KR20230071609A - Vehicle control system and navigating method using vehicle control system - Google Patents

Abstract

The present invention relates to a vehicle control system and a vehicle navigation method using the vehicle control system to set accurate driving routes for various situations. The vehicle control system according to the present invention includes a processing unit for processing data related to the driving of a vehicle and a photographing unit for detecting and photographing the external environment of the vehicle, wherein the processing unit selects landmarks based on image information obtained by the photographing unit, divides sections between the selected landmarks into designated distance units, reflects longitudinal and lateral gradient information received from a server to each of the divided distance units, and sets sections including the divided distance units, to which the longitudinal and lateral gradient information is reflected, as new landmarks.

Description

Vehicle control system and vehicle driving method using the vehicle control system

The present invention relates to a vehicle control system and a vehicle driving method using the vehicle control system, and more particularly, to an autonomous driving technology for improving the accuracy of a driving path to be driven.

Autonomous driving technology that sets a driving route of a vehicle and drives the vehicle along the set driving route without a driver directly driving is appearing. Autonomous driving technology may be implemented by acquiring route information on a driving route, setting a driving route based on the acquired route information, and driving along the set route.

According to existing autonomous driving technologies, it may not be easy to set accurate driving routes for various situations.

An embodiment of the present invention is intended to provide a technology capable of setting an accurate driving route for various situations.

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

A vehicle control system according to an embodiment of the present invention includes a processing unit for processing data related to driving of a vehicle and a photographing unit for detecting and photographing an external environment of the vehicle, wherein the processing unit includes image information acquired by the photographing unit. Selecting landmarks based on, dividing sections between the selected landmarks into designated distance units, reflecting longitudinal and lateral gradient information received from the server in each of the divided distance units, and the vertical gradient And the sections including the divided distance units reflecting the horizontal slope information may be set as new landmarks.

A vehicle driving method using a vehicle control system according to an embodiment of the present invention includes an operation of selecting landmarks based on image information acquired by a photographing unit of the vehicle control system, and designating sections between the selected landmarks. Operation of classifying by distance unit, operation of reflecting longitudinal gradient and lateral gradient information received from the server in each of the divided distance units, and the segment including the divided distance unit reflecting the vertical gradient and the lateral gradient information It may include an operation of setting them as new landmarks.

The vehicle control system according to the present technology can improve the accuracy of a driving path to be driven.

In addition to this, various effects identified directly or indirectly through this document may be provided.

1 is a block diagram illustrating a vehicle control system according to an embodiment of the present invention.
FIG. 2 is a view showing positions where cameras of a vehicle control system according to an embodiment of the present invention are disposed in a vehicle.
3 is a view showing a location where a camera of a vehicle control system according to an embodiment of the present invention is disposed in a vehicle.
4 is a diagram illustrating a location where a camera of a vehicle control system according to an embodiment of the present invention is disposed in a vehicle.
5 is a diagram illustrating a location where a camera of a vehicle control system according to an embodiment of the present invention is disposed in a vehicle.
6 is a diagram illustrating a plurality of camera devices of a vehicle control system according to an embodiment of the present invention.
7 is a diagram illustrating a plurality of camera devices of a vehicle control system according to an embodiment of the present invention.
8 is a block diagram illustrating a sparse map of a processing unit according to an embodiment of the present invention.
9 is a diagram showing a polynomial expression of a trajectory according to an embodiment of the present invention.
10 is a diagram showing landmarks according to an embodiment of the present invention.
11 is a flowchart illustrating a method of generating a sparse map by a vehicle control system according to an embodiment of the present invention.
12 is a flowchart illustrating a method of anonymously processing navigation information by a vehicle control system according to an embodiment of the present invention.
13 is a flowchart illustrating a method of comparing a trajectory generated by a road navigation model with sensing data by a vehicle control system according to an embodiment of the present invention and controlling an autonomous driving mode.
14 is a flowchart illustrating a method of comparing a path recognized using a sparse map with a path recognized using sensing data and controlling an autonomous driving mode by a vehicle control system according to an embodiment of the present invention.
15 is a diagram illustrating correction of a path generated by a vehicle control system based on a target trajectory according to a vehicle position according to an embodiment of the present invention.
16 is a flowchart illustrating a method of controlling driving of a vehicle by including a beacon of a dangerous section in a sparse map by a vehicle control system according to an embodiment of the present invention.
17 is a flowchart illustrating a method of selecting data to be clustered according to reliability of a driving trajectory when the vehicle control system clusters driving trajectories according to an embodiment of the present invention.
18 is a diagram illustrating driving trajectories excluded from clustering according to an embodiment of the present invention.
19 is a diagram illustrating driving trajectories excluded from clustering according to an embodiment of the present invention.
20 is a flowchart illustrating a method of correcting an offset of a sensor by using a longitudinal gradient and a lateral gradient in a vehicle control system according to an embodiment of the present invention.
21 is a flowchart illustrating a method of controlling a traveling speed and an amount of a gradient by using a vertical gradient and a lateral gradient by a vehicle control system according to an embodiment of the present invention.
22 is a flowchart illustrating a method for calculating an effective trajectory by a vehicle control system according to an embodiment of the present invention.
23 is a flowchart illustrating a method of setting landmarks by reflecting vertical and lateral gradient information in a vehicle control system according to an embodiment of the present invention.
24 is a flowchart illustrating a method for a vehicle control system to control driving and functions of a vehicle using a beacon in an area where a navigation satellite signal is not received according to an embodiment of the present invention.
25 is a diagram illustrating that a vehicle control system according to an embodiment of the present invention creates a virtual lane in a section where a lane is broken on a vehicle movement path.
26 is a flowchart illustrating a method of controlling driving of a vehicle by creating a virtual lane by a vehicle control system according to an embodiment of the present invention.
27 is a flowchart illustrating a method of adjusting a driving control right range according to trajectory accuracy by a vehicle control system according to an embodiment of the present invention.
28 is a flowchart illustrating a method of improving reliability of a trajectory when a vehicle control system calculates a trajectory of a vehicle using a server according to an embodiment of the present invention.
29 is a flowchart illustrating a method of adjusting a control target point for a cut-in vehicle according to a camera position by a vehicle control system according to an embodiment of the present invention.
30 is a flowchart illustrating a method of setting an entrance of a tunnel as a landmark by a vehicle control system according to an embodiment of the present invention.
31 is a flowchart illustrating a method of setting the interior of a tunnel as a landmark by a vehicle control system according to an embodiment of the present invention.
32 is a flowchart illustrating a method of correcting a sensor unit by using a portable communication device in a vehicle control system according to an embodiment of the present invention.
33 is a flowchart illustrating a method of correcting a GPS signal in consideration of characteristics of a portable communication device and a road by a vehicle control system according to an embodiment of the present invention.

1 is a block diagram illustrating a vehicle control system according to an embodiment of the present invention.

A vehicle control system according to an embodiment may include a processing unit 110 , an input unit 120 , a sensor unit 130 , a photographing unit 140 , an output unit 150 , and a vehicle control unit 160 .

The processing unit 110 may realize autonomous driving by processing data related to vehicle driving. The processing unit 110 may include a monocular image analysis module 111 , a stereoscopic image analysis module 112 , a speed and acceleration module 113 , and a navigation response module 114 .

The monocular image analysis module 111 may analyze a monocular image of an image set acquired by the photographing unit 140 . The monocular image analysis module 111 may perform monocular image analysis by merging data included in the image set with other types of data acquired by the photographing unit 140 . The monocular image analysis module 111 may detect features within a set of images, such as lane markings, vehicles, pedestrians, road signs, highway interchanges, traffic lights, hazardous objects, and other characteristics associated with the vehicle's surroundings. The vehicle control system may cause one or more navigational responses, such as turning the vehicle, changing lanes, or changing acceleration, via processing unit 110 based on the analysis of monocular image analysis module 111 .

The 3D image analysis module 112 may perform analysis by combining the data acquired from the photographing unit 140 and the data acquired from the sensor unit 130 . The stereoscopic image analysis module 112 may perform stereoscopic image analysis. The stereoscopic image analysis module 112 may implement a method related to a neural network learning system, a deep neural network learning system, or a non-learning system for detecting and/or labeling objects by utilizing computer vision algorithms in the context of capturing and processing sensing information. there is. The stereoscopic image analysis module 112 may be used by combining a learning system and a non-learning system.

The speed and acceleration module 113 may control changes in speed and/or acceleration of the vehicle. The speed and acceleration module 113 may calculate a target speed of the vehicle based on data acquired from the monocular image analysis module 111 and/or the stereoscopic image analysis module 112 . The data obtained from the monocular image analysis module 111 and/or the stereoscopic image analysis module 112 is the target position, speed, acceleration, position and/or speed of the vehicle relative to surrounding vehicles, pedestrians or objects on the road, and the speed of the road. Vehicle location information for lane marking may be included. The speed and acceleration module 113 may transmit a speed control signal to the vehicle controller 160 based on the calculated target speed.

The navigation response module 114 may determine a necessary navigation response based on data obtained from the monocular image analysis module 111 , the stereoscopic image analysis module 112 , and the input unit 120 . The data acquired from the monocular image analysis module 111, the stereoscopic image analysis module 112, and the input unit 120 may include information on the location and speed of surrounding vehicles, pedestrians, and objects on the road, and target location information of the vehicle. there is. The navigational response may be determined based on map data, a predetermined position of the vehicle, and relative velocity or relative acceleration between the vehicle and one or more objects. The navigation response module 114 may transmit a navigation control signal to the vehicle controller 160 based on a navigation response determined to be necessary. For example, the navigation response module 114 may generate a necessary navigation response by inducing a rotation of a preset angle by rotating the steering wheel of the vehicle. The navigation response determined to be necessary by the navigation response module 114 may be used as data input to the speed and acceleration module 113 to calculate the speed change of the vehicle.

The input unit 120 may receive a user input for controlling a driving function. The input unit 120 may include a driving mode switch 121 , a navigation 122 , a steering wheel 123 , an accelerator pedal 124 , and a brake pedal 125 . The input unit 120 may transmit a user input to the processing unit 110 through the driving information input interface 126 .

