Position estimation device and driving support system

Abstract

This invention provides a position estimation device and driving assistance system that enable high-precision position detection based on images from cameras that capture surrounding objects, even for vehicles that are not equipped with high-precision positioning devices. [Solution] A location estimation device mounted on a vehicle requesting location information comprises a server communicator that receives a selected model transmitted from a server having a server communicator that receives reference image data, which is an image of an object around a vehicle to which location information obtained by a positioning device mounted on the vehicle has been assigned; a model generation unit that generates multiple models for estimating the position of a photographed object from the position and angle of a stationary object extracted from the reference image data; and a model selection unit that selects a model suitable for the destination and transmits it via the server communicator; a camera; and a position estimation unit, wherein the position estimation unit uses the received selected model to estimate the position of the vehicle requesting location information based on the position and angle of a stationary object extracted from the image data.

Description

Technical Field

[0001] This application relates to a position estimation device and a driving support system.

Background Art

[0002] In recent years, the introduction of autonomous driving technology into transportation vehicles has been desired. In the efforts towards autonomous driving technology, high-precision position detection of vehicles has been desired. GPS (Global Positioning System) operated by the United States has become inexpensive due to mass production of its receivers and is widely spread. However, positioning by GPS has an error of about 1.5 m. For this reason, a technique has been disclosed for detecting surrounding objects from an image obtained by a camera mounted on a vehicle and determining the correctness of the current position by referring to map information (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the technique disclosed in Patent Document 1, it is possible to determine that the detection position by a positioning device that has received a signal from a GPS satellite is incorrect when it is clear that the detection position is incorrect. When it is recognized that the vehicle is traveling in a straight lane from the detection position by the positioning device and a road wall is detected ahead, it can be determined that the vehicle is traveling in a branching lane. In such a case, an example of determining that the detection position by the positioning device is incorrect is shown.

[0005] However, Patent Document 1 does not disclose a technology for detecting surrounding objects from images obtained by a camera mounted on a vehicle and precisely recognizing the current location by referring to map information. Errors occur even when detecting surrounding objects from images obtained by a camera. Errors also occur when calculating the vehicle's position from the information of the detected surrounding objects. Furthermore, there are cases where the detection of surrounding objects fails. The technology in Reference Document 1 can only determine when the error in the position information from the GPS positioning device is clear.

[0006] As a more precise positioning method, it is possible to detect a position with an accuracy of about 0.5m by receiving and processing signals from GPS satellites and augmentation signals from quasi-zenith satellites (for example, Japan's QZSS (Quasi-Zenith Satellite System)). With such high-precision positioning devices, it is possible to determine the lane position on the road being traveled. However, high-precision positioning devices that receive signals from both GPS satellites and quasi-zenith satellites in combination are not widely available, are expensive, and difficult to introduce into all vehicles.

[0007] The position estimation device and driver assistance system described herein were developed to solve the problems mentioned above. The objective is to obtain a position estimation device and driver assistance system that enables high-precision position detection based on images from a camera that captures surrounding objects, even for vehicles that are not equipped with a high-precision positioning device. [Means for solving the problem]

[0008] The position estimation device relating to this disclosure is A server communication device that receives reference image data, which is an image of an object around a vehicle, to which location information determined by a positioning device mounted on the vehicle has been added. A data storage unit that stores reference image data to which location information has been added. A model generation unit generates multiple models that estimate the position of a photograph based on the position and angle of a stationary object extracted from reference image data, based on reference image data to which position information is attached and stored in a data storage unit, and A location estimation device mounted on a vehicle requesting location information comprises a model selection unit that selects a model suitable for the destination from among the models generated by a model generation unit and transmits it as a selected model via a server communicator, a communicator that receives the selected model transmitted from a server, a camera, and a location estimation unit, The camera captures images of objects around the vehicle requesting location information and outputs image data. The position estimation unit is characterized by using a selected model received via a communication device to estimate the position of the vehicle requesting location information based on the position and angle of stationary objects extracted from image data.

[0009] The driver assistance system related to this disclosure is A positioning device mounted on a vehicle that acquires the vehicle's location information, A reference camera that captures images of objects around the vehicle and outputs reference image data, A reference data providing device having a reference data communicator that transmits reference image data to which the location information of the vehicle is attached, A server communicator that receives reference image data to which the aforementioned location information is attached, A data storage unit that stores reference image data to which the aforementioned location information is attached, A model generation unit generates multiple models that estimate the position of a photographed object from the position and angle of a stationary object extracted from the reference image data, based on the reference image data to which the position information is attached and stored in the data storage unit. A server having a model selection unit that selects a model suitable for the destination from among the models generated by the model generation unit and transmits it as a selected model via the server communicator, and A camera that captures images of objects around the vehicle requesting location information and outputs image data, A communication device receives the selected model transmitted from the server, A position estimation device comprising a position estimation unit that estimates the position of the position information request vehicle based on the position and angle of a stationary object extracted from the image data using the selected model received by the communicator.

Advantages of the Invention

[0010] According to the position estimation device and the driving support system according to the present disclosure, it is possible to obtain a position estimation device and a driving support system that enable highly accurate position detection based on an image from a camera that captures surrounding objects, even for a vehicle not equipped with a high-precision positioning device.

Brief Description of the Drawings

[0011] [Figure 1] It is a configuration diagram of a reference data providing device according to Embodiment 1. [Figure 2] It is a configuration diagram of a position estimation device according to Embodiment 1. [Figure 3] It is a configuration diagram of a server according to Embodiment 1. [Figure 4] It is a hardware configuration diagram of a control device according to Embodiment 1. [Figure 5] It is a flowchart showing the processing of the control unit of the reference data providing device according to Embodiment 1. [Figure 6] It is a first flowchart showing the processing of the model management unit of the server according to Embodiment 1. [Figure 7] It is a second flowchart showing the processing of the model management unit of the server according to Embodiment 1. [Figure 8] It is a first flowchart showing the processing of the position estimation unit of the position estimation device according to Embodiment 1. [Figure 9] It is a second flowchart showing the processing of the position estimation unit of the position estimation device according to Embodiment 1. [Figure 10] It is a configuration diagram of a reference data providing device according to Embodiment 2. [Figure 11] It is a configuration diagram of a position estimation device according to Embodiment 2. [Figure 12] It is a flowchart showing the processing of the control unit of the reference data providing apparatus according to Embodiment 2. [Figure 13] It is a first flowchart showing the processing of the model management unit of the server according to Embodiment 2. [Figure 14] It is a second flowchart showing the processing of the model management unit of the server according to Embodiment 2. [Figure 15] It is a first flowchart showing the processing of the position estimation unit of the position estimation apparatus according to Embodiment 2. [Figure 16] It is a second flowchart showing the processing of the position estimation unit of the position estimation apparatus according to Embodiment 2.

