阅读说明：本技术 道路种类判定装置以及驾驶支援装置 (Road type determination device and driving support device ) 是由 土佐隆敏 近藤隆幸 田中信 池泽佑太 森本一广 于 2020-04-10 设计创作，主要内容包括：道路种类判定装置(20)具备：图像信息获取部(23),获取对车辆行驶的行驶路进行拍摄的图像信息；场景判定部(35),基于上述图像信息判定上述行驶路的行驶场景；一般道路似然计算部(41),基于上述行驶场景计算表示上述车辆的行驶路的种类为一般道路的一般道路似然；汽车道似然计算部(43),基于上述行驶场景计算表示上述车辆的行驶路的种类为汽车专用道路的汽车道似然；以及种类判定部(45),基于上述一般道路似然和上述汽车道似然,判定上述车辆的行驶路的种类。(A road type determination device (20) is provided with: an image information acquisition unit (23) that acquires image information obtained by imaging a travel path on which the vehicle is traveling; a scene determination unit (35) that determines a travel scene of the travel path based on the image information; a general road likelihood calculation unit (41) that calculates a general road likelihood indicating that the type of the travel path of the vehicle is a general road, based on the travel scene; a lane likelihood calculation unit (43) that calculates, based on the travel scene, a lane likelihood indicating that the type of travel path of the vehicle is an exclusive-car road; and a type determination unit (45) that determines the type of the travel path of the vehicle on the basis of the general road likelihood and the lane likelihood.)

1. A road type determination device (20) is provided with:

an image information acquisition unit (23) that acquires image information obtained by imaging a travel path on which the vehicle is traveling;

a scene determination unit (35) that determines a travel scene of the travel path based on the image information;

a general road likelihood calculation unit (41) that calculates a general road likelihood indicating that the type of the travel path of the vehicle is a general road, based on the travel scene;

a lane likelihood calculation unit (43) that calculates, based on the travel scene, a lane likelihood indicating that the type of travel path of the vehicle is an exclusive-car road; and

and a type determination unit (45) for determining the type of the travel path of the vehicle on the basis of the general road likelihood and the lane likelihood.

2. The road kind determination device according to claim 1,

the type determination unit determines the type of the travel path of the vehicle as either a general road or a vehicle-only road based on the general road likelihood and the driveway likelihood.

3. The road kind determination device according to claim 1,

the type determination unit determines the type of the traveling path of the vehicle as any one of a general road, a vehicle-specific road, and an unknown road based on the general road likelihood and the driveway likelihood.

4. The road kind determination device according to any one of claims 1 to 3,

the vehicle further includes a ramp likelihood calculation unit (42) that calculates a ramp likelihood indicating that the traveling road of the vehicle is a ramp based on the traveling scene.

5. The road kind determination device according to claim 4,

the type determination unit determines the type of the traveling path of the vehicle as any one of a general road, a motorway, and a ramp based on the general road likelihood, the motorway likelihood, and the ramp likelihood.

6. The road kind determination device according to claim 4,

the type determination unit determines the type of the traveling path of the vehicle as any one of a general road, a motorway, a ramp, and an unknown road based on the general road likelihood, the motorway likelihood, and the ramp likelihood.

7. The road kind determination device according to any one of claims 1 to 6,

when the lane likelihood is equal to or greater than a predetermined first threshold, the type determination unit determines that the travel path of the vehicle is an exclusive road.

8. The road kind determination device according to claim 7,

when the time when the lane likelihood is equal to or longer than a predetermined first threshold value is equal to or longer than a predetermined time, the type determination unit determines that the travel path of the vehicle is an exclusive road.

9. The road kind determination device according to any one of claims 1 to 8,

a landmark identifying unit (31) for identifying a landmark set as a geographical feature that is a landmark or a marker of a road of a different type based on the image information,

the scene determination unit determines the driving scene based on landmark information on landmarks identified by the landmark identification unit.

10. The road kind determination device according to any one of claims 1 to 9,

a white line recognition unit (32) for recognizing a white line on the traveling road of the vehicle based on the image information,

the scene determination unit determines the driving scene based on white line information on the white line recognized by the white line recognition unit.

11. The road kind determination device according to any one of claims 1 to 10,

further comprises an object recognition unit (33) for recognizing an object around the vehicle based on the image information,

the scene determination unit determines the driving scene based on the object information about the object recognized by the object recognition unit.

12. A driving support device (20) for performing driving support of a vehicle based on a type of a traveling path of the vehicle determined by a road type determination device according to any one of claims 1 to 11,

when it is determined that the traveling path of the vehicle is a vehicle-only path, execution of driving assistance for causing the vehicle to automatically perform a lane change is permitted,

when it is not determined that the traveling path of the vehicle is the exclusive road for the vehicle, execution of the driving assistance for automatically changing the lane of the vehicle is prohibited.

Technical Field

The present disclosure relates to a road type determination device that determines a type of a road on which a vehicle is traveling, and a driving support device that performs driving support of the vehicle based on the type of the road determined by the road type determination device.

Background

A technique of determining whether a road on which a vehicle is traveling is an expressway or a general road is known. Patent document 1 describes a road determination as to whether a road in motion is a normal road or an expressway based on the size of a traffic light recognized from two image data acquired from two cameras.

Patent document 1: japanese laid-open patent publication 2014-215698

A driving support device capable of executing a Lane Change System (LCS) for automatically changing a Lane of a vehicle is known. In general roads where pedestrians, bicycles, and the like are present, there is a risk of collision with pedestrians, bicycles, and the like present in blind spots of vehicles at the time of lane change. Therefore, there is a demand for a technique of prohibiting support by the LCS on a general road and permitting support by the LCS only on a car-dedicated road (which may be referred to as a "motorway" for short in this specification).

Patent document 1 describes a distinction between a general road and an expressway, but does not describe a distinction between a general road and an automobile lane. In addition, in the technique of patent document 1, when the traffic light cannot be detected, the type of road cannot be determined.

Disclosure of Invention

In view of the above circumstances, an object of the present disclosure is to provide a technique relating to a road type determination device that can more reliably determine the type of road on which a vehicle is traveling.

The road type determination device provided by the present disclosure includes: an image information acquisition unit that acquires image information obtained by imaging a traveling path on which a vehicle travels; a scene determination unit configured to determine a driving scene of the driving road based on the image information; a general road likelihood calculation unit that calculates a general road likelihood indicating that the type of a travel path of the vehicle is a general road, based on the travel scene; a lane likelihood calculation unit that calculates a lane likelihood indicating that the type of a travel path of the vehicle is an exclusive road based on the travel scene; and a type determination unit that determines a type of a travel path of the vehicle based on the general road likelihood and the lane likelihood.

According to the road type determination device of the present disclosure, the scene determination unit determines the travel scene of the travel path of the vehicle based on the image information acquired by the image information acquisition unit. Then, the general road likelihood calculation unit and the driveway likelihood calculation unit calculate general road likelihood and driveway likelihood, respectively, based on the determined driving scene. Then, the type determination unit determines the type of the road based on the general road likelihood and the driveway likelihood. The type of road can be determined from the general road likelihood and the driveway likelihood calculated based on the driving scene, and therefore the type of road on which the vehicle is driving can be determined more reliably.

The present disclosure also provides a driving support apparatus that performs driving support of a vehicle based on the type of the traveling path of the vehicle determined by the road type determination apparatus. The driving support device permits execution of driving support for automatically making the vehicle perform a lane change when it is determined that the traveling path of the vehicle is an exclusive-car road, and prohibits execution of driving support for automatically making the vehicle perform a lane change when it is not determined that the traveling path of the vehicle is an exclusive-car road.

