JP2019132762A - Vehicle position estimating device - Google Patents

Abstract

To provide a vehicle position estimating device with which it is possible to recognize that a demarcation line has changed from when map information was generated.SOLUTION: The vehicle position estimating device comprises: a lateral position estimation unit for estimating a first lateral position of a vehicle by verification of the left demarcation line of a travel lane of a road surface image with the left demarcation line of a travel lane of map information, and estimating a second lateral position by verification of the right demarcation line of the travel lane of the road surface image with the right demarcation line of the travel lane of the map information; a difference determination unit for determining whether or not a difference between the first and the second lateral positions is greater than or equal to a determination threshold; a match determination unit for determining whether or not there is a portion among map width shapes in a comparison target range that matches a recognition width shape recognized from the road surface image when it is determined that the difference between the first and the second lateral positions is greater than or equal to the determination threshold; and a demarcation line change recognition unit for recognizing that the demarcation line has been changed from when the map information was generated when it is determined that there is no portion among the map width shapes that matches the recognition width shape.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a host vehicle position estimation device.

  Conventionally, JP, 2007-101690, A is known as technical literature about a self-vehicle position estimating device. This publication discloses an apparatus that estimates a position of a host vehicle by detecting a landmark around the host vehicle as a landmark and matching it with a position of a landmark included in map information.

JP 2007-101690 A

  In the vehicle position estimation as described above, road lane markings are often used as landmarks. However, if the lane markings are repainted from the time when the map information is generated (updated) and the lane marking position is changed, the vehicle position is erroneously estimated using the lane marking position in the old map information. There is a fear.

  Therefore, in the present technical field, it is desired to provide a vehicle position estimation device that can recognize that the lane marking has been changed since the generation of the map information.

  In order to solve the above-described problem, an aspect of the present invention is an own vehicle position estimation device that estimates a position of a vehicle on a map, and is based on a measurement result of a measurement unit mounted on the vehicle. Vehicle position recognition unit for recognizing the position of the vehicle, a road surface image generation unit for generating a road surface image including the left and right lane markings of the traveling lane of the vehicle based on the captured image of the camera of the vehicle, and position information of the lane markings The first position which is the lateral position on the map of the vehicle by collating the map database storing the map information with the left lane line of the traveling lane included in the road image and the left lane line of the traveling lane included in the map information. The second position which is the lateral position on the map of the vehicle is verified by comparing the right lane line of the traveling lane included in the road image with the right lane line of the traveling lane included in the map information. A lateral position estimating unit for estimating a lateral position, and a first lateral A map used for matching based on a position on the map of the vehicle recognized by the vehicle position recognition unit and a difference determination unit that determines whether or not the difference between the position and the second lateral position is equal to or greater than a determination threshold The difference between the first lateral position and the second lateral position is greater than or equal to the determination threshold by the comparison target range setting unit that includes a range on the upper lane and that includes a comparison target range that is larger than the range. If it is determined, the part that matches the recognized width shape that is the width shape of the left and right lane lines recognized from the road surface image in the map width shape that is the width shape of the left and right lane lines of the map information in the comparison target range A match determination unit that determines whether or not there is a match, and if the match determination unit determines that there is a portion that matches the recognition width shape in the map width shape, it matches the recognition width shape in the map width shape Position error calculation unit that calculates the position error of the vehicle A lane line change recognizing unit that recognizes that the lane line has been changed since the generation of the map information when the coincidence determining unit determines that the map width shape does not have a portion that matches the recognized width shape. .

  As described above, according to one aspect of the present invention, it can be recognized that the lane marking has been changed since the map information was generated.

