JP5152063B2 - Driving behavior estimation device - Google Patents

Description

  The present invention relates to a driving behavior estimation device that estimates a driving behavior of a driver.

  In a driving support system that estimates a driver's driving behavior and provides assistance according to the driving behavior, if the driver's driving behavior is not adequately supported, the assistance itself becomes a burden on the driver. It is also possible that For this reason, it is important to accurately estimate the driving behavior of the driver and provide support according to the driving behavior. Therefore, a driving behavior estimation device that estimates the driving behavior of the driver has been proposed.

  For example, in recognition of driving behavior using a hidden Markov model, parameters are estimated and recognized in units of driving behavior sub-patterns that make up driving behavior patterns, and temporally between driving behavior sub-patterns that make up driving behavior patterns There is known a driving behavior estimation device configured to recognize a driving behavior pattern by providing a driving behavior pattern template that describes a context and comparing a recognition result with a sub-pattern sequence and a sub-pattern template ( For example, see Patent Document 1). In the driving behavior estimation apparatus configured as described above, “recognition start → deceleration → right steering → acceleration → recognition end” in the case of right turn recognition, “recognition start → deceleration → vehicle stop” in the case of stop recognition. → "Recognition end", which means that driving behavior can be estimated when driving behavior is completed or just before it is completed, and support (for example, warning or guidance) that matches the driver's situation is possible. It's hard to say.

  Therefore, specializing in the estimation of driving behavior that turns at a target point (for example, an intersection), catching the deceleration for turning at the target point, and estimating the driving behavior at an early time when there is time to reach the target point A driving behavior estimation device to be completed has been proposed (see, for example, Patent Document 2). Specifically, when the driver bends at the target point, the accelerator amount, brake amount, and speed data until the target point is reached are classified by the clustering method, and the averaged data is bent. A driving action template is recognized, and a driving action pattern is recognized by comparing this template with actual driving data. In the driving behavior estimation device configured in this way, whether or not to perform route guidance guidance at an early stage before reaching the target point, depending on whether or not the driver is going to turn at the target point. It is possible to make decisions and to provide assistance according to the driver's situation.

Japanese Patent Laid-Open No. 11-99846 JP 2006-133216 A

  However, in the driving behavior estimation device described in Patent Document 2, the template and the actual driving behavior are collated by paying attention only to time-series changes in the driving data such as the accelerator amount, the brake amount, and the speed. If there are portions having similar time series changes in different driving behaviors, there is a problem that it is difficult to estimate the driving behavior of the driver with high accuracy. For example, in a driving action that turns and a driving action that stops, the driver performs a driving action that decelerates the vehicle by stepping on the brake before reaching the target point. For this reason, as shown in FIG. 10, in the driving action template TP11 that turns and the driving action template TP12 that stops, there are time areas in which the braking amount increases and the accelerator amount and the speed decrease in different time regions ( (See sub-pattern SP11 and sub-pattern SP12 in FIG. 10). As a result, the driving behavior estimation device may estimate that the driving behavior is stopped even though the driving behavior is turning.

  The present invention has been made in view of these problems, and an object thereof is to provide a technique for improving the estimation accuracy of driving behavior.

  In the driving behavior estimation apparatus according to claim 1, which is made to achieve the above object, the current position information acquisition unit acquires current position information indicating the current position of the host vehicle, and the target point position information acquisition unit includes The target point position information indicating the position of the target point, which is a target for estimating the driving behavior of the host vehicle, is acquired.

  The driving data acquisition means acquires driving data including speed information of the own vehicle and operation information related to the vehicle operation of the driver that affects the traveling state of the own vehicle. Further, the remaining time calculating means calculates the current position information acquired by the current position information acquiring means, the target point position information acquired by the target point position information acquiring means, and a speed calculation for calculating the speed of the host vehicle. Based on the information, a point arrival remaining time which is a time until the host vehicle reaches the target point is calculated.

  In addition, the template storage means is a typical example of the driver from a point before the point arrival remaining time to reach the target point to a preset fixed period before the point arrival remaining time for each of a plurality of different point arrival remaining times. A template for each driving action indicating the driving action is stored.

  Then, the template acquisition means acquires a template corresponding to the point arrival remaining time from the template storage means based on the point arrival remaining time calculated by the remaining time calculation means, and the driving action estimation means serves as the driving data acquisition means. The driving behavior of the driver at the target point is estimated by collating the history of driving data acquired in this way with the template acquired by the template acquisition means.

