Information provision system, information provision method, and computer program product

Abstract

The invention provides an information provision system, an information provision method, and a computer program product, which can restrain a user from feeling uncomfortable when the user riding a moving body is reminded of attention through an HMI using an intelligent agent. An information provision system (1) is provided with a risk identification unit (14) for identifying a prescribed risk of a moving body pair, and a smart agent control unit (13) that sets the emotion of a smart agent on the basis of an emotion level associated with the degree of happiness and the degree of sadness, and that, when the emotion of the smart agent is set to a first emotion level, determines that the risk of the moving body pair is set to a second emotion level. When switching from a state in which the predetermined risk is not recognized by the risk recognition unit (14) to a state in which the predetermined risk is recognized by the risk recognition unit, the control unit changes the emotion of the smart agent to a second emotion level in which the degree of happiness is lower than the first emotion level, and outputs the attention presentation voice by the smart agent.

Description

Technical Field

[0001] This invention relates to information provision systems, information provision methods, and computer program products. Background Technology

[0002] In recent years, technologies have been proposed that incorporate human-like intelligent agents that communicate with users into HMIs (Human Machine Interfaces) used in information-providing devices for mobile entities such as vehicles. For example, Patent Document 1 discloses a technology that estimates the driver's mood, and when the driver's mood is a negative emotion that they do not want to receive a response from others, outputs response content that can change the driver's mood to a negative emotion below a predetermined level or a positive emotion, based on the level of the negative emotion.

[0003] [Existing technical documents]

[0004] [Patent Literature]

[0005] [Patent Document 1]

[0006] Japanese Patent Application Publication No. 2022-038423 Summary of the Invention

[0007] [The problem the invention aims to solve]

[0008] However, when using the HMI's intelligent agent to provide attention reminders to avoid contact with obstacles based on the vehicle's driver assistance functions, it's worth considering changing the mode of the intelligent agent's voice output to improve the effectiveness of these reminders. However, due to individual differences in human perception, changing the mode of the intelligent agent's voice output might create a sense of unease for the driver.

[0009] Therefore, the objective of this application is to provide an information delivery system that, while providing attention reminders to passengers via an HMI employing a smart agent during movement, suppresses situations that cause discomfort to the user. Furthermore, this contributes to the development of a sustainable transportation system that further improves traffic safety.

[0010] [Methods used to solve problems]

[0011] As a first method for achieving the above objectives, an information providing system is provided, comprising: a risk identification unit that identifies a specified risk faced by a mobile vehicle; and an intelligent agent control unit that provides a human-machine interface to a user riding in the mobile vehicle. This human-machine interface uses a human-like intelligent agent with emotions to output voice messages in a pattern corresponding to the emotions of the intelligent agent. When the risk identification unit identifies the specified risk, the intelligent agent outputs a attention reminder voice message. The intelligent agent control unit sets the emotions of the intelligent agent according to emotion levels corresponding to levels of happiness and sadness. When the emotions of the intelligent agent are set to a first emotion level, and the state where the specified risk has never been identified by the risk identification unit is changed to a state where the specified risk is identified by the risk identification unit, the emotions of the intelligent agent are changed to a second emotion level where the level of happiness is lower than the first emotion level, and the intelligent agent outputs the attention reminder voice message.

[0012] In the aforementioned information provision system, it can also be configured to include: a user state recognition unit that recognizes the user's state; and a user emotion estimation unit that estimates the user's emotion based on the user state recognition result of the user state recognition unit, wherein the intelligent agent control unit sets the intelligent agent's emotion based on the user's emotion estimated by the user emotion estimation unit when the risk recognition unit does not recognize the specified risk.

[0013] In the aforementioned information providing system, it can also be configured such that, when the risk identification unit does not identify the specified risk, the intelligent agent control unit sets the intelligent agent's emotion to a level of happiness that is higher than the user's emotion level estimated by the user emotion estimation unit.

[0014] In the aforementioned information provision system, it can also be configured such that the user emotion estimation unit uses emotion value as an estimation element of the user's emotion, and the intelligent agent control unit uses emotion value as a control element of the intelligent agent's emotion.