It is a block diagram which shows the own vehicle position estimation apparatus which concerns on one Embodiment. (A) It is a graph which shows the example in case the peak of a collation value is one. (B) It is a graph which shows the example in case the peak of a collation value is two. (A) It is a top view which shows the case where the left lane line of a road surface image and the left lane line of map information are collated. (B) It is a top view which shows the case where the right lane line of a road surface image and the right lane line of map information are collated. (C) It is a top view for demonstrating recognition of the width shape of the left and right division lines based on a road surface image. (A) It is a figure which shows an example of the recognition width shape of a road surface image. (B) It is a figure which shows the 1st example of the map width shape in a comparison object range. (C) It is a figure which shows the 2nd example of the map width shape in the comparison object range. It is a flowchart which shows the memory start process of a lane width. It is a flowchart which shows the recognition process of a lane marking change.

  Embodiments of the present invention will be described below with reference to the drawings.

[Configuration of the vehicle position estimation device]
A host vehicle position estimation apparatus 100 shown in FIG. 1 includes an ECU [Electronic Control Unit] 10 that comprehensively manages the system. The ECU 10 is an electronic control unit having a CPU [Central Processing Unit], a ROM [Read Only Memory], a RAM [Random Access Memory], and the like. In the ECU 10, for example, various functions are realized by loading a program stored in the ROM into the RAM and executing the program loaded in the RAM by the CPU. The ECU 10 may be composed of a plurality of electronic units.

  The ECU 10 is connected to the GPS receiver 1, the camera 2, the vehicle speed sensor 3, the yaw rate 4, and the map database 5.

  The GPS receiving unit 1 is a measuring unit that is mounted on a vehicle and measures the position of the vehicle on the map (for example, the latitude and longitude of the vehicle) by receiving signals from three or more GPS satellites. The GPS receiving unit 1 transmits the measurement result to the ECU 10.

  The camera 2 is an imaging device that captures an external situation of the vehicle. The camera 2 is provided on the back side of the windshield of the vehicle. The camera 2 transmits a captured image in front of the vehicle to the ECU 10. The camera 2 may be a monocular camera or a stereo camera. The camera 2 may be provided so as to capture the back of the vehicle, or may be configured as a plurality of cameras that capture images in a plurality of directions.

  The vehicle speed sensor 3 is a detector that is mounted on the vehicle and detects the speed of the vehicle. As the vehicle speed sensor 3, for example, a wheel speed sensor that is provided on a vehicle wheel or a drive shaft that rotates integrally with the wheel and detects the rotation speed of the wheel is used. The vehicle speed sensor 3 transmits the detected vehicle speed information to the ECU 10.

  The yaw rate sensor 4 is a detector that is mounted on the vehicle and detects the yaw rate (rotational angular velocity) around the vertical axis of the center of gravity of the vehicle. As the yaw rate sensor 4, for example, a gyro sensor can be used. The yaw rate sensor 4 transmits the detected yaw rate information of the vehicle to the ECU 10.

  The map database 5 is a database that stores map information. The map database 5 is formed in, for example, an HDD [Hard Disk Drive] mounted on the vehicle. The map information includes road position information, road shape information (eg, curves, straight line types, curve curvature, etc.), intersection and branch point position information, and structure position information. The map database 5 may be formed on a server that can communicate with the vehicle.

  A functional configuration of the ECU 10 will be described. The ECU 10 includes a vehicle position recognition unit 11, a road surface image generation unit 12, a lateral position estimation unit 13, a difference determination unit 14, a comparison target range setting unit 15, a coincidence determination unit 16, a vertical position error calculation unit 17, and a lane marking change recognition. A portion 18 is provided. Some of the functions of the ECU 10 described below may be performed by a server that can communicate with the vehicle.

  The vehicle position recognition unit 11 recognizes the position of the vehicle on the map based on the position information of the GPS reception unit (measurement unit) 1 and the map information of the map database 5. Based on the vehicle speed information of the vehicle speed sensor 3 and the yaw rate information of the yaw rate sensor 4, the vehicle position recognition unit 11 displays the vehicle speed history (or the wheel rotation speed history) and the vehicle yaw rate history on the map of the vehicle. May be recognized. In this case, the vehicle speed sensor 3 and the yaw rate sensor 4 function as an in-vehicle measuring unit. The vehicle position recognition unit 11 may recognize the position of the vehicle on the map by so-called odometry using a known method.

