JP5664533B2 - Vehicle driver identifying learning device, vehicle driver identifying learning method, vehicle driver identifying device, and vehicle driver identifying method - Google Patents

Description

  The present invention relates to a learning device for specifying a vehicle driver, a learning method for specifying a vehicle driver, and a vehicle driver specifying device using the learning device and the learning method. The present invention relates to a vehicle driver identification method.

  In recent years, in order to enhance vehicle security, a system has been developed in which a vehicle driver is specified and whether or not the vehicle driver is an authorized user based on the specified result. Development of a system that automatically adjusts the position of the rearview mirror, seat position, etc. according to the driver of the vehicle based on this specific result is also under consideration.

  On the other hand, as a method for identifying such a driver, the ID key or ID card owner, that is, the vehicle, is verified by comparing the ID information given to the ID key or ID card with the ID information registered in advance in the in-vehicle device. There is a way to identify drivers. That is, in this method, a physical authentication medium such as an authentication ID key or ID card is distributed in advance to an authorized user, and ID information given to the ID key or ID card is registered in the vehicle-mounted device in advance. Will be.

  By the way, in such a method of specifying a driver using a physical medium such as an ID key or an ID card, the possibility that the ID key or ID card is illegally obtained cannot be denied. If the ID key or ID card is illegally obtained, it is difficult to ensure security, for example, a driver who is not a regular user is mistakenly authenticated as if it is a regular user. Furthermore, when trying to identify a driver using such a physical medium, the driver needs to be authenticated using the same medium every time the vehicle is used, which is inconvenient. .

  Therefore, for example, in a conventional system including the system described in Patent Document 1, feature amounts such as a door lock procedure performed by each driver, an accelerator pedal depression cycle, a steering operation speed, and the like are learned. I am trying to register in the database separately. Then, the feature quantity registered in the database is compared with the feature quantity of the driving operation actually performed by the driver, and the driving operation having the same feature quantity as the feature quantity registered in the database is registered in the database. The driver is identified as being a driving operation by the driver. As a result, the driver of the vehicle can be identified without using a physical medium such as an ID key or an ID card, or without requiring the driver to perform a predetermined operation for identifying the driver. .

JP 2001-63400 A

  By the way, when the data indicating the feature value of the driving operation is registered in the database, for example, the data having different feature values are registered in the database as data acquired based on the driving operation of different drivers. However, for example, the amount of driving operations performed on a vehicle traveling on the road, such as the accelerator pedal depression cycle and the steering operation speed, and the operation amount of various operation elements are not It varies greatly depending on the driving environment. That is, even for the same driver, the accelerator pedal depression cycle, the steering operation speed, and the like do not always have a constant tendency depending on the traveling environment around the vehicle. For this reason, if the amount of operation of various operating elements by the driver changes due to changes in the driving environment while the same driver is driving, data indicating this characteristic amount is acquired based on the driving operation of different drivers. Will be registered in the database as processed data. As a result, it becomes difficult not only to register data based on the driving operation of the driver in the database separately for each driver, but also to reduce the accuracy of specifying the driver using the data registered in this database.

  The present invention has been made in view of such circumstances, and its purpose is to learn for driver identification based on the driving operation of the driver through accurate management of data acquired based on the driving operation of the driver. Vehicle driver identification learning device and vehicle driver identification learning method that can be performed with higher accuracy, and vehicle driver identification device and vehicle driver identification method using the vehicle driver identification learning device and vehicle driver identification learning method It is to provide.

Hereinafter, means for solving the above-described problems and the effects thereof will be described.
According to one aspect of the present invention, a learning vehicle driver specific learning apparatus that performs for a particular driver based on the driving operation mode of the driver to a driving operation of the vehicle, the driving operation mode of the driver, driving operation A learning unit that learns by classifying each driver based on a feature amount including at least one of the operation timing and operation amount of the element, and a trip that is a period from when the driver starts driving operation to when the driving operation ends when driving operation mode in units of are those which are classified into two or more drivers through learning by the learning section, a learning result of any one of the driver's driving operation mode classification rate in the trip is in the upper And a management unit that unifies and manages the driving operation mode with the trip as a unit .

The invention according to claim 7 is a vehicle driver specifying learning method for performing learning for specifying a driver based on a driving operation mode of a driver who drives the vehicle, wherein the driving operation mode of the driver is changed to a driving operation. A step of learning while classifying by driver based on a feature amount including at least one of the operation timing and operation amount of the element, and a trip that is a period from when the driver starts driving operation to when the driving operation ends when driving operation mode in units are those which are classified into two or more drivers through the step of the learning, the learning result of the driving operation mode of any one of the drivers of the classification ratio of in the trip upper, And a step of unifying and managing the driving operation mode in units of the trip .

  In a trip, which is a period from when the vehicle starts to travel until it stops traveling, the probability that the driver will change is very low, and the driving operation is usually performed by the same driver. Moreover, many of the driving operation modes of the driver performed in one trip reflect patterns unique to the driver. However, in some driving operation modes, the driving operation mode of other drivers may be approximated due to the driving environment of the vehicle, or the driver-specific wrinkles may be difficult to be reflected. For this reason, if it is going to classify each driving operation mode appropriately based on each driving operation performed appropriately in the trip, each driving operation mode performed in one trip is specified to belong to a plurality of drivers. It can also be a thing.

  Therefore, as in the above configuration or method, the driving operation mode is classified according to the driver based on the feature amount. As a result of this classification, if the driving operation mode performed during the trip was classified as that of multiple drivers, the driver with the highest classification ratio in the trip operated the trip. Identify the driver. Then, assuming that the specified driver is a driver who has performed each driving operation mode in the trip, each driving operation mode is unified and managed as that of the driver. For this reason, each driving operation mode performed by one driver in a certain trip is not subdivided as being based on a plurality of drivers. That is, it becomes possible to specify the driver in trip units. In addition, a driver whose classification ratio is higher in a certain trip is very likely to be a driver who actually operated the trip. Therefore, by specifying the driving operation mode in the trip as that of the driver whose classification ratio is higher in the trip, it is possible to specify with high accuracy which driver the driving operation mode is.

According to a second aspect of the present invention, in the vehicle driver identification learning device according to the first aspect, a driving operation mode of a driver having a higher classification ratio in the trip is classified in the trip. The driving operation mode of the driver having the highest number of times, and the driving operation mode satisfying at least one of the driving operation modes of the driver having the higher driving time length of the vehicle classified in the trip The gist.

The invention according to claim 8 is the vehicle driver identification learning method according to claim 7, wherein the number of times the classification ratio in the trip is classified in the trip as a driving operation mode of a higher-order driver. and a brief but driving operation mode of upper driver, and that the length of the travel time of the vehicle classified in the trip selects at least one condition is satisfied driving operation mode of driving operation mode of higher driver To do.

  The classification result of the driver based on the feature value of the driving operation mode is classified more frequently as the driver actually performs the driving operation. Similarly, among the drivers identified as having traveled on a certain trip, the probability that the trip was actually traveled is higher as the driver whose driving time is higher in the vehicle.