  In the driving behavior estimation device according to claim 1 configured as described above, the typical driving behavior of the driver from before the point arrival remaining time to a preset fixed period before the point arrival remaining time is obtained. A template for each driving action shown is provided for each of a plurality of different point arrival remaining times, and the driving action is estimated using a template corresponding to the point arrival remaining time and a history of driving data.

  Even in a plurality of different driving actions, even if there is a similar part in the time change of the vehicle speed and the driver's vehicle operation before reaching the target point, the time region where this similar part occurs is the driving Different for each action. For example, when comparing a driving action that turns and a driving action that stops, there is a time region in which both the brake amount increases and the accelerator amount and speed decrease before the intersection. However, the driving action that stops is faster than the driving action that turns, as the brake amount increases and the accelerator amount and speed decrease (see FIG. 10).

  For this reason, when comparing different driving behavior templates corresponding to the same point arrival remaining time, a long time until the target point is reached after approaching the target point as in the driving behavior estimation device described in Patent Document 2. There are fewer similar parts than when comparing different driving behavior templates for regions. That is, since the driving behavior can be estimated using a template that clearly shows the difference between different driving behaviors, the estimation accuracy of the driving behavior can be improved.

  Further, in the driving behavior estimation device according to claim 1, as described in claim 2, at least one of the speed and acceleration of the host vehicle is used as the speed calculation information used by the remaining time calculation means to calculate the point arrival remaining time. One should be used.

  In addition, in the driving behavior estimation device according to claim 1 or 2, since it is a driving behavior that stops and a driving behavior that has a similar part in the time change of the vehicle speed and the driver's vehicle operation. As described in claim 3, the driving behavior may be a left / right turn behavior and a stop behavior.

  In addition, since it is an intersection that needs to be distinguished from the driving behavior that bends and the driving behavior that has a similar portion due to the time change of the vehicle speed and the driver's vehicle operation, it is the intersection, so that the claims 1 to claim In the driving behavior estimation apparatus according to any one of claims 3 to 4, the target point may be the next intersection on the travel route of the host vehicle.

  Further, in the driving behavior estimation apparatus according to any one of claims 1 to 4, specifically, the traveling state of the host vehicle may be acceleration / deceleration of the vehicle. In this case, as the operation information, for example, information relating to an accelerator pedal and a brake pedal may be used as described in claim 5. However, the traveling state of the host vehicle is not limited to the acceleration / deceleration of the vehicle. For example, the direction of the vehicle may be considered, and as operation information in that case, for example, information related to steering (such as a steering operation amount) may be used. .

1 is a block diagram illustrating a configuration of a navigation device 10 and a schematic configuration of an in-vehicle LAN to which the navigation device 10 is connected. It is a flowchart which shows a template preparation procedure. It is explanatory drawing which shows the structure of a template. It is a flowchart which shows an intersection guidance process. It is a flowchart which shows a driving action estimation process. It is a flowchart which shows a voice guidance execution process. It is a figure explaining selection of a stop action template according to remaining time (tau). It is a figure explaining selection of the left-right turn action template according to remaining time (tau). It is a graph which shows the execution conditions of voice guidance. It is a figure explaining the template used by patent document 2. FIG.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of the navigation device 10 of the present embodiment and a schematic configuration of an in-vehicle LAN 25 to which the navigation device 10 is connected.

  As shown in FIG. 1, the navigation device 10 is mounted on a vehicle and connected to various ECUs and in-vehicle devices including an engine ECU 31, a brake ECU 32, and a steering ECU 33 via an in-vehicle LAN 25.

  Of these, the engine ECU 31 is configured to control the rotation of the engine based on at least a detection signal from an accelerator opening sensor that detects an accelerator opening corresponding to the amount of depression of the accelerator pedal of the driver. The brake ECU 32 is based on at least a detection signal from a master cylinder pressure sensor that detects a brake operation amount from a hydraulic pressure of a master cylinder that pumps brake oil in response to a driver's brake pedal operation and a vehicle speed sensor that detects a vehicle speed. Thus, ABS control, traction control, and the like are executed. Further, the steering ECU 33 executes power steering control for generating an assist force at the time of changing the steering angle of the steered wheel based on at least a detection signal from a steering angle sensor that detects the steering angle of the front wheel at the time of the steering operation of the driver. It is configured as follows.

  Various vehicle information (accelerator opening, brake operation amount, vehicle speed, etc.) detected by these ECUs 31 to 33 and the like can be arbitrarily transmitted and received via the in-vehicle LAN 25.

