WO2024111966A1 - Device for tracking path of autonomous vehicle by using improved variable look-ahead distance and method therefor - Google Patents

Device for tracking path of autonomous vehicle by using improved variable look-ahead distance and method therefor Download PDF

Info

Publication number
WO2024111966A1
WO2024111966A1 PCT/KR2023/018036 KR2023018036W WO2024111966A1 WO 2024111966 A1 WO2024111966 A1 WO 2024111966A1 KR 2023018036 W KR2023018036 W KR 2023018036W WO 2024111966 A1 WO2024111966 A1 WO 2024111966A1
Authority
WO
WIPO (PCT)
Prior art keywords
curvature
vehicle
path
forward looking
distance
Prior art date
Application number
PCT/KR2023/018036
Other languages
French (fr)
Korean (ko)
Inventor
김지훈
한동석
Original Assignee
경북대학교 산학협력단
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 경북대학교 산학협력단 filed Critical 경북대학교 산학협력단
Publication of WO2024111966A1 publication Critical patent/WO2024111966A1/en

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/20—Control system inputs
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2510/00—Input parameters relating to a particular sub-units
    • B60W2510/20—Steering systems
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2520/00—Input parameters relating to overall vehicle dynamics
    • B60W2520/10—Longitudinal speed
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/30—Road curve radius

Definitions

  • the present invention is an improved path-following performance of an autonomous vehicle with an Ackerman-type steering device by varying the forward looking distance in the Pure Pursuit algorithm, a geometric path-following method, according to the current speed of the autonomous vehicle and the curvature of the reference path.
  • This relates to a path tracking device and method for an autonomous vehicle using variable forward looking distance.
  • an autonomous vehicle drives safely to its destination by recognizing, judging, route planning, and controlling the road environment on its own based on data obtained from various sensors installed in the vehicle or in already established infrastructure. It is a vehicle that does. Moreover, path following is an essential element in autonomous vehicles, along with recognition and judgment of the surrounding environment.
  • the dynamic method is a method of following a planned path in an autonomous vehicle (Path Tracking).
  • the dynamic method has excellent tracking performance by considering various dynamics of the vehicle, tires, and ground, but it is difficult to apply in real time due to complex and numerous variables.
  • a dynamic method is Model Predictive Control (MPC).
  • MPC Model Predictive Control
  • the geometric method uses only geometric considerations for the reference path and the position and direction of the vehicle, so the algorithm is simple and easy to apply, so it is widely used and studied, but has the disadvantage of relatively poor tracking performance.
  • geometric methods include the Pure Pursuit method and Stanley method.
  • the Pure Pursuit method uses information about the current location, direction, and planned path of the autonomous vehicle and calculates the front wheel steering angle of the autonomous vehicle only by geometric considerations, so it is simple and easy to apply, so it is widely used as a path-following method for autonomous driving. .
  • look-ahead distance is an important factor in determining path-following performance, and a fixed value or a value proportional only to speed is mainly used.
  • Patent Document 1 Republic of Korea Patent Publication No. 10-2394465 (2022.05.06., notice)
  • Patent Document 2 Republic of Korea Patent Publication No. 10-2349654 (2021.01.13., notice)
  • Patent Document 3 Republic of Korea Patent Publication No. 10-2017-0119463 (October 27, 2017, published)
  • the present invention is intended to solve the above problem. Based on the Pure Pursuit algorithm, the purpose of the present invention is to improve the path-following performance of an autonomous vehicle by varying the forward looking distance according to the current speed of the autonomous vehicle and the curvature of the reference path.
  • the path tracking device for an autonomous vehicle using an improved variable forward looking distance is derived from sensing data including reference path information on the planned route of the autonomous vehicle and the speed, direction, and location of the vehicle according to the driving environment.
  • Forward looking distance based on the derived vehicle path curvature (w) and vehicle speed (v) Variable forward looking distance that determines Selection module, the reference path information and the variable forward looking distance Forward looking distance determined in the selection module From the target point selection module that determines the target point, the direction and location and the target point determined in the target point selection module yielding a value calculation module; and the forward viewing distance and Steering angle from the value Steering angle that calculates and outputs the calculated steering angle data Includes calculation module.
  • variable forward looking distance The selection module includes the path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path information and the preset first curvature ( ) and second curvature ( ), a curvature comparison unit that compares the vehicle speed (v) of the autonomous vehicle and the preset first speed ( ) and second speed ( ), and the forward looking distance based on the results compared by the speed comparison unit and the curvature comparison unit and the speed comparison unit, respectively. It may include a forward viewing distance variable unit that varies and outputs .
  • the calculation module can calculate the front wheel steering angle for a moving object with an Ackerman-type steering system using the following equation.
  • the path-following method of an autonomous moving object using an improved variable forward-looking distance includes (a) variable forward-looking distance;
  • the selection module determines the forward looking distance based on the path curvature (w) and vehicle speed (v) derived from sensing data including reference path information on the autonomous vehicle's planned route and the vehicle's speed, direction of travel, and location according to the driving environment.
  • a step of determining, (b) the target point selection module uses the reference path information and the forward looking distance Step of determining the target point from, (c)
  • the calculation module is calculated from the direction and position and the target point from step (b). Step for calculating the value, (d) steering angle
  • the forward looking distance of the calculation module from step (a) above and from step (c) above. Steering angle from value It includes calculating and outputting steering angle data.
  • step (a) the path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path information is a preset first curvature ( ), if it is greater than the above forward looking distance
  • the existing forward looking distance Shorter variable step, (a-2) where the path curvature (w) is changed to the first curvature ( ), and the first curvature ( The second curvature ( ), if it is smaller than the above forward looking distance
  • the existing forward looking distance Shorter variable step, (a-4) where the vehicle speed (v) is changed to the first speed ( ), but the first speed ( The second speed set greater than ) ( ), if it is greater than the
  • forward viewing distance may be defined by the following equation depending on the vehicle speed (v) and path curvature (w).
  • the advantage of the Pure Pursuit algorithm which is a geometric path tracking method, is that the number of parameters is significantly smaller, it is easy to apply, and it can be used in autonomous vehicles in real time.
  • the speed and reference path of autonomous vehicles are determined according to the characteristics of the forward looking distance. Path-following performance can be improved by varying the appropriate forward looking distance value by considering the curvature of the vehicle, enabling stable driving of autonomous vehicles, and it can be used in various autonomous vehicles such as cars and robots, which has the advantage of being marketable. there is.
  • Figure 1 is a block diagram showing a path-following system for an autonomous mobile vehicle using an improved variable forward looking distance, as an example according to the present invention.
  • Figure 2 is a schematic diagram illustrating an example of a method for deriving the steering angle of a vehicle.
  • Figure 3 is a flowchart showing a path tracking method for an autonomous mobile vehicle using an improved variable forward looking distance according to the present invention.
  • Figure 4 shows the path tracking method of an autonomous moving object using the improved variable front gaze distance according to the present invention, and the front gaze distance is calculated from the variable front gaze distance selection module. This is a flowchart showing how to derive .
  • FIG. 1 shows a path tracking system for an autonomous vehicle, which includes a path planning unit 10, a sensor unit 20, a receiver 30, a path tracking device 100 for the autonomous vehicle, and a steering device 40 for the behavior of the autonomous vehicle. Includes.
  • the route planning unit 10 transmits reference path information on the planned route of the autonomous vehicle to the receiving unit 30.
  • the sensor unit 20 transmits sensing data including vehicle speed, direction of travel, and location according to the driving environment to the receiver 30.
  • the sensor unit 20 is a sensor installed in the vehicle to determine the path and location of the vehicle, for example, a GPS receiver, Lidar for omnidirectional recognition of the vehicle's surrounding environment, traffic lights, vehicles, and pedestrians. It can be composed of a camera to recognize and distinguish vehicles, a radar to recognize vehicles in front and behind, and an ultrasonic sensor to recognize nearby vehicles.
  • the sensing data of the sensor unit 20 can be used as sensing data for driving the vehicle by recognizing signs, vehicles, pedestrians, traffic lights, etc. around the vehicle. Additionally, the sensing data of the sensor unit 20 can be used as sensing data to recognize and distinguish a plurality of vehicles that are stopped or running on the vehicle's set driving path.
  • the sensed data of the sensor unit 20 according to this embodiment can be used for path tracking through an analysis process of an algorithm installed in the path tracking device 100 of an autonomous mobile vehicle.
  • the receiver 30 transmits information for path tracking, such as path curvature, vehicle speed, reference path, direction of travel, and location, to the path tracking device 100 of the autonomous vehicle based on this reference path information and sensed data.
  • path tracking such as path curvature, vehicle speed, reference path, direction of travel, and location
  • the path tracking device 100 for an autonomous vehicle is provided as a computer system for vehicle control, calculates data for path tracking based on information received from the receiver 30, and outputs this to the steering device 40.
  • the path tracking device 100 for an autonomous vehicle outputs data for path tracking of an autonomous vehicle based on the Pure Pursuit algorithm.
  • the Pure Pursuit algorithm selects a target point on the planned path using an appropriate forward looking distance when the current location and direction of travel of the autonomous vehicle, that is, the vehicle, is known, and the central axis of the vehicle's rear wheels follows the target point. This is to obtain the steering angle of the front wheels.
  • the path tracking device 100 for an autonomous vehicle has a forward looking distance based on the vehicle's path curvature and vehicle speed.
  • Variable forward looking distance that determines Selection module (110), reference path and forward looking distance From the target point selection module 120, which determines the target point, the direction and location, and the target point from the target point selection module 120 yielding a value Calculation module (130) and variable forward looking distance Forward viewing distance from selection module (110) and of the calculation module 130 Steering angle from value A steering angle that calculates and outputs the calculated steering angle data to the steering device 40 side.
  • a calculation module 140 includes a calculation module 140.
  • variable forward looking distance The selection module 110 includes a curvature comparison unit 111, a speed comparison unit 112, and a forward gaze distance variable unit 113.
  • the curvature comparison unit 111 compares the path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path with preset reference data. More specifically, the curvature comparison unit 111 is configured to determine the path curvature (w) and the preset first curvature ( ) and second curvature ( ) are compared respectively.
  • the speed comparison unit 112 compares the vehicle speed (v) of the autonomous vehicle with preset reference data. More specifically, the speed comparison unit 112 determines the vehicle speed (v) of the autonomous vehicle and the preset first speed ( ) and second speed ( ) are compared respectively.
  • the forward gaze distance variable unit 113 determines the forward gaze distance based on the results compared by the curvature comparison unit 111 and the speed comparison unit 112. Variable.
  • forward looking distance is a very important parameter for path following, and is the forward looking distance.
  • the curvature through which the target point passes and the central axis of the rear wheel changes. More specifically, forward looking distance If is long, the target point is selected far away, so the curvature becomes small, and the change in steering angle is small, making the autonomous vehicle's behavior stable.
  • forward viewing distance If is excessively long large corner cutting occurs and path following performance is significantly deteriorated.
  • forward looking distance If it is short the curvature increases as the target point is closer, and you can quickly enter the path, improving path tracking performance.
  • the change in steering angle becomes severe and excessive lateral control may diverge and vehicle movement may become unstable, so it is necessary to select an appropriate forward looking distance.
  • the disclosed point is the target point.
  • the Pure Pursuit algorithm is based on the central axis of the rear wheels, which is the current position of the vehicle. Find the steering angle to select and follow the target point on the planned route using the heading.
  • steering angle In order to obtain , must be obtained.
  • the target point is can be selected.
  • Direction (Yaw Angle, ), rear wheel central axis and target point from The value is as shown in Equation 3 below.
  • Wheelbase L is a fixed value once the vehicle is determined, so it is measured and entered directly.
  • Forward viewing distance can be thought of as the driver's line of sight, and generally at high speeds the driver looks far away, so the forward looking distance is large. At low speeds, the driver looks closer, so the forward looking distance is small. It has value.
  • Variable forward looking distance that selects an appropriate forward looking distance according to the above-mentioned principles
  • the path-following device 100 for an autonomous vehicle including the selection module 110, improves path-following performance by varying the look-ahead distance according to the driving environment such as vehicle speed and curvature. It can be improved.
  • the path-following device for an autonomous vehicle receives reference path information on the planned route and sensed data including vehicle speed, direction, and location generated by detecting the surrounding environment of the autonomous vehicle from a plurality of sensors installed on the autonomous vehicle (S10 ).
  • the selection module determines the forward looking distance based on the path curvature and vehicle speed. Decide (S20).
  • the target point selection module provides reference path information and forward looking distance. Determine the target point from (S30).
  • the calculation module is based on the direction and location information and the target point of the target point selection module. Calculate the value (S40).
  • the steering angle Calculation module is variable forward looking distance Forward looking distance selected in the selection module and calculated from the calculation module Steering angle from value Calculate (S50).
  • the calculation module outputs the calculated steering angle data to the steering device to ensure stable behavior of the autonomous vehicle, thereby improving the path-following performance of the autonomous vehicle equipped with a steering device (S60).
  • Figure 4 shows the variable forward looking distance of the path tracking device 100. Forward looking distance based on path curvature and vehicle speed in selection module 110 This is a flowchart showing how to determine.
  • the selection module uses the first curvature (w) with the path curvature (w) set in advance. ), (S22), forward looking distance The existing forward looking distance decision to be shorter j) and (S28), the path curvature (w) is the first curvature ( ), but the first curvature ( The second curvature ( ), if it is smaller than (S23), the forward looking distance The existing forward looking distance Make a longer decision ( ) (S27). And the path curvature (w) is the first curvature ( ) and the second curvature ( ), the vehicle speed (v) of the autonomous vehicle is further considered for comparison.
  • variable forward looking distance The selection module 110 determines that the input path curvature (w) is a preset first curvature ( ) and the second curvature (
  • the vehicle speed (v) of an autonomous vehicle moving on a path with a curvature (w) greater than ) is set at a preset first speed ( ), if it is less than (S24), the forward looking distance
  • the existing forward looking distance Shorter decision ( ) and (S28)
  • the vehicle speed (v) is the first speed ( ), but the first speed ( The second speed set greater than ) ( ), if it is greater than (S25), the forward looking distance
  • the existing forward looking distance Make a longer decision ( ) (S27).
  • variable forward looking distance The selection module 110 sets the vehicle speed (v) to the first speed ( ) and the second speed ( ), the forward looking distance depends on the path curvature (w) and the vehicle speed (v) Decide ( ) (S26).
  • the selection module 110 determines the relationship between vehicle speed (v) and path curvature (w) and determines the forward looking distance according to Equation 4 below: Decide.
  • Equation 5 is determined as a function of vehicle speed (v) and path curvature (w), and can be defined as Equation 5 below.
  • variable forward looking distance The selection module 110 can obtain a variable forward looking distance by repeating several to tens of times per second.
  • the path tracking method of an autonomous vehicle using the improved variable forward gaze distance of the present invention uses not only the speed of the autonomous vehicle but also the curvature of the reference path to determine the forward gaze distance.
  • Path following performance can be improved by varying .
  • Forward viewing distance Appropriate forward looking distance considering the vehicle speed and curvature of the reference path depending on the characteristics of Let it have a value. If the curvature is below a certain level or exceeds a certain speed, the forward looking distance is There is an advantage in improving the driving stability of the vehicle and the tracking performance of the reference path by setting it to a relatively large value and having a small value when it exceeds a certain curvature or is below a certain speed.
  • Path planning unit 20 Sensor unit 30: Receiving unit 40: Steering device
  • Path following device 110 Variable forward looking distance Selection module 111: Curvature comparison unit 112: Speed comparison unit 113: Forward viewing distance variable unit 120: Target point selection module 130: Calculation module 140: steering angle Calculation module