Therefore, as in the above configuration or method, as the driving operation mode at the top of the classification ratio,
(B) A driving operation mode in which the number of times classified in the trip is higher and (b) a driving operation mode that satisfies at least one of the driving operation modes in which the travel time of the vehicle in the trip is higher is specified. To do. Thereby, it becomes possible to unify the driving operation mode based on the classification ratio with higher accuracy.

  According to a third aspect of the present invention, in the vehicle driver identifying learning device according to the first or second aspect, the database further includes a database in which learning results from the learning unit are registered, and the management unit includes: a. In registering the learning result in the database, when the driving operation mode in the trip is classified into a plurality of drivers, the learning result is unified as the learning result based on the driving operation of one driver, A process of registering a unified learning result in the database; and b. When the driving operation mode in the trip registered in the database is classified into a plurality of drivers, the learning result registered in the database is unified as a learning result based on the driving operation of one driver. The gist is to execute any one of the processes.

  According to a ninth aspect of the present invention, in the vehicle driver specifying learning method according to the seventh or eighth aspect, as the learning step, the learning result is registered in a database and the managing step includes: a. In registering the learning result in the database, when the driving operation mode in the trip is classified into a plurality of drivers, the learning result is unified as the learning result based on the driving operation of one driver, Registering unified learning results in the database; and b. When the driving operation mode during the trip registered in the database is classified into a plurality of drivers, the learning result registered in the database is unified as a learning result based on the driving operation of one driver, The gist is to execute any one of the steps.

  According to the above configuration or method, the number of learning results is limited to the necessary minimum number before the learning results are registered in the database by executing the process or step a. As a result, pre-unified learning results are appropriately registered, and only the minimum necessary learning results can be registered in the database.

  In addition, according to the above configuration or method, even if a learning result that is not unified is once registered in the database by executing the process or step b, the learning result is unified afterwards. You can plan. As a result, it is possible to unify the learning results already registered in the database, and it is possible to accurately manage the learning results regardless of the registration timing of the learning results.

  According to a fourth aspect of the present invention, in the vehicle driver specifying learning device according to any one of the first to third aspects, the driving operation mode includes one or more traffic elements through which the vehicle passes during a trip. It is the operation mode of a plurality of types of driving elements performed originally, and the learning unit classifies the driving operation mode according to the driver for each of the one to a plurality of traffic elements.

  A tenth aspect of the present invention is the vehicle driver identification learning method according to any one of the seventh to ninth aspects, wherein the driving operation mode includes one or more traffic elements through which the vehicle passes during a trip. The gist is to select the operation modes of a plurality of types of driving elements performed originally and classify the driving operation modes according to the driver for each of the one or more traffic elements in the learning step.

  Normally, the vehicle passes through one or more traffic elements that exist between the departure point and the destination point between the start and stop of the vehicle. During this time, a driving operation corresponding to the traffic element is performed by the driver of the vehicle. In addition, driver-specific habits are easily reflected in the driving operation performed under this traffic element, while the characteristic amount of the driving operation mode for the traffic element also changes depending on the characteristics of the traffic element and the surrounding environment. That is, when one or more traffic elements are included in the route traveled by the vehicle during the trip, the driver is classified for each traffic element, and as a result, the classification result of the driving operation elements may be different for each traffic element. is there.

  In this regard, according to the above-described configuration or method, each operation is performed while learning the operation modes of a plurality of types of driving elements performed under one or more traffic elements through which the vehicle passes during a trip as the driving operation modes. While classifying the mode for each driver, it is possible to specify the once-classified operation mode as that of a driver having a very high probability of actually performing the driving operation. Thereby, while learning the driving operation mode by the driver of the vehicle that has traveled on the road where various traffic elements exist, it is possible to suitably improve the learning accuracy.

  According to a fifth aspect of the present invention, in the vehicle driver identifying learning device according to the fourth aspect, the driving operation mode includes operation modes of at least two elements of an accelerator pedal, a turn lamp, and a brake pedal. The gist.

  According to an eleventh aspect of the present invention, in the vehicle driver identification learning method according to the tenth aspect, the operation mode of at least two elements of an accelerator pedal, a turn lamp, and a brake pedal is selected as the driving operation mode. Is the gist.

  Elements such as accelerator pedals, turn lamps, and brake pedals are operated according to their characteristics at each point of a predetermined traffic element such as an intersection or a curve, and these elements are frequently operated. There is a tendency that the operation mode is particularly different among drivers.

  Therefore, according to the above configuration or method, the operation mode of elements that are easily used in normal driving operations such as an accelerator pedal, a turn lamp, and a brake pedal that are performed under a series of points, and that are likely to reflect driver-specific wrinkles. It is possible to suitably improve the learning accuracy while making the learning object. As a result, it is easy to reflect driver-specific wrinkles as learning targets, so it is possible to use elements that have a high frequency of operation by the driver, in other words, elements that have many opportunities to learn. become.

  The invention according to claim 6 is a vehicle driver specifying device that specifies a driver based on a driving operation mode of a driver who drives the vehicle, and the vehicle driver according to any one of claims 1 to 5. A driver that learns a driving operation mode for specifying a driver using a learning device for specifying, and performs processing for specifying a driver to be specified through collation between the learned result and the driving operation mode of the driver obtained each time The gist is to include a specific processing unit.

  The invention according to claim 12 is a vehicle driver specifying method for specifying a driver based on a driving operation mode of a driver who drives the vehicle, and the vehicle driver according to any one of claims 7 to 11. Including a step of learning a driving operation mode for specifying a driver using a learning method for specifying, and specifying a driver to be specified through collation between the learned result and the driving operation mode of the driver obtained each time. Is the gist.

  According to the above configuration or method, the driving operation mode for specifying the driver is learned using the vehicle driver specifying learning device. Thereby, the driving operation mode performed in the vehicle is unified and managed in units of trips. When the driver is specified, the driver is specified by collating the driving operation mode managed in a unified manner with the driving operation mode of the driver obtained each time. For this reason, it becomes possible to unify and manage the driving operation modes necessary for specifying the driver in units of trips, and to minimize the number of driving operation modes that need to be managed to identify the driver based on the driving operation modes. It becomes possible to limit.

Schematic configuration of a vehicle to which each device and each method are applied in the first embodiment of the vehicle driver specifying learning device, the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method according to the present invention. FIG. The figure which shows an example of the driving operation mode of the vehicle performed in 1 trip. The figure which shows the feature-value of a driving | operation operation aspect according to a driver. The figure which shows an example of the classification ratio of the driving operation aspect classified according to the driver. The flowchart which shows an example of the unification / management procedure of the driving operation aspect by both the driver specific learning apparatus and vehicle driver specific learning method of the embodiment, and the vehicle driver specific apparatus and vehicle driver specific method. The flowchart which shows an example of the driver specific procedure by the both driver specific learning apparatus and vehicle driver specific learning method of the embodiment, and the vehicle driver specific apparatus and vehicle driver specific method. Schematic configuration of a vehicle to which each device and each method are applied in the second embodiment of the vehicle driver specifying learning device and the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method according to the present invention. FIG. The figure which shows the feature-value of the driving operation aspect classified according to the presence or absence of a front vehicle according to a driver. The flowchart which shows an example of the unification / management procedure of the driving operation aspect by both the driver specific learning apparatus and vehicle driver specific learning method of the embodiment, and the vehicle driver specific apparatus and vehicle driver specific method. The flowchart which shows an example of the driver specific procedure by the both driver specific learning apparatus and vehicle driver specific learning method of the embodiment, and the vehicle driver specific apparatus and vehicle driver specific method.

