Driving assistance device

Abstract

Driving assistance device (15) which is installed on a vehicle and operational, to provide driving assistance information to a driver, comprising: a white line detection element (151) for detecting a lane marking which is attached to at least one right side and one left side of a lane in which the vehicle is traveling; a fork determination part (152) for determining whether or not there is a fork point ahead of the vehicle on a road on which the vehicle is traveling, based on information indicating the vehicle's position and map information; a direction estimation part (154) for estimating the direction of travel of the vehicle at the fork in the road; a virtual lane marking estimator part (155) for generating, at the fork point on the road, a virtual lane marking that is used to output the driving assistance information to cross a forked road that is not in the estimated direction of travel, based on the direction of travel of the vehicle estimated by the direction estimator part in a case where the fork determination part determines that the fork point is present on the road; and an information output part (158) for outputting the driving assistance information as an alarm about the deviation from the lane in which the vehicle is traveling, in a case where it is determined that the probability of a deviation from the lane is high, based on a position of the virtual lane marking estimated by the virtual lane marking estimator part and the lane marking detected by the white line detection part.

Description

BACKGROUND OF THE INVENTIONField of the invention

[0001] The present invention relates to a driving assistance device installed on, for example, a vehicle and operable to output driving assistance information to a driver. Description of the background technology

[0002] Conventionally, a technique has been known in which lane markings (so-called white lines) provided on the right and left sides of a lane in which a vehicle is traveling are detected using a CCD (charge-coupled device) camera or the like, and driving assistance information is reported to a driver based on the detected lane markings to prevent lane departure. However, in a case where the lane markings cannot be detected, the driving assistance information may not be reported accurately or without error. To solve this problem, various devices, methods, and the like have been proposed (see, for example, JP 2003-44836 A).

[0003] A dividing line detection device disclosed in JP 2003-44836 A includes: an image processing part for extracting white lines (corresponding to lane markings) painted on a road surface from an image captured by a camera; and a white line estimation part for estimating, when the image processing part extracts one of the white lines on the right or left side, a position of the other of the white lines not extracted, based on coordinates of a plurality of sampling points set on the extracted white line, road width calculation parameters corresponding to the sampling points, and road width data of the lane on which the vehicle is traveling.This dividing line detection device enables the other dividing line to be estimated from the one dividing line (white line or the like) that could be detected.

[0004] However, since the dividing line detection device disclosed in JP 2003-44836A estimates the position of the lane marking on the non-detected side based on the road width data of the lane in which the vehicle is traveling, etc., the position of the lane marking on the non-detected side may not be detected accurately. Specifically, the dividing line detection device disclosed in JP 2003-44836A estimates the position of the lane marking on the non-detected side based on the coordinates of the plurality of sampling points set on the extracted lane marking, the lane width calculation parameters corresponding to the sampling points, and the road width data of the lane in which the vehicle is traveling.Therefore, the position of the lane marking on the side that is not detected cannot be estimated accurately or without error in a case where the road width calculation parameters, the road width data, and the like are not accurate or without error.

[0005] Furthermore, even in a state where the lane marking can be accurately detected, there may be a case where it is preferable that the position of the lane marking used for the purpose of preventing lane departure, for example, be set to be further inward than the position estimated based on the position of the detected lane marking (namely, the actual lane marking).In particular, in a case where a sidewall made of concrete or the like, which is assumed to cause disastrous damage to a vehicle if the vehicle comes into contact with the sidewall, is installed outside the detected lane marking, it is preferable that the position of the lane marking used for the purpose of preventing the lane deviation be set to be further inward than the position of the detected lane marking.

[0006] From EP 1 959 236 A1, a lane determination device is known, comprising: an image information acquisition unit that obtains image information of a lane marking; a vehicle position information acquisition unit that obtains vehicle position information indicating a current position of a vehicle; a road information acquisition unit that obtains road information containing position and shape information regarding the lane marking around the vehicle; an image recognition unit that recognizes an image of the lane marking in the image information;and a lane determination unit that determines a vehicle lane on which the vehicle is traveling based on the image recognition result and the road information, and determines the vehicle lane based on an imaging order, shapes, and positions of a specific lane marker that is imaged after recognizing that a branch point exists in a forward direction of the vehicle and the lane marker is not imaged at all or not partially imaged in the image information and the road information.

[0007] From EP 1 089 231 A2, a lane recognition device is known which is configured to recognize lane markings on a road surface based on a pair of stereo images of a scene in front of a host vehicle obtained by a stereoscopic camera, and which comprises: a stereo image processing device for calculating distance information of an image based on the pair of images, a lane detection device for detecting a lane marking of a lane in the image based on brightness information of the image and distance information calculated by the stereo image processing device, and a lane position detection device for detecting a position of the lane marking in real space based on a position of the lane marking detected by the lane detection device and distance information calculated by the stereo image processing device. SUMMARY OF THE INVENTION

[0008] The present invention was made in view of the above-described situation. An object of the present invention is to provide a driving assistance device capable of appropriately estimating lane marking positions and outputting appropriate driving assistance information.

[0009] To achieve the aforementioned object, the present invention provides a driving assistance device as set forth in the claims. The driving assistance device according to the invention is particularly directed to the seventeenth to twenty-sixth aspects of the present disclosure, and thus to the fifth embodiment described in detail below.

[0010] Notwithstanding the foregoing, the present disclosure has the following features.

[0011] A first aspect of the present disclosure is directed to a driving assistance device installed on a vehicle and operable to output driving assistance information to a driver, comprising: a white line detection part for detecting a lane marking provided on at least one of a right side and a left side of a lane in which the vehicle is traveling; a roadside object detection part for detecting a roadside object on the other of the right side and the left side in a case where one of the lane markings on one of the right side and the left side is detected by the white line detection part and the other of the lane markings on the other of the right side and the left side is not detected;a white line estimation part for estimating a position of the other of the lane markings on the other of the right side and the left side based on the roadside object detected by the roadside object detection part; and an information output part for outputting the driving support information based on the position of the other of the lane markings on the other of the right side and the left side, the position being estimated by the white line estimation part.

[0012] In a second aspect of the present disclosure based on the aforementioned first aspect, the roadside object detecting part detects, as the roadside object on the other of the right side and the left side, a roadside object that is parallel to the one of the lane markings on the one of the right side and the left side or has a radius of curvature substantially the same as a radius of curvature that the one of the lane markings on the one of the right side and the left side has.

[0013] In a third aspect of the present disclosure based on the aforementioned first aspect, the white line estimation part estimates, as a position of the other of the lane markings on the other of the right side and the left side, a position that is a predetermined first distance inside the lane on which the vehicle is traveling from the position of the roadside object detected by the roadside object detection part.

[0014] In a fourth aspect of the present disclosure based on the aforementioned first aspect, the information output part outputs the driving support information for preventing deviation from the lane in which the vehicle is traveling.

[0015] A fifth aspect of the present disclosure is directed to a driving assistance device installed on a vehicle and operable to output driving assistance information to a driver, comprising: a white line detection part for detecting a lane marker installed on at least one of a right side and a left side of a lane in which the vehicle is traveling; a roadside object detection part for detecting a roadside object installed on at least one of the right side and the left side of the lane in which the vehicle is traveling; a virtual lane marker estimation part for estimating a position of a virtual lane marker used to output the driving assistance information based on the lane marker detected by the white line detection part and the roadside object detected by the roadside object detection part;and an information output part for outputting the driving support information based on the position of the virtual lane marking, the position being estimated by the virtual lane marking estimation part;

[0016] In a sixth aspect of the present disclosure based on the aforementioned fifth aspect, in a case where the lane marking is detected on at least one of the right side and the left side of the lane in which the vehicle is traveling, and on at least one of the right side and the left side the roadside object is detected by the roadside object detecting part, the virtual lane marking estimating part estimates, as the position of the virtual lane marking, a position that is a predetermined second distance from the position of the lane marking inside the lane in which the vehicle is traveling.

[0017] In a seventh aspect of the present disclosure based on the aforementioned sixth aspect, the driving assistance device includes a distance setting part for setting the second distance, the roadside object detecting part judges a category of the roadside object, the distance setting part sets the second distance based on the category of the roadside object judged by the roadside object detecting part, and the virtual lane marking estimating part estimates, as the position of the virtual lane marking, a position that is within the lane on which the vehicle is traveling from the position of the lane marking at the second distance, the second distance being set by the distance setting part.

[0018] In an eighth aspect of the present disclosure based on the aforementioned seventh aspect, the distance setting part estimates a degree of damage in a case of collision with the roadside object based on the category of the roadside object, and sets, as the second distance, a value whose magnitude increases according to an increase in the degree of damage, the category being judged by the roadside object detecting part.

[0019] In a ninth aspect of the present disclosure based on the aforementioned fifth aspect, the information output part outputs the driving support information for preventing deviation from the lane in which the vehicle is traveling.

[0020] A tenth aspect of the present disclosure is directed to a driving assistance device installed on a vehicle and operable to output driving assistance information to a driver, comprising: a white line detection part for detecting a lane marking provided on at least one of a right side and a left side of a lane on which the vehicle is traveling; a lane determination part for determining whether or not a road on which the vehicle is traveling has a center line based on information indicating a position of the vehicle and map information; a center line estimation part for estimating a position of a virtual center line based on the lane marking detected by the white line detection part in a case where the lane determination part determines that the road has no center line;and an information output part for outputting the driving support information based on the position of the virtual center line estimated by the center line estimation part;

[0021] In an eleventh aspect of the present disclosure based on the aforementioned tenth aspect, the center line estimation part estimates, as the position of the virtual center line, a substantially central position between lane markings on both the right side and the left side, the lane markings being detected by the white line detection part.

[0022] In a twelfth aspect of the present disclosure based on the aforementioned tenth aspect, the information output part outputs the driving support information for preventing deviation from the lane in which the vehicle is traveling.

[0023] A thirteenth aspect of the present disclosure is directed to a driving assistance device installed on a vehicle and operable to output driving assistance information to a driver, comprising: a white line detection part for detecting a lane marker provided on at least one of a right side and a left side of a lane in which the vehicle is traveling; a stopping lane determination part for determining whether or not there is a vehicle stopping bay in front of the vehicle whose lane marker is shifted outward by a predetermined width of a stopping lane, based on information indicating a position of the vehicle and map information;a lane marking correction part for correcting a position of the lane marking on at least one of the right side and the left side on which the vehicle stop bay is located, based on the lane marking detected by the white line detection part in a case where the stop lane determination part determines that the vehicle stop bay is present; and an information output part for outputting the driving support information based on the position of the lane marking corrected by the lane marking correction part.

[0024] In a fourteenth aspect of the present disclosure based on the aforementioned thirteenth aspect, the lane marking correcting part corrects a position of a lane marking on a side where the vehicle stop bay is installed, based on a position of a lane marking installed behind the vehicle stop bay.

[0025] In a fifteenth aspect of the present disclosure based on the aforementioned thirteenth aspect, the lane marking correcting part corrects the position of the lane marking on the side where the vehicle stop bay is established, based on a position of a lane marking on a side where the vehicle stop bay is not established.

[0026] In a sixteenth aspect of the present disclosure based on the aforementioned thirteenth aspect, the information output part outputs the driving support information for preventing deviation from the lane in which the vehicle is traveling.

[0027] A seventeenth aspect of the present disclosure is directed to a driving assistance device installed on a vehicle and operable to output driving assistance information to a driver, comprising: a white line detection part for detecting a lane marker provided on at least one of a right side and a left side of a lane on which the vehicle is traveling; a fork determination part for determining whether or not a fork point exists ahead of the vehicle on a road on which the vehicle is traveling based on information indicating a position of the vehicle and map information; a direction estimation part for estimating a traveling direction of the vehicle at the fork point on the road;a virtual lane marking estimation part for generating, at the fork point on the road, a virtual lane marking used to output the driving support information for crossing a forked road that does not run in the traveling direction estimated by the direction estimation part, in a case where the fork determination part determines that the fork point on the road exists; and an information output part for outputting the driving support information based on a position of the virtual lane marking estimated by the virtual lane marking estimation part;

[0028] In an eighteenth aspect of the present disclosure based on the aforementioned seventeenth aspect, the driving support device includes an operation acquiring part for acquiring operation information indicating information about an operation performed by a driver, and the direction estimating part estimates the traveling direction of the vehicle at the fork in the road based on the operation information acquired by the operation acquiring part.

[0029] In a nineteenth aspect of the present disclosure based on the aforementioned eighteenth aspect, the operation acquiring part acquires, as the operation information, steering operation information and / or turn signal operation information.

[0030] In a twentieth aspect of the present disclosure based on the aforementioned seventeenth aspect, the direction estimating part estimates the traveling direction of the vehicle at the fork in the road based on route planning assistance information from a navigation system.

[0031] In a twenty-first aspect of the present disclosure based on the aforementioned seventeenth aspect, in a case where the direction estimation part estimates that the vehicle is heading toward a right side of the forked road, the virtual lane marking estimation part generates the virtual lane marking along the lane marking on the right side detected by the white line detection part, and in a case where the direction estimation part estimates that the vehicle is heading toward a left side of the forked road, the virtual lane marking generates the virtual lane marking along the lane marking on the left side detected by the white line detection part.

