JP6989480B2 - Guidance robot system and guidance robot control method - Google Patents

Description

The present invention relates to a guidance robot system and a guidance robot control method.

In recent years, guidance robots have been developed in which guidance robots are installed in facilities such as shops and offices to guide facility users to destinations and the like. For example, it is conceivable to place a guide robot near the entrance of a shopping mall and guide the user to the target store by having a conversation between the user who has passed the entrance of the shopping mall and the guide robot. There is.

When the guide robot talks with the user, it talks according to a scenario prepared in advance. For example, when a user approaches a guidance robot, the guidance robot first says "Welcome. What are you looking for?" And sells the desired product in response to the response from the user. A scenario that guides the store is conceivable.
The content that the guidance robot speaks in the conversation with the user is selected from those registered in advance as a scenario according to the user's question and situation at that time.

In Patent Document 1, when creating an electronic publication containing text data, the input data is divided into word units and collated with the terms registered in the public order and morals conformity dictionary. A system for determining whether it is against public order and morals is described.

Japanese Unexamined Patent Publication No. 8-153106

When registering the scenario of the utterance content in the guidance robot, it is necessary to prevent the prohibited words that are offensive to public order and morals from being registered. In the case of processing in which prohibited words that are offensive to public order and morals are not registered in the utterance content, for example, the technique described in Patent Document 1 may be applied. However, in the case of a guidance robot, it is necessary not only to exclude such prohibited terms that are offensive to public order and morals from the utterance sentences included in the scenario, but also to set more advanced prohibited terms as shown below.

That is, in the case of a guidance robot installed in a shopping mall, the scenarios in which the guidance robot talks are prepared by the robot provider, the scenario prepared by the shopping mall operator, and the individual tenants (shops). It can be considered a scenario.
The utterances set in these scenarios are not only for checking prohibited words that are offensive to public order and morals, but also for various scenarios such as whether each scenario meets the policy of the shopping mall as a whole and whether there is any conflict between tenants. It is necessary to check from various points of view.

However, in the creation and checking of the scenario of the conventional guidance robot, it is basically assumed that the robot is executed by one control terminal, and the utterance content when there are multiple operators as described above. Advanced checks were not considered.

An object of the present invention is to provide a guidance robot system and a guidance robot control method capable of checking an appropriate utterance content according to the operation status of the guidance robot.

In order to solve the above problems, for example, the configuration described in the claims is adopted.
The present application includes a plurality of means for solving the above problems. For example, the guidance robot system of the present invention includes a guidance robot that speaks to surrounding people according to a scenario, and a guidance robot. It is equipped with a robot control device for controlling the above.
Here, the robot control device has a scenario storage unit that stores a scenario including the content spoken by the guiding robot, a prohibited term storage unit that stores prohibited terms that are prohibited from being included in the scenario, and a rule for executing the scenario. The rule storage unit to be stored, the registration unit to register new prohibited words or rules in the prohibited term storage unit and the rule storage unit by the operation of the editor, the prohibited words stored in the prohibited term storage unit, and the rule storage. It is provided with a notification unit for notifying and confirming a specific preset editor about the rules stored in the unit.
Then, the notification unit confirms the newly registered prohibited words or rules to one or a plurality of specific editors determined for each editor who registered the prohibited words or rules in the registration unit. Was notified.

According to the present invention, when a prohibited term or a rule is newly registered in a scenario of the content spoken by the guiding robot, a specific editor determined for each editor who performed the registration work is notified and confirmed. Then, you will be able to set an appropriate scenario according to the actual operation.
Issues, configurations and effects other than those described above will be clarified by the description of the following embodiments.

It is a block diagram which shows the configuration example of the whole system by one Embodiment of this invention. It is a figure which shows the arrangement example of the guide robot in a facility by one Embodiment of this invention. It is a block diagram which shows the structural example of the guidance robot according to the example of one Embodiment of this invention. It is a block diagram which shows the structural example of the robot control apparatus according to one Embodiment of this invention. It is a block diagram which shows the configuration example of the robot management server by the example of one Embodiment of this invention. It is a figure which shows the application scope of the prohibited term and the rule by one Embodiment of this invention. It is a figure which shows the setting example of the prohibited term by one Embodiment of this invention. It is a figure which shows the setting example of the rule of the utterance content by the example of one Embodiment of this invention. It is a flowchart which shows the example of the confirmation processing of the utterance content by the example of one Embodiment of this invention. It is a flowchart which shows the example of the scenario selection process by one Embodiment of this invention.

Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to FIGS. 1 to 10.

[1. Overall system configuration]
FIG. 1 shows a configuration example of the entire guidance robot system of the present embodiment.
The guidance robot system 1 is composed of a robot management server 2, a robot control device 3, and a guidance robot 4.
The guidance robot 4 and the robot control device 3 are connected by wireless communication and are arranged in a building 5 which is a facility that provides guidance services. The guidance robot 4 can move (self-propell) in the building 5.

The guidance robot 4 receives a control command from the robot control device 3, and gives the user who uses the building 5 the facilities in the building, the location of the tenant, the products and services provided by the tenant, and the like (a person around the guidance robot 4). We will provide a guidance service that introduces voices to (face-to-face). In the present specification, a person (guided person) who talks (dialogues) with the voice output by the guidance robot 4 is referred to as a user.

In FIG. 1, one robot control device 3 and one guide robot 4 connected to the robot control device 3 are prepared, but a plurality of guide robots 4 are arranged depending on the scale and type of the building 5. Services may be provided. When a plurality of guidance robots 4 are arranged, one robot control device 3 may control the plurality of guidance robots 4 in a unified manner, but each guidance robot 4 may be individually controlled by the robot control device 3. It may be configured to prepare.
When a plurality of guidance robots 4 are installed, it is possible to change the service content for each guidance robot 4.

The robot control device 3 installed in the building 5 is connected to the robot management server 2 via the network 6. In the embodiment of the present embodiment, one robot management server 2 manages one building 5, but it is also possible that one robot management server 2 manages a plurality of buildings 5.
The robot management server 2 manages which guide robot 4 is arranged in which building 5. In addition, the robot management server 2 manages whether each guidance robot 4 is operating normally and whether maintenance is required, and if necessary, substitute robots and parts to the building 5 together with the technician in charge. Perform management processing to send.

Depending on the scale of the guidance robot system 1, the robot management server 2 and the robot control device 3 may be integrated as one device. Further, the guidance robot 4 itself may have a function as a robot control device 3 or a robot management server 2.

FIG. 2 shows an example of operating the guide robot 4 in the building 5.
Here, the building 5 is a facility like a shopping mall in which a plurality of tenants (shops) 7a, 7b, 7c, ... Are arranged. For example, in the building 5, the first tenant 7a, the second tenant 7b, and the third tenant 7c are arranged from the side closer to the doorway 5a. Each tenant 7a, 7b, 7c, ... Is a store selling different types of products, a restaurant, or the like.

The guidance robot 4 stands by, for example, in the vicinity of the entrance / exit 5a of the building 5, moves (self-propelled) so as to pass in front of each tenant 7a, 7b, 7c, ... Guide the users of the mall. However, the traveling route M is an example, and when actually moving, the traveling route M moves on the optimum route avoiding people and obstacles.
Further, the guidance robot 4 stops at various positions such as a standby position m1 near the doorway 5a, a standby position m2 in front of the first tenant 7a, and a standby position m3 in front of the second tenant 7b, and at each position. , You can guide the inside of the shopping mall while talking with the user.

[2. Guidance robot configuration]
FIG. 3 shows a configuration example of the guidance robot 4.
The guidance robot 4 includes a CPU (Central Processing Unit) 10, a storage unit 11, an input / output unit 12, and a communication interface 13.
By reading and executing the software module stored in the storage unit 11, the CPU 10 reads and executes the movement of each part (face, arm, leg, etc.) of the guidance robot 4, which is each function of the guidance robot 4, and with the user. Realize the conversation.

The storage unit 11 stores a drive control unit 20, a conversation control unit 21, and a software module and data for executing processing as an input / output unit 22.
The drive control unit 20 is connected to the CPU 10 and controls a drive mechanism 35 described later.
The conversation control unit 21 controls the conversation with the user by the guidance robot 4.
The input / output unit 22 inputs / outputs data to / from the input / output unit 12.

