Device, method and system for generating vehicle data

Abstract

The present invention relates to an apparatus, method and system for generating vehicle data. A vehicle may include: a camera for photographing an area around the vehicle; a positioning sensor for measuring the position of the vehicle; a database for storing an accurate map; and a learning data generating device for generating data for learning based on the photographed area, the position of the vehicle and the accurate map.

Description

[0001] CROSS REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of Korean Patent Application No. 10-2019-0111712, filed on September 9, 2019, which is hereby incorporated by reference in its entirety. Technical Field

[0003] The present invention relates to an apparatus, a method and a system for generating vehicle data. Background Art

[0004] The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.

[0005] Since deep learning studies a large amount of data to obtain results, deep learning can obtain learning data with high reliability.

[0006] When deep learning is applied to identify lanes, roads, or center lines, the accuracy of identifying lanes, roads, or center lines can be improved.

[0007] However, this deep learning should generate data for learning by collecting a large amount of data using the network.

[0008] Furthermore, generating ground truth data is very important for deep learning. In most cases, people create data for learning by directly displaying ground truth on each frame of the image. Therefore, it takes a lot of time and cost to generate a large amount of data for learning whenever learning needs to be started or improved. Summary of the invention

[0009] The present invention has been made to solve the above-mentioned problems arising in the prior art, while completely retaining the advantages achieved by the prior art.

[0010] One aspect of the present invention provides an apparatus, method, and system for generating vehicle data, which utilize information from a camera image taken at a vehicle position calculated through precise positioning and a precise map to generate and provide data for learning about lanes, road center lines, etc., so as to minimize the time and cost of generating a large amount of data for learning and improve learning performance.

[0011] Another aspect of the present invention provides an apparatus, method, and system for generating vehicle data, which transmit information of a learning result using data for learning from a server to each vehicle to improve driving performance of the vehicle.

[0012] The technical problems to be solved by the inventive concept are not limited to the above-mentioned problems, and any other technical problems not mentioned herein will be clearly understood by those skilled in the art in the technical field to which the present invention pertains from the subsequent description.

[0013] According to one aspect of the present invention, a vehicle data generating device includes: a camera for photographing an area around a vehicle; a positioning sensor for measuring the position of the vehicle; a database for storing precise maps; and a learning data generating device for generating data for learning based on the captured camera image, the position of the vehicle and the precise map.

[0014] According to another aspect of the present invention, a method for generating vehicle data includes: collecting camera image information; collecting precise positioning information; acquiring shape information from a precise map based on the precise positioning information; mapping the shape information to a corresponding area of ​​the camera image information; and generating data for learning based on the mapping result of the shape information.

[0015] According to another aspect of the present invention, a system for generating vehicle data includes: a server, which is used to receive data for learning and perform deep learning machine learning based on the data for learning to generate learning result information; and a first vehicle, which includes a vehicle data generating device, which is used to generate data for learning and receive learning result information.

[0016] Further areas of applicability will become apparent from the description provided herein.It should be understood that the description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the present invention will become more apparent through the following detailed description in conjunction with the accompanying drawings, in which:

[0018] Figure 1 is a schematic diagram showing the configuration of a system to which a vehicle data generating apparatus according to an embodiment of the present invention is applied;

[0019] Figure 2 is a schematic diagram showing a configuration of a vehicle data generating device according to an embodiment of the present invention;

[0020] Figure 3 is a schematic diagram showing an embodiment referred to for describing an operation of generating learning data of a vehicle data generating apparatus according to an embodiment of the present invention;

[0021] Figure 4 is a schematic diagram showing the configuration of a system according to one embodiment of the present invention;

[0022] Figure 5 is a schematic diagram showing an operation flow of a method for generating vehicle data according to an embodiment of the present invention; and

[0023] Figure 6 A computing system is shown that implements a method according to one embodiment of the invention.

[0024] The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way. DETAILED DESCRIPTION

[0025] The following description is merely exemplary in nature and is not intended to limit the invention, application, or uses. It should be understood that throughout the drawings, corresponding reference numerals indicate the same or corresponding parts and features.

