A vehicle positioning method, positioning system and vehicle based on ground marking grid

Abstract

The present invention provides a vehicle positioning method, a positioning system and a vehicle based on ground identification grids, and relates to the technical field of vehicle information management. The vehicle positioning method based on ground identification grids of the present invention comprises: obtaining real-time image information of a vehicle in an area with identification grids, wherein the identification grids are identification grids pre-drawn on the road surface, and the identification grids have meta-coordinates; correcting the real-time image information to obtain first corrected image information; obtaining the relative position information of the vehicle relative to the meta-coordinates of the identification grid at the corresponding position according to the first corrected image information; obtaining the actual position information of the vehicle according to the relative position information and the actual position information of the meta-coordinates pre-stored. The positioning method of the present invention is simple in positioning and accurate in positioning. The positioning system has a simple structure and high positioning accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle information management, and in particular to a vehicle positioning method, a positioning system and a vehicle based on a ground identification grid. Background Art

[0002] At present, there are mainly the following types of vehicle positioning technologies: 1. Vehicle positioning technology based on satellite and ground base station positioning technology, such as using satellite systems such as Beidou, GPS, GLONASS, Galileo, and Qianxun Location. 2. Vehicle positioning technology based on near-field wireless communication, using Bluetooth beacons, Wi-Fi hotspots, ZigBee, UWB and other systems for positioning; 3. Vehicle inertial navigation-assisted vehicle positioning technology, using the gyroscope and speed sensor installed on the vehicle to estimate the vehicle's position after a period of driving based on the original vehicle position; 4. Vehicle positioning technology based on mobile operator network base stations; 5. Vehicle positioning technology based on high-precision sensors installed on the vehicle and video recognition landmarks; 6. Vehicle positioning technology that integrates multiple positioning technologies.

[0003] For these existing positioning technologies, such as vehicle positioning technology based on satellite and ground base station positioning technology, the positioning accuracy is generally 5-10m, and the traditional positioning delay is 30-50s. Even the new Qianxun positioning system based on multi-satellite and multi-base station fusion positioning has a vehicle-level positioning accuracy of about 5cm and a positioning delay of 3-5s (the positioning delay is still too long for high-speed vehicle control scenarios). In addition, the Qianxun positioning system needs to build more ground base stations, which requires huge initial investment, relatively high cost of use, and cannot be positioned in obstructed areas and basements. For vehicle positioning technology based on near-field wireless communication, this technology requires special positioning landmarks to be built in advance, and multiple landmarks need to be coordinated to improve positioning accuracy. Professional high-definition maps with landmark location data are required. For vehicle inertial navigation-assisted vehicle positioning technology, the gyroscope and speed sensor installed on the vehicle are used. The position of the vehicle after a period of driving is estimated based on the original vehicle position; the vehicle inertial navigation-assisted vehicle positioning technology requires an initial positioning position, and driving direction deviation and frequent lane changes will bring more errors, and the positioning accuracy will become lower and lower as time accumulates. For vehicle positioning technology based on mobile operator network base stations; even in the future 5G stage, the positioning accuracy requirement of this technology is 1m. For vehicle positioning technology based on high-precision sensors installed on the vehicle and video recognition landmarks, sensors with high position detection accuracy such as laser radar, millimeter wave radar, binocular camera, etc. are relatively expensive. A high-precision map based on the corresponding sensor information at the landmark is required, and the map needs to be updated as real-time as possible. For vehicle positioning technology based on the fusion of multiple positioning technologies, the fusion requires complex control logic and higher costs. Summary of the invention

[0004] An object of the present invention is to provide a vehicle positioning method based on ground marker grids to solve the problems of low positioning accuracy and long positioning delay time of existing positioning technologies.

[0005] Another object of the present invention is to solve the problem of high technical requirements and high costs of existing positioning technologies.

[0006] Another object of the present invention is to provide a vehicle positioning system based on a marker grid to solve the problems of low positioning accuracy, long positioning delay time and high cost of existing vehicles.

[0007] Another object of the present invention is to provide a vehicle comprising the above-mentioned vehicle positioning system based on the identification grid.

[0008] In particular, the present invention provides a vehicle positioning method based on a ground identification grid, comprising:

[0009] Acquire real-time image information of the vehicle in an area with a marking grid, wherein the marking grid is a marking grid pre-drawn on the road surface, and the marking grid has a meta coordinate;

[0010] Correcting the real-time image information to obtain first corrected image information;

[0011] Obtaining relative position information of the vehicle relative to the meta-coordinates of the identification grid at the corresponding position according to the first corrected image information;

[0012] The actual position information of the vehicle is obtained according to the relative position information and the actual position information of the meta-coordinates stored in advance.