  Next, the navigation device 10 can input various instructions in the same way as the position detector 11 for detecting the current position of the vehicle, the operation switch group 12 for inputting various instructions from the user, and the operation switch group 12. A remote control terminal (hereinafter referred to as “remote control”) 13a that is separate from the navigation device 10, a remote control sensor 13b that inputs a signal from the remote control 13a, and a map storage medium that records map data and various types of information A map data input device 14 for inputting map data and the like, a display unit 15 for displaying a map and various information, a voice output unit 16 for outputting various guide voices, and a voice uttered by the user That exchanges various vehicle information and the like with other devices via the in-vehicle LAN 25 and the microphone 17 that outputs electrical signals based on The LAN communication unit 18, the above-described position detector 11, operation switch group 12, remote control sensor 13b, map data input device 14, microphone 17, and various processes are executed according to inputs from the in-vehicle LAN communication unit 18, and the display unit 15, a voice output unit 16, and a control unit 19 that controls the in-vehicle LAN communication unit 18.

  Among them, the position detector 11 receives a radio wave from a GPS (Global Positioning System) artificial satellite via a GPS antenna (not shown) and outputs a received signal, and a rotational motion applied to the vehicle. A gyroscope 11b for detecting the magnitude of the vehicle, a distance sensor 11c for detecting the distance traveled from the longitudinal acceleration of the vehicle, etc., and a geomagnetic sensor 11d for detecting the traveling direction from the geomagnetism. And the control part 19 calculates the position, direction, speed, etc. of a vehicle based on the output signal from these each sensors 11a-11d.

  There are various methods for obtaining the current position based on the output signal from the GPS receiver 11a. However, any of a single positioning method and a relative positioning method (D-GPS method, interference positioning method) may be used. . In particular, the RTK-GPS (Real Time Kinematics Global Positioning System) method among interference positioning methods is preferably used.

  The operation switch group 12 includes a touch panel configured integrally with the display surface of the display unit 15 and mechanical key switches provided around the display unit 15. The touch panel and the display unit 15 are laminated and integrated. There are various types of touch panels, such as a pressure-sensitive method, an electromagnetic induction method, a capacitance method, or a combination of these methods. It may be used.

  The map data input device 14 is a device for inputting various data stored in a map storage medium (not shown). Map storage media includes map data (node data, link data, cost data, road data, terrain data, mark data, intersection data, temporary stop data, facility data, etc.), target voice data, voice recognition data Etc. are stored. As a type of storage medium for storing such data, a storage medium such as a hard disk or a memory card may be used in addition to a CD-ROM or DVD-ROM.

  The display unit 15 is a color display device such as a liquid crystal display, an organic EL display, or a CRT, and any of them may be used. On the display screen of the display unit 15, a map image based on the map data input from the map data input device 14 is displayed, and a mark indicating the current position of the vehicle detected by the position detector 11 over the map image, Additional data such as guidance routes to destinations, names, landmarks, and various facility marks are also displayed.

The sound output unit 16 outputs facility guides and various target sounds input from the map data input device 14.
The microphone 17 outputs an electric signal (voice signal) based on the input voice to the control unit 19 when the user inputs (speaks) voice. The navigation device 10 can be operated by a voice command input via the microphone 17.

The in-vehicle LAN communication unit 18 communicates with various devices (such as the engine ECU 31) connected to the in-vehicle LAN 25 via the in-vehicle LAN 25.
The control unit 19 is composed mainly of a well-known microcomputer comprising a CPU 19a, a ROM 19b, a RAM 19c, an I / O, and a bus line for connecting these components, and detects a position based on a program stored in the ROM 19b. A map display function for calculating the current position of the vehicle as a set of coordinates and traveling directions based on each detection signal from the device 11 and displaying a map near the current position read from the map data input device 14 on the display unit 15; Based on the point data read from the map data input device 14, the destination facility is selected in accordance with the operation of the operation switch group 12 or voice input via the microphone 17, and the optimum route from the current position to the destination is automatically selected. Route guidance function to perform route guidance by performing desired route calculation (eg Dijkstra method) and voice output 16 and interactive function of controlling the navigation device 10 by entering a voice command from the driver while interaction with the driver via the microphone 17 comprises a (agent function).

In the navigation apparatus 10 configured as described above, the control unit 19 executes an intersection guidance process for guiding the driver when the vehicle approaches the intersection.
The ROM 19b constituting the control unit 19 stores an action pattern template (hereinafter also simply referred to as a template) used when estimating the driver's action.

  Here, a procedure for creating an action pattern template stored in the ROM 19b will be described with reference to a flowchart shown in FIG. 2 and an explanatory diagram shown in FIG. The behavior pattern template is created by processing executed on a computer outside the navigation device 10. In addition, FIGS. 2 to 3 describe a procedure for creating an action pattern template in which the remaining time until reaching the intersection is τ for the driving action that stops at the intersection.

