JP2021144086A - Agent system and computer program - Google Patents

Abstract

To increase convenience of an agent system for carrying out transfer of information with a user through a dialog.SOLUTION: A first agent device 11 has portability and provides a first agent 31 for acquiring first information I1 through a first dialog with a user 20. A second agent device 12 is composed to be loaded to a vehicle 40 and provides a second agent 32 for acquiring second information I2 through a second dialog with the user 20. A memory device 14 stores the first information I1 and the second information I2. A control device 13 executes at least one of a process to have a first phrase P1 generated based on at least one of the first information I1 and the second information I2 stored to the memory device 14 to the first agent 31 in the first dialog, and a process to output a first phrase P2 generated based on at least one of the first information I1 and the second information I2 stored in the memory device 14 to the second agent 32 in the second dialog.SELECTED DRAWING: Figure 1

Description

The present invention relates to an agent system that exchanges information with a user through dialogue. The present invention also relates to a computer program executed by the agent system.

Patent Document 1 discloses an example of an agent system that exchanges information with a user through dialogue.

Japanese Unexamined Patent Publication No. 2016-076007

An object of the present invention is to enhance the convenience of an agent system that exchanges information with a user through dialogue.

One aspect of achieving the above objectives is an agent system.
A first agent device that is configured to be portable by a user and provides a first agent that acquires first information from the user through the first dialogue with the user.
A second agent device that is configured to be mounted on a mobile body and provides a second agent that acquires second information from the user through a second dialogue with the user.
A storage device that stores the first information and the second information,
Based on at least one of the first information and the second information stored in the storage device, the first phrase output from the first agent in the first dialogue and the second agent in the second dialogue. A control device that generates at least one of the output second phrases,
It has.

One aspect of achieving the above objectives includes a first agent device that is portable, a second agent device that is configured to be mounted on a moving body, a storage device, and a control device. A computer program run by an agent system
By being executed
The first agent device is provided with a first agent that acquires the first information through the first dialogue with the user.
The second agent device is provided with a second agent that acquires the second information through the second dialogue with the user.
The first information and the second information are stored in the storage device, and the first information and the second information are stored in the storage device.
A process of causing the first agent to output a first phrase generated based on at least one of the first information and the second information stored in the storage device in the first dialogue, and storage in the storage device. The control device is made to execute at least one of the process of outputting the second phrase generated based on at least one of the first information and the second information to the second agent in the second dialogue.

According to the above configuration, it is possible to collect information through dialogue with the agent while the user is on board or not on the moving object, and obtains information on the user's interests and preferences. Opportunities increase. In addition, since the phrases in the subsequent dialogues are generated through the sharing of information acquired from both the first agent and the second agent, the amount of information that can be reflected in the contents of the dialogues with the agents is increased. be able to. Therefore, it is possible to enhance the convenience of the agent system that exchanges information with the user through dialogue.

The functional configuration of the agent system according to one embodiment is illustrated. The configuration of the first agent device in FIG. 1 is illustrated. The configuration of the second agent device in FIG. 1 is illustrated. The configuration of the second agent device in FIG. 1 is illustrated.

An example of the embodiment will be described in detail below with reference to the accompanying drawings. FIG. 1 illustrates the functional configuration of the agent system 10 according to the embodiment. The agent system 10 is configured to exchange information through dialogue with the user 20.

As used herein, the term "agent" is meant to include hardware with which user 20 speaks and software for operating the computer as hardware with which user 20 speaks. The "agent" does not necessarily have to have a visible form. For example, an "agent" can establish a dialogue only through voice.

The agent system 10 includes a first agent device 11. The first agent device 11 is configured to be portable by the user 20. The first agent device 11 is configured to provide the first agent 31.

The first agent 31 is configured to establish a first dialogue with the user 20 by returning an appropriate first phrase P1 in response to the utterance of the user 20. The starting point of the first dialogue may be the utterance of the first phrase P1 by the first agent 31. The first agent device 11 is configured to acquire the first information I1 through the first dialogue with the user 20. The information may be acquired based on the content of the information retrieval requested by the user 20 to the first agent 31, or based on the response of the user 20 to the active question from the first agent 31. good.

The agent system 10 includes a second agent device 12. The second agent device 12 is configured to be mounted on the vehicle 40. The second agent device 12 is configured to provide the second agent 32.

The second agent 32 is configured to establish a second dialogue with the user 20 by returning an appropriate second phrase P2 in response to the utterance of the user 20. The starting point of the second dialogue may be the utterance of the second phrase P2 by the second agent 32. The second agent device 12 is configured to acquire the second information I2 through a second dialogue with the user 20. The information may be acquired based on the content of the information retrieval requested by the user 20 to the second agent 32, or based on the user 20's response to the active question from the second agent 32. good.

