location of the action scenario ( H2) is changed from the air conditioner section (h22) to the electric fan section (h23), and h11 × h23 × h33 × h43 × h55 is defined as an action scenario C and created (S606). The behavior scenario determination program 415 stores the created behavior scenario C in the behavior scenario information DB 428 and updates it (S608). It is confirmed whether there is an instruction to set all action IDs to h00. If there is no problem in action, the action scenario C is determined (S608), and the action scenario is transmitted to the log collection server 10 (S609). Exit.

  FIG. 10 is a diagram showing a list of log information to be acquired. The log information list 200 is stored in the target purpose information DB 121 of the log collection server 10 and the target purpose information DB 421 of the operation management server 40. The items of device and log 1 and log 2 indicate the types of log information acquired from the robot 20, external device 30, backbone server 50, and external system 60.

  For example, when the goal / purpose is set to the sales amount, a conversation log with the robot is acquired from the voice data of the robot 20, a face-to-face log is acquired from the image data, and the robot detection is detected from the failure detection data. Indicates that a log is acquired. From the external device 30, the number of people stopped by the conversation with the robot is acquired from the environment camera 1, and the location of the product where the person stopped is acquired from the environment camera 2. From the backbone server 50 (customer management server), the number of points of the store visitor and the purchase history of the store visitor are acquired, and from the backbone server 50 (business management server), the sales amount for each day, the sales amount for each product, and other stores It shows that the sales amount of is acquired. The external system 60 indicates that today's weather is acquired from the weather forecast DB, and the flyer information of competitors is acquired from an external Web site.

  FIG. 11 is a diagram illustrating examples of identifiers to be added to log information. A 2-digit to 3-digit identifier is added to the header portion of each log information defined in the list of log information shown in FIG. FIG. 11 shows that, for example, when the device used to provide the service is a robot, an identifier An corresponding to the device is added as a header portion of log information.

  FIG. 12 shows the structure of log data. An identifier (FIG. 11) is added to the log data 220 by the log collection program 111, and the data structure of the log data 220 is formed. As described above, the log data 220 has a configuration in which an identifier added as a header portion of log information is associated with the log information, and an identifier that matches the target / purpose, device, and log information is added. (See FIG. 11).

  FIG. 13 is a diagram showing a concept of operation of the action scenario determination program. The action scenario determination program 415 compares the received on-site model with the goal / purpose information to determine whether it has been achieved. If not, the action scenario determination program 415 compares the correlation between the on-site models to determine the action scenario. Update. For example, the behavior scenario determination program 415 is an example of the on-site model created from the collected log information when the goal / purpose is set to achieve sales quota and the behavior scenario is set to correspond in the air conditioner corner. Is shown. The action scenario determination program 415 has a positive correlation between the fan said in the created persona information and the temperature / humidity increase tendency of the field model and the sales increase of the fan, and is analyzed in the persona information. Consider a case where there is a strong negative correlation between an air conditioner and purchase abandonment, which is an action analyzed in persona information, and trouble, which is an emotion analyzed in persona information.

  The action scenario determination program 415 has a strong negative correlation with the air conditioner corner set in the action scenario (place (H2): air conditioner section (h22)). By replacing the air conditioner, the action scenario is updated to the place corresponding to the electric fan corner (place (H2): electric fan section (h23)). The behavior scenario determination program 415 updates the behavior scenario until the goal / purpose is achieved by updating the on-site model and comparing the correlation between the goal / purpose and the behavior scenario. FIG. 14A is a diagram illustrating an example of the target purpose information DB 121/421. As shown in FIG. 14A, the target / purpose value name and its ID are stored in association in the target purpose information DB 121/421. In FIG. 14A, for example, it can be seen that the sales amount, which is the target / target value name, is set as the target / target value ID0001.

  FIG. 14B is a diagram illustrating an example of the robot information DB 123/423. As shown in FIG. 14B, the robot information DB 123/423 stores a robot information ID and an actual machine identified by the ID in association with each other. In FIG. 14B, for example, it can be seen that the bodies of the robot whose robot information ID is A1 are a11, a12, and a13.

  FIG. 14C is a diagram illustrating an example of the external device information 124/424. As shown in FIG. 14C, the external device information 124/424 stores the external device information ID, the type of the external device identified by the ID, and the actual aircraft of that type in association with each other. In FIG. 14C, for example, it can be seen that the bodies of the environmental camera whose external device information ID is B1 are b11, b12, and b13.

  FIG. 15A is a diagram illustrating an example of the core system information DB 125/425. As shown in FIG. 15A, the basic system information DB 125/425 stores a basic server ID, the type of the basic server identified by the ID, and items managed in the type in association with each other. In FIG. 15A, for example, a server that performs customer management of its own store among core servers manages items such as the presence / absence of point cards, the number of points, the purchase history, the number of store visits, and the values are c11, c12, and c13. , C14.