As shown in FIG. 2, when the template creation procedure is started, first, in S110, driving behaviors that stop at a plurality of intersections from a database in which data obtained by sampling various vehicle information representing vehicle states in time series are accumulated. , The stop accumulation data group UG _stop ^τ until the remaining time τ until reaching the intersection is acquired. The remaining time τ is calculated as Lc / Vc, where Lc is the distance to the intersection and Vc is the travel speed of the vehicle.

The stop accumulation data group UG _stop ^τ includes a plurality of vehicle information designated in advance (accelerator opening A, brake operation amount B, vehicle speed V in this embodiment). Further, the stop accumulation data group UG _stop ^τ is driving data from a point where the remaining time τ is required until the vehicle reaches the intersection to a preset template time range (for example, 5 seconds in the present embodiment). Are collected for a plurality of drivers and a plurality of intersections. The driving data accumulated in the database may be collected by measurement using an actual vehicle or may be collected using a driving simulator.

Subsequently, in S120, each element A, B, V of the stop accumulation data group UG _stop ^τ acquired in S110 is normalized so as to have a value of 0 to 1.
Subsequently, in S130, the stop accumulation data group UG _stop ^τ is classified using a clustering method. Specifically, a distance between the _stop accumulation data group UG _stop ^τ is defined, and those having a short distance are classified into the same cluster. Note that such a clustering method is well known, and for example, the K-means method can be used, but the present invention is not limited to this.

Subsequently, in S140, according to the classification result in S130, as shown in FIG. 3, the average value of the _stop accumulation data group UG _stop ^τ is calculated for each of the classified M clusters (m = 1 to M).
Hereinafter, an average value of the stop accumulation data group UG _stop ^τ of the cluster m is referred to as a stop average data group UG _stop ^τm, and is expressed by the equation (1). However, U _stop ^τm (i) in equation (1) is the i (i = 1 to Tp / S) -th average operation data constituting the _stop average data group UG _stop ^τm , and is expressed by equation (2). And Further, A _stop ^τm (i) in equation (2) is the accelerator opening obtained by extracting the accelerator opening A of the i-th operation data from all the stop accumulation data groups UG _stop ^τ included in the same cluster m. The average value of degree A, and B _stop ^τm (i) and V _stop ^τm (i) are average values of the brake operation amount B and the vehicle speed V obtained in the same manner.

UG _stop ^τm
= {U _stop ^τm (1), U _stop ^τm (2),..., U _stop ^τm (Tp / S)} (1)
U _stop ^τm (i)
= {A _stop ^τm (i), B _stop ^τm (i), V _stop ^τm (i)} (2)
After that, in S150, the stop average data group UG _stop ^τm is ^set as a stop action template T _stop ^τm (hereinafter also referred to as a stop action template T _stop ^τm ), and the template creation procedure ends.

A driving action template T _turn ^τm that turns at an intersection (hereinafter also referred to as a right / left turn action template T _turn ^τm ) and a driving action template T _straight ^{τm that travels} _{straight at an} ^intersection (hereinafter also referred to as a _straight ^travel template T _straight ^τm ) Is created in the same procedure as described above.

Accordingly, the ROM 19b constituting the control unit 19 includes a stop action template T _stop ^τm indicating a typical driving action of the driver between the remaining time τ and the template time range for each of a plurality of different remaining times τ. A right / left turn action template T _turn ^τm and a straight ^traveling motion template T _straight ^τm are stored.

  Next, the procedure of intersection guidance processing executed by the control unit 19 of the navigation device 10 will be described with reference to FIG. FIG. 4 is a flowchart showing the intersection guidance process. This intersection guidance process is a process that is repeatedly executed while the control unit 19 is activated (powered on).

  When this intersection guidance process is executed, the control unit 19 first calculates a distance to the nearest intersection (hereinafter referred to as a target intersection) in S210. Specifically, the position of the nearest intersection (target intersection) from the current position of the host vehicle detected by the position detector 11 is read from the map data input device 14. Then, a distance Cd from the current position of the host vehicle to the position of the target intersection is calculated.

  Thereafter, in S220, it is determined whether or not the distance Cd is equal to or less than a predetermined determination start distance (300 m in the present embodiment). If the distance Cd is greater than the determination start distance (S220: NO), the process returns to S210 and the above processing is repeated. On the other hand, if the distance Cd is equal to or less than the determination start distance (S220: YES), the host vehicle is the target intersection. In S230, a driving behavior estimation process (described later) for estimating the driver's behavior is executed, and in S240, guidance regarding the target intersection is voiced via the voice output unit 16. A voice guidance execution process (described later) is performed.

