Dialogue system for a vehicle

Abstract

Dialogue system to provide information to a driver in a vehicle (1), wherein the dialogue system performs automatic dialogue control, in which a dialogue is performed between a first agent and a second agent, where, in the case where it is determined that information exists which requires transmission to the driver during the execution of the automatic dialogue control, it is determined, on the basis of a result from a comparison of a measure of discretion that the driver has with a decision value linked to the information, whether or not the dialogue system should transmit an input containing the information to the driver, and the dialogue system transmits the input containing the information to the driver in the case where the degree of discretion is greater than the decision value associated with the information, whereas in the case where the degree of discretion is no greater than the decision value, the dialogue system continues to carry out the automatic dialogue control and indirectly provides information to the driver by conducting the dialogue between the first agent and the second agent.

Description

[Technical field]

[0001] The present invention relates to dialogue systems for vehicles. [Previous state of the art]

[0002] JP 2016 - 133 557 A discloses a dialogue system that includes multiple agents capable of conducting a dialogue with a user. In this known dialogue system, the agents can transmit an input to the user and receive a response from the user linked to the input without interrupting the dialogue between them. [State of the art]

[0003] JP 2012 - 198 052 A describes a mode setting procedure executed by an HMI control unit, in which the driver's resting state is estimated based on input information, and the mode setting is performed accordingly. Depending on the selected mode, functions are enabled in addition to driving the vehicle.

[0004] JP 2006 - 330 833 A describes a warning notification system configured to compare a vehicle's current location information from a navigation device with pre-stored location information of a hazardous location. If this location information matches and the driver's experience meets a risk decision condition, the driver is warned via the navigation system.

[0005] DE 601 24 937 T2 describes a method for improving the driver performance of a motor vehicle through performance feedback. Vehicle operating data, driver activity data, vehicle environment data, and driver state data are received, and based on this, a driver performance assessment is recorded. This assessment evaluates the driver's ability to operate the vehicle in comparison to practical experience. This driver performance assessment is stored on board the vehicle and communicated to the driver.

[0006] JP 2005 - 100 382 A describes a dialogue system for displaying a dialogue between a user and a virtual figure on a screen. Changes in the user's state or changes in the user's situation are detected, and a suitable dialogue scenario between a user avatar and the virtual figure is selected and displayed.

[0007] US 2018 / 0244187A1 describes a monitoring system for the cargo space of a vehicle. The monitoring system monitors the condition of the cargo in the cargo space using a pressure measuring device and notifies the vehicle user if it detects that the cargo is in an unacceptable condition. [Brief description of the invention][Technical problem]

[0008] However, the existing dialogue system does not include details regarding the timing of input transmission to the user. Installing such a dialogue system in a vehicle could lead to driving errors and reduced concentration.

[0009] One object of the present invention is to provide a dialogue system that is able to transmit an input to a user in a vehicle without causing a driving error or a deterioration of concentration. [Solution to the problem]

[0010] A dialogue system is provided to supply information to a driver in a vehicle. This dialogue system performs automatic dialogue control, where a dialogue is conducted between a first agent and a second agent. In cases where it is determined that information is present that requires transmission to the driver during the execution of the automatic dialogue control, the system determines, based on a comparison of the driver's discretion with a decision value associated with the information, whether or not to transmit an input containing the information to the driver. The dialogue system transmits the input containing the information to the driver if the driver's discretion is greater than the decision value associated with the information, whereas in cases where the driver's discretion is less than the decision value associated with the information, the system does not transmit the input.in which the degree of discretion is no greater than the decision value, the dialogue system continues the execution of the automatic dialogue control and indirectly provides information to the driver by executing the dialogue between the first agent and the second agent. [Advantageous effect of the invention]

