JP2015018447A - Lane line estimation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a lane line estimation device and a method capable of easily estimating a lane line.SOLUTION: A lane line estimation device for estimating the lane line comprises: an image acquisition section installed in a vehicle for photographing the front of the vehicle to acquire image information; a line detection section for detecting a line from the image information; a storage section for storing a map in which lane width information on a road associated with speed is defined; a lane width determination section for referring to the map to determine lane width of the road on which the vehicle travels, when the line detection section detects one of right and left lines; and a first line estimation section for estimating a position of other line from the one line detected by the line detection section and the lane width determined by the lane width determination section.

Description

  The present invention relates to a lane boundary estimation device that estimates a lane boundary.

  As a method for estimating a lane boundary line, Patent Document 1 discloses a parameter W representing a road width and a parameter H representing a positional relationship between a road surface and a camera when one of the left and right white lines cannot be detected. And a method of predicting the position of the white line using the average of a certain period in the past.

JP-A-5-334595

  In the method described in Patent Document 1, it is necessary to accumulate the data of the parameters W and H for a certain period when predicting the white line position. Therefore, it cannot be said that the prediction of the white line position is easy.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a lane boundary estimation device that can easily estimate a lane boundary.

  The present invention is a lane boundary line estimation device for estimating a lane boundary line, which is mounted on a vehicle and captures image information by photographing the front of the vehicle, and detects the boundary line from the image information A boundary line detection unit, a storage unit that stores a map that defines road lane width information related to speed, and the boundary line detection unit detects one of the left and right boundary lines, A lane width determining unit that determines a lane width of a road that is currently running with reference to the map, and one of the boundary line detected by the boundary line detecting unit and the lane determined by the lane width determining unit And a first boundary line estimation unit that estimates the position of the other boundary line from the width.

  According to the present invention, when one of the left and right boundary lines is detected, the lane width of the currently traveling road is determined with reference to the map in which the lane width information of the road related to the speed is defined. Since the position of the other boundary line that cannot be detected is estimated from the lane width and the detected one boundary line, the boundary line of the lane can be easily estimated.

1 is a schematic configuration diagram of a lane boundary estimation device according to a first embodiment of the present invention. 1 is a block diagram of a lane boundary estimation device according to a first embodiment of the present invention. It is a flowchart which shows the procedure of the lane boundary estimation method which concerns on 1st Embodiment of this invention. It is a map figure in which the relationship between vehicle speed and lane width was defined. It is a flowchart which shows the procedure of the lane boundary estimation method which concerns on the modification of 1st Embodiment of this invention. It is a map figure in which the relationship between speed limit and lane width was defined. It is a flowchart which shows the procedure of the lane boundary estimation method which concerns on the other modification of 1st Embodiment of this invention. It is a flowchart which shows the procedure of the lane boundary estimation method which concerns on 2nd Embodiment of this invention. It is a flowchart which shows the procedure of the lane boundary estimation method which concerns on the modification of 2nd Embodiment of this invention.

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

<First Embodiment>
Below, with reference to FIGS. 1-4, the lane boundary estimation apparatus 100 which concerns on 1st Embodiment of this invention is demonstrated.

  The lane boundary estimation apparatus 100 is an apparatus that estimates the position of the other white line that cannot be detected when one of the left and right white lines, which is the boundary of the lane on the road, is detected. In this embodiment, a case will be described in which the lane boundary estimation device 100 is applied to a lane departure warning system that predicts that a vehicle will depart from a lane and issues a warning to a driver.

  As shown in FIGS. 1 and 2, a lane boundary estimation device 100 is mounted on a vehicle and captures the front of the vehicle and acquires image data (image information), and a camera 1 mounted on the vehicle. And a controller 2 for processing image data taken by the camera 1.

  The camera 1 is a video camera composed of a CCD image sensor or the like, and is connected to the controller 2. The camera 1 captures the situation in front of the vehicle including the road surface.