[0015] In the aforementioned information provision system, it can also be configured such that the user emotion estimation unit uses alertness as an estimation element for the user's emotions, and the intelligent agent control unit uses alertness as a control element for the intelligent agent's emotions.

[0016] As a second approach to achieving the aforementioned objective, an information provision method is provided, in which a computer performs the following steps: a risk identification step, identifying a specified risk faced by a mobile vehicle; and an intelligent agent control step, providing a human-machine interface (HMI) to a user riding in the mobile vehicle. This HMI uses a human-like intelligent agent with emotions to output voice messages in a pattern corresponding to the emotions of the intelligent agent. When the specified risk is identified through the risk identification step, the intelligent agent outputs a attention reminder voice message. In the intelligent agent control step, the intelligent agent's emotions are set according to emotion levels corresponding to happiness and sadness. When the intelligent agent's emotions are set to a first emotion level, if the state of not having identified the specified risk through the risk identification step switches to a state of having identified the specified risk through the risk identification step, the intelligent agent's emotions are changed to a second emotion level where the degree of happiness is lower than the first emotion level, and the intelligent agent outputs the attention reminder voice message.

[0017] As a third approach to achieving the aforementioned objectives, a computer program product is provided, which enables a computer to function as: a risk identification unit that identifies a specified risk faced by a mobile vehicle; and an intelligent agent control unit that provides a human-machine interface (HMI) to a user riding in the mobile vehicle. This HMI uses a human-like intelligent agent with emotions to output voice messages in a pattern corresponding to the emotions of the intelligent agent. When the risk identification unit identifies the specified risk, the intelligent agent outputs a warning voice message. The intelligent agent control unit sets the emotions of the intelligent agent according to emotion levels corresponding to happiness and sadness. When the emotions of the intelligent agent are set to a first emotion level, if the state from which the risk identification unit has never identified the specified risk to the state from which the risk identification unit has identified the specified risk changes, the emotions of the intelligent agent are changed to a second emotion level where the degree of happiness is lower than the first emotion level, and the intelligent agent outputs the warning voice message.

[0018] [Invention Effects]

[0019] Based on the aforementioned information provision system, information provision method, and computer program product, when using an HMI with an intelligent agent to provide attention reminders to users riding in a mobile vehicle, it is possible to suppress situations that cause users to feel uncomfortable. Attached Figure Description

[0020] Figure 1 This is a diagram illustrating how to use an information provision system.

[0021] Figure 2 This is a structural diagram of an information provision system.

[0022] Figure 3 This is an explanatory diagram of an emotion mapping chart based on emotion value and alertness.

[0023] Figure 4 This is a flowchart of the emotion control process of an intelligent agent.

[0024] Figure 5 This is an illustration of the emotional levels of an intelligent agent based on the user's emotional level.

[0025] Figure 6 It is a table for setting the emotional level of an intelligent agent based on the user's emotional level.

[0026] Figure 7 It is an assessment data showing the difference in user acceptability when the emotional level of the intelligent agent changes in response to the user's emotions.

[0027] Figure 8 This is an assessment of the difference in user acceptability when attention is alerted by decreasing the emotional level of the intelligent agent by one level. Detailed Implementation

[0028] [1. How to use an information provision system]

[0029] Reference Figure 1 The usage of the information providing system 1 in this embodiment will be described. The information providing system 1 provides the user U, who is the driver of the vehicle 100, with an HMI (Human Machine Interface) that uses an anthropomorphic, emotionally-enabled intelligent agent. The information providing system 1 can be configured as part of the functionality of a vehicle-mounted device such as a navigation system installed in the vehicle 100, or it can be configured as a dedicated vehicle-mounted device. The vehicle 100 corresponds to the mobile body of this disclosure.

[0030] The information providing system 1 has communication capabilities, enabling wireless communication with wearable devices 70 (such as smartwatches) worn by user U and portable terminals 60 (such as smartphones, tablets, and mobile phones) brought into vehicle 100 by user U. Additionally, the information providing system 1 communicates with external communication systems such as content server 210 via communication network 200.

