JP6798331B2 - Driving support device - Google Patents

Description

The present disclosure relates to a driving support device.

Conventionally, there is known a technique of acquiring an image using an in-vehicle camera and recognizing a desired object in the image. Patent Document 1 discloses a problem that it becomes difficult to recognize an object because when sunlight enters the frame of an in-vehicle camera, it is backlit. In addition, the same document discloses searching for a route that minimizes the influence of backlight in order to solve the problem.

JP-A-2007-315799

A technique for recognizing a road lane marking using an image of an in-vehicle camera or the like is known. In this technology, not only sunlight but also various factors may cause misrecognition of road lane markings.

It is an object of the present disclosure to provide a driving support device capable of suppressing an error in recognizing a road lane marking.

One aspect of the present disclosure is a road lane marking unit that recognizes a road lane marking in front of a vehicle, a reliability estimation unit that estimates reliability in recognizing a road lane marking by the road lane marking unit, and the vehicle. The reliability in recognizing the road lane marking that divides the first lane in progress is compared with the reliability in recognizing the road lane marking that divides the second lane adjacent to the first lane. On condition that the reliability in recognizing the contrast unit and the road lane marking that divides the second lane is higher than the reliability in recognizing the road lane marking that divides the first lane. It is a driving support device including an instruction unit for instructing a lane change to the second lane.

The driving support device, which is one aspect of the present disclosure, compares the reliability in recognizing the road lane marking that divides the first lane with the reliability in recognizing the road lane marking that divides the second lane. If the reliability in recognizing the road lane marking that divides the second lane is higher than the reliability in recognizing the road lane marking that divides the first lane, the lane from the first lane to the second lane Instruct a change. When the lane is changed to the second lane, the reliability in recognizing the road marking line becomes higher than when traveling in the first lane. As a result, it is possible to prevent the road marking line from being misrecognized.

Another aspect of the present disclosure is a road lane marking unit that recognizes a road lane marking in front of a vehicle, a position acquisition unit that acquires the position of the vehicle, a map database that stores map information, and the position acquisition unit. The road environment acquisition unit that acquires the road environment around the vehicle and the road environment acquired by the road environment acquisition unit are specified in advance by using the acquired position of the vehicle and the map information. It is a driving support device including an instruction unit for instructing a lane change from a lane in which the vehicle is traveling to another lane, provided that the road environment is the same.

The driving support device, which is another aspect of the present disclosure, acquires the road environment around the vehicle, and if the road environment is a specific preset road environment, instructs a lane change to another lane. To do. When changing lanes to another lane, the reliability in recognizing the road lane marking is higher than when driving in the original lane which is a specific road environment. As a result, it is possible to prevent the road marking line from being misrecognized.

It is a block diagram which shows the structure of the driving support device 1. It is a block diagram which shows the functional structure of the control part 3. 3A to 3D are explanatory views showing an example of a specific road environment, respectively. It is explanatory drawing which shows the region 81, 83, 105. It is a flowchart which shows the process which the driving support apparatus 1 executes. 6A and 6B are explanatory views showing an example of a state in which the first reliability tends to be low, respectively. It is explanatory drawing which shows the case where the other lane is present on both sides of the 1st lane, and the case where there is no other lane on at least one side of both sides of the 1st lane.

The embodiments of the present disclosure will be described with reference to the drawings.
<First Embodiment>
1. 1. Configuration of the driving support device 1 The configuration of the driving support device 1 will be described with reference to FIGS. 1 to 4. The driving support device 1 is an in-vehicle device mounted on a vehicle. In the following, the vehicle equipped with the driving support device 1 will be referred to as the own vehicle. As shown in FIG. 1, the driving support device 1 includes a control unit 3 and a map database 5.

The control unit 3 is mainly composed of a well-known microcomputer having a CPU 7 and a semiconductor memory (hereinafter referred to as a memory 9) such as a RAM, a ROM, and a flash memory. Various functions of the control unit 3 are realized by the CPU 7 executing a program stored in a non-transitional substantive recording medium. In this example, the memory 9 corresponds to a non-transitional substantive recording medium in which a program is stored. Moreover, when this program is executed, the method corresponding to the program is executed. The number of microcomputers constituting the control unit 3 may be one or a plurality.

