A vehicle positioning method, device and storage medium

Abstract

The present invention discloses a vehicle positioning method, device, and storage medium, relating to the technical field of vehicle positioning, and capable of improving the accuracy and robustness of vehicle pose information. The method includes: obtaining first road structured information and corresponding recognition accuracy through a vision sensor; determining a first pose of the vehicle when the vision sensor obtains the first road structured information, and obtaining a target map according to the first pose, the target map including second road structured information; performing visual environment modeling on reference information to determine third road structured information, the reference information including the first road structured information and the second road structured information; matching the second road structured information and the third road structured information to determine a first matching pair, the first matching pair including the second road structured information and target road structured information that are successfully matched; and determining a target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structured information.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle positioning, and in particular, to a vehicle positioning method, device, and storage medium. Background Art

[0002] The positioning function is one of the core functions of driverless vehicles, and can provide the pose information of the vehicle to modules such as planning and control of driverless vehicles. The accuracy and robustness of the positioning function are directly related to the output results of modules such as planning and control.

[0003] Currently, driverless vehicles can adopt a visual matching positioning method to perform pixel matching between visual recognition information and the reprojection of a map in the pixel plane, so as to determine the pose information of the driverless vehicle. However, the accuracy and robustness of the pose information of the vehicle determined in this way are relatively low. Summary of the Invention

[0004] The present invention provides a vehicle positioning method, device, and storage medium, which can improve the accuracy and robustness of the pose information of the vehicle.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a vehicle positioning method, and the method includes:

[0007] Obtain first road structured information and the corresponding recognition accuracy through a vision sensor;

[0008] Determine the first pose when the vision sensor obtains the first road structured information;

[0009] According to the first pose, obtain a target map, where the target map is a local map corresponding to the first pose, and the target map includes second road structured information;

[0010] Perform visual environment modeling on the reference information to determine third road structured information, where the reference information includes the first road structured information and the second road structured information;

[0011] Match the second road structured information and the third road structured information to determine a first matching pair, where the first matching pair includes the second road structured information and the target road structured information that are successfully matched;

[0012] According to the first pose, the first matching pair, and the recognition accuracy of the target road structured information, determine the target pose of the vehicle in the target map.

[0013] Using the vehicle positioning method provided by the present invention, the in-vehicle terminal obtains the first road structure information and the corresponding recognition accuracy through a vision sensor, and determines the first pose of the vehicle when the vision sensor obtains the first road structure information. The in-vehicle terminal is also used to obtain a target map including the second road structure information according to the first pose, and combine the first road structure information and the second road structure information to determine more accurate third road structure information. When determining the target pose of the vehicle, two matching processes are performed. First, the in-vehicle terminal directly matches the second road structure information and the third road structure information to determine the first matching pair. Compared with the prior art where the road structure information in the map needs to be projected into the coordinate system of the vision sensor and then matched based on pixel data, the amount of calculation is reduced, and the accuracy of the first matching pair is improved. Then, the in-vehicle terminal performs secondary matching according to the first matching pair, the first pose, and the recognition accuracy of the target road structure information in the first matching pair, so as to determine the target pose, further improving the accuracy and robustness of the target pose.

[0014] In a possible implementation manner, the above vehicle positioning method further includes:

[0015] Obtain a historical image according to the first pose, where the historical image includes an image obtained at the position corresponding to the first pose, and the historical image includes the fourth road structure information;

[0016] The reference information further includes the fourth road structure information.

[0017] In a possible implementation manner, determining the target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structure information includes:

[0018] Determine a first coordinate set of the target road structure information in the coordinate system of the vision sensor, where the first coordinate set includes the coordinates of a plurality of first discrete points;

[0019] Determine a second coordinate set of the second road structure information in the first matching pair in the coordinate system of the vision sensor, where the second coordinate set includes the coordinates of a plurality of second discrete points;

[0020] According to the first coordinate set and the second coordinate set, determine the target distance between the target road structure information and the corresponding second road structure information;

[0021] According to the recognition accuracy and the target distance, determine a pose determination model;

[0022] Adopt a preset iteration rule, and determine the target pose according to the first pose and the pose determination model.

