Vehicle operation support system

Abstract

To provide a vehicle operation support system 10 capable of establishing a good psychological relation between a driver and a vehicle 1.SOLUTION: A vehicle operation support system 10 comprises: driver psychology calculation means (ECU 27) which calculates driver psychological information 293 indicating a psychological state of a driver with respect to a vehicle 1; a vehicle psychology calculation means (ECU 27) which calculates vehicle psychological information 294 indicating a pseudo psychological state of the vehicle 1 with respect to the driver; language data 28 which stores a plurality of pieces of language information indicating contents of words and actions that a vehicle agent V performs; language information selection means (ECU 27) which selects the language information bringing the driver psychological information 293 and the vehicle psychological information 294 close to a reciprocal state; agent control means (ECU 27) which controls the words and actions of the vehicle agent V based on the language information; and psychological information update means (ECU 27) which updates the driver psychological information 293 and the vehicle psychological information 294 by repeatedly calculating the same.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to a vehicle driving assistance system that assists a driver in driving a vehicle by communicating with the driver through the words and actions of a vehicle agent, which is an avatar of the vehicle. [Background technology]

[0002] In automobiles and other vehicles, known vehicle driving assistance systems that assist the driver in driving operations include, for example, advanced driving assistance systems that partially intervene in the driver's driving operations, and autonomous driving systems that take over driving operations from the driver under certain conditions and perform autonomous driving.

[0003] Furthermore, in vehicle driving assistance systems, there are also known technologies that provide driving assistance by providing guidance on driving operations according to the conditions around the vehicle, and guidance on handing over driving operations between the driver and the vehicle, not only through voice and display but also through a vehicle agent, which is an avatar of the vehicle.

[0004] However, if the driver does not trust the driving assistance provided via the vehicle agent, for example, the driving assistance provided by the vehicle driving assistance system may not function effectively to ensure the safety of the driver while driving. Therefore, a technology has been proposed that simulates emotions in a vehicle and builds a good psychological relationship between the driver and the vehicle via a vehicle agent.

[0005] For example, in Patent Document 1, when the vehicle driving assistance system is not intervening and the shift position is in the parking range, a vehicle agent that personifies the vehicle faces the driver and behaves in accordance with voice guidance, thereby building a good psychological relationship between the driver and the vehicle.

[0006] However, in Patent Document 1, the vehicle agent builds a one-sided psychological relationship with the driver, regardless of whether the driver accepts the vehicle agent's words and actions or the intervention of the vehicle's driving assistance. For this reason, Patent Document 1 does not necessarily build a good psychological relationship between the driver and the vehicle via the vehicle agent, and there is room for improvement. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2023-19243 Summary of the Invention [Problem to be solved by the invention]

[0008] In view of the above-mentioned problems, an object of the present invention is to provide a vehicle driving assistance system that can build a good psychological relationship between the driver and the vehicle. [Means for solving the problem]

[0009] The present invention is a vehicle driving assistance system that assists a driver's driving operation via a vehicle agent that communicates with the driver of the vehicle, and includes a driver psychology calculation means that calculates driver psychology information that indicates the driver's psychological state with respect to the vehicle using at least one of a driver state detection means that detects the state of the driver and history information in which a history of driving operations by the driver is registered, a vehicle psychology calculation means that calculates vehicle psychology information that indicates a pseudo-psychological state of the vehicle with respect to the driver using at least one of the driver state detection means and the history information, and a plurality of speech and behavior information that indicate the content of the speech and behavior that the vehicle agent is to perform. The system is characterized by comprising: a behavioral information storage means for storing behavioral information; a behavioral information selection means for selecting from the behavioral information storage means, based on the current driver psychological information and the current vehicle psychological information, the behavioral information that brings the driver psychological information and the vehicle psychological information closer to a reciprocal state; an agent control means for controlling the behavior of the vehicle agent based on the behavioral information selected by the behavioral information selection means; and a psychological information update means for repeatedly calculating the driver psychological information by the driver psychology calculation means and the vehicle psychological information by the vehicle psychology calculation means, thereby updating the current driver psychological information and the current vehicle psychological information.

[0010] The vehicle agent refers to, for example, a three-dimensional model displayed on a display means inside the vehicle, or a robot or the like placed inside the vehicle. Assisting the driver's driving operation means partially intervening in the driver's driving operation to control the operation of the vehicle, or taking over the driving operation from the driver to control the operation of the vehicle.

[0011] Detecting the driver's state means detecting the driver's vocalizations, detecting the driver's movements, detecting the driver's heart rate, detecting the driver's amount of sweat, and detecting the driver's amount of exercise. The history of the driver's driving operations refers to a history of the driver's driving dates and times, a history of the driving behavior of the driver based on the driver's driving operations, and a history of compliance with traffic rules.

[0012] Controlling the behavior of the vehicle agent includes not only controlling the voice output and the operation based on the behavior information, but also controlling the timing of the control based on the behavior information. The above-mentioned mutually beneficial state refers to a state in which the driver and the vehicle are mutually friendly and dependent on each other.

[0013] According to this invention, the current driver psychological information and the current vehicle psychological information are successively updated, so that the speech and behavior information selected by the speech and behavior information selection means can be successively changed. Therefore, the vehicle agent can vary its speech and behavior toward the driver over time, thereby inducing psychological changes in the driver's psychological state and the vehicle's simulated psychological state.

[0014] In this case, the behavior of the vehicle agent is controlled based on speech and behavior information that brings the driver's psychological information and the vehicle's psychological information closer to a reciprocal state, so the vehicle driving assistance system can bring about psychological changes in the driver's psychological state and the vehicle's pseudo-psychological state that bring them closer to a reciprocal state.

[0015] Therefore, the vehicle driving assistance system can prevent the psychological relationship between the driver and the vehicle from becoming a one-sided relationship built between the vehicle and the driver, and can build a good psychological relationship between the driver and the vehicle. Therefore, the vehicle driving assistance system can, for example, encourage the driver to favorably accept the intervention of driving assistance, thereby ensuring the safety of the driver while driving.

[0016] As an aspect of the present invention, the driver psychology calculation means may calculate the driver's degree of intimacy with the vehicle and the driver's degree of dependence on the vehicle as the driver psychological information, and the vehicle psychology calculation means may calculate the vehicle's degree of intimacy with the driver and the vehicle's degree of dependence on the driver as the vehicle psychological information.

[0017] According to this configuration, the psychological state of the driver and the simulated psychological state of the vehicle can be calculated with high accuracy. Furthermore, since the driver's psychological state and the vehicle's pseudo-psychological state are calculated using the same index, the vehicle driving assistance system can easily compare the driver's psychological state with the vehicle's pseudo-psychological state.

[0018] This makes it easier for the vehicle driving assistance system to select speech and behavior information that brings the driver's psychological information and the vehicle's psychological information closer to a mutually beneficial state, thereby enabling a good psychological relationship to be built between the driver and the vehicle accurately and efficiently.

[0019] As another aspect of the present invention, the behavioral information selection means may be configured to select the behavioral information that increases the driver's dependence on the vehicle when the driver's current dependence on the vehicle is lower than the reciprocal state.

