KR20210103911A - Method, apparatus and computer program for generating vehicle path to avoid collision - Google Patents

Abstract

Provided are a method for generating a collision avoidance path of a vehicle, which designs collision avoidance paths in each of a plurality of vehicles, and an apparatus and a computer program thereof. According to various embodiments of the present invention, a method for generating a collision avoidance path of a vehicle is executed by a computing device. The method comprises the following steps: acquiring location information on a plurality of vehicles for a predetermined period in the future with respect to a current time point on the basis of a driving route preset for each of the plurality of vehicles; calculating a loss value for the plurality of vehicles by using the location information on the plurality of vehicles; performing a process of optimizing the loss value to acquire a variable for optimizing the loss value; and generating a collision avoidance path for each of the plurality of vehicles by using the obtained variable.

Description

METHOD, APPARATUS AND COMPUTER PROGRAM FOR GENERATING VEHICLE PATH TO AVOID COLLISION

Various embodiments of the present invention relate to a method, an apparatus, and a computer program for generating a collision avoidance path of a vehicle for preventing a collision between a plurality of vehicles.

For the convenience of users driving a vehicle, various sensors and electronic devices (eg, an Advanced Driver Assistance System (ADAS)) are being provided, and in particular, to recognize the surrounding environment without driver intervention, Technological development for an autonomous driving system of a vehicle that automatically drives to a given destination according to a recognized surrounding environment is being actively developed.

On the other hand, since the autonomous driving system automatically drives by judging by the system itself without the intervention of the driver, there is a problem in that there is a high possibility of a collision between the autonomous vehicle and adjacent objects (eg, vehicles, people, and other structures). .

In the related art, in order to prevent such a collision accident, a spatial and temporal collision avoidance path for preventing a collision is designed, and the vehicle is driven according to the designed collision avoidance path.

However, in the conventional collision avoidance path generation method, when the number of vehicles that need to generate collision avoidance paths increases, temporal and spatial collision avoidance paths must be designed for a large number of vehicles, respectively, so the amount of computation increases and collision avoidance increases. It may take a lot of time to generate the path, and as the amount of computation increases, the probability of error occurrence increases, so that an incorrect collision avoidance path may be generated.

In addition, since the conventional collision avoidance path generation method generates only a spatial collision avoidance path, there is a problem in that the vehicle travels somewhat inefficiently, such as a long way to avoid a collision.

Korea Patent No. 10-1304625 (2013.08.30)

SUMMARY OF THE INVENTION An object of the present invention is to provide a method, an apparatus and a computer program for generating a collision avoidance path of a vehicle capable of designing a collision avoidance path for each of a plurality of vehicles in order to prevent a collision between a plurality of vehicles.

Another problem to be solved by the present invention is a method of generating a collision avoidance path for a vehicle that can generate not only a spatial avoidance path but also a temporal avoidance path by using various variables in generating a collision avoidance path using the gradient descent method, To provide devices and computer programs.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

In the method for generating a collision avoidance path of a vehicle according to an embodiment of the present invention for solving the above-mentioned problems, in a method performed by a computing device, a current time point is determined based on a driving path preset for each of a plurality of vehicles. Obtaining location information on the plurality of vehicles for a predetermined period in the future as a reference, calculating loss values for the plurality of vehicles using the location information on the plurality of vehicles, optimizing the loss values and obtaining a variable for optimizing the loss value by performing , and generating a collision avoidance path for each of the plurality of vehicles by using the obtained variable.

In various embodiments, performing the optimization process may include obtaining a variable for optimizing the loss value using a gradient descent method.

In various embodiments, the obtaining of the location information for the plurality of vehicles includes obtaining reference coordinates indicating the current location of each of the plurality of vehicles, dividing the predetermined period in the future into a plurality of unit times, and each unit time The method may include obtaining a plurality of offset coordinates indicating deviations between the positions of the plurality of vehicles.

In various embodiments, the calculating of the loss value includes calculating a plurality of cost values for the plurality of vehicles and calculating the loss value using the plurality of calculated cost values, and setting a preset weight to each of the cost values, wherein the plurality of cost values are calculated based on a distance between the plurality of vehicles and a difference from the existing routes of the plurality of vehicles. The calculated cost value may be included.

