JP2014046820A - Driver's property estimation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driver's property estimation system capable of estimating a fundamental property of a driver.SOLUTION: A scene where a vehicle is rolling is estimated based on information representing an ambient situation imaged by a camera 10, and a driver's driving action is detected based on vehicle information detected by a vehicle signal detection unit 11. Based on the driving action which the driver has performed in the estimated scene, the driver's property is estimated through correlation with a property tendency label stored in a driver's property database 20. Eventually, the driver's property can be properly estimated.

Description

  The present invention relates to a driver characteristic estimation device that estimates driver characteristics based on driver operation information during vehicle driving and vehicle behavior.

  Japanese Patent Application Laid-Open No. 2011-34430 (Patent Document 1) describes an apparatus for estimating driver characteristics (generic name such as personality, tendency, and temperament) based on behavior when a driver drives a vehicle. Things are known. Patent Document 1 discloses that a driver recognizes a driver's heel based on a driving operation when driving the vehicle, and supports driving based on the driver's heel.

JP 2011-34430 A

  However, the conventional example disclosed in Patent Literature 1 described above recognizes a driver's habit and assists in driving, and the basic characteristics of the driver cannot be estimated.

  The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide a driver characteristic estimation device capable of estimating a basic characteristic of a driver. There is.

  To achieve the above object, the present invention provides a scene estimation unit that estimates a scene in which a vehicle travels based on ambient information acquired by the ambient information acquisition unit, a vehicle information acquisition unit that acquires vehicle information, Based on the driving scene and vehicle information, driving behavior detection means for detecting the driving behavior of the driver, driver characteristics storage means for storing the relationship between the characteristics of the driver and the driving behavior that the driver can take, and driving Basic characteristic estimation means for obtaining the characteristics of the driver with reference to the driver characteristic storage means based on the driving behavior of the driver.

  In the driver characteristic estimation device according to the present invention, the driver characteristics are estimated by referring to the data stored in the driver characteristic storage means based on the driving behavior taken by the driver. Basic characteristics can be estimated.

It is a block diagram which shows the structure of the driver characteristic estimation apparatus which concerns on one Embodiment of this invention. It is a figure which shows the correlation coefficient of the other person-oriented emotion reaction and self-oriented emotion reaction which are the items of a basic characteristic, and the other person consideration type, non-sympathetic type, and self-oriented type which are driving | operation high-order characteristics. The figure which shows the correlation coefficient of the other person consideration type which is a characteristic tendency label, self-oriented type, and a non-sympathetic type, and the other person consideration type which is a driving-related signal, a self-oriented type, a rule compliance type, and a non-sympathetic type It is. It is a flowchart which shows the process sequence of the driver | operator characteristic estimation apparatus which concerns on one Embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a driver characteristic estimation device according to an embodiment of the present invention. As shown in FIG. 1, a driver characteristic estimation apparatus 100 according to the present embodiment includes a vehicle signal detection unit (vehicle information acquisition unit) 11, a scene detection unit (scene estimation unit) 12, and a driving behavior detection unit (driving). (Behavior detecting means) 13, basic characteristic estimating section (basic characteristic estimating means) 14, characteristic tendency output section 15, controller (correlation calculating means) 25, driver characteristic database (driver characteristic storage means) 20, It has.

  The vehicle signal detection unit 11 detects various vehicle signals transmitted via, for example, a CAN (Controller Area Network). For example, signals such as the accelerator opening, the presence / absence of a brake operation, the steering angle of the steering wheel, the yaw rate of the vehicle, and the traveling speed of the vehicle are detected.

  The scene detection unit 12 acquires image data captured by a camera (ambient information acquisition unit) 10 mounted on the vehicle, and estimates a traveling scene of the vehicle based on the image data. For example, a scene in which the vehicle is traveling on a curved road, a scene turning at an intersection, a scene in which another vehicle is entering the lane of the host vehicle, and the like are estimated. It is also possible to estimate a traveling scene based on road information acquired from GPS (not shown) instead of the camera 10.