[0020] With this configuration, even if the driver is in a driver-driven psychological state where he or she prioritizes his or her own judgment without relying on driving assistance via the vehicle agent, the words and actions of the vehicle agent can bring the driver's psychological state closer to a reciprocal state. Therefore, the vehicle driving assistance system can reliably build a good psychological relationship between the driver and the vehicle.

[0021] In addition, as an aspect of the present invention, a mapping means may be provided for mapping the current driver psychological information and the current vehicle psychological information on a virtual two-dimensional coordinate plane formed by a first axis indicating the degree of dependence and a second axis indicating the degree of intimacy.

[0022] According to this configuration, for example, by outputting driver psychological information and vehicle psychological information mapped on a virtual two-dimensional coordinate plane to the vehicle's display means, the driver can be objectively informed of the current psychological relationship between the driver and the vehicle. Therefore, the vehicle driving assistance system can bring about psychological changes in the driver that bring the driver closer to a reciprocal state by using the driver psychological information and vehicle psychological information mapped onto a virtual two-dimensional coordinate plane. [Effects of the Invention]

[0023] The present invention can provide a vehicle driving assistance system that can build a good psychological relationship between the driver and the vehicle. [Brief explanation of the drawings]

[0024] [Figure 1] FIG. 2 is a front view showing the exterior of the vehicle driving assistance system as seen from inside the vehicle cabin. [Figure 2] FIG. 1 is a block diagram showing the internal configuration of a vehicle driving assistance system. [Figure 3] FIG. 2 is a schematic explanatory diagram illustrating an outline of speech and behavior information. [Figure 4] FIG. 2 is an explanatory diagram illustrating the relationship between the driver's psychological state and the pseudo-psychological state of the vehicle agent. [Figure 5] 10 is a flowchart showing a processing operation relating to the control of a vehicle agent. [Figure 6] 10 is a flowchart showing the processing operations of psychological information update processing. [Figure 7] 5 is a flowchart showing the processing operation of a driver's psychology estimation process. [Figure 8] FIG. 10 is an explanatory diagram illustrating an outline of a closeness calculation process. [Figure 9] 10 is a flowchart showing the processing operation of a vehicle pseudo-mental state estimation process. [Figure 10] FIG. 10 is a schematic explanatory diagram of a driver's psychology estimation means according to another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0025] An embodiment of the present invention will be described below with reference to the drawings. In this embodiment, a vehicle driving assistance system 10 that assists the driver's driving operation via a vehicle agent V that communicates with the driver of a vehicle 1 will be described with reference to FIGS. 1 to 4. FIG.

[0026] Note that Figure 1 shows a front view of the vehicle driving assistance system 10 as seen from inside the cabin of the vehicle 1, Figure 2 shows a block diagram of the vehicle driving assistance system 10, Figure 3 shows a schematic diagram outlining the speech and behavior data 28, and Figure 4 shows an explanatory diagram explaining the relationship between the driver's psychological state and the pseudo-psychological state of the vehicle agent V.

[0027] First, as shown in FIG. 1, the interior of a vehicle 1 is provided with an instrument panel 2 extending in the vehicle width direction (reference numerals omitted), and a center console 3 extending from approximately the center of the instrument panel 2 in the vehicle width direction toward the rear of the vehicle. Furthermore, the vehicle 1 is provided with a steering wheel 4 in the cabin, which is disposed in front of a driver's seat (not shown) and receives steering operations from a passenger seated in the driver's seat.

[0028] As shown in FIG. 2, the vehicle driving assistance system 10 in such a vehicle 1 includes an engine 11, a brake device 12, and a steering device 13 arranged outside the vehicle cabin, a display unit 14 provided at approximately the center of the instrument panel 2 in the vehicle width direction, and a speaker 15 arranged inside the vehicle cabin.

[0029] Furthermore, as shown in FIG. 2, the vehicle driving assistance system 10 is equipped with an accelerator opening sensor 16, a brake pedal stroke sensor 17, a vehicle speed sensor 18, a steering angle sensor 19, an acceleration sensor 20, and a yaw rate sensor 21 for detecting the driving behavior of the vehicle.

[0030] In addition, as shown in FIG. 2, the vehicle driving assistance system 10 includes an exterior camera 22 and a millimeter wave radar 23 for acquiring information about the area in front of the vehicle, an interior camera 24 for detecting the driver's condition, an activity sensor 25, and a microphone 26, and an electronic control unit (EC) for controlling the operation of each of the above-mentioned components. The ECU 27 is a control unit (hereinafter referred to as "ECU") that controls the speed of the vehicle.

[0031] More specifically, the engine 11 is made up of an engine body (not shown in detail) and a fuel injection device that supplies fuel to the engine body, etc. The engine output and engine speed of the engine 11 are controlled based on an output signal from the ECU 27.

[0032] The braking device 12 is configured with a brake caliper, a brake actuator, etc. (details of which are omitted from the drawing). The braking device 12 is controlled based on an output signal from the ECU 27 so as to generate a braking force.

[0033] The steering device 13 is composed of the above-mentioned steering wheel 4, an electric motor (not shown) that assists the driver in steering, etc. The steering device 13 is controlled to change the traveling direction of the vehicle based on an output signal from the ECU 27.

[0034] The display unit 14 is configured, for example, with a liquid crystal display, and has a function of displaying various information to a passenger, such as a driver, seated in the driver's seat, based on control signals from the ECU 27. For example, as shown in Fig. 1, the display unit 14 displays a vehicle agent V that communicates with the driver as an entity of the vehicle 1.

[0035] The speaker 15 is disposed inside the vehicle cabin and has the function of outputting a warning sound or voice based on a notification signal from the ECU 27 to notify the occupants. For example, the speaker 15 outputs a voice corresponding to the operation of the vehicle agent V displayed on the display unit 14.

[0036] The accelerator position sensor 16 is a sensor that is integrally provided on an accelerator pedal operated by the driver. The accelerator position sensor 16 has a function of detecting the displacement angle of the accelerator pedal caused by the driver's operation and a function of outputting the detected displacement angle to the ECU 27 as an accelerator position signal.

[0037] The brake pedal stroke sensor 17 is a sensor that is integrally provided on the brake pedal operated by the driver. The brake pedal stroke sensor 17 has a function of detecting the displacement angle of the brake pedal operated by the driver and a function of outputting the detected displacement angle to the ECU 27 as a brake signal.

[0038] The vehicle speed sensor 18 also has a function of detecting the rotation speed of a drive shaft connected to the wheels, for example, and a function of outputting the detected rotation speed to the ECU 27 as a rotation speed signal. The steering angle sensor 19 has a function of detecting the steering angle of the steering wheel 4 operated by the driver, and a function of outputting the detected steering angle to the ECU 27 as a steering angle signal.

[0039] In addition, the acceleration sensor 20 is arranged, for example, inside the center console 3, and has the function of detecting the longitudinal acceleration along the fore-and-aft direction of the vehicle 1 and the lateral acceleration along the width direction of the vehicle 1, and the function of outputting the detected longitudinal acceleration and lateral acceleration to the ECU 27 as acceleration signals.