[0032] In a twenty-second aspect of the present disclosure based on the aforementioned seventeenth aspect, the driving assistance device includes: a reliability estimation part for estimating a reliability of a position of the lane marking detected by the white line detection part; and a position correction part for correcting a position of the virtual lane marking generated by the virtual lane marking estimation part based on the reliability of the position of the lane marking, the reliability being estimated by the reliability estimation part, and the information output part outputs the driving assistance information based on the position of the virtual lane marking corrected by the position correction part.

[0033] In a twenty-third aspect of the present disclosure based on the aforementioned twenty-second aspect, the reliability estimation part estimates the reliability of the position of the lane marking based on a category of the lane marking detected by the white line detection part.

[0034] In a twenty-fourth aspect of the present disclosure based on the aforementioned twenty-third aspect, as the category of the lane marking, at least one of a white solid line, a white dashed line, a yellow dashed line, Botts' dots, and cat's eyes is included, and the reliability estimation part estimates that the reliability decreases in an order of the white solid line, the white dashed line, the yellow dashed line, the Botts' dots, and the cat's eyes.

[0035] In a twenty-fifth aspect of the present disclosure based on the aforementioned twenty-second aspect, the position correcting part further outwardly corrects the position of the virtual lane marking generated by the virtual lane marking estimating part according to a decrease in the reliability of the position of the lane marking estimated by the reliability estimating part.

[0036] In a twenty-sixth aspect of the present disclosure based on the aforementioned seventeenth aspect based on the aforementioned twenty-second aspect, the information output part outputs the driving support information for preventing deviation from the lane in which the vehicle is traveling.

[0037] According to the aforementioned first aspect, the lane markings provided on the right and left sides of the lane in which the vehicle is traveling are detected. In the case where one of the lane markings on one side is detected and the other of the lane markings on the other side is not detected, the roadside object on the other side is detected. In addition, based on the detected roadside object, the position of the other of the lane markings on the other side is estimated. Further, based on the estimated position of the other of the lane markings on the other side, the driving assistance information is output. Accordingly, the position of the lane marking can be appropriately estimated, and the appropriate driving assistance information can be output.

[0038] In other words, in the case where one of the lane markings on one side is detected and the other of the lane markings on the other side is not detected, the roadside object GL on the other side is detected, and based on the detected roadside object GL, the position of the lane marking VL1 on the other side is estimated, whereby the position of the lane marking VL1 can be appropriately estimated (see Fig. 3). In addition, based on the appropriately estimated position of the lane marking VL1, the driving support information is output, whereby the appropriate driving support information can be output.

[0039] According to the aforementioned second aspect, the roadside object detection part detects, as the roadside object on the other of the right side and the left side, the roadside object that is parallel to the lane marking on one of the right side and the left side or has a curvature radius substantially the same as a curvature radius of the one of the lane markings on the one of the right side and the left side. Accordingly, the position of the lane marking can be more appropriately estimated.

[0040] In other words, since the roadside object GL is detected which is parallel to the lane marking on one of the right side and the left side or has the substantially same radius of curvature as the radius of curvature of the one of the lane markings on one of the right side and the left side, the roadside object GL is detected, such as a guardrail, a median strip, a side wall, a curb, which is useful for estimating the position of the lane marking VL1 on the other of the right side and the left side, whereby the position of the lane marking VL1 can be estimated more appropriately (see Fig. 3).

[0041] According to the above aspect, the position of the other lane marking on the right side and the left side is estimated as the position within the lane in which the vehicle is traveling from the position of the roadside object at the predetermined first distance. Accordingly, the position of the lane marking can be estimated more appropriately.

[0042] In other words, since the position located at the predetermined first distance ΔL1 within the lane on which the vehicle is traveling from the position of the roadside object GL is estimated as the position of the lane marking VL1 on the other of the right side and the left side, the position of the lane marking VL1 can be more appropriately estimated by setting the first distance ΔL1 to an appropriate value (see Fig. 3).

[0043] For example, a category or type of the roadside object GL may be assessed, and based on the assessed category or type of the roadside object GL, an extent of damage caused in the event of a collision with the roadside object GL may be estimated, and a value of the first distance ΔL1 may be set to increase according to an increase in the estimated extent of damage (see Fig. 3). In this case, since a position of the lane marking VL1 is estimated with respect to a roadside object GL that causes a high degree of damage in the event of a collision, leaving a clearance (that is, estimating a position away from the roadside object GL), a risk of collision with the roadside object GL is reduced, whereby the position of the lane marking VL1 can be estimated more appropriately.

[0044] According to the aforementioned fourth aspect, the driving assistance information for preventing the vehicle from deviating from the lane in which it is traveling is output. Accordingly, the appropriate driving assistance information for preventing the vehicle from deviating from the lane in which it is traveling can be output.

[0045] According to the aforementioned fifth aspect, the lane marking provided on the right side and / or the left side of the lane in which the vehicle is traveling is detected. The roadside object provided on at least one of the right side and the left side of the lane in which the vehicle is traveling is detected. In addition, based on the lane marking and the detected roadside object, the position of the virtual lane marking used to output the driving support information is estimated. Further, based on the estimated position of the virtual lane marking, the driving support information is output. Accordingly, the position of the virtual lane marking used to output the driving support information can be appropriately estimated, and the appropriate driving support information can be output.

[0046] In other words, based on the detected lane marking WL and the detected roadside object WA, the position of the virtual lane marking VL2 used to output the driving support information is estimated, whereby the position of the virtual lane marking VL2 can be appropriately estimated (see Fig. 6). In addition, based on the appropriately estimated position of the virtual lane marking VL2, the driving support information is output, whereby the appropriate driving support information can be output.

[0047] According to the aforementioned sixth aspect, in a case where one of the lane markings is detected on at least one of the right side and the left side of the lane in which the vehicle is traveling, and the roadside object is detected on at least one of the right side and the left side, the position located within the lane in which the vehicle is traveling from the position of the lane marking at the predetermined second distance is estimated as the position of the virtual lane marking. Accordingly, the position of the virtual lane marking can be estimated more appropriately.

[0048] In other words, in the case where the roadside object WA is detected on the side where the lane marking WL is detected, the position which is within the lane on which the vehicle is traveling from the position of the lane marking WL at the predetermined second distance ΔL2 is estimated as the position of the virtual lane marking VL2, and the position of the virtual lane marking VL2 can be more appropriately estimated by setting the second distance ΔL2 to an appropriate value (see Fig. 6).

[0049] According to the aforementioned seventh aspect, the category or type of the roadside object is judged. Based on the judged category or type of the roadside object, the second distance is set. In addition, the position located within the lane in which the vehicle is traveling, starting from the position of the lane marking at the set second distance, is estimated as the position of the virtual lane marking. Accordingly, the position of the virtual lane marking can be more appropriately estimated.

[0050] In other words, the category or type of the roadside object WA is judged, and the second distance ΔL2 is set based on the judged category or type of the roadside object WA, thereby enabling the second distance ΔL2 to be set to an appropriate value. Thus, the position of the virtual lane marking VL2 can be estimated more appropriately (see Fig. 6).

[0051] According to the aforementioned eighth aspect, the extent of damage in the event of a collision with the roadside object is estimated based on the assessed category or type of the roadside object, and the second distance is set to a value whose magnitude increases in accordance with the increase in the estimated extent of damage. Accordingly, the position of the virtual lane marking can be more appropriately estimated.

[0052] In other words, the position of the virtual lane marking VL2 with respect to a roadside object WA that causes a high degree of damage in the event of a collision is estimated in such a way that a clearance is left (that is, a position distant from the roadside object WA is estimated), thereby reducing the risk of a collision with the roadside object WA. Thus, the position of the lane marking VL2 can be estimated in a more appropriate manner (see Fig. 6).

[0053] According to the aforementioned ninth aspect, the driving support information for preventing the vehicle from deviating from the lane in which it is traveling is output. Accordingly, the appropriate driving support information for preventing the vehicle from deviating from the lane in which it is traveling can be output.

[0054] According to the aforementioned tenth aspect, the lane marking provided on the right side and / or the left side of the lane in which the vehicle is traveling is detected. Based on the vehicle position information and the map information, it is determined whether or not the road on which the vehicle is traveling has the center line. In addition, if it is determined that the road does not have the center line, the position of the virtual center line is estimated based on the detected lane marking. Further, the driving support information is output based on the estimated position of the virtual center line. Accordingly, the position of the virtual center line used to output the driving support information is appropriately estimated, and the appropriate driving support information can be output.

[0055] In other words, whether the road on which the vehicle is traveling has the center line or not is determined based on the vehicle position information and the map information, thereby enabling appropriate determination of whether the road on which the vehicle is traveling has the center line or not. In addition, in the case where it is determined that the road does not have the center line, the position of the virtual center line is estimated based on the detected lane marking, thereby enabling the appropriate position of the virtual center line to be estimated. Further, the driving support information is output based on the appropriately estimated position of the virtual center line, thereby enabling the appropriate driving support information to be output.

[0056] According to the aforementioned eleventh aspect, the position of the virtual center line VL3 is estimated to be the substantially central position between the lane markings WR and WL on the right side and the left side (see Fig. 9). Accordingly, the position of the virtual center line VL3 used to output the driving support information can be estimated more appropriately.

[0057] According to the aforementioned twelfth aspect, the driving support information for preventing the vehicle from deviating from the lane in which it is traveling is output. Accordingly, the appropriate driving support information for preventing the vehicle from deviating from the lane in which it is traveling can be output.

[0058] According to the aforementioned thirteenth aspect, the lane marker provided on the right side and / or the left side of the lane in which the vehicle is traveling is detected. Based on the vehicle position information and the map information, it is determined whether or not the vehicle stop bay whose lane marker is shifted outward by the predetermined width of the stop lane is present in front of the vehicle. If it is determined that the vehicle stop bay is present, the position of the lane marker on at least one of the right side and the left side on which the vehicle stop bay is provided is corrected based on the detected lane marker. Further, the driving support information is output based on the corrected position of the lane marker.Accordingly, the position of the lane marking is corrected appropriately, whereby the appropriate driving support information can be output.

[0059] In other words, based on the vehicle position information and the map information, it is determined whether or not the vehicle stop bay whose lane marker is shifted outward by the predetermined width of the stop lane is present in front of the vehicle, whereby it is possible to appropriately determine whether or not the vehicle stop bay is present. In addition, in the case where it is determined that the vehicle stop bay is present, the position of the lane marker on the side where the vehicle stop bay is established is corrected based on the detected lane marker, thereby enabling the position of the lane marker to be appropriately corrected. Further, the driving support information is outputted based on the corrected position of the lane marker, thereby enabling the appropriate driving support information to be outputted.

[0060] According to the aforementioned fourteenth aspect, the position of the lane marker on the side where the vehicle stop bay is installed is corrected based on the position of the lane marker installed behind the vehicle stop bay. Accordingly, the position of the lane marker can be corrected more appropriately.

[0061] In other words, for example, if the position of the lane marking WL installed behind the vehicle stopping bay is taken as a starting point, by extending the lane marking VWL so that it is parallel to the lane marking WR on the side where the vehicle stopping bay is not installed, the position of the lane marking WL on the side where the vehicle stopping bay is installed can be corrected in an appropriate manner (that is, the position of the lane marking VWL that has been corrected is obtained) (see Fig. 12).

[0062] According to the aforementioned fifteenth aspect, the position of the lane marking on the side where the vehicle stop bay is provided is corrected based on the position of the lane marking on the side where the vehicle stop bay is not provided. Accordingly, the position of the lane marking can be corrected more appropriately.

[0063] In other words, for example, if the position of the lane marking WL installed behind the vehicle stopping bay is taken as a starting point, by extending the lane marking VWL so that it is parallel to the lane marking WR on the side where the vehicle stopping bay is not installed, the position of the lane marking WL on the side where the vehicle stopping bay is installed can be corrected in an appropriate manner (that is, the position of the lane marking VWL that has been corrected is obtained) (see Fig. 12).

[0064] According to the aforementioned sixteenth aspect, the driving support information for preventing the vehicle from deviating from the lane in which it is traveling is output. Accordingly, the appropriate driving support information for preventing the vehicle from deviating from the lane in which it is traveling can be output.

[0065] According to the aforementioned seventeenth aspect, the lane marker provided on the right side and / or the left side of the lane in which the vehicle is traveling is detected. Based on the vehicle position information and the map information, it is determined whether or not the fork point on the road on which the vehicle is traveling exists ahead of the vehicle. In addition, a traveling direction of the vehicle at the fork point on the road is estimated. In addition, if it is determined that the fork point on the road exists, the virtual lane marker used to output the driving support information to cross the forked road that is not in the estimated traveling direction is generated at the fork point on the road.In addition, the driving assistance information is output based on the estimated position of the virtual lane marking. Accordingly, the position of the virtual lane marking can be appropriately estimated, and the appropriate driving assistance information can be output.