(First embodiment)
Hereinafter, a vehicle driver specifying learning device and a vehicle driver specifying learning method, and a vehicle driver specifying device and a vehicle driver specifying method according to a first embodiment of the present invention will be described with reference to FIGS. explain. Note that the vehicle driver specifying learning device and the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method of the present embodiment are applied to, for example, a vehicle shared by a plurality of users.

As shown in FIG. 1, the vehicle 100 is provided with driving operation detection means 110 that detects operation modes of various driving operation elements provided in the vehicle 100.
The driving operation detection unit 110 includes, for example, an accelerator sensor 111, a brake sensor 112, a winker switch 113, a steering angle sensor 114, and the like. The driving operation detection means 110 outputs the detection results of the sensors 111 to 114 to the in-vehicle control device 130 that controls various in-vehicle devices based on the detection results.

  The accelerator sensor 111 detects the amount of depression of the accelerator pedal by the driver of the vehicle 100 and outputs the detected result to the in-vehicle control device 130. The brake sensor 112 detects the depression amount of the brake pedal by the driver of the vehicle 100 and outputs the detected result to the in-vehicle control device 130. The blinker switch 113 detects a use state of a turn lamp by the driver of the vehicle 100, that is, a direction indicator, and outputs the detection result to the in-vehicle control device 130. The steering angle sensor 114 detects the steering operation angle operated by the driver of the vehicle 100 and outputs the detection result to the in-vehicle control device 130.

  In addition, the vehicle 100 is provided with a GPS 120 that receives a satellite signal and detects the absolute position of the vehicle 100, that is, the latitude and longitude based on the received satellite signal. The GPS 120 detects the latitude / longitude of the vehicle 100 that changes as the vehicle 100 moves, in other words, the latitude / longitude of each point where the vehicle 100 travels, and sends the latitude / longitude data indicating the detection result to the in-vehicle control device 130. Output.

  As described above, the vehicle-mounted control device 130 according to the present embodiment appropriately receives data indicating operation modes of various driving operation elements detected by the driving operation detection unit 110 and latitude / longitude data indicating the latitude / longitude of the vehicle 100. Is done.

  The in-vehicle controller 130 detects various in-vehicle devices such as an engine, a brake, a turn lamp, and a steering on the basis of the detection result input from the driving operation detection unit 110, that is, data indicating the operation mode of various driving operation elements by the driver. A system control unit 131 for controlling is provided. The in-vehicle control device 130 also includes a database processing unit 132 that records data input from the driving operation detection unit 110 and the GPS 120 in the database 140. Furthermore, the in-vehicle control device 130 according to the present embodiment includes a driver identification processing unit 133 that performs processing for identifying the driver of the vehicle 100.

  Among these, when the detection result of each driving operation element is input from the driving operation detection unit 110, the system control unit 131 controls various in-vehicle devices based on the detection result. Thereby, for example, when the detection result of the accelerator sensor 111 changes due to the accelerator pedal being depressed by the driver, the engine control amount is calculated according to the detection result, and the engine is controlled according to the calculation result. .

  The database processing unit 132 of the present embodiment includes a learning unit 132a that learns the operation modes of driving operation elements such as an accelerator pedal and a brake pedal by the driver of the vehicle 100 while classifying them according to the feature amount. Yes. In addition, the database processing unit 132 includes a management unit 132b that manages the learning result of the learning unit 132a. The learning unit 132a performs a process of learning while classifying the driving operation mode of the driver of the vehicle 100 by the driver based on the feature amount. Moreover, the management part 132b performs the process which unify | regulates and manages the learning result of the learning part 132a for every trip which is a period from the time the driver starts the driving operation of the vehicle 100 to the end of the driving operation.

  When the detection result of each driving operation element is input from the driving operation detection unit 110, the database processing unit 132 including the learning unit 132a and the management unit 132b is input from the GPS 120 to the driving operation data indicating the detection result. Associate latitude and longitude data. Then, the database processing unit 132 records the associated data in the database 140 separately from the points indicated by the latitude and longitude data.

  When the detection result of each driving operation element and the latitude / longitude data of the vehicle 100 are input from the driving operation detection unit 110 and the GPS 120, the driver identification processing unit 133 according to the present embodiment stores past data already recorded in the database 140. Based on the data indicating the operation mode of each driving operation element and the corresponding latitude / longitude data, a process of specifying the driver is performed. Further, the driver identification processing unit 133 outputs the driver identification result to the driving assistance unit 150 that performs various types of driving assistance based on the identification result.

  On the other hand, each time a driving operation element is driven by a driving operation by a driver, data indicating the operation mode of each driving operation element is recorded in the database 140 together with latitude / longitude data indicating a point where the vehicle 100 has traveled.

  Note that the database processing unit 132 according to the present embodiment, for example, data indicating the operation contents of driving operation elements such as an accelerator pedal, a turn lamp, a brake pedal, and a steering wheel, and when each driving operation element is operated. The travel point of the vehicle 100, that is, the data indicating the latitude and longitude is registered in the database 140 for each traffic element.

  Further, the database processing unit 132 according to the present embodiment, for example, data indicating the operation mode of the driving operation element acquired from when the ignition key of the vehicle 100 is turned on to when it is turned off, and latitude and longitude The data is registered in the database 140 as one driving history, that is, the driving history of the same driver. Thereby, in the database 140, data indicating the operation mode of the driving operation element performed for each point is registered for each driver. In the database 140 of the present embodiment, for example, data indicating the operation mode of the driving operation element is recorded in association with data indicating the seat position of each driver, the position of the door mirror, and the like.

  When the driver identification result is input from the driver identification processing unit 133, the driving support unit 150 extracts data indicating the identified driver's seat position, door mirror position, and the like from the database 140. Then, the driving support unit 150 adjusts the seat position and the position of the door mirror to the position indicated by the extracted data. Thus, the seat position and the door mirror are automatically adjusted according to the wrinkle unique to the driver of the vehicle 100. Further, when the driver identification result is input from the driver identification processing unit 133, the driving support unit 150 identifies the operation mode of the driving operation element of the identified driver, in other words, the driving operation pattern of the driver. . Then, the driving support unit 150 performs driving support according to the identified driving operation pattern of the driver through voice guidance, image guidance, semi-automatic driving, and the like. As this driving support, for example, when the identified driver has a late timing to turn off the accelerator pedal at a point that requires deceleration such as an intersection or a curve, and the amount of depression of the brake pedal tends to be excessive, it is more than usual. Driving assistance that prompts the accelerator pedal to turn off and the brake pedal to turn on at the earliest timing is performed. Further, for example, driving assistance for guiding an appropriate depression amount of the brake pedal when the vehicle 100 is decelerated, and driving assistance for guiding an appropriate depression amount of the accelerator pedal when the vehicle 100 is started are performed. Conversely, when the identified driver can be identified as having a tendency to perform a stable driving operation with less vehicle behavior or a fuel-saving driving, the driving support unit 150 does not perform driving support for the driver. As a result, driving assistance according to the driver-specific bag and driving skill specified respectively is automatically performed.

