JP7705343B2 - Information provision support method, information provision support device, and information provision support program - Google Patents

Description

The present invention relates to an information provision support method, an information provision support device, and an information provision support program that support the provision of information from a vehicle to a driver.

Patent Document 1 proposes a driving support device that places a personified agent on the dashboard inside the vehicle and controls the behavior of the agent to simulate the danger avoidance actions that the driver should take.

JP 2018-32204 A

By notifying the driver of driving assistance information through the actions and voices of a personified agent as described above, it is expected that the driver's understanding of the driving assistance information will improve compared to when the information is mechanically displayed as text or output as voice.

On the other hand, depending on the vehicle's driving conditions, it is expected that the driver's focus on driving operations will narrow their field of vision, reducing awareness of the agent.

In view of these circumstances, the present invention aims to improve the driver's recognition of the agent.

According to one aspect of the present invention, there is provided an information provision support method for providing information to a vehicle driver through the actions of an agent based on a predetermined character motif. In this information provision support method, vehicle state information including the vehicle speed is acquired, and based on the vehicle state information, a distinction is made between when the vehicle is moving and when the vehicle is stopped. Agent position control is performed to adjust the agent position based on the distinction result. Then, in the agent position control, the agent position is adjusted to a predetermined first position when the vehicle is stopped, and when the vehicle is moving, the agent position is adjusted to a second position that is closer to the driver's gaze point than the first position.

The present invention can improve the driver's recognition of the agent.

FIG. 1 is a block diagram showing a schematic configuration of an information provision support system. FIG. 2 is a block diagram showing the configuration of the agent control device. FIG. 3 is a flow chart illustrating one aspect of agent position control. FIG. 4 is a diagram for explaining an example of adjustment of the robot position during forward travel and forward AR display. FIG. 5 is a diagram for explaining an example of a front screen display and rear position adjustment during reverse driving. FIG. 6 is a diagram illustrating an example of position control during turning. FIG. 7 is a diagram for explaining the action and effect of adjusting the robot position when moving forward and displaying the forward AR. FIG. 8 is a diagram illustrating an example of the relationship between the separation distance and the reaction sensitivity.

The following describes an embodiment of the present invention with reference to the drawings.

[First embodiment]
1 is a block diagram showing a schematic configuration of an information provision support system 101. As shown in the figure, the information provision support system 101 mainly includes various components for supporting the driving operation of a driver 10 of a vehicle 100.

In particular, the information provision support system 101 includes sensors 12, a navigation system 14, an agent control device 18, and an agent 16.

The sensors 12 are composed of one or more sensors that measure various state parameters inside or outside the vehicle 100. In particular, the sensors 12 in this embodiment include an in-vehicle camera 12a for capturing images of the inside of the vehicle, a vehicle speed sensor 12b for detecting the forward or reverse speed (vehicle speed V), a steering angle sensor 12c for detecting the steering angle θ, a microphone 12d for collecting sounds from inside the vehicle, and a shift sensor 12e for detecting the shift position (parking range, neutral range, drive range, etc.). The sensors 12 output various detection signals to the agent control device 18.

The navigation system 14 is a device realized by various hardware and software for guiding the driver 10 with information such as the route to the destination. In particular, the navigation system 14 is equipped with a unique input interface (such as a touch panel or operation buttons) that is expected to be operated by the occupants of the vehicle 100, including the driver 10. In response to operations by the occupants via the input interface, the navigation system 14 provides the occupants with information such as the route to the destination in a predetermined output form (screen display and/or audio output).

The agent 16 is a user interface that assists in providing information from the vehicle 100 to the driver 10. In particular, the agent 16 in this embodiment includes a robot 16a as an actual structure based on a specific real or virtual character, and an AR (Augmented Reality) display body 16b and a screen display body 16c as virtual display bodies that virtually display the character.

Examples of characters that serve as motifs for the robot 16a include real or virtual people, animals, plants, and anthropomorphized objects. The robot 16a is disposed in a position in front of the driver 10 in the vehicle cabin (e.g., on the dashboard). Furthermore, the robot 16a of this embodiment is equipped with a driving unit 19 for displacing the robot 16a within a predetermined range along the front-rear direction in the vehicle cabin (hereinafter referred to as the "front-rear movable range R _m1 "). The driving unit 19 can be configured, for example, with a rail extending in the direction in which the robot 16a should be displaced, and a predetermined driving source such as a motor for moving the robot 16a on the rail.

On the other hand, the AR display body 16b and the screen display body 16c are configured by two-dimensional or three-dimensional digital images imitating a predetermined character, similar to the robot 16a. In particular, the AR display body 16b is displayed in a predetermined forward AR display area R _arf or a rearward AR display area R _arr in the vehicle cabin. The forward AR display area R _arf is configured, for example, as at least a part of the area on the front windshield of the vehicle 100. The rearward AR display area R _arr is configured, for example, as at least a part of the area on the rear windshield of the vehicle 100. The forward AR display area R _arf and the rearward AR display area R _arr may be configured as display areas in a head-up display disposed in front and rear positions of the driver 10, respectively. Moreover, the screen display body 16c is displayed in a predetermined forward screen display area R _dif . The forward screen display area R _dif is configured, for example, as a display area in an in-vehicle display disposed in front of the driver 10.