According to the driving support apparatus of the present disclosure, when it is determined by the road type determination means that the traveling path of the vehicle is the lane, the execution of the LCS for automatically performing the lane change is permitted as the driving support, and when it is not determined that the lane is the lane, the execution of the LCS is prohibited. With the road type determination device of the present disclosure, it is possible to decide whether to permit or prohibit execution of LCS based on the road type determined more reliably. As a result, for example, erroneous execution of LCS by a vehicle traveling on a general road can be more reliably avoided.

Drawings

The above object and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description with reference to the accompanying drawings. The attached figures are such that,

FIG. 1 is a schematic diagram of a vehicle equipped with an ECU functioning as a road type determination device and a driving support device according to an embodiment,

fig. 2 is a flowchart of the road kind determination processing of the first embodiment,

FIG. 3 is a flowchart of the driving support process according to the first embodiment,

FIG. 4 is a view showing an example of scene determination between a motorway and a general road,

fig. 5 is a schematic diagram of the road kind determination processing of the first embodiment,

FIG. 6 is a flowchart of a road kind determination process of the second embodiment,

FIG. 7 is a diagram illustrating a form of a white line,

fig. 8 is a schematic diagram of the road kind determination processing of the second embodiment,

FIG. 9 is a flowchart of a road type determination process of the third embodiment,

fig. 10 is a schematic diagram of the road kind determination processing of the third embodiment,

fig. 11(a) is a diagram illustrating object recognition on a motorway. Fig. 11(b) is a diagram illustrating object recognition on a general road,

FIG. 12 is a flowchart of a road type determination process of the fourth embodiment,

fig. 13 is a flowchart of the road kind determination processing of the fifth embodiment,

fig. 14 is a schematic diagram of the road type determination processing of the fifth embodiment.

Detailed Description

(first embodiment)

Fig. 1 shows a vehicle control system 10 mounted on a vehicle. The vehicle control system 10 includes the imaging device 12, the ECU20, and the control object 60. The ECU20 functions as a road type determination device that determines the type of a traveling path that is a road on which the vehicle travels, and functions as a driving support device that performs driving support of the vehicle based on the type of the traveling path of the vehicle determined by the road type determination device.

The imaging device 12 may be a monocular camera such as a CCD camera, a CMOS image sensor, and a near infrared camera, or may be a stereo camera. Only one imaging device 12 may be provided in the vehicle, or a plurality of imaging devices may be provided. The imaging device 12 is attached to, for example, the vehicle at a predetermined height at the center in the vehicle width direction of the vehicle, and images an area extending in a predetermined angular range toward the front of the vehicle from an overhead viewpoint. The imaging device 12 extracts a feature point indicating the presence of an object in the captured image. As an example, edge points are extracted based on luminance information of a captured image, and the extracted edge points are subjected to hough transform. In the hough transform, points on a straight line in which a plurality of edge points are continuously arranged, and points at which straight lines are orthogonal to each other are extracted as feature points. The imaging device 12 sequentially outputs the sequentially captured images to the ECU20 as sensing information.

The ECU20 is mainly configured by a microcomputer including a CPU, a ROM, a RAM, a backup RAM, an I/O, and the like (none of which is shown), and is capable of realizing each function described in the present specification by executing various control programs stored in the ROM.

The ECU20 includes a road type determination unit 21 and a driving support unit 22. The road type determination unit 21 determines the type of the traveling road of the vehicle. The road type determination unit 21 includes an image information acquisition unit 23, an image recognition unit 30, a scene determination unit 35, a likelihood calculation unit 40, a type determination unit 45, and a road type state management unit 46.

The image information acquiring unit 23 acquires image information obtained by imaging the periphery of the vehicle from the imaging device 12. The image recognition unit 30 performs image recognition on the image information acquired by the image information acquisition unit 23. The image recognition unit 30 includes a Landmark (LM) recognition unit 31, a white line recognition unit 32, and an object recognition unit 33.

The image recognition unit 30 acquires, for example, the image information acquired by the image information acquisition unit 23, identifies the object existing around the vehicle based on the image information and dictionary information for identifying the object, which is prepared in advance, and determines the type of the object. The image recognition unit 30 compares the image information with the dictionary information by pattern matching to determine the type of the object. The image recognition unit 30 acquires the traveling direction of the vehicle and the position information of the object in the lateral direction (including the lateral width of the object) substantially orthogonal to the traveling direction based on the image information and the dictionary information. Dictionary information for identification is prepared individually from the type of an object such as a landmark, a white line, or another object, and is stored in advance in the ECU 20.

The LM identification unit 31 identifies landmark information related to landmarks by referring to a landmark identification dictionary (LM identification dictionary). Specifically, the landmark information is obtained by identifying the presence or absence of a landmark in the image information, and identifying the type, position, size, and the like of the landmark. The landmark is a geographical feature that is a landmark or a mark of a road of different types, and specifically, a road sign, a signal, an intersection, and the like that indicate an entrance of a motorway can be exemplified.

The white line recognition unit 32 extracts and connects edge points of the lane marking paint to recognize white line information on a white line on the traveling road of the vehicle. Specifically, the presence or absence of a white line in the image information is recognized, and the form thereof (specifically, the type, position, width, length, curvature, color, and the like of the white line) is recognized as white line information. Further, the ECU20 may store a recognition dictionary (road surface marker recognition dictionary) for road surface markers such as a limit marker and an instruction marker, and the white line recognition unit 32 may be configured to be able to recognize information relating to the road surface markers. The road surface identification dictionary contains information on the road surface identification pattern and the setting method determined based on the road sign, the lane line and the road identification command.

The object recognition unit 33 refers to an identification dictionary (object identification dictionary) of an object that does not correspond to a landmark, from among objects around the vehicle, and recognizes object information on the objects around the vehicle. Specifically, the type, position, size, moving speed, moving direction, and the like of the object in the image information are recognized as the object information. Specifically, the object recognition unit 33 recognizes on-road structures such as center separation belts, sound barriers, guard rails, signboards, trees, and buildings, and moving objects such as automobiles, motorcycles, bicycles, and pedestrians as objects that do not correspond to landmarks.

The scene determination unit 35 determines the driving scene of the vehicle based on a predetermined algorithm based on the result of the image recognition by the image recognition unit 30. As an example, when the signal light, pedestrian, and parked vehicle are recognized by the object recognition unit 33 while the vehicle is traveling in a vehicle speed region of 60km/h or less, it is determined as the intersection scene. Further, when the object recognition unit 33 continuously detects the rubber pillar (lane separation mark) after passing through the motorway start mark recognized by the LM recognition unit 31, it is determined as a temporary use scene. As other examples of the scene determination, lane increase/decrease, merge/branch, toll booth/Service Area (SA), temporary use, and the like are considered.

The likelihood calculation unit 40 calculates the likelihood of the road type of the traveling path of the vehicle based on the traveling scene of the vehicle determined by the scene determination unit 35. The likelihood calculation unit 40 includes a general road likelihood calculation unit 41, a Ramp Way (RW) likelihood calculation unit 42, and a driveway likelihood calculation unit 43. The likelihood means a likelihood that each road type can be calculated as a probability that the scene determined by the scene determination unit 35 can be obtained under the condition that the traveling path of the vehicle is the road type.

The general road likelihood calculation unit 41 calculates a general road likelihood Fg indicating that the type of the traveling road of the vehicle is a general road, based on the traveling scene. For example, when a result of scene determination such as "an intersection with a traffic signal", "a circular intersection (roundabout)", "a road connected to a no-entry road", "an no-entry zone", "a road with a narrow lane width", "a two-way-passing road", and "a parked vehicle on the road" is obtained, the calculated value of the general road likelihood Fg increases.

