KR20210058350A - Apparatus for calculating distance between vehicles and method thereof - Google Patents

Abstract

The present invention relates to a device and method for calculating a distance between vehicles, which includes: a camera sensor which photographs the front image of the own vehicle; and a distance calculation unit which obtains lane width information based on a lane detected from the front image photographed through the camera sensor, and applies position information of the camera sensor installed in the own vehicle and the obtained lane width information in a predefined geometric model to calculate the inter-vehicle distance from the own vehicle to the front vehicle.

Description

Device and method for calculating the distance between vehicles {APPARATUS FOR CALCULATING DISTANCE BETWEEN VEHICLES AND METHOD THEREOF}

The present invention relates to an apparatus and method for calculating an inter-vehicle distance, and more particularly, to an apparatus and method for calculating an inter-vehicle distance to a vehicle in front using a camera sensor mounted on a vehicle.

An autonomous vehicle refers to a vehicle that independently determines a driving route by recognizing the surrounding environment through a function of detecting and processing external information during driving, and driving independently using its own power. Autonomous vehicles can travel to their destination by preventing collisions with obstacles on the driving path and adjusting the vehicle speed and driving direction according to the shape of the road, even if the driver does not operate the steering wheel, accelerator pedal, or brake. have. For example, acceleration may be performed on a straight road, and deceleration may be performed while changing a driving direction in response to a curvature of the road on a curved road.

In such autonomous vehicles, a plurality of driver assistance systems (ADAS: Advanced Driver Assistance System) are applied to assist the driver's driving, and driver assistance systems include ACC (Adaptive Cruise Control), LDWS (Lane Departure Warning System), and There are Lane Keeping Assist System (LKAS), High Beam Assist (HBA), and Autonomous Emergency Braking (AEB).

Meanwhile, sensors used in autonomous driving technology include a camera sensor, a radar sensor, and a lidar sensor. Among them, since the camera sensor is easy to process data, detects an object, and is inexpensive, prior studies are being conducted to calculate the distance between cars based on the camera sensor. However, there is a difficulty in extracting an accurate inter-vehicle distance with only the camera sensor, and the accuracy is very low. In addition, researches on obtaining the position and distance of a front object by utilizing stereo vision technology using two cameras or by fusion with other sensors are active, but studies on obtaining the distance to the front object with a single camera sensor or a monocular camera sensor are insufficient. Actually.

The background technology of the present invention is disclosed in Korean Patent Application Publication No. 10-2017-0117855 (published on October 24, 2017).

An object according to an aspect of the present invention is to obtain a lane width in an image based on a lane detected from a front image of a monocular camera sensor and a vanishing point, and calculate an inter-vehicle distance from the host vehicle to the vehicle in front based on geometric information, Even when a monocular camera sensor is used, an apparatus and method for calculating a distance between vehicles capable of improving the accuracy of the distance are provided.

The vehicle-to-vehicle distance calculation apparatus according to an aspect of the present invention acquires lane width information based on a camera sensor for photographing a front image of an own vehicle, and a lane detected from a front image photographed through the camera sensor, and And a distance calculator configured to calculate an inter-vehicle distance from the host vehicle to a vehicle in front by applying the location information of the camera sensor installed in and the obtained lane width information to a predefined geometric model.

In the present invention, the lane width information includes a first lane width L1 based on a wheel of the front vehicle and a second lane width L2 based on the bonnet of the host vehicle. do.

In the present invention, the location information of the camera sensor includes a first height h1 of the camera sensor based on the ground, a second height h2 of the camera sensor based on the bonnet of the host vehicle, and the ground. It includes a first distance (b) from the bonnet to the camera sensor based on a parallel direction, and a second distance (c) from the front end of the host vehicle to the camera sensor based on a direction parallel to the ground. Characterized in that.

In the present invention, the distance calculation unit is a distance from the foremost end of the host vehicle to the end point of the blind spot of the camera sensor formed by the bonnet of the host vehicle based on the longitudinal direction of the host vehicle (a) _{And, after calculating the distance (d image} ) from the terminal point to the wheel of the front vehicle based on the geometric model, the distance _{between the vehicles using the calculated distance (a, d image} ) and the specification information of the front vehicle It is characterized in that the distance from the frontmost end of the host vehicle to the rearmost end of the front vehicle is calculated as a distance (d _{real ).}

In the present invention, the distance calculation unit calculates the distance (a) using the first and second heights (h1, h2) and the first and second distances (b, c) based on the geometric model. Characterized in that.

