KR20200128470A - Collision Prevention Program for Autonomous Vehicles - Google Patents

Abstract

Disclosed is a program for preventing a collision for an autonomous vehicle, capable of preventing collision by issuing an alarm when there is a risk of collision based on the probability of a collision with a counterpart vehicle while the autonomous vehicle is underway. The program of the present invention comprises the functions of: predicting, by a counterpart vehicle location predicting unit, the driving route of a counterpart vehicle, and predicting the subsequent location followed by the current location of the counterpart vehicle on the predicted driving route; recognizing, by a user vehicle location recognizing unit, a current location and a subsequent location of an autonomously driving user vehicle; calculating, by a relative distance calculating unit, the distance to subsequent locations between the counterpart vehicle and the user vehicle; calculating, by a control unit, a collision probability between the user vehicle and the counterpart vehicle by reflecting the distance of the subsequent locations between the counterpart vehicle and the user vehicle and one or more consideration factors to a pre-stored collision probability function; and issuing, by the control unit, a collision alarm through an alarm unit on the basis of the calculated collision probability.

Description

Collision Prevention Program for Autonomous Vehicles

The present invention relates to a collision prevention program for an autonomous vehicle, and more particularly, to a collision prevention program for an autonomous vehicle that enables a warning due to the risk of collision with an opponent vehicle while the autonomous vehicle is driving. will be.

As industry advances and users use vehicles more frequently, modern people's vehicle use time has become longer, and automobiles have become a living space, and research on vehicles that can provide more comfortable and convenient vehicles is continuously being conducted.

And, recently, research on an autonomous vehicle that allows the vehicle to drive itself without the driver's direct driving has been actively conducted. Since autonomous vehicles must travel without driver control, they recognize the surrounding situation and drive through sensors such as cameras and radars. And, if an obstacle exists in the recognized situation, it determines the risk of collision with the obstacle and performs an operation for collision avoidance.

Conventionally, in order to avoid most stationary obstacles, spatial information (coordinates, heading angles, curvatures, etc.) of the driving route is expressed as a polynomial for the moving distance, and it is determined whether or not there is a collision with the obstacle.

However, the autonomous vehicle according to the prior art cannot accurately determine whether or not a dynamic obstacle is collided with a dynamic obstacle exists in the driving path.

Prior Art 1: Korean Patent Laid-Open No. 10-2013-0056655 (Method for collision avoidance of unmanned autonomous vehicles) Prior Art 2: Korean Patent Registration No. 10-1049906 (Autonomous moving device and collision avoidance method thereof) Prior Art 3: Korean Patent Registration No. 10-1664582 (device and method for generating driving paths for autonomous vehicles)

The present invention has been proposed to solve the above-described conventional problem, and is for an autonomous vehicle that warns of a collision risk based on a collision probability with an opponent vehicle while driving of an autonomous vehicle to prevent a collision in advance. Its purpose is to provide a collision prevention program.

In order to achieve the above object, in the collision prevention program for an autonomous vehicle according to a preferred embodiment of the present invention, the opponent vehicle position prediction unit predicts the driving route of the opponent vehicle and determines the current position of the opponent vehicle in the predicted driving route. The function of predicting the subsequent position involved in; A function of the own vehicle location determining unit to determine a current location and a subsequent location of the self-driving vehicle; The relative distance calculation unit includes a function of calculating a distance between the opponent vehicle and the host vehicle at a subsequent position; A function of the control unit calculating a collision probability between the host vehicle and the opponent vehicle by reflecting a distance at a subsequent position between the opponent vehicle and the host vehicle and one or more factors to be considered in a previously stored collision probability function; And a function of the control unit performing a collision warning through a warning unit based on the calculated collision probability.

According to the present invention having such a configuration, a collision can be prevented in advance by calculating a collision probability according to the distance between the host vehicle and the opponent vehicle while the autonomous vehicle is driving, and warning when the collision risk is high.

1 is a block diagram of a collision prevention apparatus for an autonomous vehicle according to an embodiment of the present invention.
2 and 3 are graphs used to describe the control unit illustrated in FIG. 1.
4 is a flowchart illustrating a collision prevention method for an autonomous vehicle according to an embodiment of the present invention.

In the present invention, various modifications may be made and various embodiments may be provided, and specific embodiments will be illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to a specific embodiment, it is to be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention.

The terms used in the present application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present application, terms such as "comprise" or "have" are intended to designate the presence of features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, but one or more other features. It is to be understood that the presence or addition of elements or numbers, steps, actions, components, parts, or combinations thereof, does not preclude in advance.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in this application. Does not.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. In describing the present invention, in order to facilitate an overall understanding, the same reference numerals are used for the same elements in the drawings, and duplicate descriptions for the same elements are omitted.

1 is a block diagram of an apparatus for preventing a collision for an autonomous vehicle according to an embodiment of the present invention, and FIGS. 2 and 3 are graphs used to describe the control unit illustrated in FIG. 1.

