JP2017129973A - Driving support apparatus and driving support method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driving support apparatus configured to improve the accuracy of estimating driving operation, and a driving support method.SOLUTION: A driving support apparatus includes a detection unit, an acquisition unit, an estimation unit, and an information generation unit. The detection unit detects preliminary behavior of a driver which is performed before operating a moving body. The acquisition unit acquires surrounding information on the surroundings of the moving body. The estimation unit estimates driving operation of the driver, on the basis of the preliminary behavior detected by the detection unit and the surrounding information acquired by the acquisition unit. The information generation unit generates support information for supporting the driving operation estimated by the estimation unit.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a driving support device and a driving support method.

  2. Description of the Related Art Conventionally, there is a driving support device that estimates a driving operation performed by the driver by detecting a series of operations performed before the driver operates the moving body and performs driving support for performing the driving operation.

  In addition, this device proposes a technique that enables estimation of a driving operation from a newly detected driver's operation by storing a series of operations before a driving operation performed in the past as a pattern. (For example, refer to Patent Document 1).

JP 2008-232912 A

  However, in the above-described prior art, if the driver performs an operation similar to the above pattern without intending a specific driving operation, there is a possibility that the specific driving operation corresponding to the pattern is erroneously estimated. Thus, the prior art has room for improvement in the estimation accuracy of the driving operation.

  The present invention has been made in view of the above, and an object of the present invention is to provide a driving support device and a driving support method that can improve the estimation accuracy of a driving operation.

  In order to solve the above-described problems and achieve the object, a driving support apparatus according to the present invention includes a detection unit, an acquisition unit, an estimation unit, and an information generation unit. The detection unit detects a preliminary operation performed before the driver operates the moving body. The acquisition unit acquires peripheral information related to the surrounding state of the moving object. The estimation unit estimates a driving operation performed by the driver based on the preliminary motion detected by the detection unit and the peripheral information acquired by the acquisition unit. The information generation unit generates support information for supporting the driving operation estimated by the estimation unit.

  According to the present invention, it is possible to improve the estimation accuracy of a driving operation.

Drawing 1 is a figure showing an outline of a driving support method concerning an embodiment. FIG. 2 is a block diagram illustrating a configuration of a driving support system including the driving support device according to the embodiment. FIG. 3A is a diagram (part 1) illustrating a method for detecting a driver's line of sight. FIG. 3B is a diagram (part 2) illustrating a method for detecting a driver's line of sight. FIG. 3C is a diagram (No. 3) illustrating the method for detecting the driver's line of sight. FIG. 4 is a diagram illustrating the estimation information. FIG. 5 is a flowchart illustrating a processing procedure of estimation processing executed by the driving support apparatus according to the embodiment. FIG. 6 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the driving support apparatus. FIG. 7 is a block diagram illustrating a configuration of a driving support system including a driving support device according to a modification of the embodiment. FIG. 8 is a diagram illustrating a modified example of the estimation information.

  Hereinafter, embodiments of a driving support apparatus and a driving support method disclosed in the present application will be described in detail with reference to the accompanying drawings. In addition, this invention is not limited by embodiment shown below.

  First, the outline | summary of the driving assistance method which concerns on embodiment is demonstrated using FIG. Drawing 1 is a figure showing an outline of a driving support method concerning an embodiment. Hereinafter, a case where the driving support method is executed by the driving support device 1 mounted on the vehicle 60 that is a moving body will be described.

  In addition, here, the vehicle 60 is shown as a moving body, but the present invention is not limited to this, and it may be operated by a driver such as a commercial vehicle such as a bus or taxi, a motorcycle, a ship, or the like. Any mobile body may be used.

  FIG. 1 shows a scene where the vehicle 60 tries to pass the front vehicle 61. It is possible to apply the driving support method according to the present embodiment other than the scene shown in FIG. 1, and examples applied in other scenes will be described later with reference to FIGS. 7 and 8.

  The driving support method according to the embodiment estimates a driving operation performed by the driver based on a series of operations (hereinafter referred to as a preliminary operation) performed before the driver operates the vehicle 60 and a surrounding situation of the vehicle 60. Thus, this driving operation is supported.

  Specifically, as shown in FIG. 1, first, the driving support method detects a preliminary operation that is performed before the driver operates the vehicle 60 (step S <b> 1). Here, the preliminary operation is an operation that accompanies the driving operation and refers to an operation of the driver that is performed unconsciously or consciously before the driving operation. For example, as shown in FIG. 1, the driver often looks at the side mirror to check whether there is another vehicle in the lane to be changed before changing the lane. Therefore, for example, in the case of FIG. 1, the driving assistance device 1 detects an operation of gazing at the side mirror as a preliminary operation.

  Next, the driving support method according to the embodiment acquires peripheral information related to the surrounding situation of the vehicle 60 (step S2), and estimates a driving operation based on the detected preliminary movement and the peripheral information (step S3).

  Here, the point that the driving support device 1 estimates the driving operation based on the preliminary operation and the peripheral information will be described. First, a case where the driving support device 1 estimates a driving operation based on a preliminary operation will be described. In this case, it is assumed that the driving support device 1 stores the driving operation and the preliminary operation in association with each other. For example, in FIG. 1, the driving support device 1 associates and stores the side mirror gaze operation and the lane change, and estimates whether or not the preliminary operation detected from the driver is the preliminary operation for changing the lane. .