[0023] In a possible implementation, determining a first coordinate set of the target road structure information in the coordinate system of the vision sensor includes:

[0024] Determining a third coordinate set of the target road structure information in the coordinate system of the vision sensor;

[0025] If the third coordinate set includes the coordinates of multiple third discrete points, using the coordinates of the multiple third discrete points as the coordinates of the first discrete points;

[0026] If the third coordinate set includes the coordinates of line information, discretizing the line information to obtain the coordinates of the first discrete points corresponding to the line information.

[0027] In a possible implementation, the line information is any one of lane boundary information, road boundary information, or lamp post information.

[0028] In a possible implementation, determining a second coordinate set of the second road structure information in the first matching pair in the coordinate system of the vision sensor includes:

[0029] Determining a fourth coordinate set of the second road structure information in the first matching pair in the coordinate system of the target map;

[0030] Converting the fourth coordinate set into a second coordinate set in the coordinate system of the vision sensor.

[0031] In a possible implementation, determining a pose determination model according to the recognition accuracy and the target distance includes:

[0032] Determining the weight corresponding to each target road structure information according to the recognition accuracy of each target road structure information and the sum of the recognition accuracies of all target road structure information;

[0033] Determining a pose determination model according to the weight corresponding to each target road structure information and the target distance between each target road structure information and the corresponding second road structure information.

[0034] In a possible implementation, obtaining the first road structure information and the corresponding recognition accuracy through the vision sensor includes:

[0035] Obtaining a captured image through the vision sensor;

[0036] Performing image recognition processing on the captured image through the vision sensor to obtain at least one first road structure information included in the captured image and the recognition accuracy corresponding to each first road structure information.

[0037] In a possible implementation, determining the first pose of the vehicle when the vision sensor obtains the first road structure information includes:

[0038] Determine the first pose based on the third pose, the first moment, the second moment, and the motion information between the first moment and the second moment. The first moment is the moment when the vehicle is in the third pose, and the second moment is the moment when the visual sensor obtains the first road structure information.

[0039] In a possible implementation, the second road structure information includes M first lane lines, and the third road structure information includes N second lane lines, where N and M are positive integers.

[0040] Match the second road structure information and the third road structure information to determine the first matching pair, including:

[0041] Based on the first pose, the positioning accuracy, the second road structure information, and the third road structure information, determine the first lane information included in the second road structure information. The lane information is used to indicate the lane where the vehicle is located.

[0042] Based on the first lane information and the second lane information included in the third road structure information, determine the second matching pair that matches successfully. The second matching pair is used to indicate the matching relationship between the second road structure information and the third road structure information.

[0043] Delete the abnormal data in the second matching pair to determine the first matching pair.

[0044] In a second aspect, the present invention provides a vehicle positioning device, which includes:

[0045] An acquisition unit, configured to obtain the first road structure information and the corresponding recognition accuracy through a visual sensor.

[0046] A determination unit, configured to determine the first pose when the visual sensor obtains the first road structure information.

[0047] The acquisition unit is further configured to obtain a target map according to the first pose. The target map is a local map corresponding to the first pose, and the target map includes the second road structure information.

[0048] The determination unit is further configured to perform visual environment modeling on the reference information to determine the third road structure information. The reference information includes the first road structure information and the second road structure information. Match the second road structure information and the third road structure information to determine the first matching pair. The first matching pair includes the second road structure information and the target road structure information that match successfully, and determine the target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structure information.

[0049] In a possible implementation, the obtaining unit is further configured to obtain a historical image according to the first pose, where the historical image includes an image obtained at a position corresponding to the first pose, and the historical image includes fourth road structure information;

[0050] The reference information further includes fourth road structure information.

[0051] In a possible implementation, the determining unit is specifically configured to:

[0052] Determine a first coordinate set of the target road structure information in the coordinate system of the vision sensor, where the first coordinate set includes coordinates of a plurality of first discrete points;

[0053] Determine a second coordinate set of the second road structure information in the first matching pair in the coordinate system of the vision sensor, where the second coordinate set includes coordinates of a plurality of second discrete points;

[0054] Determine a target distance between the target road structure information and the corresponding second road structure information according to the first coordinate set and the second coordinate set;

[0055] Determine the second pose according to the recognition accuracy and the target distance;

[0056] Adopt a preset iteration rule to determine the target pose according to the first pose and the second pose.