  The controller 2 includes a CPU 11, a ROM 12 that stores control programs, setting values, maps, and the like necessary for processing operations of the CPU 11, and a RAM 13 that stores image data captured by the camera 1.

  The RAM 13 also stores information output from a vehicle speed sensor 15 as a speed detector, a steering angle sensor 16 as a steering angle detector, and a GPS receiver 17 as a position information acquisition unit. The vehicle speed sensor 15 is a sensor that detects the vehicle speed. The steering angle sensor 16 is a sensor that detects the steering angle (absolute steering angle) of the steering wheel of the vehicle. The GPS receiver 17 receives signals from a plurality of GPS satellites and generates current position information (latitude and longitude information) of the vehicle.

  Next, a white line estimation method will be described with reference to FIG. FIG. 3 is a flowchart showing the procedure of the white line estimation method executed by the controller 2.

  The controller 2 executes the following processing when an ON signal is received from the ignition switch, that is, when the engine is started by operating the ignition switch.

  In step 10, white lines are detected from image data acquired by the camera 1 (boundary line detection unit). Specifically, the white line is detected by performing image processing on the image data. Then, it progresses to step 11.

  In step 11, it is determined whether left and right white lines are detected simultaneously. When it is determined in step 11 that the left and right white lines are detected at the same time, the process proceeds to step 12 to recognize the lane. When the vehicle is about to deviate from the recognized lane, an alarm is issued to the driver from the lane departure warning system. After step 12, the process returns to step 10.

  If it is determined in step 11 that the left and right white lines are not detected simultaneously, the process proceeds to step 20.

  In step 20, it is determined whether or not a white line on one side is detected. If it is determined in step 20 that the white line on one side has not been detected, both the left and right white lines have not been detected, so the process returns to step 10 to detect the white line again.

  If it is determined in step 20 that one white line is detected, only one white line is detected among the left and right white lines. In such a situation, the position of the opposite white line not detected by the following procedure is estimated.

  In step 21, the current vehicle speed is acquired from the vehicle speed sensor 15.

  In step 22, the width of the currently traveling lane is determined based on the vehicle speed acquired in step 21 and the map (see FIG. 4) stored in the ROM 12 (lane width determining unit).

  The map shown in FIG. 4 will be described. The map defines the relationship between vehicle speed and lane width. Specifically, when the vehicle speed is 61 km / h or more, the lane width is 3.50 m, when the vehicle speed is 51 km / h or more and 60 km / h or less, the lane width is 3.25 m, and the vehicle speed is 41 km / h or more and 50 km / h or less. Sometimes the lane width is 3.00 m, and when the vehicle speed is 40 km / h or less, the lane width is 2.75 m. In step 22, the lane width corresponding to the vehicle speed acquired by the vehicle speed sensor 15 is determined with reference to the map shown in FIG. The lane width specified in the map is 2.75 to 3.5 m which is the lane width defined in Article 3 of the Road Structure Ordinance. The road design speed is set to be larger as the lane width is larger. Therefore, the map is defined so that the lane width increases as the vehicle speed increases. Thus, the map for determining the lane width uses the lane width stipulated by laws and regulations, and is variably defined according to the vehicle speed, so it predicts the lane width during travel more accurately. Can be determined.

  In step 23, the position of the white line on the opposite side is estimated from the white line on one side detected in step 20 and the lane width determined in step 22 (first boundary line estimation unit). Specifically, it is estimated that the white line on the opposite side is present at the position of the lane width determined in step 22 from the white line on one side detected in step 20. For example, when the vehicle is traveling at a vehicle speed of 45 km / h, if the detected white line is only the right side, it is estimated that the white line on the left side is at a position 3.00 m away from the right side white line.

  As shown in steps 21 to 23 above, when only one white line can be detected, the lane width of the currently running road is determined with reference to the map, and the determined lane width and the detected white line on one side are determined. From the above, the position of the white line on the opposite side is estimated.