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • Mechanical Engineering (AREA)
  • Remote Sensing (AREA)
  • General Physics & Mathematics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Transportation (AREA)
  • Mathematical Physics (AREA)
  • Robotics (AREA)
  • Steering Control In Accordance With Driving Conditions (AREA)
  • Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Game Theory and Decision Science (AREA)
  • Medical Informatics (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The present invention relates to a device for tracking the path of an autonomous vehicle by using an improved variable look-ahead distance and a method therefor. The device comprises: a variable look-ahead distance L d selection module for determining a look-ahead distance L d on the basis of a vehicle path curvature (w) and a vehicle velocity (v) derived from sensing data comprising the velocity, direction of movement, and position of a vehicle, which follow driving environments and reference path information on a planned path of the autonomous vehicle; a target point selection module for determining a target point from the reference path information and the look-ahead distance L d determined by the variable look-ahead distance L d selection module; an α calculation module for determining an α value from the direction of movement and the position and the target point determined by the target point selection module; and a steering angle δ calculation module for calculating a steering angle δ from the look-ahead distance L d and the α value and outputting the calculated steering angle data. Therefore, according to the present invention, an appropriate look-ahead distance L d value is provided in consideration of the current velocity of the autonomous vehicle and the curvature of the reference path according to characteristics of the look-ahead distance L d , thereby improving the vehicle's driving stability and the reference path tracking performance.

