KR102387603B1 - Apparatus and method for generating an obstacle avoidance path in an autonomous vehicle - Google Patents

Abstract

An apparatus and method for generating an obstacle avoidance path in an autonomous driving vehicle are provided so that, when there is an obstacle in front of a running autonomous vehicle, an obstacle avoidance path is generated and provided, thereby safely avoiding the obstacle and driving. The proposed device includes an avoidance route point information generator that generates information on points of different avoidance routes based on different input conditions and stores them in advance in a table form, an obstacle recognition module that recognizes obstacles in front of an autonomous driving vehicle in motion, The monitoring unit that monitors the driving behavior of the autonomous driving vehicle in motion and the driving behavior of other vehicles around the autonomous driving vehicle, the obstacle recognition unit, and the monitoring unit as the entry into the avoidance mode is determined based on information from the monitoring unit and a control unit that reads and outputs information on points of the avoidance route matching the current monitoring result from a table, and an avoidance route generator that generates an actual avoidance route based on information on points of the avoidance route from the control unit.

Description

Apparatus and method for generating an obstacle avoidance path in an autonomous vehicle}

The present invention relates to an apparatus and method for generating an obstacle avoidance path in an autonomous vehicle, and more particularly, to an apparatus and method for generating an avoidance path when an obstacle is found while the autonomous vehicle is driving.

Today's automobile industry is moving towards implementing autonomous driving that minimizes driver intervention in vehicle driving. An autonomous vehicle refers to a vehicle that recognizes the surrounding environment through external information detection and processing functions during driving, determines a driving route by itself, and drives independently using its own power.

Autonomous vehicles can drive themselves to their destinations by preventing collisions with obstacles on the driving path and adjusting vehicle speed and driving direction according to the shape of the road without the driver operating the steering wheel, accelerator pedal, or brake. there is. For example, acceleration may be performed on a straight road, and deceleration may be performed on a curved road while changing a driving direction in response to the curvature of the road.

In order to ensure stable driving of an autonomous vehicle, the driving environment must be accurately measured through each sensor mounted on the vehicle, and the driving state of the vehicle must be continuously monitored and driving must be controlled according to the measured driving environment. In addition, it is important to prevent path deviation or collision with obstacles in advance using communication technology between vehicles or between vehicles and infrastructure.

Conventionally, when an obstacle such as a vehicle or a person located in front of an autonomous driving vehicle is found, braking control is used to avoid collision with the obstacle. In other words, it calculates the risk of collision with an obstacle ahead through radar and camera and compares the current distance with the distance to the obstacle required to avoid the collision, and warns the driver if a sufficient distance is not secured. If the driver does not show an appropriate reaction despite the warning, a method of automatically generating or assisting braking force through the brake control system was used.

However, in general, since the braking distance is proportional to the square of the vehicle speed, there is a problem in that collision avoidance is difficult only by braking control when the vehicle is traveling at a high speed. In addition, evasion through braking control alone has no choice but to simplify the evasion strategy, making it difficult to effectively evade according to the situation.

Prior Art 1: Republic of Korea Patent Registration No. 10-1581286 (Route-following system and method for autonomous driving of driverless vehicle) Prior art 2: Republic of Korea Patent Publication No. 10-2018-0062503 (Autonomous driving control device and method) Prior art 3: Republic of Korea Patent Publication No. 10-2017-0119842 (Autonomous vehicle control apparatus and method)

The present invention has been proposed in order to solve the above-mentioned problems in the prior art, and when there is an obstacle in front of a traveling autonomous vehicle, an avoidance route is created and provided, thereby safely avoiding the obstacle. An object of the present invention is to provide an apparatus and method for generating an obstacle avoidance path.

In order to achieve the above object, an apparatus for generating an obstacle avoidance path in an autonomous driving vehicle according to a preferred embodiment of the present invention generates information on points of different avoidance paths based on different input conditions to form a table. an avoidance route point information generating unit stored in advance; an obstacle recognition unit for recognizing an obstacle in front of the autonomous driving vehicle in motion; a monitoring unit configured to monitor the driving behavior of the autonomous driving vehicle and the driving behavior of other vehicles in the vicinity of the autonomous driving vehicle; A control unit that reads and outputs information on points of an avoidance path matching the current monitoring result from the monitoring unit from the table as the entry into the avoidance mode is determined based on the information from the obstacle recognition unit and the monitoring unit ; and an evasion path generator configured to generate an actual evasion path based on information on points of the evasion path from the control unit.

