JP4832489B2 - Lane judgment device - Google Patents

Description

  The present invention relates to a lane determination device that determines a lane in which a host vehicle travels.

  A navigation system mounted on an automobile has a function of displaying a vehicle position detected by an autonomous method such as GPS (Global Positioning System) or a gyro system together with map information in the vicinity thereof.

  The closer the vehicle position displayed in the navigation system is to the actual vehicle position, the higher the position accuracy, and by outputting a highly accurate vehicle position, the occupant can select the appropriate road at the actual vehicle position. Information can be grasped, and comfort for passengers can be improved.

  The conventional navigation system has low estimation accuracy of the own vehicle position, and it is difficult to determine the lane in which the own vehicle is traveling on a road with multiple lanes on one side. Therefore, when performing guidance on a branch on an expressway or guidance in the direction of travel at an intersection, it is difficult to perform different route guidance for each lane and it is difficult to improve comfort for passengers. That is, in order to realize advanced route guidance, it is necessary to accurately determine the lane in which the vehicle is traveling.

  Here, the lane change is determined by the winker operation signal and the signal from the lane line detection unit (the lane line crossing), the lane position where the host vehicle is traveling is determined, and the front branch is detected and determined. There is an in-vehicle navigation device that provides branch guidance to a driver at a position a predetermined distance before a lane (see Patent Document 1). In addition, there is a vehicle control device that determines a lane that is traveling from the type of lane marking (solid line or broken line) (see Patent Document 2).

JP 2006-23278 A JP 2000-105898 A

  In Patent Document 1, since both the blinker operation and the lane line crossing are detected and the lane change is determined, there is a possibility that the lane position where the host vehicle is traveling may be lost due to forgetting to turn the blinker or the lane line crossing not being detected. In fact, in the case of a road with multiple lanes on one side, the lane marking is often a broken line, and the lane marking is often lost during lane change. For this reason, undetected crossing between lane markings occurs.

  Moreover, in patent document 2, in order to recognize the division line type (a solid line, a broken line, a dotted line, etc.), it is necessary to detect some paints, and there exists a problem that it takes time to determine a lane. . Also, if the paint is faint, it may cause undetected or erroneous detection. Furthermore, since the lane marking standards differ depending on the country or region, this method requires correspondence for each destination and is not always practical.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a lane determination device that can quickly and accurately determine the traveling lane of a host vehicle traveling on a road having a plurality of lanes on one side.

  In order to solve the above problems, one of the desirable embodiments of the present invention is as follows.

A lane determination device that determines a traveling lane of a host vehicle traveling on a one-sided multiple lane road includes an imaging unit that images a road, a lane line recognition unit that recognizes a lane line from an image captured by the imaging unit, A first trajectory calculating means for calculating a trajectory of the own vehicle by repeatedly calculating a distance from the lane line recognized by the lane marking recognizing means, a vehicle speed detecting means for detecting the speed of the own vehicle, and the own vehicle The vehicle's trajectory is calculated based on the autonomous navigation that calculates the vehicle's trajectory based on the azimuth detecting device that detects the azimuth of the vehicle, the vehicle speed detected by the vehicle speed detecting device, and the azimuth detected by the azimuth detecting device. A second trajectory calculating means for calculating, a first trajectory calculated by the first trajectory calculating means, and a second trajectory calculated by the second trajectory calculating means; Three trajectory calculating means, and Self based and third determining driving lane determining means lane running of the vehicle based on the trajectory calculated by the trajectory calculating means, the first trajectory calculated by the first path calculation means Initializing at least one of the direction and position of autonomous navigation based on the traveling direction estimation means for estimating the traveling direction of the vehicle and the traveling direction of the own vehicle estimated by the traveling direction estimation means and the first trajectory And a third trajectory calculating means that calculates a third trajectory based on the first trajectory when the lane marking recognition means can recognize the lane marking, and the lane marking recognition means If the lane line could not be recognized, a third trajectory is calculated based on the second trajectory, and the initialization means is recognized by the lane line recognition means and the traveling direction of the host vehicle estimated by the traveling direction estimation means. The difference from the direction of the lane marking For, or if the change in direction of the difference is smaller than a predetermined value for a predetermined period of time or running distance, and the initial value of the trajectory of the autonomous navigation of the first trajectory calculated by the first trajectory calculation means.