A collision prevention apparatus for an autonomous vehicle according to an embodiment of the present invention includes an opponent vehicle position prediction unit 10, an own vehicle position determination unit 12, a relative distance calculation unit 14, a vehicle speed detection unit 16, and wake up It may include an information collection unit 18, a road condition grasping unit 20, a control unit 22, and a warning unit 24.

The opponent vehicle position prediction unit 10 predicts the position of the opponent vehicle in an autonomous driving situation.

To this end, the relative vehicle position predictor 10 may include a stereo camera.

That is, the relative vehicle position prediction unit 10 photographs a relative vehicle around the host vehicle in real time with a stereo camera, identifies the location of the opponent vehicle in the current frame and one or more previous frames among the photographed frames, and Based on the positions, the driving route of the opponent vehicle is predicted, and the subsequent position of the opponent vehicle is predicted in the predicted driving route.

For example, the relative vehicle position predictor 10 may predict the driving route of the opponent vehicle using a polynomial of 3rd or less. Since the predicted driving route may be converted into map data, the opponent vehicle position predictor 10 may predict a subsequent position of the opponent vehicle for each frame on the driving path of the opponent vehicle. Here, predicting the driving path of the opponent vehicle by using a polynomial of the third order or lower can be sufficiently understood by a person engaged in the same industry as a well-known technology, and thus further description will be omitted.

Of course, in order to predict the position of the opposite vehicle, detection of the opposite vehicle must be preceded, so the opposite vehicle position prediction unit 10 may additionally include a radar sensor or a LiDAR (Light Detection And Range) sensor. have.

If necessary, the relative vehicle position predictor 10 may predict the position of the counterpart vehicle frame by frame through deep learning. That is, if learning is performed based on data of previous frames, it is possible to predict whether the other vehicle will turn left, turn right, or go straight. Then, the position of the opponent vehicle can be predicted for each frame using the predicted path.

The own vehicle location identification unit 12 identifies the location of the self-driving vehicle (self-driving vehicle) on a driving path. The own vehicle location finding unit 12 may include a Global Navigation Satellite System (GNSS) sensor to obtain high-precision location information.

In other words, the own vehicle position grasping unit 12 may grasp the current position and the subsequent position of the own vehicle under autonomous driving based on a preset driving route of the own vehicle before autonomous driving. Here, it is assumed that the host vehicle position determining unit 12 grasps the current position and the subsequent position of the host vehicle in the same frame unit as in the relative vehicle position predicting unit 10.

For example, the occupant of the own vehicle sets a departure point and a destination through navigation before autonomous driving, and starts autonomous driving after setting a driving route based on this in advance. Accordingly, since the own vehicle is autonomously driving along the set driving route, the own vehicle location determining unit 12 can not only grasp the current position of the self-driving self-driving vehicle along the driving route, but also determine the subsequent position.

The relative distance calculation unit 14 calculates a distance between the opponent vehicle and the host vehicle at a subsequent position.

For example, the relative distance calculation unit 14 may determine the relative vehicle and the host vehicle based on the subsequent position of the opponent vehicle predicted by the relative vehicle position prediction unit 10 and the subsequent position of the host vehicle identified by the host vehicle position determination unit 12. The distance from the subsequent location of the liver can be calculated. Here, information on the subsequent position of the opponent vehicle and the subsequent position of the host vehicle is provided in real time.

The vehicle speed detection unit 16 detects the vehicle speed of the self-driving vehicle.

The meteorological information collection unit 18 receives weather information from an external server (not shown) that provides various types of weather information.

The road condition grasping unit 20 detects the condition of the autonomously driving road.

For example, the road condition grasping unit 20 may determine whether the currently driving road is slippery. The road condition detection unit 20 may determine that the currently autonomous driving road is slippery when the number of revolutions of one or more of the four wheels of the host vehicle is significantly different from the number of revolutions of other wheels.

The control unit 22 reflects the distance at a subsequent position between the relative vehicle and the host vehicle from the relative distance calculation unit 14 and one or more factors (ie, vehicle speed, weather information, and road conditions) to a previously stored collision probability function. Calculate the probability of collision between and the opponent vehicle.

For example, the collision probability function may be a normal distribution function such as the following equation.

In the above equation, parameter r is the distance between the relative vehicle and the host vehicle at a subsequent position, parameter σ is the variance value, parameter π is 3.14, parameter e is a natural constant, and parameter x is an integral variable.

The control unit 22 may adjust the variance value σ of the collision probability function (eg, a normal distribution function) to be directly proportional to the vehicle speed, weather information, and road conditions. For example, as the running speed of the host vehicle increases, the dispersion value σ may increase. In addition, even when it snows, rains, or fogs, the variance value σ can be increased. In addition, even when the autonomous driving road is slippery, the dispersion value σ can be increased.

Referring to the drawings, FIGS. 2 and 3 are graphs of a normal distribution function according to distance. In FIG. 2, a variance value (σ) is adjusted to "1" (low speed), and in FIG. 3, a variance value (σ) Is set to "3" (high speed). In this way, the variance value (σ) of the collision probability function can be adjusted according to the vehicle speed. As a result, if the vehicle speed is high, snowy, and fog is present, the variance value σ of the collision probability function will be larger than "3" illustrated in FIG. 3.