The avoidance route point information generating unit may be configured to set a driving avoidance point, a target driving avoidance point, and an intermediate point corresponding thereto for each input condition, and an avoidance path connecting the driving avoidance point and the target driving avoidance point via the intermediate point It is possible to generate information on the points constituting

The driving avoidance point may be a point meaning a position of the left or right end in front of the autonomous vehicle, and the target driving avoidance point may be a point meaning a position separated to the left or right by a safe distance from an obstacle, The safety distance may be a distance from the position of the rear left end of the obstacle to the left by a predetermined distance or a distance from the position of the right end of the rear of the obstacle to the right by a predetermined distance.

The midpoint may be set according to a driving environment preset for each input condition, and the driving environment may include at least one of a surrounding situation and a vehicle speed.

The midpoint may exist within a square range with diagonals (0.1, 0.1), (0.9, 0.9).

The avoidance route point information generating unit sets the intermediate point between the driving avoidance point and the target driving avoidance point, and then uses a filter to create a new point on the avoidance route connecting the driving avoidance point and the target driving avoidance point. their positions can be calculated.

The filter is mask4 = {-w, 1/2+w, 1/2+w, -w} or mask6 = {θ, -(1/16+3θ), 9/16+2θ, 9/16+ 2θ, -(1/16+3θ), θ}, where w and θ are variables having a value so that the avoidance path is gently curved, and w is -1/2 ≤ w ≤ 1/ 2, and θ may be -1/2 ≤ θ ≤ 1/2.

The avoidance path point information generating unit may determine a position by applying any one of continuous, mirror, periodic, and zero padding to a point located at a boundary in calculating the positions of the new points to be taken.

The avoidance route point information generating unit may generate information on points of the avoidance route taking a continuous curve of a gentle S-shape.

It may further include; a display unit for displaying the actual avoidance route generated by the avoidance route generator.

Meanwhile, a method for generating an obstacle avoidance path in an autonomous vehicle according to a preferred embodiment of the present invention includes the steps of generating information on points of different avoidance paths based on different input conditions and storing the information in advance in a table form; recognizing an obstacle in front of the autonomous vehicle in motion; monitoring the driving behavior of the autonomous driving vehicle and the driving behavior of other vehicles in the vicinity of the autonomous driving vehicle; As the entry into the avoidance mode is determined based on the information in the recognizing and monitoring steps, information on the points of the avoidance path matching the current monitoring result by the monitoring step is read from the table and output to do; and generating an actual avoidance route based on the output information on points of the avoidance route.

According to the present invention having such a configuration, when there is an obstacle in front of the autonomous vehicle while driving, a safe avoidance route is generated and provided, so that the driver of the autonomous vehicle can easily check the displayed actual avoidance route with the naked eye, The autonomous vehicle may perform obstacle avoidance driving along the displayed avoidance path.

1 is a block diagram of an apparatus for generating an obstacle avoidance path in an autonomous vehicle according to an embodiment of the present invention.
2 to 14 are diagrams for explaining the operation of the avoidance route point information generator shown in FIG. 1 .
15 is a flowchart illustrating a method for generating an obstacle avoidance path in an autonomous vehicle according to an embodiment of the present invention.
16 is a flowchart illustrating a more detailed description of step S10 shown in FIG. 15 .
17 is a flowchart illustrating a more detailed description of step S30 shown in FIG. 15 .

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, 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. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as “comprise” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this 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 art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present 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 the overall understanding, the same reference numerals are used for the same components in the drawings, and duplicate descriptions of the same components are omitted.

1 is a block diagram of an apparatus for generating an obstacle avoidance path in an autonomous driving vehicle according to an embodiment of the present invention, and FIGS. 2 to 14 are diagrams for explaining the operation of the avoidance path point information generating unit shown in FIG. 1 .

An apparatus for generating an obstacle avoidance path in an autonomous vehicle according to an embodiment of the present invention includes an avoidance path point information generating unit 10 , a storage unit 20 , an obstacle recognition unit 30 , a monitoring unit 40 , and a control unit 50 . ), an avoidance path generation unit 60 , and a display unit 70 .

The avoidance route point information generating unit 10 generates information (eg, coordinates) on points forming an avoidance route that can be generated according to different input conditions.

Here, the different input conditions are provided arbitrarily set from the outside, and may be input conditions for the driving environment of the autonomous vehicle (eg, at least one of the surrounding environment and the vehicle speed). For example, the surrounding environment of an autonomous vehicle is an input value indicating that another vehicle is driving or not driving in the lane next to the lane in which the autonomous vehicle is driving, and if another vehicle is driving in the adjacent lane, the corresponding other vehicle It may be an input value of how close the vehicle is to an obstacle or a corresponding autonomous vehicle. Accordingly, when various surrounding environments and/or vehicle speed are input conditions, information (eg, coordinates) on points forming an avoidance path will be changed correspondingly.