As shown in FIG. 2, the control unit 3 has a road lane marking unit 11, a reliability calculation unit 13, a comparison unit 15, and an instruction unit as a configuration of functions realized by the CPU 7 executing a program. The vehicle includes 17, a position acquisition unit 19, a road environment acquisition unit 21, a double-sided lane determination unit 23, and another vehicle determination unit 25. The road marking line recognition unit 11 is composed of an image acquisition unit 27, an extraction unit 29, and a recognition processing unit 31. The reliability calculation unit 13 is composed of a feature point likelihood calculation unit 33 and an estimation processing unit 35.

The method for realizing these elements constituting the control unit 3 is not limited to software, and some or all of the elements may be realized by using one or a plurality of hardware. For example, when the above function is realized by an electronic circuit which is hardware, the electronic circuit may be realized by a digital circuit including a large number of logic circuits, an analog circuit, or a combination thereof.

The map database 5 stores map information. The map information remembers the road environment in association with the location. The road environment includes the structure of the road, the shape of the road, and the like. As a road structure, for example, as shown in FIG. 3B, among lanes 47, 49, and 51, there is a structure in which the number of lanes decreases when the lane 47 disappears in the middle. Further, as a road structure, for example, as shown in FIG. 3C, there is a structure in which lanes 53 and 57 branch off on a road in which lanes 53 and 55 exist. Further, as a road structure, for example, as shown in FIG. 3D, there is a structure in which lane 63 joins lane 59 on a road in which lanes 59 and 61 exist. As the shape of the road, for example, as shown in FIG. 3A, there is a curved shape. In this example, the road comprises lanes 65, 67.

As shown in FIG. 1, the own vehicle includes a GPS 69, a camera 71, a vehicle detection sensor 73, a vehicle control unit 75, and a speaker 77 in addition to the driving support device 1.
GPS69 acquires the position of its own vehicle. As shown in FIG. 4, the camera 71 captures the area 81 in front of the own vehicle 79 to create an image. The area 81 includes a road in front of the own vehicle 79. The camera 71 outputs an image to the driving support device 1. The camera 71 may be a stereo camera or a monocular camera. Further, a lidar may be used instead of the camera 71.

As shown in FIG. 4, the vehicle detection sensor 73 can detect another vehicle in the area 83 around the own vehicle 79. The area 83 includes the front of the own vehicle 79, the rear of the own vehicle 79, and a part of the second lane 85. The second lane 85 is a lane adjacent to the first lane 87, which is the lane in which the own vehicle 79 is traveling. The vehicle detection sensor 73 outputs the detection result to the driving support device 1. Examples of the vehicle detection sensor 73 include a millimeter wave radar and a rider.

The vehicle control unit 75 executes a lane change of the own vehicle 79 in response to an instruction sent from the driving support device 1. Further, the vehicle control unit 75 performs processing such as steering and warning so that the own vehicle 79 does not deviate from the lane in which the vehicle is traveling, based on the road marking line recognized by the driving support device 1. Examples of the road marking line include a white line and bots dots. Further, the vehicle control unit 75 has a function of following the vehicle in front and traveling.

The speaker 77 is installed in the passenger compartment of the own vehicle 79, and outputs voice in response to a signal sent from the driving support device 1.
The driving support device 1 is connected to the vehicle-mounted network 89. The driving support device 1 can acquire vehicle information from the in-vehicle network 89. Examples of vehicle information include the vehicle speed and yaw rate of the own vehicle 79. The vehicle information is appropriately used in the processing described later.

2. 2. Processing executed by the driving support device 1 The processing executed by the driving support device 1 repeatedly at predetermined time intervals will be described with reference to FIGS. 3 to 7. In step 1 of FIG. 5, another vehicle determination unit 25 uses the vehicle detection sensor 73 to execute a process of recognizing another vehicle (hereinafter referred to as a peripheral vehicle) existing in the vicinity of the own vehicle 79. It should be noted that this peripheral vehicle does not include other vehicles (hereinafter referred to as front vehicles) that are traveling in the same lane as the own vehicle 79 and exist in front of the own vehicle 79 among other vehicles.

In step 2, the other vehicle determination unit 25 determines whether or not there are surrounding vehicles based on the result of the process of step 1. If there are no peripheral vehicles, the process proceeds to step 3, and if there are peripheral vehicles, this process ends.

In step 3, another vehicle determination unit 25 uses the vehicle detection sensor 73 to execute a process of recognizing the vehicle in front.
In step 4, the other vehicle determination unit 25 determines whether or not the vehicle in front exists based on the result of the process in step 3. If there is no vehicle in front, the process proceeds to step 5, and if there is a vehicle in front, this process ends.