The agent system 10 includes a control device 13 and a storage device 14. The control device 13 is configured to receive the first information I1 and the second information I2 from the first agent device 11 and the second agent device 12, respectively, and store them in the storage device 14. That is, the storage device 14 is configured to store the first information I1 and the second information I2.

The control device 13 may be mounted on the vehicle 40, or may be provided as an external device capable of communicating with the vehicle 40 via a wireless communication network. In either case, the control device 13 accepts the first information I1 from the first agent device 11 via wireless communication. When the control device 13 is provided as an external processing device, the reception of the second information I2 from the second agent device 12 is also performed via wireless communication.

The storage device 14 may be mounted on the vehicle 40, or may be provided as an external device capable of communicating with the control device 13 via a wireless communication network.

The control device 13 executes a process of causing the first agent 31 to output the first phrase P1 generated based on at least one of the first information I1 and the second information I2 stored in the storage device 14 in the first dialogue. It is configured to be possible. Specifically, when the first dialogue between the user 20 and the first agent 31 is started, the control device 13 has the first information I1 and the second information I2 stored in the storage device 14 as needed. At least one of the above is referenced or read, and the first phrase information PI1 corresponding to the first phrase P1 is generated.

Subsequently, the control device 13 transmits the first phrase information PI1 to the first agent device 11 via wireless communication. The first agent 31 provided by the first agent device 11 outputs the first phrase P1 corresponding to the first phrase information PI1 in the first dialogue.

In addition to or instead, the control device 13 transfers the second phrase P2 generated based on at least one of the first information I1 and the second information I2 stored in the storage device 14 in the second dialogue. It is configured to be able to execute the process of outputting to the agent 32. Specifically, when the second dialogue is started between the user 20 and the second agent 32, the control device 13 has the first information I1 and the second information I2 stored in the storage device 14 as needed. At least one of the above is referenced or read, and the second phrase information PI2 corresponding to the second phrase P2 is generated.

Subsequently, the control device 13 transmits the second phrase information PI2 to the second agent device 12. The second agent 32 provided by the second agent device 12 outputs the second phrase P2 corresponding to the second phrase information PI2 in the second dialogue.

That is, the first information I1 acquired from the user 20 in the first dialogue with the first agent 31 can be reflected in the content of the second dialogue between the user 20 and the second agent 32 thereafter. Alternatively, the second information I2 acquired from the user 20 in the second dialogue with the second agent 32 can be reflected in the content of the first dialogue between the user 20 and the first agent 31 thereafter.

Since the algorithm itself for generating a phrase for dialogue from the collected information is well known, detailed description thereof will be omitted.

For example, the portable first agent device 11 is generally not used while the user 20 is driving the vehicle 40. During this time, information cannot be collected through the first agent 31. On the other hand, while the user 20 is getting off the vehicle 40, the second agent device 12 mounted on the vehicle 40 is not used. During this time, information cannot be collected through the second agent 32. Therefore, in a configuration in which only one of the agent devices is used, the opportunity to acquire information about the interests and preferences of the user 20 is limited, and the information that can be reflected in the content of the dialogue with the agent is also included. Also limited.

According to the above configuration, it is possible to collect information through dialogue while the user 20 is getting on and off the vehicle 40, and is an opportunity to acquire information about the interests and preferences of the user 20. Increases. In addition, since the phrase in the subsequent dialogue is generated through the sharing of the information acquired from both the first agent 31 and the second agent 32, the amount of information that can be reflected in the content of the dialogue with the agent is increased. Can be increased. Therefore, the convenience of the agent system 10 that exchanges information with the user 20 through dialogue can be enhanced.

As illustrated in FIG. 2, the first agent device 11 may include a display unit 111. In this case, the first agent 31 may be an image displayed on the display unit 111. That is, the first agent 31 has a visible form.

According to such a configuration, a general-purpose mobile device such as a smartphone can be used as the first agent device 11. Therefore, it is not necessary to prepare a special device for collecting information while not in the vehicle 40. Further, since the first agent 31 has a visible form, it becomes easy to enhance the feeling of attachment and trust of the user 20 to the first agent 31.

As illustrated in FIG. 3, the second agent device 12 can be arranged in the passenger compartment 41 of the vehicle 40. In this example, the second agent 32 provided by the second agent device 12 has a visible form. More specifically, as is clear from the comparison with FIG. 2, the first agent 31 and the second agent 32 have a form that can be recognized as the same agent. In addition, the second agent 32 has a three-dimensional structure.