[0057] In a possible implementation, the determining unit is specifically configured to:

[0058] Determine a third coordinate set of the target road structure information in the coordinate system of the vision sensor;

[0059] If the third coordinate set includes coordinates of a plurality of third discrete points, use the coordinates of the plurality of third discrete points as the coordinates of the first discrete points;

[0060] If the third coordinate set includes coordinates of line information, perform discrete processing on the line information to obtain coordinates of first discrete points corresponding to the line information.

[0061] In a possible implementation, the line information is any one of lane boundary information, road boundary information, or lamp post information.

[0062] In a possible implementation, the determining unit is specifically configured to:

[0063] Determine a fourth coordinate set of the second road structure information in the first matching pair in the coordinate system of the target map;

[0064] Convert the fourth coordinate set into the second coordinate set in the coordinate system of the vision sensor.

[0065] In a possible implementation, the determination unit is specifically configured to: The determination unit is specifically configured to:

[0066] Determine the weight corresponding to each target road structure information according to the recognition accuracy of each target road structure information and the sum of the recognition accuracies of all target road structure information;

[0067] Determine the second pose according to the weight corresponding to each target road structure information and the target distance between each target road structure information and the corresponding second road structure information.

[0068] In a possible implementation, the acquisition unit is specifically configured to:

[0069] Acquire a captured image through a vision sensor;

[0070] Perform image recognition processing on the captured image through a vision sensor to obtain at least one first road structure information included in the captured image and the recognition accuracy corresponding to each first road structure information.

[0071] In a possible implementation, the determination unit is specifically configured to:

[0072] Determine the first pose according to the third pose, the first moment, the second moment, and the motion information between the first moment and the second moment, where the first moment is the moment when the vehicle is in the third pose, and the second moment is the moment when the vision sensor acquires the first road structure information.

[0073] In a possible implementation, the second road structure information includes M first lane lines, and the third road structure information includes N second lane lines, where N and M are positive integers;

[0074] The determination unit is specifically configured to:

[0075] Determine the first lane information included in the second road structure information according to the first pose, the positioning accuracy, the second road structure information, and the third road structure information, where the lane information is used to indicate the lane where the vehicle is located;

[0076] Determine the successfully matched second matching pair according to the first lane information and the second lane information included in the third road structure information, where the second matching pair is used to indicate the matching relationship between the second road structure information and the third road structure information;

[0077] Delete the abnormal data in the second matching pair and determine the first matching pair.

[0078] In a third aspect, the present invention provides a vehicle positioning device, which includes: a processor and a memory. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the vehicle positioning device executes the vehicle positioning method as described in the first aspect and any possible implementation manner thereof.

[0079] In a fourth aspect, the present invention provides a vehicle, which includes: a processor and a memory. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the vehicle executes the vehicle positioning method as described in the first aspect and any possible implementation manner thereof.

[0080] In a fifth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored. When the computer instructions run on the vehicle positioning device, the vehicle positioning device is caused to execute the vehicle positioning method as described in any item of the first aspect or the possible implementation manners of the first aspect. Alternatively, when the computer instructions run on the vehicle, the vehicle is caused to execute the vehicle positioning method as described in the first aspect and any possible implementation manner thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 is a schematic structural diagram of a vehicle positioning system provided by an embodiment of the present invention;

[0082] Figure 2 is one of the schematic structural diagrams of the vehicle positioning device provided by an embodiment of the present invention;

[0083] Figure 3 is one of the schematic flowcharts of the vehicle positioning method provided by an embodiment of the present invention;

[0084] Figure 4 is the second schematic flowchart of the vehicle positioning method provided by an embodiment of the present invention;

[0085] Figure 5 is the third schematic flowchart of the vehicle positioning method provided by an embodiment of the present invention;

[0086] Figure 6 is the fourth schematic flowchart of the vehicle positioning method provided by an embodiment of the present invention;

[0087] Figure 7 is the second schematic structural diagram of the vehicle positioning device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0088] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0089] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present disclosure, unless otherwise stated, the meaning of "a plurality" is two or more. Additionally, the use of "based on" or "in accordance with" implies openness and inclusiveness, as a process, step, calculation, or other action "based on" or "in accordance with" one or more of the stated conditions or values may, in practice, be based on additional conditions or values beyond those stated.