  However, if the driving operation is estimated based on the preliminary operation as described above, the estimation accuracy of the driving operation may be lowered. For example, the side mirror may be watched even when the driver tries to turn right or left. In this case, if the driving operation is estimated based on the preliminary movement as described above, the driving operation such as a right turn or a left turn may be erroneously estimated as a lane change.

  Therefore, in the driving support method according to the embodiment, the driving operation is estimated based on the preliminary operation and the peripheral information. Thereby, the erroneous estimation of the driving operation can be reduced.

  Specifically, the driving assistance device 1 acquires object information such as the forward vehicle 61 existing around the vehicle 60 as, for example, peripheral information. For example, in FIG. 1, the driving support device 1 acquires information related to the forward vehicle 61 (for example, an inter-vehicle distance from the forward vehicle 61) as the object information, and performs a driving operation based on the information and the gaze operation of the side mirror. Is a lane change.

  As described above, when a similar preliminary motion is detected from the driver, it is determined whether or not the periphery of the vehicle 60 is in a situation where it is necessary to perform the preliminary motion. The estimation accuracy of the driving operation can be improved.

  And the driving assistance method which concerns on embodiment produces | generates assistance information according to the estimated driving operation (step S4). Specifically, when the driving support device 1 estimates a lane change as a driving operation, for example, the vehicle assists the vehicle with a voice such as “Is the lane changed? Please turn out the turn signal and pay attention to the surrounding vehicle.” The support information is presented to the driver by outputting from 60 speakers (not shown).

  Here, the support information is information for supporting the driving operation performed by the driver, and includes, for example, content for prompting the estimated driving operation, content for calling attention, and the like. Further, the support information to be generated is not limited to voice, and for example, the support information may be displayed on a display disposed in the vehicle 60.

  In addition, the preliminary | backup operation | movement which the driving assistance device 1 memorize | stores is good also as the pattern produced | generated using machine learning, for example. The driving support device 1 extracts the feature of the preliminary motion from a large amount of past preliminary motion data by machine learning, and stores the feature as a preliminary motion pattern.

  The past preliminary motion data is so-called teacher data, and includes preliminary motion under various conditions such as a driver's mind-body condition and road characteristics in a specific driving operation. A detailed description of machine learning will be described later.

  As a result, the driving support device 1 can perform estimation if the preliminary motion features match even if the preliminary motion detected by the driver does not match the stored preliminary motion. For this reason, even if the preliminary movement changes depending on the mind and body state of the driver, road characteristics, and the like, the driving operation can be estimated, and the estimation accuracy of the driving operation can be further improved.

  Further, the preliminary motion detected from the driver is not limited to the movement of the line of sight, and may be an operation associated with a specific driving operation such as an accelerator and brake operation or a facial movement by the driver.

  In addition to the forward vehicle 61, the peripheral information may be map information including object information such as a rear vehicle and a fallen object, and a road configuration such as an intersection, a branch point, a parking lot, and a traffic signal.

  Next, with reference to FIG. 2, the structure of the driving assistance apparatus 1 in which the driving assistance method which concerns on embodiment is performed is demonstrated concretely. FIG. 2 is a block diagram illustrating a configuration of the driving support system S including the driving support device 1 according to the embodiment.

  As shown in FIG. 2, the driving support system S is mounted on, for example, a vehicle 60, and includes a driving support device 1, an output unit 2, an infrared camera 30, an infrared LED 31, a camera 40, and a position detection unit 41. Prepare.

  The output unit 2 is a device that presents support information output from the driving support device 1 to the driver. The output unit 2 may be, for example, a speaker or a display of a navigation device (not shown). In this case, for example, the support information is presented to the driver by displaying the support information on the display. Alternatively, the output unit 2 may be an instrument panel or the like.

  The infrared camera 30 and the infrared LED 31 are provided, for example, in the upper part of the steering column of the driver's seat, and continuously capture an area including the driver's face illuminated by the infrared LED 31 with a predetermined interval. The infrared image obtained from the infrared camera 30 includes the driver's eyeballs, the corners of the eyes and the eyes.

  The camera 40 is a camera including an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor), for example, and is installed at a position where the rear of the vehicle 60 is imaged. An image captured by the camera 40 is output to the driving support apparatus 1.

  In addition, the camera 40 is not limited to the position which images the back of the vehicle 60, What is necessary is just the position which images at least 1 direction of front and rear, right and left. Further, a plurality of cameras 40 may be provided so as to capture the surroundings of the vehicle 60.

  The position detection unit 41 detects the current location (for example, latitude and longitude) of the vehicle 60. The position detection unit 41 receives signals output from a plurality of GPS (Global Positioning System) satellites, and acquires current location information of the vehicle 60 based on the received signals. The position detection unit 41 can also acquire accurate current time information from GPS satellites. The position detection unit 41 notifies the driving support device 1 of the acquired information. In addition, when the navigation apparatus which has a GPS function is mounted in the vehicle 60, the position detection part 41 may acquire the information of the position and time of the vehicle 60 which GPS of this navigation apparatus acquired.