In addition, in order to be "a form that can be recognized as the same agent", when an image of a form in which the first agent 31 is viewed from a specific direction is displayed on the display unit 111 of the first agent device 11, the same direction. It is necessary that the form of the second agent 32 viewed from the above side looks substantially the same as the image.

According to such a configuration, the user 20 can be given the feeling of being accompanied by the same agent both inside and outside the vehicle 40. This makes it easier to further increase the feeling of attachment and trust in the agent.

If the first agent 31 and the second agent 32 have a form that can be recognized as the same agent, the second agent 32 is displayed on the display unit 42 installed in the center cluster illustrated in FIG. It may be an image.

However, since the second agent 32 has a three-dimensional structure, it is possible to give the user 20 the feeling that the agent actually accompanies. In particular, by boarding the vehicle 40 and letting go of the portable first agent device 11, it is possible to obtain the feeling that the agent displayed as an image on the display unit 111 appears in the passenger compartment 41, and the ride experience is achieved. It can add value.

As illustrated in FIG. 4, in the second agent 32 in this example, each of the head portion 32a, the left arm portion 32b, and the right arm portion 32c is a movable portion. In this case, the second agent device 12 includes a drive device 121 for driving each movable portion. Examples of the drive device 121 include a motor and an actuator.

The control device 13 can output a control signal CS that controls the operation of the drive device 121 according to the content of the second dialogue. As an example, when the second agent 32 outputs the second phrase P2 having a positive content, the control device 13 controls the drive device 121 so that the head 32a of the second agent 32 rotates in the left-right direction. Output the control signal CS. As another example, when the second agent 32 outputs the second phrase P2 expressing the feeling of joy, the control device 13 controls the drive device 121 so that the second agent 32 raises the left arm portion 32b and the right arm portion 32c. Output the control signal CS.

In addition to or instead of this, the control device 13 may output a control signal CS that controls the operation of the drive device 121 according to the state of the vehicle 40. Examples of the state of the vehicle 40 include speed, acceleration, steering direction, temperature in the vehicle interior 41, and the like. For example, when the vehicle 40 suddenly decelerates, the control device 13 controls the drive device 121 so as to reproduce how the second agent 32 behaves in a balanced manner while swinging the left arm portion 32b and the right arm portion 32c so as not to fall. Output the control signal CS.

With such a configuration, the sense of reality of the second agent 32 is strengthened, and it becomes easier to further enhance the sense of attachment and trust to the agent.

The same operation may be performed on the image of the second agent 32 displayed on the display unit 42. That is, the control device 13 can output a control signal CS that controls the movement of the image of the second agent 32 displayed on the display unit 42 according to at least one of the content of the second dialogue and the state of the vehicle 40.

As illustrated in FIG. 4, the second agent device 12 may include a touch sensor 122. The touch sensor 122 is configured to detect contact with at least one predetermined location in the second agent 32 and output a detection signal DS. The detection signal DS is input to the control device 13. The control device 13 can output a control signal CS that displaces the movable portion of the second agent 32 based on the detection signal DS. For example, when the touch sensor 122 detects that the user 20 has stroked the head 32a of the second agent 32, the control device 13 drives the second agent 32 to raise the left arm portion 32b and the right arm portion 32c. The control signal CS that controls 121 is output.

With such a configuration, it is possible to realize highly realistic communication with the second agent 32, and it becomes easier to further enhance the feeling of attachment and trust to the agent.

The control device 13 may be configured to be able to determine whether the user 20 has boarded the vehicle 40. Specifically, as illustrated in FIG. 4, the control device 13 can be configured to accept boarding information BI indicating that the user 20 has boarded the vehicle 40. The boarding information BI includes a sensor for detecting the opening and closing of the door of the vehicle 40, a seating sensor provided on the seat of the vehicle 40, a motion sensor arranged in the passenger compartment 41, and an engine start switch of the vehicle 40 being operated. Can be detected by a sensor or the like that detects. When the control device 13 receives the boarding information BI, it determines that the user 20 has boarded the vehicle 40, and outputs a control signal CS for activating the second agent device 12.

There is no dialogue between the user 20 and the second agent 32 while the user 20 is not in the vehicle 40. According to the above configuration, the second agent device 12 is in the standby state while the user 20 is not in the vehicle 40, so that unnecessary power consumption can be suppressed. Further, by boarding the vehicle 40 and letting go of the portable first agent device 11, it is as if the first agent 31 displayed as an image on the display unit 111 has moved to the second agent 32 in the passenger compartment 41. It gives you a sense and can add value to your riding experience.

