A vehicle fusion positioning method and device based on road administration facility recognition

Abstract

The present invention discloses a vehicle fusion positioning method based on road administration facility identification. Road administration facilities are set around the road, and the road administration facilities include a characteristic part, a characteristic part area and a first coordinate; the vehicle obtains the characteristic part through a first camera unit, and obtains the characteristic part area and the first coordinate according to the characteristic part; the vehicle is positioned according to the first imaging position of the characteristic part in the first camera unit, the first imaging characteristic part area and the first coordinate. A vehicle fusion positioning device and a computer readable medium are also disclosed. By utilizing the high-precision positioning road administration facilities on the road, high-precision positioning data can be obtained by identifying the road administration facilities during the driving of the vehicle positioned using ordinary maps. At the same time, a more accurate and reliable coordinate basis is also provided for unmanned driving technology.

Description

Technical Field

[0001] The present invention relates to the technical field of big data vehicle fusion positioning, and in particular to a vehicle fusion positioning method and device based on road administration facility identification. Background Art

[0002] In vehicle positioning technology, the use of inertial navigation, satellite navigation and wheel speedometer to form a fusion positioning system is a very common method, and relatively satisfactory positioning performance can be obtained in many scenarios. Compared with the use of a single navigation device, it has the advantages of full autonomy, all-weather, and no interference from external information. However, the development of autonomous driving technology has put forward higher requirements on the performance of the fusion positioning system. The fusion positioning device needs to achieve real-time and uninterrupted centimeter-level positioning in all scenarios. The fusion positioning device composed of inertial navigation, satellite navigation and wheel speedometer is difficult to meet the above requirements. In the case of long-term loss of satellite signals, especially in places such as tunnels or viaducts, the positioning error of the inertial / wheel speedometer fusion positioning will accumulate with the increase of vehicle mileage, causing the positioning result to gradually deviate from the actual position of the vehicle, resulting in the inability to continue autonomous driving.

[0003] To solve this problem, commonly used technical solutions include adding map matching, adding lidar positioning, adding visual navigation and other technical means or the superposition of several means. However, the continuous high-precision positioning of map matching requires obvious geometric features of the driving route, and the actual driving route of the vehicle is difficult to guarantee the feature requirements; the relative positioning of lidar also has the problem of error accumulation, and absolute positioning requires prior mapping; visual navigation also has the problem of error accumulation, and a closed-loop path needs to be verified through loopback to eliminate the accumulated error, and the actual driving route of the vehicle is difficult to guarantee a closed-loop path. At the same time, the existing positioning correction methods are limited, and it is difficult for vehicles that originally have low-precision positioning to obtain high-precision positioning data. Summary of the invention

[0004] Based on the above situation, the present invention proposes a vehicle fusion positioning method based on road facilities identification, which utilizes the high-precision positioning road facilities on the road, so that vehicles that do not use ordinary map positioning can obtain high-precision positioning data by identifying road facilities during driving. On the other hand, it can also provide more accurate and reliable coordinate basis for autonomous driving technology and adjust the inertial navigation parameters in time.

[0005] The present invention provides a vehicle fusion positioning method based on road maintenance facility identification, wherein road maintenance facilities are arranged around a road, and the road maintenance facilities include a characteristic part, an area of ​​the characteristic part, and a first coordinate; a vehicle obtains the characteristic part through a first camera unit, and obtains the area of ​​the characteristic part and the first coordinate based on the characteristic part; the vehicle is positioned based on a first imaging position of the characteristic part in the first camera unit, a first imaging characteristic part area, and the first coordinate.

[0006] The process of the vehicle obtaining the characteristic part area and the first coordinate includes: uploading the current rough coordinates of the vehicle, obtaining the characteristic part of the road facilities around the rough coordinates of the first coordinate, and obtaining the characteristic part area and the first coordinate from the server according to the characteristic part.

[0007] The step of positioning the vehicle includes: calculating a first distance between the vehicle and the road maintenance facility based on the area of ​​the characteristic part and the numerical value of the area of ​​the first imaging characteristic part; acquiring a first angle between the vehicle and the road maintenance facility based on the imaging position of the characteristic part in the first camera unit; and calculating the vehicle position coordinates using the first distance, the first angle and the first coordinates of the road maintenance facility.