[0040] The yaw rate sensor 21 is disposed, for example, inside the center console 3, and has a function of detecting the yaw rate of the vehicle 1 and a function of outputting the detected yaw rate to the ECU 27 as a yaw rate signal. The vehicle exterior camera 22 is disposed, for example, at an upper portion of the windshield on the passenger compartment side, and has a function of capturing an image in front of the vehicle and a function of outputting the captured video image to the ECU 27 .

[0041] The millimeter wave radar 23 is disposed, for example, in the front bumper, and has the functions of emitting millimeter waves as irradiated waves in front of the vehicle, receiving reflected waves reflected by preceding vehicles, parked vehicles, pedestrians, etc., and outputting information obtained from the irradiated waves and reflected waves to the ECU 27 as a radar signal.

[0042] 1, the in-vehicle camera 24 is provided on the instrument panel 2 or a front pillar (reference numeral omitted) facing the driver seated in the driver's seat. The in-vehicle camera 24 has a function of capturing an image of the driver seated in the driver's seat, focusing on the head, and a function of outputting the captured video to the ECU 27.

[0043] In addition, the activity sensor 25 is built into, for example, the steering wheel 4, and has the function of detecting the heart rate, amount of sweating, and amount of exercise of the driver holding the steering wheel 4, and the function of outputting an activity signal indicating the detected heart rate, amount of sweating, and amount of exercise to the ECU 27.

[0044] In addition, the microphone 26 is arranged on the top of the column cover (reference numeral omitted), for example, as shown in FIG. 1, and has the function of detecting the voice emitted by the driver seated in the driver's seat and the function of outputting the detected voice to the ECU 27 as an audio signal.

[0045] The ECU 27 is made up of hardware components such as a CPU, a memory, and a storage unit for storing various information, and software components such as programs and data. This ECU 27 has the function of acquiring various signals output by the above-mentioned sensors, the exterior camera 22, the millimeter-wave radar 23, the interior camera 24, the activity sensor 25, and the microphone 26, and the function of performing various information processing based on the acquired signals. Furthermore, the ECU 27 has a function of controlling the operation of the display unit 14 and the speaker 15 , and a function of controlling the operation of the engine 11 , the brake device 12 and the steering device 13 .

[0046] More specifically, the ECU 27 realizes the functions of calculating the accelerator opening in cooperation with the accelerator opening sensor 16, calculating the amount of brake pedal depression in cooperation with the brake pedal stroke sensor 17, and calculating the vehicle speed of the vehicle 1 in cooperation with the vehicle speed sensor 18.

[0047] Furthermore, the ECU 27 realizes a function of calculating the steering angle of the steering wheel 4 in cooperation with the steering angle sensor 19, and a function of detecting the driving behavior of the vehicle 1 in cooperation with the acceleration sensor 20 and the yaw rate sensor 21.

[0048] In addition, the ECU 27 has the function of identifying the situation in front of the vehicle in cooperation with the exterior camera 22 and millimeter wave radar 23, and the function of identifying the driver seated in the driver's seat and identifying the driver's condition in cooperation with the interior camera 24, activity sensor 25 and microphone 26.

[0049] Such an ECU 27 is configured to alert the driver and notify them of the need for driving assistance intervention through the words and actions of the vehicle agent V depending on the driving behavior and the situation ahead of the vehicle, and to control the operation of the engine 11, braking device 12, and steering device 13 to provide driving assistance that assists the driver in driving operations.

[0050] The vehicle agent V is an entity of the vehicle 1 that is set with pseudo-emotions, and its speech and behavior are controlled by the ECU 27 so as to act and speak in accordance with the driving behavior and the situation ahead of the vehicle.

[0051] In order to realize driving assistance via this vehicle agent V, the ECU 27 stores speech and behavior data 28 in which the lines and actions of the vehicle agent V are registered, and driver data 29 in which various information for each driver is registered, as shown in Figure 2.

[0052] As shown in FIG. 3, the speech and behavior data 28 includes a plurality of pieces of speech and behavior information that associate the lines to be output to the speaker 15 as the voice of the vehicle agent V with the actions that the vehicle agent V takes in response to the lines. In the speech and behavior data 28, even if the content indicates a greeting at the start, for example, a plurality of pieces of speech and behavior information are registered, which are expressed differently depending on the degree of dependence and the degree of intimacy, which will be described later.

[0053] Furthermore, although detailed illustration is omitted, the speech and behavior information is assigned a flag indicating whether the expression increases or decreases intimacy, as described below, and a flag indicating whether the expression increases or decreases dependence.

[0054] Specifically, the behavioral data 28 includes the behavioral and verbal information of the vehicle agent V at the time of starting or during normal times when driving assistance is not being provided, the behavioral and verbal information of the vehicle agent V regarding driving assistance, and the behavioral and verbal information of the vehicle agent V in response to the driver's driving operations.

[0055] For example, as shown in Figure 3, the speech and behavior data 28 includes registered speech and behaviors such as "Good morning..." spoken by the vehicle agent V to a driver with a low level of intimacy and "Stand still" as speech and behaviors at the time of starting, etc., and speech and behaviors such as "Good morning! Thank you for today" and "Expression of joy" spoken by the vehicle agent V to a driver with a high level of intimacy.

[0056] Furthermore, for example, the speech and behavior data 28 stores, as speech and behavior when sudden braking is performed under the control of the ECU 27, lines such as "You're driving carelessly" and an action of "expressing anger" for drivers with low intimacy, and lines such as "I'm surprised, I'm nervous" and an action of "expressing surprise" for drivers with high intimacy.

[0057] Furthermore, for example, speech and behavior data 28 includes registered speech and behaviors to request driving operations by the driver, such as "There is construction ahead, so I need help changing lanes," and an action called "expression of distress."

[0058] In addition, for example, in the speech and behavior data 28, when the driver's response to the speech and behavior of the vehicle agent V is detected, the line "Thank you for operating it" and the action "Stay still" are registered as speech and behaviors for a driver with a low level of intimacy, and the line "Thank you for helping me" and the action "Expression of joy" are registered as speech and behaviors for a driver with a high level of intimacy.

[0059] On the other hand, the driver data 29 stores behavior history information 291 in which the driving behavior during driving and the like are registered, driving history information 292 in which the date and time when the driver drove and the like are registered, driver psychological information 293 indicative of the driver's psychological state regarding the vehicle 1, and vehicle psychological information 294 indicative of the simulated psychological state of the driver of the vehicle 1.

[0060] Specifically, the behavior history information 291 chronologically registers the angular velocity of the accelerator pedal, the angular velocity of the brake pedal, the angular velocity of the steering wheel 4, the acceleration change, the yaw rate, etc. Furthermore, the behavior history information 291 also registers the number of sudden operations (for example, sudden acceleration, sudden braking, or sudden steering) calculated by the ECU 27 based on the various registered information.