  The road surface image generation unit 12 generates a road surface image including the left and right lane markings of the travel lane on which the vehicle travels based on the captured image of the camera 2. The road surface image generation unit 12 generates a road surface image by, for example, projecting and converting the captured image of the camera 2. The road surface image generation unit 12 recognizes left and right lane markings included in the road surface image by image recognition. The road surface image generation unit 12 extends the road surface image by superimposing the newly generated road surface image on the road surface image generated immediately before (performs image matching using a lane marking or the like). The position of the lane marking in the road surface image is a relative position with respect to the vehicle (camera 2).

  The lateral position estimation unit 13 determines whether the vehicle is based on the left and right lane lines of the traveling lane included in the road surface image generated by the road surface image generation unit 12 and the left and right lane lines of the traveling lane included in the map information of the map database 5. The lateral position of is estimated. The lateral position estimating unit 13 refers to the position on the map of the vehicle recognized by the vehicle position recognizing unit 11 to obtain left and right dividing lines of the map information to be collated. The lateral position estimation unit 13 estimates the lateral position of the vehicle by collating the left and right lane markings of the traveling lane included in the road image with the left and right lane markings of the traveling lane included in the map information.

  Here, FIG. 2A is a graph showing an example in the case where the peak of the matching value is one. In FIG. 2A, the vertical axis represents the collation value, and the horizontal axis represents the horizontal position. The collation value is a value corresponding to the degree of coincidence between the left and right lane markings of the road surface image and the map information. FIG. 2A shows a case where the peak of the verification value is one. The lateral position estimation unit 13 calculates a collation value (degree of coincidence) while shifting the left and right lane markings of the road surface image laterally with respect to the left and right lane markings of the map information. When the collation value has one peak as shown in FIG. 2A, the lateral position estimation unit 13 determines the left and right lane lines of the road image and the left and right lane lines of the map information so that the collation value becomes a peak. The lateral position of the vehicle on the map is estimated.

  FIG. 2B is a graph showing an example in which there are two verification value peaks. For example, if the lane marking has changed since the generation of the map information due to the repainting of the lane marking, the left and right lane markings in the road image do not match the left and right lane markings in the map information, so there are multiple matching value peaks. Arise. Alternatively, if the left and right lane lines of the map information to be collated are erroneous due to a large vertical position error included in the position on the map of the vehicle recognized by the vehicle position recognition unit 11, Since the lane markings and the lane markings on the left and right of the map information do not match, a plurality of matching value peaks occur.

  The lateral position estimation unit 13 determines whether or not there are two or more matching value peaks based on a matching result between the left and right lane markings of the road surface image and the left and right lane markings of the map information. The lateral position estimation unit 13 does not necessarily need to graph the verification values and count the peaks, and can determine the values by a known method.

  If it is determined that there are two or more matching value peaks, the lateral position estimation unit 13 performs lateral position estimation separately for the left lane line and the right lane line. In the case of false lane markings, there are fewer lane marking candidates than true lane markings, so they can be excluded by dividing them into left and right. The lateral position estimation unit 13 estimates the first lateral position, which is the lateral position of the vehicle, by comparing the left lane line of the road surface image with the left lane line of the map information.

  Here, FIG. 3A is a plan view showing a case where the left lane line of the road surface image is collated with the left lane line of the map information. In FIG. 3A, the vehicle M, the left lane line candidate point PA included in the road image, the right lane line candidate point PB included in the road image, the left lane line LA included in the map information, and The right lane marking LB included in the map information is shown. In addition, a first lateral position W1 that is the lateral position of the vehicle M when the left lane line of the road surface image and the left lane line of the map information are collated is shown. In FIG. 3A, the lane markings recognized from the road surface image are represented as candidate points PA and PB, but may be recognized as lines instead of points.