[0008] At a first moment, the first imaging position and the first imaging characteristic part area of ​​the road maintenance facility in the first camera unit are obtained; at a second moment, the second imaging position and the second imaging characteristic part area of ​​the two-dimensional code in the first camera unit are obtained; the offset angle of the vehicle is calculated according to the first imaging position and the second imaging position, and the offset distance of the vehicle is calculated according to the first imaging characteristic part area and the second imaging characteristic part area; according to the time difference between the first moment and the second moment, the angular acceleration and linear acceleration of the vehicle are calculated, and the IMU data of the vehicle is corrected.

[0009] When the area of ​​the first imaging feature part is smaller than a preset threshold, the vehicle may be provided with a second camera unit; a distance is set between the first camera unit and the second camera unit; a first angle between the first camera unit and the road maintenance facility is obtained, and a second angle between the second camera unit and the road maintenance facility is obtained, and the relative position information between the vehicle and the road maintenance facility is calculated based on the distance, the first angle and the second angle, and the vehicle position coordinates are calculated using the first coordinates and the relative position information.

[0010] Road facilities may also include characteristic line segments and characteristic line segment lengths; when the vehicle obtains the characteristic part through the first camera unit and obtains the characteristic line segment and characteristic line segment length through the server; the characteristic line segment is used to replace the characteristic part and the characteristic line segment length is used to replace the characteristic part area.

[0011] At the same time, the present invention also provides a vehicle fusion positioning device based on road facilities identification, which is characterized by comprising: a vehicle camera module, a server and a positioning module, and each module is data connected;

[0012] The vehicle camera module comprises at least a first camera unit installed on the vehicle; and is used to obtain a characteristic part of a road facility set beside the road through the first camera unit; the road facility comprises a characteristic part, an area of ​​the characteristic part and a first coordinate;

[0013] The server is used to store the characteristic part, the area of ​​the characteristic part and the first coordinate information of the road facilities; when the vehicle camera module obtains the characteristic part of the road facilities, the characteristic part area and the first coordinate corresponding to the road facilities are obtained according to the characteristic part; the positioning module is used to position the vehicle according to the first imaging position, the first imaging characteristic part area and the first coordinate of the road facilities in the first camera unit.

[0014] In addition, the present disclosure proposes a computer-readable medium, in which a computer program is stored. The computer program is loaded and executed by a processing module to implement a vehicle fusion positioning method.

[0015] Some technical effects of the present disclosure are: by using the road facilities with high precision positioning on the roadside, the vehicle can obtain its own high-precision positioning information by scanning the road facilities during the vehicle's driving. In this way, on the one hand, the vehicle can obtain high-precision positioning data without using high-precision equipment, and on the other hand, it can provide more accurate and reliable coordinate basis for unmanned driving technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To better understand the technical solution of the present disclosure, you may refer to the following drawings for auxiliary explanation of the prior art or embodiments. These drawings will selectively display the products or methods involved in the prior art or some embodiments of the present disclosure. The basic information of these drawings is as follows:

[0017] Figure 1 It is a flow chart of an embodiment of a vehicle fusion positioning method based on road facilities identification of the present invention.

[0018] Figure 2 It is a schematic diagram of the characteristic parts corresponding to the conventional road facilities in the embodiment of the present invention.

[0019] Figure 3 It is a schematic diagram from a first angle in an embodiment of the present invention.

[0020] Figure 4 It is a flow chart of an embodiment of a vehicle fusion positioning device based on road facilities identification of the present invention. DETAILED DESCRIPTION

[0021] The technical means or technical effects involved in the present disclosure will be further described below. Obviously, the embodiments provided are only some of the embodiments of the present disclosure, not all of them. Based on the embodiments in the present disclosure and the explicit or implicit descriptions of the figures and texts, all other embodiments that can be obtained by those skilled in the art without making creative efforts will be within the scope of protection of the present disclosure.

[0022] like Figure 1 As shown, the method in this embodiment includes the steps of:

[0023] S1: Setting up road facilities around the road, wherein the road facilities include a characteristic part, an area of ​​the characteristic part and a first coordinate.