In step 5, the road lane marking unit 11 recognizes a portion of the road lane marking that divides the first lane, which is ahead of the own vehicle 79. The method of recognition is as follows. First, the image acquisition unit 27 acquires an image using the camera 71. The image includes the road in front of the own vehicle 79. Next, the extraction unit 29 extracts feature points in the image. The feature point is a point where the brightness changes greatly in the image. Next, the recognition processing unit 31 recognizes the road marking line that divides the first lane by a known method based on the extracted feature points.

In step 6, the road lane marking unit 11 recognizes a portion of the road lane marking that divides the second lane, which is ahead of the own vehicle 79. The method of recognition is the same as the method of recognizing the road lane marking that divides the first lane.

In step 7, the reliability calculation unit 13 determines the reliability in recognizing the road lane marking that divides the first lane (hereinafter referred to as the first reliability) and the road lane marking that divides the second lane. The reliability in recognition (hereinafter referred to as the second reliability) is estimated.

The method of estimating the first reliability is as follows. First, the feature point likelihood calculation unit 33 calculates the likelihood that the feature point is a feature point caused by the road marking line for each feature point. This feature point is extracted when recognizing the road lane marking line on the side opposite to the second lane among the road lane marking lines that divide the first lane. Likelihood is calculated comprehensively from a plurality of factors. The elements include, for example, whether the width of the feature points is close to the line width of the road lane marking, whether the color at the feature point is close to the color of the road lane marking, the feature point and its surroundings. There are factors such as whether or not the brightness difference between the two is close to the brightness difference inside and outside the road marking line, and whether or not the feature point is close to the extension line of the road marking line recognized in the past.

The closer the width between the feature points is to the line width of the road marking line, the higher the likelihood of those feature points. The closer the color at the feature point is to the color of the road marking line, the higher the likelihood of the feature point. The closer the luminance difference between the feature point and its surroundings is to the brightness difference inside and outside the road marking line, the higher the likelihood of the feature point. The closer the feature point is to the extension of the road marking line recognized in the past, the higher the likelihood of the feature point.

Next, the estimation processing unit 35 has a feature point whose likelihood calculated as described above is equal to or higher than a predetermined threshold value (hereinafter referred to as a high likelihood feature point) and a feature point whose likelihood is less than the threshold value. Each feature point (hereinafter referred to as a low-likelihood feature point) is counted. Next, the estimation processing unit 35 estimates the first reliability based on the count number of the high likelihood feature points and the count number of the low likelihood feature points. The first reliability is higher as the number of counts of high-likelihood feature points is larger, and is higher as the number of counts of low-likelihood feature points is smaller.

For example, as shown in FIG. 6A, when a part 93 of the road marking line 91 that divides the first lane 87 is faint or disappears, the count number of the high likelihood feature points decreases. The first reliability is low. Further, for example, as shown in FIG. 6B, when the feature point 95 caused by an event other than the road lane marking 91 exists inside the road lane marking 91, the count number of the low likelihood feature points increases, and the number of counts becomes higher. The reliability of 1 becomes low. Examples of events other than the road marking line 91 include construction traces, falling objects, and the like. Examples of construction marks include tar, repair marks, and forgetting to erase road markings.

On the other hand, when there is no faint or disappeared part in the road lane marking 91 and the number of feature points 95 caused by an event other than the road lane marking 91 is small, the count number of the high likelihood feature points is large. Therefore, the number of counts of the low-likelihood feature points is small, so that the first reliability is high.

The method of estimating the second reliability is basically the same as the method of estimating the first reliability. However, when estimating the second reliability, the feature points extracted when recognizing the road marking line that divides the second lane are used.

In step 8, the contrast unit 15 determines whether or not the second reliability estimated in step 7 is higher than the first reliability. If the second reliability is higher than the first reliability, the process proceeds to step 9, and if not, the process proceeds to step 10.

In step 9, the instruction unit 17 instructs the vehicle control unit 75 to change lanes. In this case, the vehicle control unit 75 executes a lane change from the first lane to the second lane. Further, the instruction unit 17 uses the speaker 77 to perform notification at a timing earlier than the execution of the lane change.

In step 10, the double-sided lane determination unit 23 determines whether or not there are other lanes on both sides of the first lane. The both-side lane determination unit 23 can determine whether or not there are other lanes on both sides of the first lane based on the images acquired in steps 5 and 6.

For example, as shown in FIG. 7, on a road having lanes 97, 99, 101, 103, when the first lane is lanes 99, 101, there are other lanes on both sides of the first lane. .. On the other hand, when the first lane is lanes 97 and 103, there are no other lanes on at least one side of both sides of the first lane.