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments will be described in detail with reference to the drawings. The drawings are schematically shown, and for convenience of explanation, configurations may be omitted or simplified as appropriate. Also, the mutual relationships of the sizes and positions of the configurations shown in different drawings are not necessarily accurately described and can be changed as appropriate. Also, in the following explanations, the same reference numerals are used to illustrate the same components, and their names and functions are also assumed to be the same. Therefore, detailed explanations thereof may be omitted to avoid duplication.

[0013] 1. Embodiment 1 <Configuration of the Reference Data Providing Apparatus> FIG. 1 is a configuration diagram of a reference data providing apparatus 110 according to Embodiment ①@1. The reference data providing apparatus 110 is arranged on a vehicle 100 equipped with a high-precision positioning device. The vehicle 100 equipped with the high-precision positioning device will hereinafter be abbreviated as the vehicle 100.

[0014] It should be noted that there seems to be an unclear "①@1" in the original text at line 29. If this is an error, please correct it for a more accurate translation.The reference data providing device 110 comprises a reference camera 2, a high-precision positioning device 3, map information 8, a control unit 20, and a reference data communication device 9. The reference camera 2 captures images of objects around the vehicle 100 and outputs them as reference image data. The high-precision positioning device 3 acquires high-precision position information of the vehicle 100. The map information 8 is a map information database that includes a high-precision map of the area around the route traveled by the vehicle 100. The control unit 20 generates reference image data with position information attached from the acquired data. The reference data communication device 9 transmits the reference image data with position information attached.

[0015] <High-precision positioning device> The GPS (Global Positioning System), operated by the United States, is used by civilians with an accuracy of approximately 1.5 to 10 meters. In contrast, by receiving and processing augmentation signals from quasi-zenith satellites (for example, Japan's "Michibiki" (QZSS (Quasi-Zenith Satellite System))), it is possible to detect positions with an accuracy of approximately 0.5 meters. High-precision position detection makes it possible to identify the lane a vehicle is traveling in, contributing to the introduction of autonomous driving technology.

[0016] The high-precision positioning device 3 is a device capable of high-precision positioning using such technology. The high-precision positioning device 3 may also be a device that improves positioning accuracy by comprehensively receiving radio waves from satellite positioning systems other than GPS and QZSS, such as Galileo.

[0017] GPS receivers are extremely widespread, used in vehicle navigation systems and built into smartphones, and are mass-produced, making them small and inexpensive. In contrast, high-precision positioning devices 3 are not as widespread and are used only for special purposes such as surveying. Therefore, it is not practical to equip all vehicles equipped with navigation systems with high-precision positioning devices 3.

[0018] Here, using a select group of vehicles 100 equipped with high-precision positioning devices 3 as a reference, a reference camera 2 installed on the vehicle 100 captures images of surrounding objects. Then, data is transmitted with high-precision position information acquired by the high-precision positioning device 3 added to the image data. We will consider constructing a model that estimates position based on the images by accumulating the transmitted data.

[0019] <Reference Camera> Reference camera 2, as exemplified by surveillance cameras, captures objects and outputs reference image data captured within a certain field of view. Stationary objects can be extracted from the reference image data, and their relative position and angle to reference camera 2 can be obtained. Reference camera 2 can be a visible light camera, an infrared camera, or the like. Reference camera 2 may also include a front camera, rear camera, and surround cameras mounted on vehicle 100. The camera images do not need to be limited to color or grayscale images.

[0020] <Map Information> Map information 8 is a map information database that includes a high-precision map of the area around the route traveled by vehicle 100. Map information 8 can be used to determine whether a stationary object that serves as a landmark is captured in the reference image data of reference camera 2. It can also be used to determine that vehicle 100 is traveling on a road and update its current position when it is traveling through underground passages, tunnels, etc., where radio waves cannot be received by the high-precision positioning device 3.

[0021] <Department Head> The control unit 20 selects reference image data containing stationary objects from the reference image data acquired by the reference camera 2. This is because reference image data without stationary objects is unnecessary for model construction. Then, high-precision position information is added to the reference image data acquired by the reference camera 2.

[0022] <Reference Data Communicator> The reference data communicator 9 is used to transmit reference image data with high-precision location information to the server 410. The server 410 extracts stationary objects from the reference image data with high-precision location information.

[0023] <Configuration of the position estimation device> Figure 2 is a configuration diagram of the position estimation device 310 according to Embodiment 1. The position estimation device 310 is installed in the vehicle requesting position information 300. The vehicle requesting position information 300 is a general vehicle that does not have a high-precision positioning device 3 and requires high-precision position information for itself.

[0024] The position estimation device 310 comprises a camera 12, map information 18, a position estimation unit 30, and a communication device 19. The camera 12 captures images of objects around the vehicle requesting position information 300 and outputs them as image data. The map information 18 is a map information database that includes a high-precision map of the area around the route traveled by the vehicle requesting position information 300.

[0025] The communication device 19 receives a selected model for position estimation from the server 410. The position estimation unit 30 can use the selected model received by the communication device 19 to estimate the high-precision position of the vehicle requesting position information 300.

[0026] <Camera> Camera 12, like a surveillance camera, captures objects and outputs image data captured within a certain field of view. Stationary objects can be extracted from the image data, and their relative position and angle to camera 12 can be obtained. Camera 12 can be a visible light camera, an infrared camera, or the like. Camera 12 may also include a front camera, rear camera, and surround cameras mounted on the vehicle 300 requesting location information. The camera images do not need to be limited to color or grayscale images.

[0027] <Map Information> Map information 18 may be a map information database that includes a high-precision map of the area around the route traveled by the location information requesting vehicle 300, similar to the map information 8 provided in the reference data providing device 110. In addition, the location estimation device 310 uses image data obtained from the camera 12. Since the vehicle's own high-precision location information is acquired, the map information 18 does not need to be a high-precision map.

[0028] Map information 18 can be used to determine whether a stationary object serving as a landmark is captured in the image data from camera 12. This is because image data that does not contain a stationary object serving as a landmark cannot be used and is therefore unnecessary.

[0029] Furthermore, if the location information requesting vehicle 300 is unable to acquire images of stationary objects using the camera 12, it can be used to determine that the vehicle is driving on a road and update its current location. For example, this applies when the camera 12's visibility is poor and images of stationary objects cannot be obtained, such as inside a tunnel, during heavy rain, or in dense fog.