  In the following step 12, the lane is recognized from the white line on one side detected in step 20 and the white line on the opposite side estimated in step 23. When the vehicle is about to deviate from the recognized lane, an alarm is issued to the driver from the lane departure warning system. After step 12, the process returns to step 10.

  Next, a modification of the first embodiment will be described.

  You may make it perform the estimation method (steps 21-23) of the white line of the other side in the said 1st Embodiment in the procedure shown in FIG. Below, step 21a-23 shown in FIG. 5 is demonstrated. Steps other than steps 21a to 23 are the same as the steps shown in FIG.

  In step 21a, the speed limit display provided on the road is identified from the image data acquired by the camera 1, and the speed limit is acquired from the speed limit display (limit speed acquisition unit). Specifically, the speed limit is acquired by performing image processing on the image data. In this way, the speed limit of the currently traveling lane is acquired. The speed limit display includes road signs placed on the side of the road or above the road, and road markings directly written on the road surface.

  In step 22, the width of the currently traveling lane is determined based on the speed limit acquired in step 21 and the map (see FIG. 6) stored in the ROM 12 (lane width determining unit).

  In the map shown in FIG. 6, the relationship between the speed limit and the lane width is defined. Specifically, when the speed limit is 61 km / h or more, the lane width is 3.50 m, when the speed limit is 51 km / h or more and 60 km / h or less, the lane width is 3.25 m, and the speed limit is 41 km / h or more and 50 km / h. The lane width is defined as 3.00 m when h or less, and the lane width is defined as 2.75 m when the speed limit is 40 km / h or less. In step 22, the lane width corresponding to the speed limit acquired in step 21 is determined with reference to the map shown in FIG. The lane width specified in the map is 2.75 to 3.5 m which is the lane width defined in Article 3 of the Road Structure Ordinance. The road design speed is set to be larger as the lane width is larger. Therefore, the map is defined such that the lane width increases as the speed limit increases. Thus, the map for determining the lane width uses the lane width stipulated by laws and regulations, and is variably defined according to the speed limit, so the lane width during travel can be predicted more accurately. Can be determined.

  In step 23, the position of the white line on the opposite side is estimated from the white line on one side detected in step 20 and the lane width determined in step 22 (first boundary line estimation unit).

  Next, another modification of the first embodiment will be described.

  You may make it perform the estimation method (step 21-23) of the white line of the other side in the said 1st Embodiment in the procedure shown in FIG. Below, step 21b-23 shown in FIG. 7 is demonstrated.

  In step 21b, the current position information of the vehicle is acquired from the GPS receiver 17.

  In step 21c, the speed limit of the currently traveling lane is acquired from the current position information of the vehicle acquired in step 21 using a database or the like stored in the car navigation system (limit speed acquisition unit). Specifically, referring to a map in which a road and a speed limit are associated with each other, the speed limit of the currently traveling lane is acquired from the current position information of the vehicle. In this way, the speed limit of the currently traveling lane is acquired.

  In step 22, the width of the currently traveling lane is determined based on the speed limit acquired in step 21c and the map (see FIG. 6) stored in the ROM 12 (lane width determining unit).

  In step 23, the position of the white line on the opposite side is estimated from the white line on one side detected in step 20 and the lane width determined in step 22 (first boundary line estimation unit).

  According to the above 1st Embodiment, there exists an effect shown below.

  When only one of the left and right white lines can be detected, the lane boundary estimation apparatus 100 determines the lane width of the road that is currently running with reference to the map associated with the speed, and the determined lane width and detection The position of the other white line that cannot be detected is estimated from the one white line. In this way, the lane width of the currently running road is determined simply by acquiring information related to the speed such as the vehicle speed and the speed limit, and the other lane width that cannot be detected from the determined lane width and one of the detected white lines. Since the position of the white line can be estimated, the white line of the lane can be easily estimated. The map for determining the lane width uses the lane width specified by laws and regulations, and is variably defined according to the vehicle speed and the speed limit of the road. Can be predicted and determined.

Second Embodiment
Hereinafter, a second embodiment of the present invention will be described with reference to FIG.