  Thereafter, in S250, a distance Cd from the current position of the host vehicle to the position of the target intersection is calculated. In S260, it is determined whether or not the target intersection has been passed based on the distance Cd calculated in S250. If the vehicle has not yet passed through the target intersection (S260: NO), the process returns to S220, and the processes of S220 to S260 are repeated. On the other hand, if the vehicle has already passed through the target intersection (S260: YES), The intersection guidance process is temporarily terminated.

Next, the procedure of the driving action estimation process executed in S230 will be described with reference to FIG. FIG. 5 is a flowchart showing the driving action estimation process.
When the driving action estimation process is started, first, in S310, driving using the accelerator opening A (t), the brake operation amount B (t), and the vehicle speed V (t) via the in-vehicle LAN communication unit 18 as elements. Data DU (t) = {A (t), B (t), V (t)} is acquired.

  In S320, each element A (t), B (t), V (t) of the operation data DU (t) acquired in S310 is normalized so as to take a value of 0 to 1. Note that t represents the time at which data was sampled, and the operation data DU (t) is read every preset sampling interval S (0.5 seconds in the present embodiment).

  Further, in S330, the operation data DU (t), DU (t−S),... DU (t−) for the past Tw (= 5 seconds) including the operation data DU (t) normalized in S320. Tw) is updated in the RAM so that the data stored in the RAM constituting the control unit 19 is stored. In the following, the operation data for Tw seconds (Tw / S) stored in the RAM is expressed as an operation data group MG by the expression (3), and each operation data MU (j) belonging to the operation data group MG is expressed as follows. (4) Expressed by the formula. However, j = 1 to Tw / S, j = 1 represents the oldest data, and j = Tw / S represents the latest data. That is, DU (t−Tw) = MU (1),... DU (t) = MU (Tw / S).

MG = {MU (1), MU (2), ..., MU (Tw / S)} (3)
MU (j) = {A (j), B (j), V (j)} (4)
In S340, the remaining time τ until reaching the target intersection is calculated. Specifically, the distance Cd from the current position of the host vehicle to the position of the target intersection is calculated, and is calculated by the equation (5) using the latest vehicle speed data Vc acquired in S310.

τ = Cd / Vc (5)
In S350, the stop action template T _stop ^τm , the right / left turn action template T _turn ^τm , and the straight ^traveling template T _straight stored in the ROM 19b constituting the control unit 19 and provided for each of a plurality of different remaining times τ. ^{From τm} , a stop action template T _stop ^τm , a right / left turn action template T _turn ^τm corresponding to the remaining time τ calculated in S340, and a _straight ^traveling action template T _straight ^τm are selected.

For example, when the remaining time τ is 4 seconds, the stop action template T _stop ^τm (see template group ^{TG01 in} FIG. 7) corresponding to τ = 4 [seconds] and the right corresponding to τ = 4 [seconds] A left turn action template T _turn ^τm (see template group TG11 in FIG. 8) and a straight motion template T _straight ^τm corresponding to τ = 4 [seconds] are selected. When the remaining time τ is 2 seconds, the stop action template T _stop ^τm corresponding to τ = 2 [seconds] (see the template group TG02 in FIG. 7) and the right corresponding to τ = 2 [seconds] A left turn action template T _turn ^τm (see template group TG12 in FIG. 8) and a straight motion template T _straight ^τm corresponding to τ = 4 [seconds] are selected.

  As shown in FIGS. 7 and 8, the number of templates to be selected varies depending on the remaining time τ even for the same driving action templates. Moreover, even if it is the same remaining time (tau), the number of the templates selected changes with driving behavior.

Thereafter, in S360, the weighted distance D _stop ^τ (t), D between the operation data group MG updated in S330 and the templates T _stop ^τm , T _turn ^τm , T _straight ^{τm of} each action selected in S350 at time t. _Turn ^τ (t), D _straight ^τ (t) are calculated.

Specifically, first, an index representing the distance from the driving data group MG for each of the action templates T _stop ^τm , T _turn ^τm , and T _straight ^{τm according to} equations (6), (7), and (8). Data d _stop ^τm (t), d _turn ^τm (t), and d _straight ^τm (t) are calculated. Note that a _stop ^τm , a _turn ^τm , and a _straight ^τm in the expressions (6), (7), and (8) are template weighting factors.