In various embodiments, the calculating of the loss value using the plurality of calculated cost values includes calculating the speed and time of each of the plurality of vehicles per unit time, and determining the location of each of the plurality of vehicles per unit time. Calculating, based on the respective positions of the plurality of vehicles per unit time, calculating the distance between the plurality of vehicles per unit time, and a collision cost value for the plurality of vehicles in inverse proportion to the distance between the plurality of vehicles per unit time It may include the step of calculating

In various embodiments, calculating the collision cost value includes dividing the distance between the plurality of vehicles by a reference size to normalize the distribution, and assigning a preset weight to the calculated collision cost value, The weight may be a value obtained by dividing a speed of a first vehicle that is a target for calculating the collision cost value by a sum of a speed of a second vehicle adjacent to the first vehicle and a speed of the first vehicle.

In various embodiments, the calculating of the loss value by using the plurality of calculated cost values may include calculating time cost values for the plurality of vehicles based on a change amount of each unit time compared to a reference time; calculating an offset cost value for the plurality of vehicles based on the sum of the offset coordinate sizes for each unit time, and an offset difference cost value for the plurality of vehicles based on a difference value of the offset coordinate sizes between each time point for each unit time It may include the step of calculating

In various embodiments, the generating of the collision avoidance path for each of the plurality of vehicles may include calculating the offset coordinates and speed of the vehicle for collision avoidance by optimizing the loss value, the offset of the vehicle for collision avoidance Generating the collision avoidance position coordinates of each of a plurality of vehicles for each unit time using the coordinates and the speed of the vehicle, and the collision avoidance for each of the plurality of vehicles using linear interpolation and the collision avoidance position coordinates It may include creating a route.

An apparatus for generating a collision avoidance path for a vehicle according to another embodiment of the present invention for solving the above-described problems includes a memory storing one or more instructions and a processor executing the one or more instructions stored in the memory, the processor may perform the method of generating a collision avoidance path of a vehicle according to an embodiment of the present invention by executing the one or more instructions.

A computer program according to another embodiment of the present invention for solving the above-described problems is combined with a computer that is hardware, so that the computer can perform the method of generating a collision avoidance path of a vehicle according to an embodiment of the present invention may be stored in a readable recording medium.

Other specific details of the invention are included in the detailed description and drawings.

According to various embodiments of the present disclosure, there is an advantage in that a collision avoidance path for each of a plurality of vehicles can be designed in order to prevent a collision between the plurality of vehicles.

In addition, in designing the collision avoidance path, it is possible to provide a collision avoidance path optimized for each of a plurality of vehicles by generating a temporal or spatial collision avoidance path by considering not only the distance between vehicles but also the difference from the existing path. There is this.

Effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a diagram illustrating a system for generating a collision avoidance path of a vehicle according to an embodiment of the present invention.
2 is a hardware configuration diagram of an apparatus for generating a collision avoidance path for a vehicle according to another embodiment of the present invention.
3 is a flowchart of a method for generating a collision avoidance path of a vehicle according to another embodiment of the present invention.
4 is a diagram illustrating a form in which an apparatus for generating a collision avoidance path of a vehicle generates location information for each unit time based on a driving path of a vehicle, according to various embodiments of the present disclosure;
5 to 7 are diagrams illustrating a form in which an apparatus for generating a collision avoidance path of a vehicle generates a spatial and temporal collision avoidance path of a vehicle using a gradient descent method, according to various embodiments of the present disclosure;
8 to 10 are diagrams illustrating a collision avoidance path of a vehicle and a preset driving path generated by an apparatus for generating a collision avoidance path of a vehicle, according to various embodiments of the present disclosure;

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully understand the scope of the present invention to those skilled in the art, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless specifically defined explicitly.

As used herein, the term “unit” or “module” refers to a hardware component such as software, FPGA, or ASIC, and “unit” or “module” performs certain roles. However, “part” or “module” is not meant to be limited to software or hardware. A “unit” or “module” may be configured to reside on an addressable storage medium or to reproduce one or more processors. Thus, by way of example, “part” or “module” refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. Components and functionality provided within “parts” or “modules” may be combined into a smaller number of components and “parts” or “modules” or as additional components and “parts” or “modules”. can be further separated.