[0030] <Position estimation part> The position estimation unit 30 uses the selected model received by the communication device 19. Based on the position and angle of stationary objects extracted from image data captured by the camera 12, the selected model can be used to estimate the high-precision position of the vehicle requesting location information 300.

[0031] <Communication device> The communication device 19 transmits a signal to the server 410 requesting a position estimation model for the area (travel section) in which the vehicle requesting location information 300 is traveling. The communication device 19 then receives the selected model for position estimation, which has been appropriately selected by the server 410, from the server 410.

[0032] <Server Configuration> Figure 3 is a configuration diagram of the server 410 according to Embodiment 1. It comprises a model management unit 40 consisting of a model generation unit 45 and a model selection unit 47, a data storage unit 48, and a server communicator 49. The server communicator 49 transmits and receives data with the reference data communicator 9 of the reference data providing device 110 and the communicator 19 of the position estimation device 310. The data storage unit 48 is a database that stores the data received by the server communicator 49.

[0033] <Server Communicator> The server communicator 49 receives reference image data with high-precision location information from the reference data communicator 9 of the reference data providing device 110. The server communicator 49 receives a signal from the communicator 19 of the position estimation unit 30 requesting the area (travel section) in which the location information requesting vehicle 300 is traveling, and a signal requesting a new position estimation model. The server communicator 49 then transmits the selected model to the communicator 19 of the position estimation unit 30.

[0034] <Data Storage Section> The data storage unit 48 sequentially stores the reference image data, which has high-precision location information attached, received by the server communicator 49. The data storage unit 48 is a database that aggregates this data and uses it for model construction.

[0035] <Model generation unit> The model generation unit 45 generates multiple models that estimate the high-precision location of a photograph based on the position and angle of a stationary object extracted from the reference image data, using a group of reference image data with high-precision position information stored in the data storage unit 48. Here, a model for estimating the high-precision location corresponds to a calculation formula for estimating the high-precision location. That is, it corresponds to generating a calculation formula that estimates the high-precision location of a photograph from the position and angle of a specific stationary object in the image. The models generated by the model generation unit 45 are stored in the data storage unit 48 and can be retrieved as needed.

[0036] From reference image data with high-precision position information, multiple stationary objects may be extracted, and their corresponding high-precision positions and angles may be used as reference data. Stationary objects may include distinctive buildings, signs, road signs, road walls, guardrails, and road markings. The locations where stationary objects exist are described in map information 8, which is a map information database containing a high-precision map provided in the reference data providing device 110. Therefore, the position and angle of a specific stationary object relative to the photographed location can be corrected by referring to map information 8, making it possible to understand it more accurately.

[0037] The appearance of stationary objects, including buildings, varies depending on factors such as the degree of overhang of roofs, eaves, balconies, and signs, as well as their colors, lighting, and shooting direction. Therefore, high-precision maps alone cannot reveal the actual appearance of individual stationary objects. Reference image data from the reference camera 2 of the reference data provision device 110 can be used to collect visual samples of individual stationary objects. By accumulating reference image data for stationary objects, it becomes possible to generate a highly distinguishable model from the image data acquired by the camera 12 of the position estimation device 310.

[0038] By integrally acquiring reference image data, a model can be generated that estimates the highly accurate location of the shooting site from the position and angle at which each stationary object is visible from the camera. These models may be generated for each stationary object. Alternatively, the region may be divided into a grid and a model may be generated for each divided area. Furthermore, a model may be generated for each predetermined area (driving section) around a road through which a vehicle travels. In addition, a model may be generated for each image capture time or for each weather condition at the time of capture.

[0039] <Model Selection Department> The model selection unit 47 selects a model suitable for the location information requesting vehicle 300, which is the destination of the model, from among the multiple models generated by the model generation unit 45. For example, the appropriate model may be selected by selecting a model generated for the area (travel section) in which the location information requesting vehicle 300 is traveling. The model selection unit 47 transmits the selected model as the selected model to the communication device 19 of the location information requesting vehicle 300 via the server communication device 49.

[0040] Regarding the generation of models for estimating the position of objects such as buildings from images, a method for position estimation using Cross-View with deep learning is disclosed in reference (a). (a)Zhedong Zheng, Yunchao Wei YiYang, University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization, arXiv:2002.12186v2 [cs.CV] 16 Aug 2020, SUSTech-UTS Joint Center of CIS, Southern University of Science and Technology ReLER, AAII, University of Technology Sydney, zhedong.zheng@student.uts.edu.au, yunchao.wei@uts.edu.au, yi.yang@uts.edu.au

[0041] Furthermore, a method for estimating one's own position from the relative position and angle of observed objects using scan matching technology is disclosed in reference (b). (b) Takashi Ienaga (Meiji University), Yoji Kuroda (Meiji University), Improving the Stability of Self-Localization Using Landmark Recognition and Scan Matching, No. 16-2. Proceedings of the 2016 JSME Conference on Robotics and Mechatronics, Yokohama, Japan, June 8-11, 2016.

[0042] Furthermore, a technique for generating multiple models for position estimation and selecting the optimal model is disclosed in reference (c). (c) Shota Nakamichi, Mitsunori Matsushita (Faculty of Informatics, Kansai University), Improving the Classification Accuracy of Signboard Images by Selecting a Location-Based Image Classification Model, Japanese Society for Artificial Intelligence, Interactive Information Access and Visualization Mining Research Group (28th Meeting), SIG-AM-28-06, 2022 / 03 / 04

[0043] The techniques disclosed in references (a), (b), and (c) may be applied to the processing method of the model generation unit 45. Furthermore, the techniques disclosed in references (a), (b), and (c) may be applied to the processing method of the model selection unit 47.

[0044] As described above, high-precision positioning information is acquired by the high-precision positioning device 3 of the reference data provision device 110 mounted on the vehicle 100. Reference image data of the area around the vehicle 100 is acquired by the reference camera 2 of the reference data provision device 110. Then, the reference image data with the high-precision position information attached is transmitted to the server 410 via the reference data communication device 9 of the reference data provision device 110.

[0045] The server communicator 49 of server 410 can receive reference image data with high-precision position information and store it in the data storage unit 48. Then, based on the reference image data with high-precision position information, the model generation unit 45 of server 410 generates multiple models that estimate the position of a stationary object extracted from the reference image data, based on the position and angle of the object. The model selection unit 47 of server 410 selects a model suitable for the destination and transmits it as the selected model to the position estimation device 310 via the server communicator 49.