  Next, the vehicle driver identification learning device and vehicle driver identification learning method of the present embodiment, and the learning principle of the driving operation mode by the vehicle driver identification device and the vehicle driver identification method will be described with reference to FIGS. To do.

  As illustrated in FIG. 2, for example, it is assumed that the vehicle 100 operated by the driver Da departs from the departure place Ps, passes through the intersections SC1 to SC3, and travels to the destination Pg. In the example here, for example, the period from when the vehicle 100 that departs from the departure point Ps where the ignition key is turned on to the destination Pg until the ignition key is turned off is treated as the trip.

  Here, when the driver who drives the vehicle 100 passes through the intersection SC1, at the timings t1, t2, and t3 before passing the intersection SC1, for example, the accelerator pedal is turned off. It is assumed that each driving operation element is operated in such a manner that the turn lamp is turned on and the brake pedal is turned on. In the present embodiment, the timing at which these driving operation elements are operated and the operation amount of each driving operation element are learned as the characteristic amount of the driving operation mode.

  Next, when the vehicle 100 passes through the intersection SC2 having a lane or road shape similar to that of the intersection SC1, the driver of the vehicle 100 will receive the timings t4, t5, passing through the points P4, P5, and P6 before the intersection SC2. At t6, for example, the accelerator pedal is turned off, the turn lamp is turned on, and the brake pedal is turned on.

  At this time, the distance from each point P4 to P6 where the accelerator pedal is turned off, the turn lamp is turned on, and the brake pedal is turned on to the intersection SC2 approximates the distance from the intersection SC1 to each point P1 to P3. It has become a thing. That is, since the intersection SC1 and the intersection SC2 have similar lanes and road shapes, the driving operation modes of the vehicle 100 at the intersections SC1 and SC2 are common or similar.

  On the other hand, when the vehicle 100 is present ahead in the traveling direction when the vehicle 100 travels the intersection SC3 having a lane or road shape similar to the intersections SC1 and SC2, the vehicle For example, the timing at which the accelerator pedal is turned off, the turn lamp is turned on, and the brake pedal is turned on is changed by 100 drivers. That is, the distance from each point P7 to P9 where the accelerator pedal is turned off, the turn lamp is turned on, and the brake pedal is turned on to the intersection SC3 is different from the distance from the intersection SC1 to each point P1 to P3. The timings t7 to t9 at which the operation is performed are different from the timings t1 to t3. As a result, although the same driver Da is traveling on similar intersections SC1 and SC3, the feature quantities of the driving operation modes at the intersections SC1 and SC3 are different. The feature amount of the driving operation mode at the intersection SC3 is, for example, approximate to the feature amount of the driver Db different from the driver Da that actually drives the vehicle 100.

  As described above, even if the driver is the same, depending on the traveling environment of the vehicle 100, the feature amount of the driving operation mode in each traffic element is different. For example, when trying to learn a driving operation mode in units of traffic elements, it may be learned as a driving operation mode of a different driver even though the driving operation mode is originally the same driver. . That is, in the example shown in FIG. 2, each driving operation mode performed until the vehicle 100 that departed from the departure point Ps arrives at the destination Pg is a driving operation mode by a plurality of drivers such as drivers Da and Db. Will be learned as.

  Therefore, in the present embodiment, in the example shown in FIG. 2, the feature quantity of the driving operation mode at the intersections SC1 and SC2 is that of the driver Da, and the feature quantity of the driving operation mode at the intersection SC3 is that of the driver Db. When it is classified as being, each driving operation mode in the trip is unified and managed as belonging to the driver Da having a higher classification ratio.

  Further, as exemplified in FIG. 3 for each driver, when the drivers Da to Dd pass, for example, a certain intersection, the drivers Da to Dd are accelerator off, tar lamp on, There is a tendency to brake on. Also, the amount of depression of the brake pedal tends to be different for each driver. That is, in this embodiment, the operation timing and the operation amount of the driving operation element illustrated in FIG. 3 indicate the characteristic amount of the driving operation element.

  And the learning part 132a of this Embodiment classifies which driver's operation mode belongs to every time the driving operation element of the vehicle 100 is operated based on such a feature amount.

  In addition, as shown in FIG. 4, the classification result of the driving operation mode in a certain trip, for example, when the driver Da drives the vehicle 100 in the trip La, each driving operation mode in the trip La It is classified based on the feature amount. In this example, if driving operation data indicating, for example, a total of “15 times” of driving operation modes is acquired in the trip La, the driving operation modes are the driver Da “11 times” and the driver Db “1”, respectively. Times ”, driver Dc“ 3 times ”, and driver Dd“ 0 times ”.

  Therefore, in this embodiment, the driver Da having the highest number of classifications is the driver who performed the driving operation on the trip La, and the unified management is performed assuming that all the driving operation modes classified into the drivers Da to Dc belong to the driver Da. Is done.

  Similarly, if the travel time of the vehicle 100 on the trip La is, for example, “36 minutes”, the travel time when each driving operation mode is performed is the driver Da “30 minutes” and the driver Db “1 minute”, respectively. , Driver Dc “5 minutes” and driver Dd “0 minutes”. Therefore, at this time, among the driving operation modes in each trip, the driver Da that has traveled for the longest time is specified as the driver who performed the driving operation mode.

  Hereinafter, the operation of the vehicle driver specifying learning device and the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method of the present embodiment will be described with reference to FIGS. 5 and 6.

  As shown in FIG. 5, when learning the driving operation mode, for example, when it is detected that the vehicle 100 has started running due to the ignition key being turned on (step S100), the trip count is started. (Step S101).

  Next, the latitude / longitude data acquired by the GPS 120 and the driving operation data indicating the driving operation mode detected by the driving operation detecting means 110 are associated (step S102). Then, each driving operation mode indicated by the driving operation data is classified as to which driver it belongs based on the feature amount (steps S103 and S104).

  After that, for example, when it is determined that the vehicle 100 has stopped and arrived at the destination due to the ignition key of the vehicle 100 being turned off (step S105: YES), the count of the trip is ended (step S106). ).

  Next, when the classification result in step S104 and the driving operation mode are classified into those of two or more drivers (step S107: YES), for example, the driver with the largest number of classifications is selected in the classification result, The selected driver is identified as the driver who has driven the vehicle 100 in the counted trip (steps S108 and S109).

  And all the driving operation modes in this trip are unified as the driving operation modes of the specified driver. Next, the driving operation mode indicating the unified driving operation mode and the latitude / longitude data are registered in the database 140 as belonging to the specified driver (step S110).