  In step 10, white lines are detected from image data acquired by the camera 1 (boundary line detection unit). The controller 2 stores white line information related to the detected white line in the RAM 13. This white line information includes the case where white lines on both the left and right sides are detected simultaneously (both sides detection), the case where only the right side white line is detected (one side detection), and the case where only the left side white line is detected ( Single-sided detection).

  In step 11, it is determined whether the left and right white lines are detected simultaneously. Specifically, it is determined whether or not the white line information stored in the RAM 13 is both-side detection. If it is determined in step 11 that the white line information is detected on both sides, the process proceeds to step 12 where lane recognition is performed. If it is determined in step 11 that the white line detection is not the double-sided detection, the process proceeds to step 20. Incidentally, the case where the white line detection is not the two-sided detection means, for example, a case where the white line detection is one-sided detection or a case where the white line information itself is not stored in the RAM 13 (when the white lines on both sides cannot be detected). Say.

  In step 20, it is determined whether only one white line is detected. Specifically, it is determined whether the white line information stored in the RAM 13 is one-side detection. If it is determined in step 20 that the white line information is not one-side detection, the process returns to step 10 to detect the white line again. If it is determined in step 20 that the white line information is single-sided detection, the process proceeds to step 30.

  In step 30, it is determined whether or not only the white line on the right side or the left side is detected from the state in which the left and right white lines are detected at the same time. Specifically, in the white line information stored in the RAM 13, it is determined whether the previously detected stored one is the two-sided detection, and the one detected and stored this time is the one-sided detection. If it is determined in step 30 that the previously detected stored one is the two-sided detection, and the one detected and stored this time is the one-sided detection, the process proceeds to step 31.

  In step 31, the “left / right → one side detection flag” is turned on and the process proceeds to step 32. The “left / right → one-side detection flag” is a flag indicating that the detection is switched from the both-side detection to the one-side detection, and is stored in the RAM 13 as ON or OFF.

  In step 32, the position of the white line on the opposite side is estimated from the lane width when the left and right white lines are detected simultaneously (the lane width immediately before only one white line can be detected) (second boundary line estimation unit). Specifically, it is estimated that there is a white line on the opposite side at the position of the lane width when the left and right white lines are simultaneously detected from the white line on one side detected in step 20. For example, when driving on a lane in which left and right white lines exist, the left and right white lines are detected at the same time and the lane width is 3.25 m. After that, when the right white line cannot be detected and only the left white line can be detected, it is estimated that the right white line exists at a position separated by 3.25 m to the right from the left white line. When the white line on the opposite side is estimated in step 32, the process proceeds to step 12.

  In step 12, lane recognition is executed. When the left and right white lines are detected at the same time in step 11, the lane recognition is executed from the detected left and right white lines in step 12. That is, the lane is recognized using the distance between the detected white line on the left side and the detected white line on the right side as the lane width.

  If one white line is detected and the other white line is estimated in step 32, the lane is recognized in step 12 from the detected one white line and the estimated white line. That is, the lane is recognized with the distance between the detected one white line and the estimated other white line as the lane width.

  If the vehicle is about to deviate from the lane recognized in step 12, a warning is issued to the driver from the lane departure warning system. After step 12, the process returns to step 10 and white line detection is performed.

  Returning to step 30, if the condition is not satisfied, specifically, if the previously detected stored one-side detection is one-side detection, and the one detected and stored this time is one-side detection, step Proceed to 33.

  In step 33, it is determined whether or not the “left / right → one-side detection flag” is ON. When the “left / right → one-side detection flag” is on, the process proceeds to step 34, and when the “left / right → one-side detection flag” is not on (when off), the process proceeds to step 21.

  The case where the “left / right → one-side detection flag” is ON means that the previously detected stored one is the two-sided detection, and the one detected and stored this time is the one-sided detection (step 30). This is a case where the “left / right → one side detection flag” is turned on.