Then, (9), (10), (11) as shown in the expression template T _stop ^.tau.m of each behavior, T _turn ^.tau.m, index data calculated for each ^{_{T straight τm d stop τm (t}} ), d turn τm (T), d _straight ^τm (t) having the smallest value (that is, the distance from the action pattern template most similar to the driving data group MG) is the weighted distance D _stop ^τ (t ), D _turn ^τ (t), D _straight ^τ (t).

In S370, as shown in equations (12), (13), and (14), the difference E ₁ ^τ between the weighted distances D _stop ^τ (t), D _turn ^τ (t), and D _straight ^τ (t) (T), E ₂ ^τ (t), E ₃ ^τ (t) are calculated.

E ₁ ^τ (t) = D _stop ^τ (t) −D _turn ^τ (t) (12)
E ₂ ^τ (t) = D _stop ^τ (t) −D _straight ^τ (t) (13)
E ₃ ^τ (t) = D _turn ^τ (t) −D _straight ^τ (t) (14)
Next, in S380, the normalized values of the differences E ₁ ^τ (t), E ₂ ^τ (t), and E ₃ ^τ (t) are obtained as shown in equations (15), (16), and (17). Similarities R ₁ (t), R ₂ (t), and R ₃ (t) are assumed.

Here, the similarity R ₁ (t) indicates whether the actual driving data is similar to a stationary driving behavior template or a curved driving behavior template. That is, the closer the similarity R ₁ (t) is to 0, the more similar it is to the driving action template that bends, and the closer to 1, the more similar the driving action template is to stop.

Similarly, the similarity R ₂ (t) indicates which of actual driving data is similar to a stationary driving action template or a straight driving action template. That is, the closer the similarity R ₂ (t) is to 0, the more similar it is to the template of the driving action that goes straight, and the closer to 1, the closer to the template of the driving action that stops.

The similarity R ₃ (t) indicates whether the actual driving data is similar to a curved driving action template or a straight driving action template. That is, the closer the similarity R ₃ (t) is to 0, the more similar it is to the driving action template that goes straight, and the closer to 1, the closer to the driving action template that turns.

In S390, as shown in Table 1, the similarity R ₁ (t), R ₂ (t), R ₃ (t) and the threshold values θ _stop-turn , θ _turn-stop , θ _{stop-straight} , θ _The driving behavior is determined by the magnitude relationship with _{turn-straight} .

Here, the threshold values θ _stop-turn and θ _turn-stop are used for comparison with the similarity R ₁ (t), and θ _stop-turn is set to be larger than θ _turn-stop. Yes. That is, when the similarity R ₁ (t) is larger than θ _stop-turn, it is easy to determine that the driving action is to be stopped, and when the similarity R ₁ (t) is smaller than θ _turn-stop , the vehicle turns. It becomes easier to determine when driving action is performed.

Θ _{stop-straight} is used for comparison with the similarity R ₂ (t). That is, when the similarity R ₂ (t) is greater than θ _{stop-straight,} it is easy to determine that the driving action is to stop, and when the similarity R ₂ (t) is equal to or less than θ _{stop-straight} , the vehicle _travels straight. It becomes easy to be determined when the driving action is performed.

Θ _{turn-straight} is used for comparison with the similarity R ₃ (t). That is, when the similarity R ₃ (t) is greater than θ _{turn-straight,} it is easy to determine that the driver is making a turn, and when the similarity R ₃ (t) is equal to or less than θ _{turn-straight} , the vehicle _travels straight. It becomes easy to be determined when the driving action is performed.

Thereby, for example, in “Case 1” and “Case 2” of Table 1, the similarity R ₁ (t) is larger than θ _stop-turn and the similarity R ₂ (t) is larger than θ _{stop-straight} . Regardless of the value of the similarity R ₃ (t), it is determined to perform a driving action that stops.

Further, in “Case 3” and “Case 4” in Table 1, the similarity R ₁ (t) is larger than θ _stop-turn and the similarity R ₂ (t) is equal to or less than θ _{stop-straight.} Regardless of the value of the degree R ₃ (t), it is determined that a driving action other than a driving action that stops and a driving action that turns (hereinafter, referred to as other driving actions) is performed.

In “Case 5” of Table 1, the similarity R ₁ (t) is smaller than θ _turn-stop , the similarity R ₂ (t) is larger than θ _{stop-straight} , and the similarity R ₃ (t) Is larger than θ _{turn-straight,} so it is decided to perform a turning driving action.

In “Case 6” in Table 1, the similarity R ₁ (t) is smaller than θ _turn-stop , the similarity R ₂ (t) is larger than θ _{stop-straight} , and the similarity R ₃ (t) Is less than θ _{turn-straight} , so it is determined to perform other driving actions.

