JP5478234B2 - Driving scene determination device, driving scene determination method and program, and workload estimation device, workload estimation method and program - Google Patents

Description

  The present invention relates to a driving scene determination apparatus having a function of determining a driving scene, and more particularly to a technique for improving the determination accuracy of a driving scene.

  Conventionally, a system having a function of determining a driving scene such as “waiting for a signal” or “rapid stop” has been proposed (see, for example, Patent Document 1). In a conventional system, a driving scene is determined based on information obtained from a vehicle state detection device (a vehicle speed sensor, a steering angle sensor, an acceleration sensor, or the like). This vehicle state detection device also includes a navigation device, and position information, route information, and the like of the host vehicle can also be used for determining a driving scene.

  In this conventional system, in order to determine the driving scene, the output from the vehicle state detection device is continuously monitored, and the output value and conditions corresponding to the registered driving scene (vehicle speed, vehicle position, etc.) Configured). As a result of the collation, when there is a driving scene that completely matches the condition, the driving scene is determined as the current driving scene. Note that if there is no driving scene in which the condition is completely satisfied as a result of the collation, a predetermined relevance ratio (degree to which the condition is satisfied. For example, the vehicle speed condition is satisfied, but the vehicle position condition is not satisfied, etc. ) Is determined as the current driving scene.

JP 2008-290584 A

  However, in the conventional system, when there is a driving scene that completely matches the conditions, the driving scene is always determined as the current driving scene. (Corresponding conditions) are not necessarily perfect. In other words, what can usually be prepared as a condition corresponding to the driving scene is such that if it almost matches the condition, it will be the driving scene. Therefore, as a result of collation, the condition is completely matched. However, what is determined as the current driving scene is not always correct. As described above, in the conventional system, no consideration has been given to “error determination of driving scene” that occurs when there is a driving scene that completely matches the conditions. Therefore, there is a problem in that a driving scene different from the actual driving scene is determined as the current driving scene (an erroneous determination of the driving scene occurs) only by meeting the conditions completely.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a driving scene determination device that can prevent erroneous determination of a driving scene and improve the determination accuracy of the driving scene. .

  The driving scene determination device of the present invention stores driving scene correspondence information in which driving information of a vehicle is associated with vehicle information relating to a vehicle being driven by a user and road information relating to a road on which the vehicle is passing. A scene storage unit, a scene transition enable / disable storage unit that stores information on whether or not a scene transition from one driving scene to another driving scene is possible, and current vehicle information and current road information Information acquisition means for acquiring the driving scene determination means for determining the current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, The driving scene determination means is configured such that a driving scene associated with the current vehicle information and the current road information is determined from a previous driving scene. Where possible transfer, determines that the a current driving scene.

  Thereby, not only using the driving scene correspondence information but also using the transition availability information, the current driving scene can be correctly determined from the current vehicle information and the current road information. That is, when the driving scene associated with the current vehicle information and the current road information can be transitioned from the previous driving scene, it is determined that the current driving scene. In other words, even if the driving scene is associated with the current vehicle information and the current road information, if it is not possible to make a transition from the previous driving scene, it is determined that the current driving scene is not the current driving scene. Therefore, it is possible to suppress the erroneous determination of the driving scene and to increase the determination accuracy of the driving scene.

  In the driving scene determination apparatus of the present invention, the driving scene storage unit stores a plurality of pieces of driving scene correspondence information prepared for each of the current driving scenes, and the driving scene determination unit includes the plurality of driving scenes. Driving scene correspondence information corresponding to the current driving scene may be selected from the scene correspondence information and used for determination of the current driving scene.

  Accordingly, the current driving scene is determined by using the driving scene corresponding information (the driving scene corresponding information corresponding to the current driving scene) selected from the plurality of driving scene corresponding information prepared for each driving scene. Can be done appropriately.

  In the driving scene determination device of the present invention, the driving scene includes a lane change or a left / right turn, and the vehicle information associated with the lane change or a right / left turn includes a steering angle. The set value of the rudder angle may be larger as the vehicle speed of the vehicle is smaller, and may be smaller as the vehicle speed of the vehicle is larger.

  Thereby, the vehicle information of the “steering angle” associated with the driving scene of “lane change” or “right / left turn” is appropriately set according to the “vehicle speed” of the vehicle. Specifically, the “steering angle” is set to increase as the “vehicle speed” decreases, and the “steering angle” decreases as the “vehicle speed” increases. Therefore, it is possible to appropriately determine the driving scene of “lane change” or “right / left turn” according to the vehicle speed of the vehicle.

  The driving scene determination device of the present invention further includes a scene end determination unit that determines whether or not the current driving scene has ended based on the current vehicle information, and the driving scene determination unit includes the current scene determination unit. The current driving scene may be determined on the assumption that the previous driving scene is continued until it is determined that the driving scene is finished.

  Thus, based on the current vehicle information, it is determined whether or not the current driving scene has ended. Until it is determined that the driving scene has ended, it is assumed that the previous driving scene is continuing and A driving scene is determined. Therefore, as long as the previous driving scene does not end (when the previous driving scene continues), even if the vehicle information or road information corresponds to another driving scene, the driving scene continues. It is determined that In this way, it is possible to suppress an erroneous determination that a transition to another driving scene occurs while the driving scene continues.

  Further, in the driving scene determination device of the present invention, the scene end determination means determines whether or not the lane change driving scene has ended based on the rudder angle and rudder angle change rate included in the vehicle information, The set values of the rudder angle and the rudder angle change rate may be larger as the vehicle speed of the vehicle is smaller, and may be smaller as the vehicle speed of the vehicle is larger.

  Thus, it is possible to appropriately determine whether or not the “lane change” driving scene has ended based on the “steer angle” and the “steer angle change rate”. For example, “steer angle” changes as “positive (left) → negative (right) → positive (left) → zero (straight)” and the absolute value of “steer angle change rate” is “positive (0 ° per second) Greater) → zero (0 ° per second) ”, it is determined that the“ left lane change ”driving scene has ended. In this case, the vehicle information of “steering angle” and “steering angle change rate” is appropriately set according to the “vehicle speed” of the vehicle. Specifically, the “steering angle” is set to increase as the “vehicle speed” decreases, and the “steering angle” decreases as the “vehicle speed” increases. Therefore, it is possible to appropriately determine whether or not to end the driving scene of “lane change” according to the vehicle speed of the vehicle.

  In the driving scene determination device of the present invention, the driving scene determination means may return from the second driving scene to the first driving scene after a scene transition from the first driving scene to the second driving scene. In addition, hysteresis control is performed to make it difficult to return to the first driving scene by changing the vehicle information or the road information associated with the first driving scene.

  Thereby, when returning from the second driving scene to the first driving scene after the scene transition from the first driving scene to the second driving scene, the vehicle information associated with the first driving scene. Or the hysteresis control which changes road information is performed and it becomes difficult to return to a 1st driving scene. Such hysteresis control makes it possible to suppress frequent occurrence of driving scene transitions.

  According to the driving scene determination method of the present invention, driving scene correspondence information in which the driving scene of the vehicle is associated with vehicle information regarding the vehicle the user is driving and road information regarding the road on which the vehicle is passing is stored. Transition information indicating whether or not a scene transition from one driving scene to another driving scene is possible for a plurality of the driving scenes, current vehicle information and current road information are acquired, and the driving scene is acquired. In the method of determining the current driving scene from the current vehicle information and the current road information based on the correspondence information and the transition availability information, the determination of the driving scene includes the current vehicle information and the current road. When the driving scene associated with the information can be changed from the previous driving scene, the current driving scene is determined.

