JP2005173917A - Device and method for detecting branching path - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a branching path detecting device/method, with which a branching path can more speedily be detected. <P>SOLUTION: The branching path detecting device loaded on a self-vehicle is provided with a laser radar 1, a branching path information acquiring part 31 for acquiring branching path information on the branching path, existing in front of the self-vehicle and a branching path calculating part 32 detecting the branching path existing in front of the vehicle by a prescribed operation method, based on a detection point which the laser radar 1 detects, when the branching path information acquiring part 31 acquires the branching path information. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a branch path detection device and a branch path detection method.

  Conventionally, for example, a technique described in Patent Literature 1 is known as a technique for detecting a branch road existing ahead of the host vehicle.

In the technique described in Patent Document 1, an image in which a state in front of the host vehicle is drawn is obtained using a camera mounted on the host vehicle, and display lines drawn on the road surface are obtained based on the acquired image. To detect. And the branch path and the approach path | route of the own vehicle are detected from the increase of the space | interval of the said detected display line. Then, based on the detection result, control of the host vehicle, for example, deceleration control is performed.
JP 2000-105898 A

  However, the technique described in Patent Document 1 has the following problems. That is, in this technique, since a branch path is detected based on an image acquired by a camera, the branch path cannot be detected unless the branch path is within the imaging range of the camera. For this reason, since the branch road cannot be detected until the own vehicle approaches the branch road, there is a possibility that the control of the own vehicle is delayed.

  The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide a branch path detection device and a branch path detection method capable of detecting a branch path more quickly. There is to do.

  In order to achieve the above object, the invention described in the claims of the present application is a branch path detection device mounted on the host vehicle, which emits while scanning the transmission wave in front of the host vehicle and exists in front of the host vehicle. Object detection means for detecting a reflected wave from the object and detecting a detection point indicating the position of the object with respect to the own vehicle based on the detected reflected wave, and branching road information relating to a branch road existing in front of the own vehicle When the branch path information acquisition means to acquire and the branch path information acquisition means acquire the branch path information, based on the detection point detected by the object detection means, it exists in front of the host vehicle by a predetermined calculation method. And a branch path calculating means for detecting the branch path.

  In the invention described in the claims of the present application, the following effects can be mainly obtained. That is, in the present invention, the branch path is detected based on the detection point detected by the object detection means. Here, the detectable range of the object detecting means is wider than the detectable range of the camera. Therefore, the branch path detection device can detect the branch path start position more quickly than the conventional one.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, a first embodiment of the invention will be described with reference to the drawings. Here, a case where a scanning laser radar (hereinafter referred to as a laser radar) is used as the object detection means will be described. The present invention is applied to, for example, a branch path detection apparatus as shown in FIG.

[Configuration of branch path detection device]
FIG. 1 is a block diagram illustrating a functional configuration of the branch path detection device, and FIG. 2 is a diagram illustrating an installation position of the laser radar 1. This branch path detection device is provided by a laser radar (object detection means) 1 provided in front of the host vehicle, a vehicle behavior detection unit 2 for detecting behavior during traveling, a laser radar 1 and a vehicle behavior detection unit 2. Based on the information, it is configured to include a calculation unit 3 that performs various processes for detecting a branch road existing ahead of the host vehicle, and a presentation unit 4.

  The vehicle behavior detection unit 2 includes a GPS reception sensor that receives a GPS (Global Positioning System) signal for detecting the position of the host vehicle 11, that is, the host vehicle position, a shift position sensor that detects a shift position, A wheel speed sensor that detects the wheel speed of the rear wheel, a steering angle sensor that detects the steering angle of the host vehicle 11, a vehicle behavior calculation device (own vehicle position detection means) 21, and a map storage device 22 that stores a map database. Is provided.

  The arithmetic unit 3 is composed of a microcomputer (CPU), a RAM (Random Access Memory), a ROM (Read Only Memory), an input / output I / F, and the like mounted in the vehicle. A road information acquisition unit (branch path information acquisition unit) 31, a branch path calculation unit (branch path calculation unit) 32, a branch structure detection unit 33 (branch structure detection unit), and a travel path determination unit (travel path determination) Means) and the functional blocks of the memory 35. The branch information acquisition unit 31 includes an object data income unit (object data acquisition unit) 311, a branch guide sign detection unit (branch guide sign detection unit) 312, and a branch point search unit (branch point search unit) 313. It comprises. The branch path calculation unit 32 includes a road boundary line segment detection unit (road boundary line segment detection unit) 321, a road boundary distance calculation unit (road boundary distance calculation unit) 322, and a branch path start position detection unit (branch path start position). Detection means) 323. Further, the branch structure detection unit 33 includes a branch confirmation line segment calculation unit (branch confirmation line segment calculation unit) 331 and a branch structure determination unit (branch structure determination unit) 332.

