Autonomous driving system and autonomous driving method

Abstract

The present application relates to an autonomous driving system and an autonomous driving method. An autonomous driving system is disclosed herein, including: a first sensor installed in a vehicle and acquiring front image data; a second sensor installed in the vehicle and acquiring forward detection data; a communicator receiving a high-definition map at a current position of the vehicle from an external server; and a controller, wherein the controller calculates a first distance between the vehicle and a right road boundary line and a second distance between the vehicle and a left road boundary line in response to processing the front image data and the forward detection data, calculates a third distance between the vehicle and the right road boundary line and a fourth distance between the vehicle and the left road boundary line in response to processing the high-definition map, determines a map quality index of the high-definition map based on the first distance, the second distance, the third distance, and the fourth distance, and determines an error of the high-definition map based on the map quality index.

Description

Technical Field

[0001] The present disclosure relates to an autonomous driving system and method, and more particularly, to an autonomous driving system and an autonomous driving method capable of determining an error of a high-definition map. Background Art

[0002] A vehicle refers to a device capable of transporting people or goods to a destination when traveling on a road or track. A vehicle can be moved to a variety of locations primarily by using one or more wheels provided with a body. Such vehicles may include three-wheeled or four-wheeled vehicles, vehicles with two wheels (e.g., motorcycles), construction machines, bicycles, and trains running on rails provided on tracks.

[0003] Recently, in order to reduce the burden on drivers and increase convenience, research on vehicles equipped with advanced driver assistance systems (ADAS) that dynamically provide information about vehicle conditions, driver conditions, and surrounding environments is being actively conducted, and research on autonomous driving systems using ADAS is also being actively conducted.

[0004] With the development of ADAS installed in vehicles, high-precision maps have been used as an essential element of autonomous driving systems.

[0005] However, when errors exist in high-precision maps, they may cause serious failures of autonomous driving systems. Summary of the invention

[0006] Therefore, an aspect of the present disclosure is to provide an autonomous driving system and an autonomous driving method capable of determining an error of a high-definition map in consideration of an actual road condition.

[0007] Additional aspects of the disclosure will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the disclosure.

[0008] According to one aspect of the present disclosure, an autonomous driving system includes: a first sensor installed in a vehicle, having a field of view facing forward of the vehicle, and configured to acquire forward image data; a second sensor selected from a group including radar and light detection and ranging (lidar), installed in the vehicle, having a detection field of view facing forward of the vehicle, and configured to acquire forward detection data; a communicator configured to receive a high-definition map at a current position of the vehicle from an external server; and a controller including a processor configured to process the high-definition map, the forward image data, and the forward detection data. Wherein, the controller is configured to calculate a first distance between the vehicle and a right road boundary line and a second distance between the vehicle and a left road boundary line in response to processing the front image data and the forward detection data, calculate a third distance between the vehicle and the right road boundary line and a fourth distance between the vehicle and the left road boundary line in response to processing the high-definition map, and determine a map quality index of the high-definition map based on the first distance, the second distance, the third distance and the fourth distance, and determine an error of the high-definition map based on the map quality index.

[0009] The controller may be configured to determine that an error of the high definition map exists when the map quality index is less than or equal to a predetermined threshold.

[0010] The controller may be configured to reduce the map quality indicator when a root mean square error (RMSE) between the first distance and the third distance is greater than or equal to a predetermined first value.

[0011] The controller may be configured to increase the map quality indicator when a RMSE between the first distance and the third distance is less than or equal to a predetermined second value that is smaller than a predetermined first value.

[0012] The controller may be configured to lower the map quality indicator when the RMSE between the second distance and the fourth distance is greater than or equal to a predetermined first value.

[0013] The controller may be configured to increase the map quality indicator when a RMSE between the second distance and the fourth distance is less than or equal to a predetermined second value that is smaller than a predetermined first value.

[0014] The controller may be configured to reduce the map quality indicator when a color or shape of a lane detected in response to processing the front image data and the forward detection data is different from a color or shape of a lane acquired in response to processing the high definition map.

[0015] The controller may be configured to determine the coordinates of the vehicle on the high definition map such that a sum of a RMSE between the first distance and the third distance and a RMSE between the second distance and the fourth distance is minimized.

[0016] The controller may be configured to transfer control of the vehicle to a driver when an error in the high definition map is determined to exist.

[0017] The controller may be configured to control the communicator to transmit at least one of the front image data and the forward detection data to a server when it is determined that an error of the high definition map exists.

[0018] According to another aspect of the present disclosure, an autonomous driving method includes the following steps: acquiring front image data from a front camera installed in a vehicle and having a field of view facing the front of the vehicle; receiving, by a communicator, a high-definition map at the current position of the vehicle from an external server; a controller calculating a first distance between the vehicle and a right road boundary line and a second distance between the vehicle and a left road boundary line in response to processing the front image data; the controller calculating a third distance between the vehicle and the right road boundary line and a fourth distance between the vehicle and the left road boundary line in response to processing the high-definition map; determining, by the controller, a map quality index of the high-definition map based on the first distance, the second distance, the third distance and the fourth distance; and determining an error of the high-definition map based on the map quality index.

[0019] The step of determining the error may include determining that an error of the high definition map exists when the map quality indicator is less than or equal to a predetermined threshold.

[0020] The step of determining the map quality indicator may include reducing the map quality indicator when a root mean square error (RMSE) between the first distance and the third distance is greater than or equal to a predetermined first value.

[0021] The step of determining the map quality indicator may include increasing the map quality indicator when the RMSE between the first distance and the third distance is less than or equal to a predetermined second value that is smaller than a predetermined first value.