[0026] Hereinafter, some embodiments of the present invention will be described in detail with reference to the exemplary drawings. When adding reference numerals to the components of each drawing, it should be noted that even if the same or equivalent components are shown on other drawings, they are represented by the same reference numerals. In addition, when describing the embodiments of the present invention, when it is determined that the relevant known configuration or function hinders the understanding of the embodiments of the present invention, the detailed description thereof will be omitted.

[0027] When describing components according to embodiments of the present invention, terms such as first, second, A, B, (a), (b) and the like may be used. These terms are intended only to distinguish one component from other components, and these terms do not limit the nature, order or sequence of the components. Unless otherwise defined, all terms used herein (including technical terms and scientific and technological terms) have the same meaning as those generally understood by those of ordinary skill in the art to which the present invention belongs. It will be further understood that, for example, terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning of the term in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense, unless clearly defined as such herein.

[0028] Figure 1 : is a schematic diagram showing the configuration of a system to which a vehicle data generating apparatus according to an embodiment of the present invention is applied.

[0029] Reference Figure 1 , the system may include a vehicle data generating device 100( Figure 2 as shown) and server 20.

[0030] In this regard, the vehicle data generating device 100 according to the present invention may be implemented in the vehicle 10. In this regard, the vehicle data generating device 100 may be formed integrally with an internal control unit of the vehicle 10, or may be implemented as a separate device and connected to the control unit of the vehicle through a separate connection means.

[0031] The vehicle data generating device 100 generates data for learning, which is used for deep learning machine learning on lanes and road center lines, and provides the data for learning to the server 20 .

[0032] In this regard, the vehicle data generating device 100 can utilize information from a camera image captured at the position of the vehicle 10 (which is calculated through precise positioning and precise mapping) to generate data for learning of lanes, road center lines, etc., and provide the generated data for learning to the server 20.

[0033] The server 20 stores the data for learning provided by the vehicle data generating device 100. In this regard, the server 20 may receive and store each data for learning from each vehicle data generating device 100 disposed in each of the plurality of vehicles 10.

[0034] The server 20 can perform deep learning machine learning using the stored data for learning, and can provide the learning result information to each vehicle 10 where each vehicle data generating device 100 is arranged. Deep learning machine learning is a technology for clustering or classifying objects or data, which can extract specific patterns of data for learning to cluster or classify objects or input data based on the extracted specific patterns. In this regard, the larger the amount of data for learning input to the deep learning algorithm, the more accurate the classification or clustering. Deep learning machine learning can include deep learning techniques such as deep neural networks, convolutional deep neural networks, and deep belief networks.

[0035] Therefore, the vehicle 10 can more accurately identify the lane and / or the road center line based on the learning result information provided by the server 20.

[0036] Therefore, reference will be made to Figure 2 The detailed configuration and operation of the vehicle data generating device 100 are described in more detail in the embodiments of the present invention.

[0037] Figure 2 is a schematic diagram showing the configuration of a vehicle data generating device according to an embodiment of the present invention.

[0038] Reference Figure 2 , the vehicle data generating apparatus 100 may include a controller 110, a camera 120, a precise positioning sensor 130, a precise map DB (database) 140, a communicator 150, a data collection device 160, and a learning data generating device 170. In this regard, the controller 110, the data collection device 160, and the learning data generating device 170 of the vehicle data generating apparatus 100 according to an embodiment of the present invention may be implemented as at least one processor.

[0039] The controller 110 may process signals transmitted and received between components of the vehicle data generating apparatus 100 .

[0040] The camera 120 is a device for capturing images of the area surrounding the vehicle.

[0041] As an example, the camera 120 may be a front camera arranged to face the area in front of the vehicle. When requested, the front camera may capture images of the area in front of the vehicle in real time or at predetermined intervals and send the captured images to the controller 110 and / or the data collection device 160.

[0042] In this regard, the controller 110 may store an image transmitted from the camera 120 (hereinafter referred to as a “camera image”) in a precise map DB (database) 140 and provide the camera image to the data collection device 160 .