  Even with this method, as described above, not only using the driving scene correspondence information, but also using the transition availability information, the current driving scene can be correctly determined from the current vehicle information and the current road information. It is possible to suppress the erroneous determination of the driving scene and increase the determination accuracy of the driving scene.

  The driving scene determination program of the present invention is a driving in which a driving scene of the vehicle is associated with vehicle information related to a vehicle that the user is driving and road information related to a road on which the vehicle is traveling, which are stored in a memory. Using the scene correspondence information and the transition availability information indicating whether or not a scene transition from one driving scene to another driving scene is possible for the plurality of driving scenes, the current vehicle information and the current road information are transmitted to the computer. In the program for executing the process of acquiring and the process of determining the current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, the driving scene In the determination, the driving scene associated with the current vehicle information and the current road information is determined from the previous driving scene. Where possible transfer, determines that the a current driving scene.

  In this program as well, the current driving scene can be correctly determined from the current vehicle information and the current road information by using not only the driving scene correspondence information but also the transition availability information, as described above. It is possible to suppress the erroneous determination of the driving scene and increase the determination accuracy of the driving scene.

  The workload estimation device of the present invention stores driving scene correspondence information in which vehicle information relating to a vehicle being driven by a user and road information relating to a road on which the vehicle is traveling are associated with a driving scene of the vehicle. A scene storage unit, a scene transition enable / disable storage unit that stores information on whether or not a scene transition from one driving scene to another driving scene is possible, and current vehicle information and current road information Information acquisition means for acquiring a current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, the current vehicle Information and the current driving information when the driving scene associated with the current road information can transition from the previous driving scene. Based on correct driving data in which the correlation between the driving scene determining means for determining that the vehicle is related to the workload related to the driving of the vehicle and the workload value is recorded. A model storage unit that stores a quantification model for calculating the workload value from the contribution behavior of the vehicle, a data acquisition unit that acquires data on the contribution behavior from a user who is driving the vehicle, and the quantification model. And a workload estimation unit that estimates the user workload value from data of the user's contribution behavior while driving the vehicle, wherein the workload estimation unit is used as the user's contribution behavior while driving the vehicle. The current driving scene determined by the driving scene determining means is used.

  This makes it possible to correctly determine the current driving scene from the current vehicle information and the current road information by using not only the driving scene correspondence information but also the transition availability information, and thus erroneous determination of the driving scene. Can be suppressed and the determination accuracy of the driving scene can be increased. Based on the current driving scene determined as described above, the workload value of the user during vehicle driving can be estimated with high accuracy using the quantification model.

  The workload estimation method of the present invention stores driving scene correspondence information in which driving information of a vehicle is associated with vehicle information related to a vehicle that the user is driving and road information related to a road on which the vehicle is passing. In addition, for a plurality of the driving scenes, transition permission / rejection information indicating whether or not a scene transition from one driving scene to another driving scene is stored is stored, and the user's contributing behavior that contributes to a workload related to driving of the vehicle A quantification model for calculating the workload value from the user's contribution behavior while driving the vehicle based on the correct data in which the correlation between the vehicle value and the workload value is recorded. Current road information is acquired, and based on the driving scene correspondence information and the transition availability information, the current vehicle information and the current road When the current driving scene is determined from the information, if the driving scene associated with the current vehicle information and the current road information can be changed from the previous driving scene, the current driving scene is the current driving scene. In the method of acquiring data regarding the contribution behavior from the user who is driving the vehicle, and estimating the workload value of the user from the data of the contribution behavior of the user while driving the vehicle, using the quantification model In the estimation of the workload value, the current driving scene obtained by the determination is used as the user's contributing behavior while driving the vehicle.

  Even with this method, as described above, not only using the driving scene correspondence information, but also using the transition availability information, the current driving scene can be correctly determined from the current vehicle information and the current road information. It is possible to suppress the erroneous determination of the driving scene and increase the determination accuracy of the driving scene. Based on the current driving scene determined as described above, the workload value of the user during vehicle driving can be estimated with high accuracy using the quantification model.

  The workload estimation program of the present invention is a driving in which a driving scene of the vehicle is associated with vehicle information related to a vehicle being driven by a user and road information related to a road on which the vehicle is traveling, which is stored in a memory. Scene correspondence information, transition allowance information indicating whether or not a scene transition from one driving scene to another driving scene is possible for the plurality of driving scenes, and user contribution actions that contribute to the workload related to driving the vehicle Based on the correct data in which the correlation with the workload value is recorded, a quantification model for calculating the workload value from the user's contribution behavior while driving the vehicle, and using the computer, the current vehicle Information and current road information, and the current vehicle information based on the driving scene correspondence information and the transition availability information. When the current driving scene is determined from the current road information, the current vehicle information and the driving scene associated with the current road information can be changed from the previous driving scene, A process for determining that the current driving scene is determined, a process for acquiring data related to the contribution action from the user who is driving the vehicle, and the user's contribution action data while driving the vehicle using the quantification model. In the program for executing the process of estimating the workload value, the current driving scene obtained by the determining process is used as the contribution behavior of the user during driving of the vehicle in the estimation of the workload value. .

  In this program as well, the current driving scene can be correctly determined from the current vehicle information and the current road information by using not only the driving scene correspondence information but also the transition availability information, as described above. It is possible to suppress the erroneous determination of the driving scene and increase the determination accuracy of the driving scene. Based on the current driving scene determined as described above, the workload value of the user during vehicle driving can be estimated with high accuracy using the quantification model.

  The service providing apparatus according to the present invention stores driving scene correspondence information in which driving information of a vehicle is associated with vehicle information relating to a vehicle being driven by a user and road information relating to a road on which the vehicle is traveling. A storage unit, a scene transition availability storage unit for storing transition availability information indicating whether or not a scene transition from one driving scene to another driving scene is possible, and current vehicle information and current road information. Information acquisition means for acquiring; means for determining a current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, wherein the current vehicle information And when the driving scene associated with the current road information can transition from the previous driving scene, And determining the operation scene determining means and that, and a, and a service providing means for providing a service corresponding to the determined said current driving scene.

  This makes it possible to correctly determine the current driving scene from the current vehicle information and the current road information by using not only the driving scene correspondence information but also the transition availability information, and thus erroneous determination of the driving scene. Can be suppressed and the determination accuracy of the driving scene can be increased. And a service can be provided appropriately based on the current driving scene determined in this way.

  The service providing method of the present invention stores driving scene correspondence information in which driving information of a vehicle is associated with vehicle information related to a vehicle being driven by a user and road information related to a road on which the vehicle is passing. The transition availability information indicating whether or not a scene transition from one driving scene to another driving scene is stored for a plurality of the driving scenes, current vehicle information and current road information are acquired, and the driving scene is supported. A driving scene associated with the current vehicle information and the current road information when determining the current driving scene from the current vehicle information and the current road information based on the information and the transition availability information Is determined to be the current driving scene when it is possible to transition from the previous driving scene, and according to the determined current driving scene To provide a service.

  Even with this method, as described above, not only using the driving scene correspondence information, but also using the transition availability information, the current driving scene can be correctly determined from the current vehicle information and the current road information. It is possible to suppress the erroneous determination of the driving scene and increase the determination accuracy of the driving scene. And a service can be provided appropriately based on the current driving scene determined in this way.