[0022] The step of determining the map quality indicator may include lowering the map quality indicator when the RMSE between the second distance and the fourth distance is greater than or equal to a predetermined first value.

[0023] The step of determining the map quality indicator may include increasing the map quality indicator when the RMSE between the second distance and the fourth distance is less than or equal to a predetermined second value which is smaller than a predetermined first value.

[0024] The method may further include the step of reducing the map quality indicator when a color or shape of a lane detected in response to processing the front image data is different from a color or shape of a lane acquired in response to processing the high definition map.

[0025] The method may further include the step of determining the coordinates of the vehicle on the high definition map such that a sum of the RMSE between the first distance and the third distance and the RMSE between the second distance and the fourth distance is minimized.

[0026] The method may further include the step of transferring control of the vehicle to a driver when it is determined that an error in the high definition map exists.

[0027] The method may further include the step of controlling the communicator to transmit the front image data to the server when it is determined that an error of the high definition map exists. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] These and / or other aspects of the present disclosure will become apparent and more readily understood through the following description of embodiments in conjunction with the accompanying drawings, in which:

[0029] Figure 1 is a diagram showing a configuration of a vehicle according to an exemplary embodiment;

[0030] Figure 2 is a diagram showing a configuration of an autonomous driving system according to an exemplary embodiment;

[0031] Figure 3 is a diagram showing a camera and a radar included in an autonomous driving system according to an exemplary embodiment;

[0032] Figure 4 is a flow chart of an autonomous driving method according to an exemplary embodiment;

[0033] Figure 5 is a diagram showing front image data and a high-definition map acquired by an autonomous driving system according to an exemplary embodiment;

[0034] Figures 6 to 10 It is a view showing a situation in which the presence of an error in a high-precision map can be determined. DETAILED DESCRIPTION

[0035] Now will refer to the embodiment of the present disclosure in detail, the example of the embodiment of the present disclosure is shown in the accompanying drawings, wherein throughout the text, the same reference numerals refer to the same elements. This specification does not describe all elements of the disclosed embodiment, and omits the detailed description of the content known in the art, or omits the redundant description of the substantially same configuration. The terms "component", "module", "member", "block" etc. used in the specification can be implemented with software or hardware. In addition, multiple "components", "modules", "members", "blocks", etc. can be embodied as a component. It is also possible that a "component", "module", "member", "block", etc. includes multiple components.

[0036] Throughout the specification, when an element is referred to as being “connected to” another element, it may be directly or indirectly connected to the other element, and “indirectly connected to” includes being connected to another element via a wireless communication network.

[0037] Furthermore, it should be understood that the terms “comprising” and “having” are intended to indicate the presence of the elements disclosed in the specification, but are not intended to exclude the possibility that one or more other elements may be present or may be added.

[0038] Throughout the specification, when a member is located 'on' another member, this includes not only a case where one member is in contact with another member but also a case where another member exists between the two members.

[0039] The terms first, second, etc. are used to distinguish one component from another component, and the component is not limited by the above terms.

[0040] An expression in the singular includes an expression in the plural unless it has an obviously different meaning in the context.

[0041] The reference numerals used in the operations are for convenience of description and are not intended to describe the order of the operations, and unless otherwise specified, the operations may be performed in a different order.

[0042] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0043] Figure 1 is a diagram showing a configuration of a vehicle according to an exemplary embodiment.

[0044] like Figure 1 As shown, the vehicle 1 includes an engine 10, a transmission 20, a brake device 30, and a steering device 40. The engine 10 includes a cylinder and a piston, and can generate power for driving the vehicle 1. The transmission 20 includes a plurality of gears, and can transmit the power generated by the engine 10 to the wheels of the vehicle. The brake device 30 can decelerate the vehicle 1 or stop the vehicle 1 by friction with the wheels. The steering device 40 can change the driving direction of the vehicle 1.

[0045] The vehicle 1 may include a plurality of electrical components, for example, an engine management system (EMS) 11 , a transmission control unit (TCU) 21 , an electronic brake control module 31 , an electronic power steering (EPS) 41 , a body control module (BCM), and an autonomous driving system 100 .

[0046] The EMS 11 may control the engine 10 in response to the driver's intention to accelerate through the accelerator pedal or a request from the autonomous driving system 100. For example, the EMS 11 may control the torque of the engine 10.

[0047] The TCU 21 may control the transmission 20 in response to a shift command from a driver through a shift lever and / or a travel speed of the vehicle 1. For example, the TCU 21 may adjust a transmission ratio from the engine 10 to wheels.

[0048] The electronic brake control module 31 can control the brake device 30 in response to the driver's intention to brake via the brake pedal and / or wheel slip. For example, the electronic brake control module 31 can temporarily release the brakes of the wheels in response to wheel slip detected when the vehicle 1 is braked (also known as an anti-lock braking system (ABS)). The electronic brake control module 31 can selectively release the brakes of the wheels in response to oversteer and / or understeer detected when the vehicle 1 is turning (also known as electronic stability control (ESC)). In addition, the electronic brake control module 31 can temporarily brake the wheels in response to wheel slip detected when the vehicle 1 is driven (also known as a traction control system (TCS)).

[0049] The EPS 41 can assist the operation of the steering device 40 so that the driver can easily manipulate the steering wheel in response to the driver's intention to steer through the steering wheel. For example, the EPS 41 can assist the operation of the steering device 40 to reduce the steering force during low-speed driving or parking, and increase the steering force during high-speed driving.

[0050] The BCM 51 may control the operation of electrical components that provide convenience to the driver or ensure the driver's safety. For example, the BCM 51 may control headlights, wipers, meters, multi-function switches, and direction indicators.

