KR100559379B1 - Traveling deviation revision data creation system via image tracking of leading vehicle and method thereof - Google Patents

Abstract

The present invention relates to an automatic driving apparatus and method for a vehicle by adjusting a driving angle of the vehicle by analyzing a road image obtained from an image photographing means installed at the front of the vehicle to be driven, and controlling the speed to allow unmanned driving. It is about.

The present invention calculates the steering angle of the vehicle using the distance between the vehicle axes (L) after calculating the traveling direction of the vehicle, the photographing angle (θ) of the image photographing apparatus, and the distance between the preceding vehicle. By analyzing the distance (d), it is possible to automatically control the vehicle by generating data to drive the brakes and the accelerator.

Steering angle, shooting angle, distance between cars, speed control, leading vehicle

Description

Traveling angle revision data creation system via image tracking of leading vehicle and method

1 is a conceptual diagram of a known radius of rotation,

2 is an outer wheel reference steering concept according to the present invention,

Figure 3 is a photographing means based steering conceptual diagram according to the present invention,

4A is a conceptual diagram illustrating initialization of an image photographing means of the present invention;

Figure 4b is a conceptual diagram of the inter-vehicle distance (d) of the present invention,

5A is a conceptual view illustrating a photographing angle correction for image tracking according to the present invention;

5B is a conceptual diagram of a photographing angle deviation Δθ of the present invention;

6 is a conceptual diagram of a preceding vehicle tracking according to the present invention.

7 is a state in which the image recording module is mounted on the vehicle

8 is a block diagram of a vehicle driving apparatus using the preceding vehicle image tracking according to the present invention;

9 is a block diagram of an H / W device according to the present invention;

* Description of the symbols for the main parts of the drawings *

10; video recording means 20; video data memory means

30; vehicle image selecting means 31; vehicle image separating means

50; steering angle calculation means 60; state value generating means

70; drive part

The present invention relates to the generation of driving data of an automatic traveling device installed in a vehicle and capable of unmanned driving. Specifically, an image capturing means is installed on a front surface of a traveling vehicle, and the image data obtained from the image capturing means is analyzed and the vehicle is generated. The present invention relates to an automatic traveling device and a method for driving a vehicle by adjusting a steering angle and controlling a speed.

The present invention calculates the steering angle of the vehicle using the distance between the vehicle axes (L) after calculating the traveling direction of the vehicle, the photographing angle (θ) of the image photographing apparatus, and the distance between the preceding vehicle. The present invention relates to a driving data generation method for automatically controlling a vehicle by generating data for driving a brake and an accelerator by analyzing a distance d.

As a conventional automatic driving device, the Patent Application Publication No. 10-0257592 discloses lane markings on both sides of a road by a lane detection sensor so as to detect a lane curve in a predetermined manner, and to travel according to the detected curve. A method is disclosed.

Patent Publication (Application No. 10-2000-0009711) discloses "a geometric mounting technique and a steering angle calculation method of an image photographing means for autonomous steering", but a specific method for calculating the steering angle is not presented, the radius ( The method of calculating R) has no choice but to use an estimation formula based on experimental values, and the method of calculating the steering angle disclosed is a known technique disclosed in a general vehicle theory.

Further, after obtaining the horizontal deviation (Δx) between the road and the optical imaging device optical axis as disclosed in the patent application No. 10-2003-0004788 filed by the applicant, the horizontal deviation (Δx) and the distance between the vehicle shafts (L) The steering angle is calculated by calculating the steering angle of the vehicle using the image capturing means and the horizontal distance d, and the brake and the accelerator are analyzed by analyzing the vertical deviation Δy and the position of the preceding vehicle among the front image data of the vehicle. An apparatus and a method for generating data driving the same have been presented.

According to the present invention, a target steering angle Φ is calculated by analyzing road image data obtained from an image capturing means installed at a front surface of a driving vehicle, and data for controlling the speed of the vehicle is generated to correct driving deviations. It is an object of the present invention to provide an automatic traveling device and method.