  The driving support device 1 according to the embodiment includes a control unit 10 and a storage unit 20. The control unit 10 includes a detection unit 11, an acquisition unit 12, a learning unit 13, an estimation unit 14, and an information generation unit 15.

  The control unit 10 is, for example, a microcomputer including a CPU (Central Processing Unit), a RAM (Random Access Memory), and a ROM (Read Only Memory). Such a CPU functions as, for example, the detection unit 11, the acquisition unit 12, the learning unit 13, the estimation unit 14, and the information generation unit 15 described above by performing arithmetic processing according to a program stored in advance in the ROM.

  The detection unit 11 detects a preliminary operation performed before the driver operates the vehicle 60. The detection unit 11 includes, for example, a line-of-sight detection unit 11a, and detects the movement of the driver's line of sight as a preliminary movement by acquiring an infrared image including the driver's face irradiated by the infrared LED 31 from the infrared camera 30. To do.

  Here, with reference to FIG. 3A and FIG. 3B, the detection method of a driver | operator's eyes | visual_axis by the eyes | visual_axis detection part 11a is demonstrated concretely. 3A to 3C are diagrams illustrating a gaze detection method performed by the gaze detection unit 11a.

  As shown in FIG. 3A, the line-of-sight detection unit 11a first includes the eyes of the driver by pattern matching, for example, from an infrared image obtained by imaging the driver's face illuminated by the infrared LED 31 with the infrared camera 30. Detect areas.

  Then, the line-of-sight detection unit 11a detects the driver's line-of-sight direction based on the positional relationship between the pupil 101 included in the eye area from the detected area and the infrared illumination reflection image 102 (corneal reflection) generated on the eyeball. .

  For example, as shown in FIG. 3A, if the pupil 101 is closer to the eye 103 than the corneal reflection 102, it can be seen that the driver is looking at the eye 103. Further, for example, if the pupil 101 is closer to the corner of the eye 104 than the corneal reflection 102, it can be seen that the driver is looking at the corner of the eye 104.

  Note that the detection of the line-of-sight direction is not limited to the method using the infrared camera 30 and the infrared LED 31, and for example, features of eyes, nose, mouth, and the like from a captured image of a camera provided at a position where the driver's face is imaged. The direction of the driver's face detected based on the extracted feature points based on the extracted points may be used as the line-of-sight direction.

  Then, as shown in the left diagram of FIG. 3B, the line-of-sight detection unit 11a uses the line-of-sight direction when viewing the infinity point ∞ in the frontal direction from the driver's viewpoint EP (for example, the infrared illumination reflected image 102). Then, an angle θ between the reference and the detected line-of-sight direction is calculated. The infinity point ∞ is, for example, the vanishing point of a flat straight road without a gradient.

  The graph on the right side of FIG. 3B is a graph in which the vertical axis represents time, and the horizontal axis represents the angle θ formed by the line-of-sight direction. The line-of-sight detection unit 11a detects the movement of the line of sight, for example, by calculating an angle θ formed continuously at a predetermined interval. The graph shown in FIG. 3B shows the movement of the line of sight in time series. Then, the line-of-sight detection unit 11a detects the preliminary movement from the movement of the line of sight based on, for example, the pattern of the preliminary movement learned by the learning unit 13.

  Specifically, as illustrated in the left diagram of FIG. 3C, for example, the movement of moving the line of sight in the order of the side mirror, the front, and the side mirror is learned by the learning unit 13 as the pattern of the preliminary operation when changing the lane. And The line-of-sight detection unit 11a detects, as a preliminary movement, a line-of-sight movement that matches the pattern, for example, by pattern recognition, etc.

  In this way, by detecting the movement of the driver's line of sight as a preliminary movement, the driving operation can be estimated without the driver performing a specific movement for the driving operation. The detection of the preliminary motion does not need to be completely coincident with the pattern generated by the learning unit 13 and may be a similar motion. 2 notifies the learning unit 13 and the estimation unit 14 of the detected preliminary movement.

  The acquisition unit 12 illustrated in FIG. 2 includes, for example, an object information acquisition unit 12a and a map information acquisition unit 12b, and acquires peripheral information related to the surrounding state of the vehicle 60.

  The object information acquisition unit 12a acquires object information related to objects existing around the vehicle 60 as the peripheral information. The object information is, for example, information such as the presence or absence of a moving body such as a front vehicle or a rear vehicle traveling around the vehicle 60, or the presence or absence of a fallen object such as a load that has fallen on the road. The object information includes information on the distance between the object and the vehicle 60. The object information is not limited to these, and may be information regarding an object that affects the driving operation.

  The object information acquisition unit 12a acquires, for example, an image of the camera 40 that captures the periphery of the vehicle 60, and uses a detection method such as pattern recognition or optical flow to detect the object, and to detect the detected object and the vehicle 60. Get object information such as relative speed and distance to the object.

  The acquisition of object information by the object information acquisition unit 12a may acquire object information using a millimeter wave radar or the like other than the image of the camera 40, and the acquisition method is not particularly limited.

  The map information acquisition unit 12b acquires map information 21 as peripheral information. The map information 21 is so-called electronic map data, and is road network data represented by nodes representing intersections and points on the road and links representing roads connecting the nodes. The node includes a node ID for identifying each node and attribute information such as information on intersections and branch points. The link includes attribute information such as a link ID for identifying each link and a road type (for example, a highway) corresponding to the link. The road network data includes facility information, and the facility information includes information such as the presence / absence of a parking lot and a facility type (for example, a restaurant).