[0051] The autonomous driving system 100 can assist the driver in operating (e.g., driving, braking, and steering, etc.) the vehicle 1. For example, the autonomous driving system 100 can detect the road environment on which the vehicle 1 is traveling (e.g., other vehicles, pedestrians, cyclists, lanes, road signs, traffic lights, etc.), and control the driving and / or braking and / or steering of the vehicle 1 in response to the detected environment.

[0052] In another exemplary embodiment, the autonomous driving system 100 may receive a high-definition map at the current position of the vehicle from an external server, and control driving and / or braking and / or steering of the vehicle 1 in response to the received high-definition map.

[0053] The autonomous driving system 100 can provide various functions to the driver. For example, the autonomous driving system 100 can provide lane departure warning (LDW), lane keeping assist (LKA), high beam assist (HBA), autonomous emergency braking (AEB), traffic sign recognition (TSR), smart cruise control (SCC), blind spot detection (BSD), and inertial driving guidance.

[0054] The autonomous driving system 100 may include a camera module 101 for acquiring image data around the vehicle 1, a radar module 102 for acquiring object data around the vehicle 1, and a light detection and ranging (lidar) module 103 for scanning around the vehicle 1 and detecting objects. The camera module 101 includes a camera 101a and an electronic control unit (ECU) 101b, and captures the front of the vehicle 1 and identifies other vehicles, pedestrians, cyclists, lanes, road signs, structures, etc. The radar module 102 includes a radar 102a and an ECU 102b, and can acquire the relative position, relative speed, etc. of objects (e.g., other vehicles, pedestrians, cyclists, structures, etc.) around the vehicle 1.

[0055] The laser radar module 103 includes a laser radar 103a and an ECU 103b, and acquires the relative position, relative speed, etc. of moving objects (e.g., other vehicles, pedestrians, cyclists, etc.) around the vehicle 1. In addition, the laser radar module 103 can acquire the shape and position of nearby fixed objects (e.g., buildings, signs, traffic lights, protrusions, etc.).

[0056] Specifically, the laser radar module 103 can obtain the shape and position of fixed objects around the vehicle 1 by acquiring point cloud data for the external view of the vehicle 1.

[0057] In other words, the autonomous driving system 100 processes the image data acquired from the camera module 101, the detection data acquired from the radar module 102, and the point cloud data acquired from the lidar module 103, and can detect the road environment on which the vehicle 1 is traveling, the front object located in front of the vehicle 1, the side object located on the side of the vehicle 1, and the rear object located behind the vehicle 1 in response to the processed image data, the detection data, and the point cloud data.

[0058] The autonomous driving system 100 is not limited to Figure 1The system shown in FIG. 1 is a schematic diagram of a vehicle 100 and a laser radar module 103. The laser radar module 103 may be replaced by the radar module 102 or used together with the radar module 102. The autonomous driving system 100 may include a communicator 150 for receiving a high-definition map (also referred to as a high-precision map) at the current position of the vehicle 1.

[0059] The communicator 150 may be implemented by a communication chip, an antenna, and related components to access a wireless communication network. In other words, the communicator 150 may be implemented as various types of communication modules capable of remotely communicating with an external server. In other words, the communicator 150 may include a wireless communication module for wirelessly sending and receiving data with an external server.

[0060] The above-mentioned electrical components can communicate with each other via a communication network (NT) for the vehicle. For example, the electrical components can send and / or receive data via Ethernet, media-oriented system transport (MOST), FlexRay, controller area network (CAN), and local interconnect network (LIN). For example, the autonomous driving system 100 can send a driving control signal, a brake signal, and a steering signal through the EMS 11, the electronic brake control module 31, and the EPS 41 via the NT.

[0061] Figure 2 is a diagram showing a configuration of an autonomous driving system according to an exemplary embodiment. Figure 3 is a view illustrating a camera and a radar included in an autonomous driving system according to an exemplary embodiment.

[0062] like Figure 2 As shown, vehicle 1 may include a braking system 32 , a steering system 42 , and an autonomous driving system 100 .

[0063] The brake system 32 may include Figure 1 The electronic brake control module 31 described in Figure 1 ) and the brake device 30 (see Figure 1 ), and the steering system 42 may include EPS 41 (see Figure 1 ) and the steering device 40 (see Figure 1 ).

[0064] The braking system 32 and the steering system 42 may control the vehicle 1 so that the vehicle 1 performs autonomous driving based on the control signal of the autonomous driving system 100 .

[0065] The autonomous driving system 100 may include a front camera 110 , a front radar 120 , multiple corner radars 130 , a lidar 135 , and a communicator 150 .

[0066] like Figure 3As shown, the front camera 110 may have a field of view 110a facing the front of the vehicle 1. The front camera 110 may be mounted on the front windshield of the vehicle 1, for example, but may be disposed at any position without limitation as long as it has a field of view facing the front of the vehicle 1.

[0067] The front camera 110 may capture the front of the vehicle 1 and acquire image data of the front of the vehicle 1. The image data of the front of the vehicle 1 may include a position relative to a road boundary line located in front of the vehicle 1.

[0068] The front camera 110 may include a plurality of lenses and an image sensor. The image sensor may include a plurality of photodiodes that convert light into an electrical signal, and the plurality of photodiodes may be arranged in a two-dimensional matrix.

[0069] The front camera 110 may be electrically connected to the controller 140. For example, the front camera 110 may be connected to the controller 140 through one of an NT, a hard wire, or a printed circuit board (PCB).

[0070] The front camera 110 may transmit image data of the front of the vehicle 1 (hereinafter referred to as front image data) to the controller 140 .

[0071] like Figure 3 As shown, the front radar 120 may have a sensing field 120a toward the front of the vehicle 1. The front radar 120 may be mounted on a grille or a bumper of the vehicle 1, for example.