The RW likelihood calculating unit 42 calculates an RW likelihood Fr indicating that the traveling road of the vehicle is a ramp based on the traveling scene. The ramp is generally an inclined road, but in this specification, it is a ramp that is provided between a general road and a motorway and connects them. For example, when the scene determination results such as "white lines on both sides of the vehicle are solid lines" and "the traveling path of the vehicle is a sharp curve (for example, a curve of a degree smaller than R100)" are obtained, the calculated value of the RW likelihood Fr increases.

The driveway likelihood calculation unit 43 calculates a driveway likelihood Fm indicating that the type of the travel path of the vehicle is a driveway, based on the travel scene. For example, when scene determination results such as "road sign passing through a lane", "service area", "white line crossing at the time of merging into a lane", and "white line marking a lane on the road surface" are obtained, the calculated value of the lane likelihood Fm increases.

The type determination unit 45 determines the road type of the traveling path of the vehicle based on the likelihoods Fg, Fr, Fm calculated by the likelihood calculation unit 40. In the determination of the road type, the preferred type determination unit 45 is configured to determine the road type based on the general road likelihood Fg and the driveway likelihood Fm, or may be configured to determine the road type based on the RW likelihood Fr. For use in the driving support processing described later, the type determination unit 45 may be configured to be able to determine at least whether or not the type of the road is "driveway", or may be configured to be able to further determine which of "ordinary road", "ramp", and "unknown" is the case other than "driveway".

The type determination unit 45 may be configured to determine the road type based on a comparison of a predetermined first threshold a1 and a second threshold a2 set for the lane likelihood Fm, a predetermined third threshold B1 and a fourth threshold B2 set for the general road likelihood Fg, and a predetermined fifth threshold C1 and a sixth threshold C2 set for the ramp likelihood Fr. The first threshold a1 is set to a lower limit value that the driveway likelihood Fm should satisfy in order to determine a driveway, and the second threshold a2 is set to an upper limit value that the driveway likelihood Fm should satisfy in order to determine not to be a driveway. The third threshold B1 is set to a lower limit value that the general road likelihood Fg should satisfy in order to determine that the road is a general road, and the fourth threshold B2 is set to an upper limit value that the general road likelihood Fg should satisfy in order to determine that the road is not a general road. The fifth threshold C1 is set to a lower limit value that the RW likelihood Fr should satisfy in order to determine as a turn, and the sixth threshold C2 is set to an upper limit value that the RW likelihood Fr should satisfy in order to determine as not a turn.

For example, the type determination unit 45 may determine that "the driveway" is "not the driveway" when Fm ≧ A1 and "the driveway" when Fm ≦ A2. Further, it may be determined as "general road" when Fg.gtoreq.B 1, and as "not general road" when Fg.gtoreq.B 2. Further, it may be determined as "on-ramp" when Fr.gtoreq.C 1, and determined as "not on-ramp" when Fr.gtoreq.C 2. For example, the type determination unit 45 may determine that the road type having a high likelihood is the road type of the traveling path of the vehicle when the likelihood of a certain road type is high and the likelihood of another road type is low. Specifically, it may be determined that the travel path of the vehicle is the motorway when Fm ≧ A1, Fg < B2, and Fr < C2. Further, the threshold values may be set to A1. gtoreq.A 2, B1. gtoreq.B 2, and C1. gtoreq.C 2, and A1 and A2, B1 and B2, and C1 and C2 may be unified to the same value.

The type determination unit 45 may determine the road type on the condition that the magnitude of the likelihood Fg, Fr, Fm satisfies the predetermined condition is equal to or more than a predetermined time. For example, when the time Tm of Fm ≧ A1 is equal to or longer than the predetermined time Tth, it may be determined that the travel path of the vehicle is the lane. The likelihood Fg or Fr may be set under the same time condition as that of Tm ≧ Tth. By setting the time condition, the accuracy of determining the road type can be further improved.

The type determination unit 45 may prioritize or complementarily use the image recognition results of the respective units of the LM recognition unit 31, the white line recognition unit 32, and the object recognition unit 33 when using them. It is preferable that the conditions for determining the road type be appropriately changed depending on which of the landmark information, the white line information, and the object information is used to perform the scene determination.

In the case of determining the road type using the landmark information, it is also possible to make the correlation such that the calculated value of the lane likelihood Fm decreases when the calculated value of the general road likelihood Fg increases, and the calculated value of the lane likelihood Fm increases when the calculated value of the general road likelihood Fg decreases. In this case, the type determination unit 45 can determine the road type by estimating that, when the likelihood of either one of the likelihoods Fg and Fm is increased, the likelihood of the other one is decreased.

On the other hand, when the road type is determined using the white line information or the object information, the above-described correlation between the magnitudes of the likelihoods Fg and Fm is not necessarily obtained. Therefore, the preferred type determination unit 45 is configured to preferentially use the landmark information when distinguishing between the "general road" and the "motorway". Further, it is preferable that the type determination unit 45 is configured to determine the "general road", the "driveway", and the "unknown" when the landmark information cannot be used and the white line information or the object information is used. By adding a type of "unknown", it is possible to suppress erroneous determination when determining as "general road" or "driveway".

The discrimination of the "ramp" can be determined with high accuracy based on the white line information, according to the characteristics of the structure thereof. Scenes such as "the white lines on both sides of the vehicle are solid lines", "the traveling path of the vehicle is a sharp turn (for example, a curve of a degree smaller than R100)" and the like can be determined based on the white line information. In such a scene, the calculated value of the ring road likelihood Fr increases, and the calculated values of the motorway likelihood Fm and the general road likelihood Fg decrease. Therefore, the preferred type determination unit 45 is configured to preferentially use the landmark information and the white line information when distinguishing the "general road", the "motorway", and the "ramp".

The road type state management unit 46 acquires and manages the road type of the traveling path of the vehicle determined by the type determination unit 45. The type determination unit 45 may be configured to be able to read the road type managed by the road type state management unit 46 and use the read road type in the determination process of the road type.

The driving support unit 22 performs driving support of the vehicle based on the type of the traveling path of the vehicle determined by the road type determination unit 21. As the driving support, for example, a control command signal for controlling the controlled object 60 is output.

The driving support unit 22 has various driving support functions (at least an LCS function), and is configured to include an ACC unit 51, an LKA unit 52, an LCS unit 53, and a PCS unit 54 as an example, so that automatic driving and automatic parking of the vehicle can be performed according to a travel plan or the like.

The ACC unit 51 has an ACC (Adaptive Cruise Control) function, and adjusts a driving force and a braking force to Control a vehicle traveling speed and automatically maintain a target vehicle-to-vehicle distance to a preceding vehicle. The LKA unit 52 has a Lane Keeping Assist function, and generates a steering force in a direction in which an approach to a traveling Lane line is prevented, thereby automatically maintaining the traveling Lane and causing the vehicle to travel. The LCS unit 53 has an LCS (Lane Change System) function and automatically executes a Lane Change for moving the vehicle to an adjacent Lane. The PCS section 54 has a PCS (Pre-Crash Safety) function, determines whether or not an object located around the vehicle has collided with the vehicle, and automatically controls to avoid the collision with the object or reduce collision damage.

When the road type determination unit 21 determines that the traveling path of the vehicle is the lane of the vehicle, the driving support unit 22 allows the vehicle to automatically perform driving support for changing the lane by the LCS unit 53. Further, when the road type determination unit 21 determines that the traveling path of the vehicle is the lane, the driving support unit 22 prohibits the execution of the driving support for automatically causing the vehicle to change lanes by the LCS unit 53. Specifically, the road type managed by the road type state management unit 46 is read, and permission or prohibition of the automatic lane change by the LCS unit 53 is determined.