[0043] The precise positioning sensor 130 refers to a device that measures the position of the vehicle using precise positioning technology.

[0044] Positioning technology refers to technology that uses information obtained through navigation sensors such as GPS to determine the location. Precise positioning technology refers to technology that detects the vehicle's location by using positioning technology to identify the vehicle's location, calls up high-precision map information around the vehicle, and compares the high-precision map information with sensor information collected in real time while driving to correct the vehicle's location.

[0045] In this regard, the precise positioning sensor 130 may be implemented in the form of a plurality of sensors coupled to each other.

[0046] The precise positioning sensor 130 may be synchronized with the camera 120 to detect the vehicle position when capturing camera images. The precise positioning sensor 130 may send the detected vehicle position information to the controller 110 and / or the data collection device 160.

[0047] In this regard, the controller 110 may store the vehicle position information transmitted from the precise positioning sensor 130 in a precise map DB (database) 140 and provide the vehicle position information to the data collection device 160 .

[0048] Preliminary accurate maps are stored in an accurate map DB (database) 140 .

[0049] In this regard, the precise map DB (database) 140 may include a storage medium such as a RAM (random access memory), an SRAM (static random access memory), a ROM (read only memory), a PROM (programmable read only memory), and an EEPROM (electrically erasable programmable read only memory). The precise map DB (database) 140 may include a storage medium, and in addition to the pre-established precise map, information required to generate data for learning may be stored in the storage medium. For example, the precise map DB (database) 140 may further store a camera image captured by the camera 120 and vehicle position information detected by the precise positioning sensor 130.

[0050] The communicator 150 may include a communication module for performing vehicle network communication with electronic devices and / or control units provided in the vehicle. As an example, the communication module may support vehicle network communication between the controller 110, the camera 120, the precise positioning sensor 130, the precise map DB (database) 140, the data collection device 160, and the learning data generation device 170, and may receive a signal output from each device and transmit the received signal.

[0051] In this regard, as the vehicle network communication technology, CAN (Controller Area Network) communication, LIN (Local Interconnect Network) communication, Flex-Ray communication, Ethernet, etc. can be included.

[0052] In addition, the communicator 150 may further include a communication module (such as a communication module) for wirelessly communicating with the server 20. Figure 1 As shown). As an example, the communication module can send / receive signals with the server 20 through wireless Internet communication or short-range communication.

[0053] In this regard, as the wireless Internet communication technology, WLAN (Wireless LAN), WiBro (Wireless Broadband), Wi-Fi, WiMAX (Worldwide Interoperability for Microwave Access), and the like may be included.

[0054] In addition, as the short-range communication technology, Bluetooth, ZigBee, UWB (Ultra Wideband), RFID (Radio Frequency Identification), IrDA (Infrared Data Association), etc. may be included.

[0055] When the vehicle position information is received from the precise positioning sensor 130 or the controller 110, the data collection device 160 may call the precise map stored in the precise map DB (database) 140, and collect precise map information corresponding to the vehicle position information from the called precise map. For example, the data collection device 160 may collect shape information about lanes and / or road center lines on the precise map corresponding to the vehicle position in the precise map stored in the precise map DB (database) 140. In this regard, the precise map information may include shape information about lanes and / or road center lines on the precise map.

[0056] The learning data generating device 170 generates data for learning using the camera image and precise map information collected by the data collecting device 160. In this regard, the learning data generating device 170 maps the precise map information to the camera image. For example, the learning data generating device 170 maps the shape information about the lane and / or the center line of the road on the precise map to the corresponding area of ​​the camera image. In this regard, the learning data generating device 170 may directly receive the collected camera image and precise map information from the data collecting device 160 when generating the learning data. In addition, when the learning data generating device 170 generates the learning data, the controller 110 may receive the collected camera image and precise map information from the data collecting device 160, and send the collected camera image and precise map information to the learning data generating device 170.

[0057] When mapping precise map information to a camera image, the learning data generating device 170 may perform a separate preprocessing process to a data format suitable for machine learning and then perform the mapping.

