KR20170070488A - Method and apparatus for automatically generating drive route - Google Patents

Abstract

According to an aspect of the present invention, there is provided an apparatus for automatically generating a traveling path according to an aspect of the present invention, including at least one processor, the processor including: A unit motion generating unit for generating a unitary motion consisting of a control point and a Bezier curve expressed by four vectors between a start point and an end point; A lattice graph converting unit dividing the traveling path into a travelable area and a non-traveling area and displaying the lane in a grid form; A path search unit searching for a shortest path of the traveling path indicated in the grid form; And an optimization unit for optimizing the shortest path by a gradient-based optimization method.

Description

TECHNICAL FIELD [0001] The present invention relates to a method and an apparatus for automatically generating a traveling route,

BACKGROUND OF THE INVENTION 1. Field of the Invention [0002] The present invention relates to an unmanned transportation system, and more particularly, to a method and apparatus for automatically generating a traveling route for an unmanned vehicle or an unmanned transportation system to move to a target point.

In general, the automatic traveling system has been applied to a vehicle that travels on a constant route without obstacles such as a ship or an aircraft. However, since the development of a computer system has enabled real-time calculation of a traveling route and automobile control technology has developed, Studies on an unmanned automatic traveling system have become active.

This unmanned automatic driving method is notable for automobile manufacturers as well as companies other than automakers such as Google. This is because, unlike a ship or an airplane, the driving environment of a vehicle runs along with other vehicles on a restricted road And it is in such a complicated situation that it is necessary to judge the driving condition in real time and optimize the traveling route such as a traffic lane or a lane change due to an unexpected obstacle.

Conventional technologies use a camera or a radar to determine obstacles such as roads, lanes, traffic lights, and pedestrians, to travel along roads or to avoid obstacles.

Korean Patent No. 10-1526606 discloses a technique for acquiring map information from outside and generating a center line of the road from the outside to drive the road along a center line generated from the map information . Also, a technique of avoiding an obstacle by regenerating a virtual curve to avoid the obstacle when the obstacle enters the generated center line is also disclosed.

However, since the automatic travel route generation method relies on map information acquired from the outside and follows the lane, there is a problem that it is difficult to generate a route for automatic travel when there is no lane such as a road under construction or an unpaved road.

SUMMARY OF THE INVENTION The present invention has been made in view of the technical background as described above, and it is an object of the present invention to provide a method and apparatus for efficiently calculating a path by using a Bezier curve-based motion vector, and reducing a variable by merging control points representing Bezier curves And it is an object of the present invention to provide an automatic traveling route generation method and apparatus which can reduce a calculation amount of a route calculation for traveling and generate a natural traveling route.

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided an apparatus for automatically generating a traveling path, including at least one processor, wherein the processor includes: a control point that forms a starting point and an end point of a traveling path; A unit motion generating unit for generating a unit motion composed of a Bezier curve expressed by a vector; A lattice graph converting unit dividing the traveling path into a travelable area and a non-traveling area and displaying the lane in a grid form; A path search unit searching for a shortest path of the traveling path indicated in the grid form; And an optimization unit for optimizing the shortest path by a gradient-based optimization method.

In accordance with another aspect of the present invention, there is provided a method for automatically generating a traveling path, comprising: generating a unitary motion consisting of a control point constituting a starting point and an end point of a traveling path and a Bezier curve expressed by four vectors between a starting point and an end point; ; Displaying the traveling route in a lattice form by dividing the traveling route into a travelable zone and a non-travelable zone; Searching for a shortest path of the traveling path in the lattice form; And optimizing the shortest path by a gradient-based optimization method.

According to the present invention, it is possible to perform fast path calculation and optimization with a small amount of computation by reducing the number of control points for path generation, and it is possible to achieve more flexible and broad path optimization by merging control points .

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a structural diagram of a traveling path automatic generating apparatus according to an embodiment of the present invention; FIG.
2 is a flowchart of a method of automatically generating a traveling route according to another embodiment of the present invention.
3 is a diagram showing a variable representing a unit motion according to the present invention;
FIG. 4 is a structural view of an apparatus for automatically generating a traveling route according to another embodiment of the present invention. FIG.
5 is a structural diagram showing a path composed of m + 1 unit motions.
6 is a structural view showing a traveling path in a lattice form according to the present invention.
FIG. 7 is a structural view showing each process of performing unit motion merging according to the present invention; FIG.
8 is a diagram showing a change of a path according to the number of merge according to the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. As used herein, the terms " comprises, " and / or "comprising" refer to the presence or absence of one or more other components, steps, operations, and / Or additions.