  Further, when the driver is specified, as shown in FIG. 6, when the vehicle 100 operated by the driver to be specified reaches a certain intersection that passes regularly (step S200), each point at the intersection is determined. It is determined whether or not the database of the operation modes of the driving operation elements has been completed (step S201). Note that, for example, when a predetermined number or more of data indicating operation modes of driving operation elements by the respective drivers driving the vehicle 100 are recorded in the database 140, it is determined that the database has been completed.

  When the creation of the database is not completed (step S201: NO), an operation of turning off the accelerator pedal by the driver of the vehicle 100, an operation of turning on the turn lamp, and an operation of turning on the brake pedal are detected. (Step S202).

  Then, this detection result is recorded in the database 140 in association with latitude / longitude data indicating each detection position (step S203). In the example here, for example, data indicating the distance before the intersection is associated as latitude and longitude data.

  Next, in step S204, it is determined whether or not data on a point common to or similar to a point where the accelerator pedal, turn lamp, and brake pedal elements are operated exists in the database 140. And when the data of a common or similar point does not exist in the database 140 (step S204: NO), the driving history in the same point of the driver who is driving the vehicle 100 at that time is not registered, and the same at the intersection is the same. The driving history of the driver is newly registered in the database 140.

  On the other hand, when data of common or similar points exists in the database 140 (step S204: YES), data indicating the operation mode of the driving operation element detected in step S102 is newly stored in the database 140 as the driving history of the driver. As a result, the driving history of the driver is updated (step S206).

  Further, when it is determined in step S201 that the database of the operation modes of the driving operation elements at the intersection has been completed (step S201: YES), first, a change from on to off of the accelerator pedal is detected. (Step S207). Then, when the point where the accelerator pedal is turned off is the same as or similar to the point recorded in the database 140, the driver operating the vehicle 100 is identified as the driver recorded in the database 140 (step). S208, S209).

  On the other hand, when the driver cannot be specified based on the operation mode of the accelerator pedal (step S209: NO), for example, whether or not data indicating that the point where the turn lamp is turned on is common or similar exists in the database 140 Based on this, the driver is specified (steps S210 and S211).

  When the driver cannot be identified based on the operation mode of the turn lamp (step S212: NO), for example, whether or not data indicating that the point where the brake pedal is turned on is common or similar exists in the database 140 Based on this, the driver is specified (steps S213 to S215).

  When the driver cannot be specified by the operation mode of any of the accelerator pedal, the turn lamp, and the brake pedal at the intersection (step S215: NO), an intersection or curve existing in a different area is used. The driver is identified based on the operation mode of the driving operation element performed in (Step S216).

  When the driver is identified in this way, for example, the seat position or the position of the door mirror is based on the data indicating the seat position or the position of the door mirror associated with the data indicating the operation mode of the driving operation element registered in the database 140. Adjustments are made. Further, for example, driving assistance is performed according to the identified driving pattern specific to the driver (step S217). When the driver cannot be specified, for example, driving assistance for a standard driver is performed.

  As described above, according to the vehicle driver specifying learning device and the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method according to the present embodiment, the following effects can be obtained.

  (1) The driving operation mode of the driver was learned while being classified according to the driver based on the feature amount. When the driving operation mode with trip as a unit is classified into two or more drivers through the above learning, the learning result of the driving operation mode with trip as a unit is classified into the higher classification ratio in the trip. It was decided to be managed as a driving operation mode of any one driver. This makes it possible to specify a driver on a trip basis, and each driving operation mode performed by one driver in a certain trip is not subdivided as being due to a plurality of drivers. In addition, in this way, by specifying the driving operation mode in the trip as that of the driver whose classification ratio is higher in the trip, it is possible to specify which driver is the driving operation mode with high accuracy. It becomes.

  (2) The driving operation mode of the driver with the higher classification ratio in the trip is the driving operation mode with the highest number of times classified in the trip, and the driving with the longest travel time of the vehicle in the trip. A driving operation mode that satisfies at least one of the operation modes was selected. This makes it possible to unify the driving operation mode based on the classification ratio with higher accuracy based on the number of classifications or the travel time of the vehicle 100.

  (3) When the learning result is registered in the database 140, when the driving operation mode in the trip is classified into a plurality of drivers, the learning result is unified as the learning result based on the driving operation of one driver. The unified learning result is registered in the database 140. For this reason, the number of learning results is limited to the necessary minimum number before the learning results are registered in the database 140. As a result, learning results unified in advance are registered as appropriate, and only the minimum necessary learning result can be registered in the database 140.

  (4) As the above-described driving operation mode, an operation mode of a plurality of types of driving elements performed under one or more traffic elements through which the vehicle 100 passes during a trip is selected. And it decided to classify a driving operation mode according to a driver for every 1 thru / or several traffic elements. Thereby, while learning the driving operation mode by the driver of the vehicle 100 that has traveled on the road where various traffic elements exist, it is possible to suitably improve the learning accuracy.

  (5) As the driving operation mode, operation modes of at least two elements of the accelerator pedal, the turn lamp, and the brake pedal were selected. Therefore, while learning the operation modes of elements that are easy to use in normal driving operations such as accelerator pedals, turn lamps, and brake pedals performed under a series of points, and that tend to reflect driver-specific wrinkles, It becomes possible to improve learning accuracy suitably. As a result, it is easy to reflect driver-specific wrinkles as learning targets, so it is possible to use elements that have a high frequency of operation by the driver, in other words, elements that have many opportunities to learn. become.

  (6) The driving operation mode for specifying the driver is learned, and the driver to be specified is specified through verification of the learned result and the driving operation mode of the driver obtained each time. For this reason, it becomes possible to unify and manage the driving operation modes necessary for specifying the driver in units of trips, and to minimize the number of driving operation modes that need to be managed to identify the driver based on the driving operation modes. It becomes possible to limit.

(Second Embodiment)
Next, the second embodiment of the vehicle driver specifying learning device and the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method according to the present invention is different from the first embodiment. Will be described with reference to FIGS. The basic configuration of the vehicle driver specifying learning device and the vehicle driver specifying learning method, the vehicle driver specifying device, and the vehicle driver specifying method according to the present embodiment is the same as that of the first embodiment. Also in the present embodiment, elements substantially the same as those in the first embodiment are denoted by the same reference numerals, and redundant descriptions are omitted.

  As shown in FIG. 7, the vehicle 100 of the present embodiment is provided with a travel environment detection unit 160 that detects the travel environment of the vehicle 100. The traveling environment detection unit 160 uses, for example, a millimeter wave radar 161 that detects the presence of a forward vehicle that travels ahead in the traveling direction of the vehicle 100 using a millimeter wave band radio wave, and an in-vehicle camera 162 that captures the situation around the vehicle 100. It is configured with. In addition, the traveling environment detection unit 160 includes, for example, an in-vehicle communication device 163 that performs vehicle-to-vehicle communication with a vehicle traveling around the vehicle 100 and road-to-vehicle communication with a roadside communication device provided on the road. Yes.