[0058] For example, when a front 3D image is completely generated as data for learning, the learning data generating device 170 may perform coordinate transformation on information on a precise map based on the 3D, and then map the corresponding information to the camera image.

[0059] Furthermore, when generating data for learning in a bird's-eye view, the learning data generating device 170 may transform the camera image based on a corresponding coordinate system and combine the transformed camera image with information on a precise map.

[0060] In addition, before mapping the precise map information to the camera image, the learning data generating device 170 may perform various preprocessing processes required for deep learning, such as contrast conversion and / or rotation in the left-right direction.

[0061] The learning data generating device 170 may generate a camera image mapped with accurate map information as data for learning, and may store the generated data for learning in an accurate map DB (database) 140 .

[0062] When the learning data generating device 170 generates the data for learning, the controller 110 transmits the generated data for learning to the server connected through the communicator 150 .

[0063] Furthermore, when receiving a learning result of the data for learning from the server 20 , the controller 110 may provide the received learning result data to a travel system for controlling travel of the vehicle, a lane recognition system for recognizing a lane, or the like.

[0064] The vehicle data generating device 100 according to the present embodiment operating as described above may be implemented in the form of an independent hardware device including a memory and a processor for processing each operation, or may operate in the form of being included in another hardware device such as a microprocessor or a general-purpose computer system.

[0065] Figure 3 1 is a schematic diagram showing an embodiment referred to for describing an operation of generating learning data of a vehicle data generating apparatus according to an embodiment of the present invention.

[0066] Reference Figure 3 The vehicle data generating device 100 may generate data 320 for learning using a camera image 311 acquired from a camera 120 , precise positioning information 313 acquired from a precise positioning sensor 130 , and precise map information 315 acquired from a precise map DB (database) 140 .

[0067] In this regard, the vehicle data generating device 100 may acquire precise map information 315 on a precise map based on the precise positioning information 313. As precise map information, lanes and / or road center lines of a corresponding area on the precise map may be included. In addition, the precise map information may include information that enables feature recognition in the corresponding area.

[0068] When the vehicle position information is received from the precise positioning sensor 130 or the controller 110, the data collection device 160 may call the precise map stored in the precise map DB (database) 140, and collect precise map information corresponding to the vehicle position information from the called precise map. For example, the data collection device 160 may collect shape information about lanes and / or road center lines on the precise map corresponding to the vehicle position in the precise map.

[0069] The learning data generating device 170 generates data for learning using the camera image and the precise map information collected by the data collecting device 160. In this regard, the learning data generating device 170 maps the precise map information to the camera image. For example, the learning data generating device 170 maps shape information about lanes and / or road center lines on the precise map to corresponding areas of the camera image.

[0070] When mapping precise map information to a camera image, the learning data generating device 170 may perform a separate preprocessing process to a data format suitable for machine learning and then perform the mapping.

[0071] For example, when a front 3D image is completely generated as data for learning, the learning data generating device 170 may perform coordinate transformation on information on a precise map based on the 3D, and then map the corresponding information to the camera image.

[0072] Furthermore, when generating data for learning in a bird's-eye view, the learning data generating device 170 may transform the camera image based on a corresponding coordinate system and combine the transformed camera image with information on a precise map.

[0073] In addition, before mapping the precise map information to the camera image, the learning data generating device 170 may perform various preprocessing processes required for deep learning, such as contrast conversion and / or rotation in the left-right direction.

[0074] The learning data generating device 170 may generate a camera image mapped with accurate map information as data for learning, and may store the generated data for learning in an accurate map DB (database) 140 .

[0075] Figure 4 is a schematic diagram showing the configuration of a system according to another embodiment of the present invention.

[0076] Reference Figure 4 , the system may include a first vehicle 10 , a server 20 , and a second vehicle 30 .

[0077] like Figure 1 and Figure 2 As shown, the first vehicle 10 may include a vehicle data generating device 100. The first vehicle 10 may include the vehicle data generating device 100 therein. The vehicle data generating device 100 arranged in the first vehicle 10 may be formed integrally with an internal control unit of the first vehicle 10, or may be implemented as a separate device and connected to the control unit of the first vehicle 10 through a separate connection device.