Description

개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 장치 및 그 방법Path tracking device and method for autonomous mobile vehicle using improved variable forward looking distance
본 발명은 기하학적 경로 추종 방법인 Pure Pursuit 알고리즘에서 전방주시거리를 자율이동체의 현재 속도와 기준 경로의 곡률에 따라 가변시켜 애커먼 방식의 조향장치를 가진 자율이동체의 경로추종 성능을 향상시키기 위한 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 장치와 그 방법에 관한 것이다.The present invention is an improved path-following performance of an autonomous vehicle with an Ackerman-type steering device by varying the forward looking distance in the Pure Pursuit algorithm, a geometric path-following method, according to the current speed of the autonomous vehicle and the curvature of the reference path. This relates to a path tracking device and method for an autonomous vehicle using variable forward looking distance.
일반적으로 자율이동체로서 자율주행차(Autonomous Vehicle)는 차량에 설치되거나 이미 구축된 인프라의 다양한 센서로부터 얻은 데이터를 바탕으로, 차량 스스로 도로 환경을 인지, 판단, 경로계획 및 제어를 통해 목적지까지 안전하게 주행하는 차량이다. 더욱이 자율이동체에서 주변 환경의 인지와 판단 등과 함께 경로추종은 필수적인 요소이다.In general, as an autonomous vehicle, an autonomous vehicle drives safely to its destination by recognizing, judging, route planning, and controlling the road environment on its own based on data obtained from various sensors installed in the vehicle or in already established infrastructure. It is a vehicle that does. Moreover, path following is an essential element in autonomous vehicles, along with recognition and judgment of the surrounding environment.
통상 자율주행차에서 계획된 경로를 추종(Path Tracking)하는 방법으로, 동역학적인 방법은 차량, 타이어와 지면 등에 대한 다양한 동역학들을 고려하여 우수한 추종 성능을 가지지만 복잡하고 많은 변수로 인해 실시간 적용이 어려운 실정이다. 동역학적 방법으로는 모델 예측제어(Model Predictive Control, MPC)가 있다. 또한, 기하학적인 방법은 기준 경로, 차량의 위치와 방향에 대한 기하학적인 고려만을 이용하므로 알고리즘이 단순하고 적용하기 쉬워 널리 사용되고 연구되지만, 추종 성능은 상대적으로 떨어지는 단점이 있다. Typically, it is a method of following a planned path in an autonomous vehicle (Path Tracking). The dynamic method has excellent tracking performance by considering various dynamics of the vehicle, tires, and ground, but it is difficult to apply in real time due to complex and numerous variables. am. A dynamic method is Model Predictive Control (MPC). In addition, the geometric method uses only geometric considerations for the reference path and the position and direction of the vehicle, so the algorithm is simple and easy to apply, so it is widely used and studied, but has the disadvantage of relatively poor tracking performance.
한편, 기하학적 방법으로는 Pure Pursuit 방식, Stanley 방식 등이 있다.Meanwhile, geometric methods include the Pure Pursuit method and Stanley method.
여기서, Pure Pursuit 방식은 자율이동체의 현재 위치와 진행 방향, 계획된 경로에 대한 정보를 이용하며 기하학적인 고려만으로 자율이동체의 전륜 조향각을 계산하므로 단순하고 적용하기 쉬워서 자율주행의 경로추종 방법으로 많이 사용되고 있다. 또한, 통상의 Pure Pursuit 방식에서 전방주시거리(Look-Ahead Distance)는 경로추종 성능을 결정짓는 중요한 요소로 고정된 값 또는 속도에만 비례하는 값이 주로 사용되었다. 그러나 기존의 전방주시거리로는 자율이동체 주행의 전영역에서 만족할만한 경로추종 성능을 기대하기에는 한계가 있다.Here, the Pure Pursuit method uses information about the current location, direction, and planned path of the autonomous vehicle and calculates the front wheel steering angle of the autonomous vehicle only by geometric considerations, so it is simple and easy to apply, so it is widely used as a path-following method for autonomous driving. . In addition, in the typical Pure Pursuit method, look-ahead distance is an important factor in determining path-following performance, and a fixed value or a value proportional only to speed is mainly used. However, there are limits to expecting satisfactory path-following performance in all areas of autonomous vehicle driving with the existing forward looking distance.
따라서, 자율이동체에서 경로추종 성능을 향상시킬 수 있는 방법이 요구되는 실정이다.Therefore, there is a need for a method to improve path-following performance in autonomous vehicles.
[선행기술문헌][Prior art literature]
[특허문헌][Patent Document]
(특허문헌 1) 대한민국 등록특허공보 제10-2394465호(2022.05.06., 공고)(Patent Document 1) Republic of Korea Patent Publication No. 10-2394465 (2022.05.06., notice)
(특허문헌 2) 대한민국 등록특허공보 제10-2349654호(2021.01.13., 공고)(Patent Document 2) Republic of Korea Patent Publication No. 10-2349654 (2021.01.13., notice)
(특허문헌 3) 대한민국 공개특허공보 제10-2017-0119463호(2017.10.27., 공개)(Patent Document 3) Republic of Korea Patent Publication No. 10-2017-0119463 (October 27, 2017, published)
본 발명은 상기 문제를 해결하기 위한 것으로, Pure Pursuit 알고리즘에 기반하여, 전방주시거리를 자율이동체의 현재 속도와 기준 경로의 곡률에 따라 가변시킴으로써 자율이동체의 경로 추종 성능을 향상시키는 것이 목적이다.The present invention is intended to solve the above problem. Based on the Pure Pursuit algorithm, the purpose of the present invention is to improve the path-following performance of an autonomous vehicle by varying the forward looking distance according to the current speed of the autonomous vehicle and the curvature of the reference path.
본 발명의 일 측면에 따른 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 장치는 자율이동체의 계획된 경로상의 기준 경로 정보와 주행환경에 따른 차량의 속도, 진행방향 및 위치를 포함하는 감지데이터로부터 도출된 차량의 경로 곡률(w)과 차량속도(v)를 기초로 전방주시거리
Figure PCTKR2023018036-appb-img-000001
를 결정하는 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000002
선정모듈, 상기 기준 경로 정보와 상기 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000003
선정모듈에서 결정된 전방주시거리
Figure PCTKR2023018036-appb-img-000004
로부터 목표점을 결정하는 목표점 선정모듈, 상기 진행방향 및 위치와 상기 목표점 선정모듈에서 결정된 목표점으로부터
Figure PCTKR2023018036-appb-img-000005
값을 산출하는
Figure PCTKR2023018036-appb-img-000006
계산모듈; 및 상기 전방주시거리
Figure PCTKR2023018036-appb-img-000007
와
Figure PCTKR2023018036-appb-img-000008
값으로부터 조향각
Figure PCTKR2023018036-appb-img-000009
를 산출하여 산출된 조향각 데이터를 출력하는 조향각
Figure PCTKR2023018036-appb-img-000010
산출모듈을 포함한다.The path tracking device for an autonomous vehicle using an improved variable forward looking distance according to one aspect of the present invention is derived from sensing data including reference path information on the planned route of the autonomous vehicle and the speed, direction, and location of the vehicle according to the driving environment. Forward looking distance based on the derived vehicle path curvature (w) and vehicle speed (v)
Figure PCTKR2023018036-appb-img-000001
Variable forward looking distance that determines
Figure PCTKR2023018036-appb-img-000002
Selection module, the reference path information and the variable forward looking distance
Figure PCTKR2023018036-appb-img-000003
Forward looking distance determined in the selection module
Figure PCTKR2023018036-appb-img-000004
From the target point selection module that determines the target point, the direction and location and the target point determined in the target point selection module
Figure PCTKR2023018036-appb-img-000005
yielding a value
Figure PCTKR2023018036-appb-img-000006
calculation module; and the forward viewing distance
Figure PCTKR2023018036-appb-img-000007
and
Figure PCTKR2023018036-appb-img-000008
Steering angle from the value
Figure PCTKR2023018036-appb-img-000009
Steering angle that calculates and outputs the calculated steering angle data
Figure PCTKR2023018036-appb-img-000010
Includes calculation module.
한편, 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000011
선정모듈은, 상기 기준 경로 정보를 기초로 산출된 자율이동체의 현재 위치에서의 경로 곡률(w)과 사전에 설정된 제1곡률(
Figure PCTKR2023018036-appb-img-000012
) 및 제2곡률(
Figure PCTKR2023018036-appb-img-000013
)을 각각 비교하는 곡률비교부, 자율이동체의 차량 속도(v)와 사전에 설정된 제1속도(
Figure PCTKR2023018036-appb-img-000014
) 및 제2속도(
Figure PCTKR2023018036-appb-img-000015
)을 각각 비교하는 속도비교부 및 상기 곡률비교부와 속도비교부에서 비교한 결과에 기초하여 전방주시거리
Figure PCTKR2023018036-appb-img-000016
를 가변하여 출력하는 전방주시거리 가변부를 포함할 수 있다.Meanwhile, variable forward looking distance
Figure PCTKR2023018036-appb-img-000011
The selection module includes the path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path information and the preset first curvature (
Figure PCTKR2023018036-appb-img-000012
) and second curvature (
Figure PCTKR2023018036-appb-img-000013
), a curvature comparison unit that compares the vehicle speed (v) of the autonomous vehicle and the preset first speed (
Figure PCTKR2023018036-appb-img-000014
) and second speed (
Figure PCTKR2023018036-appb-img-000015
), and the forward looking distance based on the results compared by the speed comparison unit and the curvature comparison unit and the speed comparison unit, respectively.
Figure PCTKR2023018036-appb-img-000016
It may include a forward viewing distance variable unit that varies and outputs .
또한, 조향각
Figure PCTKR2023018036-appb-img-000017
산출모듈은, 애커먼 방식의 조향시스템을 가진 이동체에 대한 앞바퀴 조향각을 다음의 수학식으로 산출할 수 있다.Also, steering angle
Figure PCTKR2023018036-appb-img-000017
The calculation module can calculate the front wheel steering angle for a moving object with an Ackerman-type steering system using the following equation.
Figure PCTKR2023018036-appb-img-000018
Figure PCTKR2023018036-appb-img-000018
본 발명의 다른 실시예에 따른 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법은, (a) 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000019
선정모듈이 자율이동체의 계획된 경로상의 기준 경로 정보와 주행환경에 따른 차량의 속도, 진행방향 및 위치를 포함하는 감지 데이터로부터 도출된 경로 곡률(w)과 차량속도(v)를 기초로 전방주시거리
Figure PCTKR2023018036-appb-img-000020
를 결정하는 단계, (b) 목표점 선정모듈이 상기 기준 경로 정보와 상기 전방주시거리
Figure PCTKR2023018036-appb-img-000021
로부터 목표점을 결정하는 단계, (c)
Figure PCTKR2023018036-appb-img-000022
계산모듈이 상기 진행방향 및 위치와 상기 (b) 단계로부터의 목표점으로부터
Figure PCTKR2023018036-appb-img-000023
값을 산출하는 단계, (d) 조향각
Figure PCTKR2023018036-appb-img-000024
산출모듈이 상기 (a) 단계로부터의 전방주시거리
Figure PCTKR2023018036-appb-img-000025
와 상기 (c)단계로부터의
Figure PCTKR2023018036-appb-img-000026
값으로부터 조향각
Figure PCTKR2023018036-appb-img-000027
를 산출하여 조향각 데이터를 출력하는 단계를 포함한다. The path-following method of an autonomous moving object using an improved variable forward-looking distance according to another embodiment of the present invention includes (a) variable forward-looking distance;
Figure PCTKR2023018036-appb-img-000019
The selection module determines the forward looking distance based on the path curvature (w) and vehicle speed (v) derived from sensing data including reference path information on the autonomous vehicle's planned route and the vehicle's speed, direction of travel, and location according to the driving environment.
Figure PCTKR2023018036-appb-img-000020
A step of determining, (b) the target point selection module uses the reference path information and the forward looking distance
Figure PCTKR2023018036-appb-img-000021
Step of determining the target point from, (c)
Figure PCTKR2023018036-appb-img-000022
The calculation module is calculated from the direction and position and the target point from step (b).
Figure PCTKR2023018036-appb-img-000023
Step for calculating the value, (d) steering angle
Figure PCTKR2023018036-appb-img-000024
The forward looking distance of the calculation module from step (a) above
Figure PCTKR2023018036-appb-img-000025
and from step (c) above.
Figure PCTKR2023018036-appb-img-000026
Steering angle from value
Figure PCTKR2023018036-appb-img-000027
It includes calculating and outputting steering angle data.
한편, (a) 단계는, (a-1) 상기 기준 경로 정보를 기초로 산출된 자율이동체의 현재위치에서의 경로 곡률(w)이 사전에 설정된 제1곡률(
Figure PCTKR2023018036-appb-img-000028
)보다 크면 상기 전방주시거리
Figure PCTKR2023018036-appb-img-000029
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000030
보다 짧게 가변하는 단계, (a-2) 상기 경로 곡률(w)이 제1곡률(