[0013] Optionally, the process of correcting the real-time image information includes:

[0014] Based on the real-time image information, obtaining an angle between a projection line of the longitudinal axis of the vehicle on the road surface and a direction of the road surface;

[0015] Correcting the real-time image information according to the angle to obtain second corrected image information;

[0016] Obtaining first information of the vehicle camera according to the second corrected image information, wherein the first information is the relative position information of the vehicle camera relative to the element coordinates of the identification grid in the second corrected image information and the angle information with the ground;

[0017] Compare the first information obtained from the vehicle camera with the second information, and obtain a correction value in combination with the distortion parameter of the vehicle camera, wherein the second information is the relative position information of the vehicle camera relative to the meta-coordinates of the reference identification grid and the angle information with the ground which are pre-calibrated;

[0018] The second corrected image information is corrected according to the correction value to obtain the first corrected image information.

[0019] Optionally, the identification grid includes a plurality of identification grids drawn on road surfaces in different regions; each group of the identification grids corresponds to a meta-coordinate.

[0020] Optionally, the identification grid is a cross grid formed by first identification lines and second identification lines, wherein the first identification lines are multiple identification lines with a preset width and a preset spacing drawn in a direction parallel to the road edge, and the second identification lines are multiple identification lines with the preset width and the preset spacing drawn in a direction perpendicular to the road edge.

[0021] Optionally, the meta-coordinates include:

[0022] On a straight road, the meta-coordinate is selected as any intersection in the road identification grid, and the identification grid at the meta-coordinate position is marked with a slash; or

[0023] At a turn or intersection, the meta-coordinate is the intersection of two marking lines on the road in two directions, which are respectively parallel to the road edges of the corresponding road.

[0024] In particular, the present invention also provides a vehicle positioning system based on a ground identification grid, comprising:

[0025] A vehicle camera, used to obtain real-time image information of a vehicle in an area having a marking grid, wherein the marking grid is a marking grid pre-drawn on a road surface, and the marking grid has a meta coordinate; and

[0026] A vehicle controller, the vehicle controller comprising:

[0027] A corrected image acquisition module, used for correcting the real-time image information to obtain first corrected image information;

[0028] A relative position information acquisition module, used for obtaining the relative position information of the vehicle relative to the meta-coordinates of the identification grid at the corresponding position according to the first corrected image information; and

[0029] The actual position information acquisition module is used to obtain the actual position information of the vehicle according to the relative position information and the actual position information of the meta-coordinates stored in advance.

[0030] Optionally, the corrected image acquisition module includes:

[0031] An angle acquisition unit, used to obtain an angle between a projection line of the longitudinal axis of the vehicle on the road surface and the road surface direction based on the real-time image information;

[0032] A second corrected image acquisition unit, configured to correct the real-time image information according to the included angle to obtain second corrected image information;

[0033] A first information acquisition unit, configured to obtain first information of the vehicle camera according to the second corrected image information, wherein the first information is relative position information of the vehicle camera relative to the element coordinates of the identification grid in the second corrected image information and angle information with the ground;

[0034] A correction value acquisition unit, used for comparing the first information obtained from the vehicle camera with the second information, and obtaining a correction value in combination with the distortion parameter of the vehicle camera, wherein the second information is the relative position information of the vehicle camera relative to the meta-coordinates of the reference identification grid and the angle information with the ground which are pre-calibrated;

[0035] The first corrected image acquisition unit is used to correct the second corrected image information according to the correction value to obtain the first corrected image information.

[0036] Optionally, the identification grid includes a plurality of identification grids drawn on road surfaces in different regions; each group of the identification grids corresponds to a meta-coordinate;

[0037] Optionally, the identification grid is a cross grid formed by first identification lines and second identification lines, wherein the first identification lines are multiple identification lines with a preset width and a preset spacing drawn in a direction parallel to the road edge, and the second identification lines are multiple identification lines with the preset width and the preset spacing drawn in a direction perpendicular to the road edge.

[0038] Optionally, the meta-coordinates include:

[0039] On a straight road, the meta-coordinate is selected as any intersection in the road identification grid, and the identification grid at the meta-coordinate position is marked with a slash; or

[0040] At a turn or intersection, the meta-coordinate is the intersection of two marking lines on the road in two directions, which are respectively parallel to the road edges of the corresponding road.

[0041] In particular, the present invention also provides a vehicle, comprising the above-mentioned vehicle positioning system based on ground marking grids.

[0042] The positioning method of the present invention requires setting up identification grids at some locations that need to be positioned. When a vehicle passes through the identification grid area, a real-time image of the vehicle in the identification grid area is obtained through a vehicle camera on the vehicle. By correcting the image, more accurate relative position information of the vehicle relative to the meta-coordinates of the identification grid can be obtained, thereby obtaining accurate actual position information of the vehicle. The positioning method of the present invention is simple and accurate in positioning.