[0011] As described, the present invention provides a dialogue system that is able to transmit an input to a driver in a vehicle without causing a driving error or a deterioration in concentration. [Brief description of the drawings] Fig. Figure 1 is a block diagram of an exemplary dialogue system. Fig. 2 is a table showing measures of leeway that a driver may have in relation to various characteristics belonging to a category of behaviors of a vehicle in which the exemplary dialogue system is installed. Fig. Figure 3 is a table showing measures of leeway that a driver may have in relation to various features belonging to a category of conditions surrounding the vehicle in which the exemplary dialogue system is installed. Fig. Figure 4 is a table showing measures of leeway that a driver may have in relation to various characteristics belonging to a category of driver states in the vehicle in which the exemplary dialogue system is installed. Fig. Figure 5 is a table showing how to determine decision values ​​linked to different types of information to be transmitted as input to a driver in the vehicle in which the exemplary dialogue system is installed. Fig. Figure 6 is a table showing how to determine whether an input should be transmitted to a driver in a vehicle in which the exemplary dialogue system is installed, based on measures of discretion that the driver may have in relation to various features belonging to a category of vehicle behaviors, and decision values ​​linked to different types of information that should be transmitted as the input to the driver. Fig. Figure 7 is a table showing how to determine whether an input should be transmitted to a driver in a vehicle in which the exemplary dialogue system is installed, based on measures of discretion that the driver may have in relation to various features belonging to a category of conditions surrounding the vehicle, and on decision values ​​linked to different types of information that should be transmitted as the input to the driver. Fig. Figure 8 is a table showing how to determine whether an input should be transmitted to a driver in a vehicle in which the exemplary dialogue system is installed, based on measures of discretion that the driver may have in relation to various characteristics belonging to a category of driver states, and decision values ​​linked to different types of information that should be transmitted as the input to the driver. Fig. Figure 9 is a flowchart showing a procedure to determine, based on measures of latitude the driver may have, whether to transmit an input to a driver in the vehicle in which the exemplary dialogue system is installed. Fig. Table 10 shows how to determine decision thresholds linked to different types of information to be transmitted as input to a driver in a vehicle in which a second exemplary dialogue system is installed, based on measures of discretion the driver may have in relation to various features belonging to a category of conditions surrounding the vehicle. Fig. Table 11 shows how to determine decision thresholds linked to different types of information to be transmitted as input to a driver in the vehicle in which the second exemplary dialogue system is installed, based on measures of discretion the driver may have in relation to various features belonging to a category of conditions surrounding the vehicle. Fig. Table 12 shows how to determine decision thresholds linked to different types of information to be transmitted as input to a driver in the vehicle in which the second exemplary dialogue system is installed, based on measures of latitude the driver may have in relation to various characteristics belonging to a category of driver states. Fig. Figure 13 is a flowchart showing a procedure to determine, based on measures of latitude a driver may have, whether to transmit an input to a driver in the vehicle in which the second exemplary dialogue system is installed. [Detailed description]

[0012] A dialogue system is disclosed for providing information to a driver in a vehicle, wherein the dialogue system performs an automatic dialogue control in which a dialogue is executed between a first agent and a second agent, wherein, in the case where it is determined that information exists which requires transmission to the driver during the execution of the automatic dialogue control, it is determined, based on a result of a comparison of a measure of discretion that the driver has with a decision value that is associated with the information, whether or not the dialogue system should transmit an input containing the information to the driver.

[0013] As described, a dialogue system is provided that is able to transmit an input to a driver in a vehicle without causing a driving error or a decrease in concentration. [Version(s)]

[0014] With reference to the attached drawings, an exemplary dialogue system for a vehicle is described.

[0015] In Fig. Vehicle 1 is a vehicle in which an exemplary dialogue system is installed, generally referred to as 1. Vehicle 1 includes a forward sensor 2, a communication unit 3, a notification unit 4, and a control unit 5.

[0016] The forward sensor 2 is configured and arranged to measure the distance from the vehicle ahead using lasers, ultrasonic waves, or millimeter waves.

[0017] Communication unit 3 is arranged in such a way that it communicates with external servers via a communication infrastructure.

[0018] The control unit 5 is arranged to communicate with the external servers via the communication unit 3 in order to obtain a variety of information, such as weather information at the current location, traffic jam information, information relating to the road profile, and traffic control information.

[0019] The notification unit 4 can include a display, a speaker, a lamp, a measuring instrument, and an acoustic display to inform the vehicle driver of various types of information through sight and hearing. The notification unit 4 can have several output formats. Specifically, it can use different output formats generated by a first, second, and third agent. The first, second, and third agents can generate different formats to inform the vehicle driver of information in such a way that the driver can distinguish between the agents sequentially. For example, the first, second, and third agents can display different visual output images and / or different voices and sounds through the speaker.

[0020] The control unit 5 consists of a computer unit that includes a central processing unit (CPU), random access memory (RAM), read-only memory (ROM), flash memory, input ports and output ports.

[0021] The computer unit's ROM, along with various types of constants and characteristic maps, stores programs that enable the computer unit to function as the control unit 5. In other words, the computer unit is enabled to function as the control unit 5 by the CPU executing the programs stored in the ROM.