[Branch path detector operation]
Next, processing by the branch path detection device will be described with reference to FIGS. 1 to 8, taking as an example a case where a branch path exists in front of the host vehicle 11. Here, FIG. 3 to FIG. 7 are explanatory diagrams showing detection positions of detection points and the like.

  As shown in FIG. 2, the laser radar 1 is installed at the front portion of the host vehicle 11. Further, the laser beam is scanned in a predetermined scanning range by changing the optical axis at a predetermined angle on the scanning surface. Thereby, the laser radar 1 irradiates an object existing in the scan range with laser light.

  Then, the laser radar 1 acquires a reflected signal based on the light intensity of the reflected laser light by detecting the reflected laser light (reflected wave) reflected by the emitted laser light irradiating an object existing ahead. To do. And based on the acquired reflected signal, the detection point which shows the position of the object with respect to the own vehicle 11, and the position of a detection point are detected. As shown in FIG. 2, the position of the detection point is as follows: the tip of the laser radar 1 is the origin, the direction parallel to the traveling direction of the host vehicle 11 is the Z axis, the direction perpendicular to the road surface is the Y axis, the Y axis, and the Z axis. It is expressed as a coordinate in a three-dimensional space with the direction perpendicular to the axis as the X axis.

  In this example, as shown in FIG. 3, the laser radar 1 (indicated by LR in the figure) detects detection points a to n that indicate the positions of the objects A to D with respect to the host vehicle 11. In the figure, the detection points a to n are plotted according to the positions obtained for the objects A to D.

  The laser radar 1 generates distance measurement information related to the position of the detection point and stores it in the memory 35 of the calculation unit 3.

  The vehicle behavior calculation device 21 calculates the own vehicle position, the own vehicle traveling direction, the direction of the own vehicle 11, and the moving distance using sensor signals from the GPS reception sensor, the shift position sensor, the wheel speed sensor, and the steering angle sensor. To do. Then, vehicle travel information relating to the calculated (detected) host vehicle position, host vehicle traveling direction, host vehicle direction, and travel distance is generated and stored in the memory 35 of the calculation unit 3.

  The object data acquisition unit 311 acquires distance measurement information and vehicle travel information from the memory 35, and performs the following processing based on the acquired information.

  That is, when a plurality of detection points are detected from the same object, the Z coordinates of these detection points (that is, the distance in the Z direction from the vehicle 11 to the detection point) are substantially the same. These detection points move in substantially the same direction. Therefore, the object data acquisition unit 311 groups detection points having the same Z coordinate and the same movement. Specifically, the following processes (a1) to (a2) are performed.

 (a1) The detection points having the same Z coordinate and existing at positions close to each other are grouped to generate a first group (first grouping).

 (a2) First groups that exist in positions close to each other and have the same movement are grouped to generate a second group, and the second group is determined as one object (second grouping). Thereby, the object data acquisition unit 311 detects an object ahead of the host vehicle 11. Then, based on the distribution of detection points, etc., the position of the object, the width of the object, and the number of detection points belonging to the object are calculated, and object data relating to the calculated position of the object is generated in the memory 35. Remember.

  In this example, as shown in FIG. 3, the object data acquisition unit 311 detects an object O (1, t) (corresponding to the object A) composed of detection points a to e, and the object O (1, The position of t) is calculated. Then, object data relating to the calculated position of the object O (1, t) is generated and stored in the memory 35.

  The branch guidance signboard detection unit 312 includes object data previously generated by the object data acquisition unit 311 (hereinafter referred to as “previous object data”) and object data generated by the object data acquisition unit 311 (hereinafter referred to as “current time”). And the threshold values Wt1 and Nt1 are acquired, and the following processing is performed based on the acquired data.

  That is, the branch guide sign detection unit 312 compares the previous object data with the current object data, and extracts object data indicating the same object as the object indicated by the previous object data from the current object data. The extraction process is performed by comparing the position of the object indicated by the current object data with the position of the object indicated by the previous object data.

  Then, the branch guide sign detection unit 312 calculates the relative speed of the object indicated by the current object data based on the extracted current object data and the previous object data, and based on the calculation result, It is determined whether the object indicated by the object data is a stop object. Thereby, the branch guide sign detection unit 312 detects the stopped object.

  The branch guide sign detection unit 312 compares the width of the object determined as a stopped object with the threshold value Wt1, and further compares the number of detection points belonging to the object determined as a stopped object with the threshold value Nt1. To do. As a result, when the width of the object is equal to or greater than the threshold value Wt1 and the number of detection points belonging to the object is equal to or greater than the threshold value Nt1, the branch guidance sign detection unit 312 determines that the object is a branch guidance signboard. . Thereby, the branch guide sign detection unit 312 detects the branch guide sign. Then, the branch guide sign detection unit 312 generates first branch path information related to the detected branch guide sign and stores it in the memory 35.