[0043] Furthermore, the positioning method of the present invention requires two corrections during the entire positioning process to correct the position movement or angle change that may occur in the vehicle camera during the actual vehicle driving process, and to correct the distortion that may occur in the camera itself, thereby ensuring the accuracy of the positioning of the entire process and improving the positioning accuracy of the vehicle.

[0044] The vehicle positioning system based on the ground identification grid of the present invention can accurately locate the vehicle through the vehicle camera and the vehicle controller. The positioning system is simple in composition and relatively low in cost. In addition, the real-time image of the vehicle in the identification grid area is first obtained through the vehicle camera, and by correcting the image, more accurate relative position information of the vehicle relative to the meta-coordinates of the identification grid can be obtained, thereby obtaining accurate actual position information of the vehicle, and the positioning is simple and accurate.

[0045] The vehicle of the present invention comprises a vehicle positioning system based on a ground marking grid. Specifically, the vehicle system having the vehicle positioning system based on a ground marking grid has a simpler system composition, a relatively low cost, and a fast and accurate vehicle positioning speed.

[0046] Based on the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will become more aware of the above and other objects, advantages and features of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Hereinafter, some specific embodiments of the present invention will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:

[0048] Figure 1 is a schematic flow chart of a vehicle positioning method based on a ground identification grid according to an embodiment of the present invention;

[0049] Figure 2 is a schematic flow chart of a process of correcting real-time image information in a vehicle positioning method based on a ground identification grid according to another embodiment of the present invention;

[0050] Figure 3 is a schematic diagram of a marking grid on a straight road according to one embodiment of the present invention;

[0051] Figure 4 is a schematic diagram of a marking grid at a right-angle turn or intersection according to one embodiment of the present invention;

[0052] Figure 5 is a schematic diagram of a marking grid at a non-right-angle turn or intersection according to another embodiment of the present invention;

[0053] Figure 6 is a schematic diagram of coordinate selection of a target vehicle on a straight road relative to a meta-coordinate according to an embodiment of the present invention;

[0054] Figure 7 is a schematic diagram of coordinate selection of relative meta-coordinates of a target vehicle on a turning or crossing road according to an embodiment of the present invention;

[0055] Figure 8 is a schematic block diagram of a vehicle positioning system based on a ground identification grid according to an embodiment of the present invention;

[0056] Fig. 9 is a schematic block diagram of a vehicle positioning system based on a ground identification grid according to another embodiment of the present invention;

[0057] Fig.10 is a schematic block diagram of a vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] Figure 1 FIG. 1 is a schematic flow chart of a vehicle positioning method based on a ground identification grid according to an embodiment of the present invention. Figure 1 As shown, the vehicle positioning method based on the ground identification grid of this embodiment may include:

[0059] S10 acquires real-time image information of the vehicle in an area with a marking grid, wherein the marking grid is a marking grid pre-drawn on the road surface, and the marking grid has a meta coordinate;

[0060] S20 corrects the real-time image information to obtain first corrected image information;

[0061] S30 obtaining relative position information of the vehicle relative to the element coordinates of the identification grid at the corresponding position according to the first corrected image information;

[0062] S40 obtains the actual position information of the vehicle according to the relative position information and the actual position information of the pre-stored meta-coordinates.

[0063] The positioning method of this embodiment requires setting up identification grids at some locations that need to be positioned. When a vehicle passes through the identification grid area, a real-time image of the vehicle in the identification grid area is obtained through a vehicle camera on the vehicle. By correcting the image, more accurate relative position information of the vehicle relative to the meta-coordinates of the identification grid can be obtained, thereby obtaining accurate actual position information of the vehicle. The positioning method of this embodiment is simple and accurate in positioning.

[0064] Figure 2 : is a schematic flow chart of a process of correcting real-time image information of a vehicle positioning method based on a ground identification grid according to another embodiment of the present invention. As another embodiment of the present invention, the process of correcting real-time image information in this embodiment includes:

[0065] Based on real-time image information, F10 obtains the angle between the projection line of the vehicle's longitudinal axis on the road surface and the road surface direction;

[0066] F20 corrects the real-time image information according to the angle to obtain second corrected image information;

[0067] F30 obtains first information of the vehicle camera according to the second corrected image information, wherein the first information is relative position information of the vehicle camera relative to the element coordinates of the identification grid in the second corrected image information and angle information with the ground;

[0068] F40 compares the first information obtained from the vehicle camera with the second information, and obtains a correction value in combination with the distortion parameter of the vehicle camera, wherein the second information is the relative position information of the pre-calibrated vehicle camera relative to the meta coordinates of the reference identification grid and the angle information with the ground;

[0069] F50 corrects the second corrected image information according to the correction value to obtain the first corrected image information.