[0061] In addition, the driving history information 292 stores the date and time when the driver drove and the distance traveled, the number of times the driver drove in the most recent month calculated by the ECU 27 based on the date and time when the driver drove and the distance traveled, and the number of days that have passed since the last time the driver drove, in association with each other.

[0062] In addition, the driver psychology information 293 registers the degree of familiarity of the driver with the vehicle 1 calculated by the ECU 27 and the degree of dependency of the driver on the vehicle 1 calculated by the ECU 27 in association with each other. In addition, the vehicle psychological information 294 registers the degree of intimacy with the driver of the vehicle 1 calculated by the ECU 27 and the degree of dependence on the driver of the vehicle 1 calculated by the ECU 27 in association with each other.

[0063] As shown in Figure 4, this driver psychological information 293 and vehicle psychological information 294 are mapped as the driver's psychological state and the pseudo-psychological state of the vehicle 1, respectively, on a virtual two-dimensional coordinate plane with the horizontal axis representing dependency and the vertical axis representing intimacy, and can be output to the display unit 14.

[0064] In addition, if the intersection P of the horizontal axis indicating the degree of dependence and the vertical axis indicating the degree of intimacy in Figure 4 is taken as the median of the degree of dependence and the median of the degree of intimacy, the initial value of the driver psychological information 293 is set so that the degree of intimacy of the driver psychological information 293 is approximately the same as the median of the intimacy, and the degree of dependence of the driver psychological information 293 is lower than the median of the degree of dependence. On the other hand, the initial value of the vehicle psychological information 294 is set so that the intimacy of the vehicle psychological information 294 is approximately the same as the median value of the intimacy, and the dependency of the vehicle psychological information 294 is higher than the median value of the dependency.

[0065] Here, the degree of familiarity of the driver with the vehicle 1 and the degree of dependency of the driver on the vehicle 1, and the degree of familiarity of the vehicle 1 with the driver and the degree of dependency of the vehicle 1 on the driver will be described in more detail.

[0066] First, the driver's intimacy with vehicle 1 is an indicator of the driver's psychological state estimated by ECU 27 based on behavior history information 291 and driving history information 292, and indicates the degree of intimacy and attachment the driver feels toward vehicle 1.

[0067] The intimacy of this driver with vehicle 1 is determined by the psychological state of the driver who feels a sense of familiarity or attachment to vehicle 1 being considered to be an intimate state, with a high intimacy level, and conversely, the psychological state of the driver who does not feel a sense of familiarity or attachment to vehicle 1 being considered to be an estranged state, with a low intimacy level (see Figure 4).

[0068] In this embodiment, it is assumed that the more familiar and attached the driver is to the vehicle 1, the less abrupt the driver's driving will be and the more opportunities to drive it will be, and the more abrupt the driver's driving will be, the more intimate the state will be, and the more abrupt the driver's driving will be and the fewer opportunities to drive it will be, the more distant the state will be.

[0069] In addition, the driver's dependence on vehicle 1 is an indicator of the driver's psychological state estimated by ECU 27 based on the driver's words and actions in response to the words and actions of vehicle agent V, and indicates the extent to which the driver depends on driving assistance via vehicle agent V.

[0070] The degree of dependency of this driver on the vehicle 1 is high when the driver's psychological state is such that he or she is easily reliant on driving assistance via the vehicle agent V, as this state is considered to be subordinate to the vehicle 1, while conversely, the degree of dependency is low when the driver's psychological state is such that he or she does not rely on driving assistance via the vehicle agent V and prioritizes his or her own judgment, as this state is considered to be dominant (see Figure 4).

[0071] In this embodiment, if the driver responds to the words and actions of the vehicle agent V by replying or nodding, the vehicle is considered to be subordinate to the vehicle 1, and if the driver does not respond to the words and actions of the vehicle agent V, the driver is considered to be dominant.

[0072] In addition, the degree of intimacy of vehicle 1 with the driver is an index showing the pseudo-psychological state of vehicle 1 estimated by ECU 27 based on the driver's words and actions, behavior history information 291 and driving history information 292, and indicates the degree of intimacy vehicle 1 feels with the driver's words and actions.

[0073] The intimacy level of the vehicle 1 with the driver is set high when the driver's driving behavior is appropriate to the situation around the vehicle and gives a sense of respect, or when the vehicle 1 is in a pseudo-psychological state where the driver treats the vehicle with affection, as an intimate state. Conversely, the intimacy level is set low when the driver's driving behavior is not favorable to the situation around the vehicle or when the vehicle 1 is not treated with affection, as an estranged state (see Figure 4).

[0074] In this embodiment, for example, when no driving assistance is being provided, the driver's driving operation is considered to be intimate with the situation around the vehicle to a degree that is favorable, and the driver's driving operation is considered to be distant with respect to the situation around the vehicle to a degree that is unfavorable.

[0075] In addition, the degree of dependence of vehicle 1 on the driver is an indicator showing the pseudo-psychological state of vehicle 1 estimated by ECU 27 based on the driver's response to the words and actions of vehicle agent V, and indicates the extent to which the driver has accepted driving assistance via vehicle agent V.

[0076] The degree of dependence of this vehicle 1 on the driver is high when the pseudo-psychological state of vehicle 1 in which the driver has not accepted the driving assistance via the vehicle agent V is considered to be a state subordinate to the driver, and low when the pseudo-psychological state of vehicle 1 in which the driver has accepted the driving assistance via the vehicle agent V is considered to be a state in which vehicle 1 is dominant (see Figure 4).

[0077] In this embodiment, when the driver does not respond to the behavior of the vehicle agent V, the vehicle is considered to be subordinate to the driver, and when the driver responds to the behavior of the vehicle agent V, the vehicle is considered to be leading.

[0078] Next, the processing operation of the ECU 27 in the vehicle driving assistance system 10 having the above-described configuration will be described with reference to FIGS. FIG. 5 shows a flowchart relating to the control of the vehicle agent V, and FIG. 6 shows a flowchart of the psychological information updating process.

[0079] Furthermore, FIG. 7 shows a flowchart of the driver's psychology estimation process, FIG. 8 shows an explanatory diagram outlining the intimacy calculation process, and FIG. 9 shows a flowchart of the vehicle pseudo-psychology estimation process.

[0080] First, the processing operation of the ECU 27 for controlling the behavior of the vehicle agent V will be described. When power is supplied to the ECU 27 from the battery, the ECU 27 executes a predetermined program stored therein and starts a processing operation for controlling the behavior of the vehicle agent V.

[0081] The ECU 27, which has started processing operations to control the behavior of the vehicle agent V, identifies the driver sitting in the driver's seat based on video images from the in-vehicle camera 24 and audio signals from the microphone 26 (step S101), as shown in Figure 5, and reads the behavior data 28 and driver data 29 corresponding to the identified driver (step S102).

[0082] Furthermore, after displaying the vehicle agent V on the display unit 14 (step S103), the ECU 27 determines whether it is time to have the vehicle agent V perform normal speech and behavior, such as greetings at start-up (step S104).