The controlled object 60 includes a driving device, a braking device, a steering device, an alarm device, and a display device. The control target 60 is configured to operate based on a control command from the ECU20 and to operate in accordance with an operation input by the driver. The operation input by the driver may be appropriately processed by the ECU20 and then input to the signal controlled object 60 as a control command to the ECU 20. The ECU20 has a function of automatically controlling a driving device, a braking device, a steering device, and the like in accordance with, for example, a traveling plan or a vehicle state of the vehicle, or for collision avoidance or lane change.

In the first embodiment, the road type determination process and the driving support process executed by the ECU20 will be described by exemplifying a case where the scene determination is performed based on the landmark information recognized by the LM recognition unit 31, the lane likelihood Fm and the general road likelihood Fg are calculated, and the road type is determined to be either "lane" or "general road". Fig. 2 shows a flowchart illustrating a road type determination process executed by the ECU 20. Fig. 3 shows a flowchart illustrating a driving assistance process executed by the ECU 20. The processing shown in fig. 2 and 3 is repeatedly executed at a predetermined cycle while the vehicle is being driven.

In step S101, image information obtained by imaging the periphery of the vehicle is acquired from the imaging device 12. Next, in step S102, the acquired image information is subjected to image recognition to recognize landmark information.

In step S103, it is determined whether or not a landmark is recognized. If a landmark is recognized, the process proceeds to step S104. If the landmark is not recognized, the process proceeds to step S113, and the road type is not updated, and the process ends.

In step S104, scene determination data corresponding to landmark information is acquired. For example, as shown in fig. 4, in the case where a road sign indicating a motorway is recognized as a landmark, scene determination data including "road sign of motorway" is acquired. For example, when "traffic light" and "intersection" are recognized as landmarks, scene determination data including "traffic light" and "intersection" is acquired.

In step S105, scene determination is performed based on the scene determination data. For example, as shown in fig. 4, it is determined that there are scenes such as "a road sign (a lane start sign) that has passed a lane", "an intersection with a traffic light", "a connection to a no-entry road", "an overtaking-prohibited section", "a characteristic sign that identifies a general road", and the like. As the characteristic mark on the general road, a low speed limit mark, a no-entry mark, a mark indicating a crossing, and the like can be exemplified.

In step S106, based on the scene determination result, the motorway likelihood Fm and the general road likelihood Fg are calculated. For example, when likelihood Fm, Fg is calculated based on a scene of "road sign of passing the driveway" shown in the left column of fig. 4, the calculated value of driveway likelihood Fm increases, and the calculated value of general road likelihood Fg decreases. For example, when the likelihoods Fm, Fg are calculated based on a scene such as "intersection with traffic light" shown in the right column of fig. 4, the calculated value of the lane likelihood Fm decreases, and the calculated value of the general road likelihood Fg increases. Since the illustrated landmarks are geographical features that are landmarks or markers of different types of roads, when the likelihoods Fm, Fg are calculated based on the landmark information, the likelihood of one of the two is increased, and the likelihood of the other is decreased.

In step S107, it is determined whether or not the driveway likelihood Fm is equal to or greater than a predetermined first threshold a 1. If Fm ≧ A1, the process proceeds to step S108. If Fm < a1, the process proceeds to step S111.

In step S108, it is determined whether or not the general road likelihood Fg is smaller than a predetermined fourth threshold value B2. If Fg < B2, the process proceeds to step S109. If Fg.gtoreq.B 2, the process proceeds to step S111.

When an affirmative determination is made in step S107 and step S108 and the process proceeds to step S109, it is determined that the road type is "motorway". Thereafter, the process proceeds to step S110, where the road type is updated to "motorway", and the process ends.

When a negative determination is made in step S107 or step S108 and the process proceeds to step S111, it is determined that the road type is "general road". Thereafter, the process proceeds to step S112, where the road type is updated to "general road" and the process ends.

When the likelihoods Fm, Fg are calculated based on the landmark information, the likelihood of one increases, and the likelihood of the other decreases. Therefore, if the set value of the first threshold value A1 is sufficiently large, Fg < B2 is clear if Fm ≧ A1, so the processing of step S108 is not necessarily required.

Fig. 3 shows a flowchart illustrating a driving assistance process executed by the ECU 20. The processing shown in fig. 3 is repeatedly executed at predetermined cycles while the vehicle is being driven.

In step S201, the road type determined by the processing shown in fig. 2 and managed by the ECU20 is acquired.

In step S202, it is determined whether the road type is "motorway". In the case of "motorway", the flow proceeds to step S203, where execution of LCS is permitted. That is, the driving support of automatically performing the lane change for moving the vehicle to the adjacent lane is permitted. If the vehicle is not "driveway", the process proceeds to step S204, and execution of LCS is prohibited. Thereafter, the process is ended.

As described above, in the road type determination process according to the first embodiment, when a landmark is recognized, the ECU20 performs scene determination based on landmark information to determine the road type and updates the stored road type. The road type is determined using the driveway likelihood Fm and the general road likelihood Fg.

Specifically, as shown in FIG. 5, in the case where Fm ≧ A1 is satisfied and Fg < B2 (condition R1-1), it is determined that the road type is "motorway". In addition, the possibility of erroneously determining as "motorway" can be reduced by setting the condition R1-1 for determining as motorway to the AND condition of Fm ≧ A1 AND Fg < B2. Therefore, the permission/prohibition of the LCS can be appropriately determined, and erroneous execution of the LCS on a vehicle traveling on a general road can be reliably avoided.

In addition, if the setting value of the first threshold A1 is set sufficiently large to calculate the likelihoods Fm, Fg based on the landmark information, Fg < B2 is set in the case where Fm ≧ A1. In such a case, it can be determined that the road type is "motorway" by satisfying Fm ≧ A1 and satisfying the condition R1-1. Further, the condition R1-2 can be satisfied by satisfying Fm < a2, and it is determined that the road type is "general road". Even if the condition R1-1 is configured to satisfy Fm ≧ A1, and the condition R1-2 is configured to satisfy Fm < A2, the road type can be appropriately determined.

Note that, when the road type determination process shown in fig. 2 is executed, the white line recognition unit 32, the object recognition unit 33, and the RW likelihood calculation unit 42 in the ECU20 are not necessarily configured in each unit.

(second embodiment)

In the second embodiment, a case will be described as an example in which scene determination is performed based on landmark information recognized by the LM recognition unit 31 and white line information recognized by the white line recognition unit 32, the lane likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr are calculated, and it is determined whether the road type is "lane", "ramp", or "general road". Fig. 6 shows a flowchart illustrating a road type determination process executed by the ECU 20. The processing shown in fig. 6 is repeatedly executed at predetermined cycles while the vehicle is being driven.

In step S301, image information obtained by photographing the periphery of the vehicle is acquired from the photographing device 12. The process advances to step S302. In step S302, image recognition is performed on the acquired image information, and landmarks and white lines are recognized. When a landmark is recognized, color, type, size, position, and the like are also recognized as landmark information. When a white line is recognized, the form of the white line is also recognized as white line information. Specifically, the width, length, curvature, color, solid line, or broken line is collectively identified, and in the case of the broken line, the interval between the broken lines, and the like. As for the length, width, and interval of the white line, for example, as shown in fig. 7, the length of the white line 70 along the traveling direction is defined as a length L, the length of the white line in the direction perpendicular to the traveling direction is defined as a width W, and the interval can be defined by the interval D between the white line 70 and the white line 71 adjacent to the white line 70 along the traveling direction. Thereafter, the process proceeds to step S304.

In step S304, scene determination data corresponding to landmark information or white line information is acquired. For example, as shown in fig. 4, in the case where a road sign indicating a motorway is recognized as a landmark, scene determination data including "road sign of motorway" is acquired. The white line has a characteristic feature in switching between a motorway, a general road, and a ramp. Scene determination data corresponding to the recognized form of the white line is acquired.