[0070] Specifically, the real-time image correction process of this embodiment includes first correcting the real-time image information using the angle between the projection line of the longitudinal axis of the vehicle on the road surface and the direction of the road surface. The longitudinal axis of the vehicle can be considered as the driving direction of the vehicle. Then, a correction value is obtained based on the relative position information of the meta-coordinates of the calibrated vehicle camera relative to the standard identification grid and the angle information with the ground, and the corrected image information is re-corrected based on the correction value to obtain the corrected image information. Two corrections are required in the whole process to correct the position movement or angle change of the vehicle camera that may occur during the actual vehicle driving process, and at the same time, the distortion that may occur in the camera itself is corrected, so as to ensure the accuracy of positioning in the whole process and improve the positioning accuracy of the vehicle.

[0071] The vehicle positioning method of this embodiment is all based on the current road surface identification grid for recognition, and after image comparison and analysis, the recognition speed is fast and the accuracy is high. In addition, the vehicle driving direction and road direction angle analysis in the vehicle positioning method of this embodiment is also based on the current road surface identification grid for recognition, and the recognition speed is fast and the accuracy is high.

[0072] As a specific embodiment of the present invention, before implementing the method, it is necessary to draw a marking grid on the road surface corresponding to the position where the vehicle positioning is required. The marking grid includes a plurality of marking grids drawn on the road surface in different areas. Each group of marking grids is provided with a meta coordinate.

[0073] Figure 3 is a schematic diagram of a marking grid on a straight road according to one embodiment of the present invention; Figure 4 is a schematic diagram of a marking grid at a right-angle turn or intersection according to one embodiment of the present invention; Figure 5 is a schematic diagram of a marking grid at a non-right-angle turn or intersection according to another embodiment of the present invention. Specifically, Figure 3-5 As shown in the figure, the identification grid 20 is a cross grid formed by the first identification lines 21 and the second identification lines 22. The first identification lines 21 are multiple identification lines with a preset width and a preset spacing drawn in a direction parallel to the road edge 70. The second identification lines 22 are multiple identification lines with a preset width and a preset spacing drawn in a direction perpendicular to the road edge 70.

[0074] Specifically, in this embodiment, the preset width of the marking line is about 10 cm, and the preset distance is about 40 cm. In addition, in one embodiment, additional marking lines can be added at the edge of the marking grid 20 to ensure that the inner marking grid 20 is clear and complete, and the parallel spacing between the marking lines and the edge 70 of the road for safe driving is about 10 cm. In this embodiment, based on the image recognition of the square marking grid 20 with a side length of 50 cm, the positioning accuracy is high and can reach the 5 cm level. The marking grid 20 is used as an image recognition reference, the recognition speed is fast, and the positioning delay is within 50 ms.

[0075] Specifically, when the vehicle camera is calibrated, the specific position of the vehicle is located by the calibration site, and the road marking grid drawn around the vehicle on the site is used to determine the equivalent vehicle camera three-coordinate reference position and the equivalent angle between the vehicle camera and the ground. Among them, the single camera is preferentially arranged on the longitudinal plane of the whole vehicle, and the relevant position parameters are determined according to the camera image conversion law. For a multi-camera system, a system equivalent position and angle are determined according to the mechanical relationship of the installation position of each camera and the image stitching algorithm. Record the basic comparison road marking grid pixel benchmark. Expand the high-precision (such as 1cm square grid on the road surface) internal auxiliary grid for road surface controller comparison based on the road marking grid pixel benchmark.

[0076] As a specific embodiment of the present invention, Figure 3-5 As shown, the selection of the meta-coordinates may include two specific methods. First, the selection of the meta-coordinates 60 on the straight road is any intersection in the identification grid 20 of the road section, and the identification grid 20 at the position of the meta-coordinates 60 is marked with a slash 23, as shown in FIG. Figure 3 As shown. Specifically, in the actual operation process, the vehicle camera system is limited by technology, and when shooting the identification grid 20, a deformed area will appear at the edge position, and the deformation of the shooting picture will directly lead to inaccurate positioning. The selection of the specific meta coordinate 60 can be selected according to actual needs. After the meta coordinate 60 is selected, in order to facilitate the identification of the meta coordinate 60, a slash 23 is generally added to the grid at the meta coordinate 60 as an identifier, and the meta coordinate 60 is selected at the outer right-angle vertex of the right triangle.