[0046] In the position estimation device 310, which is located in the vehicle requesting location information 300, the communication device 19 receives the selected model. Then, image data of the area around the vehicle requesting location information 300 is acquired by the camera 12. The position estimation unit 30 extracts stationary objects from the image data and determines the position and angle of the stationary objects. Using the selected model, the position estimation unit 30 estimates the high-precision position of the vehicle requesting location information 300 based on the position and angle of the stationary objects.

[0047] In this way, it becomes possible to estimate the high-precision position of a vehicle 300 requesting location information that is not equipped with a high-precision positioning device 3. It becomes possible to obtain a position estimation device that enables high-precision position detection based on images from a camera 12 that captures surrounding objects.

[0048] <Driving assistance system> The reference data providing device 110, position estimation device 310, and server 410 according to Embodiment 1, as shown in Figures 1 to 3, constitute a driver support system 500 (the driver support system 500 is not shown). The driver support system 500 may include a plurality of reference data providing devices 110. The driver support system 500 may include a plurality of position estimation devices 310.

[0049] By configuring the driver assistance system 500, it becomes possible to estimate the high-precision position of a location information requesting vehicle 300, which is a general vehicle that does not have a high-precision positioning device 3. By using the selected model, the high-precision position can be estimated based on image data of objects around the location information requesting vehicle 300 captured by the camera 12. This makes it possible to promote driver assistance, including autonomous driving.

[0050] Furthermore, by having the vehicle 100 equipped with the high-precision positioning device 3 travel on the road, the latest reference image data of stationary objects can be acquired. As a result, the latest reference image data of stationary objects can be stored in the data storage unit 48 of the server 410, and the model can be updated. Even in the event of building renovations or changes in the appearance of stationary objects, the model can be flexibly updated to accommodate these changes.

[0051] Furthermore, by performing model generation and model selection on the server 410, the processing load on the reference data providing device 110 mounted on the vehicle 100 and the position estimation device 310 mounted on the position information requesting vehicle 300 can be reduced. Since the reference data providing device 110 and the position estimation device 310 are in-vehicle devices, it is desirable that they be small, lightweight, and low-cost. Therefore, the advantage of offloading the software processing load to the server 410 is significant.

[0052] <Control device hardware configuration> Figure 4 is a hardware configuration diagram of the control unit 20. The hardware configuration shown in Figure 4 is also applicable to the position estimation unit 30 and the model management unit 40. Here, we will describe the case where it is applied to the control unit 20 as a representative example. In this embodiment, the control unit 20 is an electronic control device mounted on the vehicle 100 to provide reference data for the vehicle 100. Each function of the control unit 20 is realized by the processing circuits provided in the control unit 20. Specifically, the control unit 20 includes, as processing circuits, a arithmetic processing unit 90 (computer) such as a CPU (Central Processing Unit), a storage device 91 that exchanges data with the arithmetic processing unit 90, an input circuit 92 that inputs external signals to the arithmetic processing unit 90, and an output circuit 93 that outputs signals from the arithmetic processing unit 90 to the outside. Each piece of hardware, such as the arithmetic processing unit 90, the storage device 91, the input circuit 92, and the output circuit 93, is connected to each other by a wired network such as a bus or a wireless network.

[0053] The arithmetic processing unit 90 may include an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), a DSP (Digital Signal Processor), a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), various logic circuits, and various signal processing circuits. Furthermore, multiple arithmetic processing units 90 of the same or different types may be provided, with each unit performing a portion of the processing. The storage device 91 may include a RAM (Random Access Memory) configured to read and write data from the arithmetic processing unit 90, or a ROM (Read-only Memory) configured to read data from the arithmetic processing unit 90. The storage device 91 may also include non-volatile or volatile semiconductor memory such as flash memory, SSD (Solid State Drive), EPROM, EEPROM, magnetic disks, flexible disks, optical disks, compact disks, minidiscs, DVDs, etc. The input circuit 92 is connected to various sensors, switches, and communication lines, and includes an A / D converter, communication circuit, etc., which inputs the output signals and communication information of these sensors and switches to the arithmetic processing unit 90. The output circuit 93 includes a drive circuit, communication circuit, etc., which outputs control signals from the arithmetic processing unit 90. The interfaces of the input circuit 92 and output circuit 93 may be based on specifications such as CAN (Control Area Network) (registered trademark), Ethernet (registered trademark), USB (Universal Serial Bus) (registered trademark), DVI (Digital Visual Interface) (registered trademark), and HDMI (High-Definition Multimedia Interface) (registered trademark). In addition, communication may be performed by directly connecting the arithmetic processing unit 90 to a communication device, separate from the input circuit 92 and output circuit 93.

[0054] Each function of the control unit 20 is realized by the arithmetic processing unit 90 executing software (programs) stored in a storage device 91 such as ROM, and cooperating with other hardware of the control unit 20, such as the storage device 91, input circuit 92, and output circuit 93. Setting data such as thresholds and judgment values ​​used by the control unit 20 are stored in the storage device 91 such as ROM as part of the software (program). Each function of the control unit 20 may be composed of software modules, or it may be composed of a combination of software and hardware.

[0055] <Handling by the Department Head> Figure 5 is a flowchart showing the processing of the control unit 20 of the reference data providing device 110 according to Embodiment 1. The processing shown in Figure 5 is executed by the arithmetic processing unit of the control unit 20. This processing may be executed at predetermined intervals (for example, every 1 ms). Alternatively, it may be executed based on events, such as when the control unit 20 acquires high-precision position information from the high-precision positioning device 3, or when the control unit 20 acquires reference image data from the reference camera 2.

[0056] The process shown in Figure 5 is initiated, and in step S101, the control unit 20 acquires high-precision position information of the vehicle 100 from the high-precision positioning device 3. Then, in step S102, the control unit 20 acquires reference image data of the area around the vehicle 100 from the reference camera 2.

[0057] In step S109, the control unit 20 transmits the reference image data, which has been given high-precision position information, to the server 410 via the reference data communicator 9. Then, the process ends.

[0058] <Processing by the Model Management Department> Figure 6 is a first flowchart showing the processing of the model management unit 40 of the server 410 according to Embodiment 1. Figure 7 is a second flowchart showing the processing of the model management unit 40 of the server 410. The processing shown in Figures 6 and 7 is executed by the arithmetic processing unit of the model management unit 40. This processing may be executed at predetermined intervals (for example, every 1 ms).

[0059] The process in Figure 7 may be executed immediately after the process in Figure 6. The processes in Figure 6 and Figure 7 may be executed alternately. Also, the processes in Figure 6 and Figure 7 may be executed at different times.