  In the service providing program of the present invention, a driving scene in which a driving scene of the vehicle is associated with vehicle information relating to a vehicle that the user is driving and road information relating to a road on which the vehicle passes is stored in a memory. The current vehicle information and the current road information are acquired by the computer using the correspondence information and the transition availability information indicating whether or not a scene transition from one driving scene to another driving scene is possible for the plurality of driving scenes. And when determining the current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, the current vehicle information and the current road When the driving scene associated with the information can be transitioned from the previous driving scene, the current driving scene is determined. A sense, the process of providing a service corresponding to the determined said current driving scene, is the execution.

  In this program as well, the current driving scene can be correctly determined from the current vehicle information and the current road information by using not only the driving scene correspondence information but also the transition availability information, as described above. It is possible to suppress the erroneous determination of the driving scene and increase the determination accuracy of the driving scene. And a service can be provided appropriately based on the current driving scene determined in this way.

  According to the present invention, not only using the driving scene correspondence information but also using the transition availability information, the current driving scene can be correctly determined from the current vehicle information and the current road information. It is possible to suppress the occurrence of erroneous determination and improve the determination accuracy of the driving scene.

The block diagram which shows the structure of the driving scene determination apparatus in embodiment of this invention. Explanatory drawing of the driving scene correspondence information (during driving) in the embodiment of the present invention Explanatory drawing of the driving scene correspondence information (at the time of a stop) in embodiment of this invention Explanatory drawing of the possibility of scene transition in the embodiment of the present invention (A) The figure which shows the relationship between the vehicle speed and the magnitude | size of a rudder angle in the driving scene of a right-left turn (b) The figure which shows the relationship between the vehicle speed and the magnitude | size of a rudder angle in the driving scene of a lane change (c) The driving scene of a lane change Of the relationship between the vehicle speed and the rate of change in rudder angle The flowchart which shows the flow of operation | movement of the driving scene determination apparatus in embodiment of this invention. The block diagram which shows the structure of the workload estimation apparatus in embodiment of this invention. The figure which shows an example (example using a formulation model) of the workload estimation in embodiment of this invention The figure which shows an example (example using a probabilistic reasoning model) of the workload estimation in embodiment of this invention The flowchart which shows the flow of operation | movement of the workload estimation apparatus in embodiment of this invention. The block diagram which shows the structure of the service provision apparatus in embodiment of this invention.

  Hereinafter, a driving scene determination device and a workload estimation device according to an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, a case of a driving scene determination device or a workload estimation device used for a car navigation device or the like is illustrated.

  In the following, an example of the driving scene determination device will be described as the first embodiment, and an example of the workload estimation device will be described as the second embodiment. As will be described below, the device of the first embodiment has a driving scene determination function, and the device of the second embodiment has a workload estimation function in addition to the driving scene determination function. However, these functions are realized by a program stored in an HDD or a memory (not shown) of the apparatus.

(First embodiment)
The configuration of the driving scene determination apparatus according to the first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the driving scene determination apparatus according to the present embodiment. As shown in FIG. 1, the driving scene determination device 1 includes an information acquisition unit 2 having a function of acquiring various types of information such as vehicle information and road information, and a control unit 3 that performs various processes related to determination of the driving scene. I have. The information acquisition unit 2 includes a vehicle information acquisition unit 4 for acquiring information (vehicle information) related to the vehicle that the user is driving, and a road for acquiring information (road information) related to the road on which the vehicle is traveling. The information acquisition unit 5 is configured. Moreover, the control part 3 is comprised by CPU, a microcomputer, etc.

The vehicle information acquisition unit 4 has a function of acquiring vehicle information such as the current position of the vehicle from the GPS device 6 mounted on the vehicle. In addition, the vehicle information acquisition unit 4 includes a vehicle speed (km / hour), an acceleration (m / second ² ), a steering angle (°) (left turn: positive, right turn: negative), left and right winkers (left and right turn SW). Vehicle information such as on / off of the vehicle, hazard lamp on / off, brake pressure, side brake on / off, etc. is obtained from the vehicle sensor 7 mounted on the vehicle. The vehicle information acquisition unit 4 also includes vehicle information such as the distance (m) between the vehicle ahead and the presence of a stop line in front of the vehicle, the presence or absence of a crosswalk in front of the vehicle, and the presence or absence of a signal in front of the vehicle. Has a function of acquiring from the in-vehicle camera 8 mounted on the camera. In this Embodiment, the vehicle information acquisition part 4 acquires these vehicle information via vehicle-mounted LAN (For example, CAN, AVC-LAN, BEAN etc.) which is not shown in figure.

  The road information acquisition unit 5 has a function of acquiring road information from a road information DB 9 (database) configured with a large capacity HDD or the like. This road information includes information such as the position of the intersection (latitude and longitude, etc.), the number of lanes (lines), the road type (highway expressway, national road, prefectural road, etc.), and the turning radius (m) of the road. The road information includes information such as the presence / absence of a signal at the intersection, the presence / absence of a stop line at the intersection, and the presence / absence of a pedestrian crossing at the intersection. Further, the road information includes a distance (m) to the intersection (calculated from the position of the intersection and the current position of the vehicle).

  The control unit 3 includes a driving scene determination unit 10 that determines the current driving scene based on the current vehicle information and road information acquired by the information acquisition unit 2. The driving scene determination unit 10 has a function of acquiring driving scene correspondence information from a driving scene DB 11 configured with an HDD, a memory, and the like. This driving scene correspondence information is information in which a driving scene is associated with vehicle information and road information, and is preset and stored in the driving scene DB 11 in the present embodiment.

  Here, the driving scene correspondence information will be described more specifically with reference to FIGS. 2 and 3. FIG. 2 is an explanatory diagram illustrating an example of driving scene correspondence information when the vehicle is traveling, and FIG. 3 is an explanatory diagram illustrating an example of driving scene correspondence information when the vehicle is stopped. For example, in the example of FIG. 2, the driving scene “turn left” is associated with vehicle information “vehicle speed” “steering angle” “left and right turn SW” and road information “distance to intersection”. In this case, it can be said that the driving scene of “left turn” is associated with the conditions “vehicle speed V> 0”, “steering angle θ ≧ 50”, “left turn SW: ON”, and “distance to intersection D = 0”. . In this case, the driving scene “turn left” is also associated with the conditions “vehicle speed V> 0”, “steering angle θ ≧ 50”, “left turn SW: ON”, and “number of lanes N = 1”.

  Such driving scene correspondence information may be prepared for each current driving scene. For example, different driving scene correspondence information may be used depending on whether the driving scene is “stable driving” or “high speed driving”. More specifically, in the driving scene correspondence information in “stable traveling”, the condition associated with “high speed traveling” is “vehicle speed V ≧ 60”, and the condition associated with “stable traveling” is “20 ≦ vehicle speed V On the other hand, in the driving scene correspondence information in “high speed running”, the condition associated with “high speed running” is “vehicle speed V ≧ 50” and the condition associated with “stable running” is “20”. ≦ Vehicle speed V <50 ”may be sufficient.

  The driving scene determination unit 10 selects driving scene correspondence information corresponding to the current driving scene from a plurality of driving scene correspondence information prepared for each current driving scene, and determines the current driving scene. Use. That is, in the above example, in the driving scene correspondence information at the time of “stable driving”, the vehicle speed condition for transitioning to the driving scene of “high speed driving” is set to “60 km / h or more”, whereas “ In the driving scene correspondence information at the time of “high speed driving”, the vehicle speed condition for transitioning to the driving scene of “stable driving” is set to “50 km / h or less”. The difference between the driving scene correspondence information at the time of “stable driving” and the driving scene correspondence information at the time of “high speed driving” is different between driving scenes that can be changed to each other (eg, “stable driving” and “high speed driving”). It is a set value (for example, a set value of “vehicle speed”) of vehicle information and road information associated with each other.