[0066] In other words, based on the information indicating the position of the vehicle and the map information, it is determined whether or not the fork point on the road on which the vehicle is traveling exists in front of the vehicle, thereby appropriately determining whether or not the fork point on the road exists. In addition, in the case where the fork point on the road exists, the virtual lane markings VWR and VWL used to output driving support information to cross the forked road that does not run in the estimated traveling direction are generated at the fork point on the road, thereby appropriately estimating the positions of the virtual lane markings VWR and VWL (see Fig. 15). Furthermore, the driving assistance information is output based on the estimated positions of the virtual lane markings VWR and VWL, thereby enabling the appropriate driving assistance information to be output.

[0067] According to the aforementioned eighteenth aspect, operation information indicating information about the operation performed by a driver is acquired. Based on the acquired operation information, the traveling direction of the vehicle at the fork in the road is estimated. Accordingly, the traveling direction of the vehicle can be appropriately estimated.

[0068] According to the aforementioned nineteenth aspect, the steering operation information and / or the turn signal operation information are acquired as the operation information. Accordingly, the traveling direction of the vehicle can be estimated more appropriately.

[0069] In other words, for example, in a case where a fork point exists where a road ahead forks in a slanting right and left direction, and steering operation information indicating a steering operation in the right direction is acquired, it can be estimated that a traveling direction is the slanting right direction. In addition, for example, in a case where a fork point exists where a road ahead forks in a slanting right and left direction, and turn signal operation information indicating a turn signal operation in the left direction is acquired, it can be estimated that a traveling direction is the slanting left direction.

[0070] According to the aforementioned twentieth aspect, the traveling direction of the vehicle at the fork in the road is estimated based on the route planning assistance information from the navigation system. Accordingly, the traveling direction of the vehicle can be estimated more appropriately.

[0071] In other words, since it is often the case that a driver performs an operation according to the route planning assistance information from the navigation system (that is, it is often the case that the vehicle is traveling toward a direction designated by the route planning assistance information), the traveling direction of the vehicle can be more appropriately estimated based on the route planning assistance information from the navigation system.

[0072] According to the aforementioned twenty-first aspect, when the vehicle is estimated to be heading toward the right side of the forked road, the virtual lane marking is generated along the detected lane marking on the right side, and when the vehicle is estimated to be heading toward the left side of the forked road, the virtual lane marking is generated along the detected lane marking on the left side. Accordingly, the position of the virtual lane marking can be appropriately estimated.

[0073] In other words, in the case where it is estimated that the vehicle is heading towards the right side of the forked road, the virtual lane marking VWL is generated along the detected lane marking WR on the right side, and in the case where it is estimated that the vehicle is heading towards the left side of the forked road, the virtual lane marking VWR is generated along the detected lane marking WL on the left side, thereby enabling the positions of the virtual lane markings VWR and VWL to be appropriately estimated (see Fig. 15).

[0074] According to the aforementioned twenty-second aspect, the reliability of the position of the detected lane marking is estimated. In addition, the position of the generated virtual lane marking is corrected based on the estimated reliability of the lane marking position. Furthermore, the driving assistance information is output based on the corrected position of the virtual lane marking. Accordingly, the position of the virtual lane marking is appropriately corrected, and more appropriate driving assistance information can be output.

[0075] In other words, for example, since the lower the reliability of the position of the detected lane marking, the lower the estimated reliability of the position of the generated virtual lane marking, it is preferable to correct the position of the generated virtual lane marking to the outside of the lane to prevent the driving assistance device from being unnecessarily operated. Thus, since the position of the generated virtual lane marking is appropriately corrected based on the reliability of the position of the detected lane marking, unnecessary operation of the driving assistance device can be prevented, enabling more appropriate driving assistance information to be output.

[0076] According to the aforementioned twenty-third aspect, the reliability or certainty of the lane marking position is estimated based on the detected lane marking category or type. Accordingly, the reliability or certainty of the position of the detected lane marking can be appropriately estimated.

[0077] According to the aforementioned twenty-fourth aspect, it is estimated that the reliability or certainty decreases in order of the white solid line, the white dashed line, the yellow dashed line, the Botts' points, and the cat's eyes. Accordingly, the reliability or certainty of the position of the detected lane marking can be estimated more appropriately.

[0078] According to the aforementioned twenty-fifth aspect, the position of the generated virtual lane marking is further corrected outward according to the decrease in the estimated reliability or certainty of the lane marking position. Accordingly, the position of the virtual lane marking can be corrected more appropriately.

[0079] In other words, since the lower the reliability of the position of the detected lane marking, the lower the estimated reliability of the position of the generated virtual lane marking, it is preferable to correct the position of the generated virtual lane marking outward of the lane to prevent the driving assistance device from being unnecessarily operated. Thus, the position of the generated virtual lane marking is further corrected outward according to the decrease in the estimated reliability of the position of the lane marking, whereby the position of the generated virtual lane marking can be corrected more appropriately.

[0080] According to the aforementioned twenty-sixth aspect, the driving support information for preventing the vehicle from deviating from the lane in which it is traveling is output. Accordingly, the appropriate driving support information for preventing the vehicle from deviating from the lane in which it is traveling can be output.

[0081] These and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings. SHORT DESCRIPTION OF THE DRAWING Fig. 1 is a block diagram illustrating an example of a configuration of a driving support device according to the present invention; Fig. 2 is a block diagram illustrating an example of a functional configuration of a driving support ECU according to a first embodiment; Fig. 3 is a plan view illustrating an example of a situation where a position of a lane marker is estimated by the drive support ECU according to the first embodiment; Fig. 4 is a flowchart showing an example of operations of the driving support ECU according to the first embodiment; Fig. 5 is a block diagram illustrating an example of a functional configuration of a driving support ECU according to a second embodiment; Fig. 6 is a plan view illustrating an example of a situation where a position of a virtual lane marker is estimated by the drive support ECU according to the second embodiment; Fig. 7 is a flowchart showing an example of operations of the driving support ECU according to the second embodiment; Fig. 8 is a block diagram illustrating an example of a functional configuration of a driving support ECU according to a third embodiment; Fig. 9 is a plan view illustrating an example of a situation where a position of a virtual center line is estimated by the driving support ECU according to the third embodiment; Fig. 10 is a flowchart showing an example of operations of the driving support ECU according to the third embodiment; Fig. 11 is a block diagram illustrating an example of a functional configuration of a driving support ECU according to a fourth embodiment; Fig. 12 is a plan view illustrating an example of a situation where a position of a lane marker is corrected by the driving support ECU according to the fourth embodiment; Fig. 13 is a flowchart showing an example of operations of the driving support ECU according to the fourth embodiment; Fig. 14 is a block diagram illustrating an example of a functional configuration of a driving support ECU according to a fifth embodiment of the present invention; Fig. 15 is a plan view illustrating an example of a situation where a position of a virtual lane marker is estimated by the drive support ECU according to the fifth embodiment; Fig. 16 is a plan view illustrating an example of a situation where the position of the virtual lane marker is corrected by the driving support ECU according to the fifth embodiment; Fig. 17 is a flowchart showing an example of operations of the driving support ECU according to the fifth embodiment; Fig. Fig. 18 is a detailed flowchart showing an example of a traveling direction estimation process executed in step S513 in the flowchart of Fig. 17 is executed; Fig. Fig. 19 is a detailed flowchart showing an example of a position correction process executed in step S517 in the flowchart of Fig. 17 is executed; and Fig. 20 is a graph for explaining an example of a method for correcting a position of a virtual lane marker executed by a position correcting section. DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0082] Embodiments of a driving assistance device according to the present invention will be described below with reference to the drawings. The driving assistance device according to the present invention is installed on a vehicle and is operable to output driving assistance information to a driver. First, with reference to Fig. 1 describes an example of a configuration of the driving support device installed on the vehicle.

[0083] Fig. 1 is a block diagram illustrating one example of the configuration of the driving support device according to the present invention. As shown in Fig. 1, a driving support ECU (ECU: electronic control unit) 1 according to the present invention is connected to an input device 2 and an output device 3 as peripheral devices in a communicable manner. Here, the driving support ECU 1 corresponds to the driving support device. Herein, a case will be described where the driving support device outputs the driving support information for preventing deviation from a lane on which the vehicle is traveling. In addition, in the following descriptions, if it is not necessary to distinguish a driving support ECU 11 to a driving support ECU 15 from each other, the driving support ECU 11 to the driving support ECU 15 are collectively referred to as the driving support ECU 1.

[0084] First, with reference to Fig. 1 describes the input device 2 of the driving support ECU 1. The input device 2 includes a CCD camera 21, a navigation system 22, a steering torque detection sensor 23, and a turn signal detection switch 24.

[0085] The CCD camera 21 includes a CCD (charge-coupled device) and generates information indicating images in front of, beside, and the like of the vehicle. In addition, the CCD camera 21 outputs the generated image information to the driving support ECU 1, and detection of lane markings, roadside objects, and the like is performed in the driving support ECU 1 based on the image information.

[0086] In the first to fifth embodiments described below, a case will be described where the driving support ECU 1 (driving support ECU 11 to driving support ECU 15) detects the lane markings, the roadside object, and the like based on the image information from the CCD camera 21. However, the driving support ECU 1 may detect the lane markings, the roadside object, and the like using another method. For example, the driving support ECU 1 may detect the roadside object and the like using a radar device. In addition, the driving support ECU 1 may detect the lane markings, the roadside object, and the like based on image information from another type of camera (for example, a CMOS (Complementary Metal Oxide Semiconductor) camera).

[0087] Here, the "lane marking" is a target object for separating lanes, and includes, in addition to a white line or the like painted in a solid line on a road, a white or yellow dashed line formed by intermittently distributing dashed lines at predetermined regular intervals and connecting adjacent dashed lines, Botts' points (points each having a diameter of approximately 10 cm), cat's eyes (reflective plates), and the like. In addition, the "roadside object" is an object installed outside the lane, and includes a guardrail, a sidewall, a median, a curb, a roadside tree, and the like.

[0088] The navigation system 22 is provided with map information, detects a vehicle position, which is a position of a vehicle on a map, via GPS (Global Positioning System) or the like, and displays the vehicle position on the map on a display. In addition, the navigation system 22 outputs the vehicle position information and the map information corresponding to the vehicle position to the drive support ECU 1.

[0089] The steering torque detection sensor 23 detects a steering torque generated when a driver performs a steering operation via a steering wheel. In addition, the steering torque detection sensor 23 outputs a steering torque signal to the drive support ECU 1.

[0090] The turn signal detection switch 24 detects a result of an operation of instructing the flashing of a turn signal lamp performed by the driver via a turn signal lever. In other words, the turn signal detection switch 24 detects whether or not an operation of instructing the flashing of the turn signal lamp, which indicates turning in a right or left direction, is performed. In addition, the turn signal detection switch 24 outputs a signal (hereinafter referred to as a "turn signal operation signal") indicating the detected operation of the turn signal lever to the drive support ECU 1.

[0091] Since the navigation system 22, the steering torque detection sensor 23, and the turn signal detection switch 24 constitute the input device 2 used in a part of the five embodiments described below, the navigation system 22, the steering torque detection sensor 23, and the turn signal detection switch 24 are shown with dashed lines. For example, the navigation system 22 is used in the third, fourth, and fifth embodiments described below (see Fig. 8, Fig. 11 and Fig. 14).

[0092] Next, with reference to Fig. 1, the output device 3 of the driving support ECU 1 is described. The output device 3 includes a display 31, a speaker 32, and a steering control ECU 33. The display 31 is composed of an LCD (liquid crystal display) provided in front of a driver's seat, and the like, and displays images, characters, and the like according to an instruction from the driving support ECU 1, thereby allowing the driver to view the image, characters, and the like. For example, in a case where the driving support ECU 1 determines that a deviation from a lane in which the vehicle is traveling to the right side is highly likely to occur, an alarm screen notifying that the deviation from the lane in which the vehicle is traveling to the right side is about to occur is displayed on the display 31 based on an instruction from the driving support ECU 1.

[0093] The speaker 32 is provided, for example, next to the driver's seat, and outputs a guidance voice or the like to the driver according to an instruction from the driving support ECU 1. For example, in a case where the driving support ECU 1 determines that a deviation from the lane in which the vehicle is traveling to the right is highly likely to occur, based on an instruction from the driving support ECU 1, the speaker 32 outputs alarm information notifying that a deviation from the lane in which the vehicle is traveling to the right is about to occur.

[0094] The steering control ECU (ECU: electronic control unit) 33 controls a steering system and transmits a predetermined torque, which has been set in advance, to the steering wheel according to an instruction from the driving support ECU 1. For example, in a case where the driving support ECU 1 determines that a deviation from the lane in which the vehicle is traveling to the right is highly likely to occur, a torque is transmitted to excite the steering wheel to turn to the left based on an instruction from the driving support ECU 1. <Erstes Ausführungsbeispiel>

[0095] Fig. 2 is a block diagram illustrating an example of a functional configuration of a driving support ECU 11 according to a first embodiment. As shown in Fig. 2, the driving support ECU 11 includes, as functional parts, a white line detecting section 111, a roadside object detecting section 112, a white line estimating section 113, and an information outputting section 114.

[0096] The driving support ECU 11 causes a microcomputer provided at a suitable location of the driving support ECU 11 to execute a control program previously stored in a ROM (Read Only Memory) or the like provided at a suitable location of the driving support ECU 11, thereby causing the microcomputer to function as the functional parts of the white line detecting section 111, the roadside object detecting section 112, the white line estimating section 113, the information outputting section 114, and the like. Referring to Fig. 3, the functional parts of the driving support ECU 11 are described below.