  The driving behavior detection unit 13 detects the driving behavior based on the traveling scene detected by the scene detection unit 12 and the traveling state of the host vehicle. For example, driving behavior such as “traveling on a curved road while decelerating while driving”, “giving the road to the vehicle to enter”, or “turning left on the T-shaped road with good visibility without loosening the accelerator” Is detected.

  Based on the driving behavior detected by the driving behavior detection unit 13, the basic characteristic estimation unit 14 estimates the driver's characteristics by referring to various data stored in the driver characteristics database 20. The characteristic tendency output unit 15 estimates driver characteristics.

  The driver characteristic database 20 stores non-real time information whose score is an index based on a questionnaire scale, and is used when referring to information necessary for real time processing. The driver characteristic database 20 includes a basic characteristic storage unit (basic characteristic storage unit) 21, a driving higher-order characteristic storage unit 22, a characteristic tendency label storage unit (characteristic tendency label storage unit) 23, and a driving behavior storage unit 24. And. Further, a controller 25 is provided between the driver characteristic database 20 and the basic characteristic estimation unit 14, and the controller 25 includes a storage unit 25a that stores a driving behavior characteristic list.

  The driver characteristic estimation apparatus 100 according to the present embodiment can be configured as an integrated computer including a central processing unit (CPU), storage means such as RAM, ROM, and hard disk. In addition, the driver characteristic database 20 and the controller 25 may be mounted on a vehicle, or may be installed in a base station and connected by network communication or the like.

  A basic characteristic storage unit 21 provided in the driver characteristic database 20 sets a static characteristic of the driver. In the present embodiment, as an example, sociality is taken as an example, and a scale for estimating the height of “sympathy” to the surroundings is set. Furthermore, various items relating to “other-oriented emotion reaction” and “self-oriented emotion reaction” are set as subordinate concepts. For example, as an example of the item of “other person-oriented emotional reaction”, there is “there is a novel being read and emotions may be transferred to the characters”. In the present embodiment, social property is exemplified as an example of basic characteristics, and in particular, “empathy” is described as an example, but the present invention is not limited to this.

  The driving high-order characteristic storage unit (driving high-order characteristic storage means) 22 performs factor analysis with the “sympathy” set in the basic characteristic storage unit 21 to show the factor load (indicating the degree of influence). Items whose (quantity) is greater than or equal to a predetermined value (items having a relationship greater than or equal to the predetermined amount) are extracted as statistically significant factors. As a result, for example, three items of “others consideration type”, “self-oriented type”, and “non-sympathetic type” are set. In the present embodiment, the above three items are exemplified, but the present invention is not limited to this.

  Further, the driving higher-order characteristic storage unit 22 displays the items of “other-oriented emotion reaction” and “self-oriented emotion reaction” set as subordinate concepts in the basic characteristic storage unit 21 during driving. "Others consideration type", "Self-oriented type", and "Non-sympathy" calculated as factors that are statistically significant by factor analysis. The correlation coefficient between each sub-item of “type” is obtained.

  Specifically, as shown in FIG. 2, “other person-oriented emotion reaction” and “self-oriented emotion reaction” set as subordinate concepts in the basic characteristic storage unit 21 are described in the vertical direction, and higher driving The “others consideration type”, “non-sympathetic type”, and “self-oriented type” set as subordinate concept items in the characteristic storage unit 22 are described in the horizontal direction, and the score obtained for each item is statistically calculated. Processing is performed to obtain correlation coefficients a1 to a6 between the items. When the correlation coefficients a1 to a6 are equal to or greater than a preset threshold value, it is determined that the item can be set as the characteristic tendency label. In the example shown in FIG. 2, the correlation coefficient a1 between the “other-oriented emotional response” and the “other-friendly emotional type”, and the “self-directed emotional response” and the “self-oriented type”. It is determined that the correlation coefficient a6 is equal to or greater than the threshold value.

  The characteristic tendency label storage unit 23 illustrated in FIG. 1 stores three items of “others consideration type”, “self-oriented type”, and “non-sympathetic type”, which are items having a large correlation coefficient.