In step S305, scene determination is performed based on the scene determination data. For example, it is determined that a scene such as "a white line for merging from a ramp to a driveway, passing a road sign of the driveway", "white lines on both sides of the vehicle are solid lines having a large curvature", or "an intersection having a traffic light" is crossed.

In step S306, based on the scene determination result, the motorway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr are calculated. For example, when likelihood Fm, Fg, Fr is calculated based on a scene such as "road sign that has passed the lane", the calculated values of lane likelihood Fm and RW likelihood Fr increase, and the calculated value of general road likelihood Fg decreases. In addition, when the likelihoods Fm, Fg, and Fr are calculated based on the scene of "white lines on both sides of the vehicle are solid lines having a large curvature" after passing through the road sign of the driveway, the calculated value of the RW likelihood Fr further increases, and the calculated value of the driveway likelihood Fm decreases. In addition, when the likelihoods Fm, Fg, and Fr are calculated based on a scene in which "white lines for merging from the ramps to the driveways" are crossed after the road signs of the driveways, the calculated value of the driveway likelihood Fm further increases, and the calculated value of the RW likelihood Fr decreases. The calculated value of the general road likelihood Fg is kept small. For example, as shown in fig. 4, when the likelihoods Fm, Fg, and Fr are calculated based on scenes such as "an intersection with a traffic signal", "a connection with a no-entry road", "a no-entry section", and "a characteristic road sign recognized on a general road", the calculated value of the general road likelihood Fg increases.

In steps S307 and S308, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy the driveway determination conditions (Fm ≧ A1, Fg < B2, and Fr < C2 in the present embodiment). First, in S307, it is determined whether or not Fm ≧ A1 is satisfied. The process proceeds to S308 if the condition is satisfied, and proceeds to S311 if the condition is not satisfied. In S308, it is determined whether or not the conditions Fg < B2 and Fr < C2 are satisfied. If the condition is satisfied, the process proceeds to step S310, and it is determined that the road type is "motorway". Thereafter, the process proceeds to step S315. If the condition in S308 is not satisfied, the process proceeds to step S309.

In step S309, it is determined whether or not the duration Tm in which the state satisfying Fm.gtoreq.A 1 continues is equal to or greater than a predetermined threshold value Tth. When Tm is not less than Tth, the process proceeds to step S310, and it is determined that the road type is "motorway". If Tm < Tth, the process proceeds to step S311.

In step S311, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy the ramp determination conditions (Fm < A2, Fg < B2, and Fr ≧ C1 in the present embodiment), respectively. If all of the three conditions of Fm < A2, Fg < B2, and Fr ≧ C1 are satisfied, the routine proceeds to step S312, and it is determined that the road type is "ramp". Thereafter, the process proceeds to step S315. If any of the three conditions Fm < A2, Fg < B2, and Fr ≧ C1 is not satisfied, the routine proceeds to step S313.

In step S313, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy general road determination conditions (Fm < A2, Fg ≧ B1, and Fr < C2 in the present embodiment), respectively. If all of the three conditions of Fm < A2, Fg ≧ B1, and Fr < C2 are satisfied, the routine proceeds to step S314, and it is determined that the road type is "general road". Thereafter, the process proceeds to step S315. If any of the three conditions Fm < A2, Fg ≧ B1, and Fr < C2 is not satisfied, the process proceeds to step S316.

In step S315, the road type is updated based on the road types determined in steps S310, S312, and S314, and the process ends. In step S316, the road type is not updated, and the process ends.

As described above, in the road type determination process according to the second embodiment, when a landmark or a white line is recognized, the ECU20 performs scene determination to determine the road type and updates the stored road type.

Since the presence or absence of a road sign indicating a driveway can be recognized as landmark information, it is possible to accurately distinguish a general road, a ramp, or a driveway. Further, since the form of the white line can be recognized as the white line information, the lane and the driveway can be accurately discriminated. Therefore, in practice, as shown in FIG. 8, for example, the state of "ordinary road" is shifted to the state of "ramp" because the condition R2-1 (Fr ≧ C1 and Fg < B2) is satisfied. In addition, the state of the "motorway" is shifted to the state of the "ramp" because the condition R2-4 (Fr ≧ C1 and Fm < A2) is satisfied. Further, the state of "ramp" is shifted from the state of "ramp" to the state of "motorway" because the condition R2-2 (Fm. gtoreq.A 1 and Fr < C2) is satisfied, and the state of "ramp" is shifted from the state of "ramp" to the state of "general road" because the condition R2-3 (Fg. gtoreq.B 1 and Fr < C2) is satisfied. As shown in conditions R2-1 to R2-4 shown in fig. 8, even if the road type is determined based on some of the conditions in steps S307, S308, S311, and S313 of fig. 6, the road type can be accurately determined.

The road type is determined to be any one of "driveway", "general road", and "ramp" using the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr. Therefore, it is possible to perform driving assistance such as vehicle speed control required to enter the motorway from the ramp corresponding to the road type.

In addition, as shown in step S309, even when the lane determination condition (Fm ≧ A1, Fg < B2, and Fr < C2) is not satisfied, if the continuation time Tm for which the state of Fm ≧ A1 continues is equal to or greater than a predetermined threshold Tth, it is determined as "lane". Thus, for example, even in the case of a road shape or a white line shape in which the boundaries between a general road and a ramp or a motorway are unclear, it is possible to determine a motorway without fixing the determination result of the road type to the previous value. In the case of the ramp determination in step S311, the conditions may be sequentially performed in the same manner as in steps S307 and S308: Fr.gtoreq.C 1, and the conditions: determination of Fm < A2 and Fg < B2, and the same as step S309, is made so as to satisfy the condition: the duration of the state continuation of the condition Fr ≧ C1 is determined by comparison with a threshold. In the general road determination in step S313, the conditions: fg is not less than B1, and the condition is as follows: determination of Fm < A2 and Fr < C2, and the same as step S309, is made so as to satisfy the condition: the duration of the state continuation of the condition Fg ≧ B1 is determined by comparison with a threshold value.

Note that, when the road type determination process shown in fig. 6 is executed, the object recognition unit 33 in the ECU20 is not necessarily configured. In another embodiment, the time condition shown in step S309 can be used as a condition for determining the road type, thereby improving the accuracy of determining the road type.

(third embodiment)

In the third embodiment, a case will be described in which scene determination is performed based on white line information recognized by the white line recognition unit 32, the lane likelihood Fm and the general road likelihood Fg are calculated, and the road type is determined to be any of "lane", "general road", and "unknown". Fig. 9 shows a flowchart illustrating a road type determination process executed by the ECU 20. The processing shown in fig. 9 is repeatedly executed at predetermined cycles while the vehicle is being driven.

In step S401, image information obtained by photographing the periphery of the vehicle is acquired from the photographing device 12. Next, in step S402, the acquired image information is subjected to image recognition, and the presence or absence of a white line and its form are recognized as white line information.

In step S404, scene determination data corresponding to the white line information is acquired. In step S405, scene determination is performed based on the scene determination data. For example, it is determined as a scene such as "traveling on a road with a narrow lane width", "traveling on a two-way road", "traveling on a white-lined road indicating a motorway", and the like. It is possible to distinguish whether the white line is a white line of a motorway or a white line of a general road based on the length L, width W, and interval D of the white line included in the white line information.

In step S406, based on the scene determination result, the motorway likelihood Fm and the general road likelihood Fg are calculated. For example, when the likelihoods Fm, Fg are calculated based on the scenes such as "traveling on a road with a narrow lane width" and "traveling on a two-way road", the calculated value of the lane likelihood Fm decreases and the calculated value of the general road likelihood Fg increases. Further, for example, when likelihood Fm, Fg is calculated based on a scene "traveling on a road marked with a white line of a lane", the calculated value of lane likelihood Fm increases and the calculated value of general road likelihood Fg decreases.