As described above, since the agent 16 is configured based on a specific character motif, having the agent 16 execute a specific action (such as a movement and output voice) has the effect of eliciting certain emotions (attachment and familiarity) from the driver 10 toward the agent 16. Therefore, by notifying the driver 10 of driving assistance information via the action of the agent 16, the driver's 10 awareness of the driving assistance information can be improved compared to a simple text display or voice output.

The agent control device 18 is a control unit for executing the actions of the agent 16 and the agent position control described below.

Figure 2 is a block diagram showing the configuration of the agent control device 18. As shown in Figure 2, the agent control device 18 includes a vehicle state recognition unit 21, a driver state recognition unit 22, an agent position control unit 24, and an action control unit 26.

The vehicle state recognition unit 21 acquires various detection values obtained by the sensors 12 and route information obtained from the navigation system 14 as vehicle state information, and judges the vehicle state from the vehicle state information. In particular, the vehicle state recognition unit 21 acquires the vehicle speed V detected by the vehicle speed sensor 12b as vehicle state information, and determines whether the vehicle 100 is stopped or traveling based on this. The vehicle state recognition unit 21 also acquires the shift position detected by the shift sensor 12e as vehicle state information (particularly travel direction information), and determines whether the vehicle 100 is in a forward traveling state or a reverse traveling state based on this. Furthermore, the vehicle state recognition unit 21 acquires the vehicle speed V during forward traveling as vehicle state information, and determines whether the vehicle 100 is in a low vehicle speed state or a high vehicle speed state, which will be described later, based on this.

The vehicle state recognition unit 21 also acquires images from the forward camera as vehicle state information (particularly route visibility information) and estimates the poor visibility state of the driver 10 (hereinafter referred to as the "poor visibility state") based on this. Furthermore, the vehicle state recognition unit 21 acquires the steering angle θ as vehicle state information (particularly turning information) and estimates the turning state of the vehicle 100 based on this.

The driver state recognition unit 22 acquires various detection values obtained by the sensors 12 as driver state information and judges the driver state from the driver state information. In particular, the driver state recognition unit 22 acquires images from the in-vehicle camera 12a as driver state information (particularly viewing direction information) and, based on this, estimates whether the driver 10 is facing forward or backward in the vehicle 100 when traveling in reverse.

The driver state recognition unit 22 also estimates the degree of intimacy of the driver 10 with the agent 16 from the driver state information. The degree of intimacy is defined as a parameter that indexes the strength of the driver's 10 attachment to the agent 16. For example, the degree of intimacy can be appropriately determined from any quantity that can suggest the strength of the driver's 10 attachment, such as the total driving range of the vehicle 100 and the total operating time of the agent 16.

The agent position control unit 24 executes agent position control based on the vehicle state and driver state estimated by the vehicle state recognition unit 21 and driver state recognition unit 22, respectively. In the agent position control, the agent position p is determined appropriately according to the estimated vehicle state and driver state. Details of the agent position control will be described later. The agent position control unit 24 then operates the agent 16 to achieve the determined agent position p. More specifically, the drive unit 19 is operated to adjust the position of the robot 16a, or the position of the AR display unit 16b or screen display unit 16c. Meanwhile, the agent position control unit 24 outputs the determined agent position p to the action control unit 26.

The functions of the agent control device 18 described above can be realized by any on-board computer, such as a vehicle controller (not shown) that comprehensively controls the operation of the vehicle 100, or a specific computer outside the vehicle, such as a smartphone owned by the driver 10. In particular, a program (information provision assistance program) that realizes the functions of the agent control device 18 is stored in any memory area accessible by the arithmetic device of the agent control device 18.

The action control unit 26 controls the action of the agent 16 based on the action command and the agent position p in response to the driving support information to be provided to the driver 10. The action of the agent 16 means a process for notifying the driver 10 of the desired driving support information through the behavior of the character represented by the robot 16a, the AR display 16b, or the screen display 16c. Specifically, the action of the agent 16 includes notification by the movement of the character, such as making the robot 16a, the AR display 16b, or the screen display 16c execute a movement simulating a danger avoidance movement, and notification by voice output simulating the voice of the character, such as outputting support information while moving the mouth of the character represented by the robot 16a, the AR display 16b, or the screen display 16c.

Furthermore, the action control unit 26 appropriately adjusts the mode of the action of the agent 16 (such as the type of movement, and the content and sound level of the audio output) according to the agent position p. Then, the action control unit 26 operates the agent 16 to realize the specified action of the agent 16. More specifically, the action control unit 26 operates various actuators (such as the drive actuator of the robot 16a, the AR display processing device, the screen display processing device, and the speaker) that realize the movement and/or audio output of the robot 16a, the AR display body 16b, or the screen display body 16c according to the specified action.

Next, we will explain the details of agent position control in this embodiment.

Figure 3 is a flowchart explaining the agent position control. Note that each process shown in Figure 3 is repeatedly executed at a predetermined calculation period.