  The driving behavior storage unit 24 stores the above three items of “others consideration type”, “self-oriented type”, and “non-sympathetic type” as driving behavior by the driver, and further subordinate items of sociality And add the “rule compliant type” item. Specifically, each vehicle signal such as accelerator opening, presence / absence of brake operation, steering angle, yaw rate, vehicle speed, or a signal indicating a driving situation detected by, for example, a moving image (hereinafter referred to as “driving”). (Referred to as "related signals") and further driving behavior based on these signals. And based on the driving situation and the vehicle signal, four items, namely “others consideration type” (label 1), “self-oriented type” (label 2), “rule compliance type” (label 3), “non-sympathy” Each item of “type” (label 4) is stored.

  The controller 25 statistically processes the various characteristic tendency labels set in the characteristic tendency label storage unit 23 and the score obtained for each item of the driving-related signal, and has a correlation coefficient equal to or greater than a predetermined value. The driving-related signal item is extracted and stored in the storage unit 25a as a driving behavior characteristic list.

  FIG. 3 shows three items set as characteristic tendency labels, that is, “other-friendliness type”, “self-oriented type”, and “non-sympathetic type” items in the vertical axis direction as driving-related signals. 4 items memorized, “Others consideration type” (label 1), “Self-oriented type” (label 2), “Rule compliance type” (label 3), “Non-sympathetic type” (label 4) These items are shown in the horizontal axis direction, and correlation coefficients b1 to b12 between the items are shown. For example, the correlation coefficient between the “other consideration type” of the characteristic tendency label and the “other consideration type (label 1)” of the driver related signal is b1.

  Then, as shown in FIG. 3, the combination of the characteristic tendency label whose driving coefficient is the absolute value of the correlation coefficient and the driving related signal is stored in the storage unit 25a as a driving action characteristic list. In the example shown in FIG. 3, b1, b3, b4, b6, b7, b11, b12 are selected.

  Here, the driving-related signal “consideration for others” means that there are pedestrians, bicycles, and other vehicles within the range that can be related to the vehicle. ) To show the action that determines the response of the vehicle. Specific driver actions include "I was worried about the existence of the following car when turning right and stepped on the accelerator", "I sent a hand signal to a pedestrian trying to cross a pedestrian crossing without a traffic light", etc. It is an action.

  In addition, “self-oriented” means that there are pedestrians, bicycles, and other vehicles within the range that can be related to the own vehicle, and the optimal view from the viewpoint of other vehicles and the overall traffic balance are taken. It is to show the action which decided the correspondence of the own vehicle in consideration. As a specific driver's action, “When passing through a narrow road, if the oncoming vehicle is congested due to reasons such as the following vehicle continuing, the vehicle stops and passes the oncoming vehicle.” And so on.

  “Rule observance type” means an action in which pedestrians, bicycles, and other vehicles exist within a range that can be related to the own vehicle, and the action of the own vehicle is determined according to the traffic rules. Or, even if it is not stipulated in the traffic rules, it is to show an action that emphasizes safety rather than consideration for pedestrians, bicycles, and other vehicles. Specific actions of the driver are actions such as “when turning right in a situation with poor visibility from a narrow road, stopping without a mistake before the temporary stop line, and then proceeding slowly”.

  “Non-sympathetic” means that even if there are pedestrians, bicycles, and other vehicles within the range that can be related to the vehicle, the behavior of the vehicle is given priority without paying attention to them. ". A specific driver's action is an action such as “Despite a pedestrian waiting at a pedestrian crossing without a traffic light, he has passed the pedestrian crossing”.

  Next, the operation of the driver characteristic estimation apparatus 100 according to the present embodiment configured as described above will be described. As described above, the driver characteristic database 20 stores characteristic tendency labels that are items related to “empathy” set as basic characteristics. And in this embodiment, the action which the driver took at the time of driving is detected, the state around the vehicle is further detected, and the characteristics of the driver are estimated based on these results. This will be described in detail below with reference to the flowchart shown in FIG.