  Further, the driving scene determination unit 10 has a function of acquiring transition permission / inhibition information from the transition permission / inhibition DB 12 configured with an HDD, a memory, or the like. This transition availability information is information indicating whether or not a scene transition from a certain driving scene to another driving scene is possible for a plurality of driving scenes. In the present embodiment, the transition availability information is preset and stored in the transition availability DB 12. .

  Here, the transition availability information will be described more specifically with reference to FIG. FIG. 4 is an explanatory diagram illustrating an example of the transition availability information. In FIG. 4, driving scenes that can be changed are indicated by “arrows”. For example, it is indicated that a transition can be made from a driving scene of “start” to a driving scene of “acceleration” and “low-speed driving” but cannot be shifted to a driving scene of “deceleration”. FIG. 4 is merely an example. That is, in the example of FIG. 4, only scene transitions between running scenes are illustrated, but it goes without saying that transition availability information regarding scene transitions between running scenes and stop scenes may be used.

  Then, the driving scene determination unit 10 determines the current driving scene from the current vehicle information and road information with reference to the driving scene correspondence information and the transition availability information. Specifically, when the driving scene associated with the current vehicle information and road information can transition from the previous driving scene, the driving scene determination unit 10 determines that the driving scene is the current driving scene. Judge that there is.

  For example, after turning right or left in an intersection, when the steering wheel is returned (when the rudder angle is reduced), if the winker returns together and the left / right turn SW is turned off, “vehicle speed V> 0 ”,“ steering angle | θ | <50 ”, and“ distance to intersection D = 0 ”. If the transition availability information is not referred to,“ passing straight through the intersection (= The driving scene to be selected when passing through is determined as the current driving scene. However, as shown in FIG. 4, the scene transition from “right / left turn” to “passing straight at an intersection” is impossible, and therefore it is not determined that the current driving scene is “passing straight at an intersection”. In such a case, the “right / left turn” that was previously determined as the “current driving scene” is determined to be the current driving scene.

  Here, if “distance to intersection D = 0”, it is assumed that the vehicle is within the intersection, but it is also possible to determine the driving scene in consideration of the moving distance within the intersection. For example, it becomes possible to discriminate between the driving scene “near the head of signal waiting” and the driving scene “stop in the intersection” according to the moving distance (m) after entering the intersection.

  Similarly, after changing the lane and returning the steering wheel to reduce the rudder angle, the conditions of “vehicle speed V> 0”, “steering angle | θ | ≦ 10”, and “number of lanes N ≧ 2” are satisfied. If the transition permission / prohibition information is not referred to, “currently to the left / immediately right (= originally a driving scene to be selected before changing lane / before turning right / left)” is determined as the current driving scene. However, as shown in Fig. 4, it is not possible to transition from "change lane" to "soon to the left" or "soon to the right", so the current driving scene is "soon to the left / soon to the right". It is not determined that there is. In such a case, the “lane change” previously determined as the “current driving scene” is determined to be the current driving scene.

  As shown in FIGS. 5A and 5B, the reference value of “steering angle” associated with “lane change” or “right / left turn” increases as the vehicle speed decreases, and the vehicle speed increases. You may set so that it may become so small. 5A and 5B illustrate the case where the relationship between the | steering angle | and the vehicle speed is a linear function, this is merely an example, and the relationship between the | steering angle | and the vehicle speed. Needless to say, may be other functions (n-order function, trigonometric function, etc.). In the example of FIG. 2, the reference value of “steering angle” associated with “right / left turn” is a constant value (θ ≧ 50) regardless of the vehicle speed, but for example, when the vehicle speed is low (V <20). The steering angle reference value may be increased (θ ≧ 60), and the steering angle reference value may be decreased (θ ≧ 40) when the vehicle speed is high (V ≧ 50).

  The control unit 3 includes a scene end determination unit 13 that determines whether the current driving scene is ended based on the current vehicle information. The driving scene determination unit 10 determines whether the current driving scene is Until it is determined that the current driving scene has been completed, the current driving scene is determined as the previous driving scene continues. Specifically, the scene end determination means determines whether or not the lane change driving scene has ended based on the “steer angle” and the “steer angle change rate”. For example, “steer angle” changes as “positive (left) → negative (right) → positive (left) → zero (straight)” and the absolute value of “steer angle change rate” is “positive (0 ° per second) Greater) → zero (0 ° per second) ”, it is determined that the“ left lane change ”driving scene has ended. A yaw rate (yaw angle change rate: radians / second) may be used instead of the rudder angle change rate. The yaw rate information can be obtained from a yaw rate sensor (not shown) mounted on the vehicle.

  In this case, as shown in FIGS. 5B and 5C, the reference values of “steer angle” and “steer angle change rate” are larger as the vehicle speed is lower and smaller as the vehicle speed is higher. It may be set as follows. For example, when the vehicle speed is low (V <20), the rudder angle reference value is increased (θ ≧ 60), and when the vehicle speed is high (V ≧ 50), the rudder angle reference value is decreased (θ ≧ 40). May be. Similarly, for example, when the vehicle speed is low (V <20), the reference value of the steering angle change rate is increased (for example, 10 ° per second), and when the vehicle speed is high (V ≧ 50), the reference value of the steering angle change rate is set. The value may be decreased (for example, 0 ° per second). FIG. 5C illustrates a case where the relationship between the | steering angle change rate | and the vehicle speed is a linear function, but this is only an example, and the | steering angle change rate | It goes without saying that the relationship may be another function (n-order function, trigonometric function, etc.).

  The driving scene determination unit 10 associates with the first driving scene when returning from the second driving scene to the first driving scene after the scene transition from the first driving scene to the second driving scene. The vehicle information or road information (conditions for transitioning to the first driving scene) is changed, and hysteresis control is performed to make it difficult to return to the first driving scene.

For example, when transitioning from a driving scene of “low speed driving” (first driving scene) to a driving scene of “acceleration” (second driving scene), the condition for transitioning to the driving scene of “acceleration” is “acceleration ≧ 3 m Per second ² ”. Thereafter, when returning from the driving scene of “acceleration” to the driving scene of “low speed running”, the condition for transition to the driving scene of “low speed running” is changed to “acceleration <2 m / sec ² ” by hysteresis control. Further, after that, when transitioning from the driving scene of “low speed running” to the driving scene of “acceleration”, the condition for transitioning to the driving scene of “acceleration” is changed to “acceleration ≧ 4 m / sec ² ” by hysteresis control. .

  Here, the case of one-dimensional hysteresis control is exemplified (when the condition (parameter) to be changed is one), but it goes without saying that multidimensional hysteresis control may be performed. In addition, in this hysteresis control, it is possible to make it difficult to return to the first driving scene by adding additional conditions in addition to changing the parameters (conditions for transitioning to the first driving scene). is there.

  The operation of the driving scene determination apparatus 1 of the present embodiment configured as described above will be described with reference to the flowchart of FIG.

  In the driving scene determination apparatus 1 according to the present embodiment, the driving scene is determined periodically (for example, once every second). When the driving scene is determined, as shown in FIG. 6, first, the current vehicle information and road information are acquired by the information acquisition unit 2 (S1, S2). Then, as preprocessing for current driving scene determination, it is determined whether or not the driving scene (lane change or the like) obtained in the previous driving scene determination has ended (S3), and the driving scene has ended The current driving scene determination is started (S4).