In “Case 9” of Table 1, the similarity R ₁ (t) is smaller than θ _{stop-turn and} larger than θ _turn-stop , and the similarity R ₂ (t) is larger than θ _{stop-straight} and similar. Since the degree R ₃ (t) is larger than θ _{turn-straight} , it is determined to perform a driving action that stops or turns.

  When the process of S390 is completed, it is determined in S400 whether or not it is determined to perform a driving action that makes a turn in S390. Here, when it is determined that the driving action to turn is performed (S400: YES), the estimation result flag Fn is set to 1 in S410, and the driving action estimation process is ended.

On the other hand, when it is not determined to perform the turning driving action (S400: NO), the estimation result flag Fn is set to 0 in S420, and the driving action estimation process is ended.
Next, the procedure of the voice guidance execution process executed in S240 will be described with reference to FIG. FIG. 6 is a flowchart showing voice guidance execution processing.

  When the voice guidance execution process is started, first, in S510, it is determined whether or not to turn at the target intersection with reference to the route set by the route guidance function. Here, when the turn should be made at the target intersection (S510: YES), it is determined whether or not the estimation result flag Fn is set to 1 in S520. Here, when the estimation result flag Fn is set to 1 (S520: YES), since the course at the target intersection based on the set route matches the driver's action, voice guidance is performed. The voice guidance execution process is terminated as it is.

  On the other hand, if the estimation result flag Fn is not set to 1 (S520: NO), the function f is input with the vehicle speed V normalized in S320 as the distance Cd to the target intersection in S530. It is determined whether or not the required distance Dneed calculated using (V) is smaller than f (V).

  As shown in FIG. 9, the function f (V) has a larger required distance Dneed as the vehicle speed V increases, and the first voice guidance can be followed with a margin for the driver. It is set to start at the timing.

  In S530, if the distance Cd to the target intersection is smaller than the required distance Dneed (see d1 in the figure) (S530: YES), the process proceeds to S540, and the target intersection should be bent (or should not go straight). ) Is executed, and then the voice guidance execution process is terminated.

  On the other hand, if the distance Cd to the target intersection is equal to or greater than the required distance Dneed in S530 (see d2 in the figure) (S530: NO), the voice guidance is executed as if it is still too early to execute the voice guidance. Then, the voice guidance execution process is terminated as it is.

  If it is determined in S510 that the vehicle should not turn at the target intersection (S510: NO), the process proceeds to S550, whether the estimation result flag Fn is set to 1, that is, the driver turns at the intersection. It is determined whether or not it is going.

  If it is determined in S550 that the estimation result flag Fn is set to 1 (S550: YES), the process proceeds to S560, and guidance is provided that the target intersection should not bend (or should go straight). The voice guidance to be executed is executed, and then the voice guidance execution process is terminated. On the other hand, if it is determined in S550 that the estimation result flag Fn is not set to 1 (S550: NO), the voice guidance execution process is terminated without executing the voice guidance.

  In the navigation device 10 configured as described above, the current position of the host vehicle is detected by the position detector 11, and the position of the nearest intersection (target intersection) from the current position of the host vehicle detected by the position detector 11 is determined. Read from the map data input device 14 (S210).

  In addition, driving data DU (t) = {A (t), B (), which includes the accelerator opening A (t), the brake operation amount B (t), and the vehicle speed V (t) via the in-vehicle LAN communication unit 25. t), V (t)} are acquired (S310). Further, based on the distance Cd from the current position of the host vehicle to the position of the target intersection and the latest vehicle speed data Vc, the remaining time τ until reaching the target intersection is calculated (S340).

Further, the ROM 19b constituting the control unit 19 includes a stop action template T _stop ^τm indicating a typical driving action of the driver between the remaining time τ and the template time range for each of a plurality of different remaining times τ. A right / left turn action template T _turn ^τm and a straight ^traveling motion template T _straight ^τm are stored.

Then, based on the calculated remaining time τ, a stop action template T _stop ^τm , a right / left turn action template T _turn ^τm and a _straight ^travel template T _straight ^τm corresponding to the remaining time τ are selected (S350), and the obtained driving data Based on the history of DU (t) and the selected template, the driving behavior of the driver at the target intersection is estimated (S360 to S390).

  In the navigation device 10 configured as described above, a template for each driving action indicating a typical driving action of the driver from the time before the remaining time τ to the preset template time range before the remaining time τ is provided. The driving behavior is estimated using a template corresponding to the remaining time τ and the history of the driving data DU (t).