The sensor unit 130 may obtain data related to driving of the vehicle from the vehicle and the external environment. The sensor unit 130 may include a wheel speed sensor 131 , a yaw rate 132 , a steering angle sensor 144 , and a G sensor 134 . The sensor unit 130 may transfer acquired data to the processing unit 110 through the vehicle information input interface 135 .

The photographing unit 140 may sense and photograph the external environment. The photographing unit 140 may include a radar 141 , a lidar 142 , an ultrasound 143 , a camera 144 , and a camera 145 inside the vehicle. The photographing unit 140 may transfer the sensed and photographed external environment to the processing unit 110 .

The output unit 150 may provide information related to driving of the vehicle to passengers including the driver. The output unit 150 may include a speaker 151 and a display 152 . The output unit 150 may provide information related to driving of the vehicle output from the processing unit 110 through the driver output interface 153 to a passenger.

The vehicle control unit 160 may control driving of the vehicle. The vehicle control unit 160 may include an engine control system 161 , a braking control system 162 , and a steering control system 163 . The vehicle control unit 160 may receive driving control information output from the processing unit 110 through the vehicle control output interface 164 to control driving of the vehicle.

FIG. 2 is a view showing positions where cameras of a vehicle control system according to an embodiment of the present invention are disposed in a vehicle.

The camera 144 may include a first camera device 144_1 , a second camera device 144_2 , and a third camera device 144_3 . The first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may be disposed side by side in the width direction of the vehicle. The first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may be disposed around the rear view mirror of the vehicle and/or adjacent to the driver's seat. At least a portion of the fields of view of each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may overlap each other.

The camera 144 may capture an external environment. The camera 144 may fuse image information captured by the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 . The camera 144 may obtain a three-dimensional image by using a difference in field of view according to a difference in positions of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 . The camera 144 may deliver image data of the external environment captured to the processing unit 110 .

3 is a view showing a location where a camera of a vehicle control system according to an embodiment of the present invention is disposed in a vehicle.

The camera 144 may include a first camera device 144_1 and a second camera device 144_2. The first camera device 144_1 and the second camera device 144_2 may be disposed side by side in the width direction of the vehicle. The first camera device 144_1 and the second camera device 144_2 may be disposed around the rear view mirror of the vehicle and/or adjacent to the driver's seat. At least a portion of the fields of view of each of the first camera device 144_1 and the second camera device 144_2 may overlap each other. The first camera device 144_1 and the second camera device 144_2 may be disposed to be spaced apart from each other by a first distance D1 in the width direction of the vehicle.

The camera 144 may capture an external environment. The camera 144 may fuse image information captured by the first camera device 144_1 and the second camera device 144_2 . The camera 144 may obtain a three-dimensional image by using a difference in field of view due to a difference in positions of the first camera device 144_1 and the second camera device 144_2 . The camera 144 may deliver image data of the external environment captured to the processing unit 110 .

4 is a diagram illustrating a location where a camera of a vehicle control system according to an embodiment of the present invention is disposed in a vehicle.

The camera 144 may include a first camera device 144_1 , a second camera device 144_2 , and a third camera device 144_3 . The first camera device 144_1 may be disposed on or inside the bumper area of the vehicle. The first camera device 144_1 may be disposed adjacent to one of the corner portions of the bumper area. The second camera device 144_2 may be disposed around the rear view mirror of the vehicle and/or adjacent to the driver's seat. At least a portion of the fields of view of each of the first camera device 144_1 and the second camera device 144_2 may overlap each other. The first camera device 144_1 and the second camera device 144_2 may be disposed to be spaced apart from each other by a second distance D2 in the width direction of the vehicle.

The camera 144 may capture an external environment. The camera 144 may fuse image information captured by the first camera device 144_1 and the second camera device 144_2 . The camera 144 may obtain a three-dimensional image by using a difference in field of view due to a difference in positions of the first camera device 144_1 and the second camera device 144_2 . The camera 144 may deliver image data of the external environment captured to the processing unit 110 .

5 is a diagram illustrating a location where a camera of a vehicle control system according to an embodiment of the present invention is disposed in a vehicle.

The camera 144 may include a first camera device 144_1 , a second camera device 144_2 , and a third camera device 144_3 . The first camera device 144_1 and the third camera device 144_3 may be disposed on or inside the bumper area of the vehicle. The first camera device 144_1 may be disposed adjacent to one of the corner portions of the bumper area. The third camera device 144_3 may be disposed adjacent to corner portions of the bumper area, except for the corner portion where the first camera device 144_1 is disposed. The second camera device 144_2 may be disposed around the rear view mirror of the vehicle and/or adjacent to the driver's seat. At least a portion of the fields of view of each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may overlap each other.

The camera 144 may capture an external environment. The camera 144 may fuse image information captured by the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 . The camera 144 may obtain a three-dimensional image by using a difference in field of view according to a difference in positions of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 . The camera 144 may deliver image data of the external environment captured to the processing unit 110 .

6 is a diagram illustrating a plurality of camera devices of a vehicle control system according to an embodiment of the present invention.

The plurality of camera devices may include a first camera device 144_1 , a second camera device 144_2 , and a third camera device 144_3 . 7 is a diagram illustrating a plurality of camera devices of a vehicle control system according to an embodiment of the present invention. The plurality of camera devices may include a first camera device 144_1 , a second camera device 144_2 , and a third camera device 144_3 .

Each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may include an appropriate type of image capture device. The image capture device may include an optical axis. The image capture device may include an Aptina M9V024 WVGA sensor with a global shutter. The image capture device may provide a resolution of 1280x960 pixels and include a rolling shutter method. An image capture device may include a variety of optical elements. An image capture device may include one or more lenses to provide a focal length and field of view required by the image capture device. The image capture device may be combined with either a 6mm lens or a 12mm lens.

Each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may have a specified viewing angle range. Each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may have a normal viewing angle range of 40 degrees or more and 56 degrees or less. Each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may have a narrow viewing angle range of 23 degrees or more and 40 degrees or less. Each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may have a wide viewing angle range of 100 degrees or more and 180 degrees or less. Each of the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 may include a wide-angle bumper camera or a camera capable of securing a 180-degree field of view. The field of view of the first camera device 144_1 may be wider, narrower, or partially overlap with the field of view of the second camera device 144_2 .

A 7.2 megapixel image capture device with an aspect ratio of about 2:1 (eg, H x V = 3800x1900 pixels) and a horizontal field of view of about 100 degrees includes a first camera device 144_1, a second camera device 144_2, and a third camera device 144_2. A configuration of a plurality of camera devices including the camera device 144_3 may be replaced. A vertical field of view of a megapixel image capture device using a radially symmetric lens may be less than 50 degrees due to lens distortion. A radially asymmetric lens can be used to achieve a vertical field of view of more than 50 degrees when the horizontal field of view is 100 degrees.

A driving support function may be provided using a multi-camera system including a plurality of camera devices. A multi-camera system may use more than one camera facing the front of the vehicle. In a multi-camera system, at least one camera may face to the side or rear of a vehicle. In the multi-camera system, the first camera device 144_1 and the second camera device 144_2 may face the front and/or side of the vehicle by using the dual camera imaging system.

A multi-camera system including a plurality of camera devices may use a three-camera imaging system in which the first camera device 144_1 , the second camera device 144_2 , and the third camera device 144_3 have different fields of view. A three-camera imaging system can make decisions based on information obtained from objects located at various distances in front of and to the side of the vehicle.

The first camera device 144_1 may be connected to the first image processor to perform monocular image analysis on an image provided by the first camera device 144_1 . The second camera device 144_2 may be connected to the second image processor to perform monocular image analysis on an image provided by the second camera device 144_2 . Information processed and output by the first and second image processors may be merged. The second image processor may perform stereoscopic analysis by receiving images from both the first camera device 144_1 and the second camera device 144_2 . Monocular image analysis may refer to image analysis performed based on images captured from a single field of view (eg, images captured from a single camera). Stereoscopic image analysis may refer to image analysis performed based on two or more images captured with one or more image capture parameters (eg, images captured by each of two or more cameras). Captured images suitable for stereoscopic image analysis may include images captured from two or more locations, images captured from different fields of view, images captured using different focal lengths, and images captured according to parallax information.

8 is a block diagram illustrating a sparse map of a processing unit according to an embodiment of the present invention.

The processing unit 110 may include the sparse map 200 . The sparse map 200 may be used for autonomous driving. The sparse map 200 may provide information for navigation of an autonomous vehicle. The sparse map 200 and data processed by the sparse map 200 may be stored in a memory of the vehicle control system or transmitted and received to a remote server. The sparse map 200 may store and use a polynomial representation of one or more trajectories along the road by the vehicle. The sparse map 200 can be recognized as an object by simplifying the characteristics of a road section. The sparse map 200 can reduce the amount of data stored and transmitted/received for navigation of an autonomous vehicle. Sparse map 200 may include a polynomial representation 210 of a trajectory and landmarks 220 .

The polynomial expression 210 of the trajectory may be a polynomial expression of a target trajectory for guiding autonomous driving along a road section. The target trajectory may represent an ideal path for a vehicle to drive on a road section. The road section may be expressed as one or more target trajectories. The number of target trajectories may be less than the number of lanes included in the road section. A vehicle driving on the road may determine navigation by using one of the target trajectories and considering a lane and a lane offset corresponding to the target trajectory.

The landmark 220 may be a place or mark associated with a specific road section or a local map. The landmark 220 may be identified and stored in the smart map 200 . Intervals between landmarks 220 may be adjusted. The landmark 220 may be used for autonomous navigation. The landmark 220 may be used to determine the current position of the vehicle relative to the stored target trajectory. The self-driving vehicle may adjust the driving direction from the current location to match the direction of the target trajectory using current location information of the vehicle.