  In the current driving scene determination, not only the driving scene correspondence information is used but also the transition availability information is used to determine the current driving scene from the current vehicle information and the current road information. That is, when the driving scene associated with the current vehicle information and the current road information can transition from the previous driving scene (S5), it is determined that the driving scene is the current driving scene (S6). ).

  According to the driving scene determination apparatus 1 of the embodiment of the present invention as described above, not only the driving scene correspondence information is used but also the transition availability information is used, so that the current driving information is obtained from the current vehicle information and the current road information. The scene can be determined correctly. That is, when the driving scene associated with the current vehicle information and the current road information can be transitioned from the previous driving scene, it is determined that the current driving scene. In other words, even if the driving scene is associated with the current vehicle information and the current road information, if it is not possible to make a transition from the previous driving scene, it is determined that the current driving scene is not the current driving scene. Therefore, it is possible to suppress the erroneous determination of the driving scene and to increase the determination accuracy of the driving scene.

  In the present embodiment, by using driving scene correspondence information (driving scene correspondence information corresponding to the current driving scene) selected from a plurality of driving scene correspondence information prepared for each driving scene, The driving scene can be properly determined.

  In this case, it can be said that the vehicle information and road information associated with the driving scene are different in the plurality of driving scene correspondence information prepared for each driving scene. The difference between the driving scene correspondence information of a certain driving scene and the driving scene correspondence information of another driving scene is a set value of vehicle information and road information that are associated with driving scenes that can be changed to each other.

  For example, “stable driving” and “high-speed driving” are driving scenes that can transition to each other. However, in the driving scene corresponding information of the driving scene of “stable driving”, the vehicle speed for changing to the driving scene of “high-speed driving” Is set to “60 km / h or more”, but in the driving scene corresponding information of the driving scene of “high speed driving”, the vehicle speed condition for transitioning to the driving scene of “stable driving” is “50 km / h Is set to “below”. As a result, when the vehicle is traveling at a vehicle speed in the vicinity of 60 km / h (sometimes when traveling at 61 km / h and sometimes at 59 km / h), when the vehicle speed exceeds 60 km / h (when it reaches 61 km / h) ) After the driving scene changes from “Stable driving” to “High speed driving”, the driving scene changes from “High speed driving” to “Stable driving” when the driving speed is slightly below 60 km / h (about 59 km / h). do not do. In this way, it is possible to suppress frequent occurrence of driving scene transitions.

  Further, in the present embodiment, the vehicle information of “steering angle” associated with the driving scene of “lane change” or “right / left turn” is appropriately set according to the “vehicle speed” of the vehicle. Specifically, the “steering angle” is set to increase as the “vehicle speed” decreases, and the “steering angle” decreases as the “vehicle speed” increases. Therefore, it is possible to appropriately determine the driving scene of “lane change” or “right / left turn” according to the vehicle speed of the vehicle.

  In the present embodiment, it is determined whether or not the current driving scene has ended based on the current vehicle information, and the previous driving scene continues until it is determined that the driving scene has ended. As a result, the current driving scene is determined. Therefore, as long as the previous driving scene does not end (when the previous driving scene continues), even if the vehicle information or road information corresponds to another driving scene, the driving scene continues. It is determined that In this way, it is possible to suppress an erroneous determination that a transition to another driving scene occurs while the driving scene continues.

  In this case, based on the “steer angle” and the “steer angle change rate”, it can be appropriately determined whether or not the “lane change” driving scene has ended. For example, “steer angle” changes as “positive (left) → negative (right) → positive (left) → zero (straight)” and the absolute value of “steer angle change rate” is “positive (0 ° per second) Greater) → zero (0 ° per second) ”, it is determined that the“ left lane change ”driving scene has ended. In this case, the vehicle information of “steering angle” and “steering angle change rate” is appropriately set according to the “vehicle speed” of the vehicle. Specifically, the “steering angle” is set to increase as the “vehicle speed” decreases, and the “steering angle” decreases as the “vehicle speed” increases. Therefore, it is possible to appropriately determine whether or not to end the driving scene of “lane change” according to the vehicle speed of the vehicle.

  Further, in the present embodiment, after the scene transition from the first driving scene (for example, the driving scene of “low speed driving”) to the second driving scene (for example, the driving scene of “acceleration”), the second driving scene is started. When returning to the first driving scene, hysteresis control is performed to change vehicle information or road information (for example, “acceleration” information) associated with the first driving scene. It becomes difficult to return. Such hysteresis control makes it possible to suppress frequent occurrence of driving scene transitions.

(Second Embodiment)
Next, the workload estimation apparatus 20 according to the second embodiment of this invention will be described. The workload estimation device 20 according to the present embodiment is a device in which a workload estimation function is added to the driving scene determination device 1 according to the first embodiment. Therefore, here, the description will focus on the differences of the workload estimation device 20 of the second embodiment from the driving scene determination device 1 of the first embodiment. Unless otherwise specified, the configuration and operation of the present embodiment are the same as those of the first embodiment.

  FIG. 7 is a block diagram illustrating a configuration of the workload estimation apparatus 20 according to the present embodiment. As shown in FIG. 7, the workload estimation device 20 includes a quantification model for calculating a workload value related to driving of the vehicle, in addition to the configuration of the driving scene determination device 1 of the first embodiment. A stored model DB 21 (database) is provided. The quantification model is a model for calculating the workload value from the user's contribution behavior while driving the vehicle using correct data in which the correlation between the contribution behavior of the user contributing to the workload and the workload value is recorded. It is. The detailed contents of the quantification model and the correct answer data will be described later with reference to the drawings.

  The information acquisition unit 2 of the workload estimation device 20 also has a function of acquiring data related to the user's contributing behavior while driving the vehicle from the vehicle sensor 7, the in-vehicle camera 8 (including the line-of-sight camera) and the biological sensor 22. . Therefore, this information acquisition part 2 is also corresponded to the data acquisition part of this invention. And the workload estimation apparatus 20 is provided with the workload estimation part 23 which estimates a user's workload value from the data of the user's contribution action while driving a vehicle using a quantification model.

(Workload type)
Here, the types of workload of the user who is driving the vehicle and the contribution data of the workload will be described. The user's workload while driving the vehicle includes a driving workload caused by the user's driving operation when driving the vehicle. For example, the information acquisition unit 2 uses data relating to driving workload contribution behavior (for example, steering angle data, vehicle speed data, inter-vehicle distance data, accelerator opening rate data, brake signal data, etc. during vehicle operation) as vehicle sensor 7. Get from.

  Moreover, the user's workload during driving of the vehicle includes a device operation workload caused by the user's device operation when operating the device provided in the vehicle. The information acquisition unit 2 acquires data (for example, navigation operation data, audio operation data, air conditioner operation data, window opening / closing operation data, etc. during driving of the vehicle) from the vehicle sensor 7 regarding the contribution behavior of the device operation workload.

  In addition, the user's workload while driving the vehicle includes an auditory workload caused by the user's auditory motion when driving the vehicle. The information acquisition unit 2 acquires, from the vehicle sensor 7, data related to the auditory workload contribution behavior (for example, voice information data or music information data included in route guidance or audio reproduction while driving the vehicle).

  In addition, the user's workload during driving of the vehicle includes a visual workload resulting from the user's visual motion when driving the vehicle. The information acquisition unit 2 acquires from the line-of-sight camera 4 data related to visual workload contribution behavior (for example, the viewpoint movement speed data and viewpoint coordinate data of the user who is driving the vehicle).

  Further, the workload of the user who is driving the vehicle includes an intrinsic workload caused by the psychological state of the user when driving the vehicle. The information acquisition unit 2 acquires data (for example, heart rate data, blood pressure data, and respiratory rate data of a user who is driving the vehicle) from the biological sensor 22 regarding the contribution behavior of the intrinsic workload.