  Even in a plurality of different driving behaviors, even if there is a similar part in the time change of the vehicle speed and the driver's vehicle operation before reaching the target intersection, the time region where this similar part occurs is the driving Different for each action. For example, when comparing a driving action that turns and a driving action that stops, there is a time region in which both the brake amount increases and the accelerator amount and speed decrease before the intersection. However, the driving action that stops is faster than the driving action that turns, as the brake amount increases and the accelerator amount and speed decrease (see FIG. 10).

  For this reason, comparing different driving behavior templates corresponding to the same remaining time τ, compared to comparing different driving behavior templates for a long time region from approaching the target intersection until reaching the target intersection However, there are fewer similar parts. That is, since the driving behavior can be estimated using a template that clearly shows the difference between different driving behaviors, the estimation accuracy of the driving behavior can be improved.

  In the embodiment described above, the navigation apparatus 10 is the driving behavior estimation apparatus in the present invention, the position detector 11 is the current position information acquisition means in the present invention, and the process in S210 is the target point position information acquisition means in the present invention, the process in S310. Is the operation data acquisition means in the present invention, the process in S340 is the remaining time calculation means in the present invention, the ROM 19b is the template storage means in the present invention, the process in S350 is the template acquisition means in the present invention, and the processes in S360 to S390 are in the present invention. The driving action estimation means and the remaining time τ are the point arrival remaining time in the present invention.

As mentioned above, although one Example of this invention was described, this invention is not limited to the said Example, As long as it belongs to the technical scope of this invention, a various form can be taken.
For example, in the above embodiment, the remaining time τ is calculated using information indicating the traveling speed of the host vehicle. However, the acceleration of the host vehicle may be used, or both the traveling speed and acceleration of the host vehicle may be used. It may be used.

  Moreover, in the said embodiment, although what estimated the driving | running | working action in an intersection was shown, it is not limited to this, The point (for example, temporary stop point) which needs to stop, or the point where a route branches ( For example, the estimation may be performed using a highway ramp).

  In the above embodiment, the accelerator opening A, the brake operation amount B, and the vehicle speed V are used as the vehicle information used for estimating the driver's behavior. However, information other than the vehicle speed V affects the traveling state of the host vehicle. If the operation information is related to the vehicle operation of the driver who gives the vehicle, other information may be used, and the number thereof is not limited to three, and may be two or less or four or more. .

  DESCRIPTION OF SYMBOLS 10 ... Navigation apparatus, 11 ... Position detector, 12 ... Operation switch group, 13a ... Remote control, 13b ... Remote control sensor, 14 ... Map data input device, 15 ... Display part, 16 ... Audio | voice output part, 17 ... Microphone, 18 ... In-vehicle LAN communication unit, 19 ... control unit, 19a ... CPU, 19b ... ROM, 19c ... RAM, 25 ... in-vehicle LAN, 31 ... engine ECU, 32 ... brake ECU, 33 ... steering ECU

Claims (5)

Current position information acquisition means for acquiring current position information indicating the current position of the host vehicle;
Target point position information acquisition means for acquiring target point position information indicating the position of the target point, which is a target point for estimating the driving behavior of the host vehicle;
Driving data acquisition means for acquiring driving data consisting of speed information of the own vehicle and operation information related to the vehicle operation of the driver that affects the running state of the own vehicle;
The current position information acquired by the current position information acquisition means, the target point position information acquired by the target point position information acquisition means, and speed calculation information for calculating the speed of the host vehicle A remaining time calculating means for calculating a point arrival remaining time, which is a time until the host vehicle reaches the target point,
Typical driving behavior of the driver from before the point arrival remaining time to reach the target point to a predetermined fixed period before the point arrival remaining time for each of a plurality of different point arrival remaining times A template storage means for storing a template for each driving action,
Based on the remaining point arrival time calculated by the remaining time calculating unit, a template acquisition unit that acquires the template corresponding to the remaining point arrival time from the template storage unit;
Driving behavior estimation means for estimating the driving behavior of the driver at the target point based on the history of the driving data acquired by the driving data acquisition means and the template acquired by the template acquisition means. A driving action estimation device comprising: and. The driving behavior estimation apparatus according to claim 1, wherein the speed calculation information is at least one of a speed and an acceleration of the host vehicle. The driving behavior estimation device according to claim 1, wherein the driving behavior is a left / right turn behavior and a stop behavior. The driving behavior estimation apparatus according to any one of claims 1 to 3, wherein the target point is a next intersection on a travel route of the host vehicle. The driving operation estimation device according to any one of claims 1 to 4, wherein the operation information relates to an accelerator pedal and a brake pedal.