  ADVANTAGE OF THE INVENTION According to this invention, the lane determination apparatus which can determine the driving lane of the own vehicle which is drive | working the road of one side multiple lanes quickly and correctly can be provided.

  Hereinafter, embodiments will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram illustrating functions of the lane determination device.

  First, the configuration and processing contents of the lane determination device 100 will be described.

  The lane determination device 100 includes an imaging means 1, a lane marking recognition means 2, a first trajectory calculation means 3, a vehicle speed detection means 4, an orientation detection means 5, a second trajectory calculation means 6, a traveling direction estimation means 7, and an initialization. The means 8, the third trajectory calculating means 9, the traveling lane determining means 10, and the information notifying means 11 are programmed in a computer (not shown) of the lane determining device 100 and repeatedly executed at a predetermined cycle.

  The image pickup means 1 acquires an image outside the host vehicle by an image pickup device such as a CCD (Charge Coupled Device) image sensor. For example, a camera can be used, and the image pickup direction is the front of the vehicle (front view camera), Any one of a vehicle side (side view camera), a vehicle rear (rear view camera), or an oblique direction may be used, and an omnidirectional camera that images all directions may be used. Moreover, regarding the kind of camera, the monocular camera imaged with one camera and the stereo camera imaged with two cameras may be sufficient, and also regarding the number of mounting, you may arrange | position in the front-back, left-right direction of a vehicle, respectively.

  The lane marking recognition means 2 processes the image information acquired by the imaging means 1 and recognizes the lane marking in the image. Here, the lane marking is a line that is set by a road manager and is made of road fences, paint, stones, etc. drawn on the road surface of the road, and is a kind of road marking. Specifically, there are a roadway center line and a lane boundary line. It also has a function (hereinafter referred to as line type discrimination) for repeatedly calculating the distance from the recognized lane line to the imaging means and discriminating the type of lane line (solid line, broken line, dotted line, etc.).

  The first trajectory calculation means 3 calculates the travel trajectory of the host vehicle based on the distance from the lane line calculated by the lane line recognition means 2 to the imaging means. The performance of the first trajectory calculation means depends on the movement of the vehicle in addition to the presence of the lane line and the state of the lane line such as fading. When traveling straight, you can accurately draw the trajectory of the vehicle in the lane, but if the vehicle turns due to lane changes, etc., the angle between the vehicle and the lane line and the change in the lane line can be recognized. Since the trajectory is interrupted, it is impossible to recognize which lane the host vehicle is traveling immediately after the lane change.

  The vehicle speed detection means 4 detects the vehicle speed of the host vehicle. For example, the vehicle speed detection means 4 detects the vehicle speed by averaging the values obtained by the wheel speed sensors mounted on the front, rear, left, and right wheels of the vehicle. For example, a method of calculating the vehicle speed by integrating the acceleration value of the own vehicle obtained by the acceleration sensor is applicable.

  The direction detection means 5 detects the direction of the own vehicle, and a method of calculating the direction of the own vehicle from a value obtained by a gyro sensor or a yaw rate sensor can be employed.

  The second locus calculating means 6 calculates the traveling locus of the own vehicle by autonomous navigation using the vehicle speed detected by the vehicle speed detecting means 4 and the direction of the own vehicle detected by the direction detecting means 5. The second trajectory calculation means draws a trajectory based on the movement of the vehicle using autonomous navigation, so the trajectory will not be interrupted, but it can be used without correction because errors in the vehicle speed and heading accumulate over time. It is a short time. Also, since the azimuth and position are relative, initialization is important.

  The traveling direction estimating means 7 estimates the traveling direction of the own vehicle based on the traveling locus of the own vehicle calculated by the first locus calculating means 3.

  The initializing means 8 initializes the autonomous navigation based on the traveling direction of the own vehicle estimated by the traveling direction estimating means 7, the vehicle speed detected by the vehicle speed detecting means 4, and the direction of the own vehicle detected by the direction detecting means 5. If the accuracy of the traveling direction of the host vehicle estimated by the traveling direction estimation means 7 is high, initialization is performed.