[0090] In order to improve the accuracy and robustness of the pose information of a vehicle, embodiments of the present invention provide a vehicle positioning method, device, and storage medium. An in-vehicle terminal acquires first road structure information and the corresponding recognition accuracy through a vision sensor, and determines the first pose of the vehicle when the vision sensor acquires the first road structure information. The in-vehicle terminal acquires a target map according to the first pose, and the target map includes second road structure information. After the in-vehicle terminal performs visual environment modeling on the first road structure information and the second road structure information, it determines third road structure information. The in-vehicle terminal directly performs a first matching on the second road structure information and the third road structure information to determine a first matching pair. The in-vehicle terminal performs a second matching according to the first pose, the first matching pair, and the recognition accuracy of the target road structure information in the first matching pair, and then the target pose of the vehicle in the target map can be obtained. Through two matches, the accuracy and robustness of the target pose can be improved.

[0091] The execution subject of the vehicle positioning method provided by the embodiments of the present invention is a vehicle positioning device. The vehicle positioning device may be an in-vehicle terminal, or a central processing unit (CPU) in the above in-vehicle terminal, or a client for vehicle positioning in the above in-vehicle terminal. In the embodiments of the present invention, the vehicle positioning method is described by taking the in-vehicle terminal as an example of executing the vehicle positioning method.

[0092] The vehicle positioning method provided by the embodiments of the present invention can be applied to a vehicle positioning system. Figure 1 A schematic structural diagram of the vehicle positioning system is shown. As Figure 1As shown in the figure, the vehicle positioning system may include a vehicle 11 and a map server 12. Among them, a vision sensor, a positioning receiver, and an in-vehicle terminal are installed on the vehicle 11. The in-vehicle terminal is respectively connected to the map server 12, the vision sensor, and the positioning receiver through wired communication or wireless communication.

[0093] The vision sensor is used to capture an image, identify the captured image, determine the road structure information included in the captured image and the corresponding recognition accuracy, and send the road structure information and the corresponding recognition accuracy to the in-vehicle terminal.

[0094] In some embodiments, the vision sensor includes a camera. During the driving of the vehicle, the vision sensor captures images in real time through the camera, performs image recognition processing on the images, determines the road structure information included in the images and the corresponding recognition accuracy, and sends the road structure information and the corresponding recognition accuracy to the in-vehicle terminal.

[0095] In some embodiments, the camera can be an ADAS camera. The road structure information may include: lane lines, road boundaries, lamp posts, traffic signs, traffic lights, speed limit signs, and other information.

[0096] The positioning receiver is used to obtain global positioning information when the vehicle starts and send the global positioning information to the in-vehicle terminal. The global positioning information may include the longitude and latitude, altitude, attitude information, and positioning accuracy when the vehicle starts. Exemplarily, the positioning receiver can be a sensor based on the Global Positioning System (GPS), or a sensor based on the Global Navigation Satellite System (GNSS).

[0097] The map server 12 is used to send the local map corresponding to the vehicle in different poses to the in-vehicle terminal.

[0098] The in-vehicle terminal is used to obtain the first road structure information and the corresponding recognition accuracy, determine the first pose of the vehicle when the vision sensor obtains the first road structure information, and obtain the target map corresponding to the first pose according to the first pose. The in-vehicle terminal is also used to determine the target pose of the vehicle in the target map after two matches according to the first road structure information, the second road structure information included in the target map, and the first pose.

[0099] Figure 2 One of the structural schematic diagrams of the vehicle positioning device, such as Figure 2As shown in the figure, the vehicle positioning device may include: a processor 21, a memory 22, a communication interface 23, and a bus 24. The processor 21, the memory 22, and the communication interface 23 may be connected through the communication bus 24.

[0100] The processor 21 is the control center of the vehicle positioning device, which may be a single processor 21 or a collective term for multiple processing elements. For example, the processor 21 may be a general-purpose CPU or other general-purpose processors 21. Among them, the general-purpose processor 21 may be a microprocessor 21 or any conventional processor 21, etc.

[0101] As an embodiment, the processor 21 may include one or more CPUs. For example, Figure 2 the illustrated CPU0 and CPU1.