As shown in FIG. 3, in step S101, the agent control device 18 estimates (recognizes) the vehicle state and the driver state.

In step S102, the agent control device 18 judges whether the vehicle 100 is running. Specifically, if the vehicle speed V is equal to or greater than a predetermined first threshold value _Vth1 suitable for distinguishing between when the vehicle 100 is stopped and when the vehicle 100 is running, it is judged to be running, and if the vehicle speed V is less than the first threshold value _Vth1 , it is judged not to be running (i.e., stopped). The first threshold value _Vth1 is set to a preferable value (e.g., 5 km/h) for distinguishing between a substantially stopped state and a running state of the vehicle 100. If the agent control device 18 judges that the vehicle 100 is running, it executes the process of step S103, and if it judges that the vehicle 100 is stopped, it executes the process of step S113.

In step S103, the agent control device 18 determines whether the vehicle 100 is traveling forward. Specifically, it is determined that the vehicle 100 is traveling forward when the shift position in the drive (D) range. If the agent control device 18 determines that the vehicle 100 is traveling forward, it executes the process of step S104, and if it determines that the vehicle 100 is not traveling forward, it executes the process of step S107.

In step S104, the agent control device 18 judges whether the vehicle 100 is in a low vehicle speed state. Specifically, when the vehicle speed V is equal to or less than a predetermined second vehicle speed threshold _Vth2 , it is judged to be in a low vehicle speed state, and when the vehicle speed V exceeds the second vehicle speed threshold _Vth2, it is judged not to be in a low vehicle speed state (i.e., in a high vehicle speed state). The second vehicle speed threshold _Vth2 is experimentally determined in advance as the vehicle speed V at which the visual field region VR of the driver 10 is narrowed to such an extent that the robot 16a is not recognized even if the robot 16a is moved to the most forward position within the forward/ _rearward movable range Rm1 of the robot 16a. The second vehicle speed threshold _Vth2 is not limited to a specific value, but is set to, for example, 100 km/h.

In step S105, the agent control device 18 determines whether the driver 10 of the vehicle 100 is in a good visibility state. Specifically, the agent control device 18 performs a predetermined image processing on the image from the front camera to set a parameter that indexes the visibility of the driver 10, and determines whether the current state is a good visibility state or a poor visibility state based on the comparison result between the parameter and a predetermined threshold. Specific examples of a state that is determined to be a poor visibility state include, for example, an intersection with a blind spot, a curved road with a certain or greater curve radius, a sloped road, bad weather, or a driving scene on a narrow alley. Specific examples of a state that is determined to be a good visibility state include, for example, a driving road with good visibility such as a main road. In order to determine the parameter that indexes the visibility of the driver 10, other information such as route information of the driving route of the vehicle 100 may be referenced.

Then, the agent control device 18 executes the process of step S106 if the judgment results of steps S104 and S105 are both positive, and executes the process of step S110 if either of these judgment results is negative.

In step S106, the agent control device 18 executes forward robot position adjustment. Specifically, in forward robot position adjustment, the position of the robot 16a (agent position p) is adjusted to a forward adjustment position _pf . Here, the forward adjustment position _pf in the forward robot position adjustment is defined as a position that is gradually shifted forward in response to an increase in the vehicle speed V with respect to the reference robot position _Pb when the vehicle is stopped, within the aforementioned longitudinal movable range _Rm1 .

The reference robot position _Pb can be set to a position closest to the driver 10 within the longitudinal movable range _Rm1 . More specifically, the reference robot position _Pb can be set within the reach of the driver 10, such as a position near an in-vehicle display disposed in front of the driver 10. It is also preferable to adjust the reference robot position _Pb based on the intimacy degree described above. In particular, it is most preferable to determine the reference robot position _Pb so that the higher the intimacy degree, the closer the distance to the driver 10 becomes.

On the other hand, in step S110, the agent control device 18 executes forward AR display. Specifically, in forward AR display, the agent control device 18 displays the AR display body 16b in the forward AR display area R _arf described above.

In particular, in the forward AR display of this embodiment, the display position of the AR display body 16b in the forward AR display area R _arf is adjusted to a forward adjustment position _pf . Here, the forward adjustment position _pf in the forward AR display is defined as a position that is gradually shifted upward from the reference display position _P'b in the forward AR display area R _arf in response to an increase in the vehicle speed V. Note that the reference display position _P'b is defined as a movement base point of the AR display body 16b defined in the forward AR display area R _arf , corresponding to the reference robot position _Pb .

In addition, it is preferable to set the end of the forward movement adjustment position _pf in the forward AR display (i.e., the upper limit of movement of the AR display body 16b) at a position that overlaps with the gaze point O of the driver 10 when driving forward within the forward AR display area R _arf . The position of the gaze point O can be specified as a position that is estimated to be gazed at by the driver 10 on average depending on the driving direction (forward or backward), etc. Alternatively, the line of sight of the driver 10 may be detected from an image of the in-vehicle camera 12a, etc., and the position of the gaze point O may be specified based on the detected line of sight.