The input / output unit 12 includes a camera 30, a microphone 31, a gyro sensor 32, a range sensor 33, a speaker 34, and a drive mechanism 35.
The image taken by the camera 30 and the output signal of the range sensor 33 are used to recognize users and obstacles around the guidance robot 4.
The microphone 31 and the speaker 34 are used to have a conversation with the user. That is, the microphone 31 acquires the voice of the user, and the speaker 34 utters a conversational sentence.

The gyro sensor 32 detects the body of the guidance robot 4.
The drive mechanism 35 performs a process of moving the body of the guide robot 4. The movement of the guide robot 4 described with reference to FIG. 2 is performed by the drive mechanism 35 under the control of the drive control unit 20.
The communication interface 13 wirelessly communicates with the robot control device 3.

The guidance robot 4 acquires an image with the camera 30 and acquires a voice from the user with the microphone 31. Then, the acquired image and sound are transmitted to the robot control device 3. Then, the guide robot 4 moves in the building 5 by the drive mechanism 35 based on the instruction from the robot control device 3. Further, the guidance robot 4 detects an obstacle based on the image of the camera 30 and the output value of the range sensor 33 while moving, and the drive control unit 20 autonomously stops the movement or avoids the obstacle. Or something.

[3. Robot control device configuration]
FIG. 4 shows a configuration example of the robot control device 3.
The robot control device 3 includes a CPU 40 that performs various processes, a storage unit 41 that stores data such as various software modules and tables, and a communication interface 42 that can communicate with the guidance robot 4 and the robot management server 2. The communication interface 42 is also used for communication with a terminal device operated by an editor, which will be described later. Further, when the robot control device 3 acquires information such as the current weather, the communication interface 42 acquires necessary information via the Internet or the like.

Each function of the robot control device 3 is realized by the CPU 40 reading and executing the program stored in the storage unit 41. That is, the storage unit 41 includes an input / output data processing unit 50, a movement instruction unit 51, a face-to-face detection unit 52, a voice recognition unit 53, an utterance content confirmation unit 54, a scenario storage unit 55, and a scenario processing unit 56. The functions of units 50 to 56 are realized by software executed under the control of the CPU 40.

The utterance content confirmation unit 54 includes a prohibited term / rule registration unit 60, a prohibited term storage unit 61, a rule storage unit 62, a part of speech decomposition processing unit 63, a prohibited term confirmation unit 64, a rule confirmation unit 65, and an utterance content confirmation processing unit 66. It includes an utterance content registration unit 67, a confirmation content notification unit 68, and a log storage unit 69.

The input / output data processing unit 50 processes the data received from the guidance robot 4. When the guidance robot 4 guides the user, the movement instruction unit 51 gives a movement instruction such as instructing the guide destination to move the guide robot 4.
The face-to-face detection unit 52 detects whether or not the user is in a face-to-face state based on the image information obtained by the camera 30 of the guidance robot 4 and the obstacle information from the range sensor 33. The guidance service is continued while the face-to-face state is detected, and when the user leaves the robot and the face-to-face detection is stopped, the guidance is stopped.
The voice recognition unit 53 processes the user's voice data acquired by the microphone 31 of the guidance robot 4 to perform voice recognition.

In the scenario storage unit 55, what kind of utterance or action is performed when the robot detects face-to-face, what kind of utterance is returned when the guidance robot 4 performs voice recognition, and where the guidance robot 4 is moved. Scenarios such as utterances are stored. The scenario is stored in the scenario storage unit 55 by the scenario storage process by the utterance content confirmation unit 54.
The scenario stored by the scenario storage unit 55 includes the utterance content described later. The scenario processing unit 56 controls the utterance and operation of the guidance robot 4 based on the scenario stored in the scenario storage unit 55.

There are a plurality of workers who perform editing work for registering prohibited words and rules using the prohibited words / rule registration unit 60. That is, there are a worker on the robot provider side, a worker on the shopping mall operator side, and a worker for each tenant. Hereinafter, these workers will be referred to as "editors". The robot provider is a manufacturer that manufactures the guidance robot 4 or a business that rents out the guidance robot 4.
Each editor individually registers rules such as prohibited terms and the utterance time of the product based on the usage pattern of each guidance robot 4. The registration of prohibited terms and rules by each editor is executed by accessing the robot control device 3 via a LAN or the Internet by a terminal device (not shown) possessed by each editor.