  In the case shown in this example, as shown in FIG. 3, the branch guidance signboard detection unit 312 has the same object (this example) as the object O (1, t−1) indicated by the previous object data among the current object data. Then, the object data indicating the object O (1, t)) is extracted. Then, based on the extracted current object data and the previous object data, it is determined whether or not the object O (1, t) is a branch guidance signboard.

  The branch point search unit 313 acquires vehicle travel information from the memory 35, and further acquires map data around the vehicle position from the map storage device 22. Based on the acquired vehicle travel information and map data, a branch road is detected from the front of the host vehicle 11. Then, second branch path information relating to the detected branch path is generated and stored in the memory 35. Further, the branch point search unit 313 stores the acquired map data in the memory 35 as road shape information. In this example, it is assumed that the branch point search unit 313 detects a branch road and the road shape information indicates a straight road.

  The road boundary line detection unit 321 acquires road shape information, distance measurement information, vehicle travel information, first branch road information, and data of threshold values Zt1, Zt2, and Xt1 from the memory 35. Based on the information or the like, a line segment indicating a branch path is detected.

  First, detection of a right curve line segment indicating a right curve will be described.

  When the road shape information indicates a right curve road, the road boundary line detection unit 321 starts from the detection point with the largest X coordinate among the detection points that are not the target of the branch guide signboard under the following conditions ( A pair of detection points satisfying b1) to (b2) (one detection point and another detection point adjacent to the left side of the one detection point) is searched.

  (B1) The Z coordinate of the other detection point decreases in the range of Zt1 or more and Zt2 or less with respect to the Z coordinate of one detection point.

  (B2) The X coordinate of another detection point is decreased by Xt1 or more with respect to the X coordinate of one detection point.

  As a result, when there is a pair of detection points that satisfy the conditions (b1) to (b2), the road boundary line segment detection unit 321 determines that these detection points constitute the same right curve line segment. And among the detection points constituting the same right curve line segment, the right curve line segment with the leftmost detection point as the closest point and the right end detection point as the farthest point, that is, the right curve line segment connecting these detection points is (Hereinafter referred to as “right curve road calculation process”). Note that detection points that do not satisfy the conditions (b1) to (b2) are detection points that constitute other line segments or independent detection points that do not constitute line segments. Then, it is determined whether or not the calculated closest point of the right curve line segment and the detection point adjacent to the left side of the closest point satisfy the conditions (b1) to (b2). A right curve road calculation process is performed to extend the right curve road segment. The above processing is repeated until there is no detection point that constitutes the right curve line segment. Thereby, the road boundary line segment detection unit 321 detects the right curve line segment.

  Then, the road boundary line segment detection unit 321 repeats the above processing until there is no detection point belonging to any right curve line segment.

  Then, the road boundary line segment detection unit 321 selects one of the calculated right curve line segments that includes three or more detection points, and relates to the nearest point and the farthest point of the selected right curve line segment. Right curve route segment information is generated and stored in the memory 35.

  Next, detection of the left curve line segment indicating the left curve road will be described.

  When the road shape information indicates a left curve road, the road boundary line segment detection unit 321 starts from the detection point having the largest X coordinate among the detection points that are not the target of the branch guidance signboard under the following condition (c1) A pair of detection points satisfying (c2) (one detection point and another detection point adjacent to the left side of the one detection point) is searched.

  (C1) The Z coordinate of another detection point increases in a range from Zt1 to Zt2 with respect to the Z coordinate of one detection point.

  (C2) The X coordinate of another detection point is decreased by Xt1 or more with respect to the X coordinate of one detection point.

  As a result, when there is a pair of detection points satisfying the conditions (c1) to (c2), it is determined that these detection points constitute the same left curve line segment. And among the detection points composing the same left curve line segment, the left curve line segment with the right end detection point as the nearest point and the left end detection point as the farthest point, that is, the left curve line segment connecting these detection points is (Hereinafter referred to as “left curve road calculation process”). The detection points that do not satisfy the conditions (c1) to (c2) are detection points that constitute other line segments or independent detection points that do not constitute a line segment. Then, it is determined whether the farthest point of the calculated left curve line segment and the detection point adjacent to the left side of the farthest point satisfy the conditions (c1) to (c2). A left curve road calculation process is performed for the point to extend the left curve road segment. The above processing is repeated until there are no detection points that constitute the left curve line segment. Thereby, the road boundary line segment detection unit 321 detects the left curve line segment.

  Then, the road boundary line segment detection unit 321 repeats the above processing until there is no detection point belonging to any left curve line segment.

  Then, the road boundary line segment detection unit 321 selects one of the calculated left curve line segments that includes three or more detection points, and the left side regarding the nearest point and the farthest point of the selected left curve line segment. Curve route segment information is generated and stored in the memory 35.