[0022] In addition to the forward sensor 2, the input terminals of the control unit 5 are connected to a speed sensor 81, an acceleration sensor 82, a steering angle sensor 83, a GNSS receiver 84 and a driver monitor 85.

[0023] The speed sensor 81 is designed to detect the vehicle speed of vehicle 1. The acceleration sensor 82 is designed to detect longitudinal and lateral acceleration of vehicle 1. The steering angle sensor 83 is designed to detect the steering angle of a steering wheel (not shown). The GNSS receiver 84 is designed to receive radio waves from the Global Navigation Satellite System (GNSS) to determine the current location of vehicle 1.

[0024] The driver monitor 85 is configured and arranged to monitor the state of a driver operating vehicle 1. The driver monitor 85 includes, for example, a camera designed to capture an image of the driver and a microphone designed to record sounds inside the vehicle. The driver monitor 85 detects the driver's state based on the image captured by the camera and the sounds recorded by the microphone. The driver monitor 85 provides the driver's state as an input to the control unit 5.

[0025] In contrast, the output terminals of the control unit 5 are connected to various types of objects to be controlled, including the aforementioned notification unit 4.

[0026] In the exemplary dialogue system, control unit 5 is configured to perform automatic dialogue control, in which a dialogue is conducted between a first agent and a second agent. Executing this dialogue indirectly provides information to the driver. Specifically, control unit 5 provides a script to instruct one of the first and second agents to input information into the other, and then to instruct the other agent to input responses directly or indirectly related to some or all of the inputs. The driver can obtain useful information by listening to the exchange of inputs and responses between the first and second agents. Thus, the exemplary dialogue system can provide the driver in vehicle 1 with some information in an automatic dialogue control mode.

[0027] Each aspect of the driver's driving behavior and the behavior of the driver's vehicle is compared with respective reference models, for example, to evaluate the driver's driving behavior and / or the behavior of the driver's vehicle.

[0028] The first agent is activated to input the evaluation results into the second agent via notification unit 4.

[0029] In the exemplary dialogue system, the degree of discretion that a driver can have in vehicle 1 is determined on the basis of one of several inputs, for example, regarding the behavior of the driver's vehicle, conditions around the driver's vehicle, and the driver's state.

[0030] Reference models are created to evaluate various inputs relating to the behavior of the driver's vehicle, the conditions surrounding the driver's vehicle, and the driver's state. The degree of leeway for each input is determined based on a comparison of each input with its reference model, which is described in detail below with reference to the Fig. 2 to 4 are described.

[0031] As in Fig. As shown in Figure 2, the exemplary dialogue system determines the measures of leeway associated with the vehicle's behavior. Reference models include: "a parking brake is engaged for a fixed period of time"; "a zero-motion behavior is maintained for a fixed period of time"; "no exceeding of the speed limit"; "low rate of change in acceleration"; "low rate of change in deceleration"; "low rate of change in steering"; and "the rate at which the driver counter-steers is not excessive."

[0032] Referring to Fig. 2 is the measure of leeway associated with parking, in the case where a parking brake is engaged for a fixed period of time, equal to 150.

[0033] The degree of leeway associated with a stop is equal to 130 in the case where zero movement is maintained for a fixed period of time.

[0034] The degree of leeway associated with vehicle speed is equal to 100 in the case where the rate of change of vehicle speed is kept below a certain level during a fixed period of time, while the vehicle speed is within or equal to the speed limit.

[0035] The current vehicle speed outside the speed limit multiplied by 100 (percent) is used as an evaluation value for the vehicle speed. The margin of error is 70 if the evaluation value is less than V1, 40 if it is not less than V1 but less than V2, and 10 if it is not less than V2.

[0036] The acceleration of vehicle 1 is used as an evaluation value for acceleration. The measure of leeway associated with the acceleration is 70 if the evaluation value is less than A1, 40 if the evaluation value is not less than A1 but less than A2, and 10 if the evaluation value is not less than A2.

[0037] The deceleration of vehicle 1 is used as an evaluation value for the deceleration. The measure of leeway associated with the deceleration is 70 if the evaluation value is less than D1, 40 if the evaluation value is not less than D1 but less than D2, and 10 if the evaluation value is not less than D2.

[0038] The vehicle's lateral acceleration is used as an evaluation value for sudden steering. The degree of leeway associated with sudden acceleration is 70 if the evaluation value is less than H1, 40 if the evaluation value is not less than H1 but less than H2, and 10 if the evaluation value is not less than H2.