Another object of the present invention is to provide an automatic driving apparatus and method for a vehicle by driving deviation correction in which an acceleration sensor such as a gyro, a magnetic world, a positioning device such as a GPS, an active marker of a road, etc. are not required.

The present invention calculates the steering angle of the vehicle using the distance between the vehicle axes (L) after calculating the traveling direction of the vehicle, the photographing angle (θ) of the image photographing apparatus, and the distance between the preceding vehicle. The object can be achieved by analyzing the distance d to generate data for driving the brakes and the accelerator and automatically controlling the vehicle.

Steering angle data generation device of the vehicle through the preceding vehicle image tracking of the present invention for achieving the above object comprises an image taking means for photographing the road image of the front of the vehicle; Image data memory means (20) for storing the captured image data; Vehicle image selecting means 30 for selecting an image of the vehicle from the image data of the road stored in the image data memory means 20; Vehicle image separating means 31 for extracting the selected vehicle image; Vehicle image photographing angle deviation inter-vehicle distance calculating means 33 for calculating a photographing angle deviation and calculating a distance between the vehicle; Photographing angle deviation storing means (35) for storing the calculated photographing angle deviation information; Photographing angle adjusting means (34) for directly adjusting the photographing angle of the image photographing means by using the photographed photographing angle deviation information; A photographing angle sensor 36 for measuring a photographing angle of the image photographing means; Steering angle calculation means (60) for calculating a steering angle by using the inter-vehicle distance (d) calculated by the photographing angle (θ) of the photographing angle sensor and the vehicle image photographing-angle deviation distance calculation means (33); Steering angle comparison means 61 for comparing the steering angle measured by the steering angle sensor 67 with the steering angle calculated by the steering angle calculation means 60; Steering information generation means 62 for generating steering information using the compared steering angle information; The generated steering information is stored in the steering information storage means. In addition, the state value generating means generates a state value by the state value generating means 50 according to the inter-vehicle distance d, and compares the speed by using the value read from the speed sensor and the state value generated by the state value generating means. Speed comparing means 51; Speed regulating device selecting means 52 for selecting a speed regulating device by using the speed compared in the speed comparing means and the state value generated in the state value generating means; Speed control information generating means (53) for generating and storing speed control information of the selected speed control device; The generated speed control information is stored in the speed control information storage means 54.

The driving unit 70 includes steering information storage means 65 for storing information generated by the steering information generation means 64; Steering drive means 66 to independently adjust the steering 67 by using the steering information stored in the steering information storage means 65; Speed control information storage means for storing the speed control information generated by the speed control information generating means 56; The brake driving means 58 and the accelerator driving means 57 which drive the brake 58a and the accelerator 57a based on the stored speed control information are characterized in that it is configured.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a conceptual diagram of a rotating radius known in the art. Referring to the drawings, the present invention can be expressed by the following equation (1) according to the geometrical conditions in order to derive the steering angle calculation formula according to the characteristics of the steering device of the vehicle.

R = L / sin Φ + r
In Equation 4, R is the minimum radius of rotation, L is the distance between the shaft, sinφ is the outermost steering angle, r is the distance between the center of the wheel stop and the king pin.
Assuming R is large enough than r
R = L / sin Φ
2 is an outer wheel reference steering concept according to the present invention. Referring to the drawing, since the outer wheel passes both points of the circle, it can be expressed by the following equation 2 according to the condition of the function.

delete

delete

delete

x ² + y ² = R ²

Assuming that the vehicle has no slip in Equation 2, since the radius of the outer wheel must pass a point B (R-d * sinθ, d * cosθ) of a constant circle,

(Rd * sinθ) ² + (d * cosθ) ² = R ²

R ² -2 * R * d * sinθ + d ² * sin ² θ + (d * cosθ) ² = R ²

2 * R * d * sinθ = d ²

R = d / (2 * sinθ)

Assuming that there is no slip in the vehicle as shown in Equation 2, the following Equation 3 can be obtained by substituting Equation 2 into Equation 1 to obtain a steering angle based on the outer wheel.

d / (2 * sinθ) = L / sin Φ

sin Φ = 2 * L * sinθ / d

Φ = sin ^-1 (2 * L * sinθ / d)

In the present invention, three condition values should be determined to calculate the steering angle as shown in Equation 3 above.