  Next, detection of a straight line segment indicating a straight line will be described.

  When the road shape information indicates a curved road, the road boundary line segment detection unit 321 starts from the detection point having the largest X coordinate among the detection points that are not targets of the curved road or the branch guide signboard. Then, a pair of detection points satisfying the following conditions (d1) to (d2) (one detection point and another detection point adjacent to the left side of the one detection point) is searched.

  On the other hand, when the road shape information indicates a straight road, the road boundary line detection unit 321 determines the following conditions from the detection point having the largest X coordinate among the detection points that are not the target of the branch guide signboard. A pair of detection points satisfying (d1) to (d2) (one detection point and another detection point adjacent to the left side of the one detection point) is searched.

  (D1) The difference between the Z coordinate of another detection point and the Z coordinate of one detection point is a value not less than Zt1 and not more than Zt2.

  (D2) The difference between the X coordinate of another detection point and the X coordinate of one detection point is a value equal to or less than Xt1.

  As a result, when there is a pair of detection points satisfying the conditions (d1) to (d2), it is determined that these detection points constitute the same straight line segment. Of the detection points constituting the same straight line segment, a straight line segment having a detection point close to the own vehicle 11 as the nearest point and a point far from the own vehicle 11 as the farthest point, that is, a straight line connecting these detection points. Minutes are calculated (hereinafter referred to as “straight path calculation processing”). The detection points that do not satisfy the conditions (d1) to (d2) are detection points that constitute other line segments or independent detection points that do not constitute a line segment. Then, it is determined whether or not the calculated detection point at the left end and the detection point adjacent to the detection point on the left side satisfy the conditions (d1) to (d2). To extend the straight line segment. The above processing is repeated until there are no detection points constituting the straight line segment. Thereby, the road boundary line segment detection part 321 detects a straight line segment.

  Then, the road boundary line segment detection unit 321 repeats the above processing until there is no detection point belonging to any straight line segment.

  Then, the road boundary line segment detection unit 321 selects a straight line segment that includes three or more detection points from the calculated straight line segment, and the straight line segment regarding the nearest point and the farthest point of the selected straight line segment. Information is generated and stored in the memory 35.

  In this example, as shown in FIG. 4, the road boundary line segment detection unit 321 detects a straight line segment L1 from the detection points g, h, i, and detects a straight line segment L2 from the detection points j, k, l, n. To do.

  The road boundary distance calculation unit 322 acquires distance measurement information, vehicle travel information, right curve route segment information, left curve route segment information, and straight route segment information from the memory 35, and based on the acquired information, the following Perform the process.

  First, the road boundary distance calculation unit 322 calculates an estimated road center line indicating the course of the host vehicle 11 using the following formulas (1) to (2) based on the vehicle travel information, and calculates the estimated The distance in the X direction from the runway center line L to the detection points constituting the line segment detected above, that is, the road boundary distance is calculated.

1 / RH = ((1 / LH) / (1 + AH * VH ^ 2)) * (δ / NH) (1)
AH =-(M / 2LH ^ 2) * ((LF * KF-LR * KR) / (KF * KR)) (2)
Here, RH is the curvature of the estimated runway center line L. Therefore, the road boundary distance calculation unit 322 calculates the estimated runway center line L based on the value of the RH, the position of the host vehicle 11, and the like.

  LH: wheel base, VH: vehicle speed, δ: rudder angle, NH: steering gear ratio, M: vehicle weight, LF / LR: front / rear wheel axle center point distance, KF / KR: front / rear tire cornering stiffness It is. The values of VH and δ are acquired from the vehicle travel information. The values of the other parameters are stored as fixed values in the memory 35, and are acquired from the memory 35 by the road boundary distance calculation unit 322 when calculating the estimated runway center.

  Then, the road boundary distance calculation unit 322 generates road boundary distance information related to the calculated road boundary distance and stores it in the memory 35. The road boundary distance may be an X-direction distance from the coordinate point on the line segment to the estimated runway center line L.

  The branch road start position detection unit 323 acquires distance measurement information, road boundary distance information, and threshold value T1 data from the memory 35, and performs the following processing based on the acquired information and the like.

  That is, the branch road start position detection unit 323 determines the road boundary distance, the road boundary distance calculated for the detection point closest to the host vehicle 11 (hereinafter referred to as “reference boundary distance”), and the threshold value T1. , Compare. As a result, when there is a detection point where the road boundary distance is greater than the reference boundary distance and the threshold value T1, the position of the detection point closest to the host vehicle 11 is determined as the branch road start position. To do. That is, it is determined that a branch road exists in front of the host vehicle 11 and an attachment road has appeared. This is because the road structure installed at the branch road start position is installed outside the road structure installed at another position. Thereby, the branch path start position detector 323 detects the branch path start position.

  In this example, as illustrated in FIG. 5, the branch path start position detection unit 323 determines the position of the detection point n as the branch path start position.