[0077] As other embodiments, Figure 4 and Figure 5 As shown, the selection of meta-coordinates can also include that at a turn or intersection, the meta-coordinate 60 is the intersection of two marking lines on the road in two directions that are parallel to the corresponding road edge 70. And an oblique line 23 is added at the position of the meta-coordinate 60 to form an isosceles triangle as a mark, and the meta-coordinate 60 is the vertex position of the isosceles triangle. In the actual operation process, the method for determining the marking lines in each direction gives priority to drawing the marking lines of the first straight section with an extra height of 2m after the road passes the turning point, the special section of the sharp turn greater than or equal to 180° or the special continuous turning section, and one or more marking lines tangent to the road arc are added between the angles of the two directions, and the section is divided into two marking grids 20 that share a part of the marking grid 20. After the drawing is determined, the actual geographical location is measured and recorded by high-precision positioning means to obtain the actual location information of the meta-coordinate. The actual geographical location generally recorded includes longitude and latitude and the angle between the road direction and the longitude. When the meta-coordinate is a turn or an intersection, all road directions and longitude angles are recorded, the turn is regarded as two roads, and a label is defined for each road. In addition, at non-90° turns, a 1m extension line is left after the marking lines cross.

[0078] Figure 6 is a schematic diagram of coordinate selection of a target vehicle on a straight road relative to a meta-coordinate 60 according to an embodiment of the present invention; Figure 7It is a schematic diagram of the coordinate selection of the target vehicle relative to the meta-coordinate 60 on a turning or crossing road according to an embodiment of the present invention. In this embodiment, when analyzing the picture information, the target vehicle in the captured picture is first replaced with a rectangle whose side length is parallel to or perpendicular to the road surface identification grid 20 as the vehicle position, and the vehicle edge is tangent to the rectangle (that is, the vehicle edge is completely contained in the rectangle, and the edge has a tangent point). The coordinates of the vehicle rectangular position are expressed as, and the coordinate system is an orthogonal coordinate system with the meta-coordinate 60 as the origin along the direction of the road edge 70 as the X-axis direction and the direction perpendicular to the road edge 70 as the Y-axis direction. X1 is the X-axis coordinate of the near meta-coordinate of the vehicle edge, X2 is the X-axis coordinate of the far meta-coordinate of the vehicle edge, Y1 is the Y-axis coordinate of the near meta-coordinate of the vehicle edge, and Y2 is the Y-axis coordinate of the far meta-coordinate of the vehicle edge.

[0079] When the target vehicle is on a straight road, one set of coordinate information can be obtained, and when the target vehicle is in a turning transition area, the vehicle body is located in different groups of identification grids 20, and two sets of coordinate position information can be obtained. Figure 5 and Figure 6 shown. Figure 5 On a straight road, the vehicle has only one set of coordinates, namely (X1, X2, Y1, Y2). When the vehicle is at a turn or intersection, Figure 6 As shown, the vehicle may have at least two sets of coordinates. Taking two sets of coordinates as an example, we obtain (X1, X2, Y1, Y2) and (X1', X2', Y1', Y2') respectively.

[0080] Specifically, in this way, the relative position information of the vehicle relative to the meta-coordinates of the vehicle relative to the corresponding position of the identification grid in the first corrected image information can be obtained. When the relative position information of the vehicle is determined, the actual position information of the vehicle can be easily obtained according to the actual position information of each meta-coordinate stored in advance.

[0081] Figure 81 is a schematic block diagram of a vehicle positioning system 100 based on a ground grid according to an embodiment of the present invention. As a specific embodiment of the present invention, this embodiment also provides a vehicle positioning system 100 based on a ground grid, and the vehicle positioning system 100 based on a ground grid may include a vehicle camera 10 and a vehicle controller 30. The vehicle camera 10 is used to obtain real-time image information of a vehicle in an area with a grid, wherein the grid is a grid pre-drawn on a road surface, and the grid has meta coordinates. The vehicle controller 30 may include a corrected image acquisition module 31, a relative position information acquisition module 32, and an actual position information acquisition module 33. The corrected image acquisition module 31 is used to correct the real-time image information and obtain the first corrected image information. The relative position information acquisition module 32 is used to obtain the relative position information of the vehicle relative to the meta coordinates of the grid at the corresponding position according to the first corrected image information. The actual position information acquisition module 33 is used to obtain the actual position information of the vehicle according to the relative position information and the actual position information of the meta coordinates stored in advance.

[0082] The vehicle positioning system 100 based on the ground identification grid of this embodiment can accurately locate the vehicle through the vehicle camera 10 and the vehicle controller 30. The positioning system is simple in composition and relatively low in cost. In addition, the real-time image of the vehicle in the identification grid area is first obtained through the vehicle camera 10, and by correcting the image, more accurate relative position information of the vehicle relative to the meta-coordinates of the identification grid can be obtained, thereby obtaining accurate actual position information of the vehicle, and the positioning is simple and accurate.