The avoidance route point information generating unit 10 stores information on the generated points of each avoidance route into a table (eg, in the form of a lookup table) and stores the information in the storage unit 20 in advance.

A more detailed description of the avoidance route point information generating unit 10 will be described with reference to FIGS. 2 to 14 .

First, referring to FIG. 2 , the avoidance route point information generating unit 10 performs a predetermined safe distance SD with respect to an obstacle 200 in front of the autonomous driving vehicle 100 (another autonomous driving vehicle or a general vehicle) for each input condition. ) and generates information (eg, location information (ie, coordinates)) of points constituting an avoidance path AP connecting the driving avoidance point P1 and the target driving avoidance point P3 .

Here, the safe distance SD means a distance sufficient to safely avoid driving in a lane on the left or right side of the obstacle 200 without colliding with the obstacle 200 . The safety distance SD refers to a distance from the avoidance point P4 of the obstacle 200 to a predetermined distance to the left or right. The avoidance point P4 may be a position of the leftmost end or the rightmost end at the rear of the obstacle 200 . In FIG. 2 , it is assumed that the rightmost end at the rear of the obstacle 200 is the avoidance point P4 .

The driving avoidance point P1 may be a position of the leftmost end or the rightmost end in front of the autonomous vehicle 100 . In FIG. 2 , it is assumed that the leftmost end in front of the autonomous vehicle 100 is the avoidance point P1 .

The target driving avoidance point P3 may refer to a position separated from the obstacle 200 to the left or right by the safety distance SD.

As a result, as shown in FIG. 2 , when the autonomous vehicle 100 avoids the obstacle 200 and intends to travel to the right, the front left end of the autonomous vehicle 100 (ie, the driving avoidance point P1) is part) should not collide with the rightmost end of the rear of the obstacle 200 (that is, the part with the avoidance point P4). Accordingly, the avoidance route point information generating unit 10 sets a safety distance SD of a predetermined distance to the right of the obstacle 200 and an avoidance route ( It generates information of points constituting AP).

Here, in generating information on points constituting the avoidance route AP connecting the driving avoidance point P1 and the target driving avoidance point P3, the avoidance route point information generating unit 10 first performs the same as in FIG. 3 . Similarly, set the position of the midpoint (P2). Here, the position of the midpoint P2 may be determined according to the set and input driving environment (eg, surrounding conditions, vehicle speed, etc.). It is preferable that the midpoint P2 exists within a square range with diagonals (0.1, 0.1) and (0.9, 0.9). The midpoint P2 may be automatically set according to a learning result using artificial intelligence or the like. Of course, if necessary, the intermediate point P2 may be directly set by a person based on his or her experience. Then, the avoidance route point information generating unit 10 creates an avoidance route by continuously interpolating the points between the points. At this time, the midpoint P2 is a point that must be passed through.

In other words, at the very beginning, there are a driving avoidance point P1 and a target driving avoidance point P3, and a midpoint P2 is set between the driving avoidance point P1 and the target driving avoidance point P3 (FIG. 4). reference). Thereafter, the avoidance route point information generating unit 10 is configured to operate between points (eg, between the driving avoidance point P1 and the midpoint P2 , and between the midpoint P2 and the target driving point) based on the location information of these points. A new point is created between the avoidance points (P3). At this time, the avoidance path point information generating unit 10 calculates the position of a new point by using the filter and surrounding points. where the filter is mask4 = {-w, 1/2+w, 1/2+w, -w} or mask6 = {θ, -(1/16+3θ), 9/16+2θ, 9/16+ 2θ, -(1/16+3θ), θ}. w and θ are variables having values for gently bending the avoidance path AP, and may be referred to as gentle variables. In mask4, w has a range that makes it a continuous function, and it is approximately -1/2 ≤ w ≤ 1/2, and in mask6, θ has a range that allows it to be a continuous function, which is approximately -1/2 ≤ θ ≤ 1/2. The avoidance route point information generating unit 10 makes the avoidance route AP form a continuous curve shape by repeatedly operating the filter N times when determining a new point to be taken.