  Then, the branch path start position detection unit 323 generates branch path start position information related to the detected branch path start position and stores it in the memory 35. Therefore, the branch path calculation unit 32 detects the branch path by detecting the line segment indicating the branch path and the branch path start position.

  The branch confirmation line segment calculation unit 331 acquires distance measurement information and branch path start position information, and performs the following processing based on the acquired information. That is, the branch confirmation line segment calculation unit 331 extracts a detection point having the same road boundary distance as the road boundary distance of the detection point closest to the host vehicle. Of the extracted detection points, the difference between the X coordinate of the detection point closest to the host vehicle 11 and the X coordinate of the detection point furthest from the host vehicle 11, that is, the X direction distance difference is a predetermined value ( For example, when it is less than 3 (m)), it is determined that the extracted detection points are arranged on a straight line. Then, the extracted detection points are subjected to regression analysis to calculate an approximate curve Ln1 represented by the following equation (3). Then, first branch confirmation line segment information regarding the calculated approximate curve Ln1 is generated and stored in the memory 35.

X = a + b * Z (3)
On the other hand, when the X-direction distance difference is greater than or equal to the predetermined value, the branch confirmation line segment calculation unit 331 determines that the extracted detection points are arranged on the curve. And the approximated curve Ln2 shown by the following formula | equation (4) is calculated by carrying out the multiple regression analysis of the said detected detection point. Then, second branch confirmation line segment information regarding the calculated approximate curve Ln2 is generated and stored in the memory 35.

X = a + b * Z + c * Z ^ 2 (4)
The branch structure determination unit 332 acquires distance measurement information and branch confirmation line segment information, and performs the following processing based on the acquired information. That is, the branch structure determination unit 332 extends the approximate curves Ln1 to Ln2 calculated by the branch confirmation line segment calculation unit 331 in the forward direction of the host vehicle 11.

  Then, the branch structure determination unit 332 determines whether there are detection points that do not belong to any of the right curve line segment, the left curve line segment, and the straight line segment on the extended approximate curves Ln1 to Ln2. As a result, when the detection point is present, the detection point closest to the host vehicle 11 among these detection points is defined as a branch structure constituting the starting point of the branch road (that is, a road structure existing at a connection portion between roads). Among these, it is determined that the road structure is located farther from the vehicle 11 than the branch road start position). This is because the branch structure constituting the starting point of the branch road exists on a curve connecting the other road structures. Thereby, the branch structure determination unit 332 detects the branch structure. Then, branch structure information related to the branch structure is generated and stored in the memory 35. In addition, when the detection point used as a branch structure does not exist, the branch structure determination part 332 detects a branch structure by performing the process mentioned above about the detection point detected later. In this example, as shown in FIG. 6, the branch structure determination unit 332 extends the approximate curve Ln1 in the forward direction of the host vehicle 11 and exists on the extended approximate curve Ln1 and the right curve route segment. The detection point f that does not belong to any of the left curve line segment and the straight line segment is determined as a branch structure.

  The traveling path determination unit 34 acquires branch structure information and vehicle travel information, and performs the following processing based on the acquired information and the like. That is, the traveling path determination unit 34 calculates the estimated traveling path center line L from the equations (1) and (2) based on the vehicle traveling information, and the calculated estimated traveling path center line L is the branch structure determination unit 332. It is determined which side of the branch road the host vehicle travels depending on which side of the branch structure detected by the above. Thereby, the course of the own vehicle 11 is determined. In this example, as shown in FIG. 7, the traveling path determination unit 34, when the estimated traveling path center line L passes the right side of the detection point f, the host vehicle 11 travels on the right road among the branch roads. If the estimated runway center line L passes the left side of the detection point f, it is determined that the host vehicle 11 travels on the left road of the branch road. Then, the traveling path determination unit 34 generates path information related to the determination result and stores it in the memory 35.

  The memory 35 stores various information generated by the above-described processing and various threshold values. The presentation unit 4 is a display or speaker mounted on the host vehicle 11 and presents the result of the object detection process performed by the calculation unit 3 to an occupant of the host vehicle 11.

[Object detection procedure using branch path detector]
FIG. 8 is a flowchart illustrating a processing procedure when the calculation unit 3 performs the object detection process. It should be noted that the branch path detection device starts the processing after step S1 when the host vehicle 11 is traveling.

  In step S1, the laser radar 1 scans the laser beam within the scan range to obtain a reflection signal, and detects a detection point indicating the position of the object with respect to the host vehicle 11 based on the obtained reflection signal (FIG. 3).

  Next, the laser radar 1 generates distance measurement information related to the position of each detection point and stores it in the memory 35 of the calculation unit 3.