[0097] Fig. 3 is a plan view illustrating an example of a situation in which a position of a lane marker is estimated by the driving support ECU 11 according to the first embodiment. An upper view according to Fig. 3 is a plan view illustrating an example of a situation in which a position of a lane marking VL1 is estimated by the driving support ECU 11, and a bottom view according to Fig. 3 is a plan view showing the position of the lane marking VL1 estimated by the driving support ECU 11. As shown in the upper illustration according to Fig. As shown in Figure 3, a vehicle VC travels in a direction shown as an upstream direction along a lane marked or demarcated by lane markings WL and WR. In addition, a guardrail (namely, a roadside object) GL is installed outside the lane marking WL along the lane marking WL. In an area NL in front of the vehicle VC, a portion of the lane marking WL has been erased (or blurred).

[0098] As shown in the illustration above, Fig. 3, in a case where the part of the lane marking WL or the lane marking WR has been erased (or has become blurred), it is difficult to appropriately determine a probability of deviation from the lane in which the vehicle VC is traveling based on the lane markings WL and WR. In view of the above problem, the driving support ECU 11 according to the first embodiment estimates a position of the lane marking VL1 in the range NL shown in the upper diagram according to Fig. 3, based on the roadside object GL, as shown in the lower illustration according to Fig. 3, and determines the probability of lane deviation based on the estimated position of the lane marking VL1.

[0099] The white line detection section 111 is the functional part that detects the lane markings WR and WL located on the right and left sides of the lane in which the vehicle VC is traveling based on the image information from the CCD camera 21. Here, the white line detection section 111 corresponds to a white line detection part.

[0100] The roadside object detecting section 112 is the functional part that detects the roadside object GL on the other side based on the image information from the CCD camera 21 when the lane marking WR on one side is detected by the white line detecting section 112 and the lane marking WL on the other side is not detected. Here, the roadside object detecting section 112 corresponds to a roadside object detecting part. Specifically, the roadside object detecting section 112 detects, among roadside objects on the other side, the roadside object GL that is parallel to one of the lane markings on one of the right side and the left side or has a curvature radius substantially the same as a curvature radius that one of the lane markings on one of the right side and the left side has.In other words, the roadside object detecting section 112 detects, among the roadside objects on the side where the lane marking WL is not detected, the roadside object GL that is parallel to the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected, or has the substantially same radius of curvature as the radius of curvature of the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected.

[0101] As shown in the illustration above, Fig. 3, since the roadside object (namely, the guardrail) GL is installed parallel to the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected, or with the substantially same radius of curvature as the radius of curvature of the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected, the roadside object GL is detected as the roadside object used when the white line estimation section 113 estimates the position of the lane marking VL1 corresponding to the area NL shown in the upper diagram according to Fig. 3. In a case where the detected lane marking WR is a straight line, the roadside object detecting section 112 detects a roadside object that is substantially parallel to the lane marking WR.

[0102] As described above, among the roadside objects GL on the side where the lane marking WL is not detected, the roadside object GL installed parallel to the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected, or with the curvature radius substantially the same as the curvature radius of the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected, is detected. Thus, the roadside object GL, such as the guardrail, the median strip, the sidewall, and the curb, which is useful for estimating the position of the lane marking VL1 on the side where the lane marking WL is not detected, can be detected.

[0103] In the first embodiment, the case where the roadside object GL is detected is installed so as to be parallel to the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected, or has a radius of curvature substantially the same as the radius of curvature of the one of the lane markings on one of the right side and the left side on which the lane marking WR is detected. However, the roadside object detecting section 112 may detect other roadside objects in a different manner. For example, the roadside object detecting section 112 may detect a roadside object that is closest to the lane on the side on which the lane marking WL is not detected. In this case, if street trees are planted along the lane (or if a pole ormast is installed), these street trees (or the mast or post) are recorded as the roadside object.

[0104] The white line estimation section 113 is the functional part that estimates the position of the lane marking VL1 on the other side based on the roadside object GL detected by the roadside object detection section 112. Here, the white line estimation section 113 corresponds to a white line estimation part. Specifically, the white line estimation section 113 estimates, as the position of the lane marking VL1 on the other side, a position located a predetermined first distance ΔL1 inside the lane or inward toward the lane on which the vehicle VC is traveling from the position of the roadside object GL detected by the roadside object detection section 112.

[0105] As described above, since the position located within the lane or inward toward the lane on which the vehicle VC is traveling from the position of the detected roadside object GL by the predetermined first distance ΔL1 is estimated as the position of the lane marking VL1 on the other of the sides, the position of the lane marking VL1 can be more appropriately estimated by setting the first distance ΔL1 to an appropriate value.

[0106] For example, a category of the roadside object GL may be assessed, and based on the assessed category of the roadside object GL, an extent of damage caused in the event of a collision with the roadside object GL may be estimated, and a value of the first distance ΔL1 may be set to increase in accordance with an increase in the estimated extent of damage. In this case, since, with respect to a roadside object GL with a high extent of damage caused in the event of a collision, a position of the lane marking VL1 is estimated to leave a clearance (that is, a position distant from the roadside object GL is estimated), a risk of a collision with the roadside object GL is reduced.

[0107] In the first embodiment, although the case where the position located within the lane from the position of the roadside object GL at the first distance ΔL1 is estimated as the position of the lane marking VL1 is described, the white line estimation section 113 only needs to estimate the position of the lane marking VL1 on the other side based on the roadside object GL detected by the roadside object detection section 112. For example, the white line estimation section 113 may estimate the position of the roadside object GL as the position of the lane marking VL1. In this case, processing is simplified. In addition, in a case where a curb or the like is provided in close proximity to the lane marking, an appropriate position can be estimated as the position of the lane marking VL1.

[0108] The information output section 114 is the functional part that outputs driving assistance information based on the position of the lane marking VL1 on the other side, which is estimated by the white line estimation section 113. Here, the information output section 114 corresponds to an information output part. Specifically, the information output section 114 determines a probability of deviation from the lane in which the vehicle VC is traveling based on the lane marking WR or WL detected by the white line detection section 111 and the lane marking VL1 estimated by the white line estimation section 113, and outputs the driving assistance information (here, an alarm or the like) via the output device 3 if it is determined that the probability of deviation is high.

[0109] Fig. Fig. 4 is a flowchart showing an example of operations of the driving support ECU 11 according to the first embodiment. For the sake of simplicity, a case will be described here in which the lane marking WR provided on the right side of the lane is detected by the white line detecting section 111. In addition, in the embodiment according to Fig. In the flowchart shown in Figure 4, the "lane marking" is referred to as a "white line" for simplicity. First, the white line detecting section 111 detects the lane markings WR and WL provided on the right and left sides of the lane in which the vehicle VC is traveling (S101). The roadside object detecting section 112 determines whether the lane marking WL on the left side is detected or not (S103).

[0110] If it is determined that the lane marking WL is detected on the left side (YES in S103), the information output section 114 determines a probability of lane departure based on the lane markings WR and WL on the right and left sides detected in step S101 (S105), and the processing proceeds to step S115. If it is determined that the lane marking WL is not detected on the left side (NO in S103), the roadside object detection section 112 determines whether or not the roadside object GL is detected on the side where the lane marking is not detected (here, the left side) (S107).In the case where it is determined that the roadside object GL is not detected (NO in S107), the determination of the lane deviation performed by the information output section 114 is terminated (S109), the processing returns to step S101, and the processes in step S101 and the subsequent steps are repeatedly executed.

[0111] If it is determined that the roadside object GL is detected (YES in S107), the white line estimation section 113 estimates a position of the lane marking VL1 on the side (here, the left side) where the lane marking is not detected in step S101 based on the detected roadside object GL (S111). Based on the lane marking WR detected in step S101 and the lane marking VL1 estimated in step S111, the information output section 114 determines a probability of lane departure (S113). When the process in step S105 or step S113 is completed, the information output section 114 determines whether the probability of lane departure is high (S115).If it is determined that the probability of lane departure is high (YES in S115), the information output section 114 issues the alarm or the like (S117), the processing returns to step S101, and the processes in step S101 and subsequent steps are repeatedly executed. If it is determined that the probability of lane departure is not high (NO in S115), the processing returns to step S101 and the processes in step S101 and subsequent steps are repeatedly executed.

[0112] As described above, in the case where the lane marking WR is detected on one of the sides and the lane marking WL is not detected on the other of the sides, the roadside object GL on the other of the sides is detected, and the position of the lane marking VL1 on the other of the sides is estimated based on the detected roadside object GL, thereby enabling the position of the lane marking VL1 to be appropriately estimated. In addition, based on the appropriately estimated position of the lane marking VL1, the alarm or the like is output, thereby enabling the appropriate alarm or the like to be output. <Zweites Ausführungsbeispiel>

[0113] Fig. 5 is a block diagram illustrating an example of a functional configuration of a driving support ECU 12 according to a second embodiment. As shown in Fig. 5, the driving support ECU 12 includes, as functional parts, a white line detecting section 121, a roadside object detecting section 122, a distance setting section 123, a virtual lane marking estimating section 124, and an information outputting section 125.

[0114] The driving support ECU 12 causes a microcomputer provided at a suitable location of the driving support ECU 12 to execute a control program previously stored in a ROM or the like provided at a suitable location of the driving support ECU 12, thereby causing the microcomputer to function as the functional parts of the white line detecting section 121, the roadside object detecting section 122, the distance setting section 123, the virtual lane marking estimating section 124, the information output section 125, and the like. Hereinafter, with reference to Fig. 6 describes the functional parts of the driving support ECU 12.

[0115] Fig. 6 is a plan view illustrating an example of a situation in which a position of a virtual lane marker is estimated by the driving support ECU 12 according to the second embodiment. An upper view according to Fig. 6 is a plan view illustrating an example of a situation where a position of a virtual lane marker VL2 is estimated by the driving support ECU 12, and a bottom view according to Fig. Fig. 6 is a plan view showing the position of the virtual lane marker VL2 estimated by the driving support ECU 12. As shown in the upper illustration according to Fig. As shown in Figure 6, a vehicle VC travels in a direction shown as an upstream direction along a lane marked or demarcated by lane markings WL and WR. In addition, a sidewall (namely, a roadside object) WA is installed outside the lane marking WL along the lane marking WL.

[0116] As shown in the illustration above, Fig. 6, in a case where the roadside object WA is installed near the lane marking WL or WR, there may be a case where it is not appropriate to determine a probability of deviation from a lane in which the vehicle VC is traveling based on the lane markings WL or WR. Considering the above problem, the driving support ECU 12 according to the second embodiment estimates the position of the virtual lane marking VL2 facing or adjacent to the side wall (namely, a roadside object) WA, as shown in the lower diagram of FIG. Fig. 6, based on the roadside object WA, and determines the probability of lane deviation based on the estimated position of the virtual lane marking VL2.

[0117] The white line detection section 121 is the functional part that detects the lane markings WR and WL provided on the right and left sides of the lane in which the vehicle VC is traveling based on image information from a CCD camera 21. Here, the white line detection section 121 corresponds to a white line detection part.

[0118] The roadside object detecting section 122 is the functional part that detects the roadside object WA installed on the right and left sides of the lane on which the vehicle VC is traveling, based on the image information from the CCD camera 21. Here, the roadside object detecting section 122 corresponds to a roadside object detecting part. In addition, the roadside object detecting section 122 judges a category or type of the detected roadside object WA. Specifically, the roadside object detecting section 122 makes the judgment by selecting a category or type corresponding to the roadside object WA from among categories or types of a guardrail, a sidewall, a median, a curb, roadside trees, and the like, which are assumed in advance as the roadside objects WA.

[0119] The distance setting section 123 is the functional part that sets a second distance ΔL2, which is a distance between the lane marking WL detected by the white line detection section 121 and the virtual lane marking VL2 estimated by the virtual lane marking estimation section 124. Here, the distance setting section 123 corresponds to a distance setting part. Specifically, the distance setting section 123 sets the second distance ΔL2 based on the category of the roadside object WA judged by the roadside object detection section 122. Here, the distance setting section 123 estimates the distance based on the category.type of the roadside object WA judged by the roadside object detecting section 122, a degree of damage in the event of a collision with the roadside object WA, and sets as the second distance ΔL2 a value whose magnitude increases according to an increase in the estimated degree of damage.

[0120] For example, in a case where the roadside object detecting section 122 judges the category of the roadside object WA by selecting the corresponding category from among the categories of the sidewall, guardrail, median strip, road trees, and curb, the distance setting section 123 sets the second distance ΔL2 as described below. The distance setting section 123 sets values ​​of the second distance ΔL2 for the sidewall, guardrail, median strip, road trees, and curb, respectively, such that a magnitude of each of these values ​​decreases in this order, corresponding to a decreasing order of the extent of damage in the event of a collision with the roadside object WA.For example, in a case where the roadside object detecting section 122 judges the category of the roadside object WA as the sidewall, the guardrail, the median strip, the road trees, and the curb, the distance setting section 123 sets the second distance ΔL2 to 1.0 m, 0.7 m, 0.6 m, 0.5 m, and 0.1 m, respectively.