For example, in a state in which there are a driving avoidance point P1 and a target driving avoidance point P3 and a midpoint P2 is set between the driving avoidance point P1 and the target driving avoidance point P3, information on the avoidance route point If the generator 10 is executed once with the filter as mask4, the position value of the point between the point and the point can be obtained by multiplying the filter and the surrounding points. That is, as in FIG. 4 , location information (ie, coordinates) of a new point ((X _2,2 , Y _2,2 ), (X _2,4 , Y _2,4 )) may be calculated. When mask4 = {-w, 1/2+w, 1/2+w, -w} is used, (X _2,1 , Y _2,1 ) = (X _1,1 , Y _{1, 1} ) becomes , (X _2,2 , Y _2,2 ) = -w×(X _1,0 , Y _1,0 )+(1/2+w)×(X _1,1 , Y _1,1 )+(1 /2+w)×(X _1,2 , Y _1,2 )-w×(X _1,3 , Y _{1, 3} ) becomes, (X _2,3 , Y _2,3 ) = (X _{1, 2} , Y _{1, 2} ) becomes (X _2,4 , Y _2,4 ) = -w×(X _1,1 , Y _1,1 )+(1/2+w)×(X _1,2 , Y _1,2 )+(1/2+w)×(X _1,3 , Y _1,3 )-w×(X _1,4 , Y _1,4 ) becomes, (X _2,5 , Y _2,5 ) = (X _1,3 , Y _{1, 3} ). and (X _1,0 , Y _1,0 ) and (X _1,4 , Y _{1, 4} ) may be further, (X _1,0 , Y _1,0 ) and (X _1,4 , Y _{1, 4} ) are coordinates located at the boundary, so (X _1,0 , Y _1,0 ) and (X _1,4 , Y _{1, 4} ) are continuous, You can determine the position by applying one of mirror, periodic, and zero padding. Accordingly, (X _1,0 , Y _1,0 ) = (X _1,1 , Y _{1, 1} ) is assumed to converge, and (X _1,4 , Y _1,4 ) = (X _1,3 , Y _{1, 3} ) is assumed to converge. Preferably, in an embodiment of the present invention, a position is determined by applying continuous to a point located at the boundary.

Here, using mask4, w = 1/16, (X _1,1 , Y _1,1 ) = (0, 0), (X _1,2 , Y _1,2 ) = (0.5, 0.3), ( If X _1,3 , Y _1,3 ) = (1, 1), when the filter is executed once, the avoidance route point information generating unit 10 generates the values of different coordinate values for the avoidance route as shown in FIG. 5 . When the points can be obtained and the filter is executed twice, the avoidance path point information generating unit 10 can obtain points of different coordinate values for the avoidance path as shown in FIG. 6 , and when the filter is executed three times, The avoidance route point information generator 10 may obtain points of different coordinate values for the avoidance route as shown in FIG. 7 , and when the filter is executed four times, the avoidance route point information generator 10 is displayed in FIG. 8 . Points of different coordinate values for the avoidance path can be obtained as shown in FIG. can be saved Of course, if necessary, the avoidance route point information generating unit 10 may repeatedly execute the filter six or more times.

In this way, the avoidance path point information generating unit 10 selects any one of FIGS. 5 to 9, for example, so as to generate information on points of the corresponding coordinate value for the corresponding input condition. Points forming an avoidance path The information on the data is tabulated (eg, in the form of a lookup table) and stored in the storage unit 20 .

This time, the case of using mask6 will be described. In a state where there is a driving avoidance point P1 and a target driving avoidance point P3 and a midpoint P2 is set between the driving avoidance point P1 and the target driving avoidance point P3, the avoidance route point information generating unit 10 ) is executed once with the filter as mask6, the value of the position between the point and the point can be obtained by multiplying the filter and the surrounding points. That is, as in FIG. 4 , location information (ie, coordinates) of a new point ((X _2,2 , Y _2,2 ), (X _2,4 , Y _2,4 )) may be calculated. When using mask6 = {θ, -(1/16+3θ), 9/16+2θ, 9/16+2θ, -(1/16+3θ), θ}, (X _2,1 , Y _{2, 1} ) = (X _1,1 , Y _{1, 1} ), (X _2,2 , Y _2,2 ) = θ×(X _1,-1 , Y _1,-1 )-(1/16+ 3θ)×(X _1,0 , Y _1,0 )+(9/16+2θ)×(X _1,1 , Y _1,1 )+(9/16+2θ)×(X _1,2 , Y _1,2 )-(1/16+3θ)×(X _1,3 , Y _1,3 )+θ×(X _1,4 , Y _{1, 4} ) becomes (X _2,3 , Y _{2, 3} ) = (X _1,2 , Y _{1, 2} ), (X _2,4 , Y _2,4 ) = θ×(X _1,0 , Y _1,0 )-(9/16+3θ) ×(X _1,1 , Y _1,1 )+(9/16+2θ)×(X _1,2 , Y _1,2 )+(9/16+2θ)×(X _1,3 , Y _{1, 3} )-(1/16+3θ)×(X _1,4 , Y _1,4 )+θ×(X _1,5 , Y _{1, 5} ) becomes, (X _2,5 , Y _2,5 ) = (X _1,3 , Y _{1, 3} ). and (X _1,-1 , Y _1,-1 ), (X _1,0 , Y _1,0 ), (X _1,5 , Y _1,5 ), and (X _1,4 , Y _{1 , 4} ) can be more, (X _1,-1 , Y _1,-1 ), (X _1,0 , Y _1,0 ), (X ₁ ) to make a continuous curved avoidance path (AP). _,5 , Y _1,5 ), and (X _1,4 , Y _{1, 4} ) can be positioned by applying one of continuous, mirror, periodic, and zero padding. Accordingly, it is assumed that (X _1,-1 , Y _1,-1 ) = (X _1,0 , Y _1,0 ) = (X _1,1 , Y _{1, 1} ), and (X _{1, 5} , Y _1,5 ) = (X _1,4 , Y _1,4 ) = (X _1,3 , Y _{1, 3} ). Preferably, in an embodiment of the present invention, a position is determined by applying continuous to a point located at the boundary.