  On the other hand, the vehicle behavior calculation device 21 calculates the own vehicle position, the own vehicle traveling direction, the direction of the own vehicle 11, and the movement distance, and generates vehicle travel information related to the calculated own vehicle position and the like in the memory 35. Remember.

  In step S2, the object data acquisition unit 311 acquires distance measurement information and vehicle travel information from the memory 35, and performs the first grouping process and the second grouping process described above based on the acquired information. As a result, when an object ahead of the host vehicle 11 is detected, the object data acquisition unit 311 calculates the position of the object, the width of the object, and the number of detection points belonging to the object based on the distribution of the detection points. calculate. Next, object data related to the calculated position of the object is generated and stored in the memory 35. Thereafter, the branch path detection device proceeds to the process of step S3. On the other hand, when an object ahead of the host vehicle 11 is not detected, the branch path detection device proceeds to the process of step S4.

  In step S3, the branch guidance signboard detection unit 312 acquires object data and threshold values Wt1 and Nt1 data from the memory 35, and performs the branch guidance signboard detection process described above based on the acquired data (step S3). (See FIG. 3). Next, first branch path information related to the detected branch guide sign is generated and stored in the memory 35.

  In step S4, the branch point search unit 313 acquires vehicle travel information from the memory 35, and further acquires map data around the vehicle position from the map storage device 22. Next, a branch road is detected from the front of the host vehicle 11 based on the acquired vehicle travel information and map data. Next, second branch path information related to the detected branch path is generated and stored in the memory 35. Further, the branch point search unit 313 stores the acquired map data in the memory 35 as road shape information. Thereafter, the branch path detection device proceeds to the process of step 5 when the branch path is detected in step S4, and returns to the process of step 1 when the branch path is not detected.

  In step S5, the road boundary line segment detection unit 321 acquires distance measurement information, road shape information, vehicle travel information, and data of threshold values Zt1, Zt2, and Xt1 from the memory 35, and the acquired information and the like. Based on the above, the right curve route segment, the left curve route segment, and the straight route segment are detected by the above-described processing. That is, when the road shape information indicates a right curve road, the road boundary line segment detection unit 321 performs detection in the order of the right curve line segment detection process and the straight line segment detection process described above, and the road shape information indicates the left curve path. In the case of indicating a road, detection is performed in the order of a left curve line segment detection process and a straight line segment detection process. On the other hand, when the road shape information indicates a straight road, the road boundary line segment detection unit 321 performs only the straight line segment detection process. Next, the road boundary line segment detection unit 321 stores the right curve line segment information, the left curve line segment information, and the straight line segment information generated by these processes in the memory 35.

  In step S6, the road boundary distance calculation unit 322 acquires distance measurement information, vehicle travel information, right curve route segment information, left curve route segment information, and straight route segment information from the memory 35, and the acquired information. Based on the above, the road boundary distance is calculated by the above-described processing. Next, road boundary distance information related to the calculated road boundary distance is generated and stored in the memory 35.

  In step S7, the branch path start position detection unit 323 acquires distance measurement information, road boundary distance information, and threshold value T1 data from the memory 35, and performs the above-described processing based on the acquired information and the like. , To detect the branch path start position. Next, the branch path start position detection unit 323 generates branch path start position information related to the detected branch path start position and stores it in the memory 35. Thereafter, the branch path detection device proceeds to step S8 when the branch path start position is detected in step S7, and returns to the process of step S1 when the branch path start position is not detected.

  In step S8, the branch confirmation line segment calculation unit 331 acquires the distance measurement information and the branch path start position information from the memory 35, and, based on the acquired information, the approximate curve Ln1 or the approximate curve by the above-described processing. Ln2 is calculated. Next, when the approximate curve Ln1 is calculated, the branch confirmation line segment calculation unit 331 generates first branch confirmation line segment information regarding the calculated approximate curve Ln1 and stores it in the memory 35. On the other hand, when the approximate curve Ln2 is calculated, the branch confirmation line segment calculation unit 331 generates second branch confirmation line segment information regarding the calculated approximate curve Ln2 and stores the second branch confirmation line segment information in the memory 35.

  In step S9, the branch structure determination unit 332 acquires the distance measurement information and the branch confirmation line segment information from the memory 35, and the branch structure from the front of the host vehicle 11 by the above-described processing based on the acquired information. Detect objects. When the distance measurement information generated by the process of step S10 described later is stored in the memory 35, the branch structure determination unit 332 generates the branch confirmation line segment information and the process of step S10. The distance measurement information obtained is acquired from the memory 35 and the same processing is performed. Thereby, a branched structure is detected. Thereafter, the branch path detection device proceeds to step S11 when the branch structure is detected in step S9, and proceeds to the process of step S10 when the branch structure is not detected.

  In step 10, the laser radar 1 performs the same process as in step S <b> 1 to generate distance measurement information and stores it in the memory 35. Thereafter, the branch path detection device returns to Step 9.