Spatially relative terms "below", "beneath", "lower", "above", "upper", etc. It can be used to easily describe the correlation between a component and other components. A spatially relative term should be understood as a term that includes different directions of components during use or operation in addition to the directions shown in the drawings. For example, when a component shown in the drawing is turned over, a component described as “beneath” or “beneath” of another component may be placed “above” of the other component. can Accordingly, the exemplary term “below” may include both directions below and above. Components may also be oriented in other orientations, and thus spatially relative terms may be interpreted according to orientation.

In this specification, a computer means all types of hardware devices including at least one processor, and may be understood as encompassing software configurations operating in the corresponding hardware device according to embodiments. For example, a computer may be understood to include, but is not limited to, smart phones, tablet PCs, desktops, notebooks, and user clients and applications running on each device.

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

Each step described in this specification is described as being performed by a computer, but the subject of each step is not limited thereto, and at least a portion of each step may be performed in different devices according to embodiments.

In addition, although the description is based on an autonomous driving vehicle that autonomously travels according to a preset driving route, the present specification is not limited thereto, and may be applied in various fields. For example, it may be applied to a vehicle that does not use an autonomous driving function or a vehicle that does not have an autonomous driving function. In addition, in the field of controlling group driving of a plurality of RC (Radio control) cars or unmanned aerial vehicles (eg drones), it is also applicable in the process of creating a collision avoidance path to avoid collisions between a plurality of RC cars or drones. possible.

1 is a diagram illustrating a system for generating a collision avoidance path of a vehicle according to an embodiment of the present invention.

Referring to FIG. 1 , a system for generating a collision avoidance path for a vehicle according to an embodiment of the present invention may include an apparatus 100 for generating a collision avoidance path for a vehicle, a user terminal 200 , and an external server 300 .

Here, the system for generating a collision avoidance path for a vehicle shown in FIG. 1 is according to an embodiment, and its components are not limited to the embodiment shown in FIG. 1 , and may be added, changed, or deleted as necessary. .

In an embodiment, the apparatus 100 for generating a collision avoidance path of a vehicle may obtain location information of each of a plurality of vehicles based on a driving path preset for each of the plurality of vehicles, and using the obtained location information for each of the plurality of vehicles You can create an escape route.

In various embodiments, the apparatus 100 for generating a collision avoidance path of a vehicle may generate a high dimensional vector (eg, FIG. 5 ) using driving paths and location information of each of a plurality of vehicles. Thereafter, the apparatus 100 for generating a collision avoidance path of a vehicle calculates a loss value for a plurality of vehicles using the generated one vector, and obtains a variable for optimizing the loss value using a gradient descent method can (eg, Figure 6). Thereafter, the apparatus 100 for generating a collision avoidance path of a vehicle may generate a temporal and spatial collision avoidance path for each of a plurality of vehicles by using the obtained variable (eg, FIG. 7 ). However, the present invention is not limited thereto.

In various embodiments, the apparatus 100 for generating a collision avoidance path for a vehicle operates according to a temporal and spatial collision avoidance path generated for each of a plurality of vehicles based on a temporal and spatial collision avoidance path generated for each of the plurality of vehicles It can transmit a control signal instructing what to do. For example, the collision avoidance path generating apparatus 100 of the vehicle calculates a control signal instructing to operate according to the temporal and spatial collision avoidance path using the master software provided therein (eg, throttling and steering control values). calculation), which is provided in the vehicle and is connected to a control module that controls the operation of the vehicle and transmits the control signal to the control module of the vehicle, so that the vehicle receives the control signal and controls the operation (eg, throttling control and steering control). However, the present invention is not limited thereto.

In an embodiment, the user terminal 200 may be connected to the apparatus 100 for generating a collision avoidance path of a vehicle through the network 400 , and the device for generating a vehicle collision avoidance path through the network 400 provides location information of the vehicle. (100) can be provided. For example, the user terminal 200 includes at least one of a smart phone, a tablet PC, a laptop computer, a desktop, and an infotainment system of a vehicle equipped with a sensor (eg, a GPS sensor) capable of acquiring location information. can do. However, the present invention is not limited thereto.

In various embodiments, the user terminal 200 may include a display in at least a portion of the user terminal 200, and various user interfaces provided from the apparatus 100 for generating a vehicle collision avoidance path through the display. UI) can be printed. For example, the user terminal 200 may be an infotainment system of a vehicle equipped with a display, and receive a UI for outputting the current driving path and collision avoidance path of the vehicle from the vehicle collision avoidance path generating apparatus 100 to receive infotainment A corresponding UI can be output through a display in the system. However, the present invention is not limited thereto.