The landmark 220 may be used as a reference point for determining the location of a vehicle on a target trajectory. While driving based on dead reckoning in which the vehicle determines its own movement and estimates its position on the target trajectory, dead reckoning uses the location of the landmark 220 appearing on the sparse map 200 It is possible to eliminate errors in position determination caused by The landmark 220 identified in the sparse map 200 may serve as an anchor to accurately determine the location of the vehicle relative to the target trajectory.

9 is a diagram showing a polynomial expression of a trajectory according to an embodiment of the present invention.

The sparse map may include information about characteristics of the road. In the sparse map, the curved shape of the sections 212 included in the road 211 may be stored. Each of the sections 212 may have a curved shape that can be expressed as a polynomial. The road 211 may be modeled as a 3D polynomial expression formed by gathering curved shapes of respective lanes including left and right sides. A plurality of polynomials may be used to express the location and shape of each of the road 211 and sections 212 included in the road 211 . A polynomial representing each of the intervals 212 may define the position and shape of the interval 212 within a specified distance.

10 is a diagram showing landmarks according to an embodiment of the present invention.

Landmarks may include traffic signs, directional signs, roadside facilities, and general signs. The traffic sign may be a sign that guides traffic conditions and matters to be observed when passing. Traffic signs may include speed limit signs 221 , yield signs 222 , road number signs 223 , traffic signal signs 224 , and pause signs 225 . A directional sign may be a sign with one or more arrows indicating one or more directions to another place. Directional signs may include highway signs 226 with arrows directing vehicles to other roads or locations and exit signs 227 with arrows directing vehicles off the road. A general sign may be a sign providing information related to a place. General signs may include signboards 228 of famous restaurants in the area.

A sparse map may include a plurality of landmarks related to a road section. Simplified images of the actual images of each of the landmarks may be stored in the sparse map. A simplified image may be composed of data describing the characteristics of a landmark. An image stored in a sparse map can be represented and recognized using a smaller amount of data than a real image requires. Data representing landmarks may include information for describing or identifying landmarks formed along the road.

11 is a flowchart illustrating a method of generating a sparse map according to an embodiment of the present invention.

The vehicle control system may receive a plurality of images from a plurality of vehicles in operation 310 . Each of a plurality of cameras disposed in the vehicle may capture a plurality of images representing a situation around the vehicle by photographing a situation around the vehicle that the vehicle encounters while driving along a road section. A plurality of images representing a situation around the vehicle may represent the shape and condition of a driving route of the vehicle. The vehicle control system may receive a plurality of images captured by a plurality of cameras.

The vehicle control system may identify at least one feature on the road surface at operation 320 . The vehicle control system may simplify the representation of at least one line based on a plurality of images of features of a road surface extending along a road section. A line representation in which the characteristics of the road surface are simplified may indicate a path proceeding along a road section substantially corresponding to the characteristics of the road surface. The vehicle control system may analyze a plurality of images received from a plurality of cameras to identify an edge of a road or a lane marking. The vehicle control system may determine a driving trajectory along a road segment associated with a road edge or lane markings. Trajectory or line representations may include splines, polynomial representations, or curves. The vehicle control system may determine the driving trajectory of the vehicle based on the movement of the cameras themselves, such as 3D translational movement and/or 3D rotational movement.

In operation 330, the vehicle control system may identify a plurality of landmarks related to the road. The vehicle control system may identify one or more landmarks on the road section by analyzing the plurality of images received from the camera. Landmarks may include traffic signs, directional signs, roadside facilities, and general signs. The analysis may include rules for accepting and rejecting possible landmarks for a road segment. The analysis is based on rules that recognize potential landmarks when the ratio of images featuring a landmark to images where the landmark does not appear exceeds a threshold and/or rules for recognizing potential landmarks and/or landmarks not appearing on images where the landmark appears. It may include a rule to reject potential landmarks if the proportion of images that do not match exceeds a threshold.

12 is a flowchart illustrating a method of anonymously processing navigation information by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may determine at least one motion description for the vehicle at operation 410 . The vehicle control system may determine at least one motion description based on the output value of the sensor. The at least one motion description may include any indicator of motion of the vehicle. For example, the at least one motion description may include vehicle acceleration, vehicle speed, longitudinal and lateral positions of the vehicle at a specific point in time, a 3D position of the vehicle, and a determined trajectory of the vehicle.

The at least one motion description may include a motion description of the vehicle itself in a predetermined coordinate system. Self movement may include rotation, translation, movement of the vehicle in a lateral, longitudinal, or other direction. The movement of the vehicle itself may be expressed using vehicle speed, yaw rate, tilt, or roll. Its motion description for the vehicle can be judged for a specified level of freedom.

In operation 420, the vehicle control system may receive at least one image representing a surrounding situation of the vehicle. The vehicle control system may receive images of the road on which the vehicle travels and the surroundings of the vehicle from the camera.

In operation 430, the vehicle control system may analyze the image to determine road characteristics. The vehicle control system may analyze one or more images according to commands stored in the image analysis module or may determine at least one road characteristic by using a learning system such as a neural network. The one or more road characteristics may include features of the road, such as centerlines of the roads, edges of the roads, landmarks along the roads, potholes of the roads, turns of the roads, and the like. The at least one road characteristic may include lane characteristics including at least one indicator of detected lane separation, lane merging, dashed lane lane marking, solid lane lane marking, road surface color within the lane, lane color, lane direction, and lane type. . The lane characteristics may include a determination that the lane is a high occupant only (HOV) lane and is separated from other lanes by a solid line. The at least one road feature may include an indicator of a road edge. The road edge may be determined based on a detected barrier along the road edge, a detected sidewalk, a line representing the edge, a curb along the road edge, or the like, or based on detection of an object along the road.

In operation 440, the vehicle control system may collect section information on each of a plurality of sections included in the road. The vehicle control system may divide the road into a plurality of sections. The vehicle control system may collect section information on each of the plurality of sections by combining each of the plurality of sections with road characteristics. The section information may include at least one motion description of the vehicle and/or at least one road characteristic relative to the section of the road. The vehicle control system may collect section information including the motion description calculated in operation 410 and road characteristics determined in operation 430 .

13 is a flowchart illustrating a method of comparing a trajectory generated by a road navigation model with sensing data by a vehicle control system according to an embodiment of the present invention and controlling an autonomous driving mode.

The vehicle control system may apply the sparse map to the autonomous vehicle road navigation model. In the case of using a previously stored sparse map, additional judgment is required for the case where the road environment has changed. For example, when a sparse map is generated on a straight road, the route may be changed to a detour due to an event on the road such as construction. In this way, the trajectory can be changed when driving on an actual road, and a notification of a change of the trajectory and/or a notification of whether or not autonomous driving is to be terminated can be provided.

The vehicle control system may initiate an autonomous driving mode at operation 510 . The vehicle control system may drive based on the navigation set based on the sparse map when driving in the autonomous driving mode. The vehicle control system may drive by referring to a lane-related signal sensed by a front camera of an actual vehicle.

In operation 520, the vehicle control system may measure a path difference value between the sparse map and the sensing data. The vehicle control system may measure the curvature of the path recognized using the sparse map and the curvature of the path recognized using sensing data, respectively. Curvature can be expressed in units of R, which is the reciprocal of m. The vehicle control system may measure a curvature value of a path recognized by the sparse map and a difference value of the curvature of the recognized path using sensing data. For example, if the curvature value of the path recognized using the sparse map is 2000R and the curvature of the path recognized using the sensing data is 1900R, the vehicle control system sets the path difference value between the sparse map and the sensing data to 100R. can be measured with

The vehicle control system may recognize the recognized path as a curved road when the curvature of the recognized path is equal to or less than a specified value. The vehicle control system may recognize the recognized path as a straight road when the curvature of the recognized path is greater than a specified value. For example, when the curvature of the recognized path has a curvature of 3000R or less, the vehicle control system may recognize the recognized path as a curved road.

The vehicle control system may output a first warning based on the difference value in operation 530 . The vehicle control system may output a first warning when the difference value is greater than or equal to the first threshold value. The first threshold value may be 500R. For example, if the curvature value of the path recognized using the sparse map is 1000R and the curvature of the path recognized using the sensing data is 500R, the vehicle control system sets the path difference value between the sparse map and the sensing data to 500R. Therefore, the first warning may be output.

The vehicle control system may output a first warning when the type of path recognized using the sparse map and the type of path recognized using the sensing data are different from each other. The first threshold value may be 500R. For example, if the path recognized using the sparse map is a straight road with a curvature of 3100R and the path recognized using sensing data is a curved road with a curvature of 2900R, the vehicle control system uses the sparse map to Since the types of the recognized path and the path recognized using the sensing data are different from each other, a first warning may be output.

The vehicle control system may provide a visual warning notification to the driver by outputting the first warning in a pop-up form to an output unit such as a cluster of the vehicle. The first warning may include content that it is difficult to maintain current autonomous driving due to environmental changes on surrounding roads. For example, the vehicle control system may output a pop-up to the cluster saying "Switching to manual driving mode is required due to changes in the surrounding road environment."

The vehicle control system may output a second warning and end the autonomous driving mode in operation 540 . After outputting the first warning, the vehicle control system may output a second warning when there is no user input. The vehicle control system may output a second warning when the current autonomous driving mode is maintained after outputting the first warning. The second warning may be a warning having a higher intensity than the first warning. For example, in order to output the second warning, the vehicle control system may output a pop-up warning saying “Convert to manual driving mode” to the cluster and simultaneously output a warning notification sound audibly. After outputting the second warning, the vehicle control system may end the autonomous driving mode and switch to the manual driving mode based on a user input. The vehicle control system may detect that the autonomous driving mode cannot be maintained due to a change in the surrounding environment, notify the driver that the autonomous driving mode should be terminated, and terminate the autonomous driving mode based on a user input.

14 is a flowchart illustrating a method of comparing a path recognized using a sparse map with a path recognized using sensing data and controlling an autonomous driving mode by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may initiate an autonomous driving mode at operation 610 .