  As shown in FIG. 3A, when the lateral position estimation unit 13 determines that there are two or more peaks of the matching value (the left and right lane lines of the road image do not match the left and right lane lines of the map information). ), The first lateral position W1 is estimated from the comparison between the candidate point PA on the left lane line of the road image and the left lane line LA in the map information. That is, the lateral position estimation unit 13 matches the position of the candidate point PA of the left lane line of the road surface image that is a relative position with respect to the vehicle M with the left lane line LA of the map information, thereby A first lateral position W1, which is a lateral position, is estimated. At this time, the lateral position estimation unit 13 need not consider collation of the right lane marking of the road surface image.

  FIG. 3B is a plan view showing a case where the right lane line of the road surface image is collated with the right lane line of the map information. As shown in FIG. 3B, when the first lateral position W1 is estimated, the lateral position estimating unit 13 determines whether the right lane line candidate point PB of the road surface image and the right lane line LB of the map information. A second lateral position W2 is estimated from the verification. At this time, the lateral position estimation unit 13 does not need to consider collation of the left marking line of the road surface image. Note that either the first lateral position W1 or the second lateral position W2 may be estimated first.

  When the first lateral position W1 and the second lateral position W2 are estimated by the lateral position estimating unit 13, the difference determining unit 14 determines the difference between the first lateral position W1 and the second lateral position W2 as a determination threshold value. It is determined whether it is above. The determination threshold is a preset threshold. The determination threshold value can be set in consideration of, for example, an allowable error that is permitted in the lane marking verification.

  When determining that the difference between the first lateral position W1 and the second lateral position W2 is equal to or greater than the determination threshold, the difference determination unit 14 recognizes a recognition width shape that is the width shape of the left and right lane markings from the road surface image. To do. The width shape is a shape of the traveling lane focusing on the width (lane width) of the left and right lane lines recognized from the road surface image, and corresponds to a change in the width of the left and right lane lines in the extending direction of the traveling lane. The recognized width shape means the width shape of the left and right lane markings recognized from the road surface image.

  Here, FIG.3 (c) is a top view for demonstrating recognition of the width shape of the left and right division lines based on a road surface image. FIG. 3C shows an approximate line FA of the left lane line recognized from the road surface image and an approximate line FB of the left lane line recognized from the road surface image. The approximate line FA of the left lane line is an approximate line corresponding to the candidate point PA of the left lane line. Similarly, the approximate line FB of the right plot line is an approximate line corresponding to the candidate point PB of the right plot line. As shown in FIG. 3C, the difference determination unit 14 recognizes the approximate lines FA and FB of the left and right lane markings from the road surface image, so that the width (lane width) of the approximate lines FA and FB of the left and right lane markings. ) Recognize WD.

  FIG. 4A is a diagram illustrating an example of a recognized width shape of a road surface image. As shown to Fig.4 (a), the difference determination part 14 recognizes a recognition width shape by making the change of the width WD in the extending direction of a driving lane into a width shape.

  For example, the difference determination unit 14 stores the width WD in the vicinity of the vehicle M recognized from the road surface image, so that the recognition width of the left and right lane markings can be determined from the change in the width WD obtained as the vehicle M travels. Recognize the shape. When the difference determination unit 14 determines that the difference between the first lateral position W1 and the second lateral position W2 is equal to or greater than the determination threshold, the difference determination unit 14 starts storing the width WD. After that, the difference determination unit 14 determines that the lateral position estimation unit 13 determines that there are not two or more matching value peaks between the left and right lane markings of the road image and the map information left or right, or the first When it is determined that the difference between the horizontal position W1 and the second horizontal position W2 is not greater than or equal to the determination threshold, the storage of the width WD is terminated. The difference determination unit 14 recognizes the recognized width shape from the stored change (transition) of the width WD.