Furthermore, while the forward AR display is being performed, the robot 16a may be maintained in the forward-most position within the forward/rearward movable range R _m1 , or may be stored in a specified space provided inside the dashboard or the like by a storage mechanism (not shown).

4 shows one embodiment of forward robot position adjustment and forward AR display during forward travel. As shown in the figure, in a low vehicle speed state ( _Vth1 ≦ V ≦ _Vth2 ), a forward adjustment position _pf that defines the position of the robot 16a is determined so as to gradually shift forward from a reference robot position _Pb when the vehicle is stopped (V < _Vth1 ) in response to an increase in the vehicle speed V. Furthermore, in a high vehicle speed state ( _Vth2 < V), a forward adjustment position _pf that defines the position of the AR display body 16b is determined so as to gradually shift upward from a reference display position _P'b in response to an increase in the vehicle speed V.

Next, we will explain the process when the result of the determination in step S103 is negative (when the vehicle is not traveling forward).

In this case, in step S107, the agent control device 18 determines whether the vehicle 100 is traveling in reverse. Specifically, it is determined that the vehicle is traveling in reverse when the shift position is in the reverse (R) range. If the agent control device 18 determines that the vehicle 100 is traveling in reverse, it executes the process of step S108, and if it determines that the vehicle 100 is not traveling in reverse, it executes the process of step S113.

In step S108, the agent control device 18 executes a reverse basic position adjustment. Specifically, in the reverse basic position adjustment, the position of the robot 16a is adjusted to a reference position (hereinafter referred to as a "reverse reference robot position _Pbr ") that is different from the reference robot position _Pb when traveling forward. Note that the reverse reference robot position _Pbr can be set to a position that is preferable from the viewpoint of allowing the driver 10 to appropriately recognize the robot 16a while traveling backward, for example. Note that the reverse reference robot position _Pbr may be the same as or different from the reference robot position _Pb when traveling forward.

In step S109, the agent control device 18 determines whether the driver 10 is facing forward. This determination is made assuming a scenario in which the driver 10 performs driving operations while facing forward while reversing the vehicle, since the vehicle 100 has a function for displaying an image of the rear of the vehicle on an in-vehicle display disposed in front of the driver 10.

Then, if the agent control device 18 determines that the driver 10 is facing forward, it executes the process of step S111, and if it determines that the driver 10 is not facing forward (the driver 10 is facing backward), it executes the process of step S112.

In step S111, the agent control device 18 executes a forward screen display. Specifically, in the forward screen display, the position of the screen display body 16c (agent position p) is adjusted to a reverse adjustment position p _r . Here, the reverse adjustment position p _r in the forward screen display is defined as a position that is gradually shifted upward from a reference display position P _br1 in the forward screen display area R _dif in response to an increase in the vehicle speed V. The reference display position P _br1 is a movement base point of the screen display body 16c that is determined in the forward screen display area R _dif and corresponds to the reverse reference robot position P _br .

On the other hand, in step S112, which is executed when the driver 10 faces backward, the agent control device 18 executes rear AR display. Specifically, in the rear AR display, the AR display body 16b is displayed in the rear AR display area R _arr . In particular, in the rear AR display of this embodiment, the display position of the AR display body 16b in the rear AR display area R _arr is adjusted to a reverse adjustment position p _r . Furthermore, the reverse adjustment position p _r in the rear AR display is defined as a position that is gradually shifted upward from the reference display position P _br2 in the rear AR display area R _arr in response to an increase in the vehicle speed V. Note that the reference display position P _br2 is a movement base point of the AR display body 16b that is determined in the rear AR display area R _arr and corresponds to the reverse reference robot position P _br .

5 shows an example of the forward screen display area _Rdif and the rear AR display during reverse driving. As shown in the figure, during reverse driving ( _Vth1 <V), the reverse driving adjustment position _pr that defines the display position of the AR display body 16b is determined to gradually shift upward from the reference display position _Pbr1 or _Pbr2 in the forward screen display area _Rdif or _the rear AR display area Rarr as the vehicle speed V increases.

Next, we will explain the processing of step S113, which is executed after a negative determination result in step S102 (when the vehicle 100 is stopped) and one of the following: adjusting the robot position when moving forward (step S106), displaying the forward AR (step S110), displaying the forward screen (step S111), and displaying the rear AR (step S112).

In step S113, the agent control device 18 judges whether the vehicle 100 is in a turning state. Specifically, the agent control device 18 judges that the vehicle 100 is in a turning state when the absolute value of the steering angle θ is equal to _or greater than a predetermined steering angle threshold θth, and judges that the vehicle 100 is in a non-turning state when the absolute value of the steering angle _θ is less than the steering angle threshold θth. The steering angle threshold _θth is set to an appropriate value taking into consideration a steering play that does not substantially contribute to the turning operation of the vehicle 100. Then, when the agent control device 18 judges that the vehicle 100 is in a turning state, it executes the turning position control in step S114.