[0121] The virtual lane marking estimation section 124 is the functional part that estimates the position of the virtual lane marking VL2 used for outputting driving support information (here, for outputting an alarm or the like notifying lane departure) based on the lane markings WR and WL detected by the white line detection section 121 and the roadside object WA detected by the roadside object detection section 122. Here, the virtual lane marking estimation section 124 corresponds to a virtual lane marking estimation part.Specifically, in a case where the white line detecting section 121 detects the lane marking WR or WL on the right side and / or the left side of the lane in which the vehicle is traveling, and the roadside object detecting section 122 detects the roadside object WA on the side (here, on the left side of the lane) on which the lane marking WR or WL is detected, the virtual lane marking estimating section 124 estimates, as the position of the virtual lane marking VL2, a position that is within the lane or inward toward the lane in which the vehicle VC is traveling from the position of the lane marking WL at the second distance ΔL2 set by the distance setting section 123.

[0122] As described above, in the case where the roadside object WA is detected on the side where the lane marking WL is detected, the position that is within the lane or inward toward the lane in which the vehicle is traveling from the position of the lane marking WL by the second distance ΔL2 is estimated as the position of the virtual lane marking VL2. Therefore, by setting the second distance ΔL2 to an appropriate value, the position of the virtual lane marking VL2 can be more appropriately estimated.

[0123] In the second embodiment, the case where the virtual lane marking estimation section 124 estimates, as the position of the virtual lane marking VL2, the position that is within the lane from the position of the lane marking WL at the second distance ΔL2, or inward toward the lane when the roadside object WA is detected, is described. However, the virtual lane marking estimation section 124 only needs to estimate the position of the virtual lane marking VL2 based on the lane marking WL and the roadside object WA.For example, in a case where the roadside object WA is installed outside the lane marking WL, the virtual lane marking estimation section 124 may estimate a position of the lane marking WL as a position of the virtual lane marking VL2, and in a case where the roadside object WA is installed inside the lane marking WL, the virtual lane marking estimation section 124 may estimate a position of the roadside object WA as a position of the virtual lane marking VL2. In this case, processing is simplified.

[0124] In addition, since the category or type of the roadside object WA is judged and the second distance ΔL2 is set based on the judged category or type of the roadside object WA, the second distance ΔL2 can be set to an appropriate value, thereby enabling the position of the virtual lane marking VL2 to be estimated more appropriately.

[0125] Furthermore, since a position of the virtual lane marking VL2 is estimated with respect to a roadside object WA with a high degree of damage caused in the event of a collision, a clearance is left (that is, a position remote from the roadside object WA is estimated), a risk of collision with the roadside object WA is reduced. Thus, the position of the lane marking VL2 can be estimated more appropriately.

[0126] In the second embodiment, the case where a value of the second distance ΔL2, the magnitude of which increases according to the increase in the extent of damage in the case of collision with the roadside object WA, is set by the distance setting section 123 will be described. However, the distance setting section 123 may set the second distance ΔL2 by employing another method. For example, the distance setting section 123 may estimate a risk in a case where the lane departure occurs based on a category or the like of the roadside object WA, and may set a value of the second distance ΔL2, the magnitude of which increases according to an increase in the estimated risk. For example, in a case where the roadside object WA is a guardrail that is robust orstably installed, in order to prevent a vehicle from falling over a cliff, a risk is large that exists in case the deviation from the lane occurs, the second distance ΔL2 is set to have a large value.

[0127] The information output section 125 is the functional part that outputs driving assistance information based on the position of the virtual lane marking VL2 estimated by the virtual lane marking estimation section 124. Here, the information output section 125 corresponds to an information output part. Specifically, the information output section 125 determines a probability of deviation from the lane in which the vehicle VC is traveling based on the lane marking WR or WL detected by the white line detection section 111 and the virtual lane marking VL2 estimated by the virtual lane marking estimation section 124. If it is determined that the probability of deviation is high, the driving assistance information (here, an alarm or the like) is output via the output device 3.

[0128] Fig. Fig. 7 is a flowchart showing an example of operations of the driving support ECU 12 according to the second embodiment. For the sake of simplicity, a case will be described here in which the lane marking WR, which is provided on the right side of the lane, is detected by the white line detecting section 111. In the case according to Fig. In the flowchart shown in Figure 7, the "lane marking" is referred to as a "white line" for simplicity. First, the white line detecting section 121 detects the lane markings WR and WL provided on the right and left sides of the lane in which the vehicle VC is traveling (S201). The roadside object detecting section 122 determines whether the lane marking WL on the left side is detected or not (S203).

[0129] If it is determined that the lane marking WL on the left side is not detected (NO in S203), the lane departure determination performed by the information output section 125 is terminated (S205), the processing returns to step S201, and the processes in step S201 and subsequent steps are repeatedly executed. If it is determined that the lane marking WL on the left side is detected (YES in S203), the roadside object detecting section 122 determines whether the roadside object WA on the left side is detected or not (S207). If it is determined that the roadside object WA is not detected (NO in S207), a probability of lane departure is determined based on the lane markings WR and WL on the right and left sides detected in step S201 (S209), and the processing proceeds to step S217.

[0130] If it is determined that the roadside object WA is detected (YES in S207), the roadside object detecting section 122 judges a category of the roadside object WA (S211). Based on the category of the roadside object WA judged in step S211, the distance setting section 123 sets a second distance ΔL2 (S213). Next, the virtual lane marking estimating section 124 estimates a position of the virtual lane marking VL2 based on the lane marking WL detected in step S201 and the second distance ΔL2 set in step S213, and a probability of lane departure is determined based on the estimated virtual lane marking VL2 and the lane marking WR detected in step S201 (S215).

[0131] When the process in step S209 or step S215 is completed, the information output section 125 determines whether a probability of lane departure is high or not (S217). If it is determined that a probability of lane departure is high (YES in S217), the information output section 125 outputs the alarm or the like (S219), the processing returns to step S201, and the processes in step S201 and subsequent steps are repeatedly executed. If it is determined that the probability of lane departure is not high (NO in S217), the processing returns to step S201 and the processes in step S201 and subsequent steps are repeatedly executed.

[0132] As described above, since the position of the virtual lane marking VL2 used to output the driving assistance information is estimated based on the detected lane marking WL and the detected roadside object WA, the position of the virtual lane marking VL2 can be appropriately estimated. In addition, the alarm or the like is output based on the appropriately estimated position of the virtual lane marking VL2. Thus, an appropriate alarm or the like can be output. <Drittes Ausführungsbeispiel>

[0133] Fig. Fig. 8 is a block diagram illustrating an example of a functional configuration of a driving support ECU 13 according to a third embodiment. As shown in Fig. 8, the driving support ECU 13 includes, as functional parts, a white line detecting section 131, a lane determining section 132, a center line estimating section 133, and an information outputting section 134.

[0134] The driving support ECU 13 causes a microcomputer provided at a suitable location of the driving support ECU 13 to execute a control program previously stored in a ROM or the like provided at a suitable location of the driving support ECU 13, thereby causing the microcomputer to function as the functional parts of the white line detecting section 131, the lane determining section 132, the center line estimating section 133, the information outputting section 134, and the like. Hereinafter, with reference to Fig. 9 describes the functional parts of the driving support ECU 13.

[0135] Fig. 9 is a plan view illustrating an example of a situation in which a position of a virtual center line is estimated by the driving support ECU 13 according to the third embodiment. An upper diagram according to Fig. 9 is a plan view illustrating an example of the situation where a position of a virtual center line VL3 is estimated by the driving support ECU 13, and a bottom view according to Fig. Fig. 9 is a plan view showing the position of the virtual center line VL3 estimated by the driving support ECU 13. As shown in the upper illustration according to Fig. 9, a vehicle VC travels in a direction shown as an upward direction along a lane marked or demarcated with lane markings WL and WR and having no center line.

[0136] There may be a case where if a probability of deviation from a lane in which a vehicle is traveling is estimated when traveling on a road without a center line as shown in the above illustration according to Fig. 9, an alarm or the like cannot be properly output. In other words, there may be a case where it is preferable to travel along the lane marking WL on a left side (or the lane marking WR on a right side) to avoid a risk of collision with an oncoming vehicle. In view of the above problem, the drive support ECU 13 according to the third embodiment estimates the position of the virtual center line VL3 based on the lane marking WL or WR, as shown in the lower diagram according to Fig. 8, and determines a probability of lane deviation based on the estimated position of the virtual center line VL3.

[0137] The white line detection section 131 is the functional part that detects the lane markings WR and WL located on the right and left sides of the lane based on image information from a CCD camera 21. Here, the white line detection section 131 corresponds to a white line detection part.

[0138] The lane determining section 132 is the functional part that determines whether or not a road on which the vehicle VC is traveling is the road that does not have a center line based on information indicating a position of the vehicle VC and map information from the navigation system 22. Here, the lane determining section 132 corresponds to a lane determining part. Specifically, the lane determining section 132 acquires lane information indicating a lane on which the vehicle VC is traveling based on the information indicating the position of the vehicle VC and the map information from the navigation system 22, and determines whether or not the road on which the vehicle VC is traveling has a center line based on the lane information.

[0139] The center line estimation section 133 is the functional part that estimates a position of the virtual center line VL3 based on the lane markings WL and WR detected by the white line detection section 131 when the lane determination section 132 determines that the road on which the vehicle VC is traveling does not have a center line. Here, the center line estimation section 133 corresponds to a center line estimation part. Specifically, the center line estimation section 133 estimates, as the position of the virtual center line VL3, a position substantially central between the lane markings WL and WR on both sides detected by the white line detection section 133.

[0140] In other words, the centerline estimation section 133 estimates, as described in Fig. 9, the position of the virtual center line VL3 is determined based on a distance RW between the lane marking WL and the lane marking WR such that a distance RW1 from the lane marking WL to the virtual center line VL3 and a distance RW2 from the lane marking WR to the virtual center line VL3 are substantially equal to each other, that is, such that the distance RW1 and the distance RW2 are substantially half of the distance RW.

[0141] Since the substantially central position between the detected lane markings WR and WL on the two sides is estimated as the position of the virtual center line VL3, the position of the virtual center line VL3 used to output the driving support information can be appropriately estimated.

[0142] In the third embodiment, the case where the center line estimation section 133 estimates the substantially central position between the lane markings WL and WR on both sides as the position of the virtual center line VL3 will be described. However, the center line estimation section 133 only needs to estimate the position of the virtual center line VL3 based on the lane markings WL and WR detected by the white line detection section 133. For example, the center line estimation section 133 may estimate the position of the virtual center line VL3 based on the distance RW between the lane markings WL and WR on both sides.

[0143] Specifically, for example, in a case where the distance RW is greater than or equal to a predetermined threshold (for example, 5 m), the center line estimation section 133 may estimate, as the position of the virtual center line VL3, a position located a predetermined distance (for example, 2 m) away from the lane marking WL on the left side. In addition, in a case where the distance RW is less than the predetermined threshold (here, 5 m), the center line estimation section 133 may estimate, as the position of the virtual center line VL3, the substantially central position between the lane markings WL and WR. In this case, the more appropriate position can be estimated as the center line VL3.

[0144] The information output section 134 is the functional part that outputs the driving support information based on the position of the virtual center line VL3 estimated by the center line estimation section 133. Here, the information output section 134 corresponds to an information output part. Specifically, the information output section 134 determines a probability of deviation from the lane in which the vehicle VC is traveling based on the lane marking WL detected by the white line detection section 131 and the virtual center line VL3 estimated by the center line estimation section 133, and outputs the driving support information (here, an alarm or the like) via the output device 3 when it is determined that the probability of deviation is high.

[0145] Fig. 10 is a flowchart showing an example of operations of the driving support ECU 13 according to the third embodiment. In the Fig. In the flowchart shown in Figure 10, for the sake of simplicity, the "lane marking" is referred to as a "white line." First, the white line detecting section 131 detects the lane markings WR and WL provided on the right and left sides of the lane in which the vehicle VC is traveling (S301). The lane determining section 132 determines whether the lane markings WR and WL on both sides are detected in step S301 (S303). If it is determined that at least one of the lane markings WR and WL is not detected (NO in S303), the lane deviation determination by the information outputting section 134 is terminated (S305), the processing returns to step S301, and the processes in step S301 and subsequent steps are repeatedly executed.

[0146] In the case where it is determined that the lane markings WR and WL are detected on both sides (YES in S303), the lane determining section 132 acquires the lane information from the navigation system 22 (S307). Based on the lane information acquired in step S307, the lane determining section 132 determines whether the road on which the vehicle VC is traveling has the center line or not (S309). In the case where it is determined that the road has the center line (NO in S309), the information outputting section 134 determines the probability of lane departure based on the lane markings WR and WL detected in step S301 (S311), and the processing proceeds to step S317.

[0147] If it is determined that the road or lane does not have a center line (YES in S309), the center line estimation section 133 estimates the position of the virtual center line VL3 based on the lane markings WR and WL detected in step S301 (S313). Based on the lane marking WL detected in step S301 and the virtual center line VL3 estimated in step S313, the probability of lane departure is determined (S315).