  First, in step S11, the vehicle signal detection unit 11 shown in FIG. 1 acquires a vehicle signal that is a signal indicating a vehicle operation signal or behavior through CAN. Specifically, the accelerator opening, presence / absence of brake operation, steering angle, yaw rate, vehicle speed, navigation operation, and the like are acquired. This vehicle signal is output to the driving behavior detector 13.

  Next, in step S12, the scene detection unit 12 detects the environment of the road on which the vehicle is currently traveling. Specifically, an image around the vehicle is captured by the camera 10 mounted on the vehicle, and a road environment where the vehicle is currently traveling is detected based on the captured image data. Alternatively, the road environment where the vehicle is currently traveling is detected based on the GPS signal. This detection signal is output to the driving action detection unit 13.

  In step S <b> 13, the driving behavior detection unit 13 detects driving behavior based on the road environment information output from the scene detection unit 12 and the vehicle signal detected by the vehicle signal detection unit 11. In this process, for example, driving behavior information in a surrounding scene such as “turning left without relaxing the accelerator on a T-shaped road with a good view” is detected. The detected driving behavior information is output to the basic characteristic estimation unit 14.

  In step S <b> 14, the basic characteristic estimation unit 14 acquires driving behavior information, and further outputs the acquired driving behavior information to the controller 25.

  In step S15, the controller 25 compares the driving action information input from the basic characteristic estimation unit 14 with the driving action stored in the storage unit 25a, and matches or resembles the input driving action information. Are extracted from the driving behavior characteristic list stored in the storage unit 25a. In this process, from the driving behavior characteristic list stored in the storage unit 25a, for example, the above-mentioned driving behavior such as “running without loosening the accelerator in front of a pedestrian crossing without a signal”, that is, “T-shape with good visibility”. Search for items similar to "Turning left without loosening the accelerator on the road".

  In step S16, the controller 25 reads a driving-related signal corresponding to the extracted driving behavior from the driving behavior storage unit 24, and searches for a trend label related thereto. At this time, if there is a driving action having a correlation coefficient between the characteristic tendency label and a predetermined value or more, the driving action is stored in the driving action storage unit 24, and the driving action and the characteristic tendency label are further stored. Is stored in the storage unit 25a. Thereby, the data amount of driving action can be increased and it can be used for the estimation of the characteristic of the driver performed after the next time.

  In step S17, the basic characteristic estimation unit 14 estimates the characteristics of the driver based on the retrieved tendency label. As a result, the driver's characteristics can be estimated, for example, as “non-sympathetic”. In this way, the characteristics of the driver can be estimated based on various data stored in the driver characteristics database 20 based on the driving state of the vehicle and the behavior of the driver.

  And when a driver | operator's characteristic is estimated, based on this characteristic, it becomes possible to operate a vehicle-mounted apparatus so that it may match a driver | operator's characteristic.

  For example, when operating the navigation device, if the driver's characteristic is estimated to be “non-sympathetic”, to avoid troubles with other vehicles or others, it is preferentially escaped. Present. In addition, if the driver's characteristics are estimated to be “other-friendly”, part of the vehicle behavior of other vehicles may be used to show that the vehicle is in cooperation with surrounding vehicles even during traffic jams. Present in the vehicle. By doing so, the navigation device can be appropriately operated according to the characteristics of the driver.

  In this way, in the driver characteristic estimation apparatus 100 according to the present embodiment, the driver characteristic database 20 stores data related to the basic characteristic (for example, “sympathy”) of the driver, and the driver actually The characteristics of the driver can be estimated based on the relationship between the vehicle operation and the behavior of the vehicle when driving the vehicle, and the surrounding situation. Therefore, when operating the navigation device or the like, it is possible to respond in a manner suitable for the characteristics of the driver.

  Further, an item having a predetermined relationship with the basic characteristic is taken up as a driving high-order characteristic, and an item having a correlation coefficient greater than a predetermined value with respect to the item of the driving high-order characteristic is set as a characteristic label. . Since the driver's characteristics are estimated based on the relationship between the set characteristic label and the driving behavior of the driver, the driver's characteristics can be estimated in detail, and various processes corresponding to the characteristics can be performed. Can be performed smoothly.