In step S114, the agent control device 18 executes turning position control. Specifically, in the turning position control, the position of the robot 16a, the position of the AR display body 16b, or the position of the screen display body 16c, which is the agent position p, is adjusted to a turning position p _s . Here, the turning position p _s is defined as a position gradually shifted in the turning direction from the turning reference display position P bs in accordance with an increase in the absolute value of the steering angle θ within the lateral direction movable range R _m2 , the forward screen display area R _dif , or the forward AR display area R _{arf or} the rear AR display area R _arr , which is determined as a moving range of the robot 16a in the lateral direction (turning direction) of the vehicle 100. Note that the lateral direction movable range R _m2 is determined as a movable range in the turning direction of the robot 16a realized by the driving mechanism, on the premise that a driving mechanism for moving the robot 16a in the turning direction is provided. The turning reference display position _Pbs is determined, for example, at a substantially central position in the left-right direction of the vehicle 100 within the forward screen display area R _dif , the forward AR display area R _arf , or the rearward AR display area R _arr .

6 shows one embodiment of the turning position control. As shown in the figure, the turning position p _s during a left turn (steering angle θ<0) is determined to be a position shifted leftward as the steering angle θ becomes smaller in the region of θ≦-θ _th with the turning reference display position P _bs as the base point. On the other hand, the turning position p _s during a right turn (steering angle θ>0) is determined to be a position shifted rightward as the steering angle θ becomes larger in the region of θ≧θ _th with the turning reference display position P _bs as the base point.

The left end _Psma and the right end _Psmi of the turning position _ps determined in the turning position control can be set to suitable positions. For example, when the front AR display area _Rarf is set to substantially the entire surface of the windshield of the vehicle 100, the left end _Psma may be set to a position near the left A-pillar, and the right end _Psmi may be set to a position near the right A-pillar.

Next, we will explain the effects of executing the above-mentioned agent position control (particularly the forward movement processing) using more specific examples.

Figure 7 is a diagram explaining the effect of this embodiment. In particular, Figure 7(a) shows the situation when the vehicle is stopped, Figure 7(b) shows the situation when the vehicle is traveling at a low speed (adjusting the robot position when traveling forward), and Figure 7(b) shows the situation when the vehicle is traveling at a high speed (front AR display).

7A, when the vehicle is stopped, the position of the robot 16a is adjusted to a reference robot position _Pb that is relatively close to the driver 10. In other words, the distance d between the driver 10 and the agent 16 is made relatively short, thereby ensuring the driver 10's recognition of the agent 16.

7B, when the vehicle speed is low, forward robot position adjustment (step S106) is executed, and the robot 16a moves forward within the longitudinal movable range _Rm1 to a forward adjustment position _pf approaching the gaze point O of the driver 10 in response to an increase in the vehicle speed V. As a result, the robot 16a (agent 16) is included in the visual field VR of the driver 10, which is narrowed during forward driving.

Furthermore, as shown in Fig. 7(c), forward AR display is performed when the vehicle is traveling at high speed. In particular, the position of the AR display body 16b in the forward AR display area R _arf changes from the reference display position _P'b to a forward adjustment position _pf approaching the gaze point O as the vehicle speed V increases. This causes the AR display body 16b (agent 16) to be included in the narrower visual field VR of the driver 10 when traveling at high speed.

The following summarizes the effects of the information provision support method according to this embodiment.

According to this embodiment, an information provision support method is provided in which information is provided to the driver 10 of the vehicle 100 by the operation of an agent 16 based on a predetermined character motif. This information provision support method distinguishes whether the vehicle 100 is moving or stopped based on vehicle state information including the vehicle speed V of the vehicle 100 (step S102), and performs agent position control (steps S103 to S112) that adjusts the agent position p based on the distinguishing result.

In the agent position control, the agent position p is adjusted to a predetermined first position (reference robot position _Pb ) when the vehicle is stopped, and when the vehicle is traveling, the agent position p is adjusted to a second position (forward adjustment position _pf or reverse adjustment position _pr ) that is closer to the driver's 10 point of gaze O than the first position.

As a result, when the vehicle 100 is stopped, the agent 16 is positioned relatively close to the driver 10, thereby increasing the driver's 10 recognition of the agent 16, and when the vehicle is traveling, the agent 16 is moved away from the driver 10, thereby maintaining the driver's 10 recognition of the agent 16 even in a situation where the visual field VR is narrowed. Therefore, the driver's 10 recognition of the agent 16 can be appropriately ensured according to the traveling state of the vehicle 100.

The vehicle state information also includes travel direction information (shift position) indicating a forward or reverse state of the vehicle 100. Then, in the agent position control, when it is estimated that the vehicle 100 is moving forward based on the travel direction information, forward travel processing (steps S104 to S106, and step S110) is executed. Then, in the forward travel processing, the second position (forward travel adjustment position p _f ) is determined to be ahead of the driver 10.

This allows the system to properly detect when the vehicle 100 is traveling forward and when the driver 10 is presumed to be viewing the road ahead, and to determine the agent position p at an appropriate position where the agent 16 is within the driver 10's field of vision.

In particular, in the forward travel processing, the second position (forward travel adjustment position _pf ) is determined so as to approach the gaze point O of the driver 10 as the vehicle speed _V increases (see Figs. 4 and 7).