[0148] When the process in step S311 or step S315 is completed, the information output section 134 determines whether the probability of lane departure is high (S317). If it is determined that the probability of lane departure is high (YES in S317), the information output section 134 issues the alarm or the like (S319), the processing returns to step S301, and the processes in step S301 and subsequent steps are repeatedly executed. If it is determined that the probability of lane departure is not high (NO in S317), the processing returns to step S301 and the processes in step S301 and subsequent steps are repeatedly executed.

[0149] As described above, since it is determined whether the road on which the vehicle VC is traveling has the center line or not based on the information indicating the position of the vehicle and the map information, it can be appropriately determined whether the road has the center line or not. In addition, since the position of the virtual center line VL3 is estimated based on the detected lane markings WR and WL in the case where it is determined that the road does not have the center line, the appropriate position of the virtual center line VL3 can be estimated. In addition, since the alarm or the like is output based on the appropriate position estimated as the position of the virtual center line VL3, the appropriate alarm or the like can be output. <Viertes Ausführungsbeispiel>

[0150] Fig. 11 is a block diagram illustrating an example of a functional configuration of a driving support ECU 14 according to a fourth embodiment. As shown in Fig. 11, the driving support ECU 14 includes, as functional parts, a white line detecting section 141, a stopping lane determining section 142, a lane marking correcting section 143, and an information outputting section 144.

[0151] The driving support ECU 14 causes a microcomputer provided at a suitable location of the driving support ECU 14 to execute a control program previously stored in a ROM or the like provided at a suitable location of the driving support ECU 14, thereby causing the microcomputer to function as the functional parts of the white line detecting section 141, the stopping lane determining section 142, the (driving) lane marking correcting section 143, the information outputting section 144, and the like. Hereinafter, with reference to Fig. 12 describes the functional parts of the driving support ECU 14.

[0152] Fig. 12 is a plan view illustrating an example of a situation in which a position of a lane marker is corrected by the driving support ECU 14 according to the fourth embodiment. An upper view according to Fig. Fig. 12 is a plan view illustrating an example of the situation where a position of a lane marker WLB is corrected by the driving support ECU 14, and a bottom view according to Fig. Fig. 12 is a plan view showing a position of a lane marker VWL corrected by the driving support ECU 14. As shown in the above illustration, Fig. 12, a vehicle VC travels in a direction shown as an upward direction along the lane without a center line marked or demarcated with the lane markings WR and WL.

[0153] As per Fig. As shown in FIG. 12, since in a vehicle stop bay ARB such as a bus stop bay, the lane marking WL on the left side is shifted outward by a width of a stopping lane, if a probability of deviating from the lane is determined based on the lane markings WR and WL, there may be a case where an alarm or the like cannot be properly issued. In other words, in a case where the vehicle stop bay ARB is configured such as the bus stop bay, there may be a case where it is not preferable to travel along the lane marking WL on the left side (or the lane marking WR on the right side). In view of the above-mentioned problem, the driving support ECU 14 according to the fourth embodiment corrects the position of the (travel) lane marking WLB to the position of the (travel) lane marking VWL, as shown in the lower diagram according to FIG. Fig. 12, and determines the probability of deviation from the lane based on the corrected (driving) lane marking VWL.

[0154] The white line detection section 141 is the functional part that detects the lane markings WR and WL provided on the right and left sides of the lane in which a vehicle VC is traveling based on image information from the CCD camera 21. Here, the white line detection section 141 corresponds to a white line detection part.

[0155] The stopping lane determining section 142 is the functional part that determines, based on information indicating a position of the vehicle VC and map information from the navigation system 22, whether or not a vehicle stopping bay ARB, whose lane marking is shifted outward by a predetermined width of a stopping lane, exists in front of the vehicle VC. Here, the stopping lane determining section 142 corresponds to a stopping lane determining part. Specifically, the stopping lane determining section 142 acquires, from the navigation system 22, stopping lane information indicating a vehicle stopping bay that is set adjacent to the lane in which the vehicle VC is traveling, and determines, based on the stopping lane information, whether or not the vehicle stopping bay ARB is in front of the vehicle VC. The vehicle stopping bay ARB is defined as the Fig. 12, a waiting area set up on a highway or the like where an accident vehicle is waiting, and the like. In the fourth embodiment, a case where the vehicle stopping bay ARB is the bus stopping bay will be described.

[0156] The lane marking correction section 143 is the functional part that, in a case where the stopping lane determination section 142 determines that the vehicle stop bay ARB is present, corrects a position of the lane marking WL on a side where the vehicle stop bay ARB is established based on the lane markings WR and WL detected by the white line detection section 141. Here, the lane marking correction section 143 corresponds to a lane marking correction part. In other words, the lane marking correction section 143 is the functional part that obtains a position of the corrected lane marking VWL.

[0157] Specifically, the lane marking correction section 143 corrects the position of the lane marking WL on the side where the vehicle stop bay ARB is installed based on the position of the lane marking WL installed behind the vehicle stop bay ARB and the position of the lane marking WR on a side where the vehicle stop bay ARB is not installed. As shown in the figure below, Fig. 12, for example, the lane marking correcting section 143 obtains the position of the lane marking VWL that has been corrected by extending the lane marking VWL, taking the position of the lane marking WL located behind the vehicle stop bay ARW as a starting point so that it is parallel to the lane marking WR.

[0158] As described above, based on the position of the lane marking WL installed behind the vehicle stop bay ARB and the position of the lane marking WR on the side where the vehicle stop bay ARB is not installed, the position of the lane marking WL on the side where the vehicle stop bay ARB is installed can be corrected (that is, the proper position of the lane marking VWL that has been corrected can be obtained).

[0159] In the fourth embodiment, the case where the lane marking correction section 143 obtains the position of the corrected lane marking VWL based on the position of the lane marking WL installed behind the vehicle-stopping bay ARB and the position of the lane marking WR will be described. However, it is only necessary that the position of the corrected lane marking VWL be obtained based on the position of the lane marking WL installed behind the vehicle-stopping bay ARB and / or the position of the lane marking WR. For example, the lane marking correction section 143 may obtain the position of the corrected lane marking VWL based on the position of the lane marking WL installed behind the vehicle-stopping bay ARB.Specifically, the lane marking correcting section 143 may, for example, obtain the position of the lane marking VWL that has been corrected by extending the lane marking WL, taking the position of the lane marking WL located behind the vehicle stop bay ARB as the starting point.

[0160] The information output section 144 is the functional part that outputs the driving support information based on the position of the lane marking VWL corrected by the lane marking correction section 143. Here, the information output section 144 corresponds to an information output part. Specifically, the information output section 144 determines the probability of deviation from the lane in which the vehicle VC is traveling based on the lane markings WR and WL detected by the white line detection section 141 and the lane marking VWL corrected by the lane marking correction section 143, and outputs the driving support information (here, the alarm or the like) via the output device 3 in the case where it is determined that the probability of deviation is high.

[0161] Fig. 13 is a flowchart showing an example of operations of the driving support ECU 14 according to the fourth embodiment. In the Fig. In the flowchart shown in Fig. 13, the "lane marking" is referred to as a "white line" for the sake of simplicity. First, the white line detecting section 141 detects the lane markings WR and WL provided on the right and left sides of the lane in which the vehicle VC is traveling (S401). The stopping lane determining section 142 determines whether or not the lane markings WR and WL on both sides are detected in step S401 (S403). In a case where at least one of the lane markings WR and WL is not detected (NO in S403), the lane departure determination by the information outputting section 144 is terminated (S405), the processing returns to step S401, and the processes in step S401 and subsequent steps are repeatedly executed.

[0162] In the case where it is determined that the lane markings WR and WL are detected on both sides (YES in S403), the stopping lane determining section 142 acquires stopping lane information from the navigation system 22 (S407). Based on the stopping lane information acquired in step S407, the stopping lane determining section 142 determines whether or not the vehicle stopping bay ARB exists in front of the vehicle VC (S409). In the case where it is determined that no vehicle stopping bay ARB exists (NO in S409), the information outputting section 144 determines the probability of lane departure based on the lane markings WR and WL detected in step S401 (S411), and the processing proceeds to step S417.

[0163] If it is determined that the vehicle stop bay ARB is present (YES in S409), the lane marking correction section 143 corrects the position of the lane marking WL on the side where the vehicle stop bay ARB is established based on the lane markings WR and WL detected in step S401, and obtains the corrected position of the lane marking VWL (S413). Based on the lane marking WR detected in step S401 and the lane marking VWL corrected in step S413, the probability of lane departure is determined (S415).

[0164] When the process in step S411 or step S415 is completed, the information output section 144 determines whether the probability of lane departure is high (S417). If it is determined that the probability of lane departure is high (YES in S417), the information output section 144 issues the alarm or the like (S419), the processing returns to step S401, and the processes in step S401 and subsequent steps are repeatedly executed. If it is determined that the probability of lane departure is not high (NO in S417), the processing returns to step S401 and the processes in step S401 and subsequent steps are repeatedly executed.

[0165] As described above, since it is determined based on the information indicating the position of the vehicle VC and the map information whether or not the vehicle stop bay ARB, whose lane marking is shifted outward by the width of the stop lane, exists in front of the vehicle VC, it can be appropriately determined whether or not the vehicle stop bay ARB exists. In the case where it is determined that the vehicle stop bay ARB exists, since the position of the lane marking WL on the side where the vehicle stop bay ARB is established is corrected based on the detected lane markings WR and WL, the position of the lane marking WL can be appropriately corrected. In addition, since the alarm or the like is output based on the corrected position of the lane marking VWL, the appropriate alarm or the like can be output. <Fünftes Ausführungsbeispiel>

[0166] Fig. 14 is a block diagram illustrating an example of a functional configuration of a driving support ECU 15 according to a fifth embodiment. As shown in Fig. 14, the driving support ECU 15 includes, as functional parts, a white line detecting section 151, a fork determining section 152, an operation acquiring section 153, a direction estimating section 154, a virtual lane marking estimating section 155, a reliability estimating section 156, a position correcting section 157, and an information outputting section 158.

[0167] The driving support ECU 15 causes a microcomputer provided at a suitable location of the driving support ECU 15 to execute a control program previously stored in a ROM or the like provided at a suitable location of the driving support ECU 15, thereby causing the microcomputer to operate as the functional parts of the white line detecting section 151, the fork determining section 152, the operation acquiring section 153, the direction estimating section 154, the virtual lane marking estimating section 155, the reliability estimating section 156, the position correcting section 157, the information outputting section 158, and the like.

[0168] Hereinafter, with reference to Fig. 15 and Fig. 16 describes the functional parts of the driving support ECU 15.

[0169] Fig. 15 and Fig. 16 are plan views illustrating an example of a situation in which positions of virtual lane markings are estimated by the driving support ECU 15 according to the fifth embodiment. An upper diagram according to Fig. Fig. 15 is a plan view illustrating one example of the situation where the positions of the virtual lane markings VWR and VWL are estimated by the driving support ECU 15, and a bottom view according to Fig. Fig. 15 is a plan view showing the positions of the virtual lane markings VWR and VWL estimated by the driving support ECU 15. As shown in the above illustration, Fig. 15, a vehicle VC travels in a direction shown as an upward direction along a lane marked or demarcated with lane markings WR and WL and in which a fork or branch is present in front.

[0170] An upper representation according to Fig. 16 is a plan view illustrating an example of a situation where positions of virtual lane markings VWL1 and VWL2 are estimated by the driving support ECU 15, and a bottom view according to Fig. Fig. 16 is a plan view showing the positions of the virtual lane markings VWL1 and VWL2 estimated by the driving support ECU 15. As shown in the upper illustration according to Fig. 16, the vehicle VC travels in a direction shown as an upward direction along the lane marked or demarcated by the lane markings WL and WR1 (or the lane markings WL and WR2) and in which a fork or branch point is present in front. Views on the left sides and views on the right sides according to Fig. 16 differ in categories or types of lane markings WR1 and WR2 on the right side. In other words, the lane markings WR1 on the right side are in the views on the left sides according to Fig. 16 solid white lines and the lane markings WR2 are on the right side in the views on the right sides according to Fig. 16 Botts' points.

[0171] As shown in each of the above illustrations according to Fig. 15 and Fig. As shown in FIG. 16, since there is no lane marking on the left side or right side at the fork point, a probability of departure from the lane in which the vehicle VC is traveling cannot be estimated (or it is difficult to estimate the probability). In view of the above problem, the drive support ECU 15 according to the fifth embodiment generates virtual lane markings VWR, VWL, VWL1, and VWL2 at the fork, and determines the probability of departure from the lane based on positions of the generated virtual lane markings VWR, VWL, VWL1, and VWL2.

[0172] The white line detection section 151 is the functional part that detects the lane markings WR and WL provided on the right and left sides of the lane in which the vehicle VC is traveling based on image information from the CCD camera 21. Here, the white line detection section 151 corresponds to a white line detection part.