[0060] The process in Figure 6 may be executed in response to an event, such as each time the server communicator 49 of server 410 receives data from the reference data providing device 110. The process in Figure 7 may also be executed in response to an event, such as each time the server communicator 49 of server 410 receives data from the position estimation device 310.

[0061] The process shown in Figure 6 is explained below. After starting the process, in step S201, it is checked whether data has been received from the reference data supply device 110. In step S202, if data has been received from the reference data supply device 110 (determination is YES), the process proceeds to step S203. If data has not been received from the reference data supply device 110 (determination is NO), the process is terminated.

[0062] In step S203, reference image data with high-precision positioning information attached is downloaded from the reference data providing device 110. In step S204, the downloaded data is stored in the data storage unit 48.

[0063] In step S205, stationary objects are extracted from the reference image data. In step S206, multiple models are generated or updated to estimate the high-precision position of the observation point, i.e., the location where the image was taken, from the position and angle of the stationary objects. Since it is sufficient to generate multiple models in total, one model may be generated or updated each time step S206 is executed. In this case, multiple models will be generated or updated by executing step S205 multiple times. Alternatively, multiple models may be generated or updated at once by executing step S205. After that, the process ends.

[0064] The process shown in Figure 7 is explained below. After starting the process, in step S211, it is checked whether a model transmission request has been received from the position estimation device 310. In step S212, if a model transmission request has been received from the position estimation device 310 (determination is YES), the process proceeds to step S213. If there is no model transmission request from the position estimation device 310 (determination is NO), the process ends.

[0065] In step S213, the driving section information of the location information requesting vehicle 300 is received. In step S214, an appropriate model corresponding to the driving section is selected. In step S215, the time to start transmitting the selected model is set. Here, the transmission time of the model may be determined by calculating backward from the time required to complete the transmission of the selected model, and taking into account the time when the location information requesting vehicle 300 is approaching the relevant driving area (driving section), the transmission of the appropriate model has been completed.

[0066] In step S216, it is determined whether the transmission time set in step S215 has arrived. If it is determined in step S216 that the transmission time has arrived (determination is YES), the process proceeds to step S217. If it is determined that the transmission time has not arrived (determination is NO), the process terminates.

[0067] In step S217, the selected model is sent to the location information requesting vehicle 300, and the process ends.

[0068] <Processing by the position estimation unit> Figure 8 is a first flowchart showing the processing of the position estimation unit 30 of the position estimation device 310 according to Embodiment 1. Figure 9 is a second flowchart showing the processing of the position estimation unit 30 of the position estimation device 310. The processing shown in Figures 8 and 9 is executed by the arithmetic processing unit of the position estimation unit 30. This processing may be executed at predetermined intervals (for example, every 1 ms).

[0069] The process in Figure 9 may be executed immediately after the process in Figure 8. The processes in Figure 8 and Figure 9 may be executed alternately. Also, the processes in Figure 8 and Figure 9 may be executed at different times.

[0070] The process shown in Figure 8 may be executed by event, such as each time the position estimation unit 30 of the position estimation device 310 estimates the position of the vehicle requesting position information 300. The process shown in Figure 9 may also be executed by event, such as each time the position estimation unit 30 of the position estimation device 310 receives data from the server 410.

[0071] The process shown in Figure 8 will now be explained. After starting the process, in step S251, the position estimation unit 30 checks whether the model currently in use is appropriate. For example, if a model generated for each travel section is being used, it checks whether the model matches the travel section that the vehicle requesting position information 300 is currently traveling on.

[0072] In step S252, if the currently used model is appropriate (determined as YES), the process terminates. If the currently used model is not appropriate (determined as NO), proceed to step S253.

[0073] In step S253, the vehicle requesting location information sends information about the current driving section to the server 410. At this time, taking into account the time it takes from sending the model request until the model download is complete, information about the planned driving section that the vehicle will be traveling after that time has elapsed may be sent to the server 410 as driving section information. Then, in step S254, the model request is sent to the server 410. After that, the process ends.

[0074] The process shown in Figure 9 will now be explained. After starting the process, in step S301, it is checked whether the position estimation device 310 has received the model from the server 410.

[0075] If the position estimation device 310 receives a model from the server 410 in step S302 (determined as YES), the process proceeds to step S303. If the position estimation device 310 does not receive a model from the server 410 (determined as NO), the process terminates.

[0076] In step S303, the model sent from server 410 is downloaded. In step S305, the image data output by camera 12 is acquired.

[0077] In step S308, stationary objects are extracted from the acquired image data. In step S309, the position and orientation of the extracted stationary objects are calculated.

[0078] In step S311, the model downloaded from server 410 is used to estimate the high-precision position of the vehicle requesting location information 300 based on the position and orientation of the extracted stationary objects. The process is then terminated.

[0079] 2. Embodiment 2 <Configuration of the reference data provision device> Figure 10 is a configuration diagram of the reference data providing device 110 according to Embodiment 2. The difference from the reference data providing device 110 according to Embodiment 1 shown in Figure 1 is that an IMU 4 and a distance measuring device 6 have been added. The hardware configuration diagram in Figure 4 can also be applied to the control unit 20 in Figure 10. The server 410 according to Embodiment 2 will use the same configuration as in Figure 3.

[0080] <imu> An IMU (Inertial Measurement Unit) 4, also known as an inertial measurement device, is a device that measures three-dimensional displacement, velocity, and acceleration, as well as three-axis rotation, angular velocity, and angular acceleration. Specifically, it consists of a gyroscope, velocity sensor, acceleration sensor, and other components.

[0081] The IMU4 shown in Figure 10 is not limited to three dimensions and three axes; it may also be a device that measures displacement, velocity, acceleration, rotation, angular velocity, and angular acceleration in two dimensions, two axes, or one dimension, one axis. Here, since velocity can be calculated by differentiating displacement, and acceleration can be calculated by differentiating velocity, it is sufficient to measure either of these parameters. Similarly, since angular velocity can be calculated by differentiating the rotation angle, and angular acceleration can be calculated by differentiating the angular velocity, it is sufficient to measure either of these parameters.

[0082] By incorporating the IMU4, the reference data provider 110 can estimate the position of the vehicle 100 even if the high-precision positioning device 3 is temporarily unavailable. Specifically, this could occur when the vehicle 100's view of the sky is obstructed by densely packed buildings, or when satellite signals are blocked while traveling through underpasses, tunnels, etc. By continuing to detect the vehicle 100's position using the IMU4, the reference data provider 110 can continue to transmit image data with high-precision position information.