In operation 620, the vehicle control system may check whether both the sparse map and the sensing data are recognized as curves. The vehicle control system may check whether the path recognized using the sparse map and the path recognized using the sensing data are all curves. When the vehicle control system recognizes both the sparse map and the sensing data as curves (operation 620 - YES), it proceeds to operation 630 . When recognizing at least one of the sparse map and the sensing data as a straight line, the vehicle control system may proceed to operation 640 (operation 620 - NO).

In operation 630, the vehicle control system may determine whether a difference between the first curvature based on the sparse map and the second curvature based on the sensing data is greater than or equal to a first threshold value. The first curvature may be the forward curvature of the sparse map. The second curvature may be a forward road curvature based on front camera sensing data. The first threshold value may be 500R. When the difference between the first curvature based on the sparse map and the second curvature based on the sensing data in operation 630 is greater than or equal to the first threshold value (operation 630 - YES), the vehicle control system may proceed to operation 660 . When the difference between the first curvature based on the sparse map and the second curvature based on the sensing data is smaller than the first threshold value (operation 630 - NO), the vehicle control system may proceed to operation 650 .

In operation 640, the vehicle control system may check whether the road is recognized in a different form from the sparse map and the sensing data. The vehicle control system may determine whether a road having the same sparse map and sensing data is recognized as different types such as a straight road and a curved road. In operation 640, the vehicle control system may proceed to operation 660 when the road is recognized in a different form from the sparse map and the sensing data (operation 640 - YES). When the vehicle control system recognizes the road in the same form as the sparse map and the sensing data in operation 640 (operation 640 - NO), the vehicle control system may proceed to operation 650 .

The vehicle control system may continue driving in an autonomous driving mode at operation 650 . The vehicle control system may determine that the path according to the sparse map matches the path according to the sensing data and maintain the current autonomous driving mode.

In operation 660, the vehicle control system may determine that the navigation model of the sparse map and the actual road environment are different. The vehicle control system may determine that the actual road environment has become different from the information stored in the sparse map due to changes in the external environment such as road construction.

The vehicle control system may output a first warning in operation 670 . The vehicle control system may display a warning visually on the vehicle's output.

The vehicle control system may output a second warning in operation 680 . The vehicle control system may visually and audibly output a warning having a higher intensity than the first warning through an output unit of the vehicle. After outputting the first warning, the vehicle control system may output a second warning when there is no user input.

The vehicle control system may switch to the manual driving mode at operation 690. The vehicle control system may end the autonomous driving mode and start manual driving based on the user input.

15 is a diagram illustrating correction of a path generated by a vehicle control system based on a target trajectory according to a vehicle position according to an embodiment of the present invention.

The vehicle control system may set a trajectory along which the vehicle travels when the vehicle is positioned at the reference position 710 as a target trajectory. The vehicle control system may generate a path for the vehicle to travel based on the target trajectory.

The vehicle control system may detect the current location 720 of the vehicle. The vehicle control system may correct the path generated based on the target trajectory according to the current location 720 of the vehicle. The vehicle control system may control the vehicle so that the vehicle travels on the calibrated route.

In particular, in existing self-driving technology, when generating multiple routes through some target trajectories on a road with multiple lanes, multiple routes are created by considering only the offset amount of lanes, so that the lane offset of the existing target trajectory has A plurality of paths may have an error as it is, and an error value may increase in a process of generating a plurality of paths.

When the vehicle control system generates multiple routes through some target trajectories on a road with multiple lanes, it occurs when a plurality of routes are created in a state in which the vehicle is positioned in the center of the lane and a plurality of routes are generated. error can be reduced.

The vehicle control system may calculate the target trajectory, the width of the lane, and the current position of the vehicle on the lane. The vehicle control system may calculate the current location of the vehicle based on vehicle specifications. The vehicle control system may calculate the distance from the center of the vehicle to the left lane, the distance from the center of the vehicle to the right lane, and the width of the road using the sensors.

The vehicle control system may perform a correction operation of locating the calculated current vehicle position to the center of the road. The vehicle control system can correct the corrected position of the vehicle to be the same value from both lanes. The vehicle control system may correct the vehicle position so that the vehicle is separated from both lanes by a value obtained by dividing the distance by 2 after adding the distance from the center of the vehicle to the left lane and the distance from the center of the vehicle to the right lane.

The vehicle control system may generate a target trajectory based on the corrected position of the vehicle, i.e., the center of the road. The vehicle control system may generate a plurality of routes based on the location of the vehicle and the target trajectory. The vehicle control system may generate a plurality of paths in which an error is reduced by following the center of a lane rather than a plurality of paths based on a target trajectory generated biased toward one lane. The vehicle control system may generate each of the plurality of routes for each lane included in the lane.

16 is a flowchart illustrating a method of controlling driving of a vehicle by including a beacon of a dangerous section in a sparse map by a vehicle control system according to an embodiment of the present invention.

In operation 810, the vehicle control system may identify the dangerous section and classify the dangerous section according to its characteristics. The risk section may include an accident-prone section, an intersection section, and a long-term construction section. The danger section may include a high risk section requiring driving mode adjustment and a low/medium risk section requiring visual and/or audible notification to the driver through a warning mode. The accident-prone section may be a high-risk section. Intersection sections and long-term construction sections may be low/medium risk sections. The vehicle control system may specify a section in which a risk section such as an accident-prone section, an intersection section, and a long-term construction section is defined in a route, and classify the specific risk section into a high-risk section and a low/medium-risk section according to characteristics.

In operation 820, the vehicle control system may include the installation location of the beacon installed in the danger zone in the sparse map. A beacon may be a structure and/or electronic device for indicating a danger zone. The installation location of the beacon may be defined using latitude and longitude values. The vehicle control system may include the latitude and longitude values of each of the beacons installed in the danger zone in the sparse map.

In operation 830, the vehicle control system may receive a signal transmitted from the beacon from the vehicle. The beacon may transmit a signal including location information and information about a dangerous section. A signal transmitted by the beacon may be a low power Bluetooth signal. The vehicle control system may receive a low power Bluetooth signal transmitted from the beacon from the vehicle through the vehicle's Bluetooth function.

In operation 840, the vehicle control system may control driving of the vehicle based on the characteristics of the dangerous section included in the signal. The vehicle control system may determine whether the vehicle is approaching a dangerous section based on location information including the latitude and longitude values of the beacon received from the vehicle and information about the dangerous section included in the signal transmitted from the beacon. there is. When the vehicle is approaching a dangerous section, the vehicle control system may provide a notification to the driver according to characteristics of the dangerous section and control a driving mode of the vehicle.

The vehicle control system may reduce the control speed in the longitudinal direction below a certain speed when approaching or arriving at a high-risk section. The vehicle control system may output the reason for the control speed deceleration through the cluster. The vehicle control system may provide guidance on changing the driving mode to the driver. For example, the vehicle control system may output a visual guide through the cluster, saying "We are going to enter an accident-prone area in a while. We will drive by limiting the driving speed to 70 km/h or less."

The vehicle control system may provide a visual or audible notification according to the characteristics of the danger zone when approaching or arriving at a low/medium danger zone. The vehicle control system may provide guidance for low/medium risk sections to the driver. For example, the vehicle control system may output a visual guide saying "We will enter the intersection in a moment" through the cluster. As another example, the vehicle control system may output an audible guidance saying “You are entering a dangerous area.” The vehicle control system may repeatedly provide guidance.

17 is a flowchart illustrating a method of selecting data to be clustered according to reliability of a driving trajectory when the vehicle control system clusters driving trajectories according to an embodiment of the present invention.

In operation 910, the vehicle control system may cluster driving trajectories. The vehicle control system may collect data on a driving trajectory constituting a sparse map. Data that should not be included in clustering, such as driving data with a large error, must be excluded to reduce the error of the finally calculated driving path by calculating the average of the trajectory clusters.

In operation 920, the vehicle control system may determine whether the curvature of the driving trajectory is greater than or equal to the second threshold value. The second threshold value may be a curvature value for distinguishing a straight road and a curved road. The second threshold value may be 3000R. Depending on whether the driving trajectory is a straight road or a curved road, a criterion for determining reliability on a straight road and a curved road may be set differently. The vehicle control system may proceed to operation 930 when the curvature of the driving trajectory is equal to or greater than the second threshold (operation 920 - YES). When the curvature of the driving trajectory is smaller than the second threshold value (operation 920 - NO), the vehicle control system may proceed to operation 940 .

The vehicle control system may determine a straight road in operation 930 . The vehicle control system may determine a curved road in operation 940 . The vehicle control system may determine whether the driving trajectory is a straight road or a curved road based on the curvature of the driving trajectory.

The vehicle control system may determine the reliability of the driving trajectory based on the lateral acceleration value and the driver's torque in operation 950 . The vehicle control system may determine the absolute direction of the vehicle and whether or not it deviate from the lane assignment by using a diagnostic communication signal (CAN signal) output from the vehicle. The vehicle control system may determine whether the lateral acceleration value is greater than or equal to a specific threshold value in the case of a straight road. The vehicle control system may determine that the vehicle is driving abnormally when the lateral acceleration value is greater than or equal to a specific threshold value on a straight road. The vehicle control system may determine that the reliability of the driving trajectory is equal to or less than the third threshold value when the value of the lateral acceleration on the straight road is greater than or equal to a specific threshold value. The vehicle control system may determine whether the driver's torque is greater than or equal to a specific threshold value when the lateral acceleration value is smaller than the specific threshold value. A certain threshold of driver torque may be 5 Nm. When the driver's torque is greater than or equal to a specific threshold, the vehicle control system may determine that excessive steering wheel rotation has occurred due to the driver's intervention while the driver is gripping the steering device. The vehicle control system may determine that the reliability of the driving trajectory is equal to or less than the third threshold value when the driver's torque is greater than or equal to a specific threshold value.