[0039] The rate at which the driver counter-steers is used as an evaluation value for a zigzag movement. The degree of leeway associated with a zigzag movement is 70 if the evaluation value is less than F1, 40 if the evaluation value is not less than F1 but less than F2, and 10 if the evaluation value is not less than F2.

[0040] As in Fig. As shown in Figure 3, the exemplary dialogue system determines the degree of leeway linked to conditions surrounding the vehicle. Reference models include: "not less than the safety distance, which is a function of the vehicle speed (safety distance xz when it is raining)"; and "the ambient light intensity is not less than G1".

[0041] Referring to Fig. 3 is the measure of leeway that is linked to the distance from the vehicle in front, in the case where there is no vehicle in front within a certain distance, equal to 100.

[0042] The distance calculated by subtracting the current distance to the vehicle ahead from the safety distance (which is a function of the current vehicle speed) is an evaluation value for the distance to the vehicle ahead. The measure of leeway associated with the distance to the vehicle ahead is 70 when the evaluation value is less than S1, 40 when the evaluation value is not less than S1 but less than S2, and 10 when the evaluation value is not less than S2.

[0043] The degree of leeway associated with darkness is equal to 100 in the case where ambient light with a sufficiently high level of illumination is present for a longer period than a fixed period.

[0044] Ambient light is used as an evaluation value for darkness.

[0045] The degree of leeway associated with darkness is 70 in the case where the evaluation value is not less than G2 but less than G1, 40 in the case where the evaluation value is not less than G3 but less than G2, and 10 in the case where the evaluation value is less than G3.

[0046] The degree of play associated with dryness of the road surface is equal to 100 in the case where the windshield wiper was at rest for a longer period than a fixed period.

[0047] The number of wiper cycles per minute is used as an evaluation value for road surface dryness and is equal to 70 if the evaluation value is less than R1, equal to 40 if the evaluation value is not less than R1 but less than R2, and equal to 10 if the evaluation value is not less than R2.

[0048] As in Fig. As shown in Figure 4, the exemplary dialogue system determines the degree of leeway associated with a driver's states. Reference models include: "low number of blinks"; "current speed is within an average of previous speeds measured over several road segments"; "current acceleration is within an average of previous accelerations measured over several road segments"; "current deceleration is within an average of previous decelerations measured over several road segments"; and "a fixed period of time has elapsed since a conversation with a passenger."

[0049] The number of eyelashes per minute is used as an evaluation value for the arousal level. The measure of leeway associated with the arousal level is 70 if the evaluation value is less than K1, 40 if the evaluation value is not less than K1 but less than K2, and 10 if the evaluation value is not less than K2.

[0050] The current vehicle speed, calculated as an average of the driver's previous speeds on the current road segment multiplied by 100 (percentage), is used as an evaluation value for the speed at that point in time. The measure of margin associated with the speed at that point in time is 70 if the evaluation value is less than SV1, 40 if it is not less than SV1 but less than SV2, and 10 if it is not less than SV2.

[0051] The acceleration of vehicle 1 is used as an evaluation value for the acceleration at a given time. The measure of margin associated with the acceleration at a given time is 70 if the evaluation value is less than SA1, 40 if the evaluation value is not less than SA1 but less than SA2, and 10 if the evaluation value is not less than SA2. The threshold values ​​SA1 and SA2, which are specified on each of several road segments, are determined based on an average of previous accelerations of the driver on each of the several road segments.

[0052] The deceleration of vehicle 1 is used as an evaluation value for deceleration at a given time. The measure of margin associated with deceleration at a given time is 70 if the evaluation value is less than SD1, 40 if the evaluation value is not less than SD1 but less than SD2, and 10 if the evaluation value is not less than SD2. The threshold values ​​SD1 and SD2, which are specified for each of several road segments, are determined based on an average of previous decelerations of the driver on each of the several road segments.

[0053] The time elapsed since a conversation with a fellow traveler is used as an evaluation value for the interpersonal environment. The degree of leeway associated with the interpersonal environment is 70 if the evaluation value is less than K1, 40 if the evaluation value is less than K1 but not less than K2, and 10 if the evaluation value is less than K2.

[0054] The exemplary dialogue system transmits an input containing information related to the driver's driving behavior or the traffic environment to the driver via notification unit 4 by activating the first agent or the second agent.

[0055] If an input, i.e., a detected value, meets a predefined condition regarding the driver's driving behavior, it is determined that information regarding the driver's driving behavior exists that is linked to the predefined condition and requires transmission to the driver.