  Note that the difference determination unit 14 does not necessarily need to continuously store the width WD. The difference determination unit 14 may recognize the recognition width shape from the road surface image obtained by one imaging. That is, the difference determination unit 14 recognizes the change in the width shape in the extending direction of the traveling lane from the left and right lane markings (a lane marking of a certain length) included in the road surface image obtained by one imaging. By doing so, the recognition width shape may be recognized.

  When the difference determination unit 14 determines that the difference between the first lateral position W1 and the second lateral position W2 is greater than or equal to the determination threshold, the comparison target range setting unit 15 recognizes the vehicle recognized by the vehicle position recognition unit 11. Based on the position of M on the map, the comparison target range including the range on the travel lane on the map used for the collation of the lateral position estimation unit 13 is set. The comparison target range is a range on the map for comparison with the recognized width shape recognized by the difference determination unit 14.

  For example, the comparison target range setting unit 15 sets a comparison target range based on the vertical position error of the vehicle M. The vertical position error of the vehicle M is the detection of the vehicle speed sensor 3 and the yaw rate sensor 4 when recognizing the position of the vehicle M on the map using the vehicle speed information detected by the vehicle speed sensor 3 and the yaw rate information detected by the yaw rate sensor 4. This is a position error in the extending direction of the traveling lane caused by the error. The vertical position error of the vehicle M is calculated from detection errors of the vehicle speed sensor 3 and the yaw rate sensor 4 by a known method. The vertical position error of the vehicle M may be obtained in consideration of the traveling locus of the vehicle M (the own vehicle estimated locus) in addition to the detection errors of the vehicle speed sensor 3 and the yaw rate sensor 4.

  The comparison target range setting unit 15 sets the comparison target range as a larger range as the vertical position error of the vehicle M is larger. For example, the comparison target range setting unit 15 sets a comparison target range by adding a length range corresponding to the vertical position error before and after the range in which the width WD is stored by the difference determination unit 14. The comparison target range setting unit 15 recognizes the map width shape that is the width shape of the set comparison target range.

  Here, FIG.4 (b) is a figure which shows the 1st example of the map width shape in a comparison object range. 4B, the comparison target range C, the reference range C1 corresponding to the range in which the width WD is stored by the difference determination unit 14, the forward additional range C2 corresponding to the vertical position error, and the backward addition corresponding to the vertical position error. A range C3 is indicated. The comparison target range C is divided into a reference range C1, a front additional range C2, and a rear additional range C3. The reference range C1 corresponds to the range on the travel lane on the map used for collation by the lateral position estimation unit 13.

  In FIG. 4B, the width WM1 in the reference range C1 and the front additional range C2 is the same length. On the other hand, the width WM2 of the rear additional range C3 is smaller than the width WM1, and has a lane shape in which the width increases to the width WM1 in the middle. The width WM2 of the rear additional range C3 is the same length as the width WD of the left and right lane markings recognized from the road surface image shown in FIG. FIG. 4B shows the coincidence range CT. The coincidence range CT is a range of a width shape that matches the recognized width shape shown in FIG. Details will be described later.

  The comparison target range setting unit 15 sets the comparison target range C by adding the front additional range C2 and the rear additional range C3 before and after the reference range C1 based on the vertical position error of the vehicle M. The comparison target range setting unit 15 adds longer ranges as the front additional range C2 and the rear additional range C3 as the vertical position error of the vehicle M increases.

  Note that the comparison target range setting unit 15 does not necessarily need to set the comparison target range C based on the vertical position error. The comparison target range setting unit 15 may set a range of a certain distance in the front-rear direction from the position on the map of the vehicle M recognized by the vehicle position recognition unit 11 as the comparison target range C.

  When the difference determination unit 14 determines that the difference between the first horizontal position W1 and the second horizontal position W2 is equal to or greater than the determination threshold, the match determination unit 16 determines whether the map information in the comparison target range C It is determined whether or not there is a portion that matches the recognized width shape recognized from the road surface image in the map width shape that is the width shape of the lane marking.