  In step 21, step 22, and step 23, the position of the white line on the opposite side is estimated by the same method as in steps 21 to 23 shown in FIG. 3 of the first embodiment. Therefore, the description is omitted.

  In step 34a, it is determined whether or not a predetermined time has elapsed since the “left / right → one side detection flag” turned on in step 31 is turned on. The predetermined time is, for example, about 10 seconds. If the predetermined time has elapsed, the process proceeds to step 35. If the predetermined time has not elapsed, the process proceeds to step 32 described above.

  In step 35, the “left / right → one side detection flag” turned on in step 31 is turned off. Thereafter, the process proceeds to step 21, and the position of the white line on the opposite side is estimated by the same method as in steps 21 to 23 shown in FIG. 3 of the first embodiment.

  In this way, from the state where the left and right white lines are detected at the same time until the predetermined time elapses after the detection of only the white line on one side, the lane width when the left and right white lines are detected at the same time is used to The position of the white line on the side is estimated. This is because there is a high possibility that the vehicle travels in the same lane as when the left and right white lines were detected at the same time until a predetermined time has elapsed since the detection of only one white line. It is more accurate to estimate the position of the white line on the opposite side using the lane width when the left and right white lines are detected simultaneously than to estimate the position of the white line on the opposite side in steps 21 to 23 in the form. This is because the position of the white line on the opposite side can be estimated. On the other hand, after a predetermined time has elapsed since the detection of only the white line on one side, there is a high possibility that the vehicle is traveling in a lane different from the lane when the left and right white lines were detected at the same time. It is more accurate to estimate the position of the white line on the opposite side in steps 21 to 23 in the first embodiment than to estimate the position of the white line on the opposite side using the lane width when it is detected at the same time. The position of the white line on the opposite side can be estimated. As described above, the switching from the white line estimation in step 32 to the white line estimation in steps 21 to 23 is performed after a predetermined time from the state in which the left and right white lines are simultaneously detected to the detection of only one white line. .

  If only one white line is detected from the state in which both the left and right white lines are not detected, the process proceeds from step 20, step 30, and step 33 to step 21, and the same method as in the first embodiment is used. The position of the white line on the opposite side is estimated by the method of steps 21 to 23.

  Next, a modification of the second embodiment will be described.

  In the second embodiment, the lane width when the left and right white lines are detected at the same time from when the left and right white lines are detected at the same time until the predetermined time elapses after the detection of only the white line on one side is used. Thus, the position of the white line on the opposite side was estimated. Instead, from the state where the left and right white lines are detected at the same time until only the white line on one side is detected, until the absolute steering angle of the steering wheel detected by the steering angle sensor 16 exceeds a predetermined angle. The position of the white line on the opposite side may be estimated using the lane width when the left and right white lines are detected simultaneously.

  In this case, as shown in FIG. 9, the absolute steering angle of the steering wheel detected by the steering angle sensor 16 after the “left / right → one-side detection flag” in step 31 is turned on in step 34b. It may be determined whether or not the angle is equal to or greater than a predetermined angle.

  If the absolute steering angle of the steering wheel is less than a predetermined angle after detecting only the white line on one side, there is a high possibility that the vehicle is traveling in the same lane as when the left and right white lines were detected simultaneously. . Therefore, rather than estimating the position of the opposite white line in steps 21 to 23 in the first embodiment, the position of the opposite white line is estimated using the lane width when the left and right white lines are simultaneously detected. It is possible to estimate the position of the white line on the opposite side more accurately. On the other hand, when the absolute steering angle of the steering wheel exceeds a predetermined angle after detecting only the white line on one side, the vehicle is traveling in a lane different from the lane when the left and right white lines are detected simultaneously. Probability is high. Therefore, rather than estimating the position of the opposite white line using the lane width when the left and right white lines are detected simultaneously, the position of the opposite white line is estimated by steps 21 to 23 in the first embodiment. It is possible to estimate the position of the white line on the opposite side more accurately.