[0031] Vehicle 100 is equipped with a driver monitoring camera 54 for capturing images of user U, a touch panel-sized display 55, a speaker 56, a microphone 57 for collecting user U's speech, a front camera 50 for capturing images of the front of vehicle 100, ADAS (Advanced Driver-Assistance Systems) 2 for performing various driver assistance controls based on images captured by the front camera 50, and a navigation device 3 for providing route guidance to the destination.

[0032] The intelligent agent integrated into the information providing system 1 is, for example, an interface intelligent agent composed of AI (Artificial Intelligence) programs. The intelligent agent is a conversational interface that interprets the user U's speech input to the microphone 57 via voice recognition, and outputs information through voice output from the speaker 56 or image display to the display 55 based on the speech content. Based on the user U's speech, the intelligent agent proposes destinations for route guidance provided by the navigation device 3, proposes content downloaded and played from the content server 210, and provides warnings regarding risks identified by ADAS2 (such as approaching obstacles).

[0033] The information providing system 1 infers the user U's emotions based on the image of the user U captured by the driver monitoring camera 54, the user U's voice input to the microphone 57, and the user U's biometric information (heart rate, body temperature, blood pressure, etc.) detected by the wearable device 70.

[0034] Furthermore, the information providing system 1 sets the emotion of the intelligent agent based on the user U's emotion, and the intelligent agent controls the mode of the voice output from the speaker 56 (speech content, tone, pitch, volume, etc.) according to the set emotion. In addition, when the information providing system 1 detects a specified risk, it changes the emotion level of the intelligent agent and outputs a attention reminder voice, as detailed later.

[0035] [2. Structure of Information Provision System]

[0036] Reference Figures 2-5 The structure of information provision system 1 will be described below. (Refer to...) Figure 2 The information providing system 1 is a control unit that includes a processor 10, a memory 20, a communication unit 30, etc. The information providing system 1 is connected to the ADAS 2, speed sensor 51, gyroscope sensor 52, GNSS (Global Navigation Satellite System) sensor 53, driver monitoring camera 54, display 55, speaker 56 and microphone 57 installed on the vehicle 100.

[0037] The communication unit 30 communicates with the portable terminal 60 and the wearable device 70 via short-range wireless communication based on standards such as Bluetooth and UWB (Ultra Wide Band). Additionally, the communication unit 30 communicates with the content server 210 and the like via the communication network 200.

[0038] The memory 20 stores the control program 21 for the information providing system 1, data such as the emotion mapping map 22, etc. Details about the emotion mapping map 22 will be described later. The processor 10 functions as the user state recognition unit 11, the user emotion estimation unit 12, the intelligent agent control unit 13, and the risk recognition unit 14 by reading and executing the program 21.

[0039] The processing performed by the user state recognition unit 11 is equivalent to the user state recognition step in the information provision method of this disclosure, and the processing performed by the user emotion estimation unit 12 is equivalent to the user emotion estimation step in the information provision method of this disclosure. The processing performed by the intelligent agent control unit 13 is equivalent to the intelligent agent control step in the information provision method of this disclosure, and the processing performed by the risk recognition unit 14 is equivalent to the risk recognition step in the information provision method of this disclosure.

[0040] The user status recognition unit 11 identifies the status of the user U as the user U riding in the vehicle 100 based on at least any one of the following: information input through touch operation of display 55, image of user U captured by driver monitoring camera 54, voice of user U input to microphone 57, and liveness information of user U detected by wearable device 70.

[0041] The user status recognition unit 11 identifies the following elements as the status of user U, for example.

[0042] The first element is the answer that the user U inputs through the touch operation of the display 55 when the user is riding in the vehicle 100 in response to questions such as "How are you feeling today?" displayed on the display 55.

[0043] The second element is…the facial expressions and actions of user U identified from user U's image.

[0044] The third element is the content, pitch, volume, and tone of the user U's voice.

[0045] The fourth element is the user U's biological information (heart rate, blood pressure, body temperature, etc.).