Based on the instructions of the editor, the prohibited term / rule registration unit 60 performs a prohibited term storage process for registering the prohibited term in the prohibited term storage unit 61, and performs a rule storage process for registering the rule in the rule storage unit 62.
The prohibited term confirmation unit 64 checks the utterance content of the guidance robot 4 based on the prohibited term stored in the prohibited term storage unit 61, and detects the prohibited term included in the utterance content and its position.

The rule confirmation unit 65 checks the utterance content of the robot based on the rules stored in the rule storage unit 62, compares the situation in which the utterance content is executed with the utterance status described in the rule, and compares the current situation. Prints an error when is a rule-restricted situation. The details of the situation in which the utterance content is executed will be described later, but there are, for example, the time, the weather, the position of the guidance robot 4, and the like.

The utterance content registration unit 67 registers the utterance content of the guidance robot 4 based on the instruction of the editor.
The utterance content confirmation processing unit 66 checks prohibited terms and rules regarding the utterance content of the scenario (or the stored scenario) stored in the scenario storage unit 55.
The confirmation content notification unit 68 notifies the editor of the check result of the utterance content confirmation processing unit 66.

Here, in the case of the present embodiment, when a plurality of editors exist as editors and registration is performed by the prohibited term / rule registration unit 60 or the utterance content registration unit 67, the registration is performed. The check result is notified by the confirmation content notification unit 68 to a specific editor determined for each person.
The details of the process of selecting this editor will be described later.

[4. Robot management server configuration]
FIG. 5 shows an example of the configuration of the robot management server 2.
The robot management server 2 includes a CPU 70 that performs various processes, a storage unit 71 that stores data such as software modules and tables of the robot arrangement management unit 73, and a communication interface 72 that can communicate with the robot control device 3. Be prepared.

The functions required for the robot management server 2 are realized by the CPU 70 reading and executing the software module (robot arrangement management unit 73, etc.) stored in the storage unit 71.
The robot placement management unit 73 monitors which guidance robot 4 is placed in which building and can provide the service normally, and if the service cannot be provided, the necessary guidance robot 4 maintenance service can be provided. , Give instructions to dispatch the body, parts, and engineers of the guidance robot 4.

[5. Editor's scope]
FIG. 6 shows the range of editors who notify the registered prohibited terms and rules when the editor of each trader who uses the guidance robot system 1 of the present embodiment registers the prohibited terms and rules. ..
Here, among the plurality of tenants in the shopping mall, the first tenant 7a and the second tenant 7b use the guidance robot 4. Therefore, as shown in FIG. 6, the editors include the editor 81 on the robot provider side, the editor 82 on the shopping mall side, the editor 83 on the first tenant side, and the editor 84 on the second tenant side. exist.

When these editors 81-84 perform editing work on the terminal device, the robot control connected to the terminal device so that data communication is possible by logging in with the identification code and password assigned to each editor 81-84. The device 3 recognizes each editor 81-84 and permits editing.
When the robot control device 3 recognizes each editor 81-84, if a new prohibited term or rule is registered, the registered content is the range 80a, 80b, 80c shown by the robot control device 3 in FIG. Notify other editors within.

That is, when the editor 81 on the robot provider side newly registers the prohibited terms and rules, the robot control device 3 notifies the editor 82 on the shopping mall side of the change, as shown as the range 80a. Further, even when the editor 82 on the shopping mall side newly registers prohibited terms or rules, the robot control device 3 notifies the editor 81 on the robot provider side of the change, as shown as the range 80a.

When the editor 83 of the first tenant 7a newly registers prohibited terms and rules, the robot control device 3 has an editor 82 on the shopping mall side and an editor 81 on the robot provider side, as shown as the range 80b. Notify the change to.
When the editor 84 of the second tenant 7b newly registers prohibited terms and rules, the robot control device 3 has an editor 82 on the shopping mall side and an editor 81 on the robot provider side, as shown as the range 80c. Notify the change to.

Therefore, when the editor 83 of the first tenant 7a newly registers the prohibited words and rules, the editor 84 of the second tenant 7b is not notified, and conversely, the prohibited words and the prohibited words by the editor 84 of the second tenant 7b At the time of new registration of the rule, the editor 83 of the first tenant 7a is not notified.