This realizes specific control logic for including the agent 16 in the field of view VR, taking into account that the field of view VR narrows as the vehicle speed V increases.

Furthermore, in this embodiment, the agent 16 includes an actual structure (robot 16a) and a virtual display body (AR display body 16b) displayed within a predetermined virtual display area (forward AR display area R _arf ). The forward movement process includes a first forward position adjustment (forward movement robot position adjustment) performed when the vehicle speed V is equal to or lower than a predetermined vehicle speed threshold (second vehicle speed threshold V _th2 ), and a second forward position adjustment (forward AR display) performed when the vehicle speed V exceeds the second vehicle speed threshold V _th2 .

In the forward robot position adjustment, the second position (forward adjustment position _pf ) is set within the movable range of the robot 16a (within the forward/rearward movable range _Rm1 ). On the other hand, in the forward AR display, the second position (forward adjustment position _pf ) is set within the forward AR display area _Rarf .

As a result, in scenes where the vehicle speed V is relatively low, the actions of the agent 16 are realized using the robot 16a, which is a physical structure, giving the driver 10 a high sense of presence (reality) and improving the driver's 10's recognition of the actions. On the other hand, in scenes where the vehicle speed V is relatively low, the actions of the agent 16 are realized using the AR display 16b, so that the driver's 10's recognition of the actions can be maintained even in situations where the driver's 10 visual field VR is narrowed.

The vehicle state information of this embodiment further includes route visibility information indicating the visibility of the driver 10 on the driving route of the vehicle 100. Then, in the forward travel processing, if it is estimated that the visibility of the driver 10 is equal to or lower than a certain level based on the route visibility information (No in step S105), the second position (forward travel adjustment position p _f ) is determined to be closer to the gaze point O of the driver 10 (No in step S105).

This realizes specific control logic that makes it easier for the driver 10 to recognize the agent 16 in driving situations where the importance of driving assistance information is greater, such as when driving in bad weather or at an intersection with blind spots.

On the other hand, in the agent position control, when it is estimated that the vehicle 100 is moving backwards based on the travel direction information, the reverse movement processing (steps S108, S109, S111, and S112) is executed.

This allows the system to properly detect the reverse driving state of the vehicle 100, where the driver 10 is presumed to be viewing the rear, and to determine the agent position p at an appropriate position where the agent 16 is within the field of vision of the driver 10.

In particular, in the reverse drive process, the direction in which the driver 10 is facing is estimated based on visual direction information (image from the in-vehicle camera 12a) that indicates the direction in which the driver 10 is facing (step S109). If it is estimated that the driver 10 is facing forward, a third forward position adjustment (forward screen display) is performed to set the second position (reverse drive adjustment position p _r ) forward of the driver 10 in the vehicle cabin. On the other hand, if it is estimated that the driver 10 is facing backward, a rearward position adjustment (rear AR display) is performed to set the reverse drive adjustment position p _r behind the driver 10.

This allows the agent position p to be more appropriately determined according to the direction the driver 10 is facing when the driver 10 is driving backwards. More specifically, by mounting a device that displays the rear area of the vehicle 100, such as a so-called rear monitor, at the front of the vehicle interior, the vehicle 100 can set a preferred agent position p according to both a scene in which the driver 10 is facing forward when driving backwards, and a scene in which the driver 10 is facing backwards to directly visually check the rear area of the vehicle 100.

Furthermore, in this embodiment, the vehicle state information includes turning information (steering angle θ) that suggests turning of the vehicle 100. Then, in the agent position control, when the turning state of the vehicle 100 is estimated based on the steering angle θ (Yes in step S113), turning position control (step S114) is further executed. In particular, in the turning position control, the agent position p is shifted a predetermined distance in the turning direction relative to the first position or the second position (see FIG. 6).

This allows the agent position p to be set to an appropriate position where the agent 16 is within the field of vision of the driver 10, depending on the turning direction of the vehicle 100. Therefore, it is possible to ensure that the driver 10 can appropriately recognize the agent 16, even when turning.

Furthermore, the agent position control of this embodiment further executes an intimacy correction process that estimates the intimacy of the driver 10 with the agent 16 and adjusts the first position (reference robot position P _b ) based on the intimacy. In particular, the intimacy correction process determines the distance (separation distance d) from the driver 10 to be closer as the intimacy is higher.

This makes it possible to determine a first position with an optimized separation distance d, taking into consideration the strength of the driver's 10 attachment to the agent 16. This eliminates a situation in which the driver's 10 feels bothered or unsatisfied with the agent 16 and the actual separation distance d does not match, and further promotes mutual communication between the driver 10 and the agent 16.

Furthermore, in this embodiment, an information provision support device (agent control device 18) for executing the above-mentioned information provision support method is provided. This agent control device 18 has a discrimination unit (vehicle state recognition unit 21) that discriminates whether the vehicle 100 is moving or stopped based on vehicle state information including the vehicle speed V of the vehicle 100, and an agent position control unit 24 that adjusts the agent position p based on the discrimination result. In particular, the agent position control unit 24 adjusts the agent position p to a predetermined first position (reference robot position P _b ) when the vehicle is stopped, and adjusts the agent position p to a second position (forward adjustment position p _f or reverse adjustment position _pr ) that is closer to the gaze point O of the driver 10 with respect to the first position when the vehicle is moving.