  In the situation shown in FIG. 4B, the coincidence determination unit 16 compares the comparison target range C with the coincidence range CT in which the recognized width shape matches the width shape shown in FIG. It is determined that there is a part in the map width shape in the target range C that matches the recognized width shape recognized from the road surface image.

  On the other hand, FIG. 4C is a diagram illustrating a second example of the map width shape in the comparison target range C. In the comparison target range C shown in FIG. 4C, there is no place where the lane width changes, and the map width shape has a uniform width WM in the extending direction of the traveling lane. In the situation shown in FIG. 4C, the coincidence determination unit 16 does not include a range (part) where the recognized width shape and the width shape shown in FIG. It is determined that there is no portion in the map width shape in C that matches the recognized width shape recognized from the road surface image.

  When the coincidence determination unit 16 determines that the map width shape in the comparison target range C has a portion that matches the recognized width shape recognized from the road surface image, the vertical position error calculation unit 17 determines the vertical position error of the vehicle M. Therefore, the vertical position error of the vehicle M is calculated. The vertical position error calculation unit 17 calculates the vertical position error of the vehicle M from a portion that matches the recognized width shape in the map width shape.

  In the situation shown in FIG. 4B, the vertical position error calculation unit 17 can be said that the vehicle M is positioned in the matching range CT instead of the reference range C1, and therefore the vertical position of the vehicle M according to the matching range CT on the map. Calculate the error. As an example, the vertical position error calculation unit 17 applies the position of the vehicle M with respect to the left and right lane markings in FIGS. 3C and 4A and the coincidence range CT in FIG. The vertical position error can be calculated. The method for calculating the vertical position error is not particularly limited.

  When the coincidence determination unit 16 does not determine that the map width shape in the comparison target range C matches the recognition width shape recognized from the road surface image, the lane marking change recognition unit 18 generates map information. It is recognized that the lane marking has been changed. The lane line change recognizing unit 18 performs collation where the vertical position error of the vehicle M and a change in the lane line due to repainting are considered as causes of inconsistency between the left and right lane lines in the road surface image and the left and right lane lines in the map information. Even when compared with the comparison target range C that is larger than the reference range C1, which is the target, there is no portion that matches the recognized width shape, so it is recognized that the lane line has been changed since the generation of the map information by repainting the lane line. To do.

  When the lane marking change recognition unit 18 recognizes that the lane marking has been changed since the generation of the map information, the lane marking change recognition unit 18 outputs a recognition result. The output target is not particularly limited, and can be output to an automatic driving system, a driving support system, a map management server that manages map information, or the like.

[Processing of own vehicle position estimation device]
Hereinafter, the process of the vehicle position estimation apparatus 100 according to the present embodiment will be described with reference to the drawings. FIG. 5 is a flowchart showing a lane width storage start process. The flowchart shown in FIG. 5 is executed while the vehicle M is traveling.

  As shown in FIG. 5, the ECU 10 of the vehicle position estimation device 100 generates a road surface image by the road surface image generation unit 12 as S10. The road surface image generation unit 12 generates a road surface image including the left and right lane markings of the travel lane on which the vehicle travels based on the captured image of the camera 2. Thereafter, the ECU 10 proceeds to S12.

  In S <b> 12, the ECU 10 collates the left and right lane markings of the traveling lane included in the road surface image with the left and right lane markings of the traveling lane included in the map information of the map database 5 by the lateral position estimation unit 13. Thereafter, the ECU 10 proceeds to S14.

  In S <b> 14, the ECU 10 determines whether the lateral position estimation unit 13 has two or more verification value peaks. The lateral position estimation unit 13 determines whether or not there are two or more matching value peaks based on a matching result between the left and right lane markings of the road surface image and the left and right lane markings of the map information. If it is not determined that there are two or more peaks of the collation value (S14: NO), the ECU 10 ends the current process. The ECU 10 repeats the process from S10 after a predetermined time has elapsed. If it is determined that there are two or more verification value peaks (S14: YES), the ECU 10 proceeds to S16.