[6. Examples of prohibited terms and rules]
FIG. 7 shows an example of a table of prohibited terms stored in the prohibited term storage unit 61.
The prohibited term table 100 includes a registration number area 101, a registrant area 102, a prohibited term area 103, a prohibited term reason area 104, and a prohibited term update time area 105.

The prohibited terms of the scenario can be registered by each editor, and which editor has registered is recorded in the registrant area 102 of FIG.

The prohibited terms registered by the editor 81 on the robot provider side are common to any business operator, such as insulting words and other words that are offensive to public order and morals, as shown in the reason area 104. Is an unfavorable term.
The prohibited terms registered by the editor 82 on the shopping mall side are prohibited terms peculiar to the shopping mall, for example, words that do not fit the image of the shopping mall.
The prohibited terms registered by the editors 83 and 84 on each tenant side include, for example, product names that have a rivalry with the products sold by each tenant.

It should be noted that the area 104 of the reason for being a prohibited term is unnecessary information for controlling the guide robot 4, and may be omitted. However, by providing the area 104 of the reason for the prohibited term shown in FIG. 7 and inputting the reason for the prohibited term, the area 104 for this reason is appropriate for the prohibited term newly registered by another editor. It is considered to be one of the judgment materials when confirming whether or not it is.

FIG. 8 shows an example of a table of rules stored in the rule storage unit 62.
In the rule table 200, the registration number area 201, the registrant area 202, the product area 203, the time area 204, the place area 205, the prohibited place area 206, the weather area 207, the event area 208, and the like. The reason area 209 and the rule update time area 210 are included.

The prohibited terms of the scenario can be registered by each editor, and which editor has registered is recorded in the registrant area 202 of FIG.
In the product area 203, a product name recommended to the conversation partner as the scenario progresses is registered. The content registered in the area 203 of this product is the word spoken by the guidance robot 4. Since the product area 203 is a word spoken by the guidance robot 4, a product name other than the product name may be registered. For example, in the example shown in FIG. 8, a case where words such as "rain outside" and "mall event" are registered is shown.

In the time area 204, information about the time when the corresponding rule is applied is registered.
For example, "June to August" is registered as the time to apply the rule of registration number 1, and guidance to promote the registered phrase "Summer One Piece A" by the guidance robot 4 will be performed only from June to August. become. If there is no input in the time area 204, the rules for that item apply to all times.

Information about the location of the guidance robot 4 to which the corresponding rule is applied is registered in the location area 205.
For example, regarding the rules of registration numbers 1, 2, and 6, when "in front of the first tenant" is registered as a place and the guidance robot 4 comes in front of the first tenant (position m2 shown in FIG. 2), guidance is provided. Guidance to advertise in the words (registered words) of the robot 4 is performed. Guidance according to the rules of registration numbers 1, 2 and 6 is not executed except in front of the first tenant. If there is no input in area 205 of the place, the rule of the corresponding item applies to all places.

In the prohibited place area 206, information about the place of the guidance robot 4 for which the application of the corresponding rule is prohibited is registered.
For example, regarding the rule of registration number 4, "in front of the first tenant" is registered as a prohibited place, and in front of the first tenant, the guidance robot 4 does not provide guidance to advertise the word (registered word). That is, the guidance by the guidance robot 4 is performed at a place other than in front of the first tenant. If there is no input in area 206 of the prohibited place, the rule of the corresponding item applies to all places.

Information about the current weather to which the applicable rule is applied is registered in the area 207 of the weather.
For example, for the rules of registration numbers 5 and 6, "rain" is registered as the weather, and when the current weather is rain, the registered words can be spoken as the scenario progresses. When processing this weather, the robot control device 3 needs to acquire information on the current weather at any time.

Information about the event period to which the corresponding rule is applied is registered in the event area 208.
For example, for the rule of registration number 8, a Valentine's Day promotion period is set, and promotion (guidance) in the words of the corresponding rule is performed within the Valentine's Day promotion period separately set for the entire shopping mall.

In the reason area 209, the reason for setting the rule is input. The area 209 for this reason is unnecessary information for controlling the guide robot 4 and may be omitted as in the area 104 for the reason for the prohibited term. The editor will be one of the judgment factors of the rule.