Figure PCTKR2023018036-appb-img-000031
)보다 작고, 상기 제1곡률(
Figure PCTKR2023018036-appb-img-000032
)보다 작게 설정된 제2곡률(
Figure PCTKR2023018036-appb-img-000033
)보다 더 작으면 상기 전방주시거리
Figure PCTKR2023018036-appb-img-000034
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000035
보다 길게 가변하는 단계, (a-3) 상기 경로 곡률(w)이 제1곡률(
Figure PCTKR2023018036-appb-img-000036
)보다 작고 제2곡률(
Figure PCTKR2023018036-appb-img-000037
)보다 크면 자율이동체의 차량 속도(v)를 더 고려하되, 상기 차량 속도(v)가 사전에 설정된 제1속도(
Figure PCTKR2023018036-appb-img-000038
)보다 작으면 상기 전방주시거리
Figure PCTKR2023018036-appb-img-000039
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000040
보다 짧게 가변하는 단계, (a-4) 상기 차량 속도(v)가 제1속도(
Figure PCTKR2023018036-appb-img-000041
)보다 크되, 제1속도(
Figure PCTKR2023018036-appb-img-000042
)보다 크게 설정된 제2속도(
Figure PCTKR2023018036-appb-img-000043
)보다 더 크면 상기 전방주시거리
Figure PCTKR2023018036-appb-img-000044
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000045
보다 길게 가변하는 단계, (a-5) 상기 차량 속도(v)가 제1속도(
Figure PCTKR2023018036-appb-img-000046
)보다 크고 제2속도(
Figure PCTKR2023018036-appb-img-000047
)보다 작으면 상기 경로 곡률(w)과 상기 차량 속도(v)에 따라 전방주시거리
Figure PCTKR2023018036-appb-img-000048
를 가변하는 단계를 포함할 수 있다.Meanwhile, in step (a), (a-1) the path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path information is a preset first curvature (
Figure PCTKR2023018036-appb-img-000028
), if it is greater than the above forward looking distance
Figure PCTKR2023018036-appb-img-000029
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000030
Shorter variable step, (a-2) where the path curvature (w) is changed to the first curvature (
Figure PCTKR2023018036-appb-img-000031
), and the first curvature (
Figure PCTKR2023018036-appb-img-000032
The second curvature (
Figure PCTKR2023018036-appb-img-000033
), if it is smaller than the above forward looking distance
Figure PCTKR2023018036-appb-img-000034
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000035
Longer variable step, (a-3) where the path curvature (w) is changed to the first curvature (
Figure PCTKR2023018036-appb-img-000036
) and the second curvature (
Figure PCTKR2023018036-appb-img-000037
), the vehicle speed (v) of the autonomous vehicle is further considered, but the vehicle speed (v) is the preset first speed (
Figure PCTKR2023018036-appb-img-000038
), if it is less than the above forward looking distance
Figure PCTKR2023018036-appb-img-000039
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000040
Shorter variable step, (a-4) where the vehicle speed (v) is changed to the first speed (
Figure PCTKR2023018036-appb-img-000041
), but the first speed (
Figure PCTKR2023018036-appb-img-000042
The second speed set greater than ) (
Figure PCTKR2023018036-appb-img-000043
), if it is greater than the above forward looking distance
Figure PCTKR2023018036-appb-img-000044
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000045
Longer variable step, (a-5) where the vehicle speed (v) is the first speed (
Figure PCTKR2023018036-appb-img-000046
) and the second speed (
Figure PCTKR2023018036-appb-img-000047
), the forward looking distance depends on the path curvature (w) and the vehicle speed (v).
Figure PCTKR2023018036-appb-img-000048
It may include a step of varying.
또한, 전방주시거리
Figure PCTKR2023018036-appb-img-000049
는, 차량 속도(v) 및 경로 곡률(w) 에 따라 다음의 수학식으로 정의되는 것일 수 있다.Also, forward viewing distance
Figure PCTKR2023018036-appb-img-000049
may be defined by the following equation depending on the vehicle speed (v) and path curvature (w).
Figure PCTKR2023018036-appb-img-000050
Figure PCTKR2023018036-appb-img-000050
여기서,
Figure PCTKR2023018036-appb-img-000051
는 차량 속도(v) 및 경로 곡률(w)의 함수로 결정하되, 다음의 수학식으로 정의될 수 있다.here,
Figure PCTKR2023018036-appb-img-000051
is determined as a function of vehicle speed (v) and path curvature (w), and can be defined by the following equation.
Figure PCTKR2023018036-appb-img-000052
Figure PCTKR2023018036-appb-img-000052
본 발명에 따르면, 기하학적 경로 추종 방법인 Pure Pursuit 알고리즘의 장점인 파라미터의 수가 현저히 적고 적용이 쉬우며 실시간으로 자율이동체에서 사용할 수 있을 뿐만 아니라, 전방주시거리의 특성에 따라 자율이동체의 속도와 기준 경로의 곡률을 고려하여 적절한 전방주시거리 값을 가지도록 가변함으로써 경로추종 성능을 향상시킬 수 있어 자율이동체의 안정된 주행이 가능하고, 자동차, 로봇 등 다양한 자율이동체에 활용할 수 있어 시장성을 기대할 수 있는 이점이 있다.According to the present invention, the advantage of the Pure Pursuit algorithm, which is a geometric path tracking method, is that the number of parameters is significantly smaller, it is easy to apply, and it can be used in autonomous vehicles in real time. In addition, the speed and reference path of autonomous vehicles are determined according to the characteristics of the forward looking distance. Path-following performance can be improved by varying the appropriate forward looking distance value by considering the curvature of the vehicle, enabling stable driving of autonomous vehicles, and it can be used in various autonomous vehicles such as cars and robots, which has the advantage of being marketable. there is.
도 1은 본 발명에 따른 실시 예로, 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 시스템을 나타낸 블록도이다.Figure 1 is a block diagram showing a path-following system for an autonomous mobile vehicle using an improved variable forward looking distance, as an example according to the present invention.
도 2는 차량의 조향각을 유도하는 방법을 예시적으로 나타낸 개략도이다.Figure 2 is a schematic diagram illustrating an example of a method for deriving the steering angle of a vehicle.
도 3은 본 발명에 따른 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법을 나타낸 흐름도이다.Figure 3 is a flowchart showing a path tracking method for an autonomous mobile vehicle using an improved variable forward looking distance according to the present invention.
도 4는 본 발명에 따른 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법으로, 가변 전방주시거리 선정모듈에서 전방주시거리
Figure PCTKR2023018036-appb-img-000053
를 도출하는 방법을 나타낸 흐름도이다.Figure 4 shows the path tracking method of an autonomous moving object using the improved variable front gaze distance according to the present invention, and the front gaze distance is calculated from the variable front gaze distance selection module.
Figure PCTKR2023018036-appb-img-000053
This is a flowchart showing how to derive .
이하, 본 발명에 따른 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 장치와 그 방법에 관한 실시 예를 첨부된 도면을 참조하여 상세히 설명한다.Hereinafter, embodiments of the path tracking device and method for an autonomous vehicle using the improved variable forward viewing distance according to the present invention will be described in detail with reference to the attached drawings.
도 1은 자율이동체의 경로추종 시스템으로, 경로계획부(10), 센서부(20), 수신부(30), 자율이동체의 경로추종 장치(100), 자율이동체의 거동을 위한 조향장치(40)를 포함한다.1 shows a path tracking system for an autonomous vehicle, which includes a path planning unit 10, a sensor unit 20, a receiver 30, a path tracking device 100 for the autonomous vehicle, and a steering device 40 for the behavior of the autonomous vehicle. Includes.
경로계획부(10)는 수신부(30) 측으로 자율이동체의 계획된 경로상의 기준경로 정보를 전송한다.The route planning unit 10 transmits reference path information on the planned route of the autonomous vehicle to the receiving unit 30.
센서부(20)는 수신부(30) 측으로 주행환경에 따른 차량속도, 진행방향, 위치를 포함하는 감지 데이터를 전송한다.The sensor unit 20 transmits sensing data including vehicle speed, direction of travel, and location according to the driving environment to the receiver 30.
이때, 센서부(20)는 차량의 경로와 위치를 판단하기 위하여 차량에 설치되는 센서들로서, 예컨대, GPS수신기, 차량의 주변 환경을 전방위로 인식하기 위한 라이다(Lidar), 신호등, 차량, 보행자 등을 인식하고 분별하기 위한 카메라, 전방과 후방의 차량을 인식하기 위한 레이더, 근접 차량을 인식하기 위한 초음파센서 등으로 구성될 수 있다. 이러한 센서부(20)의 감지데이터는 차량 주변의 표지판, 차량, 보행자, 신호등 등을 인식하여 차량의 주행을 위한 감지데이터로 활용될 수 있다. 또한, 센서부(20)의 감지데이터는 차량의 설정된 주행 경로상에 정차하거나 주행하고 있는 복수의 차량을 인식하고 분별하기 위한 감지데이터로서 활용될 수 있다. 특히, 본 실시예에 따른 센서부(20)의 감지데이터는 자율이동체의 경로추종 장치(100)에 설치된 알고리즘의 분석과정을 거쳐 경로 추종에 사용될 수 있다.At this time, the sensor unit 20 is a sensor installed in the vehicle to determine the path and location of the vehicle, for example, a GPS receiver, Lidar for omnidirectional recognition of the vehicle's surrounding environment, traffic lights, vehicles, and pedestrians. It can be composed of a camera to recognize and distinguish vehicles, a radar to recognize vehicles in front and behind, and an ultrasonic sensor to recognize nearby vehicles. The sensing data of the sensor unit 20 can be used as sensing data for driving the vehicle by recognizing signs, vehicles, pedestrians, traffic lights, etc. around the vehicle. Additionally, the sensing data of the sensor unit 20 can be used as sensing data to recognize and distinguish a plurality of vehicles that are stopped or running on the vehicle's set driving path. In particular, the sensed data of the sensor unit 20 according to this embodiment can be used for path tracking through an analysis process of an algorithm installed in the path tracking device 100 of an autonomous mobile vehicle.
수신부(30)는 이러한 기준경로정보와 감지데이터를 기초로 경로곡률, 차량 속도, 기준경로, 진행방향, 위치 등의 경로추종을 위한 정보를 자율이동체의 경로추종 장치(100) 측으로 전달한다.The receiver 30 transmits information for path tracking, such as path curvature, vehicle speed, reference path, direction of travel, and location, to the path tracking device 100 of the autonomous vehicle based on this reference path information and sensed data.
자율이동체의 경로추종 장치(100)는 차량 제어용 컴퓨터 시스템으로 마련되어, 수신부(30)로부터 수신한 정보를 기초로 경로 추종을 위한 데이터를 산출하고, 이를 조향장치(40) 측에 출력한다. The path tracking device 100 for an autonomous vehicle is provided as a computer system for vehicle control, calculates data for path tracking based on information received from the receiver 30, and outputs this to the steering device 40.
특히, 본 실시예에 따른 자율이동체의 경로추종 장치(100)는 Pure Pursuit 알고리즘을 기반으로 자율이동체의 경로추종을 위한 데이터를 출력한다. In particular, the path tracking device 100 for an autonomous vehicle according to this embodiment outputs data for path tracking of an autonomous vehicle based on the Pure Pursuit algorithm.
여기서, Pure Pursuit 알고리즘은 자율이동체, 즉, 차량의 현재 위치와 진행 방향을 알고 있을 때, 적절한 전방주시거리를 이용하여 계획된 경로상에 목표점을 선정하고, 차량의 뒷바퀴 중심축이 목표점을 추종하기 위한 앞바퀴의 조향각을 얻을 수 있도록 하는 것이다.Here, the Pure Pursuit algorithm selects a target point on the planned path using an appropriate forward looking distance when the current location and direction of travel of the autonomous vehicle, that is, the vehicle, is known, and the central axis of the vehicle's rear wheels follows the target point. This is to obtain the steering angle of the front wheels.
따라서, 본 실시예에 따른 자율이동체의 경로추종 장치(100)는 차량의 경로곡률과 차량속도로부터 전방주시거리
Figure PCTKR2023018036-appb-img-000054
를 결정하는 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000055
선정모듈(110), 기준 경로와 전방주시거리
Figure PCTKR2023018036-appb-img-000056
로부터 목표점을 결정하는 목표점 선정모듈(120), 진행방향 및 위치와 목표점 선정모듈(120)로부터의 목표점으로부터
Figure PCTKR2023018036-appb-img-000057
값을 산출하는
Figure PCTKR2023018036-appb-img-000058
계산모듈(130) 및 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000059
선정모듈(110)로부터의 전방주시거리
Figure PCTKR2023018036-appb-img-000060
와
Figure PCTKR2023018036-appb-img-000061
계산모듈(130)의
Figure PCTKR2023018036-appb-img-000062
값으로부터 조향각
Figure PCTKR2023018036-appb-img-000063
를 산출하고, 조향장치(40) 측으로 산출된 조향각 데이터를 출력하는 조향각
Figure PCTKR2023018036-appb-img-000064
산출모듈(140)를 포함한다.Therefore, the path tracking device 100 for an autonomous vehicle according to this embodiment has a forward looking distance based on the vehicle's path curvature and vehicle speed.