In an embodiment, the external server 300 may be connected to the apparatus 100 for generating a collision avoidance path for a vehicle through the network 400 , and various kinds of information from the apparatus 100 for generating a collision avoidance path for a vehicle through the network 400 . Data can be received and stored. For example, the external server 300 may receive and store location information of a plurality of vehicles from the apparatus 100 for generating a collision avoidance path of a vehicle, and data indicating a preset driving path and a collision avoidance path.

Here, in the system for generating a collision avoidance path for a vehicle shown in FIG. 1 , various data generated by the device for generating a collision avoidance path for a vehicle 100 are transmitted and stored to the external server 300, but is not limited thereto. Instead, various data generated by the apparatus 100 for generating a collision avoidance path for a vehicle may be stored in a storage device (not shown) separately provided in the apparatus 100 for generating a collision avoidance path for a vehicle. Hereinafter, a hardware configuration of the apparatus 100 for generating a vehicle collision avoidance path will be described with reference to FIG. 2 .

2 is a hardware configuration diagram of an apparatus for generating a collision avoidance path for a vehicle according to another embodiment of the present invention.

Referring to FIG. 2 , an apparatus 100 for generating a collision avoidance path for a vehicle (hereinafter, referred to as “computing device 100”) according to another embodiment of the present invention may include a processor 110 and a memory 120 . . In various embodiments, the computing device 100 may further include a network interface (or communication interface) (not shown), storage (not shown), and a bus (not shown).

In an embodiment, the processor 110 may control the overall operation of each component of the computing device 100 . The processor 110 may include a central processing unit (CPU), a micro processor unit (MPU), a micro controller unit (MCU), or any type of processor well known in the art.

In various embodiments, the processor 110 may perform an operation on at least one application or program for executing the method according to the embodiments of the present invention. In various embodiments, the processor 110 includes one or more cores (not shown) and a graphic processing unit (not shown) and/or a connection path (eg, a bus, etc.) for transmitting and receiving signals to and from other components. can do.

In various embodiments, the processor 110 temporarily and/or permanently stores a signal (or data) processed inside the processor 110 , a random access memory (RAM) and a read access memory (ROM). -Only Memory, not shown) may be further included. In addition, the processor 110 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processing unit, a RAM, and a ROM.

In one embodiment, the processor 110 executes one or more instructions stored in the memory 120 to perform a method (eg, a method for generating a collision avoidance path of a vehicle) to be described with reference to FIGS. 3 to 10 . can For example, the processor 110 executes one or more instructions stored in the memory 120, so that the plurality of vehicles for a predetermined period in the future based on the current time point based on the driving route preset for each of the plurality of vehicles. obtaining location information for the vehicle, calculating loss values for a plurality of vehicles by using the location information for the plurality of vehicles, performing a loss value optimization process to obtain a variable for optimizing the loss value, and A method of generating a collision avoidance path for a vehicle including generating a collision avoidance path for each of a plurality of vehicles by using a variable may be performed.

In one embodiment, memory 120 may store various data, commands, and/or information. The memory 120 may store programs (one or more instructions) for processing and controlling the processor 110 . Programs stored in the memory 120 may be divided into a plurality of modules according to functions.

In various embodiments, steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, as a software module executed by hardware, or by a combination thereof. A software module may contain random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any type of computer-readable recording medium well known in the art to which the present invention pertains.

The components of the present invention may be implemented as a program (or application) to be executed in combination with a computer, which is hardware, and stored in a medium. Components of the present invention may be implemented as software programming or software components, and similarly, embodiments may include various algorithms implemented as data structures, processes, routines, or combinations of other programming constructs, including C, C++ , Java, assembler, etc. may be implemented in a programming or scripting language. Functional aspects may be implemented in an algorithm running on one or more processors. Hereinafter, a method of generating a vehicle collision avoidance path performed by the computing device 100 will be described with reference to FIG. 3 .

3 is a flowchart of a method for generating a collision avoidance path of a vehicle according to another embodiment of the present invention.

Referring to FIG. 3 , in step S110 , the computing device 100 may obtain location information from each of a plurality of vehicles. For example, when the plurality of vehicles is an autonomous driving vehicle capable of being driven by an actual driver, the computing device 100 collects vehicle location information from an infotainment system in the plurality of vehicles or is provided in the driver's smartphone in the vehicle. It is possible to collect vehicle location information from the GPS sensor. However, the present invention is not limited thereto, and any method by which the computing device 100 collects vehicle location information may be applied.