In step S407, it is determined whether or not the driveway likelihood Fm and the general road likelihood Fg satisfy the driveway determination conditions (Fm ≧ A1 and Fg < B2 in the present embodiment). When both the conditions of Fm ≧ A1 and Fg < B2 are satisfied, the routine proceeds to step S409, and it is determined that the road type is "motorway". Thereafter, the process proceeds to step S418. If at least one of the two conditions Fm ≧ A1 and Fg < B2 is not satisfied, the routine proceeds to step S411.

In step S411, it is determined whether or not the driveway likelihood Fm and the general road likelihood Fg satisfy general road determination conditions (Fm < A2 and Fg ≧ B1 in the present embodiment). When both the conditions Fm < A2 and Fg ≧ B1 are satisfied, the routine proceeds to step S412, and it is determined that the road type is "general road". Thereafter, the process proceeds to step S418. If at least one of the two conditions Fm < A2 and Fg ≧ B1 is not satisfied, the routine proceeds to step S413.

In step S413, it is determined whether or not the driveway likelihood Fm and the general road likelihood Fg satisfy the first unclear condition (in the present embodiment, a2 ≦ Fm < a1 and B2 ≦ Fg < B1), respectively. When both of the conditions of A2. ltoreq. Fm < A1 and B2. ltoreq. Fg < B1 are satisfied, the routine proceeds to step S414, and it is determined that the road type is "unknown". Thereafter, the process proceeds to step S418. If at least one of the two conditions of A2. ltoreq. Fm < A1 and B2. ltoreq. Fg < B1 is not satisfied, the flow proceeds to step S415.

In step S415, it is determined whether or not the driveway likelihood Fm and the general road likelihood Fg satisfy the second unclear condition (Fm < a2 and Fg < B2 in the present embodiment), respectively. When both the conditions Fm < a2 and Fg < B2 are satisfied, the routine proceeds to step S414, and it is determined that the road type is "unknown". Thereafter, the process proceeds to step S418. If at least one of the two conditions Fm < a2 and Fg < B2 is not satisfied, the routine proceeds to step S416, and it is determined that the result is unknown. When it is determined to be "unknown" in step S416, this corresponds to a case where the determination result is particularly blurred, compared to a case where it is determined to be "unknown" in step S414. Therefore, when it is determined that "unknown" is present in step S416, the process may proceed to step S418 as shown in fig. 9, or the road type may not be updated as in step S113 of fig. 2 of the first embodiment.

In step S418, the road type is updated based on the road type determined in steps S409, S412, S414, and S416, and the process is terminated.

As described above, in the road type determination process according to the third embodiment, the ECU20 performs scene determination based on white line information, determines the road type, and updates the stored road type.

The road type is determined using the driveway likelihood Fm and the general road likelihood Fg. Specifically, as shown in FIG. 10, when the condition R3-2 (Fm ≧ A1 AND Fg < B2) set as the AND condition is satisfied, it is determined that the road type is "motorway". When the condition R3-3 (Fm < A2 AND Fg ≧ B1) set as the AND condition is satisfied, it is determined that the road type is "ordinary road". Further, when the condition R3-1 or the condition R3-4 is satisfied, that is, when either one of the conditions R3-2 and R3-3 is not satisfied, it is determined that the road type is "unknown".

When the white line information recognized by the white line recognition unit 32 is used, it is determined as "motorway" or "general road" when the conditions R3-2 AND R3-3 set as AND conditions are satisfied, respectively. When either of the conditions R3-2 and R3-3 is not satisfied, it is determined that the road type is "unknown". This can ensure the accuracy of determining the road type, and reduce the possibility of an erroneous determination as "motorway" and the possibility of an erroneous determination as "general road". Therefore, the permission/prohibition of the LCS can be appropriately determined, and erroneous execution of the LCS on a vehicle traveling on a general road can be reliably avoided.

Note that, when the road type determination process shown in fig. 9 is executed, the LM recognition unit 31, the object recognition unit 33, and the RW likelihood calculation unit 42 in the ECU20 are not necessarily configured in each unit.

(fourth embodiment)

In the fourth embodiment, a case will be described in which scene determination is performed based on object information around a vehicle recognized by the object recognition unit 33, the lane likelihood Fm and the general road likelihood Fg are calculated, and the road type is determined to be any of "lane", "general road", and "unknown".

The object recognition unit 33 recognizes, as object information, the type, size, position, moving direction, moving speed, and the like of an object other than the object recognized as a landmark by the LM recognition unit 31.

For example, as shown in fig. 11(a), when the center separation zone 72 and the sound insulating wall 73 are recognized as objects, it can be determined that the object is a scene related to a driveway. As shown in fig. 11(b), when the vehicle 80, the pedestrian 81, the parked vehicle 82 on the road, or the like is recognized as the object, it can be determined as a scene related to the general road.

Fig. 12 shows a flowchart illustrating a road type determination process executed by the ECU 20. The processing shown in fig. 12 is repeatedly executed at predetermined cycles while the vehicle is being driven.

In step S501, image information obtained by photographing the periphery of the vehicle is acquired from the photographing device 12. Next, in step S502, image recognition is performed on the acquired image information to recognize object information.

In step S503, it is determined whether or not an object that does not correspond to the landmark (an object stored in the object identification dictionary) is identified around the vehicle. If such an object is recognized, the process proceeds to step S504. If not, the process is ended.

In step S504, scene determination data corresponding to the object information is acquired. For example, as shown in fig. 11(a), scene determination data including a center separation band, a sound insulating wall, a bicycle, a parked vehicle on the road, a pedestrian, and the like is acquired.

In step S505, scene determination is performed based on the scene determination data. For example, the vehicle is determined to be in a scene such as "traveling on a lane between the sound insulating wall and the center separation belt", "traveling on a road on which a bicycle travels", "traveling on a road on which a parked vehicle is present" or the like.

In step S506, based on the scene determination result, the motorway likelihood Fm and the general road likelihood Fg are calculated. For example, when the likelihoods Fm, Fg are calculated based on a scene "traveling at a high speed region of 80km/h or more on the lane between the sound insulating wall and the center separation band", the calculated value of the lane likelihood Fm increases and the calculated value of the general road likelihood Fg decreases. And calculating the motorway likelihood Fm. Further, for example, when likelihood Fm, Fg is calculated based on a scene such as "traveling on a road on which a bicycle is traveling", the calculated value of lane likelihood Fm decreases, and the calculated value of general road likelihood Fg increases.

In step S507, it is determined whether or not the driveway likelihood Fm and the general road likelihood Fg satisfy the driveway determination conditions (Fm ≧ A1 and Fg < B2 in the present embodiment). When both the conditions of Fm ≧ A1 and Fg < B2 are satisfied, the routine proceeds to step S509, and it is determined that the road type is "motorway". Thereafter, the process proceeds to step S518. If at least one of the two conditions Fm ≧ A1 and Fg < B2 is not satisfied, the routine proceeds to step S511.

In step S511, it is determined whether or not the driveway likelihood Fm and the general road likelihood Fg satisfy general road determination conditions (Fm < A2 and Fg ≧ B1 in the present embodiment). When both the conditions Fm < A2 and Fg ≧ B1 are satisfied, the routine proceeds to step S512, and it is determined that the road type is "general road". Thereafter, the process proceeds to step S518. If at least one of the two conditions Fm < A2 and Fg ≧ B1 is not satisfied, the process proceeds to step S513.