[0078] The vehicle data generating device 100 disposed in the first vehicle 10 can generate data for learning, which is used for deep learning machine learning of lanes and road center lines, and provide the data for learning to the server 20. In this regard, the vehicle data generating device 100 can generate data for learning of lanes, road center lines, etc. using information of a camera image taken at the position of the first vehicle 10 (which is calculated by precise positioning and precise maps), and provide the generated data for learning to the server 20.

[0079] Reference Figure 2 The vehicle data generating device 100 arranged in the first vehicle 10 may include a controller 110, a camera 120, a precise positioning sensor 130, a precise map DB (database) 140, a communicator 150, a data collection device 160 and a learning data generating device 170.

[0080] The controller 110 may process signals transmitted and received between the components 120 to 170 of the vehicle data generating apparatus 100 .

[0081] The camera 120 is a device for capturing images of the area around the vehicle. As an example, the camera 120 may be a front camera arranged to face the area in front of the vehicle. When requested, the front camera may capture images of the area in front of the vehicle in real time or at predetermined intervals, and transmit the captured images to the controller 110 and / or the data collection device 160. In this regard, the controller 110 may store the images captured by the camera 120 in the precise map DB (database) 140, and provide the images to the data collection device 160.

[0082] The precise positioning sensor 130 refers to a device that uses precise positioning technology to measure the position of the vehicle. For example, the precise positioning sensor 130 can be implemented in the form of multiple sensors connected to each other. The precise positioning sensor 130 can be synchronized with the camera 120 to detect the vehicle position when the camera image is captured. The precise positioning sensor 130 can send the detected vehicle position information to the controller 110 and / or the data collection device 160. In this regard, the controller 110 can store the vehicle position information sent from the precise positioning sensor 130 in the precise map DB (database) 140 and provide the vehicle position information to the data collection device 160.

[0083] The pre-established precise map may be stored in the precise map DB (database) 140. In this regard, the precise map DB (database) 140 may include a storage medium such as a RAM (random access memory), an SRAM (static random access memory), a ROM (read only memory), a PROM (programmable read only memory), and an EEPROM (electrically erasable programmable read only memory). The precise map DB (database) 140 may include a storage medium, and in addition to the pre-established precise map, information required to generate data for learning may be stored in the storage medium. For example, the precise map DB (database) 140 may further store a camera image captured by the camera 120 and vehicle position information detected by the precise positioning sensor 130, and the like.

[0084] The communicator 150 may include a communication module for performing vehicle network communication with electronic devices and / or control units provided in the first vehicle 10. For example, the communicator 150 may support vehicle network communication between the controller 110, the camera 120, the precise positioning sensor 130, the precise map DB (database) 140, the data collection device 160, and the learning data generation device 170. In addition, the communicator 150 may further include a communication module for wirelessly communicating with the server 20.

[0085] When the vehicle position information is received from the precise positioning sensor 130 or the controller 110, the data collection device 160 may call the precise map stored in the precise map DB (database) 140, and collect precise map information corresponding to the vehicle position information from the called precise map. For example, the data collection device 160 may collect shape information about lanes and / or road center lines on the precise map corresponding to the vehicle position.

[0086] The learning data generating device 170 can generate data for learning using the camera image and precise map information collected by the data collection device 160. For example, the learning data generating device 170 can map the shape information about the lane and / or the center line of the road on the precise map (i.e., the precise map information) to the corresponding area of ​​the camera image. In addition, the learning data generating device 170 can perform a separate preprocessing process on the precise map information to a data format suitable for machine learning, and then map the preprocessed precise map information with the camera image. In this regard, when the data for learning is generated by the learning data generating device 170, the controller 110 can store the generated data for learning in the precise map DB (database) 140, and send the data for learning to the server 20 through the communicator 150.