[0072] The front radar 120 may include a transmitting antenna (or a transmitting antenna array) that radiates a transmission wave to the front of the vehicle 1 and a receiving antenna (or a receiving antenna array) that receives a reflected wave reflected by an object. The front radar 120 may obtain forward detection data from the transmission wave emitted by the transmitting antenna and the reflected wave received by the receiving antenna. The forward detection data may include distance information and speed degree about other vehicles or pedestrians or cyclists located in front of the vehicle 1. The front radar 120 may calculate the state distance to the object based on the phase difference (or time difference) between the transmission wave and the reflected wave, and calculate the relative speed of the object based on the frequency difference between the transmission wave and the reflected wave.

[0073] The front radar 120 may be connected to the controller 140 through one of NT, hard wire, or PCB. The front radar 120 may transmit forward detection data (hereinafter referred to as forward detection data) to the controller 140.

[0074] Multiple corner radars 130 include a first corner radar 131 installed on the right front side of the vehicle 1, a second corner radar 132 installed on the left front side of the vehicle 1, a third corner radar 133 installed on the right rear side of the vehicle 1, and a fourth corner radar 134 installed on the left rear side of the vehicle 1.

[0075] like Figure 3 As shown, the first corner radar 131 may have a detection field of view 131a facing the right front side of the vehicle 1. The front radar 120 may be installed, for example, on the right side of the front bumper of the vehicle 1. The second corner radar 132 may have a detection field of view 132a facing the left front side of the vehicle 1, and may be installed on the left side of the front bumper of the vehicle 1. The third corner radar 133 may have a detection field of view 133a facing the right rear side of the vehicle 1, and may be installed, for example, on the right side of the rear bumper of the vehicle 1. The fourth corner radar 134 may have a detection field of view 134a facing the left rear of the vehicle 1, and may be installed, for example, on the left side of the rear bumper of the vehicle 1.

[0076] Each of the first corner radar 131, the second corner radar 132, the third corner radar 133 and the fourth corner radar 134 may include a transmitting antenna and a receiving antenna, respectively. The first corner radar 131, the second corner radar 132, the third corner radar 133 and the fourth corner radar 134 may respectively acquire first corner detection data, second corner detection data, third corner detection data and fourth corner detection data. The first corner detection data may include distance information and speed degree of other vehicles or pedestrians or cyclists or structures (hereinafter referred to as objects) located on the right front side of the vehicle 1. The second corner detection data may include distance information and speed degree of objects located on the left front side of the vehicle 1. The third corner detection data and the fourth corner detection data may include distance information and relative speed of objects located on the right rear side of the vehicle 1 and the left rear side of the vehicle 1, respectively.

[0077] Each of the first corner radar 131, the second corner radar 132, the third corner radar 133, and the fourth corner radar 134 may be connected to the controller 140, for example, through one of NT, hard wire, or PCB. The first corner radar 131, the second corner radar 132, the third corner radar 133, and the fourth corner radar 134 may send first corner detection data, second corner detection data, third corner detection data, and fourth corner detection data to the controller 140, respectively.

[0078] The laser radar 135 can obtain the relative position, relative speed, etc. of moving objects (e.g., other vehicles, pedestrians, cyclists, etc.) near the vehicle 1. In addition, the laser radar 135 can obtain the shape and position of nearby fixed objects (e.g., buildings, signs, traffic lights, protrusions, etc.). The laser radar 135 can be installed in the vehicle 1 to have an external field of view 135a of the vehicle 1, and can obtain point cloud data of the external field of view 135a of the vehicle 1.

[0079] For example, Figure 3 As shown, the laser radar 135 can be set outside the vehicle 1 to have an external field of view 135a of the vehicle 1, and more specifically, can be set on a loop of the vehicle 1.

[0080] The laser radar 135 includes a light transmitter that emits light, a light receiver configured to receive light in a predetermined direction among reflected light when the light emitted from the light transmitter is reflected by an obstacle, and a PCB to which the light transmitter and the light receiver are fixed. In this case, the PCB may be set on a support plate rotated by a rotary drive so that the PCB can be rotated 360 degrees in a clockwise or counterclockwise direction.

[0081] In other words, the support plate can rotate around its axis according to the power transmitted from the rotation driver, and the light emitter and the light receiver are fixed to the PCB and can rotate 360 ​​degrees clockwise or counterclockwise as the PCB rotates. Therefore, the laser radar 135 can detect objects from all directions 135a by emitting and receiving light at 360 degrees.

[0082] The light emitter is a component that emits light (eg, an infrared laser), and according to an exemplary embodiment, the light emitter may be provided in single or plural.

[0083] When light emitted from the light emitter is reflected by an obstacle, the light receiver is disposed to receive light in a predetermined direction among the reflected light. An output signal generated by the light receiver receiving the light may be provided to the object detection process of the controller 140.

[0084] The optical receiver may include a condenser lens for condensing the received light and an optical sensor for detecting the received light. According to an embodiment, the optical receiver may include an amplifier for amplifying the light detected by the optical sensor.

[0085] The laser radar 135 can receive data about many points on the outer surface of the object and can obtain point cloud data, which is a group of data for these points.

[0086] The controller 140 may include a camera module 101 (see Figure 1 ) of ECU 101b (see Figure 1 ) and / or radar module 102 (see Figure 1 ) of the ECU 102b (see Figure 1 ) and / or the ECU 103b of the laser radar module 103 (see Figure 1 ) and / or a separate integrated ECU.

[0087] In the following, for ease of description, it is assumed that the forward sensing data acquired from the front radar 120 may be replaced by or used interchangeably with the point cloud data acquired from the laser radar 135. In other words, the forward sensing data acquired from the front radar 120 may refer to the point cloud data acquired from the laser radar 135 for the front of the vehicle 1.