[0102] The memory 22 may be a read-only memory 22 (ROM) or other types of static storage devices that can store static information and instructions, a random access memory 22 (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory 22 (EEPROM), a magnetic disk storage medium, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0103] In a possible implementation, the memory 22 may exist independently of the processor 21. The memory 22 may be connected to the processor 21 through the bus 24 for storing instructions or program codes. When the processor 21 calls and executes the instructions or program codes stored in the memory 22, the vehicle positioning method provided in the following embodiments of the present invention can be implemented.

[0104] In another possible implementation, the memory 22 may also be integrated with the processor 21.

[0105] The communication interface 23 is used for the vehicle positioning device to connect to other devices through a communication network. The communication network may be an Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc. The communication interface 23 may include a receiving unit for receiving data and a sending unit for sending data.

[0106] The bus 24 can be an Industry Standard Architecture (ISA) bus 24, a Peripheral Component Interconnect (PCI) bus 24, an Extended Industry Standard Architecture (EISA) bus 24, etc. The bus 24 can be divided into an address bus 24, a data bus 24, a control bus 24, etc. For the sake of convenience of representation, Figure 2 it is only represented by a thick line in the figure, but it does not mean that there is only one bus 24 or one type of bus 24.

[0107] It should be noted that, Figure 2 the structure shown in the figure does not constitute a limitation on the vehicle positioning device. Except Figure 2 for the components shown, the vehicle positioning device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0108] The vehicle positioning method provided by the embodiments of the present invention will be described below with reference to the accompanying drawings.

[0109] As Figure 3 shown, the vehicle positioning method provided by the embodiments of the present invention includes the following steps 301-step 306.

[0110] 301. The in-vehicle terminal obtains first road structured information and the corresponding recognition accuracy through a vision sensor.

[0111] During the driving process of the driverless vehicle, the in-vehicle terminal needs to determine the real-time pose of the vehicle in real time. At the current moment, when the in-vehicle terminal needs to determine the current pose of the vehicle, the in-vehicle terminal can obtain the first road structured information and the corresponding recognition accuracy through a vision sensor. Among them, the first road structured information is the road structured information in the image captured by the camera in the vision sensor at the current moment. The road structured information may include: lane lines, road boundaries, lamp posts, traffic signs, traffic lights, speed limit signs, etc. information. Adjacent lane lines can form a lane.

[0112] Optionally, the in-vehicle terminal obtains the first road structured information and the corresponding recognition accuracy through a vision sensor, which may include: the in-vehicle terminal obtains a captured image through the camera in the vision sensor, and performs image recognition processing on the captured image through the vision sensor to obtain at least one first road structured information included in the captured image, and the recognition accuracy corresponding to each first road structured information.

[0113] Optionally, after the vehicle is started, the vehicle-mounted terminal can obtain global positioning information from the positioning receiver and initialize the pre-stored pose information in the vehicle-mounted terminal according to the global positioning information, so as to obtain the initial pose of the vehicle. The global positioning information may include: when the vehicle is started, the position information, attitude information and positioning accuracy of the vehicle, where the position information may include longitude and latitude, and altitude.

[0114] Exemplarily, the vehicle-mounted terminal can use the following formula (1) to initialize the pre-stored pose information in the vehicle-mounted terminal according to the global positioning information.

[0115] P(x, y, z, θ x , θ y , θ z ) = G(x, y, z, roll, pitch, yaw) (1)

[0116] where G(x, y, z, roll, pitch, yaw) is the position information and attitude information included in the global positioning information, and P(x, y, z, θ x , θ y , θ z ) is the initial position information and initial attitude information of the vehicle.

[0117] 302. The vehicle-mounted terminal determines the first pose of the vehicle when the vision sensor obtains the first road structure information.

[0118] Since the time stamp of the camera is inconsistent with the moment corresponding to the vehicle-mounted terminal when the vehicle is in the initial pose when the camera obtains the captured image, the vehicle-mounted terminal needs to predict the pose of the vehicle when the camera obtains the captured image according to the speed information and acceleration information of the vehicle, that is, determine the first pose of the vehicle when the vision sensor obtains the first road structure information.