In various embodiments, the computing device 100 is an RC car in which a plurality of vehicles are externally controlled by a person, and when the space in which the plurality of vehicles travel is an indoor space in which a GPS sensor is difficult to accurately recognize, the computing device 100 is installed indoors ( For example: installation on a ceiling) It is possible to obtain location information including location coordinates for each of a plurality of vehicles based on an image captured by at least one camera. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 may obtain location information on a plurality of vehicles for a predetermined period in the future based on a current time point based on current location information of each of the plurality of vehicles and a preset driving route. .

In various embodiments, the computing device 100 sets the locations and routes of the plurality of vehicles for a predetermined period in the future based on the current time point based on the current location information of each of the plurality of vehicles and the preset driving route as one It can be expressed as a vector (eg, FIG. 5 ). It will be described with reference to FIG. 4 .

4 is a diagram illustrating a form in which an apparatus for generating a collision avoidance path of a vehicle generates location information for each unit time based on a driving path of a vehicle, according to various embodiments of the present disclosure;

Referring to FIG. 4 , in various embodiments, the computing device 100 may obtain location information for a plurality of vehicles for a predetermined period in the future based on current location information of each of the plurality of vehicles and a preset driving route. .

First, the computing device 100 may obtain coordinates (eg, (X, Y)) indicating the current location of the vehicle 10 based on the current location information of the vehicle 10 obtained through step S110, The coordinates indicating the current location of (10) can be set as the reference coordinates.

Thereafter, the computing device 100 may divide a predetermined future period into a plurality of unit times. For example, the computing device 100 may divide the future 2 seconds based on the current time into 25 ms, which is a preset unit time. However, the present invention is not limited thereto, and the predetermined period and unit time in the future may be variously set.

Thereafter, the computing device 100 may obtain a plurality of offset coordinates indicating a deviation between positions of a plurality of vehicles _{for each unit time (T 1} , T ₂ , T ₃ , T _{4 ). For example, when the location of the vehicle 10 at the time point T 1} is (X+X _offset1 , Y+Y _offset1 ), the computing device 100 determines the location of the vehicle 10 and the current location of the vehicle 10 at _{T 1 .} You can get the _{(X offset1} , Y _offset1 ) coordinates that are the deviation of the position. In addition, when the position at the time point _{T 2 is (X+X offset1} +X _offset2 , Y+Y _offset1 +Y _offset2 ), the computing device 100 is the position (X) of the vehicle 10 at the time point _{T 1 .} +X _offset1 , Y+Y _offset1 ) and _{(X offset2} , Y _offset2 ) coordinates that are the deviation of the position of the vehicle 10 at the time point _{T 2} may be obtained. In addition, computing device 100 is T ₃ in the case of the position of the point (X + X _offset1 + X _offset2 + X _offset3, Y + Y _offset1 + Y _offset2 + Y _offset3), T ₂ of the vehicle at the time ( 10) (X+X _offset1 +X _offset2 , Y+Y _offset1 +Y _offset2 ) and _{(X offset3} , Y _offset3 ) coordinates that are the deviation of the position of the vehicle 10 at the time point _{T 3} can be obtained.

In various embodiments, the computing device 100 provides a plurality of devices indicating a deviation between the current position (reference coordinate) of each of the plurality of vehicles and the positions of the plurality of vehicles per unit time (T ₁ , T ₂ , T ₃ , T _{4 ).} You can get the offset coordinates. _{For example, when the location of the vehicle 10 at the time point T 1} is (X+X _offset1 , Y+Y _offset1 ), the computing device 100 determines the location of the vehicle 10 and the current location of the vehicle 10 at _{T 1 .} You can get the _{(X offset1} , Y _offset1 ) coordinates that are the deviation of the position. In addition, when the location at the time point _{T 2 is (X+X offset1} +X _offset2 , Y+Y _offset1 +Y _offset2 ), the computing device 100 is the location and the current location of the vehicle 10 in _{T 2 .} (X _offset1 +X _offset2 , Y _offset1 +Y _offset2 ) coordinates can be obtained.