[0083] <Distance measuring device> The rangefinder 6 is a device for measuring the distance to a stationary object. A millimeter-wave radiometer (MMWR) utilizing the 24-79 GHz frequency band can be used as the rangefinder 6. It is possible to detect the object's position using radio waves and to detect its speed through the Doppler effect. Laser radar, LiDAR (Light Detection and Ranging), etc., can also be used as the rangefinder 6. By irradiating a laser beam within a certain field of view and detecting point cloud data obtained from the reflection of the laser beam from the object, the object's position and shape can be determined.

[0084] By including the distance measuring device 6, the reference data providing device 110 can further add information regarding the distance of a stationary object to the image data with high-precision position information and transmit it. Thus, by adding the measured distance information by the distance measuring device 6, the accuracy of the information regarding the position and direction of the stationary object extracted from the image data can be improved. As a result, more accurate model generation at the server 410 becomes possible.

[0085] <Configuration of the position estimation device> FIG. 11 is a configuration diagram of the position estimation device 310 according to Embodiment 2. The difference from the position estimation device 310 according to Embodiment 1 shown in FIG. 2 is that a GPS receiver 13, an IMU 14, and a distance measuring device 16 are added. The hardware configuration diagram of FIG. 4 is also applicable to the position estimation unit 30 of FIG. 11.

[0086] <GPS receiver> Since the position estimation device 310 includes the GPS receiver 13, the position estimation device 310 can obtain approximate position information of the position information requesting vehicle 300 as an initial state. For this purpose, the running section information during the running of the position information requesting vehicle 300 can be transmitted to the server 410. The selected model received from the server 410 can be set as a model of an appropriate running section from the beginning. Therefore, even in the initial state, it is possible to perform high-precision position detection using the selected model from the beginning.

[0087] <imu> It is also possible to detect the driving speed of the location information requesting vehicle 300 using the IMU 14. When the model generation unit 45 of the server 410 generates a model, if different models are generated for each driving speed range, it becomes possible to select and use the model for the driving speed range that is suitable for the driving speed of the location information requesting vehicle 300.

[0088] Furthermore, by incorporating the IMU 14, the position estimation device 310 can estimate the position of the vehicle requesting location information 300 even if the camera 12 is temporarily unavailable. Specifically, this applies to situations such as when the vehicle is surrounded by dense fog or heavy rain, or when it is traveling through a tunnel. The IMU 14 makes it possible to continue detecting the position of the vehicle requesting location information 300.

[0089] <Distance measuring device> The rangefinder 16 is a device for measuring the distance to a stationary object. A millimeter-wave radar can be used as the rangefinder 16. It is possible to detect the position of an object using radio waves and to detect the object's speed of movement using the Doppler effect. Laser radar, LiDAR (Light Detection and Ranging), etc. can also be used as the rangefinder 16. By irradiating a laser beam within a certain field of view and detecting point cloud data obtained from the reflection of the laser beam from the object, the position and shape of the object can be determined.

[0090] By incorporating the distance measuring device 6, the position estimation device 310 can more accurately correct the values ​​when extracting stationary objects from images acquired by the camera 12 and calculating the distance and angle of stationary objects. By more accurately calculating the distance and angle of stationary objects, it becomes possible to estimate more accurate, high-precision position information using the selected model.

[0091] <Other parameters> Although not shown in Figures 10 and 11, the reference data providing device 110 and the position estimation device 310 may also be equipped with a weather sensor to detect the surrounding weather conditions or a weather information provider to obtain weather information. In this case, the model management unit 40 of the server 410 can generate and select different models depending on the weather conditions, and use them in the position estimation device 310.

[0092] Furthermore, the reference data providing device 110 and the position estimation device 310 may be equipped with clocks capable of recognizing the time and season. In this case, the model management unit 40 of the server 410 can generate and select different models depending on the time and season, and use them in the position estimation device 310.

[0093] Furthermore, the model management unit of server 410 may generate and select different models for each vehicle type and camera installation location, and use them in the position estimation device 310.

[0094] <Driving assistance system> The driver assistance system 500 according to Embodiment 2 consists of a reference data providing device 110 shown in Figure 10, a position estimation device 310 shown in Figure 11, and a server 410 shown in Figure 3. The driver assistance system 500 may include a plurality of reference data providing devices 110.

[0095] The driver assistance system 500 may include multiple position estimation devices 310. The driver assistance system 500 can select an appropriate model to estimate the high-precision position of the vehicle requesting position information 300. This will enable driver assistance, including autonomous driving.

[0096] <Handling by the Department Head> Figure 12 is a flowchart showing the processing of the control unit 20 of the reference data providing device 110 according to Embodiment 2. The difference between the processing shown in Figure 12 and the processing shown in Figure 5 according to Embodiment 1 is that steps S103 to S108 are added after step S102, and steps S119 and S110 are added instead of step S109.

[0097] Let me explain the added parts. In step S103, the position information of the reference camera 2 of the reference data providing device 110 of the vehicle 100 is acquired. Of the position of the reference camera 2, the most important is the ground height at which the camera is positioned. This is because the field of view of the image data differs depending on the ground height of the reference camera. Therefore, the model generation unit 45 of the server 410 may generate different models according to the ground height of the reference camera 2.

[0098] In step S104, time information is acquired. Even when traveling on the same road in the same direction, the reference image will differ as the time changes due to the influence of brightness, direction of sunlight, electric lights, and streetlights. Therefore, the model generation unit 45 of the server 410 may generate different models depending on the time (time period) captured by the reference camera 2.

[0099] In step S105, weather information is acquired. The reference data providing device 110 may be equipped with a weather sensor that detects the surrounding weather conditions or a weather information device that can obtain weather information. This is because the reference image is greatly affected by the weather conditions. Therefore, the model generation unit 45 of the server 410 may generate different models depending on the weather information captured by the reference camera 2.

[0100] In step S106, seasonal information is acquired. Seasonal information can be obtained from the date of the time information. Seasonal information is acquired because the reference image is greatly affected by the season. Therefore, the model generation unit 45 of the server 410 may generate different models depending on the seasonal information at the time of capture by the reference camera 2.

[0101] In step S107, vehicle speed information is acquired. Vehicle speed information is acquired because the reference image is affected by the vehicle speed. In particular, when the vehicle speed is high, the reference image changes in a short time. Therefore, it becomes difficult to extract stationary objects in the vicinity of the vehicle 100 from the reference image. Thus, when the vehicle speed is high, it is necessary to select stationary objects such as large buildings that can be identified in a short time and do not easily go out of the field of view to generate a model. The model generation unit 45 of the server 410 may generate different models depending on the vehicle speed information captured by the reference camera 2.