  FIG. 15B is a diagram illustrating an example of the external system information DB 126/426. As shown in FIG. 15B, the external system information DB 126/426 stores the external server ID, the type of the external server identified by the ID, and the increase / decrease of items managed in the type in association with each other. . In FIG. 15B, for example, in the server that notifies the weather forecast among the external servers, an increase (rise), a decrease (decrease), and a constant (no change) are predicted for the temperature, and any one of these values is e11, e12, It can be seen that e13 is set.

  FIG. 16 is a diagram illustrating an example of the on-site model DB 127/427. In the field model DB 127/427, the information acquired from the core server or the external server is classified by increasing, constant, decreasing, and stored together with specific numerical information like the external system information DB 126/426 shown in 15B. The In FIG. 16, for example, in the site model in which the site model ID generated by the core server that performs customer management of the own store is identified by F <b> 1, for example, increases, decreases, and constants regarding the number of visits are counted. It can be seen that f11, f12, and f13 are set as values. Similarly, in the field model in which the field model ID generated by the temperature sensor that detects the temperature of the field is identified by G2, for example, the increase (rise), decrease (fall), and constant (no change) are aggregated. It can be seen that g21, g22, and g23 are set as any one of the values.

  FIG. 17 is a diagram illustrating an example of the behavior scenario information DB 128/428. The behavior scenario is composed of a combination of these behavior scenario IDs and values corresponding to the type of behavior identified by the ID, and a target, place, purpose, time, and behavior are defined. Since the specific example of FIG. 17 is already the same as that in each flowchart, description thereof is omitted here.

  FIG. 18A is a diagram showing an example of question item information 1001 stored in the on-site model information DB 127/427. As shown in FIG. 18A, the question item information 1001 stores question items and log information in association with each other. The question items constituting the question item information 1001 are determined in advance, and in FIG. 18A, it can be seen that the items of sex, age, occupation, family structure, remarks, behavior, and emotion are set. Each of these question items is analyzed using image data and audio data included in the face-to-face log and conversation log stored as log information. FIG. 18A shows that, for example, gender is obtained by analyzing a face image of a person obtained by analyzing image data of a face-to-face log.

  FIG. 18B is a diagram showing an example of persona information 1002 stored in the on-site model information DB 127/427. As shown in FIG. 18B, persona information 1002 indicates persona information 1002 obtained as a result of analysis using the question item information 1001 shown in FIG. 18A. In FIG. 18B, as a result of analysis for each item of gender, age, occupation, family composition, remarks, behavior, and emotion, a single male office worker in the 20s to 40s visits the store, and “air conditioner” “fan” Abandoning the purchase of the persona, it is shown that the persona information that has a troubled expression was obtained. As described with reference to FIG. 7, the persona information is obtained by analyzing image data and audio data, which are log data for one week, for each question item, and the results obtained as the main trends are the persona information. Set as information.

  FIG. 19 is a diagram illustrating an example of a site model stored in the site model information DB 127/427. As shown in FIG. 19, the field model 1003 stores a field model ID for identifying a field model, field model log information, and a field model in association with each other. The on-site model ID and on-site model log information are the same as the on-site model ID and type shown in FIG. 16, respectively. The on-site model represents the numerical values of the reduced and constant items shown in FIG. 16 as character strings. The on-site model identified by the on-site model ID F3, F4, G2, G3, and G6 indicates a model in which the on-site model has been changed from the previous time.

  FIG. 20 is a diagram illustrating an example of persona information stored in the on-site model information DB 127/427 and the on-site model. FIG. 20 shows that, in the persona information 1004, each question item of age and remark has a difference from the previous time (for example, a difference from FIG. 18B). Further, among the on-site models 1005, the models with the on-site model IDs F3, F4, G2, G3, and G6 indicate that a difference from the previous time (for example, a difference from FIG. 19) has occurred.

  FIG. 21 is a diagram illustrating an example of previous information and post-processing information about the persona information and the field model stored in the field model information DB 127/427. As described with reference to FIG. 8, the on-site model update processing program 414 extracts a portion having a difference from the previous information, and selects information having a majority decision and a strong correlation as information after processing. In FIG. 21, it can be seen that the processed persona information 1006 is updated to the “fan” that has a strong correlation between the speech “air conditioner” included in the previous persona information and the speech. In addition, as the on-site model 1007 after processing, the model whose on-site model ID is F4 is updated to “increased sales”, and the on-site model IDs of G2 and G3 are changed to “temperature rise” and “humidity rise”, respectively. It can be seen that the model whose field model ID is G6 is updated to “temperature rise”. That is, it can be seen that in the latest on-site model, when the temperature rises and the temperature and humidity detected by the sensor rise, the sales of electric fans are increasing.