[0056] If an input relating to the traffic environment, received via communication unit 3 from an external server, meets a predefined condition, it is determined that information relating to the traffic environment exists that is linked to this condition and requires transmission to the driver.

[0057] If an input relating to a preference, received via communication unit 3 from an external server, meets a predefined condition, it is determined that information relating to the preference exists that is linked to this predefined condition and requires transmission to the driver.

[0058] The exemplary dialogue system provides decision values ​​for various pieces of information linked to driving, the traffic environment, preferences, etc. (see Fig. 5), which are to be transmitted to the driver.

[0059] Based on a measure of the leeway the driver has for the decision value associated with the information, it is determined whether the dialogue system should transmit an input, which in each case contains information, to the driver via notification unit 4 or not.

[0060] For example, the dialog system transmits the input containing the information to the driver via notification unit 4 if the degree of discretion is greater than the decision value associated with the information.

[0061] The input containing the information to be transmitted to the driver differs from the information the driver receives during automatic dialogue control and is presented to the driver in such a way that the driver can recognize the difference. The type of input to be transmitted can be a statement, a question, or a conversation requiring a response.

[0062] Again referring to the Fig. 2 to 4 will be the degree of discretion for each of the characteristics that belong to a category “vehicle behavior” (see Fig. 2) for each of the features that belong to a category “Conditions relating to the vehicle” (see Fig. 3), and for each of the characteristics that belong to a category ‘driver’s conditions’ (see Fig. 4) Each of the specified margin measures is compared successively with all of the decision values. For example, it can be determined that the dialog system should transmit an input to the driver if each of the decision values ​​is less than the specified margin measure. It is determined that the dialog system should not transmit an input to the driver if there is at least one decision value that is not less than the specified margin measure. In other words, the margin measure that has the minimum value of all the specified measures is compared successively with all of the decision values. If all of the decision values ​​are less than the margin measure with the minimum value, it can be determined that the dialog system should transmit an input that contains each of all the pieces of information that would require transmission to the driver.This configuration allows the dialog system to transmit an input containing all the information associated with each of the features when the driver has a margin of error associated with each of the features.

[0063] As in Fig. As shown in Figure 5, the exemplary dialogue system determines varying decision values ​​using information of different types and levels, each contained in an input to be transmitted to the driver. Types of information relating to "driving" include: information relating to "excessive speed"; information relating to "negligent lane keeping"; and information relating to "arousal".

[0064] The duration of the speed limit violation is detected and used as an evaluation value for excessive speed. The decision value associated with excessive speed is 30 if the evaluation value (i.e., the detected duration) is not less than A seconds, 5 if the evaluation value is not less than B seconds, and 1 if the evaluation value is not less than C seconds, where A < B < C.

[0065] The degree of deviation and the number of repeated counts per unit of time are used as an evaluation value for careless lane keeping. This evaluation value is indicative of zigzag movement. The decision value associated with careless lane keeping is 60 if the zigzag movement is minor, 30 if the zigzag movement is average, and 1 if the zigzag movement is large.

[0066] The degree of decline in the driver's biological information is determined and used as an evaluation value for arousal. The decision value associated with arousal is 90 if the degree of decline is low, 60 if the degree of decline is average, and 30 if the degree of decline is high. The driver's biological information is transmitted by a wearable device and received via communication unit 3. The degree of decline in the driver's biological information can be determined using information from driver monitor 85.

[0067] The types of traffic environment include information relating to dynamic contact and information relating to traffic regulations and traffic jam information.

[0068] The probability that people or vehicles will cross a predicted future location of a driver's vehicle, either ahead of or behind it, or alongside it, is predicted and used as an evaluation value for dynamic contact. The decision value associated with dynamic contact is 60 if the probability is low, 30 if the probability is average, and 1 if the probability is high.

[0069] The relationship between traffic regulations and traffic congestion, as well as a predicted future location of the driver's vehicle, is determined from the distance between the predicted future location and a location for the traffic regulations and the traffic congestion. The decision value associated with the traffic regulations and the traffic congestion is 110 if the relationship is weak, 80 if the relationship is average, and 50 if the relationship is strong.

[0070] A driver's preference is inferred from the driver's trip schedule and vehicle navigation. The types of information related to this preference include a spatiotemporally effective service.

[0071] Regarding the spatiotemporally effective service, the duration(s) of various services available within a given area around the predicted future location are determined. The decision value is 140 if the service duration is within D weeks, 120 if it is within E days, and 80 if it is within F hours.