If there is no other lane on at least one of the two sides of the first lane, the process proceeds to step 11. If there are other lanes on both sides of the first lane, this process ends.

In step 11, the position acquisition unit 19 acquires the position of the own vehicle using GPS69.
In step 12, the road environment acquisition unit 21 acquires the road environment around the own vehicle by using the position of the own vehicle acquired in the step 11 and the map information stored in the map database 5. As shown in FIG. 4, the road environment to be acquired is the road environment in the area 105 in front of the own vehicle 79.

In step 13, the instruction unit 17 determines whether or not the road environment acquired in step 12 is a specific road environment set in advance. The specific road environment is a road environment in which the reliability when recognizing the road lane marking that divides the first lane tends to be low.

As a specific road environment, for example, as shown in FIG. 3A, there is a sharp curve in front of the own vehicle 79, and the first lane (lane 67 in the example shown in FIG. 3A) is an in-course of the sharp curve. There is a road environment. In this road environment, the in-course area 107 tends to deviate from the angle of view of the camera 71, so that the reliability when recognizing the road marking line that divides the first lane tends to be low.

Further, as a specific road environment, for example, as shown in FIG. 3B, there is a road environment in which the number of lanes decreases due to the lane 47 disappearing in the middle, and the lane 47 is the first lane. .. In this road environment, since the shape of the road lane marking is complicated in the area 109 in front of the lane 47, the reliability when recognizing the road lane marking that divides the first lane tends to be low.

Further, as a specific road environment, for example, as shown in FIG. 3C, there is a road environment in which lane 53 branches off from lane 57, and lane 53 is the first lane. In this road environment, since the shape of the road lane marking is complicated in the region 111 in front of the lane 53, the reliability when recognizing the road lane marking that divides the first lane tends to be low.

Further, as a specific road environment, for example, as shown in FIG. 3D, there is a road environment in which the lane 63 joins the lane 59 and the lane 59 is the first lane. In this road environment, since the shape of the road lane marking is complicated in the region 113 in front of the lane 59, the reliability when recognizing the road lane marking that divides the first lane tends to be low.

Further, as a specific road environment, for example, a road environment in which the left and right road lane markings for dividing the first lane are not parallel, a road environment in which the number of lanes increases, and at least one of the road lane markings for dividing the first lane Examples include a road environment in which the part is botts, a road environment in which a zebra zone exists in the first lane, and a road environment in which road surface paint exists in the first lane. These specific road environments are also road environments in which the reliability when recognizing the road lane markings that demarcate the first lane tends to be low.

If the road environment is a specific road environment, the process proceeds to step 9, otherwise the process ends.
3. 3. Effects of the driving support device 1 (1A) The driving support device 1 compares the first reliability and the second reliability, and when the second reliability is higher than the first reliability, the second reliability is higher than the first reliability. Instruct a lane change from the first lane to the second lane. When the lane is changed to the second lane, the reliability in recognizing the road marking line becomes higher than when traveling in the first lane. As a result, it is possible to prevent the road marking line from being misrecognized.

(1B) The driving support device 1 acquires an image representing an area including a road, and extracts feature points from the acquired image. Further, the driving support device 1 calculates the likelihood that the feature point is a feature point caused by the road marking line for each extracted feature point, and based on the likelihood, the first reliability and the second Estimate reliability. Therefore, the first reliability and the second reliability can be estimated more accurately.

(1C) The driving support device 1 acquires the road environment around the own vehicle, and when the road environment is a specific preset road environment, the lane from the first lane to the second lane. Instruct a change. When the lane is changed to the second lane which is not a specific road environment, the reliability in recognizing the road lane marking is higher than when traveling in the first lane. As a result, it is possible to prevent the road marking line from being misrecognized.

(1D) When there are other lanes on both sides of the first lane, it is unlikely that the reliability in recognizing the road lane marking will be lowered due to the structure of the road and the shape of the road. The driving support device 1 does not instruct a lane change when there are other lanes on both sides of the first lane. Therefore, it is possible to suppress instructing a lane change that is less necessary.

(1E) The driving support device 1 does not instruct a lane change when there are surrounding vehicles. Therefore, it is possible to prevent the own vehicle 79 from coming into contact with neighboring vehicles at the time of lane change.

(1F) When there is a vehicle in front, the own vehicle 79 may follow the vehicle in front, so that even if the reliability in recognizing the road lane marking is lowered, the problem is unlikely to occur. The driving support device 1 does not instruct a lane change when a vehicle in front is present. Therefore, it is possible to suppress instructing a lane change that is less necessary.