In various embodiments, the computing device 100 may obtain the coordinates themselves indicating the positions of the plurality of vehicles _per unit time (T 1 , T ₂ , T ₃ , T _{4 ).} However, the present invention is not limited thereto, and various methods for calculating the position coordinates of the vehicle per unit time may be applied.

In various embodiments, the computing device 100 may represent the position coordinates (offset coordinates) of the vehicle 10 for each unit time obtained through the above method as one vector. For example, when the offset coordinates obtained by the above method are (0.0, 0.5), (0.5, 0.6), and (1.0, 0.7), the computing device 100 sets the coordinates to (0.0, 0.5, 0.5, 0.6). , 1.0, 0.7) to create one path vector. However, the present invention is not limited thereto.

Referring again to FIG. 3 , in operation S120 , the computing device 100 may calculate a loss value for the plurality of vehicles by using location information on the plurality of vehicles.

In various embodiments, the computing device 100 may calculate a plurality of cost values for a plurality of vehicles, and may calculate a loss value using the plurality of calculated cost values.

Here, the plurality of cost values may include a cost value calculated based on a distance between the plurality of vehicles and a cost value calculated based on a difference between the plurality of vehicles and an existing route. For example, the plurality of cost values may include a collision cost value, a time cost value, an offset cost value, and an offset differential cost value. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 may calculate a loss value by summing all of a plurality of cost values. For example, the computing device 100 sums the collision cost value, the time cost value, the offset cost value, and the offset difference cost value if a plurality of cost values are a collision cost value, a time cost value, an offset cost value, and an offset difference cost value. Thus, the loss value can be calculated.

In various embodiments, the computing device 100 may calculate a loss value using a plurality of cost values, but may set a preset weight to each of the plurality of calculated cost values.

In various embodiments, computing device 100 sets weights for each of a collision cost value, a time cost value, an offset cost value, and an offset difference cost value, with an offset difference cost value, a collision cost value, a time cost value, and an offset cost. You can set higher weights in the order of values. For example, the computing device 100 may set the weight of the offset differential cost value to 64, the weight of the collision cost value to 32, the weight to the time cost value to 4, and the weight to the offset cost value to 0.1. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 may generate a loss function to calculate a loss value of the vehicle. For example, the computing device 100 is configured to calculate a collision cost calculation function for calculating a collision cost value, a time cost calculation function for calculating a time cost value, an offset cost calculation function for calculating an offset cost value, and a method for calculating an offset difference cost value The offset difference cost calculation function may be generated, respectively, and the loss function may be generated by combining the generated collision cost calculation function, the time cost calculation function, the offset cost calculation function, and the offset difference cost calculation function.

In various embodiments, the computing device 100 may calculate a collision cost value of a vehicle per unit time by generating a collision cost calculation function. For example, the computing device 100 may calculate the speed and time of each of the plurality of vehicles per unit time, and calculate the position of each of the plurality of vehicles per unit time based on the calculated speed and time.

Thereafter, the computing device 100 may calculate a distance between the plurality of vehicles per unit time based on the respective positions of the plurality of vehicles per unit time.

Thereafter, the computing device 100 may generate a collision cost calculation function that calculates a collision cost value according to the distance between a plurality of vehicles per unit time, and may calculate a collision cost value using the generated collision cost calculation function. .

In various embodiments, the computing device 100 may design a collision cost calculation function to calculate a collision cost value with a magnitude inversely proportional to a distance between a plurality of vehicles per unit time.

In various embodiments, the computing device 100 may set a weight in order to consider a case in which a vehicle collides by rapidly approaching a stopped or parked vehicle. For example, the computing device 100 divides the distance between a plurality of vehicles into a reference size and normalizes the distribution (eg, calculates a numerical value to be similar to the graph form of the normal distribution in the form of ^{exp(-ｘ 2 )),} A preset weight may be assigned to the calculated collision cost value. In this case, the computing device 100 sets a value obtained by dividing the speed of the first vehicle, which is the target for calculating the collision cost value, by the sum of the speed of the second vehicle adjacent to the first vehicle and the speed of the first vehicle as a weight, Alternatively, a weight in case of a collision by rapidly approaching a parked vehicle may be increased. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 may generate a time cost calculation function to calculate a time cost value for the vehicle. For example, the computing device 100 may generate a time cost calculation function that calculates a time cost value based on a change amount of each unit time compared to a reference time (eg, a value of real number 1), and calculates the generated time cost The function can be used to calculate time cost values for a plurality of vehicles. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 may calculate an offset cost value for a vehicle by generating an offset cost calculation function. For example, the computing device 100 may generate an offset cost calculation function that calculates an offset cost value based on the sum of the sizes of the offset coordinates per unit time (the sum of the sizes of the offset coordinates themselves), and calculates the generated offset cost. The function may be used to calculate offset cost values for a plurality of vehicles. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 may calculate an offset difference cost value for a vehicle by generating an offset difference cost calculation function. For example, the computing device 100 may generate an offset difference cost calculation function that calculates an offset difference cost value based on a difference value of the offset coordinate size between each time point for each unit time, and the generated offset difference cost calculation function can be used to calculate offset difference cost values for a plurality of vehicles. However, the present invention is not limited thereto.