Here, using mask6, θ = 0.014, (X _1,1 , Y _1,1 ) = (0, 0), (X _1,2 , Y _1,2 ) = (0.5, 0.3), (X _{1 ,3} , Y _1,3 ) = (1, 1), when the filter is executed once, the avoidance route point information generating unit 10 generates points of different coordinate values for the avoidance route as shown in FIG. 10 . and if the filter is executed twice, the avoidance route point information generating unit 10 can obtain points of different coordinate values for the avoidance route as shown in FIG. 11 , and if the filter is executed three times, the avoidance route point information generation unit 10 The point information generating unit 10 can obtain points of different coordinate values for the avoidance path as shown in FIG. 12 , and when the filter is executed 4 times, the avoiding path point information generating unit 10 is configured as shown in FIG. 13 . Points of different coordinate values for the avoidance route can be obtained, and when the filter is executed 5 times, the avoidance route point information generating unit 10 can obtain points of different coordinate values for the avoidance route as shown in FIG. 14 . there is. Of course, if necessary, the avoidance route point information generating unit 10 may repeatedly execute the filter six or more times.

In this way, the avoidance path point information generating unit 10 selects any one of FIGS. 10 to 14, for example, so as to generate information on points of the corresponding coordinate value for the corresponding input condition. Points forming an avoidance path The information on the data is tabulated (eg, in the form of a lookup table) and stored in the storage unit 20 .

In the manner described above, the avoidance path point information generating unit 10 may generate information (eg, coordinates) on points forming different avoidance paths by receiving different input conditions from the outside.

As a result, the avoidance path point information generating unit 10 uses a subdivision algorithm, but by appropriately applying a value by a filter, a continuous curve of a gentle S-shaped (S) shape is used for the points of the avoidance path. Information can be created in advance.

The storage unit 20 stores a table (eg, in the form of a lookup table) in which information on points of an avoidance path for each different input condition is recorded. Here, the information in the table can be updated any number of times.

In FIG. 1 described above, although the avoidance route point information generating unit 10 and the storage unit 20 are separate components, even if the storage unit 20 is included in the avoiding route point information generating unit 10 . free

The obstacle recognizing unit 30 recognizes an obstacle existing in front of the autonomous driving vehicle in motion.

For example, the obstacle recognition unit 30 may include one or more of a camera, a radar unit, a LIDAR unit, and the like. The camera may capture an image of an external environment surrounding the autonomous vehicle. The camera may be a still image camera and/or a video camera. For example, the camera may be mounted on a rotating and/or tilting platform to be mechanically movable. The radar unit may utilize radio signals to detect obstacles in the autonomous vehicle's local environment. The lidar unit can use the laser to detect obstacles in front of and around the autonomous vehicle in motion.

The monitoring unit 40 may monitor the driving behavior of the autonomous driving vehicle being driven. For example, the monitoring unit 40 monitors the lateral displacement, lateral speed, vehicle speed, steering angle, steering torque, radius of curvature, etc. according to the actual driving of the autonomous driving vehicle, thereby confirming the actual driving behavior of the currently driving autonomous vehicle. can

The monitoring unit 40 may monitor the distance between the currently driving autonomous driving vehicle 100 and the obstacle 200 in front and the relative speed of the obstacle 200 in front through the driving behavior monitoring of the autonomous driving vehicle in motion. .

In addition, the monitoring unit 40 may monitor the environment around the currently driving autonomous driving vehicle and the driving behavior of other vehicles around the currently driving autonomous driving vehicle by using a camera module, a lidar module, or the like. For example, the monitoring unit 40 may monitor whether another vehicle driving in a lane next to the lane in which the corresponding autonomous vehicle is traveling exists or not, and if there is another vehicle driving in the next lane, corresponding It is possible to monitor how close another vehicle is to the obstacle 200 or the corresponding autonomous vehicle 100 .