[0072] The Fig. Sections 6 to 8 present in tabular form the conditions that must be met for the exemplary dialogue system to transmit an input to the driver, as indicated by the symbol o, and the conditions that cause the exemplary dialogue system to continue carrying out the automatic dialogue control, as indicated by the symbol x.

[0073] Fig. Figure 9 is a process flow diagram that shows a procedure for determining whether an input should be transmitted to a driver in a vehicle. The procedure can be carried out using process logic, which may include hardware (e.g., decision logic, dedicated logic, programmable logic, and microcode), software (such as software running on a general-purpose computer system or a dedicated machine), or a combination of both. In one exemplary embodiment, the process logic relates to the dialogue system or its components.

[0074] The process begins with step S1, which instructs the dialogue system to determine a degree of leeway that the driver may have, based on a calculation using one of several methods described above with reference to the Fig. 2 to 4 were discussed.

[0075] Next, in step S2, it is determined whether there is information that requires transmission to the driver.

[0076] If, in step S2, it is determined that information exists which requires transmission to the driver, the procedure proceeds to step S3, in which the dialogue system is instructed to determine a decision value associated with a type of information after a calculation using the procedure described above with reference to Fig. 5 was discussed.

[0077] In step S4, it is determined whether the degree of discretion is greater than the decision value associated with the information. If step S4 determines that the degree of discretion is greater than the decision value, the procedure proceeds to step S5, in which the dialog system is instructed to transmit an input containing the information to the driver before this routine is completed.

[0078] If in step S2 it is determined that there is no information that requires transmission to the driver, or if in step S4 it is determined that the margin of error is not greater than the decision value, the procedure proceeds to step S6, in which the dialogue system is instructed to continue the execution of the automatic dialogue control, in which a dialogue is carried out between a first agent and a second agent before this routine is terminated.

[0079] As described, in the exemplary dialogue system, if it is determined that information exists which requires transmission to the driver during the execution of the automatic dialogue control, the system determines whether or not to transmit an input containing the information to the driver based on the result of comparing the degree of discretion the driver may have with a decision value linked to the information.

[0080] The aforementioned configuration allows the exemplary dialogue system to transmit an input containing information to a driver in a vehicle without causing a driving error or a decrease in concentration, since, based on a result of comparing a measure of leeway the driver may have with a decision value linked to the information, it is determined whether the dialogue system should transmit the input to the driver or not, while the driver is provided with some information in an automatic dialogue control mode, in which a dialogue is carried out between a first agent and a second agent.

[0081] Furthermore, the exemplary dialogue system transmits the input containing the information to the driver in the case where the degree of discretion is greater than the decision value associated with the information, while the dialogue system continues the execution of the automatic dialogue control in the case where the degree of discretion is not greater than the decision value.

[0082] The aforementioned configuration allows a dialogue system to transmit an input containing information to a driver in a vehicle without causing a driving error or a decrease in concentration, since the dialogue system transmits the input containing the information directly to the driver when the degree of discretion is greater than the decision value associated with the information, while the dialogue system continues to execute the automatic dialogue control when the degree of discretion is not greater than the decision value.

[0083] In the exemplary dialogue system, the degree of leeway is calculated based on the behavior of the vehicle, the conditions around the vehicle, and the driver's state.

[0084] The aforementioned configuration allows a dialogue system to transmit an input containing information to a driver in a vehicle without causing a driving error or a decrease in concentration, since the degree of leeway is calculated based on the behavior of the vehicle, the conditions around the vehicle, and the driver's state, and whether or not the dialogue system should transmit the input to the driver is determined based on a comparison of the degree of leeway with a decision value associated with the information.

[0085] Furthermore, the decision value associated with the information contained in an input transmitted to a driver is determined based on the type and level of the information. This configuration allows a dialogue system to transmit an input containing information to a driver in a vehicle without causing a driving error or a decrease in concentration, as the conditions that must be met for the dialogue system to transmit the input to the driver are modified by changing the decision value depending on the type and level of information.

[0086] The exemplary dialogue system is modified. In the modified form, the exemplary dialogue system determines varying decision thresholds, linked to each of the measures of discretion, for different types of information contained in the input, and the comparison of a measure of discretion that the driver has with a decision value linked to the information contained in the input involves comparing the decision value linked to the information contained in the input with the decision thresholds linked to the measure of discretion.