In step S130 , the computing device 100 may obtain a variable for optimizing the loss value by performing the process of optimizing the loss value.

In various embodiments, the computing device 100 may obtain a variable for optimizing a loss value using gradient descent.

Here, the variable for optimizing the loss value may be a value calculated by optimizing the loss function generated in step S120 using gradient descent, and as shown in FIG. 6 , the smallest value (eg, collision cost value, time cost) a value in which the sum of the value, the offset cost value, and the offset difference cost value is the smallest). However, the present invention is not limited thereto, and any method for obtaining a variable optimizing the loss value may be applied.

Here, various literatures such as "An overview of gradient descent optimization algorithms", Sebastian Ruder, 2017, etc. may be referred to with respect to gradient descent.

In various embodiments, the computing device 100 obtains a variable for optimizing the loss value using gradient descent, but repeatedly performs the process of obtaining the variable for optimizing the loss value by a preset number of times, and repeatedly performs the acquisition A variable having the lowest value among a plurality of variables may be selected.

In operation S140 , the computing device 100 may generate collision avoidance paths (eg, FIGS. 8 to 10 ) for each of the plurality of vehicles by using the variables obtained in operation S130 .

Here, the collision avoidance path includes a temporal avoidance path that changes the time passing through the preset driving path by controlling the speed of the vehicle as shown in FIG. 7 and a collision avoidance path regenerated by deviating from the preset driving path It may include a spatial avoidance path through

In various embodiments, the computing device 100 calculates the offset coordinates and speed of the vehicle for collision avoidance by using the variables obtained by optimizing the loss value, and calculates the offset coordinates and the speed of the vehicle for collision avoidance. It is possible to generate the collision avoidance position coordinates of each of the plurality of vehicles for each unit time by using the . Thereafter, the computing device 100 may generate a collision avoidance path for each of the plurality of vehicles using linear interpolation and collision avoidance position coordinates. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 configures a temporal collision avoidance path or At least one collision avoidance path among spatial avoidance paths may be generated. For example, the computing device 100 may generate a collision avoidance path of the corresponding vehicle as a temporal collision avoidance path when a time cost value is the largest among a plurality of cost values constituting a variable. However, the present invention is not limited thereto.

In various embodiments, the computing device 100 displays a UI 20 that outputs preset driving paths and collision avoidance paths for a plurality of vehicles to the user terminal 200 (eg, a driver of each vehicle or collision of a plurality of vehicles). It can be provided to the terminal of the administrator who monitors whether or not. For example, as shown in FIGS. 9 and 10 , the computing device 100 may display a driving path preset for each of a plurality of vehicles and a UI ( 20), by providing information such as whether the vehicle moves according to a preset path and in which direction the collision avoidance path is generated according to the possibility of collision, the user can be guided to easily grasp the movement of a plurality of vehicles. .

In various embodiments, the computing device 100 may output a preset driving path and a collision avoidance path of the vehicle together as shown in FIGS. 9 and 10 , and in this case, to distinguish the preset driving path from the collision avoidance path It can be displayed in different colors for ease of use (eg, a preset driving route can be displayed in white and a collision avoidance route can be displayed in orange).

The above-described method for generating a collision avoidance path of a vehicle has been described with reference to the flowchart shown in the drawings. For the sake of simplicity, the method for generating a collision avoidance path of a vehicle has been described by showing a series of blocks, but the present invention is not limited to the order of the blocks, and some blocks may be performed in an order different from that shown and operated in this specification, or or simultaneously. In addition, new blocks not described in this specification and drawings may be added, or some blocks may be deleted or changed.