[0087] The server 20 may store data for learning provided by the vehicle data generating device 100 disposed in the first vehicle 10. In this regard, the server 20 may receive each data for learning from each vehicle data generating device 100 disposed in each of the plurality of vehicles 10.

[0088] The server 20 may perform deep learning machine learning using the stored data for learning, and may provide learning result information to the first vehicle 10 and the second vehicle 30 .

[0089] The second vehicle 30 may receive the result information of the deep learning machine learning performed by the server 20. The second vehicle 30 may process the image captured by the camera arranged in the second vehicle 30 based on the result information of the deep learning machine learning performed by the server 20. Therefore, the second vehicle 30 processes the image captured by the camera using the deep learning machine learning result information of the deep learning machine learning performed based on the data for learning provided by the first vehicle 10, so that the performance of the autonomous driving may be further improved.

[0090] Unlike the first vehicle 10 in which the vehicle data generating device 100 is disposed, the second vehicle 30 may not include a precise positioning sensor and a precise map DB. That is, unlike the first vehicle 10, the second vehicle 30 may not include a precise positioning sensor and a precise map DB for generating data for learning, and may include only components required for autonomous driving. For example, referring to Figure 2 , the second vehicle 30 may include therein a controller 110 , a camera 120 , a communicator 150 , a data collection device 160 , and a learning data generating device 170 .

[0091] Figure 4 The system according to another embodiment of the present invention shown in discloses: a first vehicle, a server, and a second vehicle, wherein the first vehicle can generate data for learning using a precise positioning sensor and a precise map DB; the server receives the data for learning from the first vehicle to perform deep learning machine learning and stores the learning result information therein; the first vehicle and the second vehicle receive the learning results from the server and drive autonomously. In this regard, the first vehicle and the second vehicle may be autonomous driving vehicles, and the first vehicle may be a vehicle capable of generating data for learning, and the data for learning may be used for deep learning machine learning of the server.

[0092] Figure 5 1 is a schematic diagram showing an operation flow of a method for generating vehicle data according to an embodiment of the present invention. Figure 2 To describe Figure 5 A method for generating vehicle data is shown.

[0093] The step of collecting camera image information (S110) may include collecting images of the area around the vehicle captured by the camera 120 disposed in the vehicle. In this regard, the controller 110 may store the images captured by the camera 120 in the precise map DB (database) 140 and provide the images to the data collection device 160.

[0094] The step of collecting precise positioning information (S120) may include measuring the position of the vehicle by the precise positioning sensor 130. For example, the step of collecting precise positioning information (S120) may be synchronized with the camera 120 to detect the vehicle position when capturing the camera image. In this regard, the controller 110 may store the vehicle position information transmitted from the precise positioning sensor 130 in the precise map DB (database) 140 and provide the vehicle position information to the data collection device 160.

[0095] The step of acquiring shape information based on the precise map ( S130 ) may include calling the precise map from the precise map DB (database) 140 storing the precise map. In this regard, the controller 110 may provide the precise map stored in the precise map DB (database) 140 to the data collection device 160 .

[0096] The step of comparing the collected information with each other (S140) may include comparing the information collected in the data collection device 160 with each other. For example, the step of comparing the collected information with each other (S140) may include comparing the information of the image photographed by the camera 120 with the information of the precise map called from the precise map DB (database) 140. In this regard, the controller 110 or the data collection device 160 may compare the information of the image photographed by the camera 120 with the precise map information.

[0097] When the collected camera image information and the precise map information have no difference from each other (No), that is, are identical to each other when the collected information is compared with each other (S140 and S150), the information mapping step (S190) may be performed.

[0098] The information mapping step (S190) may include mapping the precise map information collected by the data collection device 160 to the corresponding area of ​​the camera image. In this regard, the learning data generating device 170 may map the precise map information to the corresponding area of ​​the camera image based on the precise map information and the camera image information collected by the data collection device 160.

[0099] The step of generating data for learning (S200) may include mapping the precise map information collected by the data collection device 160 to the corresponding area of ​​the camera image to generate data for learning. In this regard, the learning data generating device 170 may generate data for learning based on the result of mutual mapping of the precise map information and the camera image information performed in the information mapping step (S190).