[0087] For example, the exemplary dialogue system determines varying decision thresholds, linked to each measure of discretion, for different types of information contained in the input. Examples of these different types include information related to hazard avoidance, information related to the traffic environment, and information related to a preference. Based on a comparison of each of the varying decision thresholds with a decision value associated with a piece of information, the system determines whether or not to transmit an input containing that information to the driver.

[0088] As in the Fig. As shown in Figures 10 to 12, the exemplary dialogue system specifies a group of decision thresholds, linked to each measure of discretion, for information related to hazard avoidance, information related to a traffic environment, and information related to a preference. Based on a comparison of each decision threshold in the corresponding group with a decision value associated with a piece of information, the system determines whether or not to transmit an input containing that information to the driver.

[0089] Referring again to Fig. 5. Information related to driving is classified as information related to avoiding danger, information related to traffic information is classified as information related to traffic information, and information related to a preference is classified as information related to preference.

[0090] In the event that information is found which requires transmission to a driver, the exemplary dialogue system determines a decision value using Fig. 5.

[0091] Furthermore, the exemplary dialogue system determines the degree of discretion the driver may have and defines a set of varying decision thresholds, linked to that degree of discretion, for different types of information, each contained in an input to be transmitted to the driver.

[0092] It is determined that the exemplary dialogue system should transmit an input containing information to the driver if the decision thresholds are greater than the decision value associated with the information.

[0093] Fig. Figure 13 is a process flow diagram showing a modified procedure for determining whether an input should be transmitted to a driver in a vehicle. The procedure can be carried out by process logic, which may include hardware (e.g., decision logic, dedicated logic, programmable logic, and microcode), software (such as software running on a general-purpose computer system or a dedicated machine), or a combination of both. In one exemplary embodiment, the process logic relates to the dialogue system or its components.

[0094] The process begins with step S1, which instructs the dialogue system to determine a degree of leeway that the driver may have, based on a calculation using one of several methods described above with reference to the Fig. 2 to 4 were discussed.

[0095] The process moves to step S11, in which the dialogue system is instructed to determine a set of varying decision thresholds, linked to the specified degree of discretion, for different types of information, each contained in an input to be transmitted to the driver.

[0096] Next, in step S2, it is determined whether there is information that requires transmission to the driver.

[0097] If, in step S2, it is determined that information exists which requires transmission to the driver, the procedure proceeds to step S3, in which the dialogue system is instructed to determine a decision value associated with a type of information after a calculation using the procedure described above with reference to Fig. 5 was discussed.

[0098] In step S12, it is determined whether the decision thresholds of the group associated with the specified degree of discretion are greater than the decision value. If step S12 determines that the degree of discretion is greater than the decision value associated with the information, the procedure proceeds to step S5, in which the dialog system is instructed to transmit an input containing the information to the driver before this routine is completed.

[0099] If in step S2 it is determined that there is no information that requires transmission to the driver, or if in step S12 it is determined that each of the group's decision thresholds is not greater than the decision value, the procedure proceeds to step S6, in which the dialogue system is instructed to continue the execution of the automatic dialogue control, in which a dialogue is performed between a first agent and a second agent before this routine is terminated.

[0100] As described, in the modified exemplary dialogue system, a comparison of decision thresholds of a group, which is linked to a specific degree of discretion, with the specific decision value determines whether the dialogue system should transmit an input containing information to a driver or not.

[0101] The aforementioned configuration allows the modified exemplary dialogue system to transmit an input containing information to a driver in a vehicle without causing a driving error or a decrease in concentration. This is because the varying decision thresholds of a group, linked to a specific degree of discretion, for different types of information contained in an input to be transmitted to the driver, facilitate the modified exemplary dialogue system's transmission of an input containing information related to hazard avoidance. Conversely, it allows the modified exemplary dialogue system to transmit an input containing information related to preference only when the driver has emotional discretion.

[0102] The exemplary dialogue system can use a single notification unit 4 through which the first and second agents transmit an input and a response associated with that input. However, it can also use different notification units 4, one for the first agent and the other for the second agent. These different notification units 4 can contain different speakers if the first and second agents are represented by different tones. In this case, the automatic dialogue control, which executes a dialogue between the first and second agents, aligns the speakers towards each other. In a mode where an input is transmitted to a driver via one of the first and second agents, the speaker for the agent can be oriented towards the driver.

[0103] If the first and second agents are represented by images on a monitor, the automatic dialogue control can align the images with each other. In a mode where input is being transmitted to a driver, the images can be aligned towards the driver.