  Further, the map information acquisition unit 12b converts the current location information of the vehicle 60 acquired from the position detection unit 41 into position information of the vehicle 60 on the electronic map. Note that the map information acquisition unit 12b is configured to store the map information 21 in the storage unit 20 in advance, but is not limited thereto. For example, the map information acquisition unit 12b is appropriately connected to an external server device or an in-vehicle navigation device. The map information 21 may be acquired.

  The acquisition unit 12 notifies the learning unit 13 and the estimation unit 14 of various information acquired by the object information acquisition unit 12a and the map information acquisition unit 12b as peripheral information.

  The learning unit 13 learns a driver's preliminary motion pattern by so-called machine learning based on the preliminary motion detected by the detection unit 11 and the peripheral information acquired by the acquisition unit 12. Here, the case where a lane change is performed will be described as an example of learning.

  Specifically, the learning unit 13 acquires the preliminary movement performed before the lane change from the line-of-sight detection unit 11a, and acquires the peripheral information of the time when the preliminary movement is started from the object information acquisition unit 12a. Here, it is assumed that the inter-vehicle distance between the vehicle 60 and the forward vehicle 61 is acquired as the peripheral information. The learning unit 13 uses the teacher data associating the preliminary motion with the peripheral information for pattern learning of the preliminary motion and stores the history data 22 in the storage unit 20.

  Here, the learning unit 13 detects a driving operation (lane change in this case) by the driver based on, for example, a detection result of a steering sensor (not shown). Or the operation detection part which is not shown in figure detects a driving operation based on detection results, such as a steering sensor, and you may make it the learning part 13 acquire this detection result. Moreover, you may make it acquire the estimation result by the estimation part 14 as driving operation.

  Then, learning by the learning unit 13 is performed using a plurality of past teacher data accumulated as the history data 22 and the latest acquired teacher data. That is, a pattern at the time of lane change is learned using a predetermined algorithm from a huge collection of teacher data at present and in the past. The learning unit 13 notifies the detection unit 11 of the learned pattern, and stores it in the storage unit 20 as the estimation information 23.

  In this way, the estimation information 23 learned by the learning unit 13 extracts common features from a large collection of past preliminary motion data, so that even if the preliminary motion of the driver is somewhat different, the preliminary motion If a common feature is included, the driving operation can be estimated, so that the estimation accuracy can be further improved.

  The peripheral information is not limited to the example described above. For example, roads such as the driver's concentration and sleepiness acquired by the acquisition unit 12 based on the movement of the line of sight detected by the line-of-sight detection unit 11a, and the road width and the number of lanes acquired by the map information acquisition unit 12b. Information on various situations such as characteristics, weather information, and traffic jam information can be acquired and used as teacher data. Such information is acquired by the acquisition unit 12, for example. Thereby, it can be made difficult to produce a difference in the estimation accuracy of the driving operation depending on the environment around the vehicle 60 and the state of the driver.

  Examples of algorithms used for machine learning include artificial neural networks, support vector machines, decision tree learning, correlation rule learning, and Bayesian networks, but other algorithms may be used.

  Moreover, the estimation part 23 can be updated to the newest version whenever the information of the detection part 11 and the acquisition part 12 is input by making the learning part 13 into the structure with which the driving assistance device 1 is provided. Note that the driving support device 1 does not necessarily include the learning unit 13. For example, the learning unit 13 may be included in an external server device.

  In such a case, the driving support device 1 and the server device are communicably connected to each other, the server device learns a preliminary operation pattern based on the history data 22 transmitted from the driving support device 1, and estimated information that is a learning result 23 is transmitted to the driving support device 1.

  This eliminates the need for the driving assistance device 1 to learn the pattern of the preliminary operation, thereby reducing the processing load on the CPU of the driving assistance device 1.

  The estimation unit 14 estimates a driving operation performed by the driver based on the preliminary movement detected by the detection unit 11 and the peripheral information acquired by the acquisition unit 12. The estimation process of the estimation unit 14 is performed according to the estimation information 23.

  Here, with reference to FIG. 4, the content of the estimation information 23 used for the estimation process of the estimation part 14 is demonstrated. FIG. 4 is a diagram showing the estimation information 23. The estimation information 23 is generated by the learning unit 13. Moreover, the content of the estimation information 23 shown in FIG. 4 is an example, and is not limited to this.

  As shown in FIG. 4, the estimation information 23 is information including items such as “driving operation”, “preliminary operation”, and “peripheral situation”. The “driving operation” is a driving operation performed by the driver. The “preliminary operation” is a preliminary operation before the corresponding driving operation is performed. The “surrounding situation” is a surrounding situation of the vehicle 60 when the driving operation is performed. The estimation unit 14 refers to the estimation information 23 to estimate whether the driving operation is a lane change based on the detected preliminary motion and the acquired peripheral information.

  For example, the preliminary movement detected by the line-of-sight detection unit 11a is the movement of the line of sight that moves in the order of the side mirror, the front, and the side mirror (see FIG. 3C), and the object information acquisition unit 12a Assume that object information indicating that the inter-vehicle distance is equal to or less than a predetermined distance is acquired. In such a case, the estimation unit 14 estimates that the driver changes lanes according to the estimation information 23.