[0024] At present, some road facilities are set up around ordinary roads and highways (including the left and right sides of the road or above the road, and some are set up on the road for better distinction, such as printed fonts or road signs). These road facilities can be positioned by high-precision positioning equipment when they are arranged. According to the role played by the road facilities and the conventional graphics used, the characteristic part of the road facilities and the corresponding characteristic part area can be known when they are arranged; the characteristic part of the road facilities, the corresponding characteristic part area and the corresponding first coordinates are uploaded to the server, so that each road facility will have a characteristic part, characteristic part area and unique coordinates. Figure 2As mentioned above, road facilities generally include traffic signs, street lamps, and traffic lights; conventional road facilities have significant features, and the feature parts and the corresponding feature part areas can be identified through image recognition technology. Road facilities are set on the roadside as traffic signs. Conventional feature parts include conventional figures such as circles, triangles, squares or trapezoids (of course, regular triangles, squares or isosceles trapezoids are better). The area of ​​the feature parts can be measured before layout, or the manufacturer can give the corresponding target area after production; the recognition of the feature parts of other road facilities cannot be easily converted into conventional figures. Then, the edge points of the identified feature parts can be used as feature points to virtually construct new conventional figures (such as triangles, squares or trapezoids), and the area of ​​the conventional figures after virtual construction can be given during layout; and the virtual area data is stored and marked in the server. That is, the user can identify the road facilities that do not have conventional figures through the feature parts, and use the virtually constructed area comparison to calculate the distance. When performing feature recognition, there will be a certain angle between the vehicle and the graphic, so the graphic may be distorted to a certain extent. The imaging position of the graphic in the camera unit can be used to know the shooting angle of the graphic and the vehicle; the graphic can be calibrated and compensated for distortion through the shooting angle; since this technology is relatively mature, this embodiment will not be described in detail here.

[0025] Generally speaking, only one road facility with coordinates is set within a distance range, so as to avoid confusion caused by too much data being obtained by the vehicle. Of course, in order to make up for the possible missed situations, multiple road facilities with different features can be set within a distance range. Of course, the cost will be higher, but the positioning information provided to the vehicle will be more, thereby reducing the error caused by the subsequent inertial navigation device.

[0026] S2: The first vehicle obtains a characteristic portion of a road facility through a first camera unit, and obtains an area of ​​the characteristic portion and a first coordinate corresponding to the road facility according to the characteristic portion.

[0027] As a special case, a characteristic part of a road facility corresponds to only one characteristic part area and a first coordinate. In this case, only one set of data of the road facility needs to be stored in the server.

[0028] In the case of large-scale use, it is inevitable that many road facilities use the same characteristic part, but the corresponding first coordinates are different. At this time, there are two ways to obtain the first coordinate: the first is to upload the current rough coordinates of the vehicle, obtain the characteristic part of the road facilities within the preset range forward of the rough coordinates, and compare the data pre-stored in the server. The characteristic part is extracted by the vehicle through the camera unit to accurately know which road facility the currently photographed road facility is stored in the server, and then determine the characteristic part area and first coordinate of the corresponding road facility. This method is suitable for the situation where there are many road facilities for positioning, which can greatly reduce the data burden of the server. The second is to obtain all road facilities corresponding to the characteristic parts obtained by all first camera units, and obtain the corresponding characteristic part area and first coordinates according to the feature; then judge whether the first coordinate is within the preset range of the current rough coordinates through the current rough coordinates uploaded by the vehicle, and if so, filter out the road facilities photographed by the current first camera unit; and obtain the corresponding characteristic part area and first coordinates. This method is generally through cloud computing and then transmit the results to the vehicle.

[0029] After the first vehicle is positioned by the positioning device and obtains its rough coordinates, it is uploaded to the server. In the server (navigation map), the characteristic part of the road facilities within the forward range of the first vehicle's forward direction (i.e., the shooting range of the camera unit installed in front of the vehicle, of course, if the camera unit is installed on the left, right or rear of the vehicle), the corresponding characteristic part area and the corresponding coordinates of the road facilities can be obtained. After the first vehicle identifies the corresponding characteristic part of the road facilities through the first camera unit, the corresponding characteristic part area of ​​the road facilities and the corresponding coordinates of the road facilities can be obtained in the server. The area of ​​the characteristic part of the road facilities can be the center of gravity of the characteristic part of the road facilities (if it is a circle, the center of the circle is used) as the point collected by the camera unit. As a daily implementation example, a traffic light appears in front of the vehicle road, and one of the traffic lights is marked with a high-precision positioning coordinate (i.e., the road facilities are traffic lights, the characteristic part is the circular area of ​​the light, and the characteristic part area is the circular area of ​​the light). The camera unit mentioned in this embodiment can use a high-definition camera device or generally refer to all equipment that can be used for shooting. If it is a traffic sign, the regular graphics in the sign and the corresponding numbers can be used as the basis for identifying different road facilities. Some street lights with regular shapes can also be used as the basis for coordinate identification. In short, as long as the traffic facilities have relatively obvious features and are easy to identify, they can be used as coordinate calibration objects in the embodiments of the present invention.