  Further, the workload of the user who is driving the vehicle includes an exogenous workload caused by the driving scene of the vehicle. The information acquisition unit 2 acquires data (for example, the current driving scene) regarding the contribution action of the extrinsic workload from the driving scene determination unit 10.

  The workload estimation unit 23 uses the plurality of workloads (driving workload, equipment operation workload, auditory workload, visual workload, intrinsic workload, extrinsic workload) as elements. Estimate the workload value of the user during operation. Specifically, the workload estimation unit 23 uses a quantification model stored in the model DB 21 to calculate each of a plurality of workloads from the data of the user's contribution behavior during vehicle driving acquired by the information acquisition unit 2. The workload value of the user who is driving the vehicle is estimated using them as elements.

(Estimation of workload using quantification model)
Here, estimation of a workload, which is a feature of the present embodiment, will be described with reference to the drawings. FIG. 8 is an explanatory diagram showing an example of workload estimation in the present embodiment. Here, as shown in FIG. 8, a case will be described in which a workload value is estimated for each of a plurality of workloads using a formulation model as one of the quantification models. For this formulation, for example, statistical methods such as linear multiple regression analysis, principal component analysis, and factor analysis are used.

As shown in FIG. 8, the data contributing to the driving workload (driving WL contribution data) includes the steering angle d1 (degrees), the vehicle speed d2 (km / h), the inter-vehicle distance d3 (m), and the accelerator Data of the opening rate d4 (%) and the brake signal d5 (ON: 1, OFF: 0) are included. By formulating the relationship between these data d1 to d5 and correct answer data (described later) by a statistical method or the like, the driving workload (DWL) is formulated as shown in the following formula 1.
DWL = α (n) × d1 / 360 + β (n) × d2 / 100 + γ (n) / d3
+ Δ (n) × d4 / 100 + ε × d5 (Formula 1)

  Here, α, β, γ, δ, and ε are weighting coefficients, and n is a familiarity coefficient that takes into account the user's familiarity with respect to vehicle driving. This familiarity coefficient is a coefficient determined based on the user's travel history. The weighting coefficient α will be described as an example. The weighting coefficient when the user travels for the first time (n = 1) is α (1), and the weighting when the user travels the vehicle for the second time (n = 2). The coefficient is α (2). In other words, it can be said that this formulation model is weighted in consideration of the user's familiarity with vehicle driving.

Further, data (device operation WL contribution data) contributing to the device operation workload includes navigation operation o1 (hardware button operation: A1, touch panel operation: A2, etc.) and audio operation o2 (skip button operation: A3, volume operation: A4, etc.), air conditioner operation o3 (with operation: 1, no operation: 0), and window opening / closing operation o4 (with operation: 1, no operation: 0) are included. By deriving the relationship between the data o1 to o4 and the correct data by a statistical method or the like, the equipment operation workload (OWL) is formulated as shown in the following Expression 2.
OWL = α (n) × o1 + β (n) × o2
+ Γ (n) × o3 + δ (n) × o4 (Formula 2)
A1 to A4 are respectively predetermined constants. In addition, α, β, γ, and δ are weighting coefficients, and n is a familiarity coefficient that takes into account the user's familiarity with respect to vehicle driving.

In addition, the audio information a1 (route guidance: B1, facility information guidance: B2, telephone: data included in route guidance and audio reproduction while driving the vehicle is included in the data contributing to the auditory workload (auditory WL contribution data). B3) and music information a2 (classic: B4, rock: B5, favorite song: B6, etc.). An auditory workload (AWL) is formulated as in the following Equation 3 by deriving the relationship between the data a1 and a2 and the correct data by a statistical method or the like.
AWL = α (n) × a1 + β (n) × a2 (Formula 3)
B1 to B6 are predetermined constants. In addition, α and β are weighting coefficients, and n is a familiarity coefficient in consideration of the user's familiarity with respect to vehicle driving.

The data contributing to the visual workload (visual WL contribution data) includes the viewpoint movement speed v1 (mm / second) during driving and the viewpoint coordinates v2 (room mirror: C1, side mirror: C2, navigation: Data such as C3). The visual workload (VWL) is formulated as shown in Equation 4 below by deriving the relationship between the data v1 and v2 and the correct answer data using a statistical method or the like.
VWL = α (n) × v1 + β (n) × v2 (Formula 4)
C1 to C6 are predetermined constants. In addition, α and β are weighting coefficients, and n is a familiarity coefficient in consideration of the user's familiarity with respect to vehicle driving.

The data contributing to the intrinsic workload (endogenous WL contribution data) includes data on the heart rate i1 (times / min), blood pressure i2 (mmHg), and respiratory rate i3 (times / min) while driving the vehicle. included. By deriving the relationship between the data i1 to i3 and the correct answer data by a statistical method or the like, the intrinsic workload (IWL) is formulated as in the following Expression 5.
IWL = α (n) × i1 + β (n) × i2 + γ (n) × i3 (Formula 5)
Α, β, and γ are weighting factors, and n is a familiarity factor that takes into account the user's familiarity with driving the vehicle.

Further, the data contributing to the exogenous workload (exogenous WL contribution data) includes data of the current driving scene. By deriving the relationship between the data s1 and the correct answer data by a statistical method or the like, the exogenous workload (SWL) is formulated as shown in Equation 6 below.
SWL = α (n) × f (s1) (Formula 6)
Note that f is a function for digitizing the driving scene (a predetermined output value may be set for each driving scene that is an input value), α is a weighting coefficient, and n is the user's vehicle. This is a habituation coefficient that takes into account driving habituation.

  In the foregoing, an example of workload estimation using a formulation model has been described, but the scope of the present invention is not limited to this. For example, as shown in FIG. 9, a probabilistic inference model may be used for estimating the workload.

  As the probability inference model, for example, a Bayesian network or a neural network is used. The driving scene data may be used when estimating each workload value. For example, the driving workload will be described in detail, and the steering angle d1 (degrees), the vehicle speed d2 (km / h), and the inter-vehicle distance d3 (m) are data that contribute to the driving workload (driving WL contribution data). ), Accelerator opening rate d4 (%), brake signal d5 (ON: 1, OFF: 0) and driving scene d6 data are input, and a Bayesian network or neural network probability model is constructed, and driving is performed as the inference result. The workload value of the workload (DWL) is estimated.

  Similarly, with respect to the device operation workload, the above-described device operation WL contribution data (o1 to o4) and the operation scene data o5 are input, and a Bayesian network or neural network probability model is constructed, and the inference result is the device operation. The workload value of the workload (OWL) is estimated.

  As for the auditory workload, the above-mentioned auditory WL contribution data (a1, a2) and driving scene data a3 are input, and a probability model of a Bayesian network or a neural network is constructed, and the inference result is an auditory work. The workload value of the load (AWL) is estimated.

  As for visual workload, the above-mentioned visual WL contribution data (v1, v2) and driving scene data v3 are input, and a probability model of a Bayesian network or a neural network is constructed. The workload value of the load (VWL) is estimated.

  As for the intrinsic workload, the above-described intrinsic WL contribution data (i1 to i3) and driving scene data i4 are input, and a probability model of a Bayesian network or a neural network is constructed, and the inference work results as the intrinsic work. The workload value of the load (IWL) is estimated.

  In constructing a quantification model (formulation model or probabilistic reasoning model), correct data may be used and a user's profile (age, gender, driving calendar), etc. may be taken into consideration.