  Here, for example, the case where the vehicle 60 changes lanes in order to pass the preceding vehicle 61 has been described, but an example of changing lanes is not limited thereto. For example, a lane change may be performed in order to avoid a rear vehicle approaching from behind the vehicle 60. As an example of such a scene, a scene in which the vehicle 60 and a rear vehicle are traveling in an overtaking lane on a highway can be cited.

  In this case, for example, the line-of-sight detection unit 11a detects the movement of the line of sight that moves in the order of the side mirror, the room mirror, and the side mirror as a preliminary operation, and the object information acquisition unit 12a acquires the peripheral information indicating the approach of the rear vehicle. The peripheral information indicating the approach of the rear vehicle includes, for example, information indicating that the distance to the rear vehicle is equal to or less than a predetermined value, information indicating that the relative speed with the rear vehicle is increasing, and the like. In such a case, the estimation unit 14 estimates that the lane change is performed according to the estimation information 23 in order to avoid the rear vehicle. The rear vehicle can use the same detection method as the front vehicle.

  Thereby, the driving assistance device 1 can accurately estimate the lane change for overtaking or avoiding other vehicles, and can appropriately assist based on the estimation result.

  Note that, even when a fallen object such as a load is acquired as object information, the estimation unit 14 detects such a spare when the distance between the vehicle 60 and the fallen object is equal to or less than a predetermined distance and a predetermined preliminary movement is detected. Estimation such as lane change or vehicle stop can be performed according to the operation.

  In this way, the estimation unit 14 can estimate an appropriate driving operation according to the content of the object information by performing estimation based on the object information acquired by the object information acquisition unit 12a.

  Further, the detection unit 11 detects an operation in which the driver gazes in the right front in a wide range, and the acquisition unit 12 determines that the distance between the vehicle 60 based on the map information 21 and the intersection where the right turn lane is equal to or less than a predetermined distance. Is acquired, it is estimated that the estimation part 14 will change a lane to the right turn lane.

  Further, the detection unit 11 detects the movement of the line of sight moving in the order of branching / confluence gazing and side mirrors, and the acquisition unit 12 determines that the distance between the vehicle 60 based on the map information 21 and the branch point or merging point is equal to or less than a predetermined distance. When the surrounding information is acquired, the estimation unit 14 estimates that the lane change for branching / merging is performed.

  Thus, the estimation unit 14 can estimate an appropriate driving operation according to the current location of the vehicle 60 by the map information acquisition unit 12b acquiring the map information 21 and the position information of the vehicle 60. Then, the estimation unit 14 notifies the information generation unit 15 of the estimated driving operation as an estimation result.

  The information generation unit 15 generates support information 24 for supporting driving operation based on the estimation result of the estimation unit 14. Specifically, support information 24 including support content corresponding to each driving operation is stored in advance in the storage unit 20, and the information generation unit 15 responds to the driving operation of the estimation result in the support information 24. The content of the support information 24 is selected.

  For example, when the output unit 2 is a speaker, the information generation unit 15 generates, as support information 24, an audio signal such as “Is the lane changed? Please turn out the turn signal and change the lane while paying attention to surrounding vehicles”. .

  For example, when the output unit 2 is a display unit such as a display, the image information including the character information indicating the voice signal is generated as the support information 24. When the output unit 2 includes a display unit such as a speaker and a display, the voice information and the character information are generated as the support information 24.

  When the output unit 2 is an instrument panel, display information for automatically displaying a winker indicator in the instrument panel at a predetermined blinking speed or a predetermined color may be generated as the support information 24, for example. In this case, it may be possible to provide assistance for prompting the driver to perform a blinker operation by setting the blinking speed or display color to be different from the blinking speed or display color of the blinker indicator displayed by the driver operating the blinker. The information generation unit 15 outputs the selected support content as support information 24 to the output unit 2.

  The storage unit 20 is a storage unit configured by a storage device such as a nonvolatile memory or a hard disk drive, and stores map information 21, history data 22, estimation information 23, and support information 24.

  Next, a processing procedure of estimation processing executed by the driving support device 1 according to the embodiment will be described with reference to FIG. FIG. 5 is a flowchart illustrating a processing procedure of estimation processing executed by the driving support device 1 according to the embodiment.

  As shown in FIG. 5, the detection unit 11 detects a preliminary operation that is performed before the driver operates the vehicle 60 (step S <b> 101). Subsequently, the acquisition unit 12 acquires peripheral information related to the surrounding situation of the vehicle 60 (step S102). In addition, about the process of step S101 and step S102, the order may be changed and you may perform simultaneously.

  Subsequently, the estimation unit 14 estimates a driving operation performed by the driver based on the preliminary motion detected by the detection unit 11 and the peripheral information acquired by the acquisition unit 12 (step S103). Subsequently, the information generation unit 15 generates support information 24 for supporting the driving operation estimated by the estimation unit 14 (step S104), and ends the process.

  Further, the driving support apparatus 1 according to the embodiment can be realized by a computer 200 having a configuration illustrated as an example in FIG. 6. FIG. 6 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the driving support device 1.