In step S513, it is determined whether the driveway likelihood Fm and the general road likelihood Fg satisfy the first unclear condition (in the present embodiment, a2 ≦ Fm < a1 and B2 ≦ Fg < B1), respectively. When both of the conditions of A2. ltoreq. Fm < A1 and B2. ltoreq. Fg < B1 are satisfied, the routine proceeds to step S514, and it is determined that the road type is "unknown". Thereafter, the process proceeds to step S518. If at least one of the two conditions of A2. ltoreq. Fm < A1 and B2. ltoreq. Fg < B1 is not satisfied, the flow proceeds to step S515.

In step S515, it is determined whether or not the driveway likelihood Fm and the general road likelihood Fg satisfy the second unclear condition (Fm < a2 and Fg < B2 in the present embodiment), respectively. When both the conditions Fm < a2 and Fg < B2 are satisfied, the process proceeds to step S514, and it is determined that the road type is "unknown". Thereafter, the process proceeds to step S518. If at least one of the two conditions Fm < a2 and Fg < B2 is not satisfied, the routine proceeds to step S516, where it is determined that the result is unknown. When it is determined to be "unknown" in step S516, the determination result is particularly more blurred than when it is determined to be "unknown" in step S514. Therefore, when it is determined that "unknown" is present in step S516, the process may proceed to step S518 as shown in fig. 12, or the road type may not be updated as in step S113 of fig. 2 of the first embodiment.

In step S518, the road type is updated based on the road type determined in steps S509, S512, S514, and S516, and the process ends.

As described above, in the road type determination process according to the fourth embodiment, the ECU20 performs scene determination based on the object information, determines the road type, and updates the stored road type.

As in the third embodiment, as shown in fig. 10, the road type is determined as "motorway" or "general road" when the conditions R3-2 AND R3-3 set as AND conditions are satisfied, respectively. When either of the conditions R3-2 and R3-3 is not satisfied, it is determined that the road type is "unknown". This can ensure the accuracy of determining the road type, and reduce the possibility of an erroneous determination as "motorway" and the possibility of an erroneous determination as "general road". Therefore, the permission/prohibition of the LCS can be appropriately determined, and erroneous execution of the LCS on a vehicle traveling on a general road can be reliably avoided.

Note that, when the road type determination process shown in fig. 12 is executed, the LM recognition unit 31, the white line recognition unit 32, and the RW likelihood calculation unit 42 in the ECU20 are not necessarily configured in each unit.

(fifth embodiment)

In the fifth embodiment, a case will be exemplified in which scene determination is performed based on landmark information recognized by the LM recognition unit 31, white line information recognized by the white line recognition unit 32, and object information around the vehicle recognized by the object recognition unit 33, the lane likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr are calculated, and the road type is determined as any one of "lane", "general road", "ramp", and "unclear". Fig. 13 shows a flowchart illustrating a road type determination process executed by the ECU 20. The processing shown in fig. 13 is repeatedly executed at predetermined cycles while the vehicle is being driven.

In step S601, image information obtained by photographing the periphery of the vehicle is acquired from the photographing device 12. Next, in step S602, the acquired image information is subjected to image recognition, and landmark information, white line information, and object information are recognized. Then, in step S604, scene determination data corresponding to landmark information, white line information, and object information is acquired.

In step S605, scene determination is performed based on the scene determination data. For example, as described in the first to fourth embodiments, it is determined that the vehicle is in a scene such as "a road sign passing a lane", "a white line crossing a junction from a ramp to a lane", "an intersection with a traffic light", "a connection to a no-entry road", "a no-entry section", "a characteristic sign on a general road", "traveling on a lane between a sound-insulating wall and a center separation band", "traveling on a road on which a bicycle travels", and the like. The scene determination may be performed using the landmark information, the white line information, and the object information in a complementary manner, or may be performed with the landmark information prioritized over the white line information or the object information.

In step S606, the motorway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr are calculated based on the scene determination result. The magnitudes of likelihoods Fm, Fg, and Fr calculated based on the scenes are the same as those in the first to fourth embodiments, and therefore the description thereof is omitted.

In step S607, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy the driveway determination conditions (Fm ≧ A1 and Fg < B2 and Fr < C2 in the present embodiment), respectively. If all of the three conditions of Fm ≧ A1, Fg < B2, and Fr < C2 are satisfied, the routine proceeds to step S610, and it is determined that the road type is "motorway". Thereafter, the process proceeds to step S619. If any of the three conditions Fm.gtoreq.A 1, Fg < B2, and Fr < C2 is not satisfied, the process proceeds to step S608.

In step S608, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy the ramp determination conditions (Fm < A2, Fg < B2, and Fr ≧ C1 in the present embodiment), respectively. If all of the three conditions of Fm < A2, Fg < B2, and Fr ≧ C1 are satisfied, the routine proceeds to step S611, and it is determined that the road type is "ramp". Thereafter, the process proceeds to step S619. If any of the three conditions Fm < A2, Fg < B2, and Fr ≧ C1 is not satisfied, the routine proceeds to step S612.

In step S612, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy general road determination conditions (Fm < A2, Fg ≧ B1, and Fr < C2 in the present embodiment), respectively. If all of the three conditions of Fm < A2, Fg ≧ B1, and Fr < C2 are satisfied, the routine proceeds to step S617, and it is determined that the road type is "general road". Thereafter, the process proceeds to step S619. If any of the three conditions Fm < A2, Fg ≧ B1, and Fr < C2 is not satisfied, the routine proceeds to step S613.

In step S413, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy the first unclear condition (in the present embodiment, a2 ≦ Fm < a1, B2 ≦ Fg < B1, and C2 ≦ Fr < C1), respectively. When all of the three conditions of a2 ≦ Fm < a1, B2 ≦ Fg < B1, and C2 ≦ Fr < C1 are satisfied, the routine proceeds to step S615, and it is determined that the road type is "unknown". Thereafter, the process proceeds to step S619. If any of the three conditions of A2. ltoreq. Fm < A1, B2. ltoreq. Fg < B1, and C2. ltoreq. Fr < C1 is not satisfied, the flow proceeds to step S614.

In step S614, it is determined whether or not the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr satisfy the second unclear condition (in the present embodiment, Fm < a2 and Fg < B2 and Fr < C2), respectively. When all of the three conditions of Fm < a2, Fg < B2, and Fr < C2 are satisfied, the routine proceeds to step S615, and it is determined that the road type is "unknown". Thereafter, the process proceeds to step S619. If any of the three conditions Fm < a2, Fg < B2, and Fr < C2 is not satisfied, the routine proceeds to step S618, where it is determined as unknown. When it is determined to be "unknown" in step S618, the determination result is particularly more blurred than when it is determined to be "unknown" in step S615. Therefore, when it is determined that "unknown" is present in step S618, the process may proceed to step S619 as shown in fig. 13, or the road type may not be updated as in step S113 of fig. 2 of the first embodiment.

In step S619, the road type is updated based on the road type determined in steps S610, S615, S617, and S618, and the process ends.

As described above, in the road type determination process according to the fifth embodiment, the ECU20 performs scene determination based on the landmark information recognized by the LM recognition unit 31, the white line information recognized by the white line recognition unit 32, and the object information around the vehicle recognized by the object recognition unit 33, determines the road type, and updates the stored road type.

The road type is determined using the driveway likelihood Fm, the general road likelihood Fg, and the RW likelihood Fr. As shown by the solid line in FIG. 14, when the road type is changed between "general road", "ramp", and "driveway", if the condition R4-2 is satisfied (Fm ≧ A1 and Fr < C2), the "ramp" is changed to "driveway". When the condition R4-4 (Fg. gtoreq.B 1 and Fr < C2) is satisfied, the "ramp" is changed to the "ordinary road". Further, when the conditions R4-1 (Fr. gtoreq.C 1 and Fg < B2) and R4-3 (Fr. gtoreq.C 1 and Fm < A2) are satisfied, the road type is changed from "ordinary road" or "motorway" to "ramp".