[0104] If the first and second agents are Averters in a monitor, the automatic dialog control can align the Averters towards each other. In a mode where input is being transmitted to a driver, the Averters can be aligned towards the driver.

[0105] In the exemplary dialogue system, a control unit performs various decisions and calculations based on input from different sensors. This information from various sensors can be sent via a communication unit to an external processing unit. The external processing unit then performs various decisions and calculations based on the information sent by the communication unit. The results of these decisions and calculations are received by the communication unit. Based on these results, the system can then perform various control actions.

[0106] The preceding description specifies that the dialog system should transmit an input to the driver if any of the decision values ​​are less than the specified margin, while it stipulates that the dialog system should not transmit an input to the driver if at least one decision value is not less than the specified margin. The exemplary dialog system can be modified such that the modified exemplary dialog system can transmit an input to the driver if any of the decision values ​​are less than the specified margin. In other words, the margin that has the maximum value of all the specified margins is compared successively with all of the decision values.This configuration allows the modified exemplary dialogue system to transmit an input to the driver in a positive manner when the driver has no emotional leeway whatsoever.

[0107] The exemplary dialogue system can determine the degree of discretion associated with one of the features, or a group of decision thresholds associated with that specific degree of discretion. This configuration allows the determination of the degree of discretion associated with one or more features considered important, without requiring consideration of all features related to vehicle behavior, the surrounding environment, and the driver's state. This avoids complex processing and reduces the number of sensors required. Thus, the exemplary dialogue system can deliver input to the driver without causing a driving error or a decrease in concentration. With this exemplary dialogue system, any of the features can be predetermined or selected by the user.The selection of any of the features by the user(s) allows the user to use the degree of flexibility associated with a suitable feature to adapt to the user's driving habit(s).

[0108] In the exemplary dialogue system, an average of the measures of discretion determined for some or all of the features, and an average of the decision thresholds determined for some or all of the features, can be used for comparison with the decision value(s). With this configuration, the user can obtain the measure of discretion based on a general assessment considering multiple factors.

[0109] In the exemplary dialogue system, the degree of discretion is represented by a coefficient between 0 and 1. The degrees of discretion (each represented by a coefficient) determined for some or all of the features are multiplied to obtain a correction coefficient. This correction coefficient is then multiplied by a base value of a given degree of discretion to produce a modified degree of discretion, which is then compared to the decision value.

[0110] In the examples described, the degree of leeway is determined based on one of the vehicle's behavior, the conditions surrounding the vehicle, and the driver's state; however, it can be determined based on all of these.

[0111] The foregoing description is purely illustrative and is not intended to limit the disclosure, its application, or its uses in any way. The comprehensive teachings of the disclosure can be realized in a multitude of forms. While this disclosure includes specific examples, the actual scope of the disclosure is therefore not limited, as further modifications will become apparent upon examination of the documents of the present application. [Description of reference symbols] 1 vehicle; 2 forward sensors; 3 communication unit; 4 notification units; 5 Control unit; 81 Speed ​​sensor; 82 Accelerometer; 83 Steering angle sensor; 84 GNSS receivers; 85 Driver monitor

Claims

[1] Dialogue system for providing information to a driver in a vehicle (1), wherein the dialogue system performs automatic dialogue control, in which a dialogue is performed between a first agent and a second agent, where, in the case where it is determined that information exists which requires transmission to the driver during the execution of the automatic dialogue control, it is determined, on the basis of a result from a comparison of a measure of discretion that the driver has with a decision value linked to the information, whether or not the dialogue system should transmit an input containing the information to the driver, and the dialogue system transmits the input containing the information to the driver in the case where the degree of discretion is greater than the decision value associated with the information, whereas in the case where the degree of discretion is no greater than the decision value, the dialogue system continues to carry out the automatic dialogue control and indirectly provides information to the driver by conducting the dialogue between the first agent and the second agent. [2] Dialogue system according to claim 1, wherein the dialogue system determines varying decision thresholds, which are linked to each of the measures of discretion, for different types of information contained in the input, and the comparison of a measure of discretion that the driver has with a decision value linked to the information contained in the input includes a comparison of the decision value linked to the information contained in the input with the decision thresholds linked to the measure of discretion. [3] Dialogue system according to claim 1 or 2, wherein the degree of leeway is calculated on the basis of at least one of the behavior of the vehicle (1), conditions around the vehicle (1) and the driver's state. [4] Dialogue system according to one of claims 1 to 3, wherein the decision value associated with the information contained in the input is determined depending on the type and level of the information contained in the input.

Images