[0100] The transmitting data for learning step (S210) may include transmitting the generated data for learning to the server 20. In this regard, the communicator 150 may transmit the data for learning generated by the learning data generating device 170 to the server 20 through wired / wireless communication.

[0101] When the collected camera image information and the precise map information are different from each other ("Yes") when the received information is compared with each other in steps S140 and S150, an updating image processing step (S170) or an updating precise map DB step (S180) may be performed.

[0102] When the camera image information and the precise map information are different from each other ("Yes"), the updating image processing step (S170) may be repeatedly performed a preset number of times (n times, n is a natural number).

[0103] Furthermore, when the updating image processing step (S170) is repeated equal to or more than a preset number of times, the updating precise map step (S180) may be performed.

[0104] Therefore, the method for generating vehicle data according to an embodiment of the present invention may further include: selecting an update step (S160), when the camera image information and the precise map information are different from each other when the received information is compared with each other in steps S140 and S150, selecting the update step (S160) to select one of the update image processing step (S170) and the update precise map step (S180).

[0105] Selecting the update step (S160) may include: first, when the camera image information and the precise map information are different from each other, repeatedly executing the update image processing step (S170) until the camera image information and the precise map information are the same as each other, and when the repeatedly executed update image processing step (S170) is repeated equal to or more than a preset number of times, selecting the update precise map DB step (S180).

[0106] The updating image processing step (S170) may include updating the camera image information as a comparison target of the precise map information. The updating image processing step (S170) may include returning to the collecting camera image information step (S110).

[0107] The updating step (S180) of the precise map may include updating the precise map that is the comparison target of the camera image information. The updating step (S180) of the precise map may include receiving and storing information about the precise map stored in the precise map DB (database) 140 from the server 20 or an external device. In addition, the updating step (S180) of the precise map may include returning to the collecting camera image information step (S110) after updating the precise map stored in the precise map DB (database) 140. In one example, the updating step (S180) of the precise map may include performing the information mapping step (S190) after updating the precise map stored in the precise map DB.

[0108] The device, method and system for generating vehicle data according to an embodiment of the present invention can automatically generate and provide data for learning about the recognition of lanes and road centerlines based on information of camera images taken at the vehicle position (which is calculated by precise positioning and precise maps), thereby minimizing the time and cost of generating a large amount of data for learning and improving learning performance. The device, method and system for generating vehicle data according to an embodiment of the present invention can send information learned using the data for learning (deep learning machine learning results) from a server to each vehicle, thereby improving the driving performance of the vehicle.

[0109] Figure 6 A computing system is shown that implements a method according to an embodiment of the invention.

[0110] Reference Figure 6 , the computing system 1000 may include at least one processor 1100 , a memory 1300 , a user interface input device 1400 , a user interface output device 1500 , a storage device 1600 , and a network interface 1700 connected via a bus 1200 .

[0111] The processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 1300 and / or the storage device 1600. The memory 1300 and the storage device 1600 may include various volatile or non-volatile storage media. For example, the memory 1300 may include a ROM (read only memory) and a RAM (random access memory).

[0112] Therefore, the operation of the method or algorithm described in conjunction with the embodiments disclosed herein can be implemented directly in hardware, a software module executed by the processor 1100, or a combination of the two. The software module can be located on a storage medium (i.e., memory 1300 and / or storage device 1600), such as RAM, flash memory, ROM, EPROM, EEPROM, register, hard disk, removable disk, and CD-ROM. An exemplary storage medium is connected to the processor 1100, and the processor 1100 can read information from the storage medium and can write information in the storage medium. In another method, the storage medium can be integrated with the processor 1100. The processor and the storage medium can be located in an application specific integrated circuit (ASIC). The ASIC can be located in a user terminal. In another method, the processor and the storage medium can exist as independent components in a user terminal.

[0113] The above description merely illustrates the technical concept of the present invention, and those skilled in the art may make various modifications and changes without departing from the basic characteristics of the present invention.