[0030] S3: Positioning the first vehicle according to the first imaging position, the first imaging characteristic part area and the first coordinates of the road facility in the first camera unit.

[0031] After the first vehicle captures the characteristic part area through the first camera unit, the first camera unit has the first imaging position and the first imaging characteristic part area of ​​the characteristic part of the road administration facility. Generally speaking, the first imaging of the characteristic part of the road administration facility obtained by the camera unit will be in the currently captured picture. The closer the vehicle is to the road administration facility, the larger its area in the camera unit will be. By using the change in the area of ​​the characteristic part of the road administration facility captured by the first vehicle during driving as an empirical value, the first distance between the first vehicle and the road administration facility can be calculated; through such empirical values, a reference list can be listed, so that the characteristic part area and the first imaging characteristic part area have a reference relationship. Specifically, the area change and the first imaging position of the first vehicle when it is 1-200 meters away from the road administration facility can be captured 1000 times (the more tests, the more accurate it is) on the test section, and the current area can be recorded. When the same area is extracted and the first imaging position is the same, the distance between the first vehicle and the road administration facility can be known. Through the angles and shooting areas formed by the vehicle in different lanes, a series of corresponding values ​​can be obtained by repeated recording. The server records these values ​​to accurately locate the first vehicle through the positioning of the road administration facility. In addition, the area of ​​the first imaging feature part refers to the area of ​​the first imaging of the feature part of the road facility in the camera unit, and its change can also be calculated by defining a related algorithm for area change. The prior art has relevant records, and specific references can be made to some existing books on visual SLAM and PnP related technologies. This embodiment will not be described in detail here.

[0032] According to the position of the characteristic part of the road facility in the photographed picture, the first imaging position of the characteristic part of the road facility is formed, and the first angle formed by the road facility and the first vehicle can be obtained. The step of positioning the first vehicle includes: obtaining the first distance between the vehicle and the road facility according to the ratio of the area of ​​the characteristic part of the road facility and the area of ​​the first imaging characteristic part; obtaining the first angle between the vehicle and the road facility according to the imaging position of the characteristic part of the road facility in the first camera unit; and calculating and generating the first vehicle position coordinates using the first distance, the first angle and the first coordinates of the road facility.

[0033] like Figure 3As shown, the angle formed by the plane extension line of the first camera unit and the straight line connecting the first camera unit to the road facilities is defined as the first angle. Then the first vehicle coordinates are calculated and generated using the first distance, the first angle and the first coordinate. Since the first coordinate is a high-precision positioning value, the distance between the road facilities and the first vehicle is an empirical or calculated conversion value, both have high data reliability. Based on the first distance, the first angle and the first coordinate, a simple coordinate system can be established, and the corresponding high-precision positioning coordinates of the first vehicle can be calculated by performing coordinate conversion. In addition, with respect to the first distance, the first angle is defined and calculated in relatively many ways according to the area and the corresponding imaging area ratio (both visual SLAM and PnP technologies have relevant records), and there may be different definitions and different calculation methods, but it should be understood that the methods for calculating the corresponding numerical values ​​and coordinate relationships have been replaced or modified on this basis, and this embodiment will not be described in detail here.