That is, when the distance L, the photographing angle θ, and the distance d of the vehicle are determined as in Equations 1 to 3, the steering angle of the target point can be calculated. At this time, assuming that there is no inflection point, the steering angle between the two points A and B of FIG. 2 is able to calculate the steering angle during driving based on the distance L, the photographing angle θ, and the distance d of the vehicle.

4 is a conceptual diagram illustrating initialization of an image photographing means according to the present invention. The vehicle is parked in the driving direction of the road in the center of the road where the straight lane is installed. Adjust the steering angle so that the steering angle sensor value is "0". Thereafter, the optical axis of the image photographing means is installed in the center of the vehicle in parallel with the center line of the vehicle, and then the image photographing means is initialized.

Figure 4a shows the correlation according to the installation position when installing the imaging module in the vehicle. If the image capturing module is installed at the left and right intermediate points of the vehicle, the reference point of the preceding vehicle also becomes the center of the vehicle. Otherwise, the equation is changed based on the same position of the preceding vehicle.

FIG. 4B is a diagram illustrating a distance d between the autonomous vehicle and the preceding vehicle, the orthogonal coordinate setting between the autonomous vehicle and the preceding vehicle, and the scanning line number of the lowermost part (a-a ') in which the color of the vehicle is displayed. When the first automatic driving device is set, even if the distance value according to the scanning line number set on the road as shown in FIG. 4A varies the relative shooting angle (θ) between the vehicle and the image photographing means, the condition mounted on the vehicle is left and right rotated by the image photographing means. Naturally, it should be installed so that there is no distance deviation. This is because this is an important distance measurement core technology of the present invention.

As such, the image photographing means is designed and operated to automatically adjust the photographing angle to the horizontal center line of the preceding vehicle. As shown in FIG. 4, when the image capturing means is installed away from the center line by a distance X0, the distance is reflected in the initial value. Initialization of the image capturing means is performed by scanning line number (Pn), the horizontal distance (d) of the actual image shown on the image capturing means and the scanning line, and the distance (m) of the parking line (m) from the center of the optical axis as shown in Table 1 below. Inputting the distance Xs from the lane s and the first scanning line Pm in which the image of the preceding vehicle appears is completed.

Referring to Table 1, the distances Xm and Xs from the center of the optical axis are represented by the number of pixels on the screen. The first scan line Pm is the first scan line number where the vehicle image of the preceding vehicle exists, and d is measured and input the actual distance to the preceding vehicle. This can be interpreted as being at a distance of 250 meters (d) when the first scan line Pm in which the image of the vehicle appears is 183.

In the present invention, as shown in FIG. 4B, the distance d to the preceding vehicle is calculated based on the floor of the vehicle and stored as shown in Table 1 below.

5A is a conceptual view illustrating a photographing angle correction for image tracking, and FIG. 5B is a conceptual view illustrating a photographing angle deviation Δθ according to the present invention.

5A and 5B, the horizontal deviation Δx represents a deviation value in the horizontal direction as a positional deviation between the optical axis of the image photographing means and the preceding vehicle. The image capturing module installed in the vehicle is designed and operated to rotate in accordance with the movement of the vehicle ahead.

Referring to the drawings, FIG. 5B is a diagram showing a relationship between the changed shooting angle θ and the shooting angle deviation Δθ to be changed in the current period. F is the current traveling direction of the autonomous vehicle, the coordinates are set in the front direction of the image photographing module, and S is the direction in which the preceding vehicle is moved relative to the image photographing means.