[0114] Therefore, the exemplary embodiments of the present invention are provided to explain the spirit and scope of the present invention, rather than to limit them, so that the spirit and scope of the present invention are not limited by the embodiments. The scope of the present invention should be interpreted based on the appended claims, and all technical concepts within the scope of the claims should be included within the scope of the present invention.

[0115] According to the present invention, information from a camera image captured at a vehicle position calculated through precise positioning and precise mapping can be used to generate and provide data for learning regarding lanes, road center lines, etc., thereby minimizing the time and cost of generating a large amount of data for learning and improving learning performance.

[0116] Furthermore, according to the present invention, information on a learning result using data for learning can be transmitted from the server to each vehicle, thereby improving the drivability of the vehicle.

[0117] Although the present invention has been described above with reference to exemplary embodiments and the accompanying drawings, the present invention is not limited thereto but may be variously modified and changed by those skilled in the art without departing from the spirit and scope of the invention as claimed in the appended claims.

Claims

1. A vehicle comprising a vehicle data generating device, the vehicle include: a camera configured to capture an area surrounding the vehicle; a positioning sensor configured to measure a position of the vehicle; a database configured to store the precise map; as well as a learning data generating device configured to generate data for learning based on the photographed area and the accurate map information, The precise map information is information corresponding to the vehicle position in the precise map, and includes shape information of lanes and road centerlines on the precise map. The learning data generating means is configured to map the shape information to a corresponding area of ​​the photographed area, and to generate data for learning based on the mapped shape information.

2. The vehicle comprising the vehicle data generating device according to claim 1, in, The positioning sensor is synchronized with the camera to measure the position of the vehicle at the time of the camera's capture.

3. The vehicle comprising the vehicle data generating device according to claim 2, in, The vehicle further comprises: A data collection device is configured to collect the precise map information.

4. A system for generating vehicle data, the system include: The server is configured as: receiving data for learning; Performing machine learning based on the data used for learning to generate learning result information; as well as a first vehicle including a vehicle data generating device configured to generate data for learning and receive learning result information, wherein data for learning is generated based on an area around the first vehicle photographed by a camera included in the vehicle data generating device and precise map information, The precise map information is information corresponding to the position of the first vehicle in the precise map, and includes shape information of lanes and road center lines on the precise map, The vehicle data generating apparatus further includes a learning data generating device configured to map the shape information to a corresponding area of ​​the photographed area and generate data for learning based on the mapped shape information.

5. The system for generating vehicle data according to claim 4, in, The vehicle data generating device further comprises: a positioning sensor configured to measure a position of the first vehicle; a database configured to store the precise map; and A data collection device is configured to collect the precise map information.

6. The system for generating vehicle data according to claim 5, in, The system further comprises: The second vehicle has the following configuration: receiving learning result information from the server; Autonomous driving.

7. A method for generating vehicle data, the method include: The camera collects camera image information; The precise positioning information is collected by the positioning sensor; The learning data generating device acquires shape information of lanes and road center lines on the precise map from the precise map based on the precise positioning information; Mapping the shape information to a corresponding area of ​​the camera image information by a learning data generating device; The learning data generating device generates data for learning based on the mapped shape information, The camera image information is the area around the vehicle photographed by the camera, and the precise positioning information is the position of the vehicle measured by the positioning sensor.

8. The method according to claim 7, in, The method further comprises: Send the data used for learning to the server.

9. The method according to claim 7, in, The method further comprises: In response to acquiring the shape information, the camera image information is compared with the precise map information, The precise map information is information corresponding to the vehicle position in the precise map and includes the shape information.

10. The method according to claim 9, in, The method further comprises: In response to comparing the camera image information with the precise map information, When the camera image information and the precise map information match, the shape information is mapped; When the camera image information and the precise map information do not match, the image processing is updated.

11. The method according to claim 10, in, Updated image processing includes: The camera image information is collected repeatedly until the camera image information and the accurate map information match.

12. The method according to claim 11, in, The method further comprises: When the image processing is repeated a predetermined number of times, the precise map is updated.

Images