[0046] The user emotion estimation unit 12 estimates the user U's emotion based on the state of the user U identified by the user state recognition unit 11. The user emotion estimation unit 12 then estimates the user U's emotion based on... Figure 3 The emotion mapping diagram 22 shown presumes the user U's emotion value and alertness as evaluation elements of emotion. The emotion value represents the user U's level of unpleasantness to pleasantness, while alertness represents the user U's level of excitement to calmness. Figure 3 Examples of the estimated mood levels of user U (awareness: 3, mood value: -4) and the mood levels of the set intelligent agent (awareness: 7, mood value: 0) are shown.

[0047] The user emotion estimation unit 12, for example, sets the emotion value and level of alertness based on the first to fourth elements mentioned above, and thereby estimates the emotion of user U.

[0048] Regarding the first element mentioned above, the user emotion estimation unit 12 sets the user U's emotion value to a relatively high value (e.g., 3) when the user's answer to the question is "as usual" and sets the alertness value to a relatively low value (e.g., -3) when the user's answer to the question is "different from usual".

[0049] Regarding the second element mentioned above, the user emotion estimation unit 12 sets a higher emotion value (e.g., 5) when it determines that user U is smiling based on their facial expression, and sets a lower emotion value (e.g., -5) when it determines that user U is showing displeasure. Furthermore, based on user U's behavior, the user emotion estimation unit 12 sets a higher alertness value (e.g., 3) when it determines that user U is acting restlessly, and a lower alertness value (e.g., -3) when it determines that user U is hardly active.

[0050] Regarding the third element mentioned above, the user emotion estimation unit 12 sets a higher emotion value (e.g., 5) based on the content of user U's speech. If the speech is judged to be positive, such as praising or expecting something, the emotion value is set to a lower value (e.g., -5). If the speech is judged to be negative, such as blaming something, the emotion value is set to a lower value (e.g., -5). In addition, if user U's speech includes specific keywords ("good", "very good", etc.), the user emotion estimation unit 12 sets an emotion value and alertness value associated with that keyword.

[0051] In addition, the user emotion estimation unit 12 sets the alertness level to a relatively high value (e.g., 3) based on the height of the user U's voice. For example, if the user U's voice is above a certain height, the alertness level is set to a relatively low value (e.g., -3).

[0052] Regarding the fourth element, the user emotion estimation unit 12 sets the emotion value and alertness by applying the detected values ​​of the user U's biological information (heart rate, blood pressure, body temperature, etc.) to a pre-prepared correspondence table between the detected values ​​and the emotion value and alertness. The correspondence table can also be created based on the user U's personal information input by the user U and the biological information of the user U detected in the past.

[0053] Furthermore, the user emotion estimation unit 12 can also identify emotion values ​​and alertness values ​​based on changes in the vehicle 100's speed detected by the speed sensor 51, the vehicle 100's movements detected by the gyroscope sensor 52, and changes in the vehicle 100's position detected by the GNSS sensor 53. Additionally, the user emotion estimation unit 12 can also estimate the user U's emotion based on either the emotion value or alertness, or other evaluation factors.

[0054] The intelligent agent control unit 13 provides the user U with a human-like intelligent agent with emotions in its HMI. The intelligent agent control unit 13 analyzes the user U's speech input to the microphone 57 and outputs a response to the speech via the intelligent agent's voice output from the speaker 56 or an image display from the intelligent agent to the display 55. The intelligent agent control unit 13 changes the intelligent agent's emotion based on the user U's emotion estimated by the user emotion estimation unit 12, and also changes the output mode (pitch, volume, etc.) of the intelligent agent's voice output from the speaker 56 according to the intelligent agent's emotion. Details of the processing performed by the intelligent agent control unit 13 will be described later.

[0055] The risk identification unit 14 identifies specified risks faced by the vehicle 100 based on risk information input from ADAS 2. Examples of specified risks faced by the vehicle 100 include, for instance, approaching monitored objects (other vehicles, pedestrians, obstacles, etc.) present around the vehicle 100. The risk level identified by the risk identification unit 14 increases as the distance between the vehicle 100 and the monitored object decreases. Furthermore, the risk identification unit 14 can also identify the user U's level of focus on driving based on the facial image of the user U obtained from the driver monitoring camera 54 and the biometric information of the user U detected by the wearable device 70, identifying a state where the user's focus level drops below a specified level as a specified risk.