[0034] During the driving process of the vehicle, through the ordinary navigation map and the first camera unit, it can be roughly identified as a road facility at a relatively far distance, but the characteristic part of the road facility is too small, and a small point is displayed on the graph. At this time, the characteristic part of the road facility can be regarded as a point, and the second camera unit can be set on the first vehicle; the first camera unit and the second camera unit are set at a distance. The first angle of the angle between the first camera unit and the characteristic part of the road facility is obtained, and the second angle of the angle between the second camera unit and the characteristic part of the road facility is obtained. According to the distance, the first angle and the second angle, the cosine theorem related method is used (some simple angle conversion is required, which is not described here), the relative position information of the first vehicle and the characteristic part of the road facility can be calculated; and the coordinates of the vehicle are calculated according to the first coordinates and the relative position information. Of course, no matter what the size of the characteristic part is, the characteristic points can be extracted or the characteristic part can be regarded as a point, and the double angle plus line segment method can be used to continue to calculate. The final calculation result is the same as that of only the first camera unit. Here, it is one of the multiple methods for positioning. In actual vehicle driving, in order to ensure accurate positioning throughout the entire process, using a binocular camera for positioning will have a better effect, but the hardware cost will be much higher. With the cooperation of the road facilities feature setting, the monocular camera has a higher cost-effectiveness and the positioning effect is not bad.

[0035] Since the positioning error of the inertial / wheel speedometer fusion positioning will accumulate as the mileage of the first vehicle increases, the positioning result will gradually deviate from the real position of the vehicle. At this time, the data of the inertial / wheel speedometer fusion positioning needs to be regularly corrected. In order to further provide more accurate correction data for the vehicle's IMU data. The first vehicle obtains the first imaging position of the QR code in the first camera unit and the first imaging characteristic partial area of ​​the road facilities at the first moment; obtains the second imaging position of the QR code in the first camera unit and the second imaging characteristic partial area of ​​the road facilities at the second moment; calculates the offset angle of the car according to the first imaging position and the second imaging position, and calculates the offset distance of the car according to the first imaging characteristic partial area and the second imaging characteristic partial area; calculates the angular acceleration and linear acceleration of the first vehicle according to the time difference between the first moment and the second moment; and corrects the IMU data of the first vehicle. In the field of fusion positioning technology, technicians can use existing fusion positioning technologies (such as particle filtering, Kalman filtering technology, etc.) to fuse these three types of position information (inertial navigation, satellite and vision), collect data from multiple vehicles, conduct big data statistics and corrections (that is, obtain an average value that is relatively close to most vehicles), and finally obtain the corrected position information and output the positioning result.

[0036] As another implementation example, when the vehicle obtains the characteristic part through the first camera unit, the characteristic line segment and characteristic line segment length information of the road facilities are obtained through the server. After the camera unit captures and identifies the characteristic part, its characteristic points can be extracted. The characteristic points here can be the corner points of the road facilities or the midpoints of a certain line segment. Two of the characteristic points are selected and connected to form a characteristic connection line segment to obtain the corresponding characteristic line segment. After the first camera unit captures the characteristic line segment of the road facilities, the first imaging characteristic line segment and the first imaging characteristic line segment length are formed in the camera unit (the more regular the graphics are, the easier it is to identify the defined extracted characteristic line segments; for example, for triangles, squares, and trapezoids, the imaging position and imaging length of one side can be directly known; for irregular shapes, corner points and connecting lines can be identified, and the corresponding data can be obtained through actual measurement, and the graphic data can be obtained through graphic proportions). Then, the first distance between the vehicle and the QR code is obtained based on the length of the characteristic line segment and the length of the first imaging characteristic line segment; the first angle between the vehicle and the road facilities is obtained based on the imaging position of the characteristic line segment in the first camera unit (i.e., the first imaging characteristic line segment); finally, the first distance, the first angle, and the first coordinates of the QR code are used to calculate and generate the vehicle position coordinates. Here, the calculation can be performed based on the relevant knowledge of the camera imaging principle, or the empirical value method mentioned in the above example can be used for multiple recordings, which will not be described in detail here.

[0037] In one embodiment, Figure 4As shown, the present disclosure proposes a vehicle fusion positioning based on road maintenance facility identification, including: a vehicle camera module, a server and a positioning module, each module is network connected; the vehicle camera module includes at least a first camera unit, which is installed on the vehicle; it is used to obtain the characteristic part of the road maintenance facilities set beside the road through the first camera unit; the road maintenance facilities include the characteristic part, the area of ​​the characteristic part and the first coordinate; the server is used to store the characteristic part, the area of ​​the characteristic part and the first coordinate information of the road maintenance facilities; when the vehicle camera module obtains the characteristic part of the road maintenance facility, the characteristic part area and the first coordinate corresponding to the road maintenance facility are obtained according to the characteristic part.