Figure PCTKR2023018036-appb-img-000054
Variable forward looking distance that determines
Figure PCTKR2023018036-appb-img-000055
Selection module (110), reference path and forward looking distance
Figure PCTKR2023018036-appb-img-000056
From the target point selection module 120, which determines the target point, the direction and location, and the target point from the target point selection module 120
Figure PCTKR2023018036-appb-img-000057
yielding a value
Figure PCTKR2023018036-appb-img-000058
Calculation module (130) and variable forward looking distance
Figure PCTKR2023018036-appb-img-000059
Forward viewing distance from selection module (110)
Figure PCTKR2023018036-appb-img-000060
and
Figure PCTKR2023018036-appb-img-000061
of the calculation module 130
Figure PCTKR2023018036-appb-img-000062
Steering angle from value
Figure PCTKR2023018036-appb-img-000063
A steering angle that calculates and outputs the calculated steering angle data to the steering device 40 side.
Figure PCTKR2023018036-appb-img-000064
Includes a calculation module 140.
보다 구체적으로, 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000065
선정모듈(110)은 곡률비교부(111), 속도비교부(112) 및 전방주시거리 가변부(113)을 포함한다.More specifically, variable forward looking distance
Figure PCTKR2023018036-appb-img-000065
The selection module 110 includes a curvature comparison unit 111, a speed comparison unit 112, and a forward gaze distance variable unit 113.
먼저, 곡률비교부(111) 는 기준경로를 기초로 산출된 자율이동체의 현재 위치에서의 경로 곡률(w)과 사전에 설정된 기준데이터를 비교한다. 보다 구체적으로, 곡률비교부(111)는 경로 곡률(w)과 사전에 설정된 제1곡률(
Figure PCTKR2023018036-appb-img-000066
) 및 제2곡률(
Figure PCTKR2023018036-appb-img-000067
)을 각각 비교한다. First, the curvature comparison unit 111 compares the path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path with preset reference data. More specifically, the curvature comparison unit 111 is configured to determine the path curvature (w) and the preset first curvature (
Figure PCTKR2023018036-appb-img-000066
) and second curvature (
Figure PCTKR2023018036-appb-img-000067
) are compared respectively.
또한, 속도비교부(112)는 자율이동체의 차량 속도(v)와 사전에 설정된 기준데이터를 비교한다. 보다 구체적으로, 속도비교부(112)는 자율이동체의 차량 속도(v)와 사전에 설정된 제1속도(
Figure PCTKR2023018036-appb-img-000068
) 및 제2속도(
Figure PCTKR2023018036-appb-img-000069
)를 각각 비교한다. Additionally, the speed comparison unit 112 compares the vehicle speed (v) of the autonomous vehicle with preset reference data. More specifically, the speed comparison unit 112 determines the vehicle speed (v) of the autonomous vehicle and the preset first speed (
Figure PCTKR2023018036-appb-img-000068
) and second speed (
Figure PCTKR2023018036-appb-img-000069
) are compared respectively.
전방주시거리 가변부(113)는 곡률비교부(111)와 속도비교부(112)에서 비교한 결과에 기초하여 전방주시거리
Figure PCTKR2023018036-appb-img-000070
를 가변한다.The forward gaze distance variable unit 113 determines the forward gaze distance based on the results compared by the curvature comparison unit 111 and the speed comparison unit 112.
Figure PCTKR2023018036-appb-img-000070
Variable.
이때, 전방주시거리
Figure PCTKR2023018036-appb-img-000071
는 경로추종을 위한 아주 중요한 파라미터로서, 전방주시거리
Figure PCTKR2023018036-appb-img-000072
가 길거나 짧음에 따라 목표점과 뒷바퀴 중심축이 지나는 곡률이 달라진다. 보다 구체적으로, 전방주시거리
Figure PCTKR2023018036-appb-img-000073
가 길면, 목표점이 멀리 선정되므로 곡률이 작아지게 되고, 조향각의 변화가 작아 자율이동체의 거동(Maneuver)이 안정적이게 된다. 그러나, 전방주시거리
Figure PCTKR2023018036-appb-img-000074
가 과도하게 긴 경우에는 큰 코너 커팅(Corner Cutting)이 발생하여 경로 추종 성능이 현저히 저하된다. 또한, 전방주시거리
Figure PCTKR2023018036-appb-img-000075
가 짧으면 목표점이 가까워 곡률이 커지게 되고, 빠르게 경로상에 들어올 수 있어 경로 추종 성능을 향상시킬 수 있다. 그러나 조향각의 변화가 심해지고 과도한 경우에는 횡방향 제어가 발산하여 차량의 움직임이 불안정해질 수 있는바, 적절한 전방주시거리의 선정이 필요하다.At this time, forward looking distance
Figure PCTKR2023018036-appb-img-000071
is a very important parameter for path following, and is the forward looking distance.
Figure PCTKR2023018036-appb-img-000072
Depending on how long or short it is, the curvature through which the target point passes and the central axis of the rear wheel changes. More specifically, forward looking distance
Figure PCTKR2023018036-appb-img-000073
If is long, the target point is selected far away, so the curvature becomes small, and the change in steering angle is small, making the autonomous vehicle's behavior stable. However, forward viewing distance
Figure PCTKR2023018036-appb-img-000074
If is excessively long, large corner cutting occurs and path following performance is significantly deteriorated. Also, forward looking distance
Figure PCTKR2023018036-appb-img-000075
If it is short, the curvature increases as the target point is closer, and you can quickly enter the path, improving path tracking performance. However, if the change in steering angle becomes severe and excessive, lateral control may diverge and vehicle movement may become unstable, so it is necessary to select an appropriate forward looking distance.
이하에서는, 도 2를 통해 적절한 전방주시거리를 선정하기 위한 방법을 설명한다.Below, a method for selecting an appropriate forward looking distance will be described with reference to FIG. 2.
도 2를 참조하면, 개시된 점
Figure PCTKR2023018036-appb-img-000076
은 목표점이다. Pure Pursuit 알고리즘은 차량의 현재 위치인 뒷바퀴의 중심축
Figure PCTKR2023018036-appb-img-000077
과 진행방향(Heading)을 이용하여 계획된 경로상의 목표점을 선정 및 추종하기 위한 조향각을 찾는다.Referring to Figure 2, the disclosed point
Figure PCTKR2023018036-appb-img-000076
is the target point. The Pure Pursuit algorithm is based on the central axis of the rear wheels, which is the current position of the vehicle.
Figure PCTKR2023018036-appb-img-000077
Find the steering angle to select and follow the target point on the planned route using the heading.
상기 에커먼 방식에서의 앞바퀴 조향각
Figure PCTKR2023018036-appb-img-000078
는
Figure PCTKR2023018036-appb-img-000079
이다.그리고 회전반경 R을 구하기 위하여 도 2 의 삼각형 OGP에 사인법칙을 적용하여 R 에 대하여 정리하면, 다음의 수학식 1과 같다.Front wheel steering angle in the Ackerman method
Figure PCTKR2023018036-appb-img-000078
Is
Figure PCTKR2023018036-appb-img-000079
And in order to find the radius of rotation R, apply the sine law to the triangle OGP of FIG. 2 and summarize R, which is as follows:
Figure PCTKR2023018036-appb-img-000080
Figure PCTKR2023018036-appb-img-000080
여기서,
Figure PCTKR2023018036-appb-img-000081
는 목표점과 뒷차축의 중심점을 이은 선분과 이동체의 진행방향 사이의 각도이다.here,
Figure PCTKR2023018036-appb-img-000081
is the angle between the line segment connecting the target point and the center point of the rear axle and the moving direction of the moving object.
상기 에커먼 방식에서 유도한 조향각
Figure PCTKR2023018036-appb-img-000082
에 상기 수학식 1에서 구한 식을 대입하면, 앞바퀴의 조향각은 다음의 수학식 2와 같다.Steering angle derived from the Ackerman method
Figure PCTKR2023018036-appb-img-000082
Substituting the equation obtained in Equation 1 above, the steering angle of the front wheel is equal to the following Equation 2.
Figure PCTKR2023018036-appb-img-000083
Figure PCTKR2023018036-appb-img-000083
여기서, 조향각
Figure PCTKR2023018036-appb-img-000084
을 구하기 위해서는 먼저
Figure PCTKR2023018036-appb-img-000085
,
Figure PCTKR2023018036-appb-img-000086
,
Figure PCTKR2023018036-appb-img-000087
을 구해야 한다.Here, steering angle
Figure PCTKR2023018036-appb-img-000084
In order to obtain
Figure PCTKR2023018036-appb-img-000085
,
Figure PCTKR2023018036-appb-img-000086
,
Figure PCTKR2023018036-appb-img-000087
must be obtained.
조향각
Figure PCTKR2023018036-appb-img-000088
을 구하는 과정으로, 먼저, 전방주시거리
Figure PCTKR2023018036-appb-img-000089
가 결정되면, 목표점
Figure PCTKR2023018036-appb-img-000090
을 선정할 수 있다. 진행방향(Yaw Angle,
Figure PCTKR2023018036-appb-img-000091
), 뒷바퀴 중심축
Figure PCTKR2023018036-appb-img-000092
과 목표점
Figure PCTKR2023018036-appb-img-000093
으로부터
Figure PCTKR2023018036-appb-img-000094
값은 다음의 수학식 3과 같다.steering angle
Figure PCTKR2023018036-appb-img-000088
In the process of finding , first, the forward looking distance
Figure PCTKR2023018036-appb-img-000089
Once determined, the target point is
Figure PCTKR2023018036-appb-img-000090
can be selected. Direction (Yaw Angle,
Figure PCTKR2023018036-appb-img-000091
), rear wheel central axis
Figure PCTKR2023018036-appb-img-000092
and target point
Figure PCTKR2023018036-appb-img-000093
from
Figure PCTKR2023018036-appb-img-000094
The value is as shown in Equation 3 below.
Figure PCTKR2023018036-appb-img-000095
Figure PCTKR2023018036-appb-img-000095
축거 L은 차량이 정해지면 고정된 값이므로 직접 측정하여 입력한다.Wheelbase L is a fixed value once the vehicle is determined, so it is measured and entered directly.
전방주시거리
Figure PCTKR2023018036-appb-img-000096
는 운전자의 시선으로 생각할 수 있으며, 일반적으로 고속에서 운전자는 시선을 멀리 두므로 큰 전방주시거리
Figure PCTKR2023018036-appb-img-000097
값을 가지고, 저속에서는 운전자가 시선을 가까이 두므로 작은 전방주시거리
Figure PCTKR2023018036-appb-img-000098
값을 가진다. Forward viewing distance
Figure PCTKR2023018036-appb-img-000096
can be thought of as the driver's line of sight, and generally at high speeds the driver looks far away, so the forward looking distance is large.
Figure PCTKR2023018036-appb-img-000097
At low speeds, the driver looks closer, so the forward looking distance is small.
Figure PCTKR2023018036-appb-img-000098
It has value.
상술한 원리에 따라 적절한 전방주시거리를 선정하는 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000099
선정모듈(110)을 포함한 본 실시예에 따른 자율이동체의 경로추종 장치(100)는, 전방주시거리(Look-Ahead Distance)를 차량의 속도, 곡률 등의 주행환경에 따라 가변시킴으로써 경로추종 성능을 향상시킬 수 있다. Variable forward looking distance that selects an appropriate forward looking distance according to the above-mentioned principles
Figure PCTKR2023018036-appb-img-000099
The path-following device 100 for an autonomous vehicle according to this embodiment, including the selection module 110, improves path-following performance by varying the look-ahead distance according to the driving environment such as vehicle speed and curvature. It can be improved.
다음으로, 본 발명에 따른 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법을 도 3의 흐름도를 참조하여 설명한다.Next, the path tracking method of an autonomous mobile vehicle using the improved variable forward looking distance according to the present invention will be described with reference to the flowchart of FIG. 3.
자율이동체의 경로추종 장치는 계획된 경로상의 기준 경로 정보와 자율이동체에 설치된 복수의 센서들로부터 자율이동체의 주변 환경을 감지하여 생성된 차량속도, 진행방향 및 위치를 포함하는 감지데이터를 수신한다(S10).The path-following device for an autonomous vehicle receives reference path information on the planned route and sensed data including vehicle speed, direction, and location generated by detecting the surrounding environment of the autonomous vehicle from a plurality of sensors installed on the autonomous vehicle (S10 ).
가변 전방주시거리
Figure PCTKR2023018036-appb-img-000100
선정모듈은 경로 곡률과 차량 속도로부터 전방주시거리
Figure PCTKR2023018036-appb-img-000101
를 결정한다(S20). Variable forward looking distance
Figure PCTKR2023018036-appb-img-000100
The selection module determines the forward looking distance based on the path curvature and vehicle speed.
Figure PCTKR2023018036-appb-img-000101
Decide (S20).
그리고 목표점 선정모듈은 기준 경로 정보와 전방주시거리
Figure PCTKR2023018036-appb-img-000102
로부터 목표점을 결정한다(S30). And the target point selection module provides reference path information and forward looking distance.
Figure PCTKR2023018036-appb-img-000102
Determine the target point from (S30).
Figure PCTKR2023018036-appb-img-000103
계산모듈은 진행방향 및 위치 정보와 목표점 선정모듈의 목표점으로부터
Figure PCTKR2023018036-appb-img-000104