  When the millimeter wave radar 161 detects the presence of a forward vehicle traveling in front of the traveling direction of the vehicle 100, the millimeter wave radar 161 outputs a signal indicating the detection result to the in-vehicle control device 130. The in-vehicle camera 162 appropriately outputs image data indicating the periphery of the captured vehicle 100 to the in-vehicle control device 130. For example, when communication with a forward vehicle traveling forward in the traveling direction of the vehicle 100 is established through inter-vehicle communication, the in-vehicle communication device 163 outputs the communication result to the in-vehicle control device 130. Further, the in-vehicle communication device 163 transmits various information such as information indicating the presence of other vehicles traveling around the vehicle 100, particularly the presence of a preceding vehicle or a traffic element existing on the road, and traffic jam information through road-to-vehicle communication. get. Then, the in-vehicle communication device 163 outputs the acquired various information to the in-vehicle control device 130.

  As described above, the vehicle-mounted control device 130 according to the present embodiment appropriately receives data indicating operation modes of various driving operation elements detected by the driving operation detection unit 110 and latitude / longitude data indicating the latitude / longitude of the vehicle 100. In addition, information indicating the surrounding environment of the vehicle 100 detected by the traveling environment detection unit 160 is input as appropriate.

  In-vehicle control device 130 according to the present embodiment identifies a forward vehicle that travels ahead in the traveling direction of vehicle 100 from the traveling environment of vehicle 100 based on each signal and image data input from traveling environment detection unit 160. A forward vehicle identification unit 134 is provided. When the forward vehicle identification unit 134 identifies whether there is a vehicle ahead of the vehicle 100 based on each signal or image data input from the traveling environment detection unit 160, the identification result is output to the database processing unit 132. To do.

The database processing unit 132 associates the identification result input from the preceding vehicle identification unit 134 with the driving operation data indicating the driving operation mode acquired by the driving operation detection unit 110.
Here, when the vehicle ahead is present and when it is not present, as illustrated in FIG. 8, the feature quantities of the operation modes of the driving operation elements such as the accelerator pedal, the turn lamp, and the brake pedal change. The changing feature amount is also unique to the driver, and the driver can be specified based on the feature amount.

  Therefore, as shown in FIG. 7, when the identification result is input from the forward vehicle identification unit 134, the learning unit 132a of the present embodiment learns the driving operation mode according to the presence / absence of the forward vehicle, and sets the feature amount. Based on this, the driving operation mode is classified for each driver. Thereby, in the aspect illustrated in FIG. 8, the driving operation modes detected by the driving operation detection unit 110 are classified according to the presence or absence of the preceding vehicle.

  Next, the operation of the vehicle driver specifying learning device and the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method of the present embodiment will be described with reference to FIGS. This will be described with reference to FIG.

  As shown in FIG. 9, in the present embodiment, when the trip count is started in step S <b> 101, whether there is a forward vehicle ahead of the traveling direction of the vehicle 100 based on the detection result of the traveling environment detection unit 160. Whether or not is identified (step S101A). Next, based on the identification result, the driving operation data indicating the driving operation mode detected by the driving operation detecting unit 110 and the latitude / longitude data acquired by the GPS 120 are associated with each other according to the presence / absence of the preceding vehicle (step). S102A).

  Thus, by appropriately registering each data in the database 140, data indicating a driving operation mode in the mode illustrated in FIG.

  When the driver is specified, as shown in FIG. 10, in step S300, latitude / longitude data having the same latitude and longitude as the point where the vehicle 100 driven by the driver is traveling is registered in the database 140. Is determined. When the latitude / longitude data having the same latitude / longitude as the point where the vehicle 100 is traveling is already registered in the database 140, the operation mode of each driving operation element by the driver is detected by the driving operation detecting means 110. . Further, even when the same latitude / longitude data does not exist in the database 140, the driving operation data acquired at a point where at least one of the point where the vehicle 100 travels and the road alignment, road gradient, and intersection shape are similar. When the latitude / longitude data exists in the database 140 (step S307: YES), the driving operation detection means 110 detects the operation mode of each driving operation element by the driver.

  Next, it is identified whether or not the forward vehicle 200 exists ahead of the traveling direction of the vehicle 100 (step S302). Then, when the preceding vehicle 200 exists, one or more of the data registered in the database 140 is acquired at a point that is the same as or similar to that when the vehicle 100 is traveling and is acquired when the preceding vehicle 200 exists. Driving operation data is extracted from the database 140 (step S303). Then, the feature amount of each driving operation element indicated by the extracted driving operation data is collated with the feature amount of the operation mode of each driving operation element by the driver detected in step S301 (step S304).

  Further, when the forward vehicle 200 does not exist ahead of the traveling direction of the vehicle 100 operated by the driver (step S302: NO), the data registered in the database 140 is a point that is common or similar to when the vehicle 100 is traveling. One or more driving operation data acquired when the vehicle 200 ahead is not present is extracted from the database 140 (step S308). Then, the feature amount of each driving operation element indicated by the extracted driving operation data is collated with the feature amount of the operation mode of each driving operation element by the driver detected in step S301 (step S304).

  Next, as a result of such collation, when data having a common or similar feature amount exists in the database 140 (step S305: YES), the driver operating the vehicle 100 is registered as the driver Da, for example, in the data. (Step S306).

  When the driver is specified in this manner, for example, the position adjustment of the seat position and the door mirror is performed based on the data indicating the seat position and the position of the door mirror registered in the database 140 as data specific to the driver Da. In addition, for example, driving assistance corresponding to the identified driving pattern specific to the driver is performed.

  On the other hand, as a result of the feature amount collation, when the data having the same or similar feature amount does not exist in the database 140 (step S305: NO), this process ends. Then, when the point where the vehicle 100 travels changes with the movement of the vehicle 100, the feature amount of the operation mode of each driving operation element performed at this point and the feature amount of the driving operation data registered in the database 140 And are matched. Thus, this process is executed until the driver is specified, and driving support for, for example, a standard driver is performed until the driver is specified. Similarly, when the driver cannot be specified in step S306, for example, driving assistance for a standard driver is performed.

  As described above, according to the vehicle driver specifying learning device and the vehicle driver specifying learning method, and the vehicle driver specifying device and the vehicle driver specifying method according to the present embodiment, the effects (1) to (6) described above. As well as the following effects.

  (7) The above-described driving operation mode is learned according to whether or not there is a vehicle ahead. Further, when specifying the driver, the driver is specified in consideration of the presence of a preceding vehicle ahead of the traveling direction of the vehicle 100 operated by the driver to be specified. For this reason, even if the driving operation mode changes according to the presence of the preceding vehicle, the database 140 can be constructed in consideration of the change. This makes it possible to learn the driving operation mode with higher accuracy. In addition, this makes it possible to specify the driver with high accuracy regardless of the presence of the preceding vehicle.