FIG. 5A is a conceptual diagram for calculating an imaging angle deviation Δθ of the current cycle shown in FIG. 5B. The distance of the preceding vehicle centered on the inter-vehicle distance d of the immediately preceding cycle and the optical axis of the image photographing means may be expressed as a horizontal deviation Δx, and the photographing angle deviation Δθ may be calculated according to Equation 5 below.

In the case where the horizontal center moving distance of the vehicle at the point where the preceding vehicle distance is d from the autonomous vehicle is Xd, and the image photographing means is installed at a distance of X0 from the vehicle center line, the horizontal deviation is converted into the actual distance by the following equation (4) do.

Δx = lane width * Xd / (Xm-Xs)-X0

Xm and Xs are the values read according to the inter-vehicle distance d of the vehicle in Table 1. The method of calculating Xd is calculated by obtaining the transverse center position of the vehicle, and the photographing angle deviation Δθ is calculated by the following equation (5).

As described above, the inter-vehicle distance d is determined based on the first scan line number Pm in which the color value of the preceding vehicle appears, with reference to Table 1.

Δθ = tan ^-1

Assuming that Δx / d is small enough

Δθ = Δx / d.

Δθ is rad and Δx, d is m units.

FIG. 6 is a conceptual diagram of calculating steering angles of an autonomous vehicle by using angle correction of an image photographing module through tracking a preceding vehicle, and automatically driving using a change in the inter-vehicle distance d. A vehicle image selecting means is provided by setting an orthogonal coordinate system on the basis of the automatic driving vehicle A and storing the image data of the preceding vehicle through the image capturing means 10 as digital data in the image data memory means 20. After the image data of the vehicle is separated, the horizontal center is obtained for each scan line, and the reference point is moved by the position deviation X0 where the image photographing means is installed to calculate the distance d and the angle θ for the autonomous vehicle to travel. The steering angle is then calculated.

7 is a diagram illustrating a state in which an image photographing module is mounted on a vehicle. Referring to the drawings, FIG. 7 is a view showing the image capturing module 11 according to the present invention in which the image capturing module installed in the traveling direction of the vehicle is located along the preceding vehicle from side to side in the vehicle front direction. It shows that the shooting angle can follow the shooting angle θ through the image tracking technique.

Hereinafter, an operation method of the present invention will be described with reference to a block diagram showing a vehicle driving apparatus using the preceding vehicle image tracking of FIGS. 8 and 9.

Referring to the drawings, the steering angle data raw growth value of the vehicle through the preceding vehicle image tracking of the present invention image capturing means (10) for photographing the road image of the vehicle; Image data memory means (20) for storing the captured image data; Vehicle image selecting means 30 for selecting an image of the vehicle from the image data of the road stored in the image data memory means 20; Vehicle image separating means 31 for extracting the selected vehicle image; Vehicle image photographing angle deviation inter-vehicle distance calculating means 33 for calculating a photographing angle deviation and calculating a distance between the vehicle; Photographing angle adjusting means (34) for directly adjusting the photographing angle of the image photographing means by using the photographed photographing angle deviation information; Photographing angle deviation storing means (35) for storing the calculated photographing angle deviation information; A photographing angle sensor 36 for measuring a photographing angle of the image photographing means; Steering angle calculation means (60) for calculating a steering angle by using the inter-vehicle distance (d) calculated by the photographing angle (θ) of the photographing angle sensor and the vehicle image photographing-angle deviation distance calculation means (33); Steering angle comparison means 61 for comparing the steering angle measured by the steering angle sensor 67 with the steering angle calculated by the steering angle calculation means 60; Steering information generating means 62 for generating steering information using the compared steering angle information, wherein the generated steering information is stored in the steering information storage means.