Furthermore, in this embodiment, an information provision assistance program is provided for causing a computer (agent control device 18) to execute the above-mentioned information provision assistance method. This information provision assistance program causes the agent control device 18 to determine whether the vehicle 100 is moving or stopped based on vehicle state information including the vehicle speed V of the vehicle 100, and executes agent position control for adjusting the agent position p based on the determination result. In particular, the agent position control adjusts the agent position p to a predetermined first position (reference robot position P _b ) when the vehicle is stopped, and adjusts the agent position p to a second position (forward adjustment position p _f or reverse adjustment position _pr ) that is closer to the gaze point O of the driver 10 with respect to the first position when the vehicle is moving.

[Second embodiment]
The second embodiment will be described below. The same elements as those in the first embodiment are given the same reference numerals and their description will be omitted. In particular, in this embodiment, in addition to the agent position control described above, a reaction sensitivity control is further executed to adjust the reaction sensitivity of the agent 16 according to the separation distance d between the agent position p and the driver 10.

8 shows one embodiment of the reaction sensitivity control. In the reaction sensitivity control of this embodiment, the reaction sensitivity of the agent 16 is defined in three stages according to the magnitude of the separation distance d. In particular, when the separation distance d is the first distance threshold _dth1 , a high sensitivity response R _H is selected, when the separation distance d is equal to or greater than the first distance threshold _dth1 and less than the second distance threshold _dth2 , a medium sensitivity response R _M is selected, and when the separation distance d is the second distance threshold _dth2 , a low sensitivity response R _L is selected.

Here, the level of reaction sensitivity can be determined based on the type and magnitude of the behavior of the driver 10 that triggers the action of the agent 16. For example, the low sensitivity response R _L is determined to be the driver's 10's vocalization (such as calling out to the agent 16). The medium sensitivity response R _M is determined to be the driver's 10's gaze (such as looking in the direction of the agent 16) in addition to the driver's 10's vocalization. The high sensitivity response R _H is determined to be the driver's 10's hand movement (such as making a gesture toward the agent 16) in addition to the driver's 10's vocalization and gaze.

In this way, by adjusting the reaction sensitivity of the agent 16 according to the size of the separation distance d, it is possible to more appropriately optimize the mutual communication between the agent 16 and the driver 10.

In particular, in this embodiment, the smaller the separation distance d, the greater the variety of actions of the driver 10 that trigger a response from the agent 16. As a result, in a situation where the driver 10 and the agent 16 are close to each other, such as when the vehicle is stopped (V< _Vth1 ), the agent 16 responds sensitively to the behavior of the driver 10. This can further promote mutual communication (interaction) between the agent 16 and the driver 10.

In addition, the greater the separation distance d, the fewer the types of behaviors of the driver 10 that trigger the action of the agent 16. As a result, during driving (V≧V _th1 ), the more the driver 10 needs to focus on driving (the higher the vehicle speed V), the more limited the types of behaviors of the driver 10 that trigger the response of the agent 16. For this reason, for example, it is possible to prevent the agent 16 from unintentionally inducing an action of the agent 16 due to the body movement associated with the driving operation of the driver 10. In particular, as described above, by defining the low sensitivity response R _L when the separation distance d is large as the voice of the driver 10, it is possible to detect the situation in which the driver 10 truly desires the response of the agent 16 with higher accuracy than when the gaze or hand movement is defined, and it is possible to more reliably prevent the occurrence of an unintentional action of the agent 16.

The above describes an embodiment of the present invention, but the configurations described in the above embodiment merely show some of the application examples of the present invention and are not intended to limit the technical scope of the present invention.

For example, the specific aspects of the forward movement processing, reverse movement processing, turning position control, intimacy correction processing, and reaction sensitivity control described in the above embodiment are merely examples, and may be modified as appropriate within the scope of the technical significance envisaged in these processes.

For example, in the forward movement processing of the above embodiment, an example has been described in which the agent position p is switched from adjusting the position of the robot 16a to adjusting the display position of the AR display body 16b at the timing when the vehicle speed V reaches the second vehicle speed threshold _Vth2 . However, instead of this, a configuration may be adopted in which the forward movement processing is realized only by adjusting the display position of the AR display body 16b, and a structure for moving the robot 16a is omitted.

In addition, the specific aspects of the reaction sensitivity control are not limited to the example described in the second embodiment (FIG. 8), and various modifications are possible. For example, a configuration may be adopted in which the steps of the separation distance d that determines the reaction sensitivity are increased from the example shown in FIG. 8, thereby changing the sensitivity of the response of the agent 16 more finely according to the separation distance d. In addition, in the second embodiment, an example has been described in which the low sensitivity response R _L is defined as the driver 10's voice, the medium sensitivity response R _M is defined as the driver 10's voice and gaze, and the high sensitivity response R _H is defined as the driver 10's voice, gaze, and hand movement. However, the specific aspects of the driver 10's behavior defined as the low sensitivity response R _L , the medium sensitivity response R _M , and the high sensitivity response R _H can be modified as appropriate. For example, a configuration may be adopted in which the low sensitivity response R _L , the medium sensitivity response R _M , and the high sensitivity response R _H are all defined as the driver 10's voice, and thresholds are set in stages for the volume level of the voice, and the threshold is made smaller as the sensitivity response becomes higher.