The control device 13 having each of the functions described so far can be realized by a general-purpose microprocessor that operates in cooperation with a general-purpose memory. Examples of general-purpose microprocessors include CPUs, MPUs, and GPUs. A ROM or RAM can be exemplified as a general-purpose memory. In this case, the ROM may store a computer program that executes the above-described processing. The ROM is an example of a storage medium that stores a computer program. The processor specifies at least a part of the computer program stored in the ROM, expands it on the RAM, and performs the above-described processing in cooperation with the RAM. The above computer program may be pre-installed in the general-purpose memory, or may be downloaded from an external server via a communication network and installed in the general-purpose memory. In this case, the external server is an example of a storage medium that stores a computer program.

The control device 13 may be realized by a dedicated integrated circuit capable of executing the above-mentioned computer programs such as a microcontroller, ASIC, and FPGA. In this case, the above computer program is pre-installed in the storage element included in the dedicated integrated circuit. The storage element is an example of a storage medium that stores a computer program. The control device 13 can also be realized by a combination of a general-purpose microprocessor and a dedicated integrated circuit.

The storage device 14 can be realized by a semiconductor memory or a hard disk device. The storage device 14 may be realized by the above-mentioned general-purpose memory or storage element.

The above embodiments are merely examples for facilitating the understanding of the present invention. The configuration according to the above embodiment may be appropriately changed or improved without departing from the spirit of the present invention.

The first information I1 that can be acquired from the user 20 in the first dialogue between the user 20 and the first agent 31 is not limited to that based on the utterance of the user 20. When the first agent device 11 has a configuration for accepting character input, the input characters may also constitute the first information I1. In addition to or instead, if the first agent device 11 has an image recognition function, the gesture (nodding, swinging, etc.) of the user 20 may also constitute the first information I1. That is, the exchange of information made through letters and gestures can also constitute the first dialogue.

The second information I2 that can be acquired from the user 20 in the second dialogue between the user 20 and the second agent 32 is not limited to the one based on the utterance of the user 20. When the second agent device 12 has a configuration for accepting character input, the input characters may also constitute the second information I2. In addition to or instead, if the second agent device 12 has an image recognition function, the gesture (nodding, swinging, etc.) of the user 20 may also constitute the second information I2. That is, the exchange of information made through letters and gestures can also constitute a second dialogue.

The first agent device 11 may be provided as a dedicated mobile device.

The first agent 31 and the second agent 32 may have different forms to the extent that they can be recognized as different agents.

The second agent device 12 can be mounted on a moving body other than the vehicle 40. Examples of mobiles include railroads, aircraft, ships and the like. The moving body does not have to require a driver.

10: Agent system, 11: First agent device, 111: Display unit, 12: Second agent device, 122: Touch sensor, 13: Control device, 14: Storage device, 20: User, 31: First agent, 32 : Second agent, 32a: Head, 32b: Left arm, 32c: Right arm, 40: Vehicle, DS: Detection signal, I1: First information, I2: Second information, P1: First phrase, P2: First Two phrases

Claims (8)

A first agent device that is portable and provides a first agent that acquires the first information through the first dialogue with the user.
A second agent device that is configured to be mounted on a mobile body and provides a second agent that acquires second information through a second dialogue with the user.
A storage device that stores the first information and the second information,
A process of causing the first agent to output a first phrase generated based on at least one of the first information and the second information stored in the storage device in the first dialogue, and storage in the storage device. A control device that executes at least one of a process of outputting the second phrase generated based on at least one of the first information and the second information to the second agent in the second dialogue.
Is equipped with
Agent system. The first agent device has a display unit that displays an image of the first agent.
The agent system according to claim 1. The first agent and the second agent have a form that can be recognized as the same agent.
The agent system according to claim 1 or 2. The second agent has a three-dimensional structure.
The agent system according to any one of claims 1 to 3. The second agent has a movable part and has a movable part.
The control device displaces the movable portion according to at least one of the second dialogue and the state of the moving body.
The agent system according to claim 4. The second agent device includes a sensor that outputs a detection signal corresponding to contact with the second agent.
The control device displaces the movable portion based on the detection signal.
The agent system according to claim 5. The control device activates the second agent device when it determines that the user has boarded the mobile body.
The agent system according to any one of claims 1 to 6. A computer program executed by an agent system including a first agent device having portability, a second agent device configured to be mounted on a mobile body, a storage device, and a control device. ,
By being executed
The first agent device is provided with a first agent that acquires the first information through the first dialogue with the user.
The second agent device is provided with a second agent that acquires the second information through the second dialogue with the user.
The first information and the second information are stored in the storage device, and the first information and the second information are stored in the storage device.
A process of causing the first agent to output a first phrase generated based on at least one of the first information and the second information stored in the storage device in the first dialogue, and storage in the storage device. The control device is made to execute at least one of the process of outputting the second phrase generated based on at least one of the first information and the second information to the second agent in the second dialogue.
Computer program.

Images