[0088] The controller 140 may include a processor 141 and a memory 142 .

[0089] The processor 141 processes the front image data of the front camera 110 and the forward detection data of the front radar 120, and generates a brake signal and a steering signal for controlling the brake system 32 and the steering system 42, respectively. In addition, the processor 141 can calculate the distance between the vehicle 1 and the right road boundary line (hereinafter referred to as the first distance) and the distance between the vehicle 1 and the left road boundary line (hereinafter referred to as the second distance) in response to the processing of the front image data of the front camera 110 and the forward detection data of the front radar 120.

[0090] As a method of calculating the first distance and the second distance, conventional image data processing technology and / or radar / lidar data processing technology may be used.

[0091] The processor 141 may process the front image data of the front camera 110 and the forward detection data of the front radar 120 , and detect objects (eg, lanes and structures) in front of the vehicle 1 in response to processing the front image data and the forward detection data.

[0092] In particular, the processor 141 may acquire the position (distance and direction) and relative speed of the object in front of the vehicle 1 based on the forward detection data acquired by the front radar 120. The processor 141 may acquire the position (direction) and type information (e.g., whether the object is another vehicle, a structure, etc.) of the object in front of the vehicle 1 based on the forward image data of the front camera 110. In addition, the processor 141 may match the object detected by the forward image data with the object detected by the forward detection data, and acquire the type information, position, and relative speed of the object in front of the vehicle 1 based on the matching result.

[0093] In addition, as described above, the processor 141 can obtain information related to the road environment on which the vehicle 1 is traveling, information related to the front object, and calculate the distance between the vehicle 1 and the right road boundary line and the distance between the vehicle 1 and the left road boundary line.

[0094] The road boundary line may refer to a boundary line of a structure such as a guardrail, an opposite end of a tunnel, an artificial wall, etc., which the vehicle 1 cannot physically pass through, or may refer to a center line which the vehicle 1 cannot pass through in principle, but is not limited thereto.

[0095] Processor 141 can process the high-definition map received from communicator 150, and in response to processing the high-definition map, calculate the distance between vehicle 1 and the right road boundary line (hereinafter referred to as the third distance) and the distance between vehicle 1 and the left road boundary line (hereinafter referred to as the fourth distance).

[0096] In particular, the processor 141 can receive a high-definition map at the current position of the vehicle 1 based on the current position of the vehicle 1 obtained from the global positioning system (GPS), and determine the position of the vehicle 1 on the high-definition map based on the front image data and the forward sensing data.

[0097] For example, the processor 141 can determine the road that the vehicle 1 is traveling on on the high-definition map based on the position information of the vehicle 1 obtained from the GPS, and determine the driving lane that the vehicle 1 is traveling on based on the front image data and the forward detection data. In other words, the processor 141 can determine the coordinates of the vehicle on the high-definition map.

[0098] For example, the processor 141 may determine the number of left lanes in response to processing the front image data and the forward detection data, and determine the position of the lane in which the vehicle is traveling on the high-definition map based on the determined number of left lanes, thereby specifically determining the coordinates of the vehicle 1 on the high-definition map. However, the method of determining the coordinates of the vehicle 1 on the high-definition map is not limited thereto.

[0099] In other words, processor 141 can determine the coordinates of vehicle 1 on the high-definition map based on the number of right lanes detected based on the front image data and the forward detection data, and determine the coordinates of vehicle 1 on the high-definition map based on the first distance and the second distance calculated based on the front image data and the forward detection data.

[0100] To this end, the processor 141 may include an image processor that processes front image data and high-definition map data of the front camera 110 and / or a digital signal processor that processes detection data of the front radar 120 and / or a microcontroller (MCU) or a domain control unit (DCU) that generates control signals for controlling the braking system 32 and the steering system 42.

[0101] The memory 142 may store programs and / or data for the processor 141 to process image data such as front image data and high-definition map data, programs and / or data for processing detection data, and programs and / or data for generating brake signals and / or steering signals.

[0102] The memory 142 may temporarily store image data received from the front camera 110 and / or detection data received from the radar 120 and a high-definition map received from the communicator 150 , and also temporarily store results of the processor 141 processing the image data and / or detection data.

[0103] In addition, the memory 142 may permanently or semi-permanently store image data received from the front camera 110 and / or detection data received from the radars 120 and 130 and / or a high-definition map received from the communicator 150 corresponding to a signal from the processor 141 .

[0104] To this end, the memory 142 may include not only volatile memories such as static random access memory (SRAM) and dynamic random access memory (DRAM), but also nonvolatile memories such as flash memory, read-only memory (ROM) and erasable programmable read-only memory (EPROM).

[0105] As described above, the radars 120 and 130 may be replaced by or used interchangeably with the laser radar 135 that scans near the vehicle 1 and detects objects.

[0106] As described above, each component included in the vehicle 1 has been described in detail. Figures 4 to 10 The process by which the controller 140 determines the error of the high definition map is described in detail.

[0107] Figure 4 is a flowchart of an autonomous driving method according to an exemplary embodiment, Figure 5 is a diagram showing front image data and a high-definition map acquired by an autonomous driving system according to an exemplary embodiment, and Figures 6 to 10 is a diagram showing a situation where it can be determined that an error of a high-definition map exists.

[0108] Reference Figure 4 According to an exemplary embodiment, the front camera can acquire front image data, the front radar and / or lidar can acquire forward detection data, and the controller 140 can receive the front image data and the forward detection data from the front camera and the front radar (1000).