In operation 960, the vehicle control system may determine the reliability of the driving trajectory based on the amount of steering angle change and the driver's torque. In the case of a curved road, the vehicle control system may determine whether the change amount of the steering angular velocity is greater than or equal to a specific threshold value. The vehicle control system may determine that the behavior of the vehicle has instantaneously and suddenly changed abnormally when the change amount of the steering angular velocity on the curved road is equal to or greater than a specific threshold value. The vehicle control system may determine that the reliability of the driving trajectory is equal to or less than the third threshold value when the variation value of the steering angular velocity on the curved road is equal to or greater than a specific threshold value. The vehicle control system may determine whether the driver's torque is greater than or equal to a specific threshold value when the change amount of the steering angular velocity is smaller than a specific threshold value. A certain threshold of driver torque may be 5 Nm. When the driver's torque is greater than or equal to a specific threshold, the vehicle control system may determine that excessive steering wheel rotation has occurred due to the driver's intervention while the driver is gripping the steering device. The vehicle control system may determine that the reliability of the driving trajectory is equal to or less than the third threshold value when the driver's torque is equal to or greater than a specific threshold value.

In operation 970, the vehicle control system may block sparse map update by excluding it from the clustering data when the reliability of the driving trajectory is equal to or less than the third threshold. The vehicle control system may exclude data having an unusual driving pattern in which the reliability of the driving trajectory is equal to or less than the third threshold value from the clustering process.

18 is a diagram illustrating driving trajectories excluded from clustering according to an embodiment of the present invention.

When the vehicle 980 drives with excessive lane changes on a multi-lane road while driving on a straight road, the lateral acceleration value may be greater than a specific threshold value. For example, when the vehicle 980 cuts in excessively and/or makes an unnecessary lane change, the lateral acceleration value may be greater than a certain threshold value. As another example, when making a left turn in an undesignated lane and/or a right turn in an undesignated lane, such as an illegal left/right turn or a left turn in a 2/3 lane, the lateral acceleration value may be greater than a specific threshold value. As another example, when making an illegal U-turn in an unspecified lane, the lateral acceleration value may be greater than a specific threshold value. The vehicle control system may exclude a driving trajectory having a lateral acceleration value greater than a specific threshold value from clustering.

19 is a diagram illustrating driving trajectories excluded from clustering according to an embodiment of the present invention.

When the vehicle 990 enters or exits a dangerous cornering rather than a typical cornering while driving on a curved road, the steering angular velocity variation value may be greater than a specific threshold value. The vehicle control system may exclude a driving trajectory in which the change amount of the steering angular velocity is greater than a specific threshold value from clustering.

20 is a flowchart illustrating a method of correcting an offset of a sensor by using a longitudinal gradient and a lateral gradient in a vehicle control system according to an embodiment of the present invention.

In operation 1010, the vehicle control system may calculate first values for longitudinal and lateral gradients of the vehicle based on the sensor. The vehicle control system may utilize acceleration sensors in longitudinal and lateral directions of the vehicle to calculate a longitudinal gradient value and a lateral gradient value at a corresponding position. The vehicle control system may additionally utilize the wheel speed sensor and/or the yaw rate sensor to calculate the longitudinal gradient value and the lateral gradient value. When constructing a 3D map, the vehicle control system may increase the accuracy of constructing a 3D map by determining a vertical gradient and a lateral gradient using a sensor of the vehicle.

In operation 1020, the vehicle control system may receive second values for longitudinal and lateral gradients of the vehicle calculated by the server based on the 3D map location information. The server may collect data about gradient information of a 3D map from a plurality of vehicles. The server may collect data on gradient information of a 3D map from a plurality of vehicles to calculate second values for longitudinal and lateral gradients. The vehicle control device may receive second values for longitudinal and lateral gradients of the vehicle from the server.

The vehicle control system may correct the offset of the sensor based on the difference between the first value and the second value in operation 1030 . In order to calibrate the acceleration sensor, a reference value or an error value may be used. A reference value or an error value may not be easy to obtain on an actual road. A reference value or an error value may be obtained using a difference from the 3D map location information. The vehicle control system may correct the offset of the acceleration sensor by using a difference between the longitudinal and lateral gradient values obtained using the sensor and the longitudinal and lateral gradient values obtained using the 3D map location information.

In operation 1040, the vehicle control system may control an advanced driver assistant system (ADAS) using the sensor whose offset is corrected. The driving assistance function may control driving of the vehicle by using information related to the longitudinal and lateral gradients of the vehicle. The vehicle control system may improve the accuracy of the driving assistance function by controlling the driving assistance function using a sensor that corrects an offset of the sensor and calculates an accurate vertical gradient value and a lateral gradient value.

21 is a flowchart illustrating a method of controlling a traveling speed and an amount of a gradient by using a vertical gradient and a lateral gradient by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may control the driving assistance function in operation 1110 . When the driving assistance function is controlled, a longitudinal gradient value and a lateral gradient value of the vehicle may be calculated using sensors provided in each of the plurality of vehicles.

In operation 1120, the vehicle control system may receive, from the server, values of a longitudinal gradient and a lateral gradient corresponding to a driving position of the vehicle, updated by the server based on the 3D map location information. The longitudinal gradient value and the lateral gradient value corresponding to the driving position of the vehicle may be updated based on the 3D map location information. When constructing the 3D map, the vehicle control system may update average values of longitudinal and lateral gradient values of each of the plurality of vehicles.

In operation 1130, the vehicle control system may control the driving speed by determining the slope of the current driving section based on the vertical gradient value. The vehicle control system may determine whether the section is uphill or downhill in the longitudinal direction. The vehicle control system may control the force used for driving to maintain the driving speed depending on whether it is an uphill or downhill section in the longitudinal direction. The vehicle control system may increase the force used for driving to maintain the driving speed in the case of an uphill section in the longitudinal direction. The vehicle control system may reduce the force used for driving to maintain the driving speed in the case of a downhill section in the longitudinal direction.

In operation 1140, the vehicle control system may control the amount of compensation for drift in a straight section or a control amount in a curved section of the driving assistance function based on the horizontal slope value. The vehicle control system may control the relative position of the vehicle on the lane based on the slope value. The vehicle control system may control the vehicle to be centered in a straight section based on the value of the horizontal slope. The vehicle control system may control the curvature of the driving path in the curved section based on the lateral slope value.

22 is a flowchart illustrating a method for calculating an effective trajectory by a vehicle control system according to an embodiment of the present invention. After calculating a plurality of trajectories in order to calculate the validity trajectory, a trajectory having finally secured validity may be calculated by removing trajectories having effectiveness lower than a certain criterion and combining the trajectories having effectiveness higher than a certain criterion.

In operation 1210, the vehicle control system may determine the noise level of the road surface of the trajectory by calculating the variance of the movement value of each trajectory measured by the wheel speed sensor. The vehicle control system may calculate a difference value between a maximum value and a minimum value for a designated time period for each of the plurality of wheel speed sensors. The vehicle control system may calculate a movement variance value for a specified time based on a difference value of each of the plurality of wheel speed sensors. The vehicle control system may determine an average noise level of the road surface based on the movement variance value. The noise level of the road surface may be divided into levels from 0 to 5 levels.

In operation 1220, the vehicle control system may calculate trajectory information in both directions of the current point of the 3D map.

In operation 1230, the vehicle control system may check whether the width of the road at the current point is greater than or equal to a fourth threshold value. The vehicle control system may proceed to operation 1240 when the road width at the current point is equal to or greater than the fourth threshold (operation 1230 - YES). The vehicle control system may return to operation 1220 when the road width at the current point is smaller than the fourth threshold (operation 1230 - NO).

In operation 1240, the vehicle control system may check whether there is no overlapping section based on the vehicle width among trajectories in both directions. When there is no overlapping section (operation 1240 - YES), the vehicle control system may proceed to operation 1250. The vehicle control system may return to operation 1220 if there is an overlapping section (operation 1240 - NO).

In operation 1250, the vehicle control system may update trajectory information in both directions of the current point and the noise level. The vehicle control system may update the trajectory values in both directions and the road noise level when there are no overlapping trajectories among the trajectories in both directions.

The vehicle control system may assign a weight to each trajectory based on the noise level in operation 1260 . When calculating the final effective trajectory, the vehicle control system may assign different weights to different positions of each trajectory according to the noise level.

In operation 1270, the vehicle control system may reflect the final trajectory whose validity is secured on the 3D map. The vehicle control system may calculate a final trajectory for which validity is secured. The vehicle control system can improve the reliability of the autonomous driving system by reflecting a more reliable trajectory when constructing a 3D map.

23 is a flowchart illustrating a method of setting landmarks by reflecting vertical and lateral gradient information in a vehicle control system according to an embodiment of the present invention.

In operation 1310, the vehicle control system may select landmarks based on camera image information. The vehicle control system may select existing landmarks and calculate longitudinal and lateral gradients using sensors of the vehicle. The vehicle control system may transmit the calculated longitudinal and lateral gradient values to the server. The server may calculate an average value of the longitudinal and lateral gradient values. The server may update the average value of the vertical gradient value and the horizontal gradient value in the 3D map location information.

In operation 1320, the vehicle control system may divide sections between the selected landmarks in units of designated distances. The vehicle control system may receive 3D map location information obtained by updating an average value of a longitudinal gradient value and a lateral gradient value. The vehicle control system may divide landmarks into unit lengths having specified distances based on the average values of the longitudinal and lateral gradient values.

In operation 1330, the vehicle control system may reflect longitudinal gradient and lateral gradient information received from the server to each divided distance unit. When the longitudinal and lateral gradients are reflected, the composition of the 3D map can be improved. The vehicle control system may reflect longitudinal gradient and lateral gradient information for each unit length having a specified distance.

In operation 1340, the vehicle control system may set sections including divided distance units reflecting longitudinal and lateral slope information as new landmarks. The vehicle control system may set a new landmark having a feature point by ordering ratios of vertical gradients and horizontal gradients based on a constant distance between landmarks. Accuracy and reliability of the 3D map may increase as the number of landmarks having feature points increases. The vehicle control system may increase accuracy and reliability of the 3D map by setting a distance between landmarks including a gradient ratio as a new landmark in addition to existing landmarks.