  In S16, the ECU 10 estimates the first lateral position W1 and the second lateral position W2 by the lateral position estimation unit 13. The lateral position estimation unit 13 estimates the first lateral position W1 from matching between the candidate point PA on the left lane line of the road image and the left lane line LA in the map information. The lateral position estimation unit 13 estimates the second lateral position W2 from collation between the right lane line candidate point PB of the road surface image and the right lane line LB of the map information. Thereafter, the ECU 10 proceeds to S18.

  In S18, the ECU 10 determines whether the difference between the first lateral position W1 and the second lateral position W2 is greater than or equal to a determination threshold by the difference determination unit 14. When it is determined that the difference between the first lateral position W1 and the second lateral position W2 is not equal to or greater than the determination threshold (S18: NO), the ECU 10 ends the current process. The ECU 10 repeats the process from S10 after a predetermined time has elapsed. When it is determined that the difference between the first lateral position W1 and the second lateral position W2 is equal to or greater than the determination threshold (S18: YES), the ECU 10 proceeds to S20.

  In S20, the ECU 10 causes the difference determination unit 14 to start storing the lane width. The storage of the lane width is continued until the storage ends in the width shape comparison process described later. After starting the storage of the lane width, in S21, the ECU 10 starts a width shape comparison process.

  FIG. 6 is a flowchart showing the width shape comparison process. S30 to S38 shown in FIG. 6 are the same processes as S10 to S18 shown in FIG.

  As shown in FIG. 6, when it is not determined in S34 that there are two or more verification value peaks (S34: NO), the ECU 10 proceeds to S40. If the ECU 10 does not determine in S38 that the difference between the first lateral position W1 and the second lateral position W2 is greater than or equal to the determination threshold (S38: NO), the ECU 10 proceeds to S40. When it is determined in S38 that the difference between the first lateral position W1 and the second lateral position W2 is equal to or greater than the determination threshold value (S38: YES), the ECU 10 returns to S30 and repeats the process.

  In S <b> 40, the ECU 10 ends the storage of the lane width by the difference determination unit 14 and sets the comparison target range C by the comparison target range setting unit 15. Based on the position of the vehicle M recognized by the vehicle position recognizing unit 11 on the map, the comparison target range setting unit 15 determines the range (reference range C1) on the travel lane on the map used for the collation of the lateral position estimating unit 13. ) And a comparison target range C larger than the range is set. The comparison target range setting unit 15 sets the comparison target range C according to the vertical position error of the vehicle M caused by the detection errors of the vehicle speed sensor 3 and the yaw rate sensor 4. Thereafter, the ECU 10 proceeds to S42.

  In S <b> 42, the ECU 10 determines whether or not there is a portion in the map width shape in the comparison target range C that matches the recognized width shape recognized from the road surface image by the match determination unit 16. When it is determined that there is a portion in the map width shape in the comparison target range C that matches the recognized width shape recognized from the road surface image (S42: YES), the ECU 10 proceeds to S44. If it is not determined that the map width shape in the comparison target range C has a portion that matches the recognized width shape recognized from the road surface image (S42: NO), the ECU 10 proceeds to S46.

  In S <b> 44, the ECU 10 calculates the vertical position error of the vehicle M from the portion that matches the recognized width shape in the map width shape by the vertical position error calculation unit 17. Thereafter, the ECU 10 ends the current process and returns to S10 in FIG. 5 after a predetermined time has elapsed to repeat the process.

  In S <b> 46, the ECU 10 recognizes that the lane line has been changed by the lane line change recognition unit 18 since the map information was generated. The ECU 10 outputs a recognition result that the lane marking has been changed since the generation of the map information to an automatic driving system or the like. Thereafter, the ECU 10 ends the current process and returns to S10 in FIG. 5 after a predetermined time has elapsed to repeat the process.