Further, as shown by the broken line in FIG. 14, when the condition R4-6 (Fm. gtoreq.A 2, Fg < B1, OR Fr. gtoreq.C 2) OR the condition R4-8 (Fm < A2, Fg. gtoreq.B 1, OR Fr. gtoreq.C 1) set as the OR condition is satisfied, the road type is determined as "unknown". Further, when the condition R4-5 (Fg. gtoreq.B 1, Fm < A2, AND Fr < C2) set as the AND condition is satisfied, the road type is changed from "unknown" to "general road". Further, the road type is changed from "unknown" to "motorway" when the condition R4-7 (Fm ≧ A1 AND Fg < B2 AND Fr < C2) set as the AND condition is satisfied.

This makes it possible to accurately determine whether the traveling road of the vehicle is a "driveway", "a ramp", or a "general road", and to reduce the possibility of an erroneous determination as a "driveway" or an erroneous determination as a "general road", and to reduce the possibility of an erroneous determination as a "ramp". Therefore, the driving assistance corresponding to each road type can be appropriately performed. Further, the permission/prohibition of the LCS can be appropriately determined, and erroneous execution of the LCS on a vehicle traveling on a general road can be reliably avoided.

According to the above embodiments, the following effects can be obtained.

The ECU20 includes a road type determination unit 21 including an image information acquisition unit 23, an image recognition unit 30, a scene determination unit 35, a likelihood calculation unit 40, and a type determination unit 45. The image information acquiring unit 23 acquires image information obtained by imaging a traveling path on which the vehicle travels. The scene determination unit 35 determines the driving scene of the driving path of the vehicle based on the image recognition result of the image information by the image recognition unit 30. The general road likelihood calculation unit 41 calculates a general road likelihood Fg indicating that the type of the traveling road of the vehicle is a general road, based on the traveling scene. The driveway likelihood calculation unit 43 calculates a driveway likelihood Fm indicating that the type of the travel path of the vehicle is a driveway, based on the travel scene. The type determination unit 45 determines the type of the travel path of the vehicle based on the general road likelihood Fg and the driveway likelihood Fm. Since the road type can be determined from the general road likelihood Fg and the driveway likelihood Fm calculated based on the traveling scene, the type of the road on which the vehicle travels can be determined more reliably.

The ECU20 further includes a RW likelihood calculating unit 42 that calculates an RW likelihood Fr indicating that the travel path of the vehicle is a ramp, from the travel scene. Thus, the road type can be determined as "ramp" in addition to "ordinary road" and "motorway".

The type determination unit 45 may be configured to determine the type of the travel path of the vehicle as either one of the "general road" and the "automobile-dedicated road" based on the general road likelihood Fg and the driveway likelihood Fm.

The type determination unit 45 may be configured to determine the type of the travel path of the vehicle as any one of "general road", "automobile-dedicated road", and "unknown" based on the general road likelihood Fg and the driveway likelihood Fm.

The type determination unit 45 may be configured to determine the type of the travel path of the vehicle as any one of the "general road", "automobile-dedicated road", and "ramp" based on the general road likelihood Fg, the driveway likelihood Fm, and the ramp likelihood Fr.

The type determination unit 45 may be configured to determine the type of the travel path of the vehicle as any one of "general road", "automobile-dedicated road", "ramp", and "unclear", based on the general road likelihood Fg, the driveway likelihood Fm, and the ramp likelihood Fr.

The type determination unit 45 may be configured to determine the road type based on a comparison with predetermined thresholds set for the general road likelihood Fg, the driveway likelihood Fm, and the ramp likelihood Fr. For example, the type determination unit 45 may determine the road type having a high likelihood as the road type of the traveling path of the vehicle. Specifically, for example, when the lane likelihood Fm is equal to or greater than the predetermined first threshold value a1, it may be determined that the travel path of the vehicle is a lane. For example, the type determination unit 45 may determine the road type having a high likelihood as the road type of the traveling path of the vehicle when the likelihood of a certain road type is high and the likelihood of another road type is low. Specifically, for example, when the lane likelihood Fm is equal to or greater than a predetermined first threshold a1 and the general road likelihood Fg is smaller than a predetermined fourth threshold B2, it may be determined that the travel path of the vehicle is the lane. For example, when the lane likelihood Fm is equal to or greater than the predetermined first threshold a1, the general road likelihood Fg is smaller than the predetermined fourth threshold B2, and the RW likelihood Fr is smaller than the predetermined sixth threshold C2, it may be determined that the traveling path of the vehicle is a lane.

Preferably, the image recognition unit 30 includes at least one of a landmark recognition unit 31, a white line recognition unit 32, and an object recognition unit 33. Preferably, the scene determination unit 35 is configured to determine the driving scene based on at least one of landmark information, white line information, and object information. That is, it is preferable that the scene determination in the scene determination unit 35 be performed by prioritizing or complementarily using landmark information, white line information, and object information of an object that does not correspond to a landmark.

The ECU20 includes a drive assisting unit 22 that performs drive assistance for the vehicle based on the type of the traveling path of the vehicle determined by the road type determining unit 21. Preferably, the drive support unit 22 includes an LCS unit 53 that performs drive support for automatically changing the lane of the vehicle, and allows the LCS unit 53 to perform the lane change when it is determined that the traveling path of the vehicle is the motorway. Further, the execution of the lane change by the LCS unit 53 may be prohibited when it is not determined that the traveling path of the vehicle is the lane. Based on the road type determined more reliably by the road type determination unit 21, it is possible to determine whether to permit or prohibit execution of the lane change by the LCS unit 53. As a result, for example, it is possible to reliably avoid erroneous execution of LCS by a vehicle traveling on a general road.

A sensor other than the imaging device 12 may be mounted on the vehicle. For example, sensors such as a radar sensor, a vehicle speed sensor, a yaw rate sensor, a steering angle sensor, an acceleration sensor, and a gyro sensor may be mounted. The ECU20 may be configured to acquire detection information of a sensor group other than the imaging device 12 and use the detection information in the determination process by the road type determination unit 21 and the driving support by the driving support unit 22.

The drive support unit 22 may be configured to change the drive support processing in the ACC unit 51, the LKA unit 52, and the PCS unit 54 in accordance with the road type determined by the road type determination unit 21. The driving support unit 22 may have a function of driving support not shown in fig. 1, or may be configured to change the processing of the driving support based on the road type determined by the road type determination unit 21.

The relationships among the likelihoods Fm, Fg, Fr, the travel scene, the landmark information, the white line information, and the object information, which are exemplified and described in the present specification, are merely examples, and can be appropriately adjusted according to the regulations and the environment of each country. For example, a road sign indicating an exit of a motorway, or a road sign indicating an entrance or exit of an expressway may be set as the landmark.

In the flowcharts of the respective embodiments, the determination process in which the plurality of conditions are set for the magnitudes of the likelihoods Fm, Fg, and Fr may be performed by independently determining a part or all of the plurality of conditions. For example, the flow chart may be configured to separately determine the conditions of Fm < A2, Fg < B2, Fr ≧ C1, and the conditions of Fm < A2, Fg < B2, and Fr ≧ C1, which are the three conditions shown in step S311 of FIG. 6, or may be divided into determination of Fr ≧ C1, and determination of Fm < A2, and Fg < B2.

The present disclosure has been described in terms of embodiments, but it should be understood that the present disclosure is not limited to the embodiments and configurations. The present disclosure also includes various modifications and equivalent variations. In addition, various combinations and modes including only one element, and other combinations and modes above or below the element are also within the scope and spirit of the present disclosure.