24 is a flowchart illustrating a method for a vehicle control system to control driving and functions of a vehicle using a beacon in an area where a navigation satellite signal is not received according to an embodiment of the present invention. The navigation satellite signal may be a signal supported by a Global Navigation Satellite System (GNSS). When a navigation satellite signal is received, it is possible to check whether the current location and driving route of the vehicle are accurate. The area where the navigation satellite signal is not received may be a closed space where the navigation satellite signal cannot reach, such as a tunnel or an underground road.

At least one beacon may be installed in the target section at designated intervals. The target section may be a section included in an area where navigation satellite signals are not received. The target section may be a tunnel section or an underground road section. Beacons can be installed in targeted areas such as tunnels or underpasses. At least one beacon may be installed at designated intervals so that a signal transmitted from the beacon can be detected with a designated strength or higher when the vehicle travels in a target section.

In operation 1410, the vehicle control system may classify at least one beacon installed in the target section at specified intervals according to the installation location. The vehicle control system may classify the at least one beacon into a tunnel beacon and an underpass beacon. The vehicle control system may classify the tunnel beacon into a tunnel entrance road beacon, a tunnel exit road beacon, and a tunnel inner beacon. The vehicle control system may classify the underpass beacon into an underpass access road beacon, an underpass access road beacon, and an underpass interior beacon. The vehicle control system may label each type of at least one beacon.

In operation 1420, the vehicle control system may include at least one beacon classified according to the installation location in the sparse map. The vehicle control system may include each of the at least one beacon whose type is labeled in the sparse map. The vehicle control system may measure the latitude and longitude of the installation location of each of the at least one beacon. The vehicle control system may include information on the latitude and longitude of the installation location of each of the at least one beacon in the sparse map.

In operation 1430, the vehicle control system may determine the accuracy of the driving trajectory based on a signal output from at least one beacon. A signal output from at least one beacon may be a Bluetooth communication signal. The vehicle control system may receive the signal of the tunnel access road beacon and check the latitude and longitude of the installation location of the tunnel access road beacon. The vehicle control system may confirm the signal of the beacon on the way out of the tunnel and end the tunnel entry mode.

The vehicle control system compares the Bluetooth intensity of a beacon representing a lane during autonomous navigation driving to determine the accuracy of a trajectory in progress and determines whether to use it for trajectory update of a sparse map. Beacons inside the tunnel may be classified for each lane. The frequency and/or waveform of a signal transmitted from a beacon may be different for each lane. The vehicle control system may correct driving accuracy by using the intensity of beacons classified for each lane. Signals transmitted from beacons may be measured and grouped for a specified period of time. When signals transmitted from beacons are grouped, the accuracy of measured signal strength can be improved.

When a vehicle drives on two lanes of a two-lane road, the vehicle control system may determine whether the intensity of a signal transmitted from a beacon installed on the second lane is greater than the intensity of a signal transmitted from a beacon installed on the first lane. When the strength of the signal transmitted from the beacon installed in the second lane is detected to be greater than the strength of the signal transmitted from the beacon installed in the first lane, the vehicle control system may maintain the current driving mode. The vehicle control system may use the driving trajectory for updating the sparse map trajectory when maintaining the current driving mode.

When a vehicle drives on two lanes of a two-lane road, the vehicle control system checks whether or not the intensity of the signal transmitted from the beacon installed on the second lane is greater than the intensity of the signal transmitted from the beacon installed on the first lane for a certain period of time or longer. can If the intensity of the signal transmitted from the beacon installed on the first lane is detected to be greater than the intensity of the signal transmitted from the beacon installed on the second lane over a certain period of time, the vehicle control system converts the currently driving trajectory of the tunnel or underpass to the trajectory of the sparse map Can be excluded from updates.

In operation 1440, the vehicle control system may determine whether to enter or leave the target section based on a signal output from at least one beacon. The vehicle control system may control convenience functions of the vehicle based on whether the vehicle enters or leaves the target section. The vehicle control system may change to a tunnel entry mode upon receiving a signal from a tunnel entry road beacon. The vehicle control system may issue a notification and/or warning notifying the driver of the vehicle that the vehicle has entered the current tunnel. The vehicle control system may indicate that the vehicle has entered the current tunnel through the cluster of the vehicle. The vehicle control system may change to a tunnel driving mode when receiving a signal from a beacon inside a tunnel. The vehicle control system can automatically close the vehicle's windows in tunnel driving mode. The vehicle control system may execute the vehicle's air cleaning function in the tunnel driving mode. The vehicle control system can automatically turn on the headlights in front of the vehicle in tunnel driving mode. The vehicle control system may confirm the signal of the beacon on the way out of the tunnel and end the tunnel entry mode.

25 is a diagram illustrating that a vehicle control system according to an embodiment of the present invention creates a virtual lane in a section where a lane is broken on a vehicle movement path.

The vehicle control system may detect a lane in a section in which the vehicle 1501 is driving. The vehicle control system may detect a section where the lane is broken on the vehicle movement path. The vehicle control system may create a virtual lane in a section where the lane is broken. The vehicle control system may proceed with autonomous driving assuming that a virtual lane is drawn on the road.

26 is a flowchart illustrating a method of controlling driving of a vehicle by creating a virtual lane by a vehicle control system according to an embodiment of the present invention.

In operation 1510, the vehicle control system may recognize both lanes on the driving route by using the sensor unit. The vehicle control system may recognize the left lane of the vehicle and the right lane of the vehicle by using a sensor for detecting the lane. Through the lane detection module constituting the sparse map, various types of lanes on actual roads can be databased.

In operation 1520, the vehicle control system may determine whether the recognition degree of any one lane among lanes on both sides of the driving route, recognized using the sensor unit, is equal to or less than a fifth threshold value. The recognition degree of the lane may refer to the degree to which the lane is visually recognized. The vehicle control system may determine that the recognition degree of the lane is equal to or less than the fifth threshold value when the lane is blurred, broken, or erased, and/or the lane is covered with foreign substances such as tar marks. The vehicle control system may proceed to operation 1530 when the recognition degree of at least one lane is equal to or less than the fifth threshold (operation 1520 - YES). The vehicle control system may return to operation 1510 when both lanes have recognition degrees higher than the fifth threshold (operation 1520 - NO).

In operation 1530, the vehicle control system generates a virtual lane based on at least one of the remaining lanes or guide lines other than any one lane, the curvature of the road ahead, and the width between both lanes of the previous driving portion of the driving route. can do. When the remaining lanes other than the one where the degree of recognition of the lane is equal to or less than the fifth threshold are recognized, the vehicle control system based on the recognized remaining lanes, the curvature of the road ahead, and the width between both lanes in the previous section A virtual lane may be generated on a side where the recognition degree of the lane is equal to or less than the fifth threshold. The vehicle control system may create a virtual lane by filling a disconnected or unclear section of any one lane. If the vehicle's driving route corresponds to a route with a guide line, such as a highway on-ramp, highway exit, or toll area, the vehicle control system determines the guide line, the curvature of the road ahead, and the width between the two lanes in the previous section. Based on , a virtual lane may be generated on a side where the recognition degree of the lane is equal to or less than the fifth threshold value. The vehicle control system may create a virtual lane by filling a disconnected or unclear section of any one lane.

The vehicle control system may control the vehicle control unit to control driving of the vehicle based on the virtual lane generated in operation 1540 . The vehicle control system may proceed with autonomous driving assuming that the created virtual lane is drawn on the road.

The vehicle control system may update the virtual lane generated in operation 1550 to the sparse map. The vehicle control system may control the vehicle control unit to drive by reflecting the additional lane when entering a driving section in which the virtual lane is updated. Vehicle control systems can improve the accuracy and reliability of autonomous navigation models in areas where lanes are not properly drawn.

27 is a flowchart illustrating a method of adjusting a driving control right range according to trajectory accuracy by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may initiate an autonomous driving mode at operation 1610 .

In operation 1620, the vehicle control system may compare a driving trajectory based on a sparse map and an actual driving trajectory. The vehicle control system compares a preset trajectory included in the sparse map with an actual driving trajectory to verify a current driving path of the vehicle and determine accuracy of the trajectory. The vehicle control system may determine an error in the driving trajectory that may occur when driving on the road based on the trajectory included in the sparse map.

In operation 1630, the vehicle control system may calculate a center offset value that is a difference between the center of the sparse map-based driving trajectory and the center of the actual driving trajectory. The center offset value may be a value representing an error between the two trajectories by comparing the sparse map and the actual driving trajectory based on the center of the vehicle.

In operation 1640, the vehicle control system may check whether the center offset value falls within the first range. The first range may be a length range of 0 or more and 20 cm or less. The vehicle control system may check whether the center offset value falls within the first range for a specified period of time. The specified time may be 10 seconds. The vehicle control system may proceed to operation 1650 when the center offset value falls within the first range (operation 1640 - YES). The vehicle control system may proceed to operation 1660 when the center offset value is out of the first range (operation 1640 - NO).

In operation 1650, the vehicle control system may set the range of driving control right of the vehicle to a range corresponding to the autonomous driving mode. When the center offset value falls within the first range, the vehicle control system may determine that an error value between the sparse map-based driving trajectory and the actual driving trajectory is acceptable. The vehicle control system may determine that since the current autonomous driving mode is reliable, safety and reliability of driving may be secured even when the driving control right of the vehicle is set to the autonomous driving mode.

In operation 1660, the vehicle control system may check whether the center offset value falls within the second range. The second range may be a length range of 20 cm or more and 30 cm or less. The vehicle control system may check whether the center offset value falls within the second range for a specified period of time. The vehicle control system may proceed to operation 1670 when the center offset value falls within the second range (operation 1660 - YES). The vehicle control system may proceed to operation 1680 when the center offset value is out of the second range (operation 1660 - NO).

In operation 1670, the vehicle control system may change the range of the driving control right of the vehicle to a range corresponding to the partial autonomous driving mode. When the center offset value falls within the second range, the vehicle control system may determine that an error value between the sparse map-based driving trajectory and the actual driving trajectory has reached a level requiring correction. The vehicle control system may partially limit the range of driving control right of the vehicle. The vehicle control system may determine that it is necessary to correct the driving trajectory applied in the autonomous driving mode. The vehicle control system can modify the driving trajectory applied in autonomous driving mode to match the actual road. The vehicle control system may guide the driver to correct the trajectory by warning the driver that the driving trajectory needs to be corrected.