(Other embodiments)
In addition, each said embodiment can also be implemented with the following forms.
In the second embodiment, the traveling environment detection unit 160 is configured by the millimeter wave radar 161, the in-vehicle camera 162, and the in-vehicle communication device 163. The traveling environment detection unit 160 may be configured by at least one of the millimeter wave radar 161, the in-vehicle camera 162, and the in-vehicle communication device 163. In addition, any device capable of detecting the presence of a forward vehicle or a person existing around the vehicle 100 can be employed as the driving operation detection means.

  In each of the above embodiments, when the driver cannot be identified using the driving operation data registered in the database 140, driving assistance for a standard driver is performed. Not limited to this, driving assistance may not be performed when a driver cannot be identified.

  In each of the above embodiments, the period from when the ignition key of the vehicle 100 is turned on to when it is turned off is determined as one trip. For example, the time from when the accessory position is turned on to when it is turned off may be determined as one trip. Further, trips may be counted based on opening / closing of the door and ON / OFF of the door lock. In addition to this, any element that can be used for counting the trip may be any element that can detect the operation of the vehicle 100 from the start to the end.

  In each of the above embodiments, the driver is specified, and driving support corresponding to the driving operation pattern of the specified driver is executed. Not limited to this, for example, when an application or content data that can be used in an in-vehicle information terminal or the like mounted on the vehicle 100 is a charging system, a driver who has been charged, that is, a driver having a valid usage right is operated by the driver. You may make it specify based on the feature-value of operation. In this case, when the driver charges the application or content data while using the vehicle 100, the data indicating that the driver has charged the application or content data is the data indicating the feature amount of the driving operation of the driver. It is associated and registered in the database 140. When the driver is specified based on the feature amount of the driving operation of the driver, it is also specified whether or not the driver has the authority to use the application and content data. In addition, for example, when it is not possible to specify a driver based on data registered in the database 140 that has been converted into a database, the driver that cannot be specified may be specified as an unauthorized user of the vehicle 100. Good. Then, a regular driver may be notified that the vehicle 100 has been illegally used, or a warning may be urged to a driver identified as an unauthorized user. In addition to this, for example, data relating to a driver's home base or a frequently visited location of the driver is associated with data indicating the characteristic amount of the driving operation of the driver, and the data is registered in the database 140. Also good. In this case, when a driver is specified based on the feature amount of the driving operation, a base or a frequently visited point associated with data indicating the feature amount of the specified driver is specified. Then, for example, the specified base or visited point is presented as a destination candidate for route guidance. For this reason, in the process of driving the vehicle 100 by the driver, a base of the driver and a frequently visited site are automatically identified, and the identified base and visited site can be presented for each driver. . As a result, even when a plurality of drivers with different bases or visit points use the vehicle 100, the bases or visit points are presented for each driver, or route guidance to these bases or visit points is performed. Is possible.

  In each of the above embodiments, at least two elements of an accelerator pedal, a turn lamp, and a brake pedal are selected as driving operation elements. Not limited to this, the driving operation element may be one element of an accelerator pedal, a turn lamp, and a brake pedal. In addition, the driving operation element used for specifying the driver may be an element operated according to the characteristics of each point on the road. For example, at least one of a steering, an engine brake, a manual transmission, etc. It may be.

  In each of the above embodiments, the operation mode of a plurality of types of driving operation elements performed under a specific traffic element is selected as the feature amount of the driving operation. Not limited to this, as a feature amount of the driving operation, for example, an operation mode of a kind of driving operation element such as an operation mode of only an accelerator pedal performed under a specific traffic element may be selected. Good. The point used to identify the driver may be a point where the feature amount of the driving operation such as the operation amount of the driving operation element and the operation timing changes according to the characteristics of the same point. It may be a point where no traffic element exists.

  -In each said embodiment, the continuous point where a series of driving operation by multiple types of driving operation elements is performed was selected as said point. Not limited to this, the point may be one of continuous points where a series of driving operations by a plurality of types of driving operation elements are performed. In this case, for example, in the example illustrated in FIG. 2, among the continuous points P1 to P3 before the intersection SC1, only the feature amount of the driving operation mode performed at the point P1 or the point P2 was performed. The driver is specified based only on the feature amount of the driving operation mode. In addition, the driving operation data used for specifying the driver may be data acquired under a situation where the driving environment is common or similar among the driving operation data acquired through the driving operation by the driver.

  In each of the above embodiments, the presence or absence of the forward vehicle 200 is identified as the traveling environment of the vehicle 100, and the driving operation data according to the identification result is used when specifying the driver. However, the present invention is not limited to this, and for example, the presence or absence of traffic congestion may be identified as the driving environment of the vehicle 100, and the driving operation data indicating the feature amount of the driving operation performed by the driver may be registered in the database 140. When the driver is specified, it is identified whether the area in which the vehicle 100 driven by the driver is traveling is congested, and according to the identification result, the driving operation data acquired at the time of congested traffic and the data acquired at the time of non-congested traffic You may make it use properly the performed driving operation data. According to this, even if the driver's driving operation is different between when there is a traffic jam and when there is no traffic jam, and there is a tendency that the driver-specific traps are reflected in each, the driver can be identified by taking into account the traffic jam situation. It becomes possible. Further, for example, as the driving environment of the vehicle 100, for example, the presence or absence of a person or an obstacle existing near the road is identified, and the driving operation data indicating the characteristic amount of the driving operation by the driver according to the presence or absence of the person or the obstacle is stored in the database 140. You may make it register. Then, when specifying the driver, it is identified whether there is a person or an obstacle near the road on which the vehicle 100 driven by the driver travels, and when there is a person or an obstacle according to the identification result. The acquired driving operation data and the driving operation data acquired when no person or obstacle exists may be used properly. According to this, even if there is a tendency for the driver's driving operation to change due to the presence of a person or an obstacle and the driver's unique habits to be reflected in this changing driving operation, the presence of a person or obstacle It becomes possible to specify the driver in consideration of. In addition, the driving environment identified when the driver is specified and the driving operation is registered may be any environment as long as the driving operation of the driver can be changed, and can be changed as appropriate.

  In each of the above embodiments, when the learning operation is registered in the database 140, when the driving operation mode in the trip is classified into a plurality of drivers, the learning result is a learning result based on the driving operation of one driver. Then, the unified learning result is registered in the database 140. Not limited to this, when the driving operation mode in the trip already registered in the database 140 is classified into a plurality of drivers, the learning result registered in the database 140 is the learning result based on the driving operation of one driver. It may be unified as there is. With this configuration or method, even if learning results that are not unified are once registered in the database 140, the learning results can be unified afterwards. Thereby, it is possible to unify the learning results already registered in the database 140, and it is possible to accurately manage the learning results regardless of the timing of registration of the learning results.

  In each of the above embodiments, as the driving operation mode of the driver having the higher classification ratio in the trip, the number of times classified in the trip is the upper driving operation mode, and the travel time of the vehicle in the trip A driving operation mode that satisfies at least one condition of the driving operation mode having a higher length is adopted. Not limited to this, as the driving operation mode of the driver with the higher classification ratio in trip, the number of times classified in the trip is only the upper driving operation mode, or the length of travel time of the vehicle in the trip However, only the higher-level driving operation mode may be adopted. In addition, the driving operation mode in which the number of classifications and the traveling time are both classified as higher may be unified as the driving operation mode. In addition, the classification ratio only needs to unify the upper driving operation modes, and the conditions used for the unification can be appropriately changed.