(Correct data)
Next, correct data used to construct a quantification model will be described. In the present embodiment, a quantification model is constructed using two types of correct answer data (biological signal index and subjective evaluation index). Note that the construction of a quantification model can also be referred to as modeling (quantification) of each workload.

  Here, a pupil reaction is described as an example of a biological signal index, but the scope of the present invention is not limited to this. The biological signal index of the present invention includes heart rate fluctuation, respiratory rate fluctuation, bioelectric potential fluctuation and the like in addition to pupil reaction.

  In general, when the pupil is mydriatic (the pupil radius increases), the sympathetic nerve is stimulated and the user is excited. Conversely, when the pupil is miotic (the pupil radius is small), the parasympathetic nerve is stimulated and the user is relaxed. The larger the pupil response (pupil radius change rate), the higher the workload applied to the user, and the lower the pupil response, the lower the workload applied to the user.

  In the present embodiment, time series data of the user's pupil reaction during driving of the vehicle when the user runs on a predetermined experimental course is acquired, and the correspondence with driving of the vehicle is analyzed. For example, the pupil response is small at the time of departure, and the pupil response is large when performing a left turn, right turn, passing or overtaking, and the like. From this correct data of pupil response, the part that contributes to each workload is extracted, and each workload is formulated based on the magnitude and rate of change of the pupil response of that part. .

  Here, the formulation of the operation workload will be described as an example. For example, from the correct data of the pupil reaction, the part that contributes to the driving workload (for example, the part of the right turn) is extracted, and the driving WL contribution data (steering angle signal, vehicle speed, inter-vehicle distance, accelerator opening) at that time is extracted. The driving workload is formulated by using statistical methods to derive the relationship with the frequency and brake signal data). That is, based on the correct answer data of the pupil reaction, a formulation model (the above formula 1) for calculating the workload value of the driving workload from the driving WL contribution data is obtained.

  In the above, the example which formulates driving | running | working workload from the correct data of the pupil reaction was demonstrated. Although explanation is omitted here, in the same way, the equipment operation workload, the auditory workload, the visual workload, and the intrinsic workload are formulated.

  Next, correct answer data of the subjective evaluation index will be described. Here, the conventional NASA-TLX will be described as an example of the subjective evaluation index, but the scope of the present invention is not limited to this. For example, instead of the conventional NASA-TLX, a continuous NASA-TLX (time series NASA-TLX) in which subjective evaluation is performed on a PC while watching a video may be used. The subjective evaluation index of the present invention includes SWAT, MCH and the like in addition to NASA-TLX.

  NASA-TLX is an index by which a user subjectively quantifies a workload. Generally, the larger the value, the higher the user's workload, and the smaller the value, the lower the user's workload.

  In the present embodiment, time series data of the NASA-TLX of the user who is driving the vehicle when the user runs on a predetermined experimental course is acquired, and the correspondence with the driving of the vehicle is analyzed. For example, the value of NASA-TLX is large at the time of departure, left turn, or right turn, and the value of NASA-TLX is small when going straight. From such NASA-TLX correct answer data, a portion that contributes to each workload is extracted, and based on the size, rate of change, etc. of the NASA-TLX of that portion, a formula for each workload is extracted. Is done.

  Here, the formulation of the operation workload will be described as an example. For example, from the correct answer data of NASA-TLX, a portion (for example, a right turn portion) performing a driving workload contribution action is extracted, and driving WL contribution data (steering angle signal, vehicle speed, inter-vehicle distance, accelerator) at that time is extracted. The operational workload is formulated by using statistical methods to derive the relationship between the opening rate and brake signal data). That is, based on the NASA-TLX correct answer data, a formulation model (the above formula 1) for calculating the workload value of the driving workload from the driving WL contribution data is obtained.

  In the above, the example which formulates a driving | working workload from the correct data of NASA-TLX was demonstrated. Although explanation is omitted here, in the same way, the equipment operation workload, the auditory workload, the visual workload, and the intrinsic workload are formulated.

  About the workload estimation apparatus 20 comprised as mentioned above, the operation | movement is demonstrated using the flowchart of FIG.

  In the workload estimation device 20 of the present embodiment, the current driving scene is determined, and the workload is estimated using the current driving scene. That is, as in the first embodiment, first, the current vehicle information and road information are acquired by the information acquisition unit 2 (S1, S2). Then, as preprocessing for current driving scene determination, it is determined whether or not the driving scene (lane change or the like) obtained in the previous driving scene determination has ended (S3), and the driving scene has ended The current driving scene determination is started (S4).

  In the current driving scene determination, not only the driving scene correspondence information is used but also the transition availability information is used to determine the current driving scene from the current vehicle information and the current road information. That is, when the driving scene associated with the current vehicle information and the current road information can transition from the previous driving scene (S5), it is determined that the driving scene is the current driving scene (S6). ). The workload is estimated using the current driving scene (S7).

  Also by the driving scene determination apparatus 1 of the second embodiment of the present invention, the same operational effects as the first embodiment are exhibited.

  That is, not only using the driving scene correspondence information but also using the transition availability information, it is possible to correctly determine the current driving scene from the current vehicle information and the current road information, and thus erroneous determination of the driving scene is possible. Generation | occurrence | production can be suppressed and the determination precision of a driving scene can be improved.

  Moreover, in the present embodiment, based on the current driving scene determined in this way, the workload value of the user during vehicle driving can be estimated with high accuracy using a quantification model.

(Third embodiment)
Next, the service providing apparatus 30 according to the third embodiment of this invention will be described. The service providing apparatus 30 according to the present embodiment is an apparatus in which a service providing function is added to the driving scene determination apparatus 1 according to the first embodiment. Therefore, here, the service providing device 30 according to the third embodiment will be described focusing on differences from the driving scene determination device 1 according to the first embodiment. Unless otherwise specified, the configuration and operation of the present embodiment are the same as those of the first embodiment.

  FIG. 11 is a block diagram showing a configuration of service providing apparatus 30 in the present embodiment. As shown in FIG. 11, the workload estimation device 20 includes a service providing unit 31 that provides a service according to the current driving scene, in addition to the configuration of the driving scene determination device 1 of the first embodiment. Yes.

  For example, the service providing unit 31 executes a service for notifying a driver or a passenger of a driving scene name (displaying on a screen, reading out by voice, etc.) as a service corresponding to the current driving scene. Such a driving scene name notification service may be executed to a person outside the company via a network. Moreover, this service provision part 31 may perform the service allocated to each driving scene. For example, in a specific scene (for example, turning right or left or changing lanes), a service for lowering the volume of audio or a service for sounding a warning sound may be executed.

  Such a service providing apparatus according to the third embodiment of the present invention also provides the same operational effects as those of the first embodiment.

  That is, not only using the driving scene correspondence information but also using the transition availability information, it is possible to correctly determine the current driving scene from the current vehicle information and the current road information, and thus erroneous determination of the driving scene is possible. Generation | occurrence | production can be suppressed and the determination precision of a driving scene can be improved.

  Moreover, in the present embodiment, services can be appropriately provided based on the current driving scene determined in this way.

  The embodiments of the present invention have been described above by way of example, but the scope of the present invention is not limited to these embodiments, and can be changed or modified according to the purpose within the scope of the claims. is there.

  As described above, the driving scene determination device according to the present invention has an effect that it can suppress the erroneous determination of the driving scene and can increase the determination accuracy of the driving scene, and is applied to, for example, a car navigation device or the like. Is useful.