[0038] The process of obtaining the characteristic part area and the first coordinate of the road maintenance facility based on the characteristic part specifically includes: uploading the current rough coordinates of the vehicle, obtaining the characteristic part of the road maintenance facility within a preset range forward of the rough coordinates, and the characteristic part is the same as the characteristic part of the road maintenance facility obtained by the vehicle through the camera unit; in this way, the characteristic part can be used to identify which road maintenance facility stored in the server is the currently photographed road maintenance facility, and then determine the characteristic part area and the first coordinate of the corresponding road maintenance facility.

[0039] After the first vehicle captures the characteristic area through the first camera unit, the first imaging position and the first imaging characteristic area of ​​the characteristic part of the road facilities are in the first camera unit. Generally speaking, the first imaging of the characteristic part of the road facilities obtained by the camera unit will be in the currently captured picture. The closer the vehicle is to the road facilities, the larger its area in the camera unit will be. By taking the change in the area of ​​the characteristic part of the road facilities captured by the first vehicle during driving as an empirical value, the first distance between the first vehicle and the road facilities can be calculated; specifically, the area change and the first imaging position of the first vehicle when it is 1-200 meters away from the road facilities can be captured 1000 times (the more tests, the more accurate it is) on the test section, and the current area can be recorded. When the same area is extracted and the first imaging position is the same, the distance between the first vehicle and the road facilities can be known. Through the angles and shooting areas formed by the vehicles in different lanes, a series of corresponding values ​​can be obtained by repeated recording. By recording these big data on the server, the first vehicle can be accurately positioned through the positioning of the road facilities. In addition, the area change of the first image of the characteristic part of the road facilities in the camera unit can also be calculated by defining a related algorithm for area change. The prior art has relevant records. For details, please refer to some existing books on visual SLAM and PnP related technologies. This embodiment will not be described in detail here.

[0040] According to the position of the characteristic part of the road facility in the photographed picture, the first imaging position of the characteristic part of the road facility is formed, and the first angle formed by the road facility and the first vehicle can be obtained. The step of positioning the first vehicle includes: obtaining the first distance between the vehicle and the road facility according to the ratio of the area of ​​the characteristic part of the road facility and the area of ​​the first imaging characteristic part; obtaining the first angle between the vehicle and the road facility according to the imaging position of the characteristic part of the road facility in the first camera unit; and calculating and generating the first vehicle position coordinates using the first distance, the first angle and the first coordinates of the road facility.

[0041] It can be understood by those skilled in the art that all or part of the steps in the embodiment can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable medium, which can include a flash drive, a mobile hard disk, a read-only memory, a random access memory, a disk or an optical disk, and other media that can store program codes. In one embodiment, the present disclosure proposes a computer-readable medium, in which a computer program is stored, and the computer program is loaded and executed by a processing module to implement a vehicle fusion positioning method based on a two-dimensional code.

[0042] Within the scope of the knowledge and ability level of those skilled in the art, the various embodiments or technical features mentioned herein may be combined with each other as other optional embodiments without conflict. These limited number of optional embodiments, which are not listed one by one and are formed by combining a limited number of technical features, still fall within the technical scope disclosed in the present disclosure and can be understood or inferred by those skilled in the art in combination with the drawings and the above text.

[0043] In addition, the description of most embodiments is based on different focuses. For details not described in detail, please refer to the content of the prior art or other relevant descriptions in this article for understanding.

[0044] It is emphasized again that the embodiments listed above are typical and preferred embodiments of the present disclosure, which are only used to explain and interpret the technical solutions of the present disclosure in detail to facilitate the understanding of the readers, and are not intended to limit the protection scope or application of the present disclosure. Any technical solutions obtained by any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A vehicle fusion positioning method based on road facilities recognition, Features: A road facility is set up around the road, the road facility includes a characteristic part, an area of ​​the characteristic part and a first coordinate; a vehicle obtains the characteristic part through a first camera unit, and obtains the area of ​​the characteristic part and the first coordinate according to the characteristic part; the vehicle is positioned according to a first imaging position of the characteristic part in the first camera unit, the area of ​​the first imaging characteristic part and the first coordinate; The step of positioning the vehicle includes: obtaining a first distance between the vehicle and the road facility according to the characteristic part area and the first imaging characteristic part area; The first angle between the vehicle and the road maintenance facility is acquired according to the first imaging position of the characteristic part in the first camera unit; and the vehicle position coordinates are calculated using the first distance, the first angle and the first coordinates of the road maintenance facility.