[0119] Optionally, when the vehicle-mounted terminal determines the first pose when the vision sensor obtains the first road structure information, it can use the following formula (2) to determine the first pose according to the third pose, the first moment, the second moment, and the motion information between the first moment and the second moment. Among them, when the vehicle is initially started, the third pose is the initial pose; after the vehicle is started, the third pose is the pose of the vehicle at the previous moment. The first moment is the moment when the vehicle is in the third pose, and the second moment is the moment when the vision sensor obtains the first road structure information.

[0120]

[0121] Where, t is the first moment, T is the second moment, P(T) is the first pose, P(t) is the third pose, v is the speed of the vehicle from the first moment to the second moment, and a is the acceleration of the vehicle from the first moment to the second moment.

[0122] 303. The in-vehicle terminal obtains a target map according to the first pose.

[0123] After the in-vehicle terminal determines the first pose of the vehicle, it can obtain the local map corresponding to the first pose, that is, the target map, from the map server according to the first position information and the first attitude information included in the first pose. The target map may include second road structured information, and the second road structured information is discrete three-dimensional point coordinates.

[0124] 304. The in-vehicle terminal performs visual environment modeling on the reference information to determine third road structured information.

[0125] Optionally, the reference information may include first road structured information and second road structured information. Since the field of view range of the camera is limited when obtaining the captured image and only part of the road structured information can be recognized, it is necessary to perform visual environment modeling by combining the first road structured information obtained by the visual sensor and the second road structured information in the target map to obtain the third road structured information. The third road structured information is more accurate road structured information than the first road structured information.

[0126] Exemplarily, the principles of visual environment modeling may include: if two lanes included in the second road structured information in the target map are merged into one lane, then one corresponding lane in the first road structured information is deleted. If one lane included in the second road structured information in the target map is divided into two lanes, then the corresponding lanes are added in the first road structured information.

[0127] 305. The in-vehicle terminal matches the second road structured information and the third road structured information to determine a first matching pair, and the first matching pair includes the second road structured information and the target road structured information that match successfully.

[0128] Optionally, the target road structured information may be the road structured information in the third road structured information that matches successfully with one second road structured information.

[0129] 306. The in-vehicle terminal determines the target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structured information.

[0130] Using the vehicle positioning method provided by the embodiments of the present invention, the in-vehicle terminal obtains the first road structure information and the corresponding recognition accuracy through a vision sensor, and determines the first pose of the vehicle when the vision sensor obtains the first road structure information. The in-vehicle terminal is further configured to obtain a target map including the second road structure information according to the first pose, and determine more accurate third road structure information by combining the first road structure information and the second road structure information. When determining the target pose of the vehicle, two matching processes are performed. First, the in-vehicle terminal directly matches the second road structure information and the third road structure information to determine the first matching pair. Compared with the prior art that needs to project the road structure information in the map into the coordinate system of the vision sensor and then perform matching based on pixel data, the amount of calculation is reduced, and the accuracy of the first matching pair is improved. Then, the in-vehicle terminal jointly determines the target pose according to the first matching pair, the first pose, and the recognition accuracy of the target road structure information in the first matching pair, further improving the accuracy and robustness of the target pose.

[0131] Combine Figure 3 , such as Figure 4 As shown, the vehicle positioning method provided by the embodiments of the present invention may further include step 401.

[0132] 401. The in-vehicle terminal obtains a historical image according to the first pose.

[0133] Optionally, the historical image includes an image obtained at the position corresponding to the first pose, and the historical image includes fourth road structure information. At this time, the above reference information may further include the fourth road structure information. The fourth road structure information may be the same as or different from the first road structure information. When the in-vehicle terminal performs visual environment modeling according to the first road structure information, the second road structure information, and the third road structure information, more accurate third road structure information can be obtained.

[0134] Exemplarily, when the camera captures the historical image, if the historical pose of the vehicle is the same as the first pose, the fourth road structure information is the same as the first road structure information; if the historical pose in the historical pose of the vehicle is different from the first pose in the first pose, the fourth road structure information is different from the first road structure information.

[0135] Optionally, the second road structure information may include M first lane lines, and the third road structure information may include N second lane lines. N and M are positive integers, and N and M are the same or N and M are different.

[0136] Combine Figure 4 , such as Figure 5 As shown, the above step 305 may include the following steps 501-step 503.