[0109] According to an exemplary embodiment, the communicator 150 may receive a high-definition map at the current position of the vehicle 1 from the cloud server, and the controller 140 may receive the high-definition map from the communicator 150 (1100). In particular, the communicator 150 may transmit the current position of the vehicle 1 detected from the GPS to the cloud server, and the cloud server may transmit the high-definition map based on the current position of the vehicle 1 to the communicator 150.

[0110] The controller 140 may process the front image data acquired from the front camera 110 and the forward detection data acquired from the front radar 120, and in response to processing the front image data and the forward detection data, calculate a first distance between the vehicle 1 and the right road boundary line and a second distance between the vehicle 1 and the left road boundary line (1200).

[0111] In addition, the controller 140 may calculate a third distance between the vehicle 1 and the right road boundary line and a fourth distance between the vehicle 1 and the left road boundary line in response to processing the high definition map ( 1300 ).

[0112] Reference Figure 5 , the front image data 300 and the high-definition map 400 can be identified. As described above, the controller 140 can determine the coordinates of the vehicle 1 on the high-definition map 400. For example, the controller 140 can determine that the coordinates of the vehicle 1 on the high-definition map 400 are the center of the second lane based on the number of the left lane detected from the front image data 300 being 1.

[0113] Thereafter, the controller 140 may match the right road boundary line RB1 and the left road boundary line LB1 detected by processing the front image data 300 and the forward detection data with the right road boundary line RB2 and the left road boundary line LB2 detected by processing the high definition map 400 .

[0114] The controller 140 may calculate the first distance d1 and the second distance d2 based on the front image data 300 and the forward detection data, and may calculate the third distance d3 and the fourth distance d4 based on the high definition map 400 .

[0115] The controller 140 may determine a map quality index of the high definition map 400 based on the first distance d1 , the second distance d2 , the third distance d3 , and the fourth distance d4 ( 1400 ).

[0116] In particular, the controller 140 may lower the map quality index when a root mean square error (RMSE) between the first distance d1 and the third distance d3 is greater than or equal to a predetermined first value, and may increase the map quality index when the RMSE between the first distance d1 and the third distance d3 is less than or equal to a predetermined second value that is smaller than the predetermined first value. In this case, the predetermined first value may be set to the width of the vehicle 1, but is not limited thereto.

[0117] In addition, when the RMSE between the second distance d2 and the fourth distance d4 is greater than or equal to a predetermined first value, the controller 140 can lower the map quality index, and when the RMSE between the second distance d2 and the fourth distance d4 is less than or equal to a predetermined second value smaller than the predetermined first value, the map quality index can be increased.

[0118] In this case, the controller 140 may correct the coordinates of the vehicle 1 on the high-definition map 400 so that the sum of the RMSE between the first distance d1 and the third distance d3 and the RMSE between the second distance d2 and the fourth distance d4 is minimized.

[0119] In other words, in order to avoid the situation where the RMSE between the first distance d1 and the third distance d3 and the RMSE between the second distance d2 and the fourth distance d4 become larger due to the coordinates incorrectly determined on the high-definition map 400, the controller 140 can determine the coordinates of the vehicle 1 on the high-definition map 400 so that the sum of the RMSE between the first distance d1 and the third distance d3 and the RMSE between the second distance d2 and the fourth distance d4 is minimized.

[0120] When the RMSE between the first distance d1 and the third distance d3 and / or the RMSE between the second distance d2 and the fourth distance d4 is less than or equal to a predetermined second value by correcting the coordinates of the vehicle 1 on the high-definition map 400, the controller 140 can improve the map quality index of the high-definition map, thereby compensating the map quality index.

[0121] However, even when the coordinates of the vehicle 1 on the high-definition map 400 are corrected, the RMSE between the first distance d1 and the third distance d3 and / or the RMSE between the second distance d2 and the fourth distance d3 is greater than or equal to a predetermined first value, the controller 140 may reduce the map quality index of the high-definition map, thereby reducing the map quality index to less than or equal to the threshold.

[0122] Thereafter, the controller 140 may determine an error of the high definition map based on the map quality index ( 1500 ).

[0123] For example, when the map quality index is less than or equal to a predetermined threshold, the controller 140 may determine that an error of the high-definition map exists.

[0124] Reference Figure 6 , it can be recognized that the number of lanes based on the front image data 300 is 4, while the number of lanes based on the high-definition map 400 is 3.

[0125] exist Figure 6 In the illustrated case, the controller 140 may determine the coordinates of the vehicle 1 on the high-definition map 400 as the rightmost lane based on the number of the left lane.

[0126] Thereafter, when it is determined that the RMSE between the first distance d1 and the third distance d3 is greater than or equal to a predetermined first value, the controller 140 may correct the coordinates of the vehicle 1 on the high-definition map 400 so that the sum of the RMSE between the first distance d1 and the third distance d3 and the RMSE between the second distance d2 and the fourth distance d4 is minimized.

[0127] However, when it is determined that the RMSE between the first distance d1 and the third distance d3 or the RMSE between the second distance d2 and the fourth distance d4 is greater than or equal to the predetermined first value, the map quality index of the high definition map 400 may be lowered.

[0128] Therefore, when information about lanes of a road that have been increased due to expansion of the road is not reflected in the high-definition map, the autonomous driving system 100 according to the exemplary embodiment may easily determine an error of the high-definition map.

[0129] Reference Figure 7 , it can be identified that the number of available lanes based on the front image data 300 is 2, while the number of available lanes based on the high-definition map 400 is 3.

[0130] exist Figure 7 In the illustrated case, the controller 140 may determine the coordinates of the vehicle 1 on the high-definition map 400 as the middle lane based on the number of the left lane.