값을 산출한다(S40).
Figure PCTKR2023018036-appb-img-000103
The calculation module is based on the direction and location information and the target point of the target point selection module.
Figure PCTKR2023018036-appb-img-000104
Calculate the value (S40).
따라서 조향각
Figure PCTKR2023018036-appb-img-000105
산출모듈은 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000106
선정모듈에서 선정된 전방주시거리
Figure PCTKR2023018036-appb-img-000107
와
Figure PCTKR2023018036-appb-img-000108
계산모듈에서 산출된
Figure PCTKR2023018036-appb-img-000109
값으로부터 조향각
Figure PCTKR2023018036-appb-img-000110
를 산출한다(S50). Therefore, the steering angle
Figure PCTKR2023018036-appb-img-000105
Calculation module is variable forward looking distance
Figure PCTKR2023018036-appb-img-000106
Forward looking distance selected in the selection module
Figure PCTKR2023018036-appb-img-000107
and
Figure PCTKR2023018036-appb-img-000108
calculated from the calculation module
Figure PCTKR2023018036-appb-img-000109
Steering angle from value
Figure PCTKR2023018036-appb-img-000110
Calculate (S50).
그리고 조향각
Figure PCTKR2023018036-appb-img-000111
산출모듈은 산출된 조향각 데이터를 조향장치로 출력하여 자율이동체의 안정적인 거동이 이루어질 수 있도록 함으로써, 조향장치를 가진 자율이동체의 경로추종 성능을 향상시킬 수 있다(S60).and steering angle
Figure PCTKR2023018036-appb-img-000111
The calculation module outputs the calculated steering angle data to the steering device to ensure stable behavior of the autonomous vehicle, thereby improving the path-following performance of the autonomous vehicle equipped with a steering device (S60).
또한, 도 4는 경로추종 장치(100)의 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000112
선정모듈(110)에서 경로 곡률과 차량 속도를 기반으로 전방주시거리
Figure PCTKR2023018036-appb-img-000113
를 결정하는 방법을 나타낸 흐름도이다.In addition, Figure 4 shows the variable forward looking distance of the path tracking device 100.
Figure PCTKR2023018036-appb-img-000112
Forward looking distance based on path curvature and vehicle speed in selection module 110
Figure PCTKR2023018036-appb-img-000113
This is a flowchart showing how to determine.
자율이동체가 이동 중인 현재 위치를 기준으로 자율이동체가 이동하는 경로 곡률(w)과 차량 속도(v)를 실시간으로 수신한다(S21).Based on the current location where the autonomous vehicle is moving, the path curvature (w) and vehicle speed (v) along which the autonomous vehicle moves are received in real time (S21).
가변 전방주시거리
Figure PCTKR2023018036-appb-img-000114
선정모듈은 경로 곡률(w)이 사전에 설정된 제1곡률(
Figure PCTKR2023018036-appb-img-000115
)보다 크면(S22), 전방주시거리
Figure PCTKR2023018036-appb-img-000116
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000117
보다 짧게 결정
Figure PCTKR2023018036-appb-img-000118
j)하고(S28), 경로 곡률(w)이 제1곡률(
Figure PCTKR2023018036-appb-img-000119
)보다 작되, 제1곡률(
Figure PCTKR2023018036-appb-img-000120
)보다 작게 설정된 제2곡률(
Figure PCTKR2023018036-appb-img-000121
)보다 더 작으면(S23), 전방주시거리
Figure PCTKR2023018036-appb-img-000122
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000123
보다 길게 결정(
Figure PCTKR2023018036-appb-img-000124
)한다(S27). 그리고 경로 곡률(w)이 제1곡률(
Figure PCTKR2023018036-appb-img-000125
)보다 작고 제2곡률(
Figure PCTKR2023018036-appb-img-000126
)보다 크면 자율이동체의 차량 속도(v)를 더 고려하여 비교한다.Variable forward looking distance
Figure PCTKR2023018036-appb-img-000114
The selection module uses the first curvature (w) with the path curvature (w) set in advance.
Figure PCTKR2023018036-appb-img-000115
), (S22), forward looking distance
Figure PCTKR2023018036-appb-img-000116
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000117
decision to be shorter
Figure PCTKR2023018036-appb-img-000118
j) and (S28), the path curvature (w) is the first curvature (
Figure PCTKR2023018036-appb-img-000119
), but the first curvature (
Figure PCTKR2023018036-appb-img-000120
The second curvature (
Figure PCTKR2023018036-appb-img-000121
), if it is smaller than (S23), the forward looking distance
Figure PCTKR2023018036-appb-img-000122
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000123
Make a longer decision (
Figure PCTKR2023018036-appb-img-000124
) (S27). And the path curvature (w) is the first curvature (
Figure PCTKR2023018036-appb-img-000125
) and the second curvature (
Figure PCTKR2023018036-appb-img-000126
), the vehicle speed (v) of the autonomous vehicle is further considered for comparison.
또한, 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000127
선정모듈(110)은 입력된 경로 곡률(w)이 사전에 설정된 제1곡률(
Figure PCTKR2023018036-appb-img-000128
)보다 작고 제2곡률(
Figure PCTKR2023018036-appb-img-000129
)보다 큰 곡률(w)의 경로를 이동하는 자율이동체의 차량 속도(v)가 사전에 설정된 제1속도(
Figure PCTKR2023018036-appb-img-000130
)보다 작으면(S24), 전방주시거리
Figure PCTKR2023018036-appb-img-000131
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000132
보다 짧게 결정(
Figure PCTKR2023018036-appb-img-000133
)하고(S28), 차량 속도(v)가 제1속도(
Figure PCTKR2023018036-appb-img-000134
)보다 크되, 제1속도(
Figure PCTKR2023018036-appb-img-000135
)보다 크게 설정된 제2속도(
Figure PCTKR2023018036-appb-img-000136
)보다 더 크면(S25), 전방주시거리
Figure PCTKR2023018036-appb-img-000137
를 기존 전방주시거리
Figure PCTKR2023018036-appb-img-000138
보다 길게 결정(
Figure PCTKR2023018036-appb-img-000139
)한다(S27).Also, variable forward looking distance
Figure PCTKR2023018036-appb-img-000127
The selection module 110 determines that the input path curvature (w) is a preset first curvature (
Figure PCTKR2023018036-appb-img-000128
) and the second curvature (
Figure PCTKR2023018036-appb-img-000129
The vehicle speed (v) of an autonomous vehicle moving on a path with a curvature (w) greater than ) is set at a preset first speed (
Figure PCTKR2023018036-appb-img-000130
), if it is less than (S24), the forward looking distance
Figure PCTKR2023018036-appb-img-000131
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000132
Shorter decision (
Figure PCTKR2023018036-appb-img-000133
) and (S28), the vehicle speed (v) is the first speed (
Figure PCTKR2023018036-appb-img-000134
), but the first speed (
Figure PCTKR2023018036-appb-img-000135
The second speed set greater than ) (
Figure PCTKR2023018036-appb-img-000136
), if it is greater than (S25), the forward looking distance
Figure PCTKR2023018036-appb-img-000137
The existing forward looking distance
Figure PCTKR2023018036-appb-img-000138
Make a longer decision (
Figure PCTKR2023018036-appb-img-000139
) (S27).
또한, 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000140
선정모듈(110)은 차량 속도(v)가 제1속도(
Figure PCTKR2023018036-appb-img-000141
)보다 크고 제2속도(
Figure PCTKR2023018036-appb-img-000142
)보다 작으면, 경로 곡률(w)과 상기 차량 속도(v)에 따라 전방주시거리
Figure PCTKR2023018036-appb-img-000143
를 결정(
Figure PCTKR2023018036-appb-img-000144
)한다(S26).Also, variable forward looking distance
Figure PCTKR2023018036-appb-img-000140
The selection module 110 sets the vehicle speed (v) to the first speed (
Figure PCTKR2023018036-appb-img-000141
) and the second speed (
Figure PCTKR2023018036-appb-img-000142
), the forward looking distance depends on the path curvature (w) and the vehicle speed (v)
Figure PCTKR2023018036-appb-img-000143
Decide (
Figure PCTKR2023018036-appb-img-000144
) (S26).
가변 전방주시거리
Figure PCTKR2023018036-appb-img-000145
선정모듈(110)은 차량 속도(v) 및 경로 곡률(w)의 관계를 판단하여 다음의 수학식 4 에 따라 전방주시거리
Figure PCTKR2023018036-appb-img-000146
를 결정한다.Variable forward looking distance
Figure PCTKR2023018036-appb-img-000145
The selection module 110 determines the relationship between vehicle speed (v) and path curvature (w) and determines the forward looking distance according to Equation 4 below:
Figure PCTKR2023018036-appb-img-000146
Decide.
Figure PCTKR2023018036-appb-img-000147
Figure PCTKR2023018036-appb-img-000147
여기서,
Figure PCTKR2023018036-appb-img-000148
는 차량 속도(v) 및 경로 곡률(w)의 함수로 결정하되, 다음의 수학식 5로 정의될 수 있다.here,
Figure PCTKR2023018036-appb-img-000148
is determined as a function of vehicle speed (v) and path curvature (w), and can be defined as Equation 5 below.
Figure PCTKR2023018036-appb-img-000149
Figure PCTKR2023018036-appb-img-000149
또한, 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000150
선정모듈(110)은 초당 수에서 수십 회 반복하여 가변 전방주시거리를 얻을 수 있다.Also, variable forward looking distance
Figure PCTKR2023018036-appb-img-000150
The selection module 110 can obtain a variable forward looking distance by repeating several to tens of times per second.
본 발명의 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법은 자율이동체의 속도뿐만 아니라 기준 경로의 곡률을 이용하여 전방주시거리
Figure PCTKR2023018036-appb-img-000151
를 가변시킴으로써 경로추종 성능을 향상시킬 수 있다. 전방주시거리
Figure PCTKR2023018036-appb-img-000152
의 특성에 따라 차량의 속도와 기준 경로의 곡률을 고려하여 적절한 전방주시거리
Figure PCTKR2023018036-appb-img-000153
값을 가지도록 한다. 만약 일정 곡률 미만이거나 일정 속도를 초과할 때는 전방주시거리
Figure PCTKR2023018036-appb-img-000154
를 상대적으로 큰 값을 가지도록 하고, 일정 곡률을 초과하거나 일정 속도 미만일 때는 작은 값을 가지게 함으로써 차량의 주행 안정성과 기준 경로에 대한 추종 성능을 향상시킬 수 있는 장점이 있다.The path tracking method of an autonomous vehicle using the improved variable forward gaze distance of the present invention uses not only the speed of the autonomous vehicle but also the curvature of the reference path to determine the forward gaze distance.
Figure PCTKR2023018036-appb-img-000151
Path following performance can be improved by varying . Forward viewing distance
Figure PCTKR2023018036-appb-img-000152
Appropriate forward looking distance considering the vehicle speed and curvature of the reference path depending on the characteristics of
Figure PCTKR2023018036-appb-img-000153
Let it have a value. If the curvature is below a certain level or exceeds a certain speed, the forward looking distance is
Figure PCTKR2023018036-appb-img-000154
There is an advantage in improving the driving stability of the vehicle and the tracking performance of the reference path by setting it to a relatively large value and having a small value when it exceeds a certain curvature or is below a certain speed.
이상에서는 본 발명의 바람직한 실시 예에 대하여 도시하고 설명하였지만, 본 발명은 상술한 특정의 실시 예에 한정되지 아니하며, 청구범위에서 청구하는 본 발명의 요지를 벗어남이 없이 당해 발명이 속하는 기술분야에서 통상의 지식을 가진 자에 의해 다양한 변형된 실시가 가능한 것은 물론이고, 이러한 변형된 실시들은 본 발명의 기술적 사상이나 전망으로부터 개별적으로 이해되어서는 안 될 것이다.In the above, preferred embodiments of the present invention have been shown and described, but the present invention is not limited to the specific embodiments described above, and may be used in the technical field to which the invention pertains without departing from the gist of the invention as claimed in the claims. Of course, various modified implementations are possible by those skilled in the art, and these modified implementations should not be understood individually from the technical idea or perspective of the present invention.
[부호의 설명][Explanation of symbols]
10: 경로계획부 20: 센서부 30: 수신부 40: 조향장치10: Path planning unit 20: Sensor unit 30: Receiving unit 40: Steering device
100: 경로추종 장치 110: 가변 전방주시거리
Figure PCTKR2023018036-appb-img-000155
선정모듈 111: 곡률비교부 112: 속도비교부 113: 전방주시거리가변부 120: 목표점 선정모듈 130:
Figure PCTKR2023018036-appb-img-000156
계산모듈 140: 조향각
Figure PCTKR2023018036-appb-img-000157
산출모듈100: Path following device 110: Variable forward looking distance
Figure PCTKR2023018036-appb-img-000155
Selection module 111: Curvature comparison unit 112: Speed comparison unit 113: Forward viewing distance variable unit 120: Target point selection module 130:
Figure PCTKR2023018036-appb-img-000156
Calculation module 140: steering angle
Figure PCTKR2023018036-appb-img-000157
Calculation module