The information provision support device (agent control device 18) of the above embodiment can be implemented by a so-called AI (Artificial Intelligence) program. The AI program may be referred to as a trained model.

10 Driver, 12 Sensors, 14 Navigation system, 15 Vehicle controller, 16 Agent, 18 Agent control device, 19 Driving unit, 21 Vehicle state recognition unit, 22 Driver state recognition unit, 24 Agent position control unit, 26 Agent operation control unit, 100 Vehicle, 101 Information provision support system

Claims (11)

An information provision support method for providing information to a driver of a vehicle by an action of an agent having a predetermined character motif, comprising:
determining whether the vehicle is moving or stopped based on vehicle state information including a vehicle speed of the vehicle;
Execute an agent position control to adjust the agent position based on the discrimination result;
In the agent position control,
adjusting the agent position to a predetermined first position when the vehicle is stopped;
adjusting the agent position to a second position closer to the driver's gaze point relative to the first position during the driving;
Information provision support method. 2. The information provision support method according to claim 1,
the vehicle state information includes travel direction information indicating a forward or reverse state of the vehicle;
In the agent position control, when a forward movement of the vehicle is estimated based on the traveling direction information, a forward movement process is executed;
In the forward movement process, the second position is set forward of the driver.
Information provision support method. 3. The information provision support method according to claim 2,
In the forward movement process,
The second position is determined so as to approach the gaze point of the driver as the vehicle speed increases.
Information provision support method. 4. The information provision support method according to claim 2, further comprising:
The agent is
The present invention includes a virtual display object that is displayed within a specific virtual display area, and
The forward movement process includes:
a first forward position adjustment that is performed when the vehicle speed is equal to or less than a predetermined vehicle speed threshold, and a second forward position adjustment that is performed when the vehicle speed exceeds the vehicle speed threshold,
In the first forward position adjustment, the second position is defined within a movable range of the substantial structure;
The second forward position adjustment defines the second position within the virtual viewing area.
Information provision support method. The information provision support method according to any one of claims 2 to 4,
The vehicle state information further includes route visibility information indicating the visibility of the driver on a driving route of the vehicle,
In the forward movement process,
When it is estimated that the visibility of the driver is equal to or lower than a certain level based on the route visibility information, the second position is determined to be closer to the gaze point of the driver.
Information provision support method. The information provision support method according to any one of claims 2 to 5,
In the agent position control, when it is estimated that the vehicle is moving backward based on the traveling direction information, a process for moving backward is executed;
In the reverse travel processing,
estimating a direction in which the driver is facing based on viewing direction information indicating a direction in which the driver is facing;
When it is estimated that the driver is facing forward, a third forward position adjustment is performed to determine the second position forward of the driver in the vehicle interior;
When it is estimated that the driver is facing rearward, a rearward position adjustment is performed to set the second position rearward of the driver.
Information provision support method. The information provision support method according to any one of claims 1 to 6,
the vehicle state information includes turning information suggesting a turning of the vehicle,
When the turning state of the vehicle is estimated based on the turning information, a turning position control is further executed;
In the turning position control,
shifting the agent position in a rotational direction relative to the first position or the second position;
Information provision support method. The information provision support method according to any one of claims 1 to 7,
In the agent position control,
further performing an intimacy correction process of estimating an intimacy level of the driver with respect to the agent and adjusting the first position based on the intimacy level;
In the intimacy correction process,
The first position is determined so that the higher the degree of intimacy, the closer the first position is to the driver.
Information provision support method. The information provision support method according to any one of claims 1 to 8,
and further performing a reaction sensitivity control for adjusting a reaction sensitivity of the agent in accordance with a distance between the second position and the driver.
Information provision support method. An information provision support device that provides information to a driver of a vehicle through the actions of an agent that has a predetermined character motif,
A discrimination unit that discriminates whether the vehicle is moving or stopped based on vehicle state information including a vehicle speed of the vehicle;
an agent position control unit that adjusts the agent position based on the determination result,
The agent position control unit:
adjusting the agent position to a predetermined first position when the vehicle is stopped;
adjusting the agent position to a second position closer to the driver's gaze point relative to the first position during the driving;
Information provision support device. An information provision assistance program that provides information to a driver of a vehicle through an action of an agent having a predetermined character motif,
On the computer,
determining whether the vehicle is moving or stopped based on vehicle state information including the vehicle speed;
Executing an agent position control to adjust the agent position based on the discrimination result;
In the agent position control,
adjusting the agent position to a predetermined first position when the vehicle is stopped;
adjusting the agent position to a second position closer to the driver's gaze point relative to the first position during the driving;
Information provision support program.