  In each of the above embodiments, the driving operation mode in which the classification ratio is the highest is adopted as the driving operation mode of the driver having the higher classification ratio in the trip. Not limited to this, the driving operation mode with the classification ratio of the second and later is selected as the driving operation mode of the driver with the higher classification ratio in the trip, and the driver who has performed the selected driving operation mode is selected in the trip You may identify as a driver who performed driving operation.

  -In each said embodiment, it decided to unify the driving | running operation mode on the basis of one trip. Not limited to this, the driving operation mode may be unified in units of two or more trips. In particular, when one trip is completed at a place other than the home where the vehicle 100 is based, the probability that the driver has not changed is high. Therefore, until one trip is completed at home or the like, the driving operation mode can be handled in a larger unit by unifying the driving operation mode in units of a plurality of trips.

  In each of the above-described embodiments, the database processing unit 132, the driver identification processing unit 133, and the database 140 are mounted on the vehicle 100. However, the present invention is not limited to this, and the database processing unit 132, the driver identification processing unit 133, and the database 140 may be provided in a management center that collects data indicating the feature amount of the driving operation of the driver. In this configuration, data indicating the feature amount of the driving operation by the driver of the vehicle 100 and the point where the driving operation has been performed is transmitted from the vehicle 100 to the management center through communication between the management center and the vehicle 100. . Then, in the management center, the driver of the vehicle 100 is specified based on the data transmitted from the vehicle 100 and the data registered in the database 140. Thereby, the structure of the vehicle-mounted control apparatus 130 can be simplified. As a result, the management center can collectively manage data necessary for driver identification, and the management center can collectively perform processing necessary for driver identification.

  DESCRIPTION OF SYMBOLS 100 ... Vehicle, 110 ... Driving operation detection means, 111 ... Accelerator sensor, 112 ... Brake sensor, 113 ... Winker switch, 114 ... Steering angle sensor, 120 ... GPS, 130 ... In-vehicle control apparatus, 131 ... System control part, 132 ... Database processing unit, 132a ... learning unit, 132b ... management unit, 133 ... driver identification processing unit, 134 ... front vehicle identification unit, 140 ... database, 150 ... driving support unit, 160 ... driving environment detection unit, 161 ... millimeter wave Radar, 162 ... vehicle-mounted camera, 163 ... vehicle-mounted communication device, 200 ... vehicle ahead.

Claims (12)

A vehicle driver identification learning device that performs learning for driver identification based on a driving operation mode of a driver who operates the vehicle,
A learning unit that learns the driver's driving operation mode while classifying the driver according to the feature amount including at least one of the operation timing and the operation amount of the driving operation element ,
When the driving operation mode in units of trips, which is a period from the start of the driving operation of the vehicle to the end of the driving operation, is classified into two or more drivers through learning by the learning unit , A management unit that unifies and manages the learning result of the driving operation mode of any one driver who has the highest classification ratio in the trip as a driving operation mode in units of the trip. A vehicle driver identifying learning device. The driving operation mode of the driver having the higher classification ratio in the trip is the driving operation mode of the driver having the higher number of times classified in the trip, and the driving of the vehicle classified in the trip The vehicle driver specifying learning device according to claim 1, wherein the driving operation mode satisfies at least one of the driving operation modes of a driver having a higher time length. In the vehicle driver specifying learning device according to claim 1 or 2,
A database in which learning results from the learning unit are registered;
The management unit
a. In registering the learning result in the database, when the driving operation mode in the trip is classified into a plurality of drivers, the learning result is unified as the learning result based on the driving operation of one driver, A process of registering the unified learning result in the database; and b. When the driving operation mode in the trip registered in the database is classified into a plurality of drivers, the learning result registered in the database is unified as a learning result based on the driving operation of one driver. processing,
A vehicle driver specifying learning device characterized by executing any one of the processes. The driving operation mode is an operation mode of a plurality of types of driving elements performed under one or more traffic elements through which a vehicle passes during a trip,
The vehicular driver specifying learning device according to any one of claims 1 to 3, wherein the learning unit classifies the driving operation mode according to a driver for each of the one or more traffic elements. The vehicle driver specifying learning device according to claim 4, wherein the driving operation mode includes operation modes of at least two elements of an accelerator pedal, a turn lamp, and a brake pedal. A vehicle driver identification device that identifies a driver based on a driving operation mode of a driver who operates the vehicle,
A driving operation mode for specifying a driver is learned using the vehicle driver specifying learning device according to any one of claims 1 to 5, the learning result and a driving operation mode of the driver obtained each time, A vehicle driver identification device comprising: a driver identification processing unit that performs a process of identifying a driver to be identified through collation. A vehicle driver identifying learning method for performing learning for driver identification based on a driving operation mode of a driver who operates a vehicle,
The driving operation mode of the driver is at least the operation timing and operation amount of the driving operation element.
Learning while classifying by driver based on features including
When the driving operation mode with the trip as a unit from the start of the driving operation of the vehicle to the end of the driving operation is classified into two or more drivers through the learning step , the trip And the step of unifying and managing the learning result of the driving operation mode of any one of the drivers with the highest classification ratio in the vehicle as the driving operation mode with the trip as a unit. Learning method for vehicle driver identification. As the driving operation mode of the driver having a higher classification ratio in the trip, the number of times of classification in the trip is the driving operation mode of the upper driver , and the driving time of the vehicle classified in the trip The vehicle driver specifying learning method according to claim 7, wherein a driving operation mode that satisfies at least one of the driving operation modes of a driver having a higher length is selected. In the vehicle driver specifying learning method according to claim 7 or 8,
As the learning step, the learning result is registered in a database,
As the managing step,
a. In registering the learning result in the database, when the driving operation mode in the trip is classified into a plurality of drivers, the learning result is unified as the learning result based on the driving operation of one driver, Registering unified learning results in the database; and b. When the driving operation mode during the trip registered in the database is classified into a plurality of drivers, the learning result registered in the database is unified as a learning result based on the driving operation of one driver,
A vehicle driver specifying learning method characterized by executing any one of the steps. As the driving operation mode, selecting an operation mode of a plurality of types of driving elements performed under one or more traffic elements through which the vehicle passes during a trip,
The vehicle driver specifying learning method according to any one of claims 7 to 9, wherein, in the learning step, the driving operation mode is classified by driver for each of the one or more traffic elements. The vehicle driver specifying learning method according to claim 10, wherein operation modes of at least two elements of an accelerator pedal, a turn lamp, and a brake pedal are selected as the driving operation mode. A vehicle driver identification method for identifying a driver based on a driving operation mode of a driver who operates the vehicle,
A driving operation mode for specifying a driver is learned using the vehicle driver specifying learning method according to any one of claims 7 to 11, and the learned result and the driving operation mode of the driver obtained each time The vehicle driver identification method characterized by including the step which identifies the driver made into a specific object through collation of.

Images