  The computer 200 includes a CPU (Central Processing Unit) 210, a ROM (Read Only Memory) 220, a RAM (Random Access Memory) 230, and an HDD (Hard Disk Drive) 240. The computer 200 also includes a media interface (I / F) 250, a communication interface (I / F) 260, and an input / output interface (I / F) 270.

  The computer 200 may include an SSD (Solid State Drive), and the SSD may execute a part or all of the functions of the HDD 240. Further, an SSD may be provided instead of the HDD 240.

  The CPU 210 operates based on a program stored in at least one of the ROM 220 and the HDD 240 and controls each unit. The ROM 220 stores a boot program executed by the CPU 210 when the computer 200 is started up, a program depending on the hardware of the computer 200, and the like. The HDD 240 stores a program executed by the CPU 210, data used by the program, and the like.

  The media I / F 250 reads programs and data stored in the storage medium 280 and provides them to the CPU 210 via the RAM 230. The CPU 210 loads such a program from the storage medium 280 onto the RAM 230 via the media I / F 250 and executes the loaded program. Alternatively, the CPU 210 executes a program using such data. The storage medium 280 is, for example, a magneto-optical recording medium such as a DVD (Digital Versatile Disc), an SD card, or a USB memory.

  The communication I / F 260 receives data from other devices via the network 290 and sends the data to the CPU 210, and transmits data generated by the CPU 210 to other devices via the network 290. Alternatively, the communication I / F 260 receives a program from another device via the network 290, sends the program to the CPU 210, and the CPU 210 executes the program.

  The CPU 210 controls the output unit 2 such as a display via the input / output I / F 270. The CPU 210 acquires data from the output unit 2 via the input / output I / F 270. Further, the CPU 210 outputs the generated data to the output unit 2 via the input / output I / F 270.

  For example, when the computer 200 functions as the driving support device 1, the CPU 210 of the computer 200 executes a program loaded on the RAM 230, thereby detecting the detection unit 11, the acquisition unit 12, the learning unit 13, the estimation unit 14, and the information. Each function of the generation unit 15 is realized.

  The CPU 210 of the computer 200 reads and executes these programs from the storage medium 280, for example. However, as another example, these programs may be acquired from other devices via the network 290. The HDD 240 can store map information 21, history data 22, estimation information 23, and support information 24 stored in the storage unit 20.

  As described above, the driving support device 1 according to the embodiment includes the detection unit 11, the learning unit 13, the acquisition unit 12, the estimation unit 14, and the information generation unit 15. The detection unit 11 detects a preliminary operation performed before the driver operates the vehicle 60. The learning unit 13 learns a driver's preliminary motion pattern based on the preliminary motion and surrounding information. The acquisition unit 12 acquires peripheral information related to the surrounding situation of the vehicle 60. The estimation unit 14 estimates a driving operation performed by the driver based on the preliminary motion detected by the detection unit 11 and the peripheral information acquired by the acquisition unit 12. The information generation unit 15 generates support information 24 for supporting the driving operation estimated by the estimation unit 14.

  The detection unit 11 includes a line-of-sight detection unit 11a. The line-of-sight detection unit 11a detects the movement of the driver's line of sight before operating the vehicle 60 as a preliminary operation.

  The acquisition unit 12 includes an object information acquisition unit 12a and a map information acquisition unit 12b. The object information acquisition unit 12a acquires object information related to objects existing around the vehicle 60 as the peripheral information. The map information acquisition unit 12b acquires map information 21 as peripheral information.

  Thereby, the driving assistance device 1 according to the embodiment can improve the estimation accuracy of the driving operation.

  Next, a modified example of the embodiment will be described with reference to FIGS. FIG. 7 is a block diagram illustrating a configuration of a driving support system S including the driving support device 1 according to a modification of the embodiment. FIG. 8 is a diagram illustrating a modified example of the estimation information 23.

  In the above-described embodiment, the case where the detection unit 11 detects the driver's line of sight as a preliminary motion has been described. For example, as illustrated in FIG. 7, the detection unit 11 acquires various types of sensors related to traveling of the vehicle 60. Preliminary movement may be detected based on signals from the group.

  The driving support device 1 according to this modification shown in FIG. 7 is the same as the driving support device 1 shown in FIG. 2 except that the detection unit 11 detects a preliminary operation from various sensor groups other than the line of sight. The same reference numerals are given and the description is omitted.

  As shown in FIG. 7, various sensor groups include, for example, a vehicle speed sensor 32 that detects the traveling speed of the vehicle 60, a door sensor 33 that detects the open / closed state of the door of the vehicle 60, and an accelerator opening degree by a driver's accelerator operation. An accelerator opening sensor 34 that detects the brake operation of the driver, an IG switch 36 that controls engine ignition of the vehicle 60, and the like. The detection unit 11 detects a preliminary operation based on each signal output from the various sensor groups. Note that the method for detecting the preliminary motion from each signal is the same as the method for detecting based on the angle θ indicating the movement of the line of sight, and thus the description thereof is omitted.

  The learning unit 13 learns a preliminary motion pattern based on the preliminary motion detected by the detection unit 11 and generates estimated information 23. Except for the estimation information 23 generated by the learning unit 13, for example, the pattern learning method and the like are the same as those in the above-described embodiment.