  In step 11, the traveling path determination unit 34 acquires the branch structure unit information and the vehicle travel information from the memory 35, and based on the acquired information, the vehicle 11 is located on either side of the branch path by the above-described processing. Judge whether to drive. Next, traveling path determination information related to the determination result is generated and stored in the memory 35. Furthermore, the presentation unit 4 acquires the first to second branch road information, the branch road start position information, and the travel path judgment information from the memory 35 and presents them to the passengers of the host vehicle 11 and the like. In addition, the presentation unit 4 outputs the acquired first to second branch road information, branch path start position information, and travel path determination information to a vehicle control device (for example, an inter-vehicle distance control device). Thereby, the vehicle control apparatus can control the own vehicle 11 based on the 1st-2nd branch road information, branch road start position information, and advancing path judgment information. For example, the host vehicle 11 can be decelerated according to the branch path start position. Thereafter, the branch path detection device returns to step S1.

  As described above, in the present embodiment, the branch path detection device detects the branch path (specifically, the line segment indicating the branch path and the branch path start position) based on the detection point detected by the laser radar 1. (Refer to step S1 to step S7 and FIG. 5). Here, the detectable range of the laser radar 1 is wider than the detectable range of the camera. Therefore, the branch path detection device can detect the branch path more quickly than in the prior art.

  Further, since the laser radar 1 also detects the position of each detection point, the branch path detection device can accurately detect the branch path start position (see step S5, step S7 and FIG. 5).

  Further, since the branch road detection device detects the branch road when the branch guide sign is detected and the branch road is detected from the map data, it is clear that there is a branch road ahead of the host vehicle 11. After confirming this, the branch path can be detected. Therefore, the branch path detection device can accurately detect the branch path. Moreover, since the branch path detection device detects the branch guide sign based on the detection point detected by the laser radar 1, the branch guide sign can be detected quickly.

  Further, the branch road detection device detects a right curve line segment, a left curve line segment, and a straight line segment, and based on the detected line segment, each detection point included in the line segment from the estimated runway center line L The road boundary distance to is calculated. Then, a position where the road boundary distance is greater than the road boundary distance calculated for the detection point closest to the host vehicle 11 and the threshold value T1 is detected as a branch path start position (see step S7 and FIG. 5). This is because the road structure installed at the branch road start position is installed outside the road structure installed at another position as described above. Therefore, the branch path detection device can accurately detect the branch path start position.

  In addition, since the branch road detection device calculates the above-described line segment by connecting the detection points according to the road shape indicated by the road shape information, the detection points that are more likely to be road structures are connected. The line segment can be calculated. Therefore, the branch path detection device can detect the branch path start position based on the detection point that is more likely to be a road structure, and therefore can accurately detect the branch path start position.

  Further, the branch path detection apparatus calculates the above-described line segment by connecting the detection points according to the road shape indicated by the road shape information, so that only the detection points that are more likely to be road structures are efficiently used. The line segment can be calculated by extracting to.

  Further, the branch path detection device approximates the curve Ln1 to the detection point having the same road boundary distance as the detection point closest to the host vehicle 11 among the detection points existing closer to the host vehicle 11 than the branch path start position. Ln2 is calculated. And the detection point which exists on the extension line of the calculated approximate curve Ln1-Ln2 is detected as a branch structure. This is because the branch structure exists on the curve connecting the other road structures as described above. Therefore, the branch path detection device can quickly detect the road structure that is the starting point of the branch. Further, since the branch road detection device determines the course of the host vehicle 11 based on the detected road structure and the estimated runway center line L, the branch path of the host vehicle 11 can be quickly determined.

  Further, since the branch path detection device outputs the branch path start position information and the like related to the branch path start position to the vehicle control device, the vehicle control device can control the host vehicle 11 more quickly than in the past. . For example, since the course of the host vehicle is detected more quickly than before, the vehicle control device can decelerate the host vehicle 11 more quickly than before.

  In the present embodiment, the laser radar 1 is used as the object detection means, but other devices such as a millimeter wave radar may be used. In addition, when the first to second branch path information is generated, the branch path detection device performs the branch path detection process, but when any of the first to second branch path information is generated. Alternatively, the branch path detection device may perform a branch path detection process.

It is a block diagram which shows the structure of the branch path detection apparatus which concerns on one embodiment of this invention. It is the side view and top view which show the installation position of a laser radar. It is explanatory drawing which shows the detection position etc. of a detection point. It is explanatory drawing which shows an example of a line segment detection process. It is explanatory drawing which shows an example of a fork road start position detection process. It is explanatory drawing which shows an example of the detection process of a road structure. It is explanatory drawing which shows an example of the process which judges the road where the own vehicle drive | works. It is a flowchart which shows the procedure of the process by a branch path detection apparatus.