In the present invention, the distance calculation unit connects the camera sensor and the wheel of the front vehicle from the ground at the end point by using the first and second lane widths L1 and L2 based on the geometric model. A second connecting the camera sensor and the end point by calculating the height (l) to the first straight line, and using the distance (a), the second distance (c), and the first height (h1) Calculate a first angle (θ) formed by a straight line perpendicular to the ground, and use the first angle (θ), the distance (a), the second distance (c), and the first height (h1) Thus, after calculating the second angle α formed by the first straight line and the second straight line, the distance is determined by using the height l, the first angle θ, and the second angle α. It is characterized in that (d _{image) is calculated.}

_{In the present invention, the distance calculation unit subtracts a rear overhang length (d OH} ) as the specification information of the front vehicle from the sum of the distance (a) and the distance (d _image ), and the inter-vehicle distance It is characterized by calculating (d _{real ).}

In the present invention, the camera sensor is a monocular camera sensor.

The present invention is characterized in that it further comprises a control unit for controlling an operation of a driver assistance system (DAS) applied to the own vehicle based on the inter-vehicle distance calculated through the distance calculation unit.

In the method for calculating a distance between vehicles according to an aspect of the present invention, the distance calculating unit obtains a front image through a camera sensor for capturing a front image of the own vehicle, and the distance calculating unit is based on a lane detected from the front image. Acquiring lane width information, and the distance calculating unit applies the position information of the camera sensor installed in the host vehicle and the obtained lane width information to a predefined geometric model, and the inter-vehicle distance from the host vehicle to the vehicle in front It characterized in that it comprises the step of calculating.

According to an aspect of the present invention, the present invention obtains a lane width in an image based on a lane detected from a front image of a camera sensor and a vanishing point, and calculates the inter-vehicle distance from the own vehicle to the vehicle in front based on a predetermined geometric model. By calculating, even when a monocular camera sensor is used, the accuracy of the distance can be improved, and the accuracy of the autonomous driving control can be improved by utilizing the calculated inter-vehicle distance for autonomous driving control.

1 is a block diagram illustrating an apparatus for calculating a distance between vehicles according to an embodiment of the present invention.
2 and 3 are exemplary views illustrating a process of calculating an inter-vehicle distance in the apparatus for calculating an inter-vehicle distance according to an embodiment of the present invention.
4 is a flowchart illustrating a method of calculating a distance between vehicles according to an embodiment of the present invention.

Hereinafter, an embodiment of an apparatus and method for calculating an inter-vehicle distance according to the present invention will be described with reference to the accompanying drawings. In this process, the thickness of the lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of description. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of users or operators. Therefore, definitions of these terms should be made based on the contents throughout the present specification.

1 is a block diagram illustrating an apparatus for calculating an inter-vehicle distance according to an embodiment of the present invention, and FIGS. 2 and 3 are for calculating an inter-vehicle distance in an inter-vehicle distance calculating apparatus according to an embodiment of the present invention. It is an exemplary diagram for explaining the process.

Referring to FIG. 1, an apparatus for calculating a distance between vehicles according to an embodiment of the present invention may include a camera sensor 100, a distance calculating unit 200, and a control unit 300.

The camera sensor 100 may capture an image in front of the own vehicle and transmit it to the distance calculator 200 to be described later. The camera sensor 100 may be a front camera sensor installed on the top of the windshield inside the own vehicle to capture a front image of the own vehicle, or may be a monocular camera sensor.

The distance calculating unit 200 may calculate an inter-vehicle distance to the vehicle in front based on the front image captured through the camera sensor 100. In order to help understand the embodiment, the distance calculation unit 200 and the control unit 300 to be described later are described as separate configurations, but the functions of the distance calculation unit 200 of the present embodiment may be implemented by being integrated into the control unit 300. . The distance calculation unit 200 and the control unit 300 may be implemented as a processor or a system on chip (SoC) that performs processing and calculation of various data, and uses instructions or data received from other components. It can be designed to load and process a memory, and to store various data in the memory.