[0083] Fig. 9It is a schematic block diagram of a vehicle positioning system based on a ground identification grid according to another embodiment of the present invention; as another specific embodiment of the present invention, the corrected image acquisition module 31 of the present invention may include an angle acquisition unit 311, a second corrected image acquisition unit 312, a first information acquisition unit 313, a correction value acquisition unit 314 and a first corrected image acquisition unit 315. Among them, the angle acquisition unit 311 is used to obtain the angle between the projection line of the longitudinal axis of the vehicle on the road surface and the road surface direction based on the real-time image information. The second corrected image acquisition unit 312 is used to obtain the second corrected image information according to the angle correction of the real-time image information. The first information acquisition unit 313 is used to obtain the first information of the vehicle camera 10 according to the second corrected image information, wherein the first information is the relative position information of the vehicle camera 10 relative to the element coordinates of the identification grid in the second corrected image information and the angle information with the ground. The correction value acquisition unit 314 is used to compare the first information obtained from the vehicle camera 10 with the second information, and obtain the correction value in combination with the distortion parameter of the vehicle camera 10, wherein the second information is the relative position information of the pre-calibrated meta-coordinates of the vehicle camera 10 relative to the reference identification grid and the angle information with the ground. The first corrected image acquisition unit 315 is used to correct the second corrected image information according to the correction value to obtain the first corrected image information.

[0084] The real-time image correction module of this embodiment includes first correcting the real-time image information using the angle between the projection line of the longitudinal axis of the vehicle on the road surface and the road surface direction, and then obtaining the correction value based on the relative position information of the element coordinates of the calibrated vehicle camera 10 relative to the standard identification grid and the angle information with the ground, and then re-correcting the corrected image information based on the correction value. Two corrections are required in the whole process, to correct the position movement or angle change of the vehicle camera 10 that may occur during the actual vehicle driving process, and at the same time to correct the distortion that will occur in the vehicle camera itself, so as to ensure the accuracy of the positioning of the whole process and improve the positioning accuracy of the vehicle.

[0085] In addition, all vehicles in this embodiment are identified based on the current road sign grid, and after image comparison and analysis, the identification speed is fast and the accuracy is high. In addition, the vehicle direction and road direction angle analysis of the vehicle positioning system 100 based on the ground sign grid in this embodiment is also based on the current road sign grid identification, and the identification speed is fast and the accuracy is high.

[0086] In this embodiment, the vehicle camera 10 is one or a group of cameras installed on the vehicle, and the vehicle controller 30 is a controller that comes with the vehicle. When the vehicle camera 10 is calibrated, the specific position of the vehicle is located by the calibration site, and the road marking grid drawn around the vehicle on the site is used to determine the equivalent three-coordinate reference position of the vehicle camera 10 and the equivalent angle between the vehicle camera 10 and the ground. Among them, a single camera is preferably arranged on the longitudinal plane of the whole vehicle, and the relevant position parameters are determined according to the camera image conversion law. When there are multiple cameras, a system equivalent position and angle are determined based on the mechanical relationship of the installation position of each camera and the image stitching algorithm. Record the basic comparison road marking grid pixel benchmark. Expand the high-precision (such as a 1cm square grid on the road) internal auxiliary grid for road controller comparison based on the road marking grid pixel benchmark.

[0087] As a specific embodiment of the present invention, the vehicle positioning system 100 based on ground grids in this embodiment needs to draw grids on the road surface corresponding to the position where the vehicle positioning is required before the vehicle camera 10 takes pictures. The grids include multiple grids drawn on the road surface in different areas. Each group represents a setting with a corresponding meta coordinate.

[0088] Figure 3 is a schematic diagram of a marking grid on a straight road according to one embodiment of the present invention; Figure 4 is a schematic diagram of a marking grid at a right-angle turn or intersection according to one embodiment of the present invention; Figure 5 is a schematic diagram of a marking grid at a non-right-angle turn or intersection according to another embodiment of the present invention. Specifically, Figure 3-5 As shown in the figure, the identification grid 20 is a cross grid formed by the first identification lines 21 and the second identification lines 22. The first identification lines 21 are multiple identification lines with a preset width and a preset spacing drawn in a direction parallel to the road edge 70. The second identification lines 22 are multiple identification lines with a preset width and a preset spacing drawn in a direction perpendicular to the road edge 70.

[0089] Specifically, in this embodiment, the preset width of the marking line is about 10 cm, and the preset distance is about 40 cm. In addition, in one embodiment, additional marking lines can be added at the edge of the marking grid 20 to ensure that the inner marking grid 20 is clear and complete, and the parallel spacing between the marking lines and the edge 70 of the road for safe driving is about 10 cm. In this embodiment, based on the image recognition of the square marking grid 20 with a side length of 50 cm, the positioning accuracy is high and can reach the 5 cm level. The marking grid 20 is used as an image recognition reference, the recognition speed is fast, and the positioning delay is within 50 ms.