The monitoring unit 40 may monitor the relative speed of another vehicle driving in the next lane through the monitoring of the driving behavior of other vehicles around the currently driving autonomous vehicle.

The control unit 50 determines whether to enter the avoidance mode based on the information from the obstacle recognition unit 30 and the monitoring unit 40 . For example, the distance between the currently running autonomous driving vehicle 100 and the obstacle 200 in front is within a predetermined distance, and the current driving speed of the autonomous driving vehicle 100 is higher than the current driving speed of the obstacle 200 . In the early case, the controller 50 may determine to enter the avoidance mode. Of course, when determining the entry into the avoidance mode, the controller 50 may determine which lane of the left lane and the right lane of the obstacle 200 should be avoided.

The control unit 50 reads (reads out) information on points of the avoidance path matching the current monitoring result from the monitoring unit 40 from the table of the storage unit 20 as it determines the entry into the avoidance mode, While transmitting the read information on the points of the avoidance route to the avoidance route generator 60 , it commands generation of the avoidance route.

The avoidance route generation unit 60 generates an actual avoidance route based on information on points of the avoidance route from the controller 50 based on the avoidance route generation command from the controller 50 . For example, the avoidance path generator 60 may generate an actual avoidance route by scaling information of points of the avoidance route from the controller 50 .

The display unit 70 displays the actual avoidance route generated by the avoidance route generator 60 on the screen.

Here, the display unit 70 is installed on the dashboard in front of the driver's seat, and is preferably installed at a position where the driver can easily check the actual avoidance route displayed on the display unit 70 with the naked eye.

15 is a flowchart for explaining a method for generating an obstacle avoidance path in an autonomous vehicle according to an embodiment of the present invention, and FIG. 16 is a flowchart showing a more detailed description of step S10 shown in FIG. 15 , FIG. 17 is a flowchart showing a more detailed description of step S30 shown in FIG. 15 .

First, the avoidance path point information generating unit 10 generates information (eg, coordinates) on points forming an avoidance path that can be generated for each different input condition ( S10 ). In other words, as shown in FIG. 16 , the avoidance route point information generating unit 10 sets a predetermined safety distance SD with respect to the obstacle 200 in front of the autonomous vehicle 100 , and sets the driving avoidance point P1 and the driving avoidance point P1 . The position of the target driving avoidance point P3 and the intermediate point P2 between the driving avoidance point P1 and the target driving avoidance point P3 is determined (S11). Here, the method for determining the driving avoidance point P1 , the intermediate point P3 , and the target driving avoidance point P3 has already been described. Then, in a state in which there are a driving avoidance point P1 and a target driving avoidance point P3 and a midpoint P2 is set between the driving avoidance point P1 and the target driving avoidance point P3, the avoidance route point information generating unit In (10), the position value of the point between the point and the point is obtained by multiplying the filter and the surrounding points (S12). The method of multiplying the filter and the surrounding points to obtain the position value of the point between the point and the point has already been described. When the operation of multiplying the filter by the surrounding points is executed up to N preset times (“Yes” in S13), the operation of step S10 is terminated.

Thereafter, the avoidance route point information generating unit 10 stores information on the points of each avoidance route generated in this way into a table (eg, in the form of a lookup table) and stores the information in advance in the storage unit 20 ( S20 ).

Then, as the control unit 50 determines to enter the avoidance mode based on information from the obstacle recognition unit 30 and the monitoring unit 40 , the avoidance path generation unit 60 matches the current monitoring result. An actual avoidance route is generated based on the points of the avoidance route and displayed on the display unit 70 (S30). In other words, as shown in FIG. 17 , in the state in which the autonomous driving vehicle is driving ( S31 ), the obstacle recognition unit 30 recognizes an obstacle existing in front of the driving autonomous vehicle, and the monitoring unit 40 recognizes the obstacle while driving. The driving behavior of the autonomous driving vehicle, the driving behavior and surrounding environment of other vehicles existing in the vicinity of the autonomous driving vehicle, and the distance between the currently driving autonomous driving vehicle 100 and the obstacle 200 in front and the obstacle in front ( 200) and monitor the relative speed (S32). Next, the control unit 50 determines whether to enter the avoidance mode based on the information from the obstacle recognition unit 30 and the monitoring unit 40 . As it is decided to enter the avoidance mode (“Yes” in S33 ), the control unit 50 stores information on points of the avoidance path matching the current monitoring result from the monitoring unit 40 to the table of the storage unit 20 . read (read out) (S34). Then, the control unit 50 commands the generation of the avoidance route while transmitting the read information on the points of the avoidance route to the avoidance route generator 60 . Thereafter, the avoidance path generation unit 60 generates an actual avoidance path by scaling information on points of the avoidance path from the control unit 50 based on the avoidance path generation command ( S35 ). Then, the display unit 70 displays the actual avoidance route generated by the avoidance route generator 60 (S36).