A process in which the distance calculating unit 200 calculates the inter-vehicle distance from the host vehicle to the vehicle in front will be described in detail.

In this embodiment, the distance calculating unit 200 acquires lane width information based on the lane detected from the front image photographed through the camera sensor 100, and obtains the position information of the camera sensor 100 installed in the own vehicle and the acquisition. By applying the determined lane width information to a predefined geometric model, the inter-vehicle distance from the host vehicle to the vehicle in front can be calculated. Here, the lane width information may include a first lane width L1 based on a wheel of the vehicle in front and a second lane width L2 based on the bonnet of the host vehicle.

2 shows an example of a process of obtaining lane width information from a front image captured through the camera sensor 100. The left and right lanes are detected from the front image, and a point where the detected left and right lanes meet forms a vanishing point.

In the front image, the first lane width (L1) refers to the distance between the left and right lanes based on the point where the wheel (rear wheel) of the vehicle in front and the ground contact, and the second lane width (L2) is the bonnet of the own vehicle. It means the distance between the left and right lanes based on (B, e.g., the top of the bonnet in the front image). Since the first lane width L1 and the second lane width L2 have distance values in the front image, the value of the first lane width L1 is smaller than the value of the second lane width L2. . In FIG. 2, the height H1 means the vertical distance between the first lane width L1 and the second lane width L2, and the height H2 is from the vanishing point to the straight line formed by the first lane width L1. Means the vertical distance. An image processing algorithm for calculating the lane widths L1 and L2 and the heights H1 and H2 from the front image captured through the camera sensor 100 may be predefined in the distance calculator 200 ( In addition, the rear overhang length (d _OH ) of the front vehicle to be described later may also be obtained through the image processing algorithm).

When the lane width information is obtained through the above-described process, the distance calculating unit 200 applies the location information and the lane width information of the camera sensor 100 installed in the own vehicle to a predefined geometric model, You can calculate the distance between cars. 3 is a schematic diagram of a geometric model to aid in understanding of the present embodiment, and specific equations follow Equations 2 to 6 to be described later. 3, the location information of the camera sensor 100 considered in the geometric model is the first height h1 of the camera sensor 100 based on the ground, and the camera sensor based on the bonnet of the own vehicle. The second height (h2) of (100), the first distance (b) from the bonnet to the camera sensor 100 based on a direction parallel to the ground, and from the foremost end of the host vehicle based on a direction parallel to the ground The second distance c to the camera sensor 100 may be included.

As shown in FIG. 3, the inter-vehicle distance d _real finally calculated in this embodiment is the distance from the frontmost end of the own vehicle (eg, front bumper) to the rearmost end of the front vehicle (eg, rear bumper), In order to calculate such an inter-vehicle distance (d _real ), in this embodiment, three distance parameters, namely, distance (a), distance (d _image ), and rear overhang length (d _OH ) Is employed.

The distance (a) is the end point of the blind spot of the camera sensor 100 formed by the bonnet of the host vehicle from the foremost end of the host vehicle (P in Fig. 3, the longitudinal direction of the host vehicle) Refers to the distance to the front end point of the blind spot of the camera sensor 100). The distance (d _image ) refers to the distance from the aforementioned termination point to the wheel (rear wheel) of the front vehicle. When following the definition of the distance (a) and the distance (d _image) as described above, in this embodiment the inter-vehicle distance is the distance (a) and the distance (d _image) the rear overhang of the front vehicle at a value adding (rear overhang) length ( d _OH ) can be calculated by subtracting. This can be expressed as Equation 1 below.

[ Equation 1 ]

Now, the process of calculating the distance (a) and the distance (d _image ), which are factors of Equation 1, will be described in detail.

First, the distance calculating unit 200 may calculate the distance a using the first and second heights h1 and h2 and the first and second distances b and c based on the geometric model. The location information of the camera sensor 100 currently installed in the own vehicle may be previously stored in the distance calculating unit 200. Using the proportional equation with reference to FIG. 3, the distance a may be calculated according to Equation 2 below.