[0056] [3. Emotion Control Processing in Intelligent Agents]

[0057] according to Figure 4 The flowchart shown illustrates the steps of emotion control processing performed by the intelligent agent of the HMI by the information providing system 1. When the information providing system 1 recognizes that user U has entered vehicle 100 and turned on the power, it begins... Figure 6The flowchart shown illustrates the processing.

[0058] exist Figure 4 In step S1, the user emotion estimation unit 12 estimates the user U's emotion based on the user U's state identified by the user state recognition unit 11. In the next step S2, the intelligent agent control unit 13 uses the user U's emotion level estimated by the user emotion estimation unit 12 in step S1 as a benchmark and sets the intelligent agent's emotion level to be higher than the user U's emotion level. Furthermore, a high emotion level indicates a state where both the emotion value and the alertness value are high.

[0059] Here, as Figure 5 , Figure 6 As shown, the intelligent agent control unit 13 takes the user U's emotional level as the origin (emotional value: 0, alertness: 0). With the seven levels representing the intelligent agent's emotional level as sadness 3, sadness 2, sadness 1, user emotion, happiness 1, happiness 2, and happiness 3, the intelligent agent's emotional level is set by the relative level (first emotional level) of any one of happiness 1, happiness 2, and happiness 3.

[0060] In the next step S3, the intelligent agent control unit 13 determines whether the risk identification unit 14 has identified a specified risk. If the risk identification unit 14 identifies a specified risk, the intelligent agent control unit 13 proceeds to step S20; if the risk identification unit 14 does not identify a specified risk, the process proceeds to step S4.

[0061] In step S4, when the intelligent agent performs voice guidance such as route guidance and content suggestions, the intelligent agent control unit 13 controls the output mode of the voice output from the speaker 56 according to the first emotion level setting, so that the processing proceeds to step S5.

[0062] In step S20, the intelligent agent control unit 13 changes the intelligent agent's emotion level to a second emotion level, which is one level lower than the first emotion level. In the next step S21, when the intelligent agent outputs a warning voice to alert the user of a specified risk, the intelligent agent control unit 13 controls the output mode of the voice output from the speaker 56 according to the second emotion level, thus proceeding to step S5.

[0063] Here, refer to Figure 7 , Figure 8 The effects of these voice prompts that trigger changes in the emotional level of intelligent agents are explained. Figure 7The data presented shows the evaluation data obtained by monitoring the degree of acceptability (discomfort) felt by subjects when the emotional level of an intelligent agent changed from the estimated emotional level of the subjects and attention reminder voice was output.

[0064] in addition, Figure 8 This paper presents evaluation data obtained by monitoring the degree of acceptability (discomfort) felt by multiple subjects when an intelligent agent changed its emotional level from a presumed emotional level (first emotional level) to a second emotional level and output an attention reminder voice. Figure 7 , Figure 8 The figure illustrates the data used to evaluate the acceptability (discomfort) perceived by the test subjects when the intelligent agent outputs a attention reminder voice after changing the emotional level of the agent. The maximum score is set to ±3.

[0065] according to Figure 7 When the change in the intelligent agent's emotional level from the first emotional level to the second emotional level is set to two levels (two stages) or more, the acceptance level is reduced. Additionally, according to... Figure 8 The evaluation showed that by setting the first emotional level higher than the subject's emotional level and lowering the second emotional level by one level (one level lower), acceptability could be further improved. In particular, the highest acceptability score of +1.5 was achieved when the first emotional level was set to "happiness 2," which is two levels higher than the user's emotional level, and the second emotional level was set to "happiness 1," which is one level lower than "happiness 2."

[0066] Based on this evaluation result, Figure 4 In step S2, the intelligent agent control unit 13 sets the first emotion level of the intelligent agent to be higher than the emotion level of the user U, and in step S20, sets the second emotion level of the intelligent agent to be one level lower than the first emotion level. Thus, the intelligent agent can output attention reminder voice while suppressing any sense of unease felt by the user U.

[0067] exist Figure 4 In step S5, the intelligent agent control unit 13 determines whether a power disconnection operation has occurred in the vehicle 100. Then, if no power disconnection operation has occurred, the intelligent agent control unit 13 proceeds to step S1, and executes steps S1-S5, S20, and S21 again. On the other hand, if a power disconnection operation has occurred, the intelligent agent control unit 13 proceeds to step S6, ending the intelligent agent's emotion control processing.