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

1 shows a structure of an apparatus for automatically generating a traveling route according to an embodiment of the present invention.

A Motion Primitive Generator (100) generates a Motion Segment that links a pair of unattended vehicle states.

로 구성되는데 x, y는 무인차의 위치를 나타내는 좌표이고

는 무인차의 헤딩각,

는 곡률을 각각 나타낸다.In FIG. 2, the autonegotiation state is expressed as (x, y,

Where x and y are coordinates indicating the position of the unmanned vehicle

The heading angle of the unmanned car,

Respectively.

In the present invention, unlike the conventional method in which six control points are used to search for an automatic path by using a Bezier curve of a fifth order, two control points (a, b, c, and d) By using the -D vector, it is possible to express the curve efficiently.

~

}는 5차 베지어 커브의 컨트롤 포인트를 나타내고 베지어 커브의 모양을 결정하며 t는 곡선의 형태를 조절하기 위한 비례변수이다.The Bezier curve is expressed by the following equation (1). {

~

} Represents the control point of the 5th Bezier curve, determines the shape of the Bezier curve, and t is a proportional variable for controlling the shape of the curve.

가 주어질 때 다음과 같은 과정을 거쳐 변환된 식으로 정의된다.In the present invention, the Bezier curve is defined as (x, y,

Is defined as a transformed expression through the following process.

Hereinafter, a method of representing a Bezier curve according to the present invention will be described in detail with reference to FIG.

First, the start and end control points are fixed at the start / end positions and can be expressed by the following equation (2).

와 평행해야 하고, 대칭적으로 다섯 번째 컨트롤 포인트는 도착방향인

와 평행해야 하므로 수학식 3과 같이 나타난다.The second control point is the departure direction

, And the fifth control point symmetrically has to be parallel to the arrival direction

And therefore it is expressed by Equation (3).

를 만족시켜야 하므로 컨트롤 포인트

를 연결한 직선으로부터의 수직거리가

가 되어야 하고, 대칭적으로 네 번째 컨트롤 포인트는

로부터 수직거리가

가 되어야 하며 이를 식으로 나타내면 수학식 4와 같다.The third control point is the start curvature

Therefore,

The vertical distance from the straight line connecting

And the fourth control point symmetrically

Lt; RTI ID = 0.0 &

And expressed by Equation (4).

Equation (2) to Equation (4) are used to calculate six control points, and substituting the six control points into Equation (1), the fifth-order Bezier curve can be rewritten as Equation (5).

의 함수로 표현되나 본 발명에 의한 수학식 5는

의 함수로 표현되는 특징이 있다.That is, the conventional equation (1)

(5) according to the present invention is expressed as a function of

As shown in Fig.

In expressing one curve, there is no difference in that the conventional method or the method according to the present invention requires 12 variables. However, since a plurality of curves are combined to generate a path in automatic path generation, .

5 shows an example of a path composed of m + 1 curves.

~

은 접점이 되는 벡터로 공통되고, 각 커브는 (a, b, c, d) 벡터에 의해 경로가 표현되기 때문에 4m+4*(m+1) 개의 변수만으로 전체 경로를 표현할 수 있다.In the conventional method, since 12 variables are required for each curve, 12 * (m + 1) variables are required to represent the entire path, but according to the present invention

~

And each curve is represented by a vector of (a, b, c, d), the entire path can be represented by only 4m + 4 * (m + 1) variables.

In addition to reducing computation time by reducing the number of variables, optimization can be performed much better because the variation for optimization can be increased.

The lattice graph transforming unit 120 transforms a path into a grid shape for searching for a path as shown in FIG. 6, so that a path can be set by distinguishing a road on which the vehicle can travel and a place where the vehicle can not travel.

The path search unit 130 finds the shortest path of the vehicle on the path converted by the lattice graph transform unit 120 using the A * algorithm (ESTAR algorithm) on the path converted to the lattice form by the lattice graph transform unit I will.

The A * algorithm is one of a graph / tree search algorithm that finds the shortest path from a given starting vertex to a target vertex, and combines the curves generated by the unit motion generator 110 to find the shortest path.

In FIG. 6, the shortest travel path of the vehicle found by the route search unit 130 by the A * algorithm is indicated by a solid black line.

, 장애물과의 근접도

등으로 구성된 비용함수(cost function) J(x)를 기울기 기반 최적화(Gradient-based optimization)를 통해 최소화하고 이는 다음 수학식 6과 같이 표현된다.In the optimization unit 140, the distance s, the curvature

, Proximity to obstacles

And the like are minimized through gradient-based optimization, which is expressed as Equation (6): " (6) "

The slope-based optimization is a method of finding the minimum value by repeatedly modifying the initial estimate in the direction of the slope where the cost function is reduced. However, when the sensitivity of J (x) changes sensitively according to the variation of the free variable, The degree of freedom that can be deformed in the value is reduced.