[ Equation 2 ]

Next, the process of calculating the distance (d _image ) is the process of calculating the height (l) in FIG. 3 based on the geometric model, the process of calculating the first angle (θ), and calculating the second angle (α). It may consist of a process of doing and a process of calculating a distance (d _image).

Specifically, the distance calculation unit 200 uses the first and second lane widths L1 and L2, and the height from the ground to the first straight line connecting the camera sensor 100 and the wheel of the vehicle in front at the end point (l) can be calculated. The second lane width (L2) obtained from the front image corresponds to the first height (h1), and the difference between the second lane width (L2) and the first lane width (L1) corresponds to the above height (l). When considering the proportional relationship, the height (l) can be calculated according to Equation 3 below.

[ Equation 3 ]

And, the distance calculation unit 200 uses the distance (a), the second distance (c), and the first height (h1), the second straight line connecting the camera sensor 100 and the end point is perpendicular to the ground. The first angle θ formed with the straight line may be calculated. Referring to FIG. 3, the first angle θ may be calculated according to Equation 4 below.

[ Equation 4 ]

In addition, the distance calculating unit 200 uses the first angle θ, the distance a, the second distance c, and the first height h1 to form a second angle formed by the first straight line and the second straight line. (α) can be calculated. Referring to FIG. 3, the second angle α may be calculated according to Equation 5 below.

[ Equation 5 ]

_{Finally, the distance calculator 200 may calculate the distance d image} using the height l, the first angle θ, and the second angle α. Referring to FIG. 3, the distance d _image may be calculated according to Equation 6 below.

[ Equation 6 ]

When the distance a and the distance d _image are calculated through the above-described process, the final inter-vehicle distance d _real may be calculated according to Equation 1 above.

Tables 1 and 2 below are experimental data showing values obtained from a front image captured through the camera sensor 100 and an error rate between the measured values. From Tables 1 and 2, it can be seen that the error rate of the inter-vehicle distance detection method according to the present embodiment has high reliability in the range of 0% to 5%.

[ Table 1 ]

[ Table 2 ]

Meanwhile, in the present embodiment, as shown in FIG. 1, a control unit 300 for controlling the operation of a driver assistance system (DAS) applied to the own vehicle based on the inter-vehicle distance calculated through the distance calculating unit 200 ) May be further included. The driving assistance system can mean a variety of systems that utilize the distance between cars as a control factor, and such as ACC (Adaptive Cruise Control), AEB (Autonomous Emergency Braking), LKAS (Lane Keeping Assist System), HDA (Highway Driving Assist), etc. It may include.

4 is a flowchart illustrating a method of calculating a distance between vehicles according to an embodiment of the present invention. A method of calculating the distance between vehicles according to the present embodiment will be described with reference to FIG. 4, and descriptions will be made focusing on the time-series configuration, omitting the content overlapping with the previously described content.

First, the distance calculating unit 200 acquires a front image through the camera sensor 100 that photographs the front image of the own vehicle (S100).

Subsequently, the distance calculating unit 200 acquires lane width information based on the lane detected from the front image (S200).

Subsequently, the distance calculation unit 200 calculates the inter-vehicle distance from the host vehicle to the vehicle in front by applying the location information and the lane width information of the camera sensor 100 installed in the host vehicle to a predefined geometric model (S300).

The above-described lane width information may include a first lane width L1 based on a wheel of a vehicle in front and a second lane width L2 based on a bonnet of the host vehicle. In addition, the location information of the camera sensor 100 includes a first height h1 of the camera sensor 100 based on the ground, a second height h2 of the camera sensor 100 based on the bonnet of the own vehicle, A first distance (b) from the bonnet to the camera sensor 100 based on a direction parallel to the ground, and a second distance from the front end of the own vehicle to the camera sensor 100 based on a direction parallel to the ground ( c).

Accordingly, in step S300, the distance calculation unit 200 is the distance from the front end of the host vehicle to the end point of the blind spot of the camera sensor 100 formed by the bonnet of the host vehicle based on the longitudinal direction of the host vehicle. After calculating (a) and the distance from the end point to the wheel of the vehicle in front (d _image ) based on the geometric model, the calculated distance (a, d _image ) and the data of the vehicle in front are used as the distance between vehicles. The distance from the front end of the vehicle to the rear end of the vehicle in front is calculated.