[Operational effect of own vehicle position estimation device]
According to the vehicle position estimation device 100 according to the present embodiment described above, when it is determined that the difference between the first lateral position W1 and the second lateral position W2 is greater than or equal to the determination threshold value, As a cause of inconsistency between the lane marking and the left and right lane markings in the map information, the vertical position error of the vehicle M and the change in the lane marking due to repainting can be considered. If there is no portion that matches the recognized width shape even when compared with C, it can be recognized that the lane line has been changed since the generation of the map information by repainting the lane line. In addition, according to the vehicle position estimation device 100, when there is a portion that matches the recognized width shape as compared with the comparison target range C, the vertical position error of the vehicle M can be calculated from the matching portion. .

  As mentioned above, although preferred embodiment of this invention was described, this invention is not limited to embodiment mentioned above. The present invention can be implemented in various forms including various modifications and improvements based on the knowledge of those skilled in the art including the above-described embodiments.

  For example, the host vehicle position estimation apparatus 100 does not necessarily have to determine whether there are two or more verification value peaks. The vehicle position estimation device 100 estimates the first lateral position W1 and the second lateral position W2 by dividing the left lane line and the right lane line regardless of the peak of the collation value, and the difference determination unit 14 The determination may be made. Further, the calculation of the vertical position error by the vertical position error calculation unit 17 is not essential. ECU10 should just recognize whether the lane line changed from the time of the generation of map information by the lane line change recognition part 18 at least.

  DESCRIPTION OF SYMBOLS 1 ... GPS receiving part, 2 ... Camera, 3 ... Vehicle speed sensor, 4 ... Yaw rate sensor, 5 ... Map database, 10 ... ECU, 11 ... Vehicle position recognition part, 12 ... Road surface image generation part, 13 ... Lateral position estimation part, DESCRIPTION OF SYMBOLS 14 ... Difference determination part, 15 ... Comparison object range setting part, 16 ... Match determination part, 17 ... Vertical position error calculation part, 18 ... Dividing line change recognition part, 100 ... Own vehicle position estimation apparatus.

Claims (1)

A vehicle position estimation device for estimating a position of a vehicle on a map,
A vehicle position recognition unit for recognizing a position of the vehicle on a map based on a measurement result of a measurement unit mounted on the vehicle;
A road surface image generation unit that generates a road surface image including left and right lane markings of the traveling lane of the vehicle, based on a captured image of the vehicle camera;
A map database for storing map information including location information of lane markings;
By collating the left lane line of the travel lane included in the road image with the left lane line of the travel lane included in the map information, a first lateral position that is a lateral position on the map of the vehicle is determined. A second position which is a lateral position on the map of the vehicle by estimating and comparing the right lane line of the travel lane included in the road surface image with the right lane line of the travel lane included in the map information. A lateral position estimation unit for estimating the lateral position of
A difference determination unit that determines whether or not a difference between the first lateral position and the second lateral position is greater than or equal to a determination threshold;
Comparison target range setting that sets a comparison target range that includes a range on the travel lane on the map used for the collation and that is larger than the range based on the position of the vehicle recognized by the vehicle position recognition unit And
When the difference determination unit determines that the difference between the first horizontal position and the second horizontal position is equal to or greater than the determination threshold, the left and right division lines of the map information in the comparison target range A match determination unit that determines whether or not there is a portion that matches the recognized width shape that is the width shape of the left and right lane markings recognized from the road surface image in the map width shape that is a width shape;
When it is determined by the coincidence determination unit that there is a portion that matches the recognized width shape in the map width shape, the vertical position of the vehicle from the portion that matches the recognized width shape in the map width shape A vertical position error calculation unit for calculating an error;
A lane line change recognition unit that recognizes that the lane line has been changed since the generation of the map information when the match determination unit determines that there is no portion that matches the recognition width shape in the map width shape. When,
A vehicle position estimation device comprising:

Images