In addition, an area of information indicating that confirmation by another editor has been completed is added to the items of each registration number in the table 100 of prohibited terms shown in FIG. 7 and the table 200 of the rules shown in FIG. , It may be possible to know whether it has been confirmed or not.
Alternatively, the prohibited terms and rules registered in the prohibited term table 100 shown in FIG. 7 and the rule table 200 shown in FIG. 8 have been confirmed by other editors within the range shown in FIG. 6 and consented. Only the information provided may be posted. In this case, for prohibited terms and rules before confirmation by other editors, a dedicated table for unapproved information is prepared, and after approval is completed, the prohibited terms table 100 shown in FIG. 7 and the figure are shown. It is registered in the table 200 of the rule shown in 8.

[7. Example of processing when registering utterance content]
FIG. 9 is a flowchart illustrating a processing example when the utterance content is registered in the utterance content confirmation processing unit 66 of the robot control device 3.
First, the utterance content confirmation processing unit 66 accepts the utterance content registration input by the terminal device of any of the editors (step S11). Here, numbers 1 to N (N is an arbitrary integer) are assigned to the utterance contents. The processing is performed for each of 1 to N.

Then, the utterance content confirmation processing unit 66 determines whether or not the utterance content number (n) currently being processed is equal to or less than the maximum value N (step S12). When it is determined in step S12 that the maximum value is N or less (YES in step S12), the part of speech decomposition processing unit 63 decomposes the part of speech of the utterance content (step S13). Then, the prohibited term confirmation unit 64 confirms whether the prohibited term is included in the utterance content (step S14), and the utterance content confirmation processing unit 66 stores the confirmation result log (step S15). Here, when a prohibited term is included, the utterance content confirmation processing unit 66 becomes the included prohibited term, its utterance content, the position of the prohibited term in the utterance content, and the registrant / prohibited term of the prohibited term. The reason, the update time of the prohibited utterance, etc. are stored in the log storage unit 69. Further, when the prohibited term is not included, the utterance content confirmation processing unit 66 stores the result of normality in the log storage unit 69.

After that, the prohibited term confirmation unit 64 confirms the rule of the utterance content (step S16), and the utterance content confirmation processing unit 66 stores the confirmation result log (step S17). Here, the product, the content of the utterance, the registrant of the rule, the reason for the rule, the update time of the rule, and the like are stored in the log storage unit 69. If there is no rule regarding the product, or if the utterance time is according to the rule, the result that it is normal is stored in the log storage unit 69.
After the processing up to this point is performed, the utterance content confirmation processing unit 66 returns to the acceptance processing in step S11.

If it is determined in step S12 that the maximum value is not N or less (NO in step S12), the confirmation content notification unit 68 notifies the terminal device of the editor who performed the utterance registration work of the registration content (step). S21). After this notification, the utterance content confirmation processing unit 66 determines whether or not the editor who performed the utterance registration work accepts the confirmation content (step S22). Here, if the confirmation content is not accepted (NO in step S22), the editor restarts from the acceptance process in step S11, corrects the utterance content, and the like.

If the confirmation content is accepted in step S22, the utterance content confirmation processing unit 66 determines whether or not there is a registration operation by the editor of the confirmed content (step S23). Here, when the confirmation content is not registered (NO in step S23), the process ends. If there is a registration operation in step S23 (YES in step S23), the utterance content confirmation processing unit 66 stores the registered scenario or the rules in the scenario in the scenario storage unit 55 or the rule storage unit 62 (step). S24). Then, the utterance content confirmation processing unit 66 stores the registration result log (step S25).
In step S24, when the registered content relates to the entire scenario, it is stored as a new scenario in the scenario storage unit 55, and when it is a rule in the scenario (that is, a rule as shown in FIG. 8), the rule storage unit Store as a new rule in 62.

After that, the utterance content confirmation processing unit 66 determines whether or not notification of the registered content is necessary (step S26). Here, if the notification is not required (NO in step S26), the registration process is terminated. If the notification of the registered contents is not required, for example, a minor scenario modification performed by the robot provider or the shopping mall operator is equivalent. In addition, as a judgment of this step S26, as a general rule, notification is required for all updates during normal times, and it is determined that notification is not required only for special updates such as updates during robot maintenance. May be good.