In addition, the above-described method for generating a collision avoidance path for a vehicle generates a collision avoidance path based on a vehicle having two-dimensional motion, such as an autonomous driving vehicle that can be operated by a person directly or an RC car that can be controlled by a human from the outside. However, the method is not limited thereto, and the above-described method for generating a collision avoidance path of a vehicle can be applied to an object having a three-dimensional movement such as a drone, and a collision using three-dimensional coordinates indicating the three-dimensional movement of the drone is used. You can create an escape route.

As mentioned above, although embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains can realize that the present invention can be embodied in other specific forms without changing its technical spirit or essential features. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

100: collision avoidance path generation device of the vehicle (computing device)
200: user terminal
300 : external server
400: network

Claims (10)

A method performed by a computing device, comprising:
obtaining location information on the plurality of vehicles for a predetermined period in the future based on a current time point based on a driving route preset for each of the plurality of vehicles;
calculating a loss value for the plurality of vehicles by using the location information on the plurality of vehicles;
obtaining a variable for optimizing the loss value by performing an optimization process for the loss value; and
generating a collision avoidance path for each of the plurality of vehicles by using the obtained variable;
How to create a collision avoidance path for a vehicle. According to claim 1,
The step of performing the optimization process is,
using gradient descent to obtain a variable that optimizes the loss value,
How to create a collision avoidance path for a vehicle. According to claim 1,
The step of obtaining location information for the plurality of vehicles includes:
obtaining reference coordinates indicating a current location of each of the plurality of vehicles; and
dividing the predetermined period in the future into a plurality of unit times, and obtaining a plurality of offset coordinates indicating deviations between positions of the plurality of vehicles per unit time,
How to create a collision avoidance path for a vehicle. 4. The method of claim 3,
Calculating the loss value comprises:
calculating a plurality of cost values for the plurality of vehicles, calculating the loss value using the plurality of calculated cost values, and setting a preset weight to each of the plurality of calculated cost values; and
The plurality of cost values are
including a cost value calculated based on the distance between the plurality of vehicles and a cost value calculated based on a difference between the plurality of vehicles and an existing route,
How to create a collision avoidance path for a vehicle. 5. The method of claim 4,
Calculating the loss value using the plurality of calculated cost values includes:
calculating the speed and time of each of the plurality of vehicles for each unit time;
calculating a location of each of the plurality of vehicles per unit time;
calculating a distance between the plurality of vehicles per unit time based on the positions of the plurality of vehicles per unit time; and
Comprising the step of calculating a collision cost value for the plurality of vehicles to be in inverse proportion to the distance between the plurality of vehicles per unit time,
How to create a collision avoidance path for a vehicle. 6. The method of claim 5,
Calculating the collision cost value comprises:
dividing the distance between the plurality of vehicles by a reference size to normalize the distribution; and
Comprising the step of giving a preset weight to the calculated collision cost value,
The weight is a value obtained by dividing the speed of the first vehicle, which is a target for calculating the collision cost value, by the sum of the speed of the second vehicle adjacent to the first vehicle and the speed of the first vehicle,
How to create a collision avoidance path for a vehicle. 5. The method of claim 4,
Calculating the loss value using the plurality of calculated cost values includes:
calculating time cost values for the plurality of vehicles based on a change amount of each unit time compared to a reference time;
calculating offset cost values for the plurality of vehicles based on the sum of the offset coordinate sizes for each unit time; and
Comprising the step of calculating the offset difference cost value for the plurality of vehicles based on the difference value of the offset coordinate size between each time point for each unit time,
How to create a collision avoidance path for a vehicle. 4. The method of claim 3,
The step of generating a collision avoidance path for each of the plurality of vehicles comprises:
calculating offset coordinates and speed of the vehicle for collision avoidance by optimizing the loss value;
generating collision avoidance position coordinates of each of a plurality of vehicles per unit time by using the vehicle's offset coordinates for collision avoidance and the speed of the vehicle; and
generating the collision avoidance path for each of the plurality of vehicles using linear interpolation and the collision avoidance location coordinates;
How to create a collision avoidance path for a vehicle. a memory storing one or more instructions; and
a processor executing the one or more instructions stored in the memory;
The processor by executing the one or more instructions,
An apparatus for performing the method of claim 1 . A computer program stored in a computer-readable recording medium in combination with a computer, which is hardware, to perform the method of claim 1.

Images