[0173] The fork determination section 152 is the functional part that determines whether or not a fork point exists ahead of the vehicle VC on a road on which the vehicle VC is traveling, based on information indicating a position of the vehicle VC and map information from the navigation system 22. Here, the fork determination section 152 corresponds to a fork determination part. In addition, the “fork point” is a point at which a traveling direction of a road forks into two directions, such as points at an intersection of an expressway, an exit from an expressway to a general road, an entrance from an expressway to a service area, or the like. Further, the fork determination section 152 specifically acquires fork information indicating the fork point.branch point, from the navigation system 22, and determines whether or not the fork point exists on the road ahead of the vehicle VC based on the acquired fork information.

[0174] The operation acquisition section 153 is the functional part that acquires operation information indicating information about an operation performed by a driver. Here, the operation acquisition section 153 corresponds to an operation acquisition part. Specifically, the operation acquisition section 153 acquires steering torque information and turn signal operation information generated by the operation performed by the driver via a steering torque detection sensor 23 and a turn signal detection switch 24.

[0175] The direction estimation section 154 is the functional part that estimates a traveling direction of the vehicle VC at the fork in the road based on the operation information (here, the steering torque information and the turn signal operation information). Here, the direction estimation section 154 corresponds to a direction estimation part. Specifically, the direction estimation section 154 determines whether or not a turn signal operation is being performed based on the turn signal operation information, and estimates that the vehicle VC is traveling in a direction according to the turn signal operation when the turn signal operation is being performed. In a case where the turn signal operation is not being performed, the direction estimation section 154 determines whether an absolute value or magnitude of a steering torque TR acquired by the operation acquisition section 153 isis greater than or equal to a threshold value TSH (for example, 1.5 Nm) used to determine whether or not an intentional steering operation is being performed, and in a case where the absolute value is greater than or equal to the determination threshold value TSH, estimates a traveling direction of the vehicle VC based on a direction of the steering torque (that is, based on whether the steering torque TR is positive or negative).

[0176] As described above, based on the steering torque information and the turn signal operation information, the traveling direction of the vehicle VC at the fork point on the road can be accurately estimated.

[0177] In the fifth embodiment, the case where the direction estimation section 154 estimates the traveling direction of the vehicle VC at the road fork based on the operation information acquired by the operation acquisition section 153 will be described. However, the direction estimation section 154 may estimate the traveling direction of the vehicle VC at the road fork using another method. For example, the direction estimation section 154 may estimate the traveling direction of the vehicle VC at the road fork based on the steering torque information and / or the turn signal operation information. In this case, processing is simplified.

[0178] In addition, the direction estimation section 154 can estimate the traveling direction of the vehicle VC at the fork in the road, for example, based on route planning assistance information from the navigation system 22. In a case where a driver sets a destination or the like in the navigation system 22 and sets the route planning assistance information to be output to travel to the destination, it is often the case that an operation is performed based on the route planning assistance information from the navigation system 22 (that is, it is often the case that the vehicle VC travels in a direction designated by the route planning assistance information). Accordingly, in this case, the direction estimation section 154 can accurately estimate the traveling direction of the vehicle VC with a simple configuration.

[0179] The virtual lane marking estimation section 155 is the functional part that generates the virtual lane markings VWL, VWR, and VWL1, which are used to determine the probability of deviating from the lane at the fork point when the fork determination section 152 determines that the fork point exists on the road, so as to traverse a forked road in a direction other than the traveling direction estimated by the direction estimation section 154. Here, the virtual lane marking estimation section 155 corresponds to a virtual lane marking estimation part.

[0180] For example, in a case where the vehicle VC is in the Fig. 15, the vehicle is heading towards a right side of the forked road, as shown in the lower illustration on the right side according to Fig. As shown in Fig. 15, the virtual lane marking estimation section 155 generates a virtual lane marking VWL to cross the forked road on a left side in a direction other than the traveling direction. In addition, in a case where the vehicle VC is heading toward a left side of the forked road, as shown in the lower illustration on the left side according to Fig. As shown in Fig. 15, the virtual lane marking estimation section 155 generates a virtual lane marking VWR to cross the forked road on a right side in a direction other than the traveling direction. Further, in a case where the vehicle is traveling in a lane determined according to Fig. 16 (that is, in a case where the vehicle VC is heading towards a right side of the forked road), as shown in the lower illustration according to Fig. 16, a virtual lane marking VWL1 is generated to cross the forked road on a left side in a direction other than the traveling direction.

[0181] In addition, in a case where the direction estimation section 154 determines that the vehicle VC is heading toward the right side of the forked road, the virtual lane marking estimation section 155 generates the virtual lane marking VWL along the lane marking WR on the right side detected by the white line detection section 151; and in a case where the direction estimation section 154 determines that the vehicle VC is heading toward the left side of the forked road, the virtual lane marking estimation section 155 generates the virtual lane markings VWL, VWR, and VWL1 along the lane marking WL on the left side detected by the white line detection section 151.

[0182] For example, in a case where the direction estimating section 154 estimates that the vehicle VC is in the direction shown in Fig. 15, the vehicle is heading towards the right side of the forked road, as shown in the lower illustration on the right side according to Fig. 15, the virtual lane marking estimation section 155 generates a virtual lane marking VWL along the lane marking WR on the right side detected by the white line detection section 151. In addition, for example, in a case where the direction estimation section 154 estimates that the vehicle VC is heading toward the left side of the forked road, as shown in the lower left-hand diagram according to Fig. 15, the virtual lane marking estimation section 155 generates a virtual lane marking VWR along the lane marking VWL on the left side detected by the white line detecting section 151. Further, in a case where the direction estimation section 154 estimates that the vehicle VC is in the direction shown in Fig. 16 (that is, in the case where the vehicle is heading towards the right side of the forked road), as shown in the illustration below according to Fig. 16, the virtual lane marking estimation section 155 generates a virtual lane marking VWL1 along the lane marking WR on the right side, which is detected by the white line detection section 151.

[0183] As described above, since in the case where the vehicle VC is heading to the right side of the forked road, the virtual lane marking VWL is generated along the detected lane marking WR on the right side, and in the case where the vehicle VC is heading to the left side of the forked road, the virtual lane marking VWR is generated along the lane marking WL on the left side, the positions of the virtual lane markings VWR and VWL can be appropriately estimated.

[0184] The reliability estimation section 156 is the functional part that estimates the reliability of the positions of the lane markings WR and WL detected by the white line detection section 151. Here, the reliability estimation section 156 corresponds to a reliability estimation part. Specifically, the reliability estimation section 156 estimates the reliability of the positions of the lane markings WR and WL based on the categories of the lane markings WR and WL detected by the white line detection section 151. Here, the lane marking categories include a white solid line, a white dashed line, a yellow dashed line, Botts' dots, and cat's eyes.The reliability estimation section 156 estimates the reliability in a decreasing order of reliability, which is the order of the white solid line, the white dashed line, the yellow dashed line, the Botts' points, and the cat's eyes.

[0185] The position correction section 157 is the functional part that corrects a position of the virtual lane marking VWL1 generated by the virtual lane marking estimation section 155 based on the reliability of the positions of the lane markings WR and WL estimated by the reliability estimation section 156. Here, the position correction section 157 corresponds to a position correction part. Specifically, the position correction section 157 further corrects the position of the virtual lane marking VWL1 generated by the virtual lane marking estimation section 155 according to a decrease in the reliability of the positions of the lane markings WR and WL estimated by the reliability estimation section 156.

[0186] For example, in accordance with Fig. 16 the lane marking WR on the right side in the illustration on the left side according to Fig. 16 a white solid line and the lane marking WR is on the right side in the illustration on the right side according to Fig. 16 Botts' points. Since the lane marking WR on the right side in the illustration on the left side according to Fig. 16 is the white solid line, it is determined accordingly that the reliability of the lane marking WR on the right side is high, and the position of the virtual lane marking VWL1 is not corrected. Since the lane marking WR on the right side in the illustration on the right side according to Fig. 16 is the Botts' points, on the other hand, it is determined that the reliability of the lane marking WR on the right side is low, and the position of the virtual lane marking VWL1 is corrected to a position of the virtual lane marking VWL2.

[0187] Fig. Fig. 20 is a graph for explaining an example of a method of correcting the position of the virtual lane marking VWL1, which is executed by the position correcting section 157. An upper graph according to Fig. 20 is a graph G1 showing a relationship between a category (horizontal axis) of each of the lane markings WR and WL detected by the white line detecting section 151 and the reliability (vertical axis) of the position of each of the lane markings WR and WL estimated by the reliability estimating section 156. As shown in the graph G1, the reliability decreases in the order of the white solid line, the white dashed line, the yellow dashed line, the Botts' points, and the cat's eyes.

[0188] A lower graph according to Fig. 20 is a graph G2 showing a relationship between the reliability (horizontal axis) of the position of each of the lane markings WR and WL estimated by the reliability estimation section 156 and a corrected distance (vertical axis) resulting from the correction performed by the position correction section 157. As shown in the graph G2, the corrected distance increases according to a decrease in reliability (that is, the position of the virtual lane marking VWL1 is further corrected outward by the position correction section 157). For example, in a case where the reliability is 50%, the position of the virtual lane marking VWL1 is corrected to be shifted outward by 0.3 m.

[0189] As described above, since the estimated reliability of the position of the generated virtual lane marking VWL1 is lower the lower the reliability of the position of each of the detected lane markings WR and WL, it is preferable to correct the position of the generated virtual lane marking VWL1 to the outside of the lane to prevent the drive support ECU 15 from being unnecessarily operated. Thus, since the position of the generated virtual lane marking VWL1 is appropriately corrected based on the reliability of the position of each of the detected lane markings WR and WL, unnecessary operation of the drive support ECU 15 can be prevented, thereby enabling an alarm or the like to be output more appropriately.

[0190] In addition, since the reliability estimation section 156 estimates the reliability in the decreasing order of reliability, which is the order of the white solid line, the white dashed line, the yellow dashed line, the Botts' points, and the cat's eyes, the reliability of the position of each of the detected lane markings WR and WL can be estimated more appropriately.

[0191] In the fifth embodiment, the case where the position correction section 157 further corrects the position of the virtual lane marking VWL1 outward according to the decrease in the reliability estimated by the reliability estimation section 156 will be described. In contrast, however, the position correction section 157 may further correct the position of the virtual lane marking VWL1 inward according to the increase in the reliability estimated by the reliability estimation section 156. In this case, the lane deviation can be prevented in a more secure manner.

[0192] In addition, in the fifth embodiment, the case where the reliability estimation section 156 estimates the reliability in the decreasing order of reliability, which is the order of the white solid line, the white dashed line, the yellow dashed line, the Botts' dots, and the cat's eyes, will be described. However, it is only necessary that the reliability estimation section 156 estimate the reliability of the position of each of the lane markings WR and WL based on the category or the like of each of the lane markings WR and WL detected by the white line detecting section 151. For example, the reliability estimation section 156 may estimate the reliability of the position of each of the lane markings WR and WL based on the category and the distinctness of each of the lane markings WR and WL detected by the white line detecting section 151.Here, "distinctness" refers to a characteristic of a clear demarcation in an image based on a color, a reflectance property, and the like of each of the lane markings WR and WL with respect to a road surface. In this case, the reliability of the position of each of the lane markings WR and WL can be estimated more accurately.

[0193] Again referring to Fig. Figure 14 describes the functional configuration of the driving support ECU 15. The information output section 158 is the functional part that outputs the driving support information based on the virtual lane markings VWL, VWR, and VWL1 estimated by the virtual lane marking estimation section 155, or based on the position of the virtual lane marking VWL2 corrected by the position correction section 157. Here, the information output section 158 corresponds to an information output part.Specifically, the information output section 158 determines a probability of lane departure based on the lane markings WR and WL detected by the white line detection section 141 and the virtual lane markings VWL, VWR, and VWL1 estimated by the virtual lane marking estimation section 155 (or the virtual lane marking VWL2 corrected by the position correction section 157), and outputs the driving support information (here, the alarm or the like) via the output device 3 when it is determined that the probability of departure is high.

[0194] Fig. Fig. 17 is a flowchart showing an example of operations of the driving support ECU 15 according to the fifth embodiment. In the Fig. In the flowchart shown in Fig. 15, for the sake of simplicity, the "lane marking" is referred to as a "white line." First, the white line detecting section 151 detects the lane markings WR and WL provided on the right and left sides of the lane in which the vehicle VC is traveling (S501). The bifurcation determining section 152 determines whether or not the lane markings WR and WL on both sides are detected in step S501 (S503). If it is determined that at least one of the lane markings WR and WL is not detected (NO in S503), the lane deviation determination by the information outputting section 158 is terminated (S505), the processing returns to step S501, and the processes in step S501 and subsequent steps are repeatedly executed.

[0195] In the case where it is determined that the lane markings WR and WL are detected on both sides (YES in S503), the fork determining section 152 acquires the fork information from the navigation system 22 (S507). Based on the fork information acquired in step S507, the fork determining section 152 determines whether or not a fork point exists ahead of the vehicle VC (S509). In a case where the fork point does not exist (NO in S509), the information output section 158 determines a probability of lane departure based on the lane markings WR and WL detected in step S501 (S511), and the processing proceeds to step S517.