In addition, the present invention ecological value generating means 50 for generating a state value according to the inter-vehicle distance (d); Speed comparing means (51) for comparing the speed by using the value read from the speed sensor (51a) and the state value generated by the state value generating means (50); Speed regulating device selecting means 52 for selecting a speed regulating device by using the speed compared in the speed comparing means and the state value generated in the state value generating means; Speed control information generating means (53) for generating and storing speed control information of the selected speed control device; Speed control information storage means 54 for storing the generated speed control information.

In addition, the driving unit indicated by 70 in the figure is a steering information storage means 63 for storing information generated by the steering information generating means 62; Steering driving means 64 which independently adjusts the steering 65 by using the steering information stored in the steering information storage means 63; Speed control information storage means 54 for storing the speed control information generated by the speed control information generating means 53; The brake driving means 57 and the accelerator driving means 55 for driving the brake 58 and the accelerator 56 based on the stored speed control information.

In the figure, reference numeral 56a denotes an accelerator sensor, 56b denotes an accelerator pedal, 58a denotes a brake sensor, and 66 denotes a steering wheel.

According to the present invention configured as described above, the image capturing means 10 is installed in the front direction of the vehicle to capture the image of the road and the preceding vehicle and is stored as digital data in the image data memory means 20. The stored image data is displayed on a monitor installed inside the vehicle, and the vehicle image is selected manually by the vehicle image selecting means 30 at the beginning of the vehicle tracking. The road image and the vehicle image are separated by the vehicle image separating means 31 using the color code of the selected vehicle image.

Using Table 1 created when the image capturing means 10 is initialized using the separated vehicle image, the horizontal deviation Δx is calculated as shown in Equation 4, and the horizontal deviation and the distance d are expressed in Equation 5 Accordingly, the photographing angle deviation Δθ is calculated by the vehicle image photographing angle deviation inter-vehicle distance calculating means 33.

The calculated shooting angle deviation Δθ is stored in the shooting angle deviation information storing means 35. The stored photographing angle deviation changes the photographing angle of the image photographing means 10 by the photographing angle adjusting means 34 by the photographing angle deviation. The photographing angle of the changed image photographing means 10 is measured by the photographing angle sensor and transmitted to the steering angle calculating means 60.

In the present invention, when the shooting angle deviation Δθ calculated as in Equation 5 is stored in the shooting angle deviation information storing means 35, the shooting angle adjusting means 34 changes the shooting angle by the stored angle. In this case, the changed photographing angle is measured by the photographing angle sensor 36 to store the measured value. The method of changing the photographing angle is simply composed of a motor and a driving unit, and the known technique may be used.

On the other hand, in the present invention, the steering angle is calculated by the steering angle calculation means 60 in accordance with Equation 3 using the value of the inter-vehicle distance d and the shooting angle information calculated by the vehicle image photographing angle deviation inter-vehicle distance calculation means 33. The calculated steering angle is compared with the steering angle measured by the steering angle sensor 67 by the steering angle comparing means 61, and the difference value is calculated by the steering information generating means 62 and stored in the steering information storing means 63.

Here, the steering angle calculation unit 60 reads the shooting angle stored in the shooting angle sensor and uses the steering angle calculation. The distance d with the preceding vehicle is determined by using the minimum scan line number in which the image of the vehicle is displayed. It is obtained by comparing with the value inputted at initialization. In this way, the inter-vehicle distance d is determined, and the photographing angle of the image capturing means is obtained from the photographing angle sensor, so that the steering angle can be calculated according to equation (3).

In addition, according to the present invention, the steering drive means 64 of the drive unit 70 drives the steering 65 in accordance with the steering information stored in the steering information storage means (63). In addition, the inter-vehicle distance (d) calculated by the vehicle image photographing angle deviation inter-vehicle distance calculation means 33 is transmitted to the state value generating means 50 and compared with the value of the inter-vehicle distance (d) of the immediately preceding cycle to generate a state value. Create a status value according to

According to the present invention, the speed comparing means 51 compares the vehicle speed by using the vehicle speed read from the speed sensor 51a and the state value generated by the state value generating means. Using the speed compared in the speed comparing means 51 and the state value generated in the state value generating means, the speed adjusting device selecting means 52 selects the speed adjusting device. By generating the speed control information of the selected speed control device, the speed control information generating means 53 stores the generated speed control information in the speed control information storage means 54.