Accordingly, the driver of the autonomous vehicle 100 can easily check the actual avoidance route displayed on the display unit 70 with the naked eye, and the autonomous vehicle 100 will perform obstacle avoidance driving along the displayed avoidance route.

Although the description of the present invention described above has been limited to an autonomous vehicle, it is considered that it can be sufficiently applied to a general vehicle as it is.

In addition, the method for generating an obstacle avoidance path in the autonomous vehicle of 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 in which data readable by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. In addition, the computer-readable recording medium is distributed in a computer system connected through a network, so that the computer-readable code can be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the method can be easily inferred by programmers in the art to which the present invention pertains.

As described above, the best embodiment has been disclosed in the drawings and the specification. Although specific terms have been used herein, they are only used for the purpose of describing the present invention and are not used to limit the meaning or scope of the present invention described in the claims. Therefore, it will be understood by those skilled in the art that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical protection scope of the present invention should be defined by the technical spirit of the appended claims.

10: avoidance route point information generation unit 20: storage unit
30: obstacle recognition unit 40: monitoring unit
50: control unit 60: avoid path generation unit
70: display

Claims (20)

an avoidance route point information generating unit that generates information on points of different avoidance routes based on different input conditions and stores the information in advance in a table format;
an obstacle recognition unit for recognizing an obstacle in front of the autonomous driving vehicle in motion;
a monitoring unit configured to monitor the driving behavior of the autonomous driving vehicle and the driving behavior of other vehicles in the vicinity of the autonomous driving vehicle;
A control unit that reads and outputs information on points of an avoidance path matching the current monitoring result from the monitoring unit from the table as the entry into the avoidance mode is determined based on the information from the obstacle recognition unit and the monitoring unit ; and
an avoidance path generator for generating an actual avoidance route based on information on points of the avoidance route from the control unit;
The avoidance route point information generating unit,
Setting a driving avoidance point, a target driving avoidance point, and an intermediate point corresponding to the input conditions for each input condition, and generating information on points constituting an avoidance path connecting the driving avoidance point and the target driving avoidance point via the intermediate point do,
After setting the midpoint between the driving avoidance point and the target driving avoidance point, using a filter to calculate the positions of new points to be taken on the avoidance path connecting the driving avoidance point and the target driving avoidance point;
An apparatus for generating an obstacle avoidance path in an autonomous vehicle, characterized in that in calculating the positions of the newly taken points, the positions are determined by applying any one of continuous, mirror, periodic, and zero padding to the points located at the boundary. . delete The method according to claim 1,
The driving avoidance point is a point meaning the position of the left end or the right end in front of the autonomous vehicle,
The target driving avoidance point is a point that means a position separated to the left or right by a safe distance from the obstacle,
The safety distance is a distance from the position of the rear left end of the obstacle to the left by a predetermined distance or the distance from the position of the right end of the rear of the obstacle to the right by a predetermined distance. Obstacle avoidance path generation device in The method according to claim 1,
The midpoint is
The apparatus for generating an obstacle avoidance path in an autonomous driving vehicle, wherein the driving environment is set according to a driving environment preset for each input condition, wherein the driving environment includes at least one of a surrounding situation and a vehicle speed. 5. The method according to claim 4,
The midpoint is
An obstacle avoidance path generating device in an autonomous vehicle, characterized in that it exists within a square range having diagonals of (0.1, 0.1) and (0.9, 0.9). delete an avoidance route point information generating unit that generates information on points of different avoidance routes based on different input conditions and stores the information in advance in a table format;
an obstacle recognition unit for recognizing an obstacle in front of the autonomous driving vehicle in motion;
a monitoring unit configured to monitor the driving behavior of the autonomous driving vehicle and the driving behavior of other vehicles in the vicinity of the autonomous driving vehicle;
A control unit that reads and outputs information on points of an avoidance path matching the current monitoring result from the monitoring unit from the table as the entry into the avoidance mode is determined based on the information from the obstacle recognition unit and the monitoring unit ; and
an avoidance path generator for generating an actual avoidance route based on information on points of the avoidance route from the control unit;
The avoidance route point information generating unit,
Setting a driving avoidance point, a target driving avoidance point, and an intermediate point corresponding to the input conditions for each input condition, and generating information on points constituting an avoidance path connecting the driving avoidance point and the target driving avoidance point via the intermediate point do,