[0083] If it is not the timing to make the vehicle agent V speak or act normally (step S104: No), the ECU 27 determines whether it is the timing to make the vehicle agent V speak or act regarding driving assistance (step S105).

[0084] For example, the ECU 27 determines whether it is the right time to warn the driver when the distance between the vehicle and the vehicle ahead becomes narrow, the right time to provide guidance to the driver regarding his driving operation, or the right time to forcibly intervene in the driver's driving operation and provide guidance after controlling the operation of the vehicle 1.

[0085] If it is not the time to have the vehicle agent V make any statements or actions related to driving assistance (step 105: No), the ECU 27 identifies the driver sitting in the driver's seat based on video images from the in-vehicle camera 24, and determines whether the identified driver is the same as the driver identified in step S101 (step S106).

[0086] If the driver is not the same (step S106: No), the ECU 27 returns the process to step S101 and repeats the processes from step S101 to step S108, which will be described later, until the supply of electric power is interrupted.

[0087] On the other hand, if the driver is the same (step S106: Yes), the ECU 27 returns the process to step S104 and repeats the processes from step S104 to step S108, which will be described later, until the supply of electric power is interrupted.

[0088] Also, in step S104 of Figure 5, if it is time to have the vehicle agent V behave in a normal manner (step S104: Yes), or if it is time to have the vehicle agent V behave in a manner related to driving assistance (step S105: Yes), the ECU 27 determines the behavior to be performed by the vehicle agent V based on the behavior data 28 (step S107).

[0089] At this time, the ECU 27 determines the lines and actions of the vehicle agent V based on the driver psychological information 293 and the vehicle psychological information 294 so as to bring the driver's current psychological state indicated by the driver psychological information 293 and the vehicle 1's current pseudo-psychological state indicated by the vehicle psychological information 294 closer to a state in which the other is viewed favorably and is not overly dependent on the other (hereinafter referred to as a reciprocal state).

[0090] In addition, the intersection P of the horizontal axis indicating the degree of dependence and the vertical axis indicating the degree of intimacy in Figure 4 is taken as the median of the degree of dependence and the median of the degree of intimacy, and an example of a reciprocal state in this embodiment is a state in which the degree of dependence is at the median and the degree of intimacy is higher than the median, as shown in Figure 4.

[0091] As an example, in step S107, it is assumed that the intimacy of the current driver to the vehicle 1 is lower than the intimacy indicating a reciprocal state, and the dependency of the current driver on the vehicle 1 is lower than the median dependency (see FIG. 4).

[0092] In this case, based on the flags assigned to the behavioral information (lines and actions) of the behavioral data 28, the ECU 27 extracts and determines from the behavioral data 28 lines and actions of the vehicle agent V that are suitable for increasing both the driver's intimacy with the vehicle 1 and the driver's dependence on the vehicle 1.

[0093] Thereafter, the ECU 27 outputs a voice based on the lines determined in step S107 from the speaker 15, and operates the vehicle agent V displayed on the display unit 14 based on the action determined in step S107 (step S108). Then, the ECU 27 proceeds to step S106 and repeats the processes from step S101 to step S108 until the supply of electric power is interrupted.

[0094] In this way, the vehicle driving assistance system 10 assists the driver in driving by having the vehicle agent V act and speak in a manner determined based on the driver psychological information 293 and the vehicle psychological information 294.

[0095] At this time, the vehicle driving assistance system 10 successively updates the driver psychological information 293 and the vehicle psychological information 294 at appropriate timing in order to bring the driver's psychological state and the pseudo-psychological state of the vehicle 1 closer to a mutually beneficial state.

[0096] Next, a psychological information updating process for updating the driver psychological information 293 and the vehicle psychological information 294 that determine the behavior of the vehicle agent V will be described. When power is supplied to the ECU 27 from the battery, the ECU 27 executes a predetermined program stored therein to start a psychological information update process for updating the driver psychological information 293 and the vehicle psychological information 294 .

[0097] As shown in FIG. 6, the ECU 27, which has started the psychological information update process, identifies the driver sitting in the driver's seat based on the video image from the in-vehicle camera 24 and the audio signal from the microphone 26 (step S111), and reads the speech and behavior data 28 and the driver data 29 corresponding to the identified driver (step S112). Note that the data in steps S111 and S112 may be obtained in steps S101 and S102 in FIG. 5 described above.

[0098] Furthermore, after starting to record the driving behavior of the vehicle 1 in the behavior history information 291 (step S113), the ECU 27 performs a driver psychology estimation process (step S114) to estimate the driver's psychological state and a vehicle pseudo-psychological estimation process (step S115) to estimate the pseudo-psychological state of the vehicle 1 in parallel processing.

[0099] Specifically, the ECU 27, which has started the driver's psychology estimation process, determines whether or not the intimacy has been calculated at least once since the engine 11 was started, for example, as shown in FIG. 7 (step S121). If the degree of intimacy has been calculated once (step S121: Yes), the ECU 27 proceeds to step S123, which will be described later.

[0100] On the other hand, if the intimacy has not been calculated (step S121: No), the ECU 27 performs an intimacy calculation process to calculate the intimacy of the driver with the vehicle 1 based on the behavior history information 291 and the driving history information 292 (step S122).

[0101] Here, the driver's familiarity with vehicle 1 is calculated based on a driving roughness index value indicating the roughness of the driver's driving operation, a driving frequency index value indicating the number of times the driver has driven per past month, and a driving interval index value indicating the number of days elapsed since the last time the driver drove, as shown in Figure 8.

[0102] The driving roughness index value is calculated by dividing the actual value obtained by dividing the number of sudden operations (such as sudden braking or sudden steering) during one drive by the distance traveled during one drive, and then dividing the actual value from the previous drive by the average actual value from the previous drive to the drive before that. Specifically, the degree of familiarity of the driver with the vehicle 1 is calculated by subtracting the driving interval index value from the sum of the roughness index value and the driving frequency index value, as shown in FIG.

[0103] After calculating the degree of familiarity of the driver with the vehicle 1, the ECU 27 determines whether the current state of the vehicle 1 is during driving assistance via the vehicle agent V, as shown in FIG. 7 (step S123). If driving assistance via the vehicle agent V is not being performed (step S123: No), the ECU 27 ends the driver's psychology estimation process and proceeds to step S116 in FIG.

[0104] On the other hand, if driving assistance is being provided via the vehicle agent V (step S123: Yes), the ECU 27 determines whether or not a response other than the driver's driving operation has been detected when the vehicle agent V makes a statement or action related to driving assistance (step S124).

[0105] For example, the ECU 27 identifies whether the driver has responded favorably to the voice of the vehicle agent V based on the audio signal from the microphone 26, and if the driver has responded favorably, determines that a response other than the driver's driving operation has been detected.

[0106] Alternatively, the ECU 27 determines whether the driver nods in response to the words and actions of the vehicle agent V based on video images from the in-vehicle camera 24, and if the driver nods, determines that a response other than the driver's driving operation has been detected.

[0107] Alternatively, the ECU 27 detects a change in the driver's activity level based on the activity level signal from the activity level sensor 25, and when the driver's activity level increases, determines that a response other than the driver's driving operation has been detected.