  Further, when the correlation coefficient between the characteristic tendency label and the driving action is a predetermined value or more, the driving action is added to the driving action characteristic list, so that data based on the action taken by the driver can be accumulated. This makes it possible to more accurately estimate the driver's characteristics with respect to driving behavior.

  Further, since the vehicle signal detection unit 11 detects at least one of the accelerator opening, the brake operation, the steering steering angle, the yaw rate, the vehicle speed, and the navigation operation of the vehicle, the vehicle operation and the behavior of the vehicle are improved. It can be recognized with high accuracy, and the driving behavior of the driver can be detected more reliably.

  As mentioned above, although the driver characteristic estimation apparatus of this invention was demonstrated based on embodiment of illustration, this invention is not limited to this, The structure of each part is set to the thing of the arbitrary structures which have the same function. Can be replaced.

  For example, in the present embodiment, “sympathy” is given as an example of the basic characteristics of the driver, and “others consideration type”, “self-oriented type” are examples of factors having a large factor load with respect to this “sympathy”. “Three items of“ non-sympathetic type ”are given as examples, but these are examples, and the present invention is not limited to these items.

  The present invention can be used to estimate the characteristics of a driver based on the driving operation of the vehicle.

DESCRIPTION OF SYMBOLS 10 Camera 11 Vehicle signal detection part 12 Scene detection part 13 Driving action detection part 14 Basic characteristic estimation part 15 Characteristic tendency estimation part 20 Driver characteristic database 21 Basic characteristic storage part 22 Driving high-order characteristic storage part 23 Characteristic tendency label storage part 24 Driving behavior storage unit 25 Controller (correlation calculation means)
25a Storage unit 100 Driver characteristic estimation device

Claims (4)

In a driver characteristic estimation device that estimates a driver's characteristics based on driving operation of a vehicle,
Surrounding information acquisition means for acquiring surrounding information of the vehicle;
Scene estimating means for estimating a scene in which the vehicle travels based on the surrounding information acquired by the surrounding information acquiring means;
Vehicle information acquisition means for acquiring vehicle information which is information related to operation by the driver or information related to vehicle behavior;
A driving behavior detection means for detecting a driving behavior of a driver based on the scene in which the vehicle travels and the vehicle information;
Driver characteristic storage means for storing the relationship between the characteristics of the driver and the driving actions that the driver can take;
Based on the driving behavior detected by the driving behavior detection means, with reference to the driver characteristic storage means, basic characteristic estimation means for obtaining the characteristics of the driver;
A driver characteristic estimation device comprising: The driver characteristic storage means is
Basic characteristic storage means for storing basic characteristics of the driver and items that are subordinate concepts thereof;
Driving high-order characteristic storage means for storing items having a relationship of a predetermined amount or more with the basic characteristic as driving high-order characteristics;
Based on the basic characteristics and the driving higher-order characteristics, an item having a correlation coefficient equal to or greater than a predetermined value is extracted as a characteristic trend label, and characteristic trend label storage means for storing the characteristic trend label;
Driving behavior storage means for storing items about the driving behavior of the driver as driving behavior items;
In addition,
Correlation calculating means for obtaining a correlation coefficient between the driving behavior item and the characteristic tendency label and storing the driving behavior item and the characteristic tendency label in which the correlation coefficient is equal to or greater than a predetermined value,
When the driving behavior of the driver is acquired, the basic characteristic estimation unit extracts the driving behavior item related to the driving behavior, and based on a correlation coefficient between the driving behavior item and the characteristic tendency label The driver characteristic estimation apparatus according to claim 1, wherein the characteristic of the driver is estimated.   The correlation calculating means drives the item as the driving action when the correlation coefficient between the characteristic tendency label and the item as the driving action detected by the driving action detecting means is a predetermined value or more. The driver characteristic estimation apparatus according to claim 1, wherein the driver characteristic estimation apparatus is added to the behavior characteristic list.   The vehicle information acquisition means is at least one of an accelerator opening, a brake operation, a steering steering angle, a yaw rate, a vehicle speed, and a navigation operation. The driver characteristic estimation device according to the item.

Images