In the prior art, the unit motion is represented by a set of points (control point), which is a control point. A change in the position of one point in the set of dense points causes a sharp increase in curvature, There are limitations to optimization because conversion can cause inconvenience to passengers.

In contrast, according to the present invention, a unitary motion Bezier curve is represented by two control points and four vectors. Since the number of control points is further reduced through the merging process, the curvature is rapidly Therefore, it is possible to achieve the optimization of the automatic traveling route with a degree of freedom that is much greater than that of the conventional invention, thereby making it possible to create a more natural route and enhance the convenience of the passenger.

In the optimization phase, the number of variables is further reduced by merging the control points. 7 is a diagram showing a step-by-step configuration in which control points are reduced by merging.

을 시작 컨트롤 포인트로 하고 반지름을 단위길이(1)로 고정한 베지어 커브의 집합을 단위모션생성기(110)에 의해 만들어낸다. 도 7의 (a)는 이렇게 생성된 베지어 커브들의 집합을 나타낸다.first

As a start control point, and sets a Bezier curve in which the radius is fixed to the unit length (1) by the unit motion generator 110. FIG. 7 (a) shows a set of Bezier curves generated in this way.

과

사이의 유클리디안 거리(Euclidean Distance)를 나타낸다.The next set of Bezier curves is scaled to fit the end of the two curves. R is

and

And the Euclidean distance between the two points.

를 건너뛰고

까지 연결될 수 있도록 단위길이의 베지어 커브를 크기 변형하는 것이다. 도 7의 (b)는 크기 변형된 베지어 커브의 집합을 나타낸다.Intermediate control point

Skip

The length of the Bezier curve of the unit length is changed. Fig. 7 (b) shows a set of size-modified Bezier curves.

와

사이,

와

사이를 연결하는 두 개의 베지어 커브와 가장 유사한 커브를 검색한다. 시작 컨트롤 포인트와 종료 컨트롤 포인트는

와

로 정해져 있으므로 전술했던 것처럼 커브의 (a, b, c, d) 벡터를 결정하여 유사한 벡터를 검색하게 된다.Size Of existing sets of deformed Bezier curves

Wow

between,

Wow

Find the curve that most closely resembles the two Bezier curves connecting the two. The start control point and the end control point

Wow

(A, b, c, d) of the curve is determined as described above, and a similar vector is searched.

,

,

,

는 병합한 커브의 (a, b, c, d)벡터를 각각 나타낸다.7 (c)

,

,

,

(A, b, c, d) vectors of the merged curve, respectively.

, R(a*, b*, c*, d*),

로 나타낼 수 있다.By using the (a, b, c, d) vector searched, two unit motion vectors are merged into a fifth unit Bezier curve, which is a unit motion vector.

, R (a *, b *, c *, d *),

.

If the collision can be avoided during the merging of the curves, the merging can be continued in a bottom-up manner.

The control point can be reduced by merging the Bezier curves through the process of FIG. 7, and the degree of freedom of optimization can be increased by reducing the number of control points, thereby making it possible to generate a more natural path.

8 shows the change of the automatic traveling route according to the merging step. FIG. 8 (a) shows a path when no merging is performed, and FIG. 8 (b) and FIG. 8 (c) each show a path after merging once and six times. The path indicated by the dotted line is the path before optimization, and the path indicated by the solid line is the path after optimization. It can be seen that the more merging, the farther away from the obstacle, the more natural curve is formed.

Through the above-described configuration, the calculation amount of the route search can be reduced by using the control point and the vector and the merge method in generating the automatic travel route by the Bezier curve, which is advantageous for the real-time route search and the degree of freedom is higher A more natural traveling route can be generated.

2 is a flowchart illustrating a method of generating an automatic traveling route according to another embodiment of the present invention.

In the unit motion generation step (S210), a motion segment is created to connect a pair of uninjured state using a Bezier curve.

의 함수로 베지어 커브를 구성하는 것은 전술한 바와 같다.Unlike conventional Bezier curves with control points only

The construction of the Bezier curve as a function of the above is as described above.

In the lattice graph generation step S220, the traveling path of the automobile is divided into a travelable space and a non-travelable space and expressed in a lattice form.

The path search step S230 is a step of finding the shortest distance in the traveling path expressed in a lattice form using the A * algorithm.

Finally, the optimization step S240 optimizes the cost function J (x) by minimizing the slope-based optimization and merging the Bezier curves, which are unit motions.