The driving unit 70 stores the information generated by the steering information generating means 62 in the steering information storage means 63 and the steering driving means 64 independently using the steering information stored in the steering information storage means 63. The steering 65 is adjusted.

In addition, the speed control information generated by the speed control information generating means 53 is stored in the speed control information storage means 54 and the brake drive means 57 and the accelerator drive means 55 are based on the stored speed control information. Respectively, the brake 58 and the accelerator 56 can be automatically controlled to adjust the speed.

As described above, the present invention enables unmanned driving by adjusting the steering angle of the vehicle and controlling the speed by analyzing the road image obtained by the image capturing means installed in the front of the driving vehicle.

In addition, the present invention calculates the steering angle of the vehicle using the distance between the vehicle axis (L) after calculating the traveling direction of the vehicle, the shooting angle (θ) of the image recording apparatus and the distance between the preceding vehicle (d), The vehicle can be automatically controlled by analyzing the distance d and generating data for driving the brake and the accelerator.

Claims (4)

Image capturing means (10) for photographing an image of a road ahead of the vehicle; Image data memory means (20) for storing the captured image data; Vehicle image selecting means (30) for selecting an image of the vehicle from the image data of the road stored in the image data memory means (20); Vehicle image separation means 31 for extracting the selected vehicle image; Vehicle image photographing angle deviation inter-vehicle distance calculating means 33 for calculating a photographing angle deviation using the separated vehicle image and calculating a distance between the vehicle; Photographing angle deviation storing means (35) for storing the calculated photographing angle deviation information; Photographing angle adjusting means (34) for directly adjusting the photographing angle of the image photographing means by using the photographed photographing angle deviation information; A photographing angle sensor 36 for measuring a photographing angle of the image photographing means; And Steering angle calculation means (60) for calculating a steering angle by using the inter-vehicle distance (d) calculated by the photographing angle (θ) of the photographing angle sensor and the vehicle image photographing-angle deviation distance calculation means (33); Steering angle data generation device of the vehicle through the preceding vehicle image tracking, comprising a. The method of claim 1, wherein the steering angle calculation means is based on the distance (d) between the straight line (d) between the preceding vehicle and the autonomous vehicle, the traveling direction of the vehicle and the image data area center of the preceding vehicle, and the distance between the axes of the vehicle. The steering angle data generation device of the vehicle through the preceding vehicle image tracking, characterized in that to calculate the target steering angle by the following equation. Φ = sin ^-1 (2 * L * sinθ / d) Photographing a road image in front of the vehicle and storing data through the image photographing means mounted on the vehicle; Selecting and separating an image of the vehicle from the image data of the stored road; Calculating and storing a photographing angle deviation and an inter-vehicle distance using the separated vehicle image; Adjusting the photographing angle of the image photographing means by using the stored photographing angle deviation information; Measuring a photographing angle of the image photographing means; And Calculating a steering angle by using the inter-vehicle distance calculated by the photographing angle of the photographing angle sensor and the vehicle image photographing-angle deviation distance calculating means; Steering angle data generation method of the vehicle through the preceding vehicle image tracking comprising a. The method of claim 3, wherein the calculating of the steering angle comprises: The target steering angle is calculated by the following equation based on the linear distance (d) between the preceding vehicle and the autonomous vehicle, the direction of the vehicle and the angle (θ) formed by the center of the image data area of the preceding vehicle and the distance between the axes of the vehicle. Steering angle data generation method of the vehicle through the preceding vehicle image tracking, characterized in that. Φ = sin ^-1 (2 * L * sinθ / d)

Images