Claims (6)

  1. 자율이동체의 계획된 경로상의 기준 경로 정보와 주행환경에 따른 차량 속도, 진행방향 및 위치를 포함하는 감지데이터로부터 도출된 차량의 경로 곡률(w)과 차량 속도(v)를 기초로 전방주시거리
    Figure PCTKR2023018036-appb-img-000158
    를 결정하는 가변 전방주시거리
    Figure PCTKR2023018036-appb-img-000159
    선정모듈;    Forward looking distance based on the vehicle's path curvature (w) and vehicle speed (v) derived from sensing data including reference path information on the autonomous vehicle's planned route and vehicle speed, direction, and location according to the driving environment.
    Figure PCTKR2023018036-appb-img-000158
    Variable forward looking distance that determines
    Figure PCTKR2023018036-appb-img-000159
    selection module;
    상기 기준 경로 정보와 상기 가변 전방주시거리
    Figure PCTKR2023018036-appb-img-000160
    선정모듈에서 결정된 전방주시거리
    Figure PCTKR2023018036-appb-img-000161
    로부터 목표점을 결정하는 목표점 선정모듈;    The reference path information and the variable forward looking distance
    Figure PCTKR2023018036-appb-img-000160
    Forward looking distance determined in the selection module
    Figure PCTKR2023018036-appb-img-000161
    A target point selection module that determines a target point from;
    상기 진행방향 및 위치와 상기 목표점 선정모듈에서 결정된 목표점으로부터
    Figure PCTKR2023018036-appb-img-000162
    값을 산출하는
    Figure PCTKR2023018036-appb-img-000163
    계산모듈; 및    From the direction and location and the target point determined in the target point selection module
    Figure PCTKR2023018036-appb-img-000162
    yielding a value
    Figure PCTKR2023018036-appb-img-000163
    calculation module; and
    상기 전방주시거리
    Figure PCTKR2023018036-appb-img-000164
    와
    Figure PCTKR2023018036-appb-img-000165
    값으로부터 조향각
    Figure PCTKR2023018036-appb-img-000166
    를 산출하여 산출된 조향각 데이터를 출력하는 조향각
    Figure PCTKR2023018036-appb-img-000167
    산출모듈을 포함하는, 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 장치.    The forward viewing distance
    Figure PCTKR2023018036-appb-img-000164
    and
    Figure PCTKR2023018036-appb-img-000165
    Steering angle from value
    Figure PCTKR2023018036-appb-img-000166
    Steering angle that calculates and outputs the calculated steering angle data
    Figure PCTKR2023018036-appb-img-000167
    A path tracking device for an autonomous vehicle using an improved variable forward looking distance, including a calculation module.
  2. 제1항에 있어서, According to paragraph 1,
    상기 가변 전방주시거리
    Figure PCTKR2023018036-appb-img-000168
    선정모듈은,    The variable forward looking distance
    Figure PCTKR2023018036-appb-img-000168
    The selection module is,
    상기 기준 경로 정보를 기초로 산출된 자율이동체의 현재 위치에서의 경로 곡률(w)과 사전에 설정된 제1곡률(
    Figure PCTKR2023018036-appb-img-000169
    ) 및 제2곡률(
    Figure PCTKR2023018036-appb-img-000170
    )을 각각 비교하는 곡률비교부;     The path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path information and the preset first curvature (
    Figure PCTKR2023018036-appb-img-000169
    ) and second curvature (
    Figure PCTKR2023018036-appb-img-000170
    ) a curvature comparison unit that compares each;
    자율이동체의 차량 속도(v)와 사전에 설정된 제1속도(
    Figure PCTKR2023018036-appb-img-000171
    )및 제2속도(
    Figure PCTKR2023018036-appb-img-000172
    )을 각각 비교하는 속도비교부; 및    The vehicle speed (v) of the autonomous vehicle and the preset first speed (
    Figure PCTKR2023018036-appb-img-000171
    ) and the second speed (
    Figure PCTKR2023018036-appb-img-000172
    ) a speed comparison unit that compares each; and
    상기 곡률비교부와 속도비교부에서 비교한 결과에 기초하여 전방주시거리
    Figure PCTKR2023018036-appb-img-000173
    를 가변하여 출력하는 전방주시거리 가변부를 포함하는, 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 장치.     Forward looking distance based on the results compared by the curvature comparison unit and the speed comparison unit
    Figure PCTKR2023018036-appb-img-000173
    A path tracking device for an autonomous mobile vehicle using an improved variable front gaze distance, including a front gaze distance variable unit that varies and outputs .
  3. 제1항에 있어서, According to paragraph 1,
    상기 조향각
    Figure PCTKR2023018036-appb-img-000174
    산출모듈은,     The steering angle
    Figure PCTKR2023018036-appb-img-000174
    The calculation module is,
    애커먼 방식의 조향시스템을 가진 이동체에 대한 앞바퀴 조향각을 다음의 수학식으로 산출하는, 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 장치.A path-following device for an autonomous mobile vehicle using an improved variable forward looking distance that calculates the front wheel steering angle for a mobile vehicle with an Ackerman-type steering system using the following equation.
    Figure PCTKR2023018036-appb-img-000175
    Figure PCTKR2023018036-appb-img-000175
  4. (a) 가변 전방주시거리
    Figure PCTKR2023018036-appb-img-000176
    선정모듈이, 자율이동체의 계획된 경로상의 기준 경로 정보와 주행환경에 따른 차량의 속도, 진행방향 및 위치를 포함하는 감지 데이터로부터 도출된 경로 곡률(w)과 차량속도(v)를 기초로 전방주시거리
    Figure PCTKR2023018036-appb-img-000177
    를 결정하는 단계;    (a) Variable forward looking distance
    Figure PCTKR2023018036-appb-img-000176
    The selection module looks ahead based on the path curvature (w) and vehicle speed (v) derived from sensing data including reference path information on the autonomous vehicle's planned route and the vehicle's speed, direction of travel, and location according to the driving environment. distance
    Figure PCTKR2023018036-appb-img-000177
    determining;
    (b) 목표점 선정모듈이, 상기 기준 경로 정보와 상기 전방주시거리
    Figure PCTKR2023018036-appb-img-000178
    로부터 목표점을 결정하는 단계;    (b) The target point selection module, the reference path information and the forward looking distance
    Figure PCTKR2023018036-appb-img-000178
    determining a target point from;
    (c)
    Figure PCTKR2023018036-appb-img-000179
    계산모듈이, 상기 진행방향 및 위치와 상기 (b) 단계로부터의 목표점으로부터
    Figure PCTKR2023018036-appb-img-000180
    값을 산출하는 단계;    (c)
    Figure PCTKR2023018036-appb-img-000179
    The calculation module determines the direction and position from the target point from step (b).
    Figure PCTKR2023018036-appb-img-000180
    calculating a value;
    (d) 조향각
    Figure PCTKR2023018036-appb-img-000181
    산출모듈이, 상기 (a) 단계로부터의 전방주시거리
    Figure PCTKR2023018036-appb-img-000182
    와 상기 (c)단계로부터의
    Figure PCTKR2023018036-appb-img-000183
    값으로부터 조향각
    Figure PCTKR2023018036-appb-img-000184
    를 산출하여 조향각 데이터를 출력하는 단계; 를 포함하여 이루어진, 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법.    (d) Steering angle
    Figure PCTKR2023018036-appb-img-000181
    Calculation module, forward looking distance from step (a)
    Figure PCTKR2023018036-appb-img-000182
    and from step (c) above.
    Figure PCTKR2023018036-appb-img-000183
    Steering angle from value
    Figure PCTKR2023018036-appb-img-000184
    Calculating and outputting steering angle data; A path-following method for an autonomous mobile vehicle using an improved variable forward looking distance, including.
  5. 제4항에 있어서, According to clause 4,
    상기 (a) 단계는,In step (a),
    (a-1) 상기 기준 경로 정보를 기초로 산출된 자율이동체의 현재 위치에서의 경로 곡률(w)이 사전에 설정된 제1곡률(
    Figure PCTKR2023018036-appb-img-000185
    )보다 크면 상기 전방주시거리
    Figure PCTKR2023018036-appb-img-000186
    를 기존 전방주시거리
    Figure PCTKR2023018036-appb-img-000187
    보다 짧게 가변하는 단계;    (a-1) The path curvature (w) at the current location of the autonomous vehicle calculated based on the reference path information is the preset first curvature (
    Figure PCTKR2023018036-appb-img-000185
    ), if it is greater than the above forward looking distance
    Figure PCTKR2023018036-appb-img-000186
    The existing forward looking distance
    Figure PCTKR2023018036-appb-img-000187
    shorter variable stages;
    (a-2) 상기 경로 곡률(w)이 제1곡률(
    Figure PCTKR2023018036-appb-img-000188
    )보다 작고, 상기 제1곡률(
    Figure PCTKR2023018036-appb-img-000189
    )보다 작게 설정된 제2곡률(
    Figure PCTKR2023018036-appb-img-000190
    )보다 더 작으면 상기 전방주시거리
    Figure PCTKR2023018036-appb-img-000191
    를 기존 전방주시거리
    Figure PCTKR2023018036-appb-img-000192
    보다 길게 가변하는 단계;    (a-2) The path curvature (w) is the first curvature (
    Figure PCTKR2023018036-appb-img-000188
    ), and the first curvature (
    Figure PCTKR2023018036-appb-img-000189
    The second curvature (
    Figure PCTKR2023018036-appb-img-000190
    ), if it is smaller than the above forward looking distance
    Figure PCTKR2023018036-appb-img-000191
    The existing forward looking distance
    Figure PCTKR2023018036-appb-img-000192
    longer variable stages;
    (a-3) 상기 경로 곡률(w)이 제1곡률(
    Figure PCTKR2023018036-appb-img-000193
    )보다 작고 제2곡률(
    Figure PCTKR2023018036-appb-img-000194
    )보다 크면 자율이동체의 차량 속도(v)를 더 고려하되, 상기 차량 속도(v)가 사전에 설정된 제1속도(
    Figure PCTKR2023018036-appb-img-000195
    )보다 작으면 상기 전방주시거리
    Figure PCTKR2023018036-appb-img-000196
    를 기존 전방주시거리
    Figure PCTKR2023018036-appb-img-000197
    보다 짧게 가변하는 단계;    (a-3) The path curvature (w) is the first curvature (
    Figure PCTKR2023018036-appb-img-000193
    ) and the second curvature (
    Figure PCTKR2023018036-appb-img-000194
    ), the vehicle speed (v) of the autonomous vehicle is further considered, but the vehicle speed (v) is the preset first speed (
    Figure PCTKR2023018036-appb-img-000195
    ), if it is less than the above forward looking distance
    Figure PCTKR2023018036-appb-img-000196
    The existing forward looking distance
    Figure PCTKR2023018036-appb-img-000197
    shorter variable stages;
    (a-4) 상기 차량 속도(v)가 제1속도(
    Figure PCTKR2023018036-appb-img-000198
    )보다 크되, 제1속도(
    Figure PCTKR2023018036-appb-img-000199
    )보다 크게 설정된 제2속도(
    Figure PCTKR2023018036-appb-img-000200
    )보다 더 크면 상기 전방주시거리
    Figure PCTKR2023018036-appb-img-000201
    를 기존 전방주시거리
    Figure PCTKR2023018036-appb-img-000202
    보다 길게 가변하는 단계;    (a-4) The vehicle speed (v) is the first speed (
    Figure PCTKR2023018036-appb-img-000198
    ), but the first speed (
    Figure PCTKR2023018036-appb-img-000199
    The second speed set greater than ) (
    Figure PCTKR2023018036-appb-img-000200
    ), if it is greater than the above forward looking distance
    Figure PCTKR2023018036-appb-img-000201
    The existing forward looking distance
    Figure PCTKR2023018036-appb-img-000202
    longer variable stages;
    (a-5) 상기 차량 속도(v)가 제1속도(
    Figure PCTKR2023018036-appb-img-000203
    )보다 크고 제2속도(
    Figure PCTKR2023018036-appb-img-000204
    )보다 작으면 상기 경로 곡률(w)과 상기 차량 속도(v)에 따라 전방주시거리
    Figure PCTKR2023018036-appb-img-000205
    를 가변하는 단계; 를 포함하는 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법.    (a-5) The vehicle speed (v) is the first speed (
    Figure PCTKR2023018036-appb-img-000203
    ) and the second speed (
    Figure PCTKR2023018036-appb-img-000204
    ), the forward looking distance depends on the path curvature (w) and the vehicle speed (v).
    Figure PCTKR2023018036-appb-img-000205
    A step of varying; A path-following method for an autonomous mobile vehicle using an improved variable forward viewing distance including.
  6. 제5항에 있어서, According to clause 5,
    상기 전방주시거리
    Figure PCTKR2023018036-appb-img-000206
    는,    The forward viewing distance
    Figure PCTKR2023018036-appb-img-000206
    Is,
    차량 속도(v) 및 경로 곡률(w) 에 따라 다음의 수학식으로 정의되는 것인, 개선된 가변 전방주시거리를 이용한 자율이동체의 경로추종 방법.A path-following method for an autonomous vehicle using an improved variable forward looking distance, which is defined by the following equation according to vehicle speed (v) and path curvature (w).
    Figure PCTKR2023018036-appb-img-000207
    Figure PCTKR2023018036-appb-img-000207
    여기서,
    Figure PCTKR2023018036-appb-img-000208
    는 차량 속도(v) 및 경로 곡률(w)의 함수로 결정하되, 다음의 수학식으로 정의된다.    here,
    Figure PCTKR2023018036-appb-img-000208
    is determined as a function of vehicle speed (v) and path curvature (w), and is defined by the following equation.
    Figure PCTKR2023018036-appb-img-000209
    Figure PCTKR2023018036-appb-img-000209
PCT/KR2023/018036 2022-11-21 2023-11-10 Device for tracking path of autonomous vehicle by using improved variable look-ahead distance and method therefor WO2024111966A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR1020220156776A KR20240074565A (en) 2022-11-21 2022-11-21 Path Tracking Device of Autonomous Driving Things Using Improved Variable Look-Ahead Distance and Method Thereof
KR10-2022-0156776 2022-11-21