2. The vehicle fusion positioning method according to claim 1, wherein the vehicle obtains the area of ​​the characteristic part and the first coordinate process include: Upload the current rough coordinates of the vehicle, obtain the characteristic part of the road facilities around the rough coordinates of the first coordinate, and obtain the area of ​​the characteristic part and the first coordinate from the server according to the characteristic part.

3. The vehicle fusion positioning method according to claim 1, Features: At a first moment, a first imaging position and an area of ​​a first imaging characteristic portion of the characteristic portion in the first camera unit are obtained; at a second moment, a second imaging position and an area of ​​a second imaging characteristic portion of the characteristic portion in the first camera unit are obtained; a displacement angle of the vehicle is calculated according to the first imaging position and the second imaging position, and a displacement distance of the vehicle is calculated according to the area of ​​the first imaging characteristic portion and the area of ​​the second imaging characteristic portion; and a time difference between the first moment and the second moment is calculated; The angular acceleration and linear acceleration of the vehicle are calculated and the IMU data of the vehicle is corrected.

4. The vehicle fusion positioning method according to claim 1, Features: The vehicle is provided with a second camera unit; a distance is set between the first camera unit and the second camera unit; a first angle formed by the first camera unit and the road maintenance facility is obtained, and a second angle formed by the second camera unit and the road maintenance facility is obtained, and the relative position information of the vehicle and the road maintenance facility is calculated according to the distance, the first angle and the second angle, and the vehicle position coordinates are calculated by the first coordinates and the relative position information.

5. The vehicle fusion positioning method according to any one of claims 1 to 4, Features: The road facilities include characteristic line segments and characteristic line segment lengths; when the vehicle obtains the characteristic part through the first camera unit and obtains the characteristic line segment and characteristic line segment length through the server; the characteristic line segment is used to replace the characteristic part and the characteristic line segment length is used to replace the area of ​​the characteristic part.

6. A vehicle fusion positioning device based on road administration facility identification, Features: include: Vehicle camera module, server and positioning module, and data connection of each module; The vehicle camera module comprises at least a first camera unit, which is installed on the vehicle; Used to obtain the characteristic part of the road facilities set beside the road through the first camera unit; the road facilities include the characteristic part, the area of ​​the characteristic part and the first coordinate; the server is used to store the characteristic part, the area of ​​the characteristic part and the first coordinate information of the road facilities; when the vehicle camera module obtains the characteristic part of the road facilities, the characteristic part area and the first coordinate corresponding to the road facilities are obtained according to the characteristic part; the positioning module is used to locate the vehicle according to the first imaging position of the road facilities in the first camera unit, the first imaging characteristic part area and the first coordinate; The step of positioning the vehicle includes: obtaining a first distance between the vehicle and the road facility according to a reference relationship between the characteristic part area and the first imaging characteristic part area; The first angle between the vehicle and the road maintenance facility is acquired according to the imaging position of the characteristic part in the first camera unit; and the vehicle position coordinates are calculated using the first distance, the first angle and the first coordinates of the road maintenance facility.

7. The vehicle fusion positioning device according to claim 6, Features: When the area of ​​the first imaging feature part is smaller than a preset threshold, the vehicle is provided with a second camera unit; a distance is set between the first camera unit and the second camera unit; a first angle between the first camera unit and the road maintenance facility is obtained, and a second angle between the second camera unit and the road maintenance facility is obtained, and the relative position information between the vehicle and the road maintenance facility is calculated based on the distance, the first angle and the second angle, and the vehicle position coordinates are calculated using the first coordinates and the relative position information.

8. Computer readable media, Features: The computer-readable medium stores a computer program, which is loaded and executed by the processing module to implement the vehicle fusion positioning method as described in any one of claims 1 to 5.

Images