  The third trajectory calculation means 9 is based on the first trajectory calculated by the first trajectory calculation means 3 and the second trajectory calculated by the second trajectory calculation means 6 and finally travels by the host vehicle. Calculate the trajectory. Specifically, based on the characteristics of the first and second trajectory calculation means, the first trajectory is valid (when the lane marking is recognized by the lane marking recognition means 2). If the trajectory is the final travel trajectory of the vehicle, and the first trajectory is invalid (if the lane line recognition means 2 does not recognize the lane line), the final vehicle is extrapolated with the second trajectory. The traveling locus of Thereby, a continuous traveling locus is obtained regardless of the traveling state of the host vehicle.

  The traveling lane determining means 10 determines the lane in which the vehicle is traveling in the multi-lane road based on the final traveling locus of the own vehicle calculated by the third locus calculating means 9.

  The information notification unit 11 performs a process of notifying the passenger of information obtained from the traveling lane determination means 10 with a voice or a monitor screen in an easily understandable manner. In addition, it is possible to provide easy-to-understand and kind guidance by switching the content to be notified to the occupant based on the lane in which the host vehicle is traveling, which is obtained from the traveling lane determining means 10.

  Next, processing contents of the entire lane determination device will be described.

  FIG. 2 is a flowchart showing the processing contents of the lane determination device in the present embodiment.

  First, in the process 201, the distance from the lane marking recognized by the lane marking recognition means 2 to the imaging means is repeatedly calculated, and the type of the lane marking (solid line, broken line, dotted line, etc.) is determined.

  Next, in process 202, it is determined whether or not the autonomous navigation has been initialized. If the autonomous navigation has been initialized, the process proceeds to process 203, and if the autonomous navigation has not been initialized, the process proceeds to process 210. .

  In process 203, the travel locus (first locus) of the host vehicle is calculated based on the distance from the lane marking calculated in process 201 to the imaging means.

  Next, in process 204, the traveling locus (second locus) of the own vehicle is calculated by autonomous navigation.

  Next, in process 205, it is determined whether or not the host vehicle is traveling straight. If the host vehicle is traveling straight, the process proceeds to process 206. If the host vehicle is not traveling straight, the process proceeds to process 207. Here, the determination as to whether or not the host vehicle is traveling straight is made based on whether the angle difference between the traveling direction of the host vehicle estimated by the traveling direction estimation unit 7 and the direction of the lane line recognized by the lane line recognition unit 2 is a predetermined value. Within an angle (for example, within 2 degrees) or a change in angle difference is within a predetermined value (for example, 1 degree / second), and the vehicle speed detected by the vehicle speed detecting means 4 is at or above a predetermined speed (for example, at least 10 km / h) This is performed under the condition that the rate of change (angular velocity) of the direction of the vehicle detected by the direction detection means 5 is within a predetermined range (for example, within 0.2 deg / s).

  In process 206, since the vehicle is traveling straight and it is considered that the direction and position estimation accuracy is high, the autonomous navigation is initialized. This initialization method substitutes the value of the first trajectory for the lateral position on the road (y in FIG. 3 to be described later), and substitutes the traveling direction of the host vehicle estimated by the traveling direction estimating means 7 for the heading. To do. Specifically, the traveling locus of the host vehicle is traced back for a predetermined time, and the inclination of the regression line based on the least square method is defined as the traveling direction.

  Next, in process 207, it is determined whether or not the first trajectory is valid. If the first trajectory is valid, the process proceeds to process 208 to calculate the first trajectory as a third trajectory. On the other hand, if the first trajectory is not valid, the process proceeds to process 209, and the second trajectory is calculated as the third trajectory. Here, the first trajectory is valid indicates that the lane marking is recognized by the lane marking recognition means 2, and the lane marking recognition means 2 recognizes the lane marking when the first trajectory is not valid. Indicates the case where it is not.

  Next, in processing 210, lane determination processing based on line type determination is performed. Here, the lane in which the host vehicle is traveling is determined using the type of lane marking recognized by the lane marking recognition means 2 (solid line, broken line, dotted line, etc.). For example, in the case of a Japanese highway with three lanes on one side, the lane markings are arranged in the order of solid line, broken line, broken line, and solid line from the left. The lane in which the vehicle is traveling can be determined.