Further, when it is determined in step S26 that notification is necessary (YES in step S26), the utterance content confirmation processing unit 66 refers to the terminal device of another editor within a predetermined range. The registered log is sent by e-mail or the like (step S27), and the process of registering the utterance content is completed. When transmitting the log in step S27, not only the words set as the rules but also the details (time, weather, place, etc.) of each item shown in FIG. 8 are transmitted.

In addition, even when the utterance content is checked, the utterance content registered in the scenario storage unit 55 or the like is periodically checked even when the registration is not performed, and the product and the utterance content whose current time is different from the utterance time are checked. You may try to do it.

As explained above, when there is a new registration for a scenario or rule, other editors who need to confirm the scenario or rule will be notified by e-mail, etc., and the editor who received the notification will be notified. You can check the registered contents. Therefore, all operators of the guidance robot system can confirm whether or not the scenario and rules are appropriate. For example, even if the rule is registered by the robot provider, when the shopping mall operator determines that the application of the rule is not appropriate, the corresponding rule can be deleted or modified. In addition, the shopping mall operator or the robot provider can determine whether the rules registered by the operator of each tenant are disadvantageous to other tenants.
Therefore, as shown in FIG. 2, the guidance robot 4 can be appropriately operated in a large-scale facility such as a shopping mall where a large number of tenants have moved in.

Also, when notifying other editors of a rule, information on the detailed conditions set as the rule, such as when to apply the rule, weather, and location conditions, is notified to other editors. At the time of confirmation in, you can also judge those detailed conditions. Furthermore, by notifying the reason for setting the rules and prohibited words based on the input information in the tables 100 and 200, it is possible to judge whether the registered rules and prohibited words are appropriate. become.

If a scenario or rule is registered, the registered content may not be valid until the operation of accepting the registered content is performed from the destination after the log is transmitted in step S27.
Further, in the flowchart of FIG. 9, the process of registering the scenario and the rule has been described, but the registration of the prohibited terms shown in the table 100 of FIG. 7 can also be executed by the same procedure. In this case as well, when a new prohibited term is registered, another editor who needs to confirm the prohibited term is notified by e-mail, etc., and the editor who receives the notification is the prohibited term. You will be able to check the registered contents of.

[8. Processing example when selecting a scenario]
FIG. 10 is a flowchart showing an example of a scenario selection process when the robot control device 3 has a conversation with a user who has approached the guide robot 4 by an utterance in the guide robot 4.
First, the face-to-face detection unit 52 of the robot control device 3 determines whether or not there is a user (guidance target person) facing the guidance robot 4 (step S31). In step S31, if there is no user facing the guide robot 4 (NO in step S31), the robot waits until the user's face-to-face is detected.

Further, when the user's face-to-face is detected in step S31 (YES in step S31), the robot control device 3 confirms the current position of the guidance robot 4 (step S32), and determines the current time (date and time) and the weather. Confirm (step S33). Then, the scenario processing unit 56 selects one scenario that seems to be the most appropriate according to the current position, time, and weather (step S34). When selecting this scenario, it may be determined other than the matters confirmed in step S33. For example, the scenario processing unit 56 may estimate the gender and age of the face-to-face user from a camera image or the like and add it to the requirements for scenario selection.

Then, the utterance process is performed based on the scenario selected in step S34 (step S35). After this utterance process is performed, the process returns to the determination in step S31.
When the scenario includes guidance by traveling (moving) the robot, the robot control device 3 causes the guiding robot 4 to travel after (or in parallel with) the utterance in step S35.

[9. Modification example]
The present invention is not limited to the above-described embodiment, and includes various modifications.
For example, in each of the above-described embodiments, it is applied to a self-propelled guide robot, but it may be applied to a stationary guide robot that does not self-propell.

Further, in the above-described embodiment, the case where the guide robot 4 is installed in the shopping mall has been described, but the present invention can also be applied to the case where the guide robot 4 is installed in various other facilities.