[0137] 501. The vehicle-mounted terminal determines the first lane information included in the second road structure information according to the first pose, the positioning accuracy, and the second road structure information, and the lane information is used to indicate the lane where the vehicle is located.

[0138] 502. The vehicle-mounted terminal determines a second matching pair according to the first lane information and the second lane information included in the third road structure information, and the second matching pair is used to indicate the matching relationship between the second road structure information and the third road structure information.

[0139] 503. The vehicle-mounted terminal deletes the abnormal data in the second matching pair to determine the first matching pair.

[0140] Optionally, if the matching relationship between a second road structure information in the second matching pair and the corresponding third road structure information is incorrect, then this matching relationship is abnormal data. Then, at least one matching relationship in the first matching pair is a possibly correct matching relationship, and a matching relationship includes multiple target road structure information and the corresponding second road structure information.

[0141] Exemplarily, when both the second road structure information and the third road structure information are lane lines, if the lane line attributes in the second road structure information are inconsistent with the lane line attributes in the third road structure information, then this matching relationship is incorrect. Among them, the lane line attributes may include the color and type of the lane line, and the type may include solid line, dashed line, straight line, curve, but is not limited thereto.

[0142] Optionally, if the third road structure information includes a road boundary, or the number of lanes in the third road structure information is the same as the number of lanes in the second road structure information, or the number of solid lines or dashed lines of the lane lines in the third road structure information is the same as that in the second road structure information, it indicates that the first matching pair is in a converged state. The first matching pair being in a converged state indicates that there is only one pair of matching relationships in the first matching pair, that is, it can be determined that the vehicle is in a specific lane. However, at this time, the pose information of the vehicle in this lane cannot be determined yet.

[0143] Optionally, if the first matching pair is already in a converged state, and the positioning accuracy exceeds the threshold, but the target road structure information in the first matching pair is inconsistent with the second road structure information, it indicates that the above matching process has degenerated, and the first matching pair is still in a non-converged state, indicating that there are multiple pairs of matching relationships in the first matching pair, that is, there may be multiple results for the vehicle's position at this time. At this time, the vehicle-mounted terminal needs to re-execute steps 501 - 503.

[0144] Optionally, if the first matching pair remains unconverged after repeating steps 501 - 503, then it is necessary to perform secondary matching on multiple pairs of matching relationships in the first matching pair to respectively determine the attitude information of the vehicle in each lane.

[0145] Combined with Figure 5 , such as Figure 6 shown, the above step 306 may include the following steps 601 - 605.

[0146] 601. The in - vehicle terminal determines a first coordinate set of the target road structure information in the coordinate system of the vision sensor.

[0147] Optionally, the first coordinate set may include the coordinates of multiple first discrete points. When determining the first coordinate set, the in - vehicle terminal needs to first determine a third coordinate set of the target road structure information in the coordinate system of the vision sensor. If the third coordinate set includes the coordinates of multiple third discrete points, the in - vehicle terminal takes the coordinates of the multiple third discrete points as the coordinates of the first discrete points. If the third coordinate set includes the coordinates of line information, the in - vehicle terminal performs discrete processing on the line information to obtain the coordinates of the first discrete points corresponding to the line information.

[0148] Optionally, the line information may be any one of lane boundary information, road boundary information, or lamp post information. Information such as traffic signs or speed limit signs may be the coordinates of discrete points.

[0149] Exemplarily, the expression of the line information can be represented by the following formulas (3) and (4).

[0150] Y = C0 + C1×X + C2×X 2 + C3×X 3 , X ∈ (start, end) (3)

[0151] Z = Z0 (4)

[0152] Wherein, C0, C1, C2, and C3 are the coefficients of the line information in the target road structure information, X is the X - axis coordinate of the line information in the target road structure information, start is the starting coordinate of the X - axis coordinate of the line information in the target road structure information, end is the ending coordinate of the X - axis coordinate of the line information in the target road structure information, and Z0 is the Z - axis coordinate.

[0153] Exemplarily, when the in - vehicle terminal performs discrete processing on the line information, it can obtain the coordinates of the first discrete points corresponding to the line information according to the following formulas (5) - (7).