[0131] Thereafter, when it is determined that the RMSE between the first distance d1 and the third distance d3 is greater than or equal to a predetermined first value, the controller 140 may correct the coordinates of the vehicle 1 on the high-definition map 400 so that the sum of the RMSE between the first distance d1 and the third distance d3 and the RMSE between the second distance d2 and the fourth distance d4 is minimized.

[0132] However, when it is determined that the RMSE between the first distance d1 and the third distance d3 or the RMSE between the second distance d2 and the fourth distance d4 is greater than or equal to the predetermined first value, the map quality index of the high definition map 400 may be lowered.

[0133] Therefore, when information about a reduction in available lanes due to road construction is not reflected in the high-definition map, the autonomous driving system 100 according to the exemplary embodiment may easily determine an error of the high-definition map.

[0134] Reference Figure 8 , it can be identified that the type of the highway toll booth based on the front image data 300 is an open type, while the type of the highway toll booth based on the high-definition map 400 is a closed type.

[0135] exist Figure 8 In the illustrated case, the controller 140 may determine the coordinates of the vehicle 1 on the high-definition map 400 as the middle lane based on the number of the left lane.

[0136] Thereafter, when the vehicle 1 passes through a toll station, when the RMSE between the first distance d1 and the third distance d3 and / or the RMSE between the second distance d2 and the fourth distance d4 is greater than or equal to a predetermined first value, the controller 140 may correct the coordinates of the vehicle 1 on the high-definition map 400 so that the sum of the RMSE between the first distance d1 and the third distance d3 and the RMSE between the second distance d2 and the fourth distance d4 is minimized.

[0137] However, when it is determined that the RMSE between the first distance d1 and the third distance d3 or the RMSE between the second distance d2 and the fourth distance d4 is greater than or equal to the predetermined first value, the map quality index of the high definition map 400 may be lowered.

[0138] Therefore, when information that a closed toll booth has been changed to an open toll booth is not reflected in the high-definition map, the autonomous driving system 100 according to the exemplary embodiment may easily determine an error of the high-definition map.

[0139] although Figure 4 Not shown, but when the color or shape of the lane detected in response to processing the front image data and the forward detection data is different from the color or shape of the lane acquired in response to processing the high-definition map, the controller 140 can reduce the map quality index of the high-definition map.

[0140] In other words, even when the RMSE between the first distance d1 and the third distance d3 and / or the RMSE between the second distance d2 and the fourth distance d4 is less than the predetermined first value, the map quality index may be lowered.

[0141] For example, refer to Fig. 9 , it can be recognized that the color and shape of the left lane LL1 based on the front image data 300 are different from the color and shape of the left lane LL2 based on the high-definition map 400 .

[0142] In other words, when the left lane LL1 based on the front image data 300 is a blue solid line and the left lane LL2 based on the high definition map 400 is a white dashed line, the controller 140 may lower the map quality index of the high definition map 400 .

[0143] Therefore, when information that a general lane has been changed to a bus-only lane is not reflected in the high-definition map, the autonomous driving system 100 according to the exemplary embodiment may easily determine an error of the high-definition map.

[0144] As another exemplary embodiment, refer to Fig.10, it can be recognized that the shape of the right lane RB1 based on the front image data 300 and the shape of the left lane RB2 based on the high-definition map 400 are different.

[0145] In other words, when the right lane RB1 based on the front image data 300 is a straight line and the right lane RB2 based on the high definition map 400 is a curve, the controller 140 may lower the map quality index of the high definition map 400 .

[0146] Therefore, when information that a curved lane has been changed into a straight lane is not reflected in the high-definition map, the autonomous driving system 100 according to the exemplary embodiment may easily determine an error of the high-definition map.

[0147] In the case where the vehicle 1 according to the exemplary embodiment is an autonomous driving vehicle, when it is determined that the error of the high-definition map exists, the controller 140 may transfer the control of the vehicle 1 to the driver. In this case, the controller 140 may request the driver to transfer the control through a user interface provided in the vehicle 1.

[0148] Therefore, the autonomous driving system 100 according to the exemplary embodiment may prevent malfunction of the autonomous driving system 100 due to an error of the high definition map.

[0149] In addition, when it is determined that an error of the high-definition map exists, the controller 140 may control the communicator 150 to transmit at least one of the front image data and the forward detection data to the cloud server.

[0150] Therefore, the cloud server can quickly eliminate errors in high-definition maps.

[0151] According to the present disclosure, by comparing the actual driving condition of the vehicle 1 with the high-definition map, it is possible to quickly determine the error of the high-definition map and minimize the risk of the error.

[0152] As is apparent from the above, embodiments of the present disclosure can quickly determine an error of a high-definition map to prevent a malfunction of an autonomous driving system due to the error of the high-definition map.

[0153] At the same time, the disclosed embodiments can be implemented in the form of a recording medium for storing instructions to be carried by a computer. The instructions can be stored in the form of program code, and when the instructions are executed by a processor, a program module can be generated to perform the operations in the embodiments of the present disclosure. The recording medium can correspond to a computer-readable recording medium.

[0154] The computer-readable recording medium includes any type of recording medium on which data can be stored and then read by a computer. For example, the computer-readable recording medium can be a read-only memory (ROM), a random access memory (RAM), a magnetic tape, a magnetic disk, a flash memory, an optical data storage device, etc.

[0155] The exemplary embodiments of the present disclosure have been described so far with reference to the accompanying drawings. It is obvious to those skilled in the art that the present disclosure can be practiced in other forms different from the above exemplary embodiments without changing the technical ideas or basic features of the present disclosure. The above exemplary embodiments are only examples and should not be interpreted in a limited sense.

[0156] CROSS-REFERENCE TO RELATED APPLICATIONS

[0157] This application is based on and claims the benefit of priority from Korean Patent Application No. 10-2020-0109296 filed on August 28, 2020, in the Korean Intellectual Property Office, the disclosure of which is incorporated herein by reference in its entirety.