[0108] If the driver's response to the vehicle agent V's words and actions cannot be detected (step S124: No), the ECU 27 determines whether the driver performed appropriate driving operations in response to the vehicle agent V's words and actions regarding driving assistance (step S125).

[0109] If the driver is not performing an appropriate driving operation (step S125: No), the ECU 27 determines that the driver has little interest in the behavior of the vehicle agent V and is ignoring the driving assistance provided through the vehicle agent V.

[0110] Therefore, the ECU 27 reduces the current degree of dependency of the driver on the vehicle 1 and temporarily stores it as a new degree of dependency (step S126), then ends the driver psychology estimation process and proceeds to step S116 in FIG.

[0111] On the other hand, if the ECU 27 detects in step S124 a response from the driver to the behavior of the vehicle agent V (step S124: Yes), or if the driver performs appropriate driving operations in step S125 (step S125: Yes), the ECU 27 increases the driver's current degree of dependence on the vehicle 1 and temporarily stores it as a new degree of dependence (step S127) because the driver is dependent on driving assistance via the vehicle agent V. Thereafter, the ECU 27 ends the driver's psychology estimation process and proceeds to step S116 in FIG.

[0112] Furthermore, when the vehicle pseudo-mental state estimation process is started in step S115 of FIG. 6, the ECU 27 determines whether the current state of the vehicle 1 is during driving assistance via the vehicle agent V, as shown in FIG. 9 (step S131). If driving assistance via the vehicle agent V is not being performed (step S131: No), the ECU 27 determines whether the driver is behaving in a manner that is favorable for the vehicle 1 (step S132).

[0113] For example, ECU 27 identifies whether the driver is operating in compliance with traffic rules based on the driving behavior of the vehicle and the situation ahead of the vehicle, and if the driver is operating in compliance with traffic rules, it determines that the driver is behaving in a manner that is favorable for vehicle 1.

[0114] Alternatively, the ECU 27 determines whether the driver has responded favorably to the voice of the vehicle agent V based on the audio signal from the microphone 26, and if the driver has responded favorably, determines that the driver has behaved in a manner that is favorable to the vehicle 1.

[0115] Alternatively, the ECU 27 identifies whether the driver has nodded in response to the words and actions of the vehicle agent V based on video images from the in-vehicle camera 24, and if the driver has nodded, determines that the driver has behaved in a manner that is favorable to the vehicle 1; or identifies whether the driver has spoken to the vehicle agent V based on audio signals from the microphone 26, and if the driver has spoken, determines that the driver has behaved in a manner that is favorable to the vehicle 1.

[0116] If the driver is behaving in a manner that is favorable to the vehicle 1 (step S132: Yes), the ECU 27 determines that the pseudo-psychological state of the vehicle 1 is an intimate state in which the driver feels intimacy with the driver's behavior. Therefore, the ECU 27 increases the current intimacy with the driver of the vehicle 1 and temporarily stores it as a new intimacy (step S133), then ends the vehicle pseudo-mentation estimation process, and proceeds to step S116 in FIG.

[0117] On the other hand, if the driver does not behave in a manner that is favorable to the vehicle 1 (step S132: No), the ECU 27 determines that the pseudo-psychological state of the vehicle 1 is an estranged state in which the driver does not feel any intimacy with the driver's behavior. Therefore, the ECU 27 lowers the current intimacy with the driver of the vehicle 1 and temporarily stores it as a new intimacy (step S134), then ends the vehicle pseudo-mentation state estimation process, and proceeds to step S116 in FIG.

[0118] Also, in step S131 of Figure 9, if driving assistance is being provided via the vehicle agent V (step S131: Yes), the ECU 27 determines whether or not a response other than the driver's driving operation has been detected when the vehicle agent V makes a statement or action related to driving assistance, as in step S124 of Figure 7 (step S135).

[0119] If the driver's response to the vehicle agent V's words and actions cannot be detected (step S135: No), the ECU 27 determines whether the driver performed appropriate driving operations in response to the vehicle agent V's words and actions regarding driving assistance (step S136), similar to step S125 in Figure 7.

[0120] If the driver is not performing appropriate driving operations (step S136: No), the ECU 27 determines that the driving assistance provided by the vehicle agent V is ignored by the driver and that the simulated psychological state of the vehicle 1 is subordinate to the driver.

[0121] Therefore, the ECU 27 increases the current degree of dependency on the driver of the vehicle 1 and temporarily stores it as a new degree of dependency (step S137), then ends the vehicle pseudo-mental state estimation process, and proceeds to step S116 in FIG.

[0122] On the other hand, if the ECU 27 detects the driver's response to the vehicle agent V's words and actions in step S135 (step S135: Yes), or if the driver performs appropriate driving operations in step S136 (step S136: Yes), the ECU 27 determines that the simulated psychological state of the vehicle 1 is not dependent on the driver. Therefore, the ECU 27 reduces the current degree of dependency of the driver on the vehicle 1 and temporarily stores it as a new degree of dependency (step S138), then ends the vehicle pseudo-mental state estimation process, and proceeds to step S116 in FIG.

[0123] Returning to FIG. 6, when the driver psychology estimation process in step S114 and the vehicle pseudo-psychology estimation process in step S115 are completed, the ECU 27 updates the driver psychology information 293 and the vehicle psychology information 294 (step S116).

[0124] Specifically, the ECU 27 updates the driver psychology information 293 using the degree of familiarity of the driver with the vehicle 1 and / or the degree of dependency of the driver on the vehicle 1 that are temporarily stored in the driver psychology estimation process in step S114.

[0125] Furthermore, the ECU 27 updates the vehicle psychological information 294 using the degree of intimacy with the driver of the vehicle 1 or the degree of dependence on the driver of the vehicle 1 that is temporarily stored in the vehicle pseudo-psychology estimation process in step S115.

[0126] Thereafter, the ECU 27 identifies the driver sitting in the driver's seat based on the video image from the in-vehicle camera 24, and determines whether the identified driver is the same as the driver identified in step S101 (step S117).

[0127] If the driver is not the same (step S117: No), the ECU 27 returns the process to step S111 and repeats the processes from step S111 to step S117 until the supply of electric power is interrupted.

[0128] On the other hand, if the driver is the same (step S117: Yes), the ECU 27 returns the process to step S114 and step S115, and repeats the processes from step S114 to step S117 until the supply of power is interrupted.

[0129] In this way, the vehicle driving assistance system 10 successively updates the driver psychological information 293 and the vehicle psychological information 294 that determine the behavior of the vehicle agent V. As a result, the vehicle driving assistance system 10 can change the behavior of the vehicle agent V over time, thereby making it possible to bring the driver's psychological state and the simulated psychological state of the vehicle 1 closer to a mutually beneficial state.

[0130] As described above, the vehicle driving assistance system 10 of this embodiment is a system that assists the driver in driving the vehicle 1 via the vehicle agent V that communicates with the driver.