[0090] As a specific embodiment of the present invention, Figure 3-5As shown, the selection of the meta-coordinates may include two specific methods. First, the selection of the meta-coordinates 60 on the straight road is any intersection in the identification grid 20 of the road section, and the identification grid 20 at the position of the meta-coordinates 60 is marked with a slash 23, as shown in FIG. Figure 3 As shown. Specifically, in the actual operation process, the vehicle camera 10 is limited by technology, and when shooting the identification grid 20, a deformed area will appear at the edge position, and the deformation of the shooting picture will directly lead to inaccurate positioning. The selection of the specific meta coordinate 60 can be selected according to actual needs. After the meta coordinate 60 is selected, in order to facilitate the identification of the meta coordinate 60, a slash 23 is generally added to the grid at the meta coordinate 60 as an identifier, and the meta coordinate 60 is selected at the outer right-angle vertex of the right triangle.

[0091] As other embodiments, Figure 4 and Figure 5 As shown, the selection of meta-coordinates can also include that at a turn or intersection, the meta-coordinate 60 is the intersection of two marking lines parallel to the corresponding road edge 70 on the road in two directions. And an oblique line 23 is added at the position of the meta-coordinate 60 to form an isosceles triangle as a mark, and the meta-coordinate 60 is the vertex position of the isosceles triangle. In the actual operation process, the method for determining the marking lines in each direction gives priority to drawing the marking lines of the first straight section with an extra height of 2m after the road passes the turning point, the special section of the sharp turn greater than or equal to 180° or the special continuous turning section, and one or more marking lines tangent to the road arc are added between the angles of the two directions, and the section is divided into two marking grids 20 that share a part of the marking grid 20. After the drawing is determined, the actual geographical location is measured and recorded by high-precision positioning means to obtain the actual location information of the meta-coordinate. The actual geographical location generally recorded includes longitude and latitude and the angle between the road direction and the longitude. When the meta-coordinate is a turning point or an intersection, all road directions and longitude angles are recorded, the turning point is regarded as two roads, and a label is defined for each road. In addition, at non-90° turns, a 1m extension line is left after the marking lines cross.

[0092] Figure 6 is a schematic diagram of coordinate selection of a target vehicle on a straight road relative to a meta-coordinate 60 according to an embodiment of the present invention; Figure 7It is a schematic diagram of the coordinate selection of the target vehicle relative to the meta-coordinate 60 on a turning or crossing road according to an embodiment of the present invention. In this embodiment, when analyzing the picture information, the target vehicle in the captured picture is first replaced with a rectangle whose side length is parallel to or perpendicular to the road surface identification grid 20 as the vehicle position, and the vehicle edge is tangent to the rectangle (that is, the vehicle edge is completely contained in the rectangle, and the edge has a tangent point). The coordinates of the vehicle rectangular position are expressed as, and the coordinate system is an orthogonal coordinate system with the meta-coordinate 60 as the origin along the direction of the road edge 70 as the X-axis direction and the direction perpendicular to the road edge 70 as the Y-axis direction. X1 is the X-axis coordinate of the near meta-coordinate of the vehicle edge, X2 is the X-axis coordinate of the far meta-coordinate of the vehicle edge, Y1 is the Y-axis coordinate of the near meta-coordinate of the vehicle edge, and Y2 is the Y-axis coordinate of the far meta-coordinate of the vehicle edge.

[0093] When the target vehicle is on a straight road, one set of coordinate information can be obtained, and when the target vehicle is in a turning transition area, the vehicle body is located in different groups of identification grids 20, and two sets of coordinate position information can be obtained. Figure 5 and Figure 6 shown. Figure 5 On a straight road, the vehicle has only one set of coordinates, namely (X1, X2, Y1, Y2). When the vehicle is at a turn or intersection, Figure 6 As shown, the vehicle may have at least two sets of coordinates, taking two sets of coordinates as an example, (X1, X2, Y1, Y2) and (X1', X2', Y1', Y2') are obtained respectively. In this way, the relative position information of the vehicle relative to the meta-coordinates of the vehicle relative to the identification grid at the corresponding position in the first corrected image information can be obtained. When the relative position information of the vehicle is determined, the actual position information of the vehicle can be easily obtained according to the actual position information of each meta-coordinate stored in advance.

[0094] Fig.10 1 is a schematic block diagram of a vehicle according to an embodiment of the present invention. As a specific embodiment of the present invention, the present embodiment also includes a vehicle 200, and the vehicle 200 includes the above-mentioned vehicle positioning system 100 based on the ground identification grid. Specifically, the vehicle positioning system 100 based on the ground identification grid can accurately locate the vehicle, the system composition is simpler, the cost is relatively low, and the vehicle positioning speed is fast and the positioning is accurate.

[0095] At this point, those skilled in the art should recognize that, although multiple exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications that conform to the principles of the present invention can still be directly determined or derived based on the content disclosed in the present invention without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should be understood and recognized as covering all these other variations or modifications.