[0102] In step S108, vehicle information is acquired. This is because different vehicle types require different camera installation locations, resulting in variations in the field of view of the reference image. The model generation unit 45 of server 410 may generate different models depending on the vehicle information captured by the reference camera 2. If some of this information is unnecessary, the corresponding steps in steps S103 to S108 may be omitted.

[0103] In step S119, the control unit 20 transmits reference image data, which includes high-precision location information and other information, to the server 410 via the reference data communication device 9. In step S110, payment for the data upload is obtained. The map generated by the server 410 is updated when the reference image data with the latest high-precision location information is transmitted from the reference data providing device 110. The image visible from the road on which the vehicle 100 is traveling changes constantly. Changes can occur due to building renovations and additions, signs, decorations, landscaping, and changes in appearance due to construction, so the model always needs to be updated. Therefore, payment may be made for data uploads from the reference data providing device 110. The payment may not be money, but may be a right such as the right to upload the latest high-precision map. The payment may be given each time reference image data is transmitted, but it may also be a contract for a fixed amount to be paid at predetermined intervals.

[0104] <Processing by the Model Management Department> Figure 13 is a first flowchart showing the processing of the model management unit 40 of the server 410 according to Embodiment 2. Figure 14 is a second flowchart showing the processing of the model management unit 40 of the server 410 according to Embodiment 2. The processing shown in Figures 13 and 14 is executed by the arithmetic processing unit of the model management unit 40. This processing may be executed at predetermined intervals (for example, every 1 ms).

[0105] The process in Figure 14 may be executed immediately after the process in Figure 13. The processes in Figure 13 and Figure 14 may be executed alternately. Also, the processes in Figure 13 and Figure 14 may be executed at different times.

[0106] The process shown in Figure 13 may be executed in response to an event, such as each time the server communicator 49 of server 410 receives data from the reference data providing device 110. The process shown in Figure 14 may also be executed in response to an event, such as each time the server communicator 49 of server 410 receives data from the position estimation device 310.

[0107] The only difference between the process in Figure 13 and the process in Figure 6 relating to Embodiment 1 is that step S203 is replaced with step S223. In step S223 of Figure 13, reference image data to which high-precision positioning information and other information are added is downloaded.

[0108] Other information includes the location information, time information, weather information, seasonal information, vehicle speed information, and vehicle type information of the reference camera 2. Some of this information may be omitted. This information, along with high-precision positioning information, is added to the reference image data and downloaded by the model management unit 40 of the server 410.

[0109] The only difference between the process in Figure 14 and the process in Figure 7 relating to Embodiment 1 is that step S213 is replaced with step S233. In step S233 of Figure 14, the vehicle type, camera mounting location, current time, season, weather, vehicle speed, and driving section information of the location information requesting vehicle 300 are received. Some of this information may be omitted. Based on this information, the optimal model is selected in step S214.

[0110] <Processing by the position estimation unit> Figure 15 is a first flowchart showing the processing of the position estimation unit 30 of the position estimation device 310 according to Embodiment 2. Figure 16 is a second flowchart showing the processing of the position estimation unit 30 of the position estimation device 310 according to Embodiment 2. The processing shown in Figures 15 and 16 is executed by the arithmetic processing unit of the position estimation unit 30. This processing may be executed at predetermined time intervals (for example, every 1 ms).

[0111] The process in Figure 16 may be executed immediately after the process in Figure 15. The processes in Figure 15 and Figure 16 may be executed alternately. Also, the processes in Figure 15 and Figure 16 may be executed at different times.

[0112] The process shown in Figure 15 may be executed by events, such as each time the position estimation unit 30 of the position estimation device 310 estimates the position of the vehicle requesting position information 300. The process shown in Figure 16 may also be executed by events, such as each time the position estimation unit 30 of the position estimation device 310 receives data from the server 410.

[0113] The only difference between the processing in Figure 15 and the processing in Figure 8 relating to Embodiment 1 is that step S253 is replaced with step S263. In step S263 of Figure 15, the vehicle type, camera mounting location, current time, season, weather, vehicle speed, and driving area information of the location information requesting vehicle 300 are transmitted to the server 410. Some of this information may be omitted. Information for model selection when sending a model request is transmitted in step S254.

[0114] The process shown in Figure 16 differs from the process shown in Figure 9 in Embodiment 1. The only difference is that step S304 is added after step S303, steps S306 and S307 are added after step S305, and step S310 is added after step S309.

[0115] Let me explain the added parts. In step S304, the position estimation device 310 of the location information requesting vehicle 300 pays the model download fee. Payment may be made each time a model is downloaded, or it may be a fixed amount paid at predetermined intervals. This financial burden allows the vehicle 100 equipped with the high-precision positioning device to provide assistance for driving. In this way, even a vehicle 100 that is not equipped with the high-precision positioning device 3 can perform high-precision position detection.

[0116] After step S305, in step S306, positioning information is acquired by the GPS receiver 13 of the vehicle requesting location information 300. In step S307, distance measurement information, which is the distance from the distance measuring device 16 to the stationary object, is acquired. After step S309, in step S310, the position and direction of the stationary object are corrected using the distance measurement information.

[0117] Regarding Figure 16, the explanation described how the position estimation device 310 downloads a model, uses the model to extract stationary objects from camera image data, and estimates the high-precision position of the vehicle requesting location information 300 based on the position and orientation of the stationary objects. However, in order to reduce the processing load on the in-vehicle equipment, the position estimation device 310, the processes of "extracting stationary objects from image data," "clarifying the position and orientation of stationary objects," and "estimating the high-precision position of the vehicle requesting location information 300 using the model" may all be performed on the server 410.

[0118] The processing unit of server 410 is relatively large and has a high execution speed compared to the in-vehicle equipment. Therefore, by transferring processing to server 410, it is expected that the process of "estimating the high-precision position of the location information requesting vehicle 300" will be accelerated. This will reduce the processing load on the in-vehicle equipment, allowing it to concentrate on tasks that require real-time processing.