After setting the midpoint between the driving avoidance point and the target driving avoidance point, using a filter to calculate the positions of new points to be taken on the avoidance path connecting the driving avoidance point and the target driving avoidance point;
The filter is
mask4 = {-w, 1/2+w, 1/2+w, -w} or mask6 = {θ, -(1/16+3θ), 9/16+2θ, 9/16+2θ, -( 1/16+3θ), θ},
Wherein w and θ are variables having a value for allowing the avoidance path to be gently curved, wherein w is -1/2 ≤ w ≤ 1/2, and θ is -1/2 ≤ θ ≤ 1/2 An apparatus for generating an obstacle avoidance path in an autonomous vehicle, characterized in that delete The method according to claim 1,
The avoidance route point information generating unit,
An apparatus for generating an obstacle avoidance path in an autonomous vehicle, characterized in that it generates information on points of an avoidance path taking a continuous curve of a gentle S-shape. The method according to claim 1,
and a display unit configured to display the actual avoidance route generated by the avoidance route generator. generating information on points of different avoidance paths based on different input conditions and storing the information in advance in the form of a table;
recognizing an obstacle in front of the autonomous vehicle in motion;
monitoring the driving behavior of the autonomous driving vehicle and the driving behavior of other vehicles in the vicinity of the autonomous driving vehicle;
As the entry into the avoidance mode is determined based on the information in the recognizing step and the monitoring step, information on the points of the avoidance path matching the current monitoring result by the monitoring step is read from the table and output to do; and
generating an actual avoidance route based on the output information on points of the avoidance route;
The step of pre-saving is,
setting a driving avoidance point, a target driving avoidance point, and an intermediate point corresponding to each of the input conditions; and
generating information on points constituting an avoidance path connecting the driving avoidance point and the target driving avoidance point via the intermediate point;
The step of generating information on points constituting an avoidance path connecting the driving avoidance point and the target driving avoidance point includes:
Calculating the positions of points to be newly taken on an avoidance path connecting the driving avoidance point and the target driving avoidance point using a filter,
The step of calculating the positions of the points to be newly taken is,
A method for generating an obstacle avoidance path in an autonomous vehicle, characterized in that the position is determined by applying any one of continuous, mirror, periodic, and zero padding to a point located at the boundary. delete 12. The method of claim 11,
The driving avoidance point is a point meaning the position of the left end or the right end in front of the autonomous vehicle,
The target driving avoidance point is a point that means a position separated to the left or right by a safe distance from the obstacle,
The safety distance is a distance from the position of the rear left end of the obstacle to the left by a predetermined distance or the distance from the position of the right end of the rear of the obstacle to the right by a predetermined distance. How to create an obstacle avoidance path in 12. The method of claim 11,
The midpoint is
The method of generating an obstacle avoidance path in an autonomous driving vehicle, wherein the driving environment is set according to a driving environment preset for each input condition, wherein the driving environment includes at least one of a surrounding situation and a vehicle speed. 15. The method of claim 14,
The midpoint is
A method for generating an obstacle avoidance path in an autonomous vehicle, characterized in that it exists within a square range having diagonals of (0.1, 0.1) and (0.9, 0.9). delete generating information on points of different avoidance paths based on different input conditions and storing the information in advance in the form of a table;
recognizing an obstacle in front of the autonomous vehicle in motion;
monitoring the driving behavior of the autonomous driving vehicle and the driving behavior of other vehicles in the vicinity of the autonomous driving vehicle;
As the entry into the avoidance mode is determined based on the information in the recognizing step and the monitoring step, information on the points of the avoidance path matching the current monitoring result by the monitoring step is read from the table and output to do; and
generating an actual avoidance route based on the output information on points of the avoidance route;
The step of pre-saving is,
setting a driving avoidance point, a target driving avoidance point, and an intermediate point corresponding to each of the input conditions; and
generating information on points constituting an avoidance path connecting the driving avoidance point and the target driving avoidance point via the intermediate point;
The step of generating information on points constituting an avoidance path connecting the driving avoidance point and the target driving avoidance point includes:
Calculating the positions of points to be newly taken on an avoidance path connecting the driving avoidance point and the target driving avoidance point using a filter,
The filter is
mask4 = {-w, 1/2+w, 1/2+w, -w} or mask6 = {θ, -(1/16+3θ), 9/16+2θ, 9/16+2θ, -( 1/16+3θ), θ},
Wherein w and θ are variables having a value for allowing the avoidance path to be gently curved, wherein w is -1/2 ≤ w ≤ 1/2, and θ is -1/2 ≤ θ ≤ 1/2 A method for generating an obstacle avoidance path in an autonomous vehicle, characterized in that delete 12. The method of claim 11,
The information about the points of the different avoidance paths in the pre-storing step is,
A method for generating an obstacle avoidance path in an autonomous vehicle, characterized in that it takes a continuous curve of a gentle S-shape. 12. The method of claim 11,
and displaying the actual avoidance route.

Images