Claims

1. A vehicle positioning method based on ground marking grid, characterized in that: include: Acquire real-time image information of the vehicle in an area with a marking grid, wherein the marking grid is a marking grid pre-drawn on the road surface, and the marking grid has a meta coordinate; Correcting the real-time image information to obtain first corrected image information; Obtaining relative position information of the vehicle relative to the meta-coordinates of the identification grid at the corresponding position according to the first corrected image information; Obtaining the actual position information of the vehicle according to the relative position information and the actual position information of the meta-coordinates stored in advance; The process of correcting the real-time image information includes: Based on the real-time image information, obtaining an angle between a projection line of the longitudinal axis of the vehicle on the road surface and a direction of the road surface; Correcting the real-time image information according to the angle to obtain second corrected image information; Obtaining first information of the vehicle camera according to the second corrected image information, wherein the first information is the relative position information of the vehicle camera relative to the element coordinates of the identification grid in the second corrected image information and the angle information with the ground; Compare the first information obtained from the vehicle camera with the second information, and obtain a correction value in combination with the distortion parameter of the vehicle camera, wherein the second information is the relative position information of the vehicle camera relative to the meta-coordinates of the reference identification grid and the angle information with the ground which are pre-calibrated; The second corrected image information is corrected according to the correction value to obtain the first corrected image information.

2. The vehicle positioning method based on ground identification grid according to claim 1, characterized in that: The identification grid includes a plurality of identification grids drawn on the road surface in different areas; each group of the identification grids corresponds to a meta-coordinate.

3. The vehicle positioning method based on ground identification grid according to claim 2 is characterized in that: The identification grid is a cross grid formed by first identification lines and second identification lines, wherein the first identification lines are multiple identification lines with a preset width and a preset spacing drawn in a direction parallel to the road edge, and the second identification lines are multiple identification lines with the preset width and the preset spacing drawn in a direction perpendicular to the road edge.

4. The vehicle positioning method based on ground identification grid according to claim 1, characterized in that: The meta-coordinates include: On a straight road, the meta-coordinate is selected as any intersection in the road identification grid, and the identification grid at the meta-coordinate position is marked with a slash; or At a turn or intersection, the meta-coordinate is the intersection of two marking lines on the road in two directions, which are respectively parallel to the road edges of the corresponding road.

5. A vehicle positioning system based on a ground grid, characterized in that: include: A vehicle camera, used to obtain real-time image information of a vehicle in an area having a marking grid, wherein the marking grid is a marking grid pre-drawn on a road surface, and the marking grid has a meta coordinate; and A vehicle controller, the vehicle controller comprising: A corrected image acquisition module, used for correcting the real-time image information to obtain first corrected image information; A relative position information acquisition module, used to obtain the relative position information of the vehicle relative to the meta-coordinates of the identification grid at the corresponding position according to the first corrected image information; and An actual position information acquisition module, used to obtain the actual position information of the vehicle according to the relative position information and the actual position information of the meta-coordinates stored in advance; The corrected image acquisition module comprises: An angle acquisition unit, used to obtain an angle between a projection line of the longitudinal axis of the vehicle on the road surface and the road surface direction based on the real-time image information; A second corrected image acquisition unit, configured to correct the real-time image information according to the included angle to obtain second corrected image information; A first information acquisition unit, configured to obtain first information of the vehicle camera according to the second corrected image information, wherein the first information is relative position information of the vehicle camera relative to the element coordinates of the identification grid in the second corrected image information and angle information with the ground; A correction value acquisition unit, used for comparing the first information obtained from the vehicle camera with the second information, and obtaining a correction value in combination with the distortion parameter of the vehicle camera, wherein the second information is the relative position information of the vehicle camera relative to the meta-coordinates of the reference identification grid and the angle information with the ground which are pre-calibrated; The first corrected image acquisition unit is used to correct the second corrected image information according to the correction value to obtain the first corrected image information.

6. The vehicle positioning system based on ground marking grid according to claim 5, characterized in that: The identification grid includes a plurality of identification grids drawn on the road surface in different areas; each group of the identification grids corresponds to a meta-coordinate.

7. The vehicle positioning system based on ground marking grid according to claim 6, characterized in that: The identification grid is a cross grid formed by first identification lines and second identification lines, wherein the first identification lines are multiple identification lines with a preset width and a preset spacing drawn in a direction parallel to the road edge, and the second identification lines are multiple identification lines with the preset width and the preset spacing drawn in a direction perpendicular to the road edge.

8. The vehicle positioning system based on ground marking grid according to claim 5, characterized in that: The meta-coordinates include: On a straight road, the meta-coordinate is selected as any intersection in the road identification grid, and the identification grid at the meta-coordinate position is marked with a slash; or At a turn or intersection, the meta-coordinate is the intersection of two marking lines on the road in two directions, which are respectively parallel to the road edges of the corresponding road.

9. A vehicle, characterized in that: A vehicle positioning system based on a ground marking grid comprising any one of claims 5-8.

Images