Further, the above-described embodiment is described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to the one including all the described configurations. Further, in the configuration diagrams and functional block diagrams of FIGS. 1, 3, 4, and 5, only the control lines and information lines considered to be necessary for explanation are shown, and not all the control lines and the control lines in the product are shown. It does not always show the information line. In practice, it can be considered that almost all configurations are interconnected. Further, in the flowcharts shown in FIGS. 9 and 10, the execution order of some processing steps is changed or some processing steps are executed at the same time within a range that does not affect the processing results of the embodiment. You may do it.

1 ... Guidance robot system, 2 ... Robot management server, 3 ... Robot control device, 4 ... Guidance robot, 5 ... Building, 5a ... Doorway, 6 ... Network, 7a ... 1st tenant, 6b ... 2nd tenant, 6c ... No. 3 tenants, 10 ... CPU, 11 ... storage unit, 12 ... input / output unit, 13 ... communication interface, 20 ... drive control unit, 21 ... conversation control unit, 22 ... input / output unit, 30 ... camera, 31 ... microphone, 32 ... Gyro sensor, 33 ... Range sensor, 34 ... Speaker, 35 ... Drive mechanism unit, 40 ... CPU, 41 ... Storage unit, 42 ... Communication interface, 50 ... Input / output data processing unit, 51 ... Movement instruction unit, 52 ... Face-to-face detection unit, 53 ... voice recognition unit, 54 ... speech content confirmation unit, 55 ... scenario storage unit, 56 ... scenario processing unit, 60 ... prohibited term / rule registration unit, 61 ... prohibited term storage unit, 62 ... rule storage unit , 63 ... part-of-speech decomposition processing unit, 64 ... prohibited term confirmation unit, 65 ... rule confirmation unit, 66 ... speech content confirmation processing unit, 67 ... speech content registration unit, 68 ... confirmation content notification unit, 69 ... log storage unit, 70 ... CPU, 71 ... Storage unit, 72 ... Communication interface, 73 ... Robot layout management unit, 81 ... Robot provider side editor, 82 ... Shopping mall side editor, 83 ... First tenant side editor, 84 … Editor on the second tenant side

Claims (5)

A guide robot that speaks to people around you according to the scenario,
A guidance robot system including a robot control device for controlling the guidance robot.
The robot control device is
A scenario storage unit that stores a scenario including the content spoken by the guidance robot, and a scenario storage unit.
A prohibited term storage unit that stores prohibited terms that are prohibited from being included in the scenario,
A rule storage unit that stores the rules that execute the scenario,
A registration unit that registers new prohibited terms or rules in the prohibited term storage unit and the rule storage unit by the operation of the editor, and
It is provided with a forbidden term stored in the prohibited term storage unit and a notification unit for notifying a specific preset editor to confirm the rule stored in the rule storage unit.
The notification unit confirms the newly registered prohibited words or rules to one or more specific editors determined for each editor who registered the prohibited words or rules in the registration unit. Guidance robot system to notify you. The registration unit is a confirmation by the specific user notified by the notification unit, and when there is a response permitting registration, the registration unit is a new prohibited term or rule registered in the prohibited term storage unit or the rule storage unit. The guidance robot system according to claim 1. As a rule stored by the rule storage unit, it is possible to set the time or weather when the scenario can be applied.
The guidance robot system according to claim 2, wherein the notification unit notifies information on the time or weather when a new prohibited term or rule is applied. The guidance robot is a self-propelled robot.
As a rule stored by the rule storage unit, it is possible to set whether or not a predetermined scenario can be applied according to the position of the guide robot.
The guidance robot system according to claim 2, wherein the notification unit notifies information on a place to which a new prohibited term or rule is applied. It is a guidance robot control method that controls a guidance robot that speaks to surrounding people according to a scenario.
A scenario storage process for storing a scenario including the content spoken by the guidance robot, and
Forbidden term storage processing that stores prohibited terms that are prohibited from being included in the scenario,
Rule storage processing that stores the rules that execute the scenario, and
In the registration process of registering a new prohibited term or rule by the operation of the editor in the execution of the prohibited term storage process or the rule storage process,
The prohibited words stored by the prohibited word storage process and the rules stored by the rule storage process are notified to a specific preset editor for confirmation, and when the notification is given, the registration is performed. Notification processing that notifies one or more specific editors determined for each editor who registered the prohibited words or rules in the process to confirm the newly registered prohibited words or rules. Guidance robot control method including.

Images