To describe the step S300 in more detail, the distance calculation unit 200 uses the first and second heights h1 and h2 and the first and second distances b and c based on the geometric model. ) Is calculated.

In addition, in step S300, the distance calculation unit 200 connects the camera sensor 100 and the wheel of the front vehicle from the ground at the end point using the first and second lane widths L1 and L2 based on the geometric model. A second straight line connecting the camera sensor 100 and the end point by calculating the height (l) to the first straight line, and using the distance (a), the second distance (c), and the first height (h1) A first angle (θ) formed with a straight line perpendicular to the ground is calculated, and a first straight line is obtained using the first angle (θ), the distance (a), the second distance (c), and the first height (h1). And after calculating the second angle α formed by the second straight line, the distance d _image is calculated using the height 1, the first angle θ, and the second angle α.

In step S300, the distance calculation unit 200 finally subtracts the _{rear overhang length d OH} as the specification information of the vehicle in front from the sum of the _{distance a and the distance d image} Yields

After step S300, step S400 in which the controller 300 controls the operation of the driving assistance system applied to the own vehicle based on the inter-vehicle distance as illustrated in FIG. 4 may be further performed.

As described above, the present embodiment acquires the lane width in the image based on the detected lane and vanishing point detected from the front image of the camera sensor 100, and calculates the inter-vehicle distance from the host vehicle to the vehicle in front based on a predetermined geometric model. By calculating, even when the monocular camera sensor 100 is used, the accuracy of the distance can be improved, and the accuracy of the autonomous driving control can be improved by utilizing the calculated inter-vehicle distance for autonomous driving control.

The implementations described herein may be implemented in, for example, a method or process, an apparatus, a software program, a data stream or a signal. Although discussed only in the context of a single form of implementation (eg, only as a method), the implementation of the discussed features may also be implemented in other forms (eg, an apparatus or program). The device may be implemented with appropriate hardware, software and firmware. The method may be implemented in an apparatus such as a processor, which generally refers to a processing device including, for example, a computer, a microprocessor, an integrated circuit or a programmable logic device, or the like. Processors also include communication devices such as computers, cell phones, personal digital assistants (“PDAs”) and other devices that facilitate communication of information between end-users.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is only exemplary, and those of ordinary skill in the art to which the present technology pertains, various modifications and other equivalent embodiments are possible. I will understand. Therefore, the true technical protection scope of the present invention should be determined by the following claims.

100: camera sensor
200: distance calculation unit
300: control unit

Claims (17)