[0131] This vehicle driving assistance system 10 is equipped with a driver psychology calculation means (ECU 27) that calculates driver psychology information 293 indicating the driver's psychological state toward the vehicle 1 using at least one of a driver state detection means (in-vehicle camera 24, activity sensor 25, microphone 26) that detects the driver's state, and history information (behavior history information 291, driving history information 292) that registers the history of driving operations by the driver.

[0132] Furthermore, the vehicle driving assistance system 10 is equipped with a vehicle psychology calculation means (ECU 27) that calculates vehicle psychology information 294 indicating a simulated psychological state for the driver of the vehicle 1 using at least one of a driver state detection means (in-vehicle camera 24, activity sensor 25, microphone 26) and history information (behavior history information 291, driving history information 292).

[0133] Furthermore, the vehicle driving assistance system 10 includes speech and behavior data 28 that stores a plurality of pieces of speech and behavior information (lines and actions) that indicate the content of speech and behavior that the vehicle agent V is to perform.

[0134] In addition, the vehicle driving assistance system 10 is equipped with a behavioral information selection means (ECU 27) that selects behavioral information from the behavioral data 28 based on the current driver psychological information 293 and the current vehicle psychological information 294, which brings the driver psychological information 293 and the vehicle psychological information 294 closer to a reciprocal state, and an agent control means (ECU 27) that controls the behavior of the vehicle agent V based on the behavioral information selected by the behavioral information selection means.

[0135] The vehicle driving assistance system 10 is equipped with a psychological information update means (ECU 27) that repeatedly calculates the driver psychological information 293 using the driver psychological calculation means and the vehicle psychological information 294 using the vehicle psychological calculation means, thereby updating the current driver psychological information 293 and the current vehicle psychological information 294, respectively.

[0136] According to this configuration, the current driver psychological information 293 and the current vehicle psychological information 294 are successively updated, so that the speech and behavior information (words and actions) selected by the speech and behavior information selection means can be successively changed. Therefore, the vehicle agent V can change its speech and behavior toward the driver over time, thereby causing psychological changes in the driver's psychological state and the pseudo-psychological state of the vehicle 1.

[0137] In this case, the behavior of the vehicle agent V is controlled based on speech and behavior information that brings the driver psychological information 293 and the vehicle psychological information 294 closer to a reciprocal state, so the vehicle driving assistance system 10 can bring about psychological changes in the driver's psychological state and the pseudo-psychological state of the vehicle 1 that bring them closer to a reciprocal state.

[0138] Therefore, the vehicle driving assistance system 10 can prevent the psychological relationship between the driver and the vehicle 1 from becoming a one-sided relationship established between the driver and the vehicle 1, and can build a good psychological relationship between the driver and the vehicle 1. Therefore, the vehicle driving assistance system 10 can ensure the safety of the driver while driving by, for example, encouraging the driver to favorably accept the intervention of driving assistance.

[0139] The driver psychology calculation means (ECU 27) is configured to calculate the degree of intimacy of the driver with the vehicle 1 and the degree of dependency of the driver on the vehicle 1 as driver psychology information 293. On the other hand, the vehicle psychology calculation means (ECU 27) is configured to calculate the degree of intimacy of the vehicle 1 with the driver and the degree of dependency of the vehicle 1 on the driver as vehicle psychology information 294.

[0140] According to this configuration, the psychological state of the driver and the pseudo psychological state of the vehicle 1 can be calculated with high accuracy. Furthermore, since the driver's psychological state and the pseudo-psychological state of the vehicle 1 are calculated using the same index, the vehicle driving assistance system 10 can easily compare the driver's psychological state with the pseudo-psychological state of the vehicle 1.

[0141] This makes it easier for the vehicle driving assistance system 10 to select speech and behavior information (words and actions) that brings the driver psychological information 293 and the vehicle psychological information 294 closer to a mutually beneficial state, thereby enabling a good psychological relationship to be built between the driver and the vehicle 1 accurately and efficiently.

[0142] In addition, the behavioral information selection means (ECU 27) is configured to select behavioral information (words and actions) that will increase the driver's dependence on the vehicle 1 if the driver's current dependence on the vehicle 1 is lower than the reciprocal state. According to this configuration, even if the driver is in a driver-led psychological state where he or she prioritizes his or her own judgment without relying on driving assistance via the vehicle agent V, the driver's psychological state can be brought closer to a reciprocal state through the words and actions of the vehicle agent V. Therefore, the vehicle driving assistance system 10 can reliably build a good psychological relationship between the driver and the vehicle 1.

[0143] The vehicle driving assistance system 10 also includes a mapping means (ECU 27) for mapping the current driver psychological information 293 and the current vehicle psychological information 294 on a virtual two-dimensional coordinate plane formed by a first axis indicating the degree of dependence and a second axis indicating the degree of intimacy.

[0144] According to this configuration, for example, by outputting driver psychological information 293 and vehicle psychological information 294 mapped on a virtual two-dimensional coordinate plane to the display means of vehicle 1, the driver can be objectively informed of the current psychological relationship between the driver and vehicle 1. Therefore, the vehicle driving assistance system 10 can bring about a psychological change in the driver that brings the driver closer to a reciprocal state by using the driver psychological information 293 and the vehicle psychological information 294 mapped onto the virtual two-dimensional coordinate plane.

[0145] In correspondence between the configuration of this invention and the above-mentioned embodiment, The driver state detection means of the present invention corresponds to the in-vehicle camera 24, the activity sensor 25, and the microphone 26 of the embodiment. Similarly, The history information corresponds to driving history information 292 and behavior history information 291, The driver psychology calculation means, the vehicle psychology calculation means, the speech and behavior information selection means, the agent control means, the psychology information update means, and the mapping means correspond to the ECU 27. The speech and behavior information corresponds to the lines and actions of the speech and behavior data 28, The speech and behavior information storage means corresponds to speech and behavior data 28, The present invention is not limited to the configurations of the above-described embodiments, and many other embodiments can be obtained.

[0146] For example, in the above embodiment, the vehicle agent V is a character displayed on the display unit 14, but the present invention is not limited to this and may be, for example, a robot placed on the instrument panel 2. Furthermore, the vehicle driving assistance system 10 is not limited to the configuration described above, and may include, for example, a camera, millimeter wave radar, or a sensor that detects the situation behind or to the sides of the vehicle. Furthermore, the vehicle driving assistance system 10 may include, for example, a navigation system, which assists the driver in driving operations based on the vehicle position and map data.

[0147] In addition, as one of the means for detecting the driver's state, an activity sensor 25 for detecting the driver's activity level is provided on the steering wheel 4, but this is not limited to this, and the activity sensor 25 may also be provided on a wearable terminal worn by the driver. Furthermore, the speech and behavior data 28 shown in FIG. 3 is an example, and is not limited to this, and may have a configuration in which appropriate speech and behavior information (lines and actions) is registered.

[0148] Furthermore, the processing operation of the driver psychology estimation process in Figure 7 is an example, and is not limited to the above-mentioned embodiment, and the driver's intimacy with the vehicle 1 and the driver's dependence on the vehicle 1 may be calculated based on appropriate conditions.