  FIG. 22 is a diagram illustrating an example of the correlation between the temperature, the store sales (number) of air conditioners, and the store sales (number) of fans. As indicated by the black circles in FIG. 22, when the temperature exceeds 25 degrees, the sales of the electric fan increase, indicating that the temperature is high and the sales of the electric fan have a strong correlation.

  Thus, according to this system, it is possible to realize a service robot system that determines the next action of a robot that operates in a complicated work scenario from the collected log information.

  In addition, the form for implementing this invention is not limited to the Example demonstrated above, Various modifications and an equivalent structure are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. A part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Moreover, you may add a part of structure of another Example to the structure of a certain Example.

10: Log collection server 20: Robot 30: External device 40: Operation management server 50: Core server 60: External server 100: Service base 110: Operation management center 120: Customer main base 130: External system 140: Network

Claims (12)

A robot control system that controls the behavior of a robot that performs autonomous operation and realizes communication with a user,
A log collection unit for collecting robot log information from the robot at a service base where the robot provides a service based on an action scenario;
An action scenario determination unit that determines an action scenario of the robot based on the robot log information collected by the log collection unit;
An action scenario update unit that updates the action scenario determined by the action scenario determination unit and transmits the updated action scenario to the robot;
A robot control system comprising: The robot control system according to claim 1,
The log collection unit collects external device log information from an external device that executes a service in cooperation with the robot,
The behavior scenario determination unit determines the behavior scenario of the robot based on the robot log information and the external device log information.
A robot control system characterized by that. The robot control system according to claim 1,
The log collection unit collects backbone server log information from a backbone server connected to the service base via a network,
The behavior scenario determination unit determines the behavior scenario of the robot based on the robot log information and the core server log information.
A robot control system characterized by that. The robot control system according to claim 1,
The log collection unit collects external server log information from an external server connected to the service base via a network,
The behavior scenario determination unit determines the behavior scenario of the robot based on the robot log information and the external server log information.
A robot control system characterized by that. The robot control system according to claim 1,
A setting unit for setting an objective or a goal to be achieved by the operation of the robot;
The behavior scenario determination unit determines the behavior scenario for achieving the purpose or goal set by the setting unit;
A robot control system characterized by that. The robot control system according to claim 5,
The log collection unit determines whether or not the purpose or target set by the setting unit matches the purpose or target obtained by analyzing the robot log information, and determines that they match. Collecting the robot log information,
A robot control system characterized by that. The robot control system according to claim 1,
An on-site model generation unit that generates persona information indicating a virtual person image providing the service and an on-site model indicating the state of an environment in which the service is provided based on log information collected by the log collection unit; Prepared,
The behavior scenario determination unit determines a behavior scenario of the robot based on the robot log information, the persona information generated by the site model generation unit, and the site model.
A robot control system characterized by that. The robot control system according to claim 7,
The behavior scenario update unit determines whether or not the correlation between the site model determined last time and the site model determined this time is more than a certain value, and when determining that the correlation between both is more than a certain value, Updating the behavior scenario;
A robot control system characterized by that. The robot control system according to claim 7,
When determining whether or not the correlation between the field model created last time by the field model creation unit and the field model created this time is a certain level or more, and when determining that the correlation between both is a certain level or more A site model update unit for updating the behavior scenario,
The behavior scenario determination unit determines the behavior scenario of the robot based on the robot log information and the persona information and the site model updated by the site model update unit.
A robot control system characterized by that. The robot control system according to claim 7,
The log collection unit determines whether or not there is a difference between the robot log information collected this time and the robot log information collected last time. Sending robot log information to the on-site model generator,
A robot control system characterized by that. The robot control system according to claim 1,
The log collecting unit includes an identifier for identifying a purpose or a target obtained by analyzing the robot log information when collecting robot log information from the robot, and a device used for providing the service. Create log data with the log information added to the identifier for identification,
The behavior scenario determination unit determines the behavior scenario based on the log data created by the log collection unit;
A robot control system characterized by that. A robot control method for controlling the behavior of a robot that performs autonomous operation and realizes communication with a user,
A log collection unit collects robot log information from the robot at a service base where the robot provides a service based on an action scenario,
An action scenario determination unit determines an action scenario of the robot based on the robot log information collected by the log collection unit,
The behavior scenario update unit updates the behavior scenario determined by the behavior scenario determination unit, and transmits the updated behavior scenario to the robot.
A robot control method characterized by the above.

Images