[0068] [4. Other Implementation Methods]

[0069] In the above embodiments, a vehicle 100 is shown as the mobile body of this disclosure, but the mobile body of this disclosure can be any mobile body that provides an HMI using a smart agent to the user in the vehicle, or it can be an aircraft, a ship, etc.

[0070] In the above embodiments, an information providing system 1 consisting of an in-vehicle unit mounted on a vehicle 100 is exemplified as the information providing system of this disclosure. As other embodiments, the information providing system of this disclosure may also consist of a portable terminal 60 brought into the vehicle 100, or a content server 210 or other external communication systems of the vehicle 100.

[0071] When the portable terminal 60 constitutes an information providing system, the camera, microphone, speaker, and display of the portable terminal 60 can be used to provide an HMI using a smart agent. Additionally, biometric information of the user U can be obtained through communication between the wearable device 70 and the portable terminal 60, and the user U's emotions can be inferred through the portable terminal 60 based on this biometric information. Furthermore, through communication between the vehicle's onboard unit 100 and the portable terminal 60, the portable terminal 60 can obtain information on prescribed risks identified by ADAS2.

[0072] In addition, in the case where an information providing system is constituted by an external communication system, for example, the system is configured to send the image of the user U captured by the driver monitoring camera 54 of the vehicle 100, the voice data input to the microphone 57, and the information on the prescribed risks identified by ADAS2 from the vehicle 100 to the external communication system, and the external communication system sets the emotions of the intelligent agent.

[0073] In the above embodiment, a user state recognition unit 11 and a user emotion estimation unit 12 are included, and the intelligent agent control unit 13 sets the intelligent agent's emotion based on the estimated value of the user U's emotion. Alternatively, the user state recognition unit 11 and the user emotion estimation unit 12 may be omitted, and the intelligent agent's emotion in a stable state where ADAS 2 has not identified a specified risk may be set to a pre-set specified level (the first emotion level of this disclosure). In this case, the intelligent agent control unit 13 is configured to change the intelligent agent's emotion level to an emotion level that reduces the degree of happiness (equivalent to the second emotion level of this disclosure) when switching from a state where ADAS 2 has never identified a specified risk to a state where ADAS 2 has identified a specified risk. Furthermore, the intelligent agent's emotion level in a stable state may also be set based on the user U's personal information, etc.

[0074] in addition, Figure 2This is a schematic diagram illustrating the structure of the information providing system 1 according to its main processing content for ease of understanding of the invention. The structure of the information providing system 1 can also be configured through other distinctions. Furthermore, the processing of each component can be performed by one hardware unit or by multiple hardware units. Figure 4 The processing of each component shown can be performed by one program or by multiple programs.

[0075] [5. Structure supported by the above embodiments]

[0076] The above implementation is a specific example of the following structure.

[0077] (Structure 1) An information providing system comprising: a risk identification unit that identifies a predetermined risk faced by a mobile body; and an intelligent agent control unit that provides a human-machine interface to a user riding in the mobile body, the human-machine interface using an anthropomorphic intelligent agent with emotions to perform voice output in a pattern corresponding to the emotions of the intelligent agent, wherein when the predetermined risk is identified by the risk identification unit, the intelligent agent outputs a attention reminder voice, the intelligent agent control unit sets the emotions of the intelligent agent according to emotion levels corresponding to happiness and sadness, and when the emotions of the intelligent agent are set to a first emotion level, if the state from which the predetermined risk has never been identified by the risk identification unit to the state from which the predetermined risk has been identified by the risk identification unit changes the emotions of the intelligent agent to a second emotion level where the degree of happiness is lower than the first emotion level, and the intelligent agent outputs the attention reminder voice.

[0078] According to the information provision system of Structure 1, when the HMI using the intelligent agent provides attention reminders to the user while in motion, it is possible to suppress situations that make the user feel uncomfortable.