In operation 1680, the vehicle control system may check whether the center offset value falls within the third range. The third range may be a length range of 30 cm or more. The vehicle control system may check whether the center offset value falls within the third range for a specified period of time. The vehicle control system may proceed to operation 1690 when the center offset value falls within the third range (operation 1680 - YES).

In operation 1690, the vehicle control system may change the driving control right range of the vehicle to a range corresponding to the manual driving mode. When the center offset value falls within the third range, the vehicle control system may determine that an error value between the sparse map-based driving trajectory and the actual driving trajectory has reached a level at which the autonomous driving mode should be terminated. The vehicle control system may end the autonomous driving mode and change to the manual driving mode by limiting the entire driving control right of the vehicle. The vehicle control system may guide the driver to manually drive by warning the driver that the autonomous driving mode changes to the manual driving mode because the trajectory does not match.

28 is a flowchart illustrating a method of improving reliability of a trajectory when a vehicle control system calculates a trajectory of a vehicle using a server according to an embodiment of the present invention.

The vehicle control system may initiate an autonomous driving mode at operation 1710 . When constructing a 3D map in autonomous driving mode, the trajectory location of each vehicle can be clustered more reliably.

In operation 1720, the vehicle control system may obtain other vehicle information and lane information using the sensor unit. The vehicle control system may obtain lane information by integrating information of vehicles recognized using the sensor unit and GPS information.

The vehicle control system may transmit other vehicle information and lane information acquired in operation 1730 to the server. The vehicle control system may transmit vehicle information, lane information, and GPS information recognized by using the sensor unit to the cloud server.

The vehicle control system may receive information related to adjacent vehicles from the server in operation 1740 . The server may determine an adjacent vehicle adjacent to the target vehicle among other vehicles. The server uses the distance between the plurality of vehicles included in the GPS information of each of the plurality of vehicles, the moving speed of each of the plurality of vehicles, and the moving direction of each of the plurality of vehicles to drive the target vehicle. It is possible to check whether a vehicle is driving in a lane adjacent to the vehicle. The server can check how many lanes the corresponding road has from the navigation map. The server may combine the recognition information of each of the plurality of vehicles on the navigation map to predict whether the lane information of each of the plurality of vehicles is a first lane or a second lane. The server may transmit information related to adjacent vehicles to the target vehicle. The vehicle control system may receive information related to adjacent vehicles from the server.

In operation 1750, the vehicle control system may determine whether an adjacent vehicle has changed lanes. The vehicle control system may determine whether an adjacent vehicle has changed lanes. When a lane on the road is recognized, the vehicle control system determines the similarity between the visual range of the lane in the direction of the position of the adjacent vehicle and the longitudinal distance of the object, and checks whether the signal of the visual range of the lane is discontinuous, so that the adjacent vehicle is in the lane. You can check whether or not it has been changed. If the lane on the road is not recognized, the vehicle control system creates a virtual lane based on the yaw rate based on the own vehicle, and checks whether the adjacent vehicle invades the virtual lane so that the adjacent vehicle changes the lane. You can check whether it has been done.

In operation 1760, the vehicle control system may transmit information related to whether the adjacent vehicle has changed lanes to the server so that the server determines whether the lane change of the adjacent vehicle is an effective lane change. The server may determine whether the lane change is valid and exclude unnecessary lane change trajectories from trajectory calculation. The server may determine whether a lane change in each lane information is an unnecessary lane change or a necessary lane change by utilizing the navigation map information. The server may determine whether the corresponding lane change is an effective trajectory by checking whether or not a lane change is made for each lane of the corresponding road. The server may determine whether the corresponding lane change is a valid trajectory by utilizing surrounding information such as toll gate information, road access road information, and road access road information. When the server determines that the corresponding lane change is an unnecessary lane change trajectory that is a lane change excluding an effective lane change, it may be excluded from calculating the effective trajectory. The server can cluster only reliable trajectories for each lane. Accordingly, when constructing a 3D map, the reliability of the autonomous driving system can be increased by calculating more reliable trajectory information.

29 is a flowchart illustrating a method of adjusting a control target point for a cut-in vehicle according to a camera position by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may initiate an autonomous driving mode at operation 1810 . In order to lower the risk of an accident in an autonomous driving mode, it may be important to quickly determine whether a preceding vehicle cuts in front of the host vehicle. If the control target point is adjusted according to the cut-in, the stability of autonomous driving can be secured.

In operation 1820, the vehicle control system may recognize a preceding vehicle using a plurality of cameras. A plurality of cameras may be included in the photographing unit of the vehicle. Each of the plurality of cameras may be mounted at different positions of the vehicle. The vehicle control system may recognize a preceding vehicle using each of a plurality of cameras mounted at different locations.

In operation 1830, the vehicle control system may predict an angle at which the preceding vehicle is directed toward the lane based on a difference between image information acquired from each of the plurality of cameras. A difference may occur in image information recognized by each of a plurality of cameras mounted at different locations. The vehicle control system may calculate a heading angle, which is an angle formed by the preceding vehicle compared to the own vehicle, from the difference in image information.

In operation 1840, the vehicle control system may calculate the time remaining until lane departure by calculating the speed of the preceding vehicle in the lateral direction based on the angle at which the preceding vehicle is directed. The time remaining until lane departure may be referred to as Time to Lane Crossing (TTLC). The vehicle control system may calculate the speed of the preceding vehicle in the lateral direction using the calculated bow angle. The vehicle control system may use the calculated lateral speed of the preceding vehicle to calculate the remaining time until the preceding vehicle leaves the lane and invades the lane of the subject vehicle.

The vehicle control system may calculate a deflection driving factor based on the time until the vehicle meets the preceding vehicle in the longitudinal direction in operation 1850 and the speed in the lateral direction. The time until it meets the preceding vehicle in the longitudinal direction may be referred to as Time to Crashing (TTC). The vehicle control system may calculate the deflection driving coefficient as a value obtained by dividing the time until the vehicle meets the preceding vehicle in the longitudinal direction by the speed in the lateral direction.

The vehicle control system may control the vehicle controller based on the deflection driving coefficient in operation 1860 . The vehicle control system may start the vehicle controller's deflection control to deviate to the opposite lane according to the deflection coefficient. The vehicle control system may initiate variable skew travel control according to the skew travel coefficient. The vehicle control system may determine whether a preceding vehicle has cut in and deflect the control point in the lateral direction to the opposite side of the preceding vehicle to cut in, thereby increasing the stability of autonomous driving.

30 is a flowchart illustrating a method of setting an entrance of a tunnel as a landmark by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may initiate an autonomous driving mode at operation 1910 . In the autonomous driving mode, the reliability and usability of the map may increase as the 3D map has more landmarks having feature points.

In operation 1920, the vehicle control system may check whether the illuminance sensor of the vehicle is activated at a time within a designated time range. The specified time range may be daytime. The vehicle control system may proceed to operation 1930 when the illuminance sensor of the vehicle is activated at a time within a specified time range (operation 1920 - Yes). When the vehicle's illuminance sensor is inactivated at a time within a specified time range (operation 1920 - No), the vehicle control system determines that the vehicle is running normally and returns to operation 1910.

In operation 1930, the vehicle control system may check whether lights of surrounding vehicles are turned on. The vehicle control system may proceed to operation 1940 when the lights of surrounding vehicles are turned on (operation 1930 - Yes). When the lights of surrounding vehicles are turned off (operation 1930 - No), the vehicle control system may determine that the illuminance sensor of the vehicle is turned on temporarily and return to operation 1910 .

In operation 1940, the vehicle control system may calculate an estimated trajectory based on the position of at least one light in the tunnel. The vehicle control system may calculate an equation representing a trajectory connecting lights in a tunnel as a lane equation corresponding to an estimated trajectory.

In operation 1950, the vehicle control system may check the degree of similarity between the estimated trajectory and the trajectory of the lane recognized using the sensor unit. The vehicle control system may proceed to operation 1960 when the estimated trajectory and the trajectory of the lane recognized using the sensor unit are more similar than the specified degree of similarity (operation 1950 - Yes). When the vehicle control system recognizes the lane trajectory using the estimated trajectory and the sensor unit and has a similarity lower than the specified similarity (Operation 1950 - No), the surroundings are temporarily darkened due to a temporary cause that occurred while driving, and the illuminance sensor of the own vehicle And surrounding vehicles may determine that the light is turned on for a temporary reason and return to operation 1910 .

In operation 1960, the vehicle control system may determine a driving situation in which the vehicle enters the tunnel. The vehicle control system may determine that the vehicle enters the tunnel when trajectories of the lanes recognized using the lane equation and the sensor unit are similar. The vehicle control system may designate a current location as location information of a tunnel entrance in a driving situation in which a vehicle enters a tunnel. The vehicle control system may transmit location information of the tunnel entrance to the server. Since GPS information is poor in a tunnel, it is not easy to accurately determine the location of a vehicle when only GPS signals are used. When the vehicle control system determines that the vehicle is entering the tunnel, the vehicle control system may set the current location as a feature point and turn the vicinity of the tunnel entrance into a landmark. The vehicle control system may landmark the vicinity of the tunnel entrance. The vehicle control system may construct a 3D map by inversely calculating the position of the entrance of the tunnel according to the GPS signal. The vehicle control system can increase the accuracy and reliability of the 3D map by setting more landmarks when constructing the 3D map by making the vicinity of the tunnel entrance a landmark.

31 is a flowchart illustrating a method of setting the inside of a tunnel as a landmark by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may initiate an autonomous driving mode at operation 1910 . In the autonomous driving mode, the reliability and usability of the map may increase as the 3D map has more landmarks having feature points.