DESCRIPTION OF SYMBOLS 1 Driving scene determination apparatus 2 Information acquisition part 3 Control part 4 Vehicle information acquisition part 5 Road information acquisition part 6 GPS apparatus 7 Vehicle sensor 8 Car-mounted camera 9 Road information DB
10 Driving Scene Judgment Unit 11 Driving Scene DB
12 Transition availability DB
13 Scene end determination unit 20 Workload estimation device 21 Model DB
22 Biosensor 23 Workload Estimation Unit 30 Service Providing Device

Claims (11)

A driving scene storage unit that stores driving scene correspondence information in which a driving scene of the vehicle is associated with vehicle information about a vehicle that the user is driving and road information about a road on which the vehicle is passing;
For a plurality of the driving scenes, a scene transition possibility storage unit that stores transition possibility information indicating whether or not a scene transition from one driving scene to another driving scene is possible;
Information acquisition means for acquiring current vehicle information and current road information;
Driving scene determination means for determining a current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information;
With
The driving scene determination means determines that the current driving scene is the current driving scene when the driving scene associated with the current vehicle information and the current road information can transition from the previous driving scene. A driving scene determination device as a feature. The driving scene storage unit stores a plurality of driving scene correspondence information prepared for each of the current driving scenes,
2. The driving scene determination unit according to claim 1, wherein the driving scene determination unit selects driving scene corresponding information corresponding to the current driving scene from the plurality of driving scene corresponding information, and uses the driving scene corresponding information for determination of the current driving scene. Driving scene determination device. The driving scene includes a lane change or a left / right turn,
The vehicle information associated with the lane change or the left / right turn includes a steering angle,
The driving scene determination device according to claim 1 or 2, wherein the set value of the rudder angle increases as the vehicle speed of the vehicle decreases and decreases as the vehicle speed of the vehicle increases. A scene end determination means for determining whether or not the current driving scene is ended based on the current vehicle information;
4. The driving scene determination unit according to claim 1, wherein the driving scene determination unit determines the current driving scene as a previous driving scene is continued until it is determined that the current driving scene has ended. The driving scene determination device according to claim 1. The scene end determination means determines whether or not the lane change driving scene has ended based on the rudder angle and rudder angle change rate included in the vehicle information,
5. The driving scene determination device according to claim 4, wherein the setting values of the rudder angle and the rudder angle change rate increase as the vehicle speed of the vehicle decreases and decrease as the vehicle speed of the vehicle increases.   The driving scene determination means associates with the first driving scene when returning from the second driving scene to the first driving scene after a scene transition from the first driving scene to the second driving scene. 6. The driving scene determination device according to claim 1, wherein hysteresis control is performed to change the vehicle information or the road information being used to make it difficult to return to the first driving scene. . Stored in memory,
Driving scene correspondence information in which the driving scene of the vehicle is associated with the vehicle information regarding the vehicle the user is driving and the road information regarding the road on which the vehicle is passing;
About the plurality of driving scenes, transition availability information indicating whether or not a scene transition from one driving scene to another driving scene,
Using,
On the computer,
Processing to obtain current vehicle information and current road information;
A process of determining a current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information;
In a program that executes
In the determination of the driving scene, when the driving scene associated with the current vehicle information and the current road information can be changed from the previous driving scene, it is determined that the driving scene is the current driving scene. Characteristic driving scene determination program. A driving scene storage unit that stores driving scene correspondence information in which a driving scene of the vehicle is associated with vehicle information about a vehicle that the user is driving and road information about a road on which the vehicle is passing;
For a plurality of the driving scenes, a scene transition possibility storage unit that stores transition possibility information indicating whether or not a scene transition from one driving scene to another driving scene is possible;
Information acquisition means for acquiring current vehicle information and current road information;
A means for determining a current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, the current vehicle information and the current road information Driving scene determination means for determining that the current driving scene is the current driving scene when the associated driving scene is transitionable from the previous driving scene;
The contribution of the user while driving the vehicle based on the correct data in which the correlation between the contribution behavior of the user contributing to the workload related to the driving of the vehicle and the workload value indicating the level of the workload is recorded. Model storage means for storing a quantification model for calculating the workload value from an action;
Data acquisition means for acquiring data related to the contributing behavior from a user who is driving the vehicle;
Using the quantification model, workload estimation means for estimating the user's workload value from data of the user's contribution behavior while driving the vehicle,
With
The workload associated with driving the vehicle is:
Driving workload due to the user's driving operation when driving the vehicle, device operation workload due to the user's device operation when operating the device provided in the vehicle, driving the vehicle Due to the auditory workload due to the user's auditory movement, due to the visual workload due to the user's visual movement when driving the vehicle, due to the user's psychological state when driving the vehicle Including an intrinsic workload, an extrinsic workload resulting from the driving scene of the vehicle,
The workload estimation unit uses the current driving scene determined by the driving scene determination unit as the user's contributing behavior while driving the vehicle. Stored in memory,
Driving scene correspondence information in which the driving scene of the vehicle is associated with the vehicle information regarding the vehicle the user is driving and the road information regarding the road on which the vehicle is passing;
About the plurality of driving scenes, transition availability information indicating whether or not a scene transition from one driving scene to another driving scene,
The contribution of the user while driving the vehicle based on the correct data in which the correlation between the contribution behavior of the user contributing to the workload related to the driving of the vehicle and the workload value indicating the level of the workload is recorded. A quantification model for calculating the workload value from an action;
Using,
On the computer,
Processing to obtain current vehicle information and current road information;
When determining the current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, the current vehicle information and the current road information are associated with each other. A process of determining that the current driving scene is the current driving scene when the given driving scene can transition from the previous driving scene;
Processing for obtaining data related to the contributing behavior from a user driving the vehicle;
Using the quantification model, a process of estimating the user workload value from data of the user's contribution behavior while driving the vehicle;
In a program that executes
The workload associated with driving the vehicle is:
Driving workload due to the user's driving operation when driving the vehicle, device operation workload due to the user's device operation when operating the device provided in the vehicle, driving the vehicle Due to the auditory workload due to the user's auditory movement, due to the visual workload due to the user's visual movement when driving the vehicle, due to the user's psychological state when driving the vehicle Including an intrinsic workload, an extrinsic workload resulting from the driving scene of the vehicle,
In the estimation of the workload value, the current driving scene obtained by the determination process is used as the user's contributing behavior while driving the vehicle. A driving scene storage unit that stores driving scene correspondence information in which a driving scene of the vehicle is associated with vehicle information about a vehicle that the user is driving and road information about a road on which the vehicle is passing;
For a plurality of the driving scenes, a scene transition possibility storage unit that stores transition possibility information indicating whether or not a scene transition from one driving scene to another driving scene is possible;
Information acquisition means for acquiring current vehicle information and current road information;
A means for determining a current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, the current vehicle information and the current road information Driving scene determination means for determining that the current driving scene is the current driving scene when the associated driving scene is transitionable from the previous driving scene;
Service providing means for providing a service according to the determined current driving scene;
A service providing apparatus comprising: Stored in memory,
Driving scene correspondence information in which the driving scene of the vehicle is associated with the vehicle information regarding the vehicle the user is driving and the road information regarding the road on which the vehicle is passing;
About the plurality of driving scenes, transition availability information indicating whether or not a scene transition from one driving scene to another driving scene,
Using,
On the computer,
Processing to obtain current vehicle information and current road information;
When determining the current driving scene from the current vehicle information and the current road information based on the driving scene correspondence information and the transition availability information, the current vehicle information and the current road information are associated with each other. A process of determining that the current driving scene is the current driving scene when the given driving scene can transition from the previous driving scene;
A process of providing a service according to the determined current driving scene;
A service providing program characterized in that the program is executed.