[0079] (Structure 2) The information providing system according to Structure 1, wherein the information providing system includes: a user state recognition unit that recognizes the state of the user; and a user emotion estimation unit that estimates the emotion of the user based on the recognition result of the user state recognition unit on the user state, wherein the intelligent agent control unit sets the emotion of the intelligent agent based on the emotion of the user estimated by the user emotion estimation unit when the risk recognition unit does not recognize the specified risk.

[0080] According to the information provision system of Structure 2, by setting the emotions of the intelligent agent based on the user's emotions, a natural dialogue between the user and the intelligent agent can be achieved, and the effect of alerting attention when a specified risk occurs can be improved.

[0081] (Structure 3) According to the information providing system of Structure 2, when the risk identification unit does not identify the specified risk, the intelligent agent control unit sets the intelligent agent's emotion to a level of happiness that is higher than the emotion level of the user estimated by the user emotion estimation unit.

[0082] According to the information provision system of Structure 3, in Structure 1, the first emotional level of the intelligent agent set by the intelligent agent control unit is an emotional level of happiness that is higher than the emotional level of the user estimated by the user's emotional estimation unit. Therefore, when reminding the user to pay attention, the effect of suppressing the user's sense of discomfort can be enhanced.

[0083] (Structure 4) An information providing system based on Structure 2 or Structure 3, wherein the user emotion estimation unit uses emotion value as an estimation element of the user's emotion, and the intelligent agent control unit uses emotion value as a control element of the intelligent agent's emotion.

[0084] According to the information provision system of Structure 4, the user's emotions can be inferred using pleasant and unpleasant evaluation scales, i.e., emotion values, and the emotions of the intelligent agent can be controlled.

[0085] (Structure 5) An information providing system based on any one of Structures 2 to 4, wherein the user emotion estimation unit uses alertness as an estimation element of the user's emotion, and the intelligent agent control unit uses alertness as a control element of the intelligent agent's emotion.

[0086] According to the information provision system of Structure 5, it is possible to use alertness as an evaluation scale for excitement and calmness to infer the user's emotions and control the emotions of the intelligent agent.

[0087] (Structure 6) An information provision method, wherein a computer performs the following steps: a risk identification step, identifying a specified risk faced by a mobile vehicle; and an intelligent agent control step, providing a human-machine interface to a user riding in the mobile vehicle, the human-machine interface using a human-like intelligent agent with emotions, performing voice output in a pattern corresponding to the emotions of the intelligent agent, wherein when the specified risk is identified through the risk identification step, the intelligent agent outputs a attention reminder voice, and in the intelligent agent control step, the emotions of the intelligent agent are set according to emotion levels corresponding to happiness and sadness, and when the emotions of the intelligent agent are set to a first emotion level, if the state of never having identified the specified risk through the risk identification step switches to the state of having identified the specified risk through the risk identification step, the emotions of the intelligent agent are changed to a second emotion level where the degree of happiness is lower than the first emotion level, and the intelligent agent outputs the attention reminder voice.

[0088] By having a computer execute the information provision method of Structure 6, the same effect as the information provision system of Structure 1 can be achieved.

[0089] (Structure 7) A computer program product that enables a computer to function as: a risk identification unit that identifies a specified risk faced by a mobile body; and an intelligent agent control unit that provides a human-machine interface to a user riding in the mobile body, the human-machine interface using an anthropomorphic intelligent agent with emotions to output voice in a pattern corresponding to the emotions of the intelligent agent, wherein when the risk identification unit identifies the specified risk, the intelligent agent outputs a attention reminder voice, the intelligent agent control unit sets the emotions of the intelligent agent according to emotion levels corresponding to happiness and sadness, and when the emotions of the intelligent agent are set to a first emotion level, if the state from which the risk identification unit has never identified the specified risk to the state from which the risk identification unit has identified the specified risk changes, the emotions of the intelligent agent are changed to a second emotion level with a happiness level lower than the first emotion level, and the intelligent agent outputs the attention reminder voice.

[0090] By executing the computer program product of structure 7 by a computer, the structure of the information provision system of structure 1 can be realized.