Publications (1)

Publication Number Publication Date
WO2024111966A1 true WO2024111966A1 (en) 2024-05-30

Family

ID=91195947

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2023/018036 WO2024111966A1 (en) 2022-11-21 2023-11-10 Device for tracking path of autonomous vehicle by using improved variable look-ahead distance and method therefor

Country Status (2)

Country Link
KR (1) KR20240074565A (en)
WO (1) WO2024111966A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120089498A (en) * 2010-12-06 2012-08-13 현대자동차주식회사 Lane keeping assist system considering driver's range of view and steering tendency by speed
KR20160066775A (en) * 2014-12-03 2016-06-13 현대모비스 주식회사 Smart cruise control system of a vehicle, and a controlling method thereof
US20170235306A1 (en) * 2016-02-16 2017-08-17 Fuji Jukogyo Kabushiki Kaisha Vehicle traveling control apparatus
KR20200017571A (en) * 2018-07-23 2020-02-19 한국자동차연구원 Lateral control parameter correction apparatus and method for autonomous vehicle
KR20200133852A (en) * 2019-05-20 2020-12-01 현대모비스 주식회사 Autonomous driving apparatus and method

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102500704B1 (en) 2016-04-19 2023-02-16 주식회사 에이치엘클레무브 Apparatus and method for controlling automatic driving of vehicle
KR102349654B1 (en) 2021-06-21 2022-01-13 주식회사 라이드플럭스 Method, apparatus and computer program for revising curvature error of automatic driving vehicle
KR102394465B1 (en) 2021-09-07 2022-05-06 주식회사 라이드플럭스 Method, server and computer program for verifying steering wheel angle command for automatic driving vehicle

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120089498A (en) * 2010-12-06 2012-08-13 현대자동차주식회사 Lane keeping assist system considering driver's range of view and steering tendency by speed
KR20160066775A (en) * 2014-12-03 2016-06-13 현대모비스 주식회사 Smart cruise control system of a vehicle, and a controlling method thereof
US20170235306A1 (en) * 2016-02-16 2017-08-17 Fuji Jukogyo Kabushiki Kaisha Vehicle traveling control apparatus
KR20200017571A (en) * 2018-07-23 2020-02-19 한국자동차연구원 Lateral control parameter correction apparatus and method for autonomous vehicle
KR20200133852A (en) * 2019-05-20 2020-12-01 현대모비스 주식회사 Autonomous driving apparatus and method

Also Published As

Publication number Publication date
KR20240074565A (en) 2024-05-28

Similar Documents

Publication Publication Date Title
CN109556615B (en) Driving map generation method based on multi-sensor fusion cognition of automatic driving
CN109263639B (en) Driving path planning method based on state grid method
WO2018143759A1 (en) Method for estimating road surface friction coefficient of tire and device for estimating road surface friction coefficient of tire in high speed normal driving state
EP3663153A1 (en) Vehicle control device
EP3678110A1 (en) Method for correcting positional error and device for correcting positional error in driving assistance vehicle
CN109606368B (en) Intelligent automobile extension vehicle speed self-adaptive change trajectory tracking control method
CN108609014A (en) The crosswise joint method and device of intelligent vehicle
CN110834630A (en) Vehicle driving control method and device, vehicle and storage medium
EP0446902A2 (en) Automatic travelling apparatus
US11914386B2 (en) Mobile object control device, mobile object control method, and storage medium
CN106292666A (en) Pilotless automobile barrier-avoiding method based on ultrasonic distance detection and system
KR20200024934A (en) Lane information management method, driving control method and lane information management device
CN114426032A (en) Automatic driving-based vehicle trajectory prediction method and system, vehicle and computer-readable storage medium
WO2024111966A1 (en) Device for tracking path of autonomous vehicle by using improved variable look-ahead distance and method therefor
CN111845737A (en) Curve target identification method of intelligent vehicle and danger level judgment mechanism thereof
US20230137911A1 (en) Vehicle motion control device and vehicle motion control method
EP3447449B1 (en) Host vehicle position confidence degree calculation device
CN102679997B (en) Guider
US20240051526A1 (en) Information processing device, information processing method, and storage medium
CN113460064B (en) Front target vehicle identification method and device and electric automobile
KR20190064229A (en) Apparatus for controlling vehicle and method thereof
US20230177846A1 (en) Mobile object control device, mobile object control method, and storage medium
CN115218902B (en) Track planning method, device, equipment and storage medium
CN113460091B (en) Unprotected crossroad unmanned vehicle rolling optimization decision method
US20220316913A1 (en) Map information assessment device, storage medium storing computer program for map information assessment, and map information assessment method