[0196] In a case where the fork point exists (YES in S509), a traveling direction estimation process in which a traveling direction of the vehicle VC at the fork point is estimated by the direction estimation section 154 or the like is executed (S513). Based on the traveling direction estimated in step S513, the virtual lane marking estimation section 155 generates the virtual lane markings VWL, VWR, and VWL1, which are used to determine a probability of lane departure (S515). Next, a position correction process in which the position of each of the virtual lane markings VWL, VWR, and VWL1 generated in step S515 is corrected by the reliability estimation section 156 and the position correction section 157 is executed (S517).Next, based on the lane markings WR and WL detected in step S501 and the virtual lane marking VWL, VWR and VWL1 generated in step S513 (or the virtual lane marking VWL2 corrected in step S515), a probability of lane departure is determined by the information output section 158 (S519).

[0197] When the process in step S511 or step S519 is completed, the information output section 158 determines whether the probability of lane departure is high (S521). If it is determined that the probability of lane departure is high (YES in S521), the information output section 158 issues the alarm or the like (S523), the processing returns to step S501, and the processes in step S501 and subsequent steps are repeatedly executed. If it is determined that the probability of lane departure is not high (NO in S521), the processing returns to step S501 and the subsequent steps are repeatedly executed.

[0198] Fig. Fig. 18 is a detailed flowchart showing an example of the traveling direction estimation process executed in step S513 in the flowchart of Fig. 17 is executed. First, the operation acquiring section 153 acquires turn signal operation information and determines whether or not a turn instruction operation has been received (S601). If it is determined that the turn instruction operation has not been received (NO in S601), the processing proceeds to step S609. If it is determined that the turn instruction operation has been received (YES in S601), the direction estimating section 154 determines whether the turn instruction operation received in step S601 instructs a right direction (S603). In the case where it is determined that the right direction is instructed (YES in S603), the direction estimation section 154 estimates that the vehicle VC is heading toward the right direction at a fork point (S607), and the processing returns to the processing according to Fig. 17. In the case where it is determined that a left direction is instructed (NO in S603), the direction estimation section 154 estimates that the vehicle VC is heading toward the left direction at the fork point (S605), and the processing returns to the step shown in Fig. 17, step S515.

[0199] If NO in S601, the operation acquisition section 153 acquires steering torque TR information via the steering torque detection sensor 23 (S609). The direction estimation section 154 determines whether or not an absolute value of the steering torque TR acquired in step S609 is greater than or equal to the determination threshold TSH (S611). If it is determined that the absolute value of the steering torque TR is less than the determination threshold TSH (NO in S611), the processing returns to step S601, and the processes in step S601 and subsequent steps are repeatedly executed. In the case where it is determined that the absolute value of the steering torque TR is greater than or equal to the determination threshold TSH (YES in S611), the direction estimating section 154 determines based on whether the steering torque TR obtained in step S609 isacquired steering torque TR is positive or negative, whether the steering torque TR corresponds to steering in the right direction or not (S613). In the case where it is determined that the steering torque TR corresponds to steering in the right direction (YES in S613), the direction estimation section 154 estimates that the vehicle VC is heading toward the right direction at the fork point (S617), and returns the processing to that according to . Fig. 17. In the case where it is determined that the steering torque TR corresponds to steering in the left direction (NO in S613), the direction estimation section 154 estimates that the vehicle VC is heading to the left direction at the fork point (S615), and the processing returns to the step shown in Fig. 17, step S515.

[0200] Fig. Fig. 19 is a detailed flowchart showing an example of the position correction process executed in step S517 in the flowchart of Fig. 17. First, the virtual lane marking estimation section 155 determines whether in step S513 in the flowchart according to Fig. 17, whether or not it is estimated that the vehicle VC is heading toward the right direction (S701). In a case where it is estimated that the vehicle VC is heading toward the left direction (NO in S701), the reliability estimation section 156 judges a category of the lane marking WL on the left side (S703). In a case where it is estimated that the vehicle VC is heading toward the right direction (YES in S701), the reliability estimation section 156 judges a category of the lane marking WR on the right side (S705).

[0201] In a case where the process in step S703 or step S705 is completed, the reliability estimation section 156 estimates the reliability of the position of the lane marking WR (or the lane marking WL) based on the category of the lane marking WR (or the lane marking WL) estimated in step S703 or step S705 (S707). The position correction section 157 corrects the position of the lane marking WR (or the lane marking WL) calculated in accordance with Fig. 17 shown in step S515 based on the reliability estimated in step S707, and the processing returns to the Fig. 17, step S519.

[0202] As described above, based on the information indicating the position of the vehicle VC and the map information, it is determined whether or not a fork point on a road on which the vehicle VC is traveling exists in front of the vehicle VC. Thus, it can be appropriately determined whether or not the fork point on the road exists. In addition, in the case where it is determined that the fork point on the road exists, the virtual lane marker VWR, VWL, and VWL1 are generated, which is used to issue the alarm or the like indicating the deviation from the lane at the fork point on the road to cross the estimated forked road that is not in the traveling direction. Thus, the position of the virtual lane marker VWR, VWL, and VWL1 can be appropriately estimated (see Fig. 15 and Fig.16). Further, the alarm or the like is issued based on the virtual lane markings VWR, VWL, and VWL1, which are estimated to be at appropriate positions. Thus, the appropriate alarm or the like can be issued.

[0203] The driving support device according to the present invention is not limited to the above-described embodiments, and can be realized as described below.

[0204] (A) In the first embodiment, the driving support ECU 11 is described as including, as functional parts, the white line detection section 111, the roadside object detection section 112, the white line estimation section 113, the information output section 114, and the like. However, at least one of the functional parts of the white line detection section 111, the roadside object detection section 112, the white line estimation section 113, and the information output section 114 may be implemented by hardware such as an electric circuit.

[0205] Similarly, in the second embodiment, the driving support ECU 12 is described as including, as functional parts, the white line detection section 121, the roadside object detection section 122, the distance setting section 123, the virtual lane marking estimation section 124, the information output section 125, and the like. However, at least one of the functional parts of the white line detection section 121, the roadside object detection section 122, the distance setting section 123, the virtual lane marking estimation section 124, and the information output section 125 may be implemented by hardware such as an electric circuit.

[0206] Similarly, in the third embodiment, the driving support ECU 13 is described as including, as functional parts, the white line detection section 131, the lane determination section 132, the center line estimation section 133, the information output section 134, and the like. However, at least one of the functional parts of the white line detection section 131, the lane determination section 132, the center line estimation section 133, and the information output section 134 may be implemented by hardware such as an electric circuit.

[0207] Similarly, in the fourth embodiment, the driving support ECU 14 is described as including, as functional parts, the white line detection section 141, the stopping lane determination section 142, the (driving) lane marking correction section 143, the information output section 144, and the like. However, at least one of the functional parts of the white line detection section 141, the stopping lane determination section 142, the (driving) lane marking correction section 143, and the information output section 144 may be implemented by hardware such as an electric circuit.

[0208] Similarly, in the fifth embodiment, the driving support ECU 15 is described, which includes, as the functional parts: the white line detecting section 151, the fork determining section 152, the operation acquiring section 153, the direction estimating section 154, the virtual lane marking estimating section 155, the reliability estimating section 156, the position correcting section 157, the information outputting section 158, and the like. However, at least one of the functional parts of the white line detection section 151, the fork determination section 152, the operation acquisition section 153, the direction estimation section 154, the virtual lane marking estimation section 155, the reliability estimation section 156, the position correction section 157, and the information output section 158 may be realized by hardware such as an electric circuit.

[0209] (B) In the first to fifth embodiments, the driving support ECU 1 (namely, the driving support ECUs 11 to 15) is described as outputting the driving support information for preventing deviation from the lane in which the vehicle VC is traveling. However, the driving support ECU 1 may output other driving support information. For example, the driving support ECU 1 may output driving support information for assisting a driver in changing lanes (or overtaking).

[0210] (C) In the first to fifth embodiments, the driving support ECU 1 (namely, the driving support ECUs 11 to 15) is described as detecting the lane markings, the roadside object, and the like based on the image information from the CCD camera 21. However, the driving support ECU 1 may detect the lane markings, the roadside object, and the like using another method. For example, the driving support ECU 1 may detect the roadside object and the like using a radar device. In addition, the driving support ECU 1 may detect the lane markings, the roadside object, and the like based on image information from another type of camera (for example, a CMOS camera).

[0211] While the invention has been described in detail, the foregoing description is in all aspects illustrative and not restrictive. It should be understood that numerous other modifications and variations may be devised without departing from the scope of the invention.

[0212] A driving support ECU (11) comprises: a roadside object detection section (112) for detecting a roadside object (GL) in a case where a lane marking (WR) on at least one of the right side and the left side is detected by a white line detection section (111) and another lane marking (WL) is not detected on the other of the right side and the left side; a white line estimation section (113) for estimating a position of a lane marking (VL1) on the other of the right side and the left side based on the detected roadside object (GL); and an information output section (114) for determining a deviation from the lane in which the vehicle (VC) is traveling based on the estimated position of the lane marking (VL1) on the other of the right side and the left side.

Claims

[1] A driving assistance device (15) installed on a vehicle and operable to output driving assistance information to a driver, comprising: a white line detecting part (151) for detecting a lane marking provided on at least one of a right side and a left side of a lane on which the vehicle is traveling; a fork determination part (152) for determining whether or not there is a fork point ahead of the vehicle on a road on which the vehicle is traveling, based on information indicating a position of the vehicle and map information; a direction estimating part (154) for estimating a traveling direction of the vehicle at the fork point on the road; a virtual lane marking estimation part (155) for generating, at the fork point on the road, a virtual lane marking used to output the driving support information to cross a forked road that does not run in the estimated traveling direction, based on the traveling direction of the vehicle estimated by the direction estimation part in a case where the fork determining section determines that the fork point on the road exists; and an information output part (158) for outputting the driving support information as an alarm about the deviation from the lane in which the vehicle is traveling, in a case where it is determined that the possibility of deviation from the lane is high, based on a position of the virtual lane marking estimated by the virtual lane marking estimation part and the lane marking detected by the white line detection part. [2] Driving assistance device according to claim 1, comprising: an operation acquiring part (153) for acquiring operation information indicating information about an operation performed by a driver, wherein the direction estimation part estimates the traveling direction of the vehicle at the fork point on the road based on the operation information acquired by the operation acquisition part. [3] The driving support device according to claim 2, wherein the operation acquiring part acquires, as the operation information, steering operation information and / or turn signal operation information. [4] The driving support device according to claim 1, wherein the direction estimating part estimates the traveling direction of the vehicle at the fork in the road based on route planning assistance information from a navigation system. [5] The driving support device according to claim 1, wherein, in a case where the direction estimation part estimates that the vehicle is heading toward a right side of the forked road, the virtual lane marking estimation part generates the virtual lane marking along the lane marking on the right side detected by the white line detection part, and in a case where the direction estimation part estimates that the vehicle is heading toward a left side of the forked road, the virtual lane marking generates the virtual lane marking along the lane marking on the left side detected by the white line detection part. [6] Driving assistance device according to claim 1, comprising: a reliability estimation part (156) for estimating a reliability of a position of the lane marking detected by the white line detection part; and a position correcting part (157) for correcting a position of the virtual lane marking generated by the virtual lane marking estimating part based on the reliability of the position of the lane marking, the reliability being estimated by the reliability estimating part, wherein the information output part outputs the driving support information based on the position of the virtual lane marking corrected by the position correction part. [7] The driving assistance device according to claim 6, wherein the reliability estimation part estimates the reliability of the position of the lane marking based on a category of the lane marking detected by the white line detection part. [8] Driving assistance device according to claim 7, wherein the lane marking category includes at least one of a white solid line, a white dashed line, a yellow dashed line, Botts' dots and cat's eyes, and The reliability estimation part estimates that reliability decreases in an order of the white solid line, the white dashed line, the yellow dashed line, the Botts' points, and the cat's eyes. [9] The driving assistance device according to claim 6, wherein the position correcting part further corrects the position of the virtual lane marking generated by the virtual lane marking estimating part outward according to a decrease in the reliability of the position of the lane marking estimated by the reliability estimating part. [10] The driving assistance device according to claim 1, wherein the virtual lane marking estimation part generates, at the fork point on the road, the virtual lane marking for crossing the forked road which is not in the traveling direction estimated by the direction estimation part, in an area on the right side or the left side where the lane marking is not detected by the white line detection part. [11] The driving support device according to claim 10, wherein, in a case where the direction estimation part estimates that the vehicle is heading toward a right side of the forked road, the virtual lane marking estimation part generates the virtual lane marking along the lane marking on the right side detected by the white line detection part, and in a case where the direction estimation part estimates that the vehicle is heading toward a left side of the forked road, the virtual lane marking generates the virtual lane marking along the lane marking on the left side detected by the white line detection part. [12] The driving support device according to claim 1, wherein, in a case where the direction estimation part estimates that the vehicle is heading to a right side of the forked road, the virtual lane marking estimation part generates the virtual lane marking based on the traveling direction of the vehicle estimated by the direction estimation part to cross the forked road on a left side, and, in a case where the direction estimation part estimates that the vehicle is heading to a left side of the forked road, the virtual lane marking estimation part generates the virtual lane marking based on the traveling direction of the vehicle estimated by the direction estimation part to cross the forked road on a right side.

Images