[0149] For example, similar to the vehicle pseudo-mental state estimation process, it may be possible to determine whether the driver is behaving in a manner that is favorable to the vehicle 1, and calculate the degree of intimacy of the driver with the vehicle 1. Alternatively, it may be possible to combine the calculation of the degree of intimacy based on the driving history information 292 and the behavior history information 291 with the calculation of the degree of intimacy based on whether the driver is behaving in a manner that is favorable to the vehicle 1.

[0150] Furthermore, for example, when a driving assistance system forcibly intervenes without any prior warning while the driver is driving, the degree of dependency of the driver on the vehicle 1 may be calculated based on the driver's behavior in response to the forcible intervention of the driving assistance system. Alternatively, in the driver psychology estimation process of FIG. 7, the driver psychology calculation means for calculating the driver psychology information 293 may be constructed using a machine learning model.

[0151] In this case, the machine learning model is constructed so that, as shown in Figure 10, which shows a schematic diagram of a driver psychology estimation means in another embodiment, behavior history information 291, driving history information 292, video images from the exterior camera 22 and the interior camera 24, and various signals from the millimeter-wave radar 23, activity sensor 25, and microphone 26 are used as input information, and the driver's degree of intimacy with the vehicle 1 and the driver's degree of dependence on the vehicle 1 are obtained as output information in response to the input information.

[0152] The machine learning model is constructed by repeatedly learning, as training data for input information, information indicating the driver's driving behavior that is appropriate for the situation in front of the vehicle, information indicating the driver's preferred driving behavior, and information indicating the driver's behavior that is appropriate for the vehicle agent V's behavior.

[0153] Furthermore, the processing operation of the vehicle pseudo-psychology estimation process in Figure 9 is an example, and is not limited to the above-mentioned embodiment, and the degree of intimacy with the driver of vehicle 1 and the degree of dependence on the driver of vehicle 1 may be calculated based on appropriate conditions.

[0154] For example, the degree of intimacy with the driver of the vehicle 1 may be calculated during driving assistance via the vehicle agent V. Furthermore, for example, when a driving assistance system forcibly intervenes without any prior warning while the driver is driving, the degree of dependency of the vehicle 1 on the driver may be calculated based on the driver's behavior in response to the forcible intervention of the driving assistance system.

[0155] Alternatively, in the vehicle pseudo-psychology estimation process of FIG. 9, the vehicle psychology calculation means for calculating the vehicle psychology information 294 may be constructed using a machine learning model.

[0156] In this case, the machine learning model, like the machine learning model constituting the driver psychology estimation means described above, is constructed so that input information is behavior history information 291 and driving history information 292, video images from the exterior camera 22 and the interior camera 24, and various signals from the millimeter-wave radar 23, activity sensor 25 and microphone 26, and the output information in response to the input information is the degree of intimacy with the driver of vehicle 1 and the degree of dependence on the driver of vehicle 1.

[0157] The machine learning model is constructed by repeatedly learning, as training data for input information, information indicating the driver's driving behavior that is appropriate for the situation in front of the vehicle, information indicating the driver's preferred driving behavior, and information indicating the driver's behavior that is appropriate for the vehicle agent V's behavior.

[0158] Also, although not mentioned in the above-described embodiment, in the driver psychology estimation process, the driver's dependence on the vehicle 1 may be calculated each time driving assistance via the vehicle agent V intervenes, or may be calculated at an appropriate timing when driving assistance is not intervening.

[0159] For example, while the driver is driving the vehicle 1, information indicating the behavior of the vehicle agent V and information indicating the behavior of the driver may be sequentially stored in the ECU 27, and the next time the driver drives the vehicle 1, the driver's dependence on the vehicle 1 may be calculated based on the stored information.

[0160] Similarly, in the vehicle pseudo-mental state estimation process, the degree of dependence on the driver of the vehicle 1 and the degree of intimacy with the driver of the vehicle 1 may also be calculated at appropriate timing. Even with this configuration, the same effects as those of the above-described embodiment can be achieved.

[0161] Although not mentioned in the above embodiment, the driver psychological information 293 and the vehicle psychological information 294 may be mapped on a virtual two-dimensional coordinate plane with the horizontal axis representing the degree of dependence and the vertical axis representing the degree of intimacy, and displayed on the display unit 14.

[0162] Furthermore, the speech and behavior information (words and actions) is selected according to the degree of intimacy and dependence of the driver on the vehicle 1 indicated by the driver psychological information 293 and the degree of intimacy and dependence of the vehicle 1 on the driver indicated by the vehicle psychological information 294, but this is not limitative. For example, the ECU 27 may control the operation timing of the vehicle agent V according to the degree of intimacy and dependence of the driver on the vehicle 1 indicated by the driver psychological information 293. [Explanation of symbols]

[0163] 1...Vehicle 10...Vehicle driving assistance system 24...In-car camera 25...Activity sensor 26...Mike 27...ECU 291...Behavior history information 292...Driving history information 293...Driver Psychological Information 294...Vehicle Psychological Information V…Vehicle Agent

Claims

1. A vehicle driving assistance system that assists a driver of a vehicle with a driving operation via a vehicle agent that communicates with the driver, a driver psychology calculation means for calculating driver psychology information indicating the driver's psychological state regarding the vehicle using at least one of a driver state detection means for detecting the state of the driver and history information in which a history of driving operations by the driver is registered; a vehicle psychology calculation means for calculating vehicle psychology information indicating a pseudo-psychological state of the driver of the vehicle using at least one of the driver state detection means and the history information; a speech and behavior information storage means for storing a plurality of speech and behavior information items indicating the content of speech and behavior to be performed by the vehicle agent; a speech and behavior information selection means for selecting, based on the current driver psychological information and the current vehicle psychological information, from the speech and behavior information storage means, the speech and behavior information that brings the driver psychological information and the vehicle psychological information closer to a reciprocal state; agent control means for controlling the behavior of the vehicle agent based on the behavior information selected by the behavior information selection means; and a psychological information updating means for repeatedly calculating the driver psychological information by the driver psychological information calculation means and the vehicle psychological information by the vehicle psychological information calculation means, and updating the current driver psychological information and the current vehicle psychological information, respectively. Vehicle driving assistance system.

2. The driver psychology calculation means calculating, as the driver psychological information, a degree of familiarity of the driver with the vehicle and a degree of dependency of the driver on the vehicle; The vehicle psychology calculation means The vehicle psychological information is calculated based on the degree of intimacy between the vehicle and the driver and the degree of dependency between the vehicle and the driver. The vehicle driving assistance system according to claim 1 .

3. The speech and behavior information selection means When the current degree of dependency of the driver on the vehicle is lower than the degree of dependency of the driver in the reciprocal state, the speech and behavior information that increases the degree of dependency of the driver on the vehicle is selected. The vehicle driving assistance system according to claim 2 .

4. a mapping means for mapping the current driver psychological information and the current vehicle psychological information on a virtual two-dimensional coordinate plane formed by a first axis indicating the degree of dependency and a second axis indicating the degree of intimacy; The vehicle driving assistance system according to claim 2 .

Images