  As described above, the switching from the white line estimation in step 32 to the white line estimation in steps 21 to 23 is performed when the left and right white lines are detected at the same time and only the white line on one side is detected, and then the absolute steering of the steering wheel is performed. This is performed after the angle is equal to or greater than a predetermined angle.

  According to the above 2nd Embodiment, there exists an effect shown below.

  When only one of the left and right white lines can be detected, as a method of estimating the position of the opposite white line, a method using a map in which the lane width is defined and a lane width when the left and right white lines are detected at the same time are used. There are methods to use. These two estimation methods are switched according to the passage of time or the absolute steering angle of the steering wheel. Therefore, since the position of the white line on the opposite side can be estimated by an optimum method according to the situation, the lane width during traveling can be estimated more accurately.

  The embodiment of the present invention has been described above. However, the above embodiment only shows a part of application examples of the present invention, and the technical scope of the present invention is limited to the specific configuration of the above embodiment. Absent.

100 Lane boundary estimation device 1 Camera 2 Controller 15 Vehicle speed sensor (speed detector)
16 Steering angle sensor (steering angle detector)
17 GPS receiver (location information acquisition device)

Claims (6)

A lane boundary estimation device for estimating a lane boundary,
An image acquisition unit mounted on a vehicle and capturing image information by photographing the front of the vehicle;
A boundary detection unit for detecting the boundary from the image information;
A storage unit for storing a map in which road lane width information related to speed is defined;
A lane width determining unit that determines a lane width of a road that is currently running with reference to the map when the boundary line detecting unit detects one of the left and right boundary lines;
A first boundary line estimation unit that estimates the position of the other boundary line from the one boundary line detected by the boundary line detection unit and the lane width determined by the lane width determination unit;
A lane boundary estimation device comprising: A speed detector for detecting the vehicle speed;
The map defines the relationship between vehicle speed and lane width,
2. The lane boundary estimation device according to claim 1, wherein the lane width determination unit determines a lane width of a currently traveling road from a current vehicle speed with reference to the map. A speed limit acquisition unit for identifying a speed limit display from the image information and acquiring a speed limit of a road that is currently running;
The map defines the relationship between speed limit and lane width,
2. The lane boundary estimation device according to claim 1, wherein the lane width determining unit determines a lane width of a currently traveling road from a speed limit of the currently traveling road with reference to the map. A position information acquisition unit for acquiring current position information of the vehicle;
A speed limit acquisition unit that refers to a map in which roads and speed limits are associated with each other, and acquires a speed limit of a road that is currently running from the current position information of the vehicle;
The map defines the relationship between speed limit and lane width,
2. The lane boundary estimation device according to claim 1, wherein the lane width determining unit determines a lane width of a currently traveling road from a speed limit of the currently traveling road with reference to the map. When the boundary line detection unit detects one of the left and right boundary lines, the one boundary line detected by the boundary line detection unit and the boundary line detection unit detect the left and right boundary lines. A second boundary line estimation unit that estimates the position of the other boundary line from the lane width when
When only one boundary line can be detected from the state in which the boundary line detection unit detects the left and right boundary lines, the second boundary line estimation unit estimates the position of the other boundary line. The position of the other boundary line is estimated by the first boundary line estimation unit after a predetermined time has elapsed since only one of the boundary lines can be detected. The lane boundary estimation apparatus according to one. A steering angle detector for detecting the steering angle of the steering wheel of the vehicle;
When the boundary line detection unit detects one of the left and right boundary lines, the one boundary line detected by the boundary line detection unit and the boundary line detection unit detect the left and right boundary lines. A second boundary line estimation unit that estimates the position of the other boundary line from the lane width when the
When only one boundary line can be detected from the state in which the boundary line detection unit detects the left and right boundary lines, the second boundary line estimation unit estimates the position of the other boundary line. The position of the other boundary line is estimated by the first boundary line estimation unit after the absolute steering angle of the steering wheel becomes equal to or greater than a predetermined angle. The lane boundary line estimation apparatus according to claim 1.

Images