In operation 1920, the vehicle control system may check whether the illuminance sensor of the vehicle is activated at a time within a designated time range. The specified time range may be daytime. The vehicle control system may proceed to operation 1930 when the illuminance sensor of the vehicle is activated at a time within a specified time range (operation 1920 - Yes). When the vehicle's illuminance sensor is inactivated at a time within a specified time range (operation 1920 - No), the vehicle control system determines that the vehicle is running normally and returns to operation 1910.

In operation 1930, the vehicle control system may check whether lights of surrounding vehicles are turned on. The vehicle control system may proceed to operation 1940 when the lights of surrounding vehicles are turned on (operation 1930 - Yes). When the lights of surrounding vehicles are turned off (operation 1930 - No), the vehicle control system may determine that the illuminance sensor of the vehicle is turned on temporarily and return to operation 1910 .

In operation 1940, the vehicle control system may calculate an estimated trajectory based on the position of at least one light in the tunnel. The vehicle control system may calculate an equation representing a trajectory connecting lights in a tunnel as a lane equation corresponding to an estimated trajectory.

In operation 1950, the vehicle control system may check the degree of similarity between the estimated trajectory and the trajectory of the lane recognized using the sensor unit. The vehicle control system may proceed to operation 1960 when the estimated trajectory and the trajectory of the lane recognized using the sensor unit are more similar than the specified degree of similarity (operation 1950 - Yes). When the vehicle control system recognizes the lane trajectory using the estimated trajectory and the sensor unit and has a similarity lower than the specified similarity (Operation 1950 - No), the surroundings are temporarily darkened due to a temporary cause that occurred while driving, and the illuminance sensor of the own vehicle And surrounding vehicles may determine that the light is turned on for a temporary reason and return to operation 1910 .

In operation 2010, the vehicle control system may calculate at least one road surface feature point in the tunnel. At least one road surface feature point may include gradient information including a noise level of the road surface and a longitudinal gradient and a lateral gradient. The noise level of the road surface can be calculated using the wheel speed sensor.

In operation 2020, the vehicle control system may detect the distance between locations where at least one light is installed and the number of locations where at least one light is installed in the tunnel. The vehicle control system may store locations where lights are installed as coordinate values.

In operation 2030, the vehicle control system may generate landmark information corresponding to a tunnel including continuous road surface feature points among at least one road surface feature point and information related to locations where lights are installed. The vehicle control system may store landmarks including road surface feature points and installation position information such as lights. If the tunnel itself is landmarked, the reliability of the 3D map in the tunnel can be improved.

32 is a flowchart illustrating a method of correcting a sensor unit by using a portable communication device in a vehicle control system according to an embodiment of the present invention. Errors such as offsets may occur in the sensor unit of the vehicle due to various factors such as production errors. In particular, in the case of an acceleration sensor, it may not be easy to calibrate after the vehicle is produced.

The vehicle control system may initiate an autonomous driving mode at operation 2110 .

In operation 2120, the portable communication device may be installed at a specific point in the vehicle. The portable communication device may be a smart phone. An accommodating part of a portable communication device such as a cradle may be installed at a specific point in a vehicle. The receptacle may be fixed at a specific point. Accordingly, the portable communication device can be fixed to a specific point in the vehicle.

In operation 2130, the vehicle control system may receive acceleration information and rotation information of the vehicle from the portable communication device. A sensor of the portable communication device may obtain acceleration information and rotation information of the vehicle. Acceleration information and rotation information of the vehicle acquired by the portable communication device may be received.

In operation 2140, the vehicle control system may correct the acceleration information and rotation information received from the portable communication device to match the position of the sensor unit as much as the difference between the position of the portable communication device and the position of the sensor unit of the vehicle. An arrangement position of the sensor unit of the vehicle and a position where the portable communication device is mounted may be different from each other. The vehicle control system may correct acceleration information and rotation information of the vehicle acquired by the portable communication device to match the location of the vehicle sensor unit.

The vehicle control system may correct the offset of the sensor unit by comparing the corrected acceleration information and rotation information in operation 2150 with the acceleration information and rotation information detected by the sensor unit. The vehicle control system may compare the calibrated acceleration information and rotation information of the vehicle obtained by the portable communication device with vehicle acceleration information and rotation information acquired by the sensor unit of the vehicle. The vehicle control system may determine that an offset of the sensor unit of the vehicle has occurred by the difference value of the comparison result. The vehicle control system may compensate for an offset of the sensor unit of the vehicle by a difference value as a result of the comparison. The vehicle control system may calibrate the sensor unit of the vehicle to correspond with the sensor of the portable communication device. The vehicle control system may improve the accuracy of the sensor unit of the vehicle. The vehicle control system may improve stability of autonomous driving by using a sensor unit of a vehicle having improved accuracy.

33 is a flowchart illustrating a method of correcting a GPS signal in consideration of characteristics of a portable communication device and a road by a vehicle control system according to an embodiment of the present invention.

The vehicle control system may initiate an autonomous driving mode in operation 2210 . Errors may occur in a GPS mounted in a vehicle due to various reasons.

In operation 2220, the vehicle control system may receive first location information from the portable communication device. The portable communication device may be a smart phone. The vehicle control system may receive GPS information from a portable communication device.

In operation 2230, the vehicle control system may determine whether a difference value between the first location information and the second location information received from the sensor unit of the vehicle is greater than or equal to a sixth threshold value. The vehicle control system may compare GPS information obtained from the GPS of the vehicle and GPS information received from the portable communication device in real time. The vehicle control system may proceed to operation 2240 when the difference between the GPS information obtained from the GPS of the vehicle and the GPS information received from the portable communication device is equal to or greater than the sixth threshold value (operation 2230 - Yes). The vehicle control system performs autonomous driving when the difference between the GPS information obtained from the GPS of the vehicle and the GPS information received from the portable communication device is smaller than the sixth threshold (Operation 2230 - No) when the accuracy of the GPS value of the vehicle is reduced. It is determined that the condition for doing this is satisfied, and operation 2220 may be returned.

In operation 2240, the vehicle control system may compare a first road feature point corresponding to the first location and a second road feature point corresponding to the second location information. The first road feature point and the second road feature point may include the curvature of the road at the location where the difference value is generated, the amount of the longitudinal gradient of the road, the amount of the lateral gradient of the road, and the noise level of the road surface. The vehicle control system may compare curvature information at the current location, a longitudinal gradient of the road, a lateral gradient of the road, and a noise level of the road surface. The vehicle control system may compare feature points of the road when a difference between the GPS information obtained from the GPS of the vehicle and the GPS information received from the portable communication device is greater than or equal to a sixth threshold value.

In operation 2250, the vehicle control system may determine information to be used for autonomous driving from among the first location information and the second location information based on the comparison result. The vehicle control system may compare the first road feature point and the second road feature point, and select location information having a large similarity value with road conditions among the first location information and the second location information as information to be used for autonomous driving. The vehicle control system may improve the accuracy and reliability of the location information by selecting location information having a more similar characteristic point among the first location information and the second location information received in different ways. The vehicle control system may correct the location of the vehicle according to the GPS by predicting the driving route of the vehicle on the GPS in consideration of the GPS information received from the portable communication device in a situation where the GPS signal of the vehicle is disconnected. The vehicle control system may correct the location of the vehicle according to the GPS by predicting the driving path of the vehicle on the GPS in consideration of the continuity of road driving characteristics in a situation where the GPS signal of the vehicle is disconnected.

Claims (11)

In the vehicle control system,
a processing unit that processes data related to driving of the vehicle; and
Including a photographing unit for detecting and photographing the external environment of the vehicle,
The processing unit,
Selecting landmarks based on image information acquired by the photographing unit;
Dividing sections between the selected landmarks in units of designated distances;
Reflecting longitudinal gradient and lateral gradient information received from the server in each of the divided distance units, and
Characterized in that, the vehicle control system sets the sections including the divided distance unit reflecting the longitudinal slope and the lateral slope information as new landmarks. The method of claim 1,
an input unit receiving a user input for controlling driving functions of the vehicle;
a sensor unit acquiring data related to driving of the vehicle from the vehicle and an external environment;
an output unit providing information related to driving of the vehicle; and
Characterized in that, the vehicle control system further comprises a vehicle control unit for controlling the driving of the vehicle. The method of claim 2,
The processing unit,
At the same time as selecting an existing landmark, the vehicle control system, characterized in that for calculating the vertical gradient and the horizontal gradient by utilizing the sensor unit. The method of claim 1,
The processing unit,
Characterized in that, based on the average value of the vertical gradient value and the horizontal gradient value, the landmarks are divided into the specified distance unit, the vehicle control system. The method of claim 4,
The processing unit,
Characterized in that the longitudinal gradient and the lateral gradient information are reflected for each specified distance unit, the vehicle control system. The method of claim 1,
The processing unit,
The vehicle control system, characterized in that by ordering the ratio of the vertical gradient and the horizontal gradient based on a constant distance between the landmarks and setting them as the new landmark having a feature point. In the driving method of a vehicle using a vehicle control system,
selecting landmarks based on image information acquired by the photographing unit of the vehicle control system;
dividing sections between the selected landmarks in units of designated distances;
reflecting longitudinal gradient and lateral gradient information received from the server to each of the divided distance units; and
and setting the sections including the divided distance units reflecting the vertical gradient and the horizontal gradient information as new landmarks. The method of claim 7,
The operation of selecting the landmarks,
A method of driving a vehicle, characterized in that the longitudinal gradient and the lateral gradient are calculated by utilizing a sensor unit of the vehicle while selecting an existing landmark. The method of claim 7,
The operation of dividing the sections into the specified distance unit,
Characterized in that, based on the average value of the vertical gradient value and the horizontal gradient value, the landmarks are divided into the specified distance unit, the vehicle driving method. The method of claim 9,
The operation of reflecting the vertical gradient and the lateral gradient information,
Characterized in that the vertical gradient and the lateral gradient information are reflected for each unit of the specified distance, a vehicle driving method. The method of claim 7,
The operation of setting the sections as the new landmarks,
The method of driving a vehicle, characterized in that by ordering the ratio of the vertical gradient and the horizontal gradient based on a constant distance between the landmarks and setting them as the new landmark having a feature point.

Images