  Here, an example of the estimation information 23 according to the present modification will be described with reference to FIG. As illustrated in FIG. 8, the estimation information 23 according to the present modification includes driving operations such as right / left turn, start, and stop in addition to the lane change as “driving operation”. Hereinafter, a case where the estimation unit 14 estimates a right / left turn as a driving operation based on the estimation information 23 will be described.

  When the map information acquisition unit 12b acquires peripheral information indicating that the distance between the vehicle 60 and the intersection is equal to or less than a predetermined distance based on the map information 21 and the position information of the vehicle 60, the estimation unit 14 determines that the vehicle 60 If the detection unit 11 detects a deceleration operation as a preliminary operation, it is estimated that the driver performs a right / left turn driving operation. In addition, the estimation part 14 can specify the direction which a driver | operator gazes as a bending direction by detecting also the motion of a driver | operator's eyes | visual_axis other than deceleration operation | movement as preliminary operation | movement.

  Then, when it is estimated that the driving operation is a right / left turn, the information generation unit 15 generates the support information 24 that allows the blind spot of the vehicle 60 to be visually confirmed. Thereby, the driving assistance apparatus 1 can reduce that a driver | operator makes a turn without visually confirming a blind spot when turning right or left at an intersection or the like.

  Next, the start will be described. When the map information acquisition unit 12b acquires peripheral information indicating that the vehicle 60 is stopped at the parking lot based on the map information 21 and the position information of the vehicle 60, the estimation unit 14 performs the preliminary operation as the detection unit 11. When the door lock release and the engine start movement are detected in order, it is estimated that the driver performs the starting driving operation.

  And when it is estimated that driving operation is a start, the information generation part 15 produces | generates the assistance information 24 of the content which makes a driver | operator confirm the periphery visually before the start operation of the vehicle 60. FIG. Thereby, the driving assistance apparatus 1 can reduce starting in a parking lot, without a driver | operator confirming a periphery.

  Next, stop will be described. First, the map information acquisition unit 12 b acquires traffic signal information related to traffic signals existing in the traveling direction of the vehicle 60 based on the map information 21 and the position information of the vehicle 60.

  Here, the information regarding the traffic light is information indicating the distance between the traffic light and the vehicle 60 and the lighting color of the traffic light, for example. Such information may be acquired based on a captured image of the camera 40 that is mounted on the vehicle 60 and images the front of the vehicle 60, for example. In this case, the acquisition unit 12 detects the distance from the captured image of the camera 40 to the traffic light and the lighting color, and generates traffic signal information.

  The estimation unit 14 stops the driving operation based on the traffic signal information acquired by the map information acquisition unit 12b or the acquisition unit 12, the upward upward gaze detected by the detection unit 11, and the preliminary operation of the deceleration operation of the vehicle 60. It is estimated that. Specifically, based on the traffic signal information, when the lighting color of the traffic signal within a predetermined distance from the vehicle 60 is a color indicating stop, the preliminary operation in which the driver gazes at the traffic signal or the standby mode for decelerating the vehicle 60 is performed. It is presumed that the driver performs the stop operation when the operation is performed.

  Thereby, the information generation part 15 produces | generates the assistance information 24, such as an appropriate brake timing until a stop, when it is estimated that driving operation is a stop. Thereby, the driving assistance apparatus 1 can make it easy for the driver to perform a smooth braking operation when the vehicle stops before the traffic light.

  Note that the “driving operation” of the estimation information 23 according to the above-described modification is not limited to lane change, right / left turn, start, and stop, and may be any driving operation that can be estimated from the preliminary operation and the surrounding situation.

  In the above-described embodiment and the modification, the information about the surrounding situation of the vehicle 60 is described as an example of the surrounding information, but the surrounding information is not limited thereto. For example, the acquisition unit 12 may acquire route information regarding the route of the vehicle 60 from the navigation device as the peripheral information. In this case, it is assumed that the estimation unit 14 estimates the driving operation using the route information. As such route information, for example, there is route guidance for guiding a route to a predetermined destination.

  Specifically, the estimation unit 14 performs a right-turn preliminary operation such as, for example, when the route guidance acquired by the acquisition unit 12 is a predetermined distance or less to an intersection scheduled to turn right and the driver gazes at the right side mirror. In this case, it is estimated that the driver turns right at the intersection. The information generation unit 15 generates support information 24 that supports, for example, a right turn as the driving operation of the estimation result.

  On the other hand, when the driver does not perform the right turn preliminary movement even when the acquisition unit 12 acquires the above-described route guidance, the estimation unit 14 goes straight or left at the intersection, that is, a preset route. Estimate that the guidance does not turn right.

  In this case, the information generation unit 15 generates support information 24 for notifying an intersection scheduled to turn right, for example, according to the estimation result. Thereby, for example, even when the driver overlooks a right turn at the intersection indicated by the route guidance, the driver can be assisted in driving operation (right turn operation here).

  The route of the route information is not limited to the route guidance route acquired from the navigation device. For example, a travel history of the vehicle 60 may be used. In this case, for example, the acquisition unit 12 may acquire history information of a route traveled by the vehicle 60 in the past from the navigation device, and the learning unit 13 may learn a route frequently used by the driver based on the history information. . An example of a frequently used route is a commute route used by a driver when commuting.