  Next, in process 211, the lane in which the host vehicle is traveling is determined using the third trajectory calculated in process 208 or 209. Specifically, it is only necessary to determine which position on the multi-lane road the horizontal position of the third trajectory is, and the determination is made using the information on the lane width. If the lane in which the vehicle is traveling is determined by the line type determination in the process 210, the lane determined by the line type determination is not adopted without adopting the lane determined using the third trajectory. Adopt with priority. At this time, if the lane determined by the line type discrimination and the lane determined by the third trajectory are different, the lateral position of the third trajectory is corrected to the lane position determined by the line type discrimination. As described above, it is necessary to travel a predetermined distance to determine the line type and the response is slow, but the reliability is high. On the other hand, if a long time has passed since the first locus becomes invalid due to repeated meandering of the vehicle, an error accumulates in the second locus based on autonomous navigation. After that, even if the lane line is recognized, the method based on the distance from the lane line to the imaging unit cannot identify the lane line, so the travel locus cannot be drawn. For this reason, the above-described correction by line type discrimination is used as a backup for special operating conditions.

  Finally, in process 212, based on the information on the lane in which the host vehicle is traveling obtained in process 211, the road guidance is switched and information is notified to the occupant using voice or a screen.

  As described above, according to the lane determination device 100 described above, a lane line installed on a multi-lane road on one side is detected, and a relative relationship between the lane line and the own vehicle and an autonomous navigation calculated by the autonomous navigation are used. Since the lane in which the vehicle is traveling is determined based on the travel locus, the travel distance until the travel lane is determined can be shortened, and the accumulated error can be reduced according to the travel distance. it can. Therefore, the traveling lane of the own vehicle can be determined quickly and accurately.

  Therefore, for example, when a navigation system guides a branch on an expressway or guides the direction of travel at an intersection located in front of the vehicle, different route guidance can be performed for each lane, and advanced route guidance for passengers It can be realized and passenger comfort can be improved.

  Next, a specific example of the lane determination process in the lane determination device 100 will be described with reference to FIG.

  FIG. 3 is a diagram illustrating a case where the own vehicle 300 changes the lane by drawing a locus of a dotted arrow 301 from the left lane to the right lane on a two-lane road on one side.

  In FIG. 3, D1 is the distance from the vehicle center to the lane line visible to the left, D2 is the distance from the vehicle center to the lane line visible to the right, and the vehicle 300 exists in the left lane and in the right lane. In some cases, the target lane markings are different. In addition, the distance from the leftmost lane line to the vehicle 300 is represented by y. Further, the lane width is assumed to be W.

  First, when the own vehicle 300 exists at the point A, it is determined that the own vehicle 300 is traveling straight (YES in processing 205), and an initialization process of autonomous navigation is performed (processing 206). At this time, if both D1 and D2 are valid, for example, the first trajectory 312 is calculated with priority given to D1, the lateral position of autonomous navigation is set to y1, and the orientation is obtained from the first trajectory 312. When set to, the locus calculated by the autonomous navigation becomes the second locus 314. Further, since the first track is valid (YES in process 207), the first track is substituted for the third track (process 208), and the lane in which the host vehicle is traveling is determined (process 211). . In this case, since y1 <W, the lane in which the vehicle is traveling is 1 (left lane) as indicated by the lane position 318. In addition, since the vehicle 300 is determined to be traveling straight ahead until the vehicle 300 reaches the point B from the point A (the autonomous navigation initialization flag 311 is ON), the subsequent processing exists at the point A. The same.

  Next, when the own vehicle 300 reaches the point B, it is determined that the own vehicle 300 is not moving straight because the lane change operation has started (NO in the process 205), and the autonomous navigation initialization flag is turned OFF. D1 and D2 change until the own vehicle 300 reaches point C. However, since the first locus is valid (YES in process 207), the first locus is substituted into the third locus. (Processing 208), the lane in which the host vehicle is traveling is determined (Processing 211). Therefore, from the points A to C, the first trajectory and the third trajectory are the same.

  Next, when the host vehicle 300 reaches the point C and D1 and D2 are lost (NO in process 207), the third trajectory is extrapolated by the second trajectory (process 209). Further, when the own vehicle 300 reaches the point D, the lateral position of the third trajectory exceeds the lane width W, so the lane position 318 in which the own vehicle is traveling changes to 2 (processing 211).