[0091] Explanation of reference numerals in the attached figures

[0092] 1…Information providing system, 2…ADAS, 10…Processor, 11…User status recognition unit, 12…User emotion estimation unit, 13…Intelligent agent control unit, 14…Risk recognition unit, 20…Memory, 21…Program, 22…Emotion map, 30…Communication unit, 50…Front-facing camera, 51…Speed ​​sensor, 52…Gyroscope sensor, 53…GNSS sensor, 54…Driver monitoring camera, 55…Display, 56…Speaker, 57…Microphone, 60…Portable terminal, 70…Wearable device, 100…Vehicle (mobile body), 200…Communication network, 210…Content server, U…User.

Claims

1. An information providing system, the information providing system comprising: The risk identification department identifies the prescribed risks faced by the mobile entity; and The intelligent agent control unit provides a human-machine interface (HMI) to the user riding in the mobile vehicle. This HMI uses a human-like intelligent agent with emotions to output voice messages in a pattern corresponding to the emotions of the intelligent agent. When the risk identification unit identifies the specified risk, the intelligent agent outputs a warning voice message. The intelligent agent control unit sets the intelligent agent's emotion according to the emotion levels corresponding to happiness and sadness. When the intelligent agent's emotion is set to the first emotion level, if the state from which the prescribed risk has never been identified by the risk identification unit to the state from which the prescribed risk has been identified by the risk identification unit, the intelligent agent's emotion is changed to the second emotion level, where the degree of happiness is lower than the first emotion level, and the intelligent agent outputs the attention reminder voice.

2. The information providing system according to claim 1, wherein, This information providing system has the following features: The user status identification unit identifies the user's status; and The user emotion estimation unit estimates the user's emotion based on the user state recognition unit's recognition result of the user's state. When the risk identification unit fails to identify the specified risk, the intelligent agent control unit sets the intelligent agent's emotion based on the user's emotion inferred by the user emotion inference unit.

3. The information providing system according to claim 2, wherein, When the risk identification unit fails to identify the specified risk, the intelligent agent control unit sets the intelligent agent's emotion to a level of happiness that is higher than the user's emotion level estimated by the user emotion estimation unit.

4. The information providing system according to claim 2 or 3, wherein, The user emotion estimation unit uses emotion values ​​as inference factors for the user's emotions. The intelligent agent control unit uses emotion values ​​as the control element for the intelligent agent's emotions.

5. The information providing system according to claim 2 or 3, wherein, The user emotion estimation unit uses alertness as a factor in estimating the user's emotion. The intelligent agent control unit uses alertness as a control element for the intelligent agent's emotions.

6. An information providing method, in which a computer performs the following steps: The risk identification process identifies the specified risks faced by the moving body; and The intelligent agent control step provides a human-machine interface (HMI) to the user riding in the mobile device. This HMI uses a human-like intelligent agent with emotions to output voice messages in a pattern corresponding to the emotions of the intelligent agent. When the risk identification step identifies the specified risk, the intelligent agent outputs a warning voice message. In the intelligent agent control step, the intelligent agent's emotion is set according to the emotion level corresponding to happiness and sadness. When the intelligent agent's emotion is set to the first emotion level, if the state of never having identified the specified risk through the risk identification step switches to the state of having identified the specified risk through the risk identification step, the intelligent agent's emotion is changed to the second emotion level, where the degree of happiness is lower than the first emotion level, and the intelligent agent outputs the attention reminder voice.

7. A computer program product that enables a computer to function as a component: The risk identification department identifies the prescribed risks faced by the mobile entity; and The intelligent agent control unit provides a human-machine interface (HMI) to the user riding in the mobile vehicle. This HMI uses a human-like intelligent agent with emotions to output voice messages in a pattern corresponding to the emotions of the intelligent agent. When the risk identification unit identifies the specified risk, the intelligent agent outputs a warning voice message. The intelligent agent control unit sets the intelligent agent's emotion according to the emotion levels corresponding to happiness and sadness. When the intelligent agent's emotion is set to the first emotion level, if the state from which the prescribed risk has never been identified by the risk identification unit to the state from which the prescribed risk has been identified by the risk identification unit, the intelligent agent's emotion is changed to the second emotion level, where the degree of happiness is lower than the first emotion level, and the intelligent agent outputs the attention reminder voice.

Images