  For example, first, the learning unit 13 learns, as the “frequently used route”, a route that the vehicle 60 has traveled a predetermined number of times or more based on the history information acquired by the acquisition unit 12 from the navigation device, and stores it as the estimation information 23. To do.

  Then, based on the estimation information 23, the estimation unit 14 is less than a predetermined distance to the intersection that makes a right turn on the “frequently used route”, and when the driver performs a right turn preliminary operation, the driver determines the intersection. Presumed to turn right.

  Information acquired by the acquisition unit 12 from the navigation device is not limited to route information. For example, the acquisition unit 12 may acquire facility information related to a specific facility from the navigation device based on the current location of the vehicle 60. In this case, the estimation unit 14 estimates the driving operation using the facility information acquired by the acquisition unit 12.

  For example, it is assumed that the acquisition unit 12 acquires facility information indicating that a commercial facility such as a gas station that the driver has used in the past is located in the vicinity of the side road on the right side of the road as the peripheral information. In this case, the estimation unit 14 estimates that the driver turns right toward the commercial facility when the driver performs the preliminary operation of watching the commercial facility and the preliminary operation of turning right.

  In addition, the acquisition unit 12 acquires information on registered points (home, company, etc.) registered by the driver stored in the navigation device as the facility information, and the estimation unit 14 drives based on the information on the registered points. The operation may be estimated.

  For example, it is assumed that the route toward the home includes an intersection that approaches the home when turning right and goes away from the home when moving straight or turning left. In this case, when the distance between the vehicle 60 and the intersection is equal to or less than a predetermined distance and the driver performs a right turn preliminary operation, the estimation unit 14 estimates that the driver turns right at the intersection in order to head home. To do.

  As described above, the estimation unit 14 can identify the route on which the driver travels by using the route information and the facility information. Therefore, the estimation unit 14 estimates the driving operation based on the identified travel route and the preliminary operation. The estimation accuracy of the operation can be improved.

  Further effects and modifications can be easily derived by those skilled in the art. Thus, the broader aspects of the present invention are not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications can be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

DESCRIPTION OF SYMBOLS 1 Driving assistance apparatus 11 Detection part 11a Gaze detection part 12 Acquisition part 12a Object information acquisition part 12b Map information acquisition part 13 Learning part 14 Estimation part 15 Information generation part

Claims (10)

A detection unit that detects a preliminary operation performed before the driver operates the moving body;
An acquisition unit for acquiring peripheral information related to the surrounding situation of the moving body;
An estimation unit that estimates a driving operation performed by the driver based on the preliminary movement detected by the detection unit and the peripheral information acquired by the acquisition unit;
An information generation unit that generates support information for supporting the driving operation estimated by the estimation unit. The detector is
A line-of-sight detection unit that detects movement of the line of sight of the driver before operating the moving body as the preliminary operation
The estimation unit includes
The driving support device according to claim 1, wherein the driving operation is estimated based on the movement of the line of sight detected by the line-of-sight detection unit and the peripheral information. A learning unit that learns the pattern of the preliminary movement of the driver based on the preliminary movement and the peripheral information;
The estimation unit includes
The driving support device according to claim 1, wherein the driving operation is estimated based on the pattern learned by the learning unit, the preliminary movement detected by the detection unit, and the peripheral information. The acquisition unit
An object information acquisition unit that acquires object information related to an object existing around the moving body as the peripheral information;
The estimation unit includes
The driving support device according to claim 1, wherein the driving operation is estimated based on the object information acquired by the object information acquisition unit and the preliminary movement. The object information acquisition unit
Obtaining a distance between the moving object and the object in front and / or behind the moving object;
The estimation unit includes
The driving support according to claim 4, wherein when the distance acquired by the object information acquisition unit is equal to or less than a predetermined distance, it is estimated whether the driving operation is a lane change based on the preliminary movement. apparatus. The acquisition unit
A map information acquisition unit that acquires map information as the peripheral information,
The estimation unit includes
4. The driving support device according to claim 1, wherein the driving operation is estimated based on the map information acquired by the map information acquisition unit, a current location of the moving body, and the preliminary operation. The map information acquisition unit
Obtaining traffic signal information relating to traffic signals present in the traveling direction of the mobile body based on the map information and the current location of the mobile body;
The estimation unit includes
The driving support device according to claim 6, wherein whether or not the driving operation is a stop is estimated based on the traffic signal information acquired by the map information acquisition unit and the preliminary operation. The estimation unit includes
The driving support device according to claim 6 or 7, wherein when the moving body is stopped in a parking lot, it is estimated whether or not the driving operation is a start based on the preliminary movement. The estimation unit includes
The driving support device according to claim 6, 7 or 8, wherein when the moving body approaches a front intersection, it is estimated whether or not the driving operation is a right / left turn based on the preliminary movement. A detection step for detecting a preliminary movement performed before the driver operates the moving body;
An acquisition step of acquiring peripheral information related to the peripheral state of the mobile body;
An estimation step of estimating a driving operation performed by the driver based on the preliminary movement detected by the detection step and the peripheral information acquired by the acquisition step;
An information generating step of generating support information for supporting the driving operation estimated by the estimating step.

Images