[0119] While this application describes various exemplary embodiments and examples, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but are applicable individually or in various combinations to the embodiments. Accordingly, countless variations not illustrated are envisioned within the scope of the art disclosed herein. These include, for example, modifying, adding or omitting at least one component, or even extracting at least one component and combining it with a component from another embodiment. [Explanation of Symbols]

[0120] 2 Reference camera, 3 High-precision positioning device, 4 IMU, 6, 16 Distance measuring device, 8, 18 Map information, 9 Reference data communicator, 12 Camera, 13 GPS receiver, 14 IMU, 19 Communicator, 20 Control unit, 30 Position estimation unit, 40 Model management unit, 45 Model generation unit, 47 Model selection unit, 48 Data storage unit, 49 Server communicator, 100 Vehicle equipped with high-precision positioning device, 110 Reference data providing device, 300 Vehicle requesting location information, 310 Position estimation device, 410 Server< / imu> < / imu>

Claims

1. A server communication device that receives reference image data, which is an image of an object around the vehicle, to which position information determined by a positioning device mounted on the vehicle has been added. A data storage unit that stores the reference image data to which the aforementioned location information is attached, A model generation unit generates multiple models that estimate the position of a photograph based on the position and angle of a stationary object extracted from the reference image data, based on the reference image data to which the position information is attached and stored in the data storage unit, and A location estimation device mounted on a vehicle requesting location information comprises a model selection unit that selects a model suitable for the destination from among the models generated by the model generation unit and transmits it as a selected model via the server communication device, a communication device that receives the selected model transmitted from a server, a camera, and a location estimation unit, The camera captures images of objects around the vehicle requesting location information and outputs image data. The position estimation device is characterized in that the position estimation unit estimates the position of the vehicle requesting position information based on the position and angle of a stationary object extracted from the image data, using the selected model received via the communication device.

2. The positioning device mounted on the vehicle is a positioning device according to claim 1, which determines the position based on signals from GPS satellites and quasi-zenith satellites.

3. The vehicle requesting location information is equipped with a rangefinder that measures the distance to stationary objects in its vicinity, The position estimation device according to claim 1, wherein the position estimation unit corrects the position and angle of a stationary object extracted from the image data using the distance measured by the distance measuring device, and estimates the position of the vehicle requesting position information using the selected model.

4. A positioning device mounted on a vehicle that acquires the vehicle's location information, A reference camera that captures images of objects around the vehicle and outputs reference image data, A reference data providing device having a reference data communicator that transmits reference image data to which the location information of the vehicle is attached, A server communicator that receives reference image data to which the aforementioned location information is attached, A data storage unit that stores reference image data to which the aforementioned location information is attached, A model generation unit generates multiple models that estimate the position of a photographed object from the position and angle of a stationary object extracted from the reference image data, based on the reference image data to which the position information is attached and stored in the data storage unit. A server having a model selection unit that selects a model suitable for the destination from among the models generated by the model generation unit and transmits it as a selected model via the server communicator, and A camera that captures images of objects around the vehicle requesting location information and outputs image data, A communication device receives the selected model transmitted from the server, A driving assistance system comprising a position estimation device having a position estimation unit that estimates the position of the vehicle requesting position information based on the position and angle of a stationary object extracted from the image data, using the selected model received by the communication device.

5. The driving support system according to claim 4, wherein the positioning device of the reference data providing device determines the position based on signals from GPS satellites and quasi-zenith satellites.

6. The aforementioned reference data providing device has a distance measuring device for measuring the distance to the stationary object, The reference data communicator of the reference data providing device transmits reference image data to which the position information and the distance to the stationary object are added. The server communication device of the server receives reference image data to which the location information and the distance to the stationary object are assigned. The data storage unit of the server stores reference image data to which the position information and the distance to the stationary object are assigned. The driving support system according to claim 4, wherein the model generation unit of the server generates a plurality of models that estimate the position of a photographed object from the position and angle of a stationary object extracted from the reference image data, based on the reference image data to which the position information and the distance to the stationary object are assigned, stored in the data storage unit.

7. The communication device of the reference data providing device transmits reference image data to which the location information and the mounting location information of the reference camera are attached. The server communication device of the server receives reference image data to which the location information and the mounting location information of the reference camera are attached. The data storage unit of the server stores reference image data to which the location information and the mounting position information of the reference camera are attached. The driver assistance system according to claim 4, wherein the model generation unit of the server generates a plurality of models that estimate the position of a photographed object from the position and angle of a stationary object extracted from the reference image data, based on the reference image data to which the position information and the mounting position information of the reference camera stored in the data storage unit are attached.

8. The position estimation device includes a distance measuring device that measures the distance to stationary objects around the vehicle requesting position information. The driving support system according to claim 4, wherein the position estimation unit corrects the position and angle of a stationary object extracted from the image data using the distance measured by the distance measuring device, and estimates the position of the vehicle requesting position information based on the selected model.

9. The model generation unit of the server generates multiple models according to the vehicle type, The driver assistance system according to claim 4, wherein the model selection unit selects a model corresponding to the vehicle of the recipient from among the models generated by the model generation unit and transmits it as the selected model via the server communication device.

10. The model generation unit of the server generates a plurality of models corresponding to the mounting position of the camera of the vehicle requesting location information, The driving support system according to claim 4, wherein the model selection unit selects a model corresponding to the mounting position of the camera at the transmission destination from among the models generated by the model generation unit and transmits it as the selected model via the server communication device.

11. The model generation unit of the server generates multiple models according to the vehicle speed, The driver assistance system according to claim 4, wherein the model selection unit selects a model corresponding to the vehicle speed of the destination from among the models generated by the model generation unit and transmits it as the selected model via the server communicator.

12. The model generation unit of the server generates multiple models according to the time, The driver support system according to claim 4, wherein the model selection unit selects a model corresponding to the time of the transmission destination from among the models generated by the model generation unit and transmits it as the selected model via the server communicator.

13. The model generation unit of the server generates multiple models according to the weather, The driver assistance system according to claim 4, wherein the model selection unit selects a model corresponding to the weather conditions at the destination from among the models generated by the model generation unit and transmits it as the selected model via the server communicator.

14. The model generation unit of the server generates multiple models according to the season, The driver assistance system according to claim 4, wherein the model selection unit selects a model corresponding to the season of the destination from among the models generated by the model generation unit and transmits it as the selected model via the server communicator.

15. The driver assistance system according to claim 4, wherein the model selection unit of the server selects a model corresponding to the planned driving section of the destination and transmits it as the selected model via the server communication device.

16. The driver assistance system according to claim 4, wherein the model selection unit of the server determines the transmission start time according to the time until the transmission of the selected model is completed.

17. The driving support system according to any one of claims 4 to 16, wherein the position estimation device is required to pay a fee when it receives the selected model via the communication device.

18. The driver assistance system according to any one of claims 4 to 16, wherein the reference data providing device is paid when it transmits the reference image data to which the location information is attached via the communication device.

Images