The merging process of the Bezier curve merges two Bezier curves into one Bezier curve through unit curve generation, size transformation, search and transformation, and continues merging if there is no collision with the obstacle.

Meanwhile, a method of automatically generating a traveling route according to an embodiment of the present invention may be implemented in a computer system or recorded on a recording medium. 4, a computer system includes at least one processor 421, a memory 423, a user input device 426, a data communication bus 422, a user output device 427, And may include a storage 428. Each of the above-described components performs data communication via the data communication bus 422. [

The computer system may further include a network interface 129 coupled to the network. The processor 421 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 423 and / or the storage 428.

The memory 423 and the storage 428 may include various forms of volatile or non-volatile storage media. For example, the memory 423 may include a ROM 424 and a RAM 425.

Therefore, the method of automatically generating the traveling route according to the embodiment of the present invention can be implemented by a method executable by a computer. When the model parameter learning method of the speech recognition apparatus according to the embodiment of the present invention is performed in a computer device, computer-readable instructions can perform the recognition method according to the present invention.

Meanwhile, the method for automatically generating a traveling route according to the present invention can be implemented as a computer-readable code on a computer-readable recording medium. The computer-readable recording medium includes all kinds of recording media storing data that can be decoded by a computer system. For example, there may be a ROM (Read Only Memory), a RAM (Random Access Memory), a magnetic tape, a magnetic disk, a flash memory, an optical data storage device and the like. The computer-readable recording medium may also be distributed and executed in a computer system connected to a computer network and stored and executed as a code that can be read in a distributed manner.

While the present invention has been described in detail with reference to the accompanying drawings, it is to be understood that the invention is not limited to the above-described embodiments. Those skilled in the art will appreciate that various modifications, Of course, this is possible. Accordingly, the scope of protection of the present invention should not be limited to the above-described embodiments, but should be determined by the description of the following claims.

Claims (8)

A traveling route automatic generation apparatus including at least one processor,
A unit motion generation unit for generating a unitary motion consisting of a control point constituting a starting point and an end point of a traveling path and a Bezier curve expressed by a vector composed of four variables between a starting point and an end point;
A lattice graph converting unit dividing the traveling path into a travelable area and a non-traveling area and displaying the lane in a grid form;
A path search unit searching for a shortest path of the traveling path indicated in the grid form; And
An optimizer for calculating the shortest path based on at least one of a travel distance, a curvature and an obstacle proximity;
Wherein the automatic route guidance apparatus is implemented by including the automatic route guidance apparatus.
2. The apparatus of claim 1, wherein the optimizing unit
And performing optimization by merging two adjacent unit motions of the unit motions generated by the unit motion generating unit
A device for automatically generating a traveling route.
3. The apparatus of claim 2, wherein the optimizer
Generating a set of Bezier curves having a unit length from a control point of a starting point of a first unit motion of two adjacent unit motions;
Transforming the set of Bezier curves to an end point control point of a second one of the adjacent two unit motions;
Searching four vectors most similar to the scaled unit motion among four vectors forming a unit motion including the control point of the start point and the end point control point; And
Replacing the adjacent two unit motions with a unit motion composed of the control point of the start point, the control point of the end point, and the four vectors;
Wherein the optimizing step includes:
2. The apparatus of claim 1, wherein the path search unit
Searching the shortest path by an A * algorithm
A device for automatically generating a traveling route.
Generating a unitary motion consisting of a control point constituting a starting point and an end point of a traveling path and a Bezier curve expressed by a vector composed of four variables between a starting point and an end point;
Displaying the traveling route in a lattice form by dividing the traveling route into a travelable zone and a non-travelable zone;
Searching for a shortest path of the traveling path in the lattice form; And
Calculating the shortest path based on at least one of a travel distance, a curvature and an obstacle proximity;
And automatically generating a traveling route.
6. The method of claim 5, wherein optimizing
And performing the optimization by merging two adjacent unit motions of the unit motions generated in the step of generating the unit motions
A method for automatically generating a traveling route.
7. The method of claim 6, wherein optimizing
Generating a set of Bezier curves having a unit length from a control point of a starting point of a first unit motion of two adjacent unit motions;
Transforming the set of Bezier curves to an end point control point of a second one of the adjacent two unit motions;
Searching four vectors most similar to the scaled unit motion among four vectors forming a unit motion including the control point of the start point and the end point control point; And
Replacing the adjacent two unit motions with a unit motion composed of the control point of the start point, the control point of the end point, and the four vectors;
Wherein the optimizing step includes:
6. The method of claim 5, wherein the step of searching for the shortest path
Searching the shortest path by an A * algorithm
A method for automatically generating a traveling route.

Images