(1G) The driving support device 1 determines whether or not a peripheral vehicle and a vehicle in front exist before the processing in step 5 and thereafter, and if it determines that a peripheral vehicle or a vehicle in front exists, that time point. Ends the process with. As a result, the calculation load of the control unit 3 can be reduced.
<Other embodiments>
Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and can be implemented in various modifications.

(1) If a negative determination is made in step 8, this process may be terminated. That is, the processes of steps 10 to 13 need not be executed.
(2) Immediately after the step 6, the process may proceed to the step 10. That is, the processes of steps 7 and 8 need not be executed.

(3) A plurality of functions possessed by one component in the above embodiment may be realized by a plurality of components, or one function possessed by one component may be realized by a plurality of components. .. Further, a plurality of functions possessed by the plurality of components may be realized by one component, or one function realized by the plurality of components may be realized by one component. Further, a part of the configuration of the above embodiment may be omitted. In addition, at least a part of the configuration of the above embodiment may be added or replaced with the configuration of the other above embodiment. It should be noted that all aspects included in the technical idea specified from the wording described in the claims are embodiments of the present disclosure.

(4) In addition to the above-mentioned driving support device 1, a system having the driving support device 1 as a component, a program for operating a computer as a control unit 3 of the driving support device 1, a semiconductor memory recording this program, and the like. The present disclosure can also be realized in various forms such as a non-transitional actual recording medium and a driving support method.

1 ... Driving support device, 3 ... Control unit, 7 ... CPU, 9 ... Memory, 11 ... Road lane marking unit, 13 ... Reliability calculation unit, 15 ... Comparison unit, 17 ... Instruction unit, 19 ... Position acquisition unit, 21 ... Road environment acquisition unit, 23 ... Both sides lane judgment unit, 25 ... Other vehicle judgment unit, 27 ... Image acquisition unit, 29 ... Extraction unit, 31 ... Recognition processing unit, 33 ... Feature point likelihood calculation unit, 35 ... Estimating processing unit, 69 ... GPS, 71 ... camera, 73 ... vehicle detection sensor, 75 ... vehicle control unit, 77 ... speaker, 79 ... own vehicle, 89 ... in-vehicle network

Claims (4)

A road lane marking unit that recognizes the road lane marking in front of the vehicle,
A reliability estimation unit that estimates the reliability of road lane marking recognition by the road lane marking unit,
The reliability in recognizing the road lane marking that divides the first lane in which the vehicle is traveling, and the reliability in recognizing the road lane marking that divides the second lane adjacent to the first lane. And the contrasting unit that contrasts
The second, provided that the reliability in recognizing the road lane marking that divides the second lane is higher than the reliability in recognizing the road lane marking that divides the first lane. An instruction unit that instructs the lane change to the lane, and
With
The road lane marking unit is
An image acquisition unit that acquires an image representing an area including a road,
An extraction unit that extracts feature points in the image,
A recognition processing unit that recognizes the road marking line based on the feature points,
With
The reliability estimation unit is
For each of the feature points, a feature point likelihood calculation unit that comprehensively calculates the likelihood that the feature point is a feature point caused by a road lane marking from a plurality of elements,
An estimation processing unit that estimates the reliability based on the likelihood, and
A driving support device equipped with. A road lane marking unit that recognizes the road lane marking in front of the vehicle,
A position acquisition unit that acquires the position of the vehicle and
A map database that stores map information and
A road environment acquisition unit that acquires the road environment around the vehicle by using the position of the vehicle acquired by the position acquisition unit and the map information.
An instruction unit that instructs a lane change from the lane in which the vehicle is traveling to another lane, provided that the road environment acquired by the road environment acquisition unit is a specific preset road environment. When,
A two-sided lane determination unit that determines whether or not there are other lanes on both sides of the lane in which the vehicle is traveling,
With
The instruction unit is a driving support device configured not to instruct the lane change when the both side lane determination unit determines that there are other lanes on both sides of the lane in which the vehicle is traveling . The driving support device according to claim 2 .
The road environment is a driving support device including a road structure or a road shape. The driving support device according to any one of claims 1 to 3 .
Further equipped with another vehicle determination unit for determining whether or not another vehicle exists in the vicinity of the vehicle.
The instruction unit is a driving support device configured not to instruct the lane change when the other vehicle determination unit determines that the other vehicle is present.

Images