  Next, when the host vehicle 300 reaches the point E and D1 and D2 become valid again (YES in process 207), the third trajectory returns to the first trajectory (process 208), so the second trajectory. If there is an error, the third trajectory jumps as shown in region 317.

  Next, when the vehicle 300 reaches the point F, the autonomous navigation initialization flag is turned ON again, and the autonomous navigation is initialized (processing 206). At this time, if there is an error in the second trajectory, the second trajectory jumps as in the region 315.

  Thereafter, it is determined that the host vehicle 300 is traveling straight ahead until the host vehicle 300 reaches the point G (the autonomous navigation initialization flag 311 is turned ON).

  As described above, by combining the distance from the own vehicle to the lane marking and autonomous navigation, it becomes possible to draw an accurate traveling locus without any interruption, and the traveling lane of the own vehicle can be determined quickly and accurately. it can.

  In addition, although the example in the one-side two-lane road was shown in FIG. 3, it cannot be overemphasized that the number of lanes is applicable even if it is three or more lanes.

  As described above, the present invention can be implemented in various modes without departing from the spirit of the present invention.

  According to the above, a lane line installed on a multi-lane road on one side is detected, and the vehicle travels based on the relative relationship between the lane line and the vehicle and the travel locus of the vehicle calculated by autonomous navigation. Since the lane is determined, the travel distance required until the travel lane is determined can be made shorter than the method of Patent Document 2, and the accumulated error can be reduced according to the travel distance. Therefore, the traveling lane of the own vehicle can be determined quickly and accurately. Therefore, advanced route guidance by the navigation system is possible.

The block diagram of a lane determination apparatus. The flowchart which shows the processing content of a lane determination apparatus. The figure explaining a specific example.

Explanation of symbols

100 lane determination device 300 own vehicle 301 travel locus

Claims (4)

In the lane determination device for determining the traveling lane of the own vehicle traveling on one side multiple lane road,
Imaging means for imaging a road;
Lane marking recognition means for recognizing lane markings from the image captured by the imaging means;
First trajectory calculation means for repeatedly calculating the distance from the lane line recognized by the lane marking recognition means to the imaging means and calculating the trajectory of the own vehicle;
Vehicle speed detection means for detecting the speed of the vehicle;
Direction detection means for detecting the direction of the vehicle;
Second trajectory calculating means for calculating the trajectory of the own vehicle based on the autonomous navigation that calculates the trajectory of the own vehicle based on the vehicle speed detected by the vehicle speed detecting means and the orientation detected by the azimuth detecting means. When,
Third trajectory calculating means for calculating the trajectory of the vehicle based on the first trajectory calculated by the first trajectory calculating means and the second trajectory calculated by the second trajectory calculating means;
Traveling lane determining means for determining the lane in which the vehicle is traveling based on the third locus calculated by the third locus calculating means;
A traveling direction estimating means for estimating a traveling direction of the host vehicle based on the first locus calculated by the first locus calculating means;
Initialization means for initializing at least one of the direction and position of autonomous navigation based on the traveling direction of the host vehicle estimated by the traveling direction estimation means and the first trajectory ,
The third trajectory calculation means calculates the third trajectory based on the first trajectory when the lane marking recognition means can recognize the lane marking, and the lane marking recognition means recognizes the lane marking. If not, calculate the third trajectory based on the second trajectory ,
The initialization means includes a traveling direction of the host vehicle estimated by the traveling direction estimation means and the section.
The difference between the direction of the lane marking recognized by the line recognition means is within a predetermined angle, or
If the change in direction difference is less than a predetermined value over a predetermined period or mileage, the first
A lane determination device using the first trajectory calculated by the trajectory calculating means as an initial value of the trajectory for autonomous navigation . The travel lane determining means determines the lane in which the host vehicle is traveling based on the type of the lane line recognized by the lane line recognition means, and the third trajectory calculated by the third trajectory calculation means. The lane determination device according to claim 1, wherein Comprising information notifying means for notifying the occupant of information using a screen or sound;
The lane determination device according to claim 2 , wherein the information notification unit switches the content of information notified to the occupant based on the lane in which the host vehicle is traveling determined by the travel lane determination unit. A navigation system having a lane determining device according to claims 1 to 3.

Images