Claims

1. An autonomous driving system, the autonomous driving system include: a first sensor mounted in the vehicle, having a field of view facing forward of the vehicle, and configured to acquire forward image data; a second sensor selected from the group consisting of radar and light detection and ranging lidar, mounted in the vehicle, having a detection field of view facing forward of the vehicle, and configured to acquire forward detection data; a communicator configured to receive a high-definition map at a current position of the vehicle from an external server; as well as A controller comprising a processor configured to process the high definition map, the front image data and the forward detection data, wherein: The controller is configured to: calculating a first distance between the vehicle and a right road boundary line and a second distance between the vehicle and a left road boundary line in response to processing the front image data and the forward detection data, calculating a third distance between the vehicle and the right road boundary line and a fourth distance between the vehicle and the left road boundary line in response to processing the high definition map, determining a map quality indicator of the high definition map based on at least one of a root mean square error between the first distance and the third distance or a root mean square error between the second distance and the fourth distance, correcting the coordinates of the vehicle on the high-definition map based on at least one of a root mean square error between the first distance and the third distance or a root mean square error between the second distance and the fourth distance to correct the determined map quality indicator, determining whether an error in the high definition map exists based on the corrected map quality indicator, and Control of the vehicle is transferred to a driver in response to determining that an error in the high definition map exists.

2. The autonomous driving system according to claim 1, in, The controller is configured to determine that an error of the high definition map exists when the map quality index is less than or equal to a predetermined threshold.

3. The autonomous driving system according to claim 1, in, The controller is configured to determine the map quality indicator by lowering the map quality indicator when a calculated root mean square error between the first distance and the third distance is greater than or equal to a predetermined first value.

4. The autonomous driving system according to claim 3, in, The controller is configured to determine the map quality indicator by increasing the map quality indicator when a calculated root mean square error between the first distance and the third distance is less than or equal to a predetermined second value that is smaller than the predetermined first value.

5. The autonomous driving system according to claim 1, in, The controller is configured to determine the map quality indicator by lowering the map quality indicator when a calculated root mean square error between the second distance and the fourth distance is greater than or equal to a predetermined first value.

6. The autonomous driving system according to claim 5, in, The controller is configured to determine the map quality indicator by increasing the map quality indicator when a calculated root mean square error between the second distance and the fourth distance is less than or equal to a predetermined second value that is smaller than the predetermined first value.

7. The autonomous driving system according to claim 1, in, The controller is configured to reduce the map quality indicator when a color or shape of a lane detected in response to processing the front image data and the forward detection data is different from a color or shape of a lane acquired in response to processing the high definition map.

8. The autonomous driving system according to claim 1, in, The controller is configured to determine the coordinates of the vehicle on the high definition map such that a sum of a root mean square error between the first distance and the third distance and a root mean square error between the second distance and the fourth distance is minimized.

9. The autonomous driving system according to claim 1, in, The controller is configured to control the communicator to transmit at least one of the front image data and the forward detection data to the server when it is determined that an error of the high-definition map exists.

10. An autonomous driving method, the method The following steps are involved: acquiring forward image data from a forward camera mounted in the vehicle and having a field of view facing forward of the vehicle; receiving, by a communicator, a high-definition map of the current position of the vehicle from an external server; a controller calculating a first distance between the vehicle and a right road boundary line and a second distance between the vehicle and a left road boundary line in response to processing the front image data; the controller calculating a third distance between the vehicle and the right road boundary line and a fourth distance between the vehicle and the left road boundary line in response to processing the high definition map; The controller determines a map quality indicator of the high definition map based on at least one of a root mean square error between the first distance and the third distance or a root mean square error between the second distance and the fourth distance; The controller corrects the coordinates of the vehicle on the high-definition map based on at least one of a root mean square error between the first distance and the third distance or a root mean square error between the second distance and the fourth distance to correct the determined map quality indicator; The controller determines whether an error of the high definition map exists based on the corrected map quality indicator; and The controller transfers control of the vehicle to a driver in response to determining that an error in the high definition map exists.

11. The method according to claim 10, in, The step of determining the error includes determining that an error of the high definition map exists when the map quality indicator is less than or equal to a predetermined threshold.

12. The method according to claim 10, in, The step of determining the map quality indicator comprises lowering the map quality indicator when a calculated root mean square error between the first distance and the third distance is greater than or equal to a predetermined first value.

13. The method according to claim 12, in, The step of determining the map quality indicator comprises increasing the map quality indicator when a calculated root mean square error between the first distance and the third distance is less than or equal to a predetermined second value that is smaller than the predetermined first value.

14. The method according to claim 10, in, The step of determining the map quality indicator comprises lowering the map quality indicator when a calculated root mean square error between the second distance and the fourth distance is greater than or equal to a predetermined first value.

15. The method according to claim 14, in, The step of determining the map quality indicator comprises increasing the map quality indicator when a calculated root mean square error between the second distance and the fourth distance is less than or equal to a predetermined second value that is smaller than the predetermined first value.

16. The method according to claim 10, further comprising: The following steps are involved: The map quality indicator is lowered when a color or shape of a lane detected in response to processing the front image data is different from a color or shape of a lane acquired in response to processing the high definition map.

17. The method according to claim 10, further comprising: The following steps are involved: The coordinates of the vehicle on the high definition map are determined so that a sum of a root mean square error between the first distance and the third distance and a root mean square error between the second distance and the fourth distance is minimized.

18. The method according to claim 10, further comprising: The following steps are involved: When it is determined that an error of the high definition map exists, the communicator is controlled to transmit the front image data to the server.

Images