A camera sensor for photographing a front image of the own vehicle; And
Lane width information is obtained based on the lane detected from the front image captured by the camera sensor, and the position information of the camera sensor installed in the host vehicle and the obtained lane width information are applied to a predefined geometric model to A distance calculating unit that calculates an inter-vehicle distance from the vehicle to the vehicle in front;
Vehicle-to-vehicle distance calculation device comprising a.
The method of claim 1,
The lane width information includes a first lane width (L1) based on a wheel of the vehicle in front and a second lane width (L2) based on a bonnet of the host vehicle. Output device.
The method of claim 2,
The location information of the camera sensor includes a first height h1 of the camera sensor based on the ground, a second height h2 of the camera sensor based on the bonnet of the host vehicle, and a direction parallel to the ground. It characterized in that it comprises a first distance (b) from the bonnet to the camera sensor as a reference, and a second distance (c) from the front end of the host vehicle to the camera sensor based on a direction parallel to the ground. A device for calculating the distance between vehicles.
The method of claim 3,
The distance calculation unit,
The distance (a) from the foremost end of the host vehicle to the end point of the blind spot of the camera sensor formed by the bonnet of the host vehicle based on the longitudinal direction of the host vehicle, and the front vehicle from the end point After calculating the distance (d _image ) to the wheel of the vehicle based on the geometric model, using the calculated distance (a, d _image ) and the specification information of the vehicle in front of the vehicle as the distance between the vehicles (d _real ) An inter-vehicle distance calculating device, characterized in that calculating a distance from the frontmost end of the vehicle to the rearmost end of the front vehicle.
The method of claim 4,
The distance calculation unit, based on the geometric model,
An apparatus for calculating a distance between vehicles, comprising calculating the distance (a) by using the first and second heights (h1, h2) and the first and second distances (b, c).
The method of claim 5,
The distance calculation unit, based on the geometric model,
Using the first and second lane widths L1 and L2, a height l from the ground to a first straight line connecting the camera sensor and the wheel of the vehicle in front is calculated from the end point,
Using the distance (a), the second distance (c), and the first height (h1), a first angle formed by a second straight line connecting the camera sensor and the end point with a straight line perpendicular to the ground ( θ),
Using the first angle (θ), the distance (a), the second distance (c), and the first height (h1), a second angle (α) formed by the first straight line and the second straight line After calculating
The distance calculation device between vehicles, characterized in that _{the distance (d image} ) is calculated using the height (l), the first angle (θ), and the second angle (α).
The method of claim 6,
The distance calculation unit,
_{The distance (d real} ) is calculated by subtracting the _{rear overhang length (d OH} ) as the specification information of the front vehicle from the sum of the distance (a) and the distance (d _{image ).} A device for calculating the distance between vehicles.
The method of claim 1,
The camera sensor is a vehicle distance calculation device, characterized in that the monocular camera sensor.
The method of claim 1,
And a control unit for controlling an operation of a driver assistance system (DAS) applied to the own vehicle based on the inter-vehicle distance calculated through the distance calculating unit.
Obtaining, by a distance calculating unit, a front image through a camera sensor that photographs a front image of the own vehicle;
Obtaining, by the distance calculating unit, lane width information based on the lane detected from the front image; And
Calculating an inter-vehicle distance from the host vehicle to a front vehicle by applying the position information of the camera sensor installed in the host vehicle and the obtained lane width information to a predefined geometric model, by the distance calculating unit;
Vehicle-to-vehicle distance calculation method comprising a.
The method of claim 10,
The lane width information includes a first lane width (L1) based on a wheel of the vehicle in front and a second lane width (L2) based on a bonnet of the host vehicle. Calculation method.
The method of claim 11,
The location information of the camera sensor includes a first height h1 of the camera sensor based on the ground, a second height h2 of the camera sensor based on the bonnet of the host vehicle, and a direction parallel to the ground. It characterized in that it comprises a first distance (b) from the bonnet to the camera sensor as a reference, and a second distance (c) from the front end of the host vehicle to the camera sensor based on a direction parallel to the ground. How to calculate the distance between vehicles.
The method of claim 12,
In the calculating step, the distance calculating unit,
The distance (a) from the foremost end of the host vehicle to the end point of the blind spot of the camera sensor formed by the bonnet of the host vehicle based on the longitudinal direction of the host vehicle, and the front vehicle from the end point After calculating the distance (d _image ) to the wheel of the vehicle based on the geometric model, using the calculated distance (a, d _image ) and the specification information of the vehicle in front of the vehicle as the distance between the vehicles (d _real ) And calculating the distance from the frontmost end of the vehicle to the rearmost end of the vehicle in front of the vehicle.
The method of claim 13,
In the calculating step, the distance calculating unit, based on the geometric model,
The distance (a) is calculated using the first and second heights (h1, h2) and the first and second distances (b, c).
The method of claim 14,
In the calculating step, the distance calculating unit, based on the geometric model,
Using the first and second lane widths L1 and L2, a height l from the ground to a first straight line connecting the camera sensor and the wheel of the vehicle in front is calculated from the end point,
Using the distance (a), the second distance (c), and the first height (h1), a first angle formed by a second straight line connecting the camera sensor and the end point with a straight line perpendicular to the ground ( θ),
Using the first angle (θ), the distance (a), the second distance (c), and the first height (h1), a second angle (α) formed by the first straight line and the second straight line After calculating
_{The distance (d image} ) is calculated using the height (l), the first angle (θ) and the second angle (α).
The method of claim 15,
In the calculating step, the distance calculating unit,
_{The distance (d real} ) is calculated by subtracting the _{rear overhang length (d OH} ) as the specification information of the front vehicle from the sum of the distance (a) and the distance (d _{image ).} How to calculate the distance between vehicles.
The method of claim 10,
Controlling, by a controller, an operation of a driver assistance system (DAS) applied to the own vehicle based on the calculated distance between vehicles.

Images