Explanation of symbols

1 ... Laser radar (object detection means)
2 ... Vehicle behavior detection unit 21 ... Vehicle behavior calculation device (own vehicle position detection means)
DESCRIPTION OF SYMBOLS 22 ... Map storage device 3 ... Calculation part 31 ... Branch road information acquisition part (branch path information acquisition means)
311: Object data acquisition unit (object data acquisition means)
312 ... Branch guide sign detection unit (branch guide sign detection means)
313: Branch point search unit (branch point search means)
32 ... Branching path calculation unit (branching path calculation means)
321... Road boundary line detection unit (road boundary line detection means)
322 ... Road boundary distance calculation unit (road boundary distance calculation means)
323 ... Branch path start position detector (branch path start position detector)
33 ... Branch structure detector (branch structure detector)
331 ... branch confirmation line segment calculation unit (branch confirmation line segment calculation means)
332 ... Branch structure determination unit (branch structure determination means)
34. Travel path determination unit (travel path determination means)
35 ... Memory 4 ... Presentation section 11 ... Own vehicle

Claims (8)

In the branch path detection device mounted on the host vehicle,
A detection point indicating the position of the object with respect to the host vehicle based on the detected reflected wave, which is detected by detecting the reflected wave from the object existing in front of the host vehicle while emitting the transmission wave in front of the host vehicle. Object detection means for detecting
Branch road information acquisition means for acquiring branch road information relating to a branch road existing ahead of the host vehicle;
When the branch road information acquisition means acquires the branch road information, a branch for detecting a branch road existing ahead of the host vehicle by a predetermined calculation method based on the detection point detected by the object detection means. A branch path detecting device. The branch path detection device according to claim 1,
A branch path detection device comprising: a branch structure detection means for detecting a branch structure constituting the starting point of the branch path existing ahead of the host vehicle based on the branch path detected by the branch path calculation means . In the branch path detection device according to claim 2,
An estimated runway centerline indicating the course of the host vehicle is calculated, and the course of the host vehicle is determined based on the calculated estimated runway centerline and the branch structure detected by the branch structure detecting means. A branch path detecting device having a travel path judging means. In the branch path detection apparatus according to any one of claims 1 to 3,
The branch path information acquisition means includes:
Among the detection points detected by the object detection means, detection points having the same distance from the own vehicle to the detection point and moving in the same manner are grouped and detected as an object existing in front of the own vehicle. Object data acquisition means;
If the object detected by the object data acquisition means is a stop object, the width of the object is a predetermined width or more, and the number of detection points included in the object is a predetermined number or more, the object is branched. A branch path detection apparatus comprising: branch guide sign detection means that determines information about an object determined to be a guide sign and acquires information related to the object determined to be the branch guide sign as the branch path information. In the branch path detection apparatus according to any one of claims 1 to 4,
Provided with own vehicle position detecting means for detecting the position of the own vehicle;
The branch path information acquisition means includes:
Based on the position of the host vehicle detected by the host vehicle position detection means and map data, a branch road is detected from the front of the host vehicle, and information on the detected branch road is acquired as the branch road information. A branch path detection device having a branch point search means. In the branch path detection device according to claim 2 or 3,
The branch path calculation means includes
A road boundary for detecting a first detection line indicating a branch road ahead of the host vehicle by connecting adjacent detection points among detection points detected by the object detection means using a predetermined calculation method. Line segment detection means;
An estimated runway center line indicating the course of the host vehicle is calculated, and a road boundary distance from the calculated estimated runway center line to a coordinate point on the first detection line detected by the road boundary line segment detecting means is calculated. Road boundary distance calculating means;
A branch road that exists in front of the host vehicle at a position where the road boundary distance calculated by the road boundary distance calculating unit is longer than the road boundary distance calculated for the coordinate point near the host vehicle from the first detection line. And a branch path start position detecting means for detecting as a branch path start position at which the start is started. In the branch path detection apparatus according to claim 6,
The branch structure detection means includes
A branch confirmation line segment calculating means for detecting a second detection line by connecting the detection points existing on the own vehicle side with respect to the branch road start position detected by the branch road start position detecting means;
The second detection line detected by the branch confirmation line segment calculation means is extended forward of the host vehicle, exists on the extended second detection line, and is detected by the branch path start position detection means. A branch path detection device comprising: a branch structure determination unit that detects, as the branch structure, a detection point that is located farther from the host vehicle than a branch path start position. A detection point indicating the position of the object with respect to the host vehicle based on the detected reflected wave, which is detected by detecting the reflected wave from the object existing in front of the host vehicle while emitting the transmission wave in front of the host vehicle. A first step of detecting
A second step of acquiring branch road information relating to a branch road existing ahead of the host vehicle;
When the branch road information is acquired in the second step, a branch road existing ahead of the host vehicle is detected by a predetermined calculation method based on the detection point detected in the first step. A branch path detection method comprising: a third step.

Images