On the other hand, when there are multiple opponent vehicles, the control unit 22 calculates a collision probability for each opponent vehicle, and sets the largest value among the calculated collision probabilities as the collision probability.

In the above description, it is illustrated that the control unit 22 uses a normal distribution function as a collision probability function, but if necessary, an Erlang Distribution function, a Potential function, and a Probability density function. It can be replaced with any one of them.

On the other hand, the control unit 22 considers the risk of collision as high if the collision probability (e.g., a value between 0 and 1) calculated through the collision probability function is higher than a preset threshold (e.g., about 0.7) and controls a warning to be issued. do.

The warning unit 24 issues a collision warning under the control of the control unit 22.

The warning unit 24 may output a predetermined warning sound or output a warning sound and warning text. For example, the warning unit 24 may warn approximately 5 to 10 seconds before a collision.

In the above description, when the collision risk is high, a collision warning is performed.If necessary, a number of collision risk classes and collision warning types for each class are previously determined, and the corresponding collision according to the value of the collision probability calculated through the collision probability function. You may want to give them a warning.

4 is a flowchart illustrating a collision prevention method for an autonomous vehicle according to an embodiment of the present invention.

First, the opponent vehicle position prediction unit 10 predicts the position of the opponent vehicle in an autonomous driving situation (S10). That is, the relative vehicle position prediction unit 10 photographs a relative vehicle around the host vehicle with a stereo camera, identifies the location of the opponent vehicle in the current frame and one or more previous frames among the photographed frames, and determines the locations of the identified vehicle. On the basis of this, it is possible to predict the driving route of the opponent vehicle and predict the subsequent position of the opponent vehicle in the predicted driving route.

In addition, the own vehicle position grasping unit 12 determines the current position and the subsequent position of the own vehicle, which is an autonomous vehicle, based on a preset driving path of the own vehicle before autonomous driving (S20).

The relative distance calculation unit 14 is based on the subsequent position of the relative vehicle from the relative vehicle position prediction unit 10 provided in real time and the subsequent position of the host vehicle from the host vehicle position determination unit 12. The distance from the location is calculated (S30).

Thereafter, the control unit 22 calculates a collision probability between the host vehicle and the opponent vehicle by reflecting various factors (vehicle speed, weather information, road conditions) to the collision probability function (S40). Here, the vehicle speed is provided from the vehicle speed detection unit 16, the weather information is provided from the meteorological information collection unit 18, and the road condition is provided from the road condition detection unit 20.

If the impulse probability calculated through the collision probability function (eg, a value between 0 and 1) is equal to or greater than a preset threshold (eg, about 0.7) (“Yes” in S50), the collision risk is considered high. A collision warning is performed through the warning unit 24 (S60).

In addition, the collision prevention method for an autonomous vehicle according to the present invention described above can be implemented as computer-readable codes on a computer-readable recording medium. The computer-readable recording medium includes all types of recording devices that store data that can be read by a computer system. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tapes, floppy disks, and optical data storage devices. In addition, the computer-readable recording medium is distributed over a computer system connected through a network, so that computer-readable codes can be stored and executed in a distributed manner. In addition, functional programs, codes and code segments for implementing the method can be easily deduced by programmers in the art to which the present invention belongs.

As described above, an optimal embodiment has been disclosed in the drawings and specifications. Although specific terms have been used herein, these are only used for the purpose of describing the present invention, and are not used to limit the meaning or the scope of the present invention described in the claims. Therefore, those of ordinary skill in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical scope of the present invention should be determined by the technical spirit of the appended claims.

10: relative vehicle position prediction unit 12: host vehicle position detection unit
14: relative distance calculation unit 16: vehicle speed detection unit
18: weather information collection unit 20: road condition identification unit
22: control unit 24: warning unit

Claims (4)

A function for predicting a driving route of the opponent vehicle and predicting a subsequent position accompanying the current position of the opponent vehicle in the predicted driving route;
A function of the own vehicle location determining unit to determine a current position and a subsequent position of the self-driving vehicle;
The relative distance calculation unit includes a function of calculating a distance between the opponent vehicle and the host vehicle at a subsequent position;
A function of the control unit calculating a collision probability between the host vehicle and the opponent vehicle by reflecting a distance at a subsequent position between the opponent vehicle and the host vehicle and one or more factors to be considered in a previously stored collision probability function; And
Including; the control unit, a function of performing a collision warning through the warning unit based on the calculated collision probability;
The collision probability function is a collision prevention program for an autonomous vehicle, wherein the collision probability function is a normal distribution function having a variance value. The method according to claim 1,
The factor of consideration is a collision prevention program for an autonomous vehicle, characterized in that it includes at least one of vehicle speed, weather information, and road conditions. The method according to claim 2,
The function of calculating the collision probability,
A collision prevention program for an autonomous vehicle, characterized in that adjusting the variance value in direct proportion to vehicle speed, weather information, and road conditions. The method according to claim 1,
The function of performing the collision warning,
A collision prevention program for an autonomous vehicle, characterized in that, if the calculated collision probability is higher than a preset threshold, the collision warning is regarded as high in collision risk.

Images