[0154]

[0155]

[0156] z k = Z0(7)

[0157] where k is the sorting of the line information in the target road structured information in the first matching pair; n is the number of the target road structured information in the first matching pair; x k is the X-axis coordinate of the k-th first discrete point corresponding to the line information; y k is the Y-axis coordinate of the k-th first discrete point corresponding to the line information; z k is the Z-axis coordinate of the k-th first discrete point corresponding to the line information.

[0158] 602. The vehicle-mounted terminal determines the second coordinate set of the second road structured information in the first matching pair in the coordinate system of the vision sensor.

[0159] Optionally, the second coordinate set may include the coordinates of multiple second discrete points. When determining the second coordinate set, the vehicle-mounted terminal may first determine the fourth coordinate set of the second road structured information in the first matching pair in the coordinate system of the target map, and then convert the fourth coordinate set into the second coordinate set in the coordinate system of the vision sensor.

[0160] Exemplarily, the vehicle-mounted terminal may use the following formulas (8)-(10) to convert the coordinates of a discrete point in the fourth coordinate set into the coordinates of a discrete point in the second coordinate set.

[0161] X tmap = (X map - X camera ) × cosθ y' + (Y map - Y camera ) × sinθ y' (8)

[0162] Y tmap = -(X map - X camera ) × sinθ y' + (Y map - Y camera ) × cosθ y' (9)

[0163] Z tmap = Z map (10)

[0164] where, (X map , Y map , Z map ) is the first global three-dimensional coordinate of the second road structured information in the coordinate system of the target map, θ y'When the vehicle is in the first pose, it is the azimuth angle of the camera in the vision sensor, (X camera , Y camera , Z camera ) is the second global three-dimensional coordinate of the second road structure information in the coordinate system of the target map, and (X tmap , Y tmap , Z tmap ) is the local three-dimensional coordinate of the second road structure information in the coordinate system of the vision sensor.

[0165] 603. The vehicle-mounted terminal determines the target distance between the target road structure information and the corresponding second road structure information according to the first coordinate set and the second coordinate set.

[0166] Optionally, when the target road structure information is the coordinate of the first discrete point and the second road structure information is the coordinate of the second discrete point, the distance between the coordinate of the first discrete point and the coordinate of the corresponding second discrete point can be calculated using the Euclidean distance calculation formula.

[0167] Optionally, when the target road structure information is line information, when determining the distance from a line information to the corresponding second road structure information, it can be first converted to the distance from all the coordinates of the first discrete points to the second road structure information, and then converted to the distance from all the coordinates of the first discrete points to the foot point of the corresponding line segment in the second road structure information.

[0168] Exemplarily, taking the endpoint coordinates of the line segment as D1(x1, y1, z1) and D2(x2, y2, z2) respectively, the vehicle-mounted terminal can use the following formulas (11)-(14) to determine the foot point D of the line segment C (x c , y c , z c ).

[0169] x c = x1 + K × (x2 - x1) (11)

[0170] y c = y1 + K × (y2 - y1) (12)

[0171] z c = z1 + K × (z2 - z1) (13)

[0172]

[0173] During the matching process of the vehicle-mounted terminal, when calculating the target distance, classification matching is achieved according to the types of the second road structure information and the third road structure information, which can improve the accuracy of the matching result. That is, the coordinates of the discrete points in the second road structure information and the third road structure information are matched, and the line information in the second road structure information and the third road structure information is matched. For example, lane lines are matched with lane lines, and traffic signs are matched with traffic signs.

[0174] 604. The vehicle-mounted terminal determines the second pose according to the recognition accuracy and the target distance.

[0175] Optionally, when the vehicle-mounted terminal determines the second pose, it can determine the weight corresponding to each target road structure information according to the recognition accuracy of each target road structure information and the sum of the recognition accuracies of all target road structure information, and then determine the second pose according to the weight corresponding to each target road structure information and the target distance between each target road structure information and the corresponding second road structure information.

[0176] Exemplarily, the vehicle-mounted terminal can use the following formula (15) to determine the weight corresponding to each target road structure information.

[0177]

[0178] Where accuracy(i) is the recognition accuracy of the i-th target road structure information in each pair of matching relationships in the first matching pair, and ∑ i accuracy(i) is the sum of the recognition accuracies of all target road structure information in each pair of matching relationships in the first matching pair. According to the recognition accuracy of each target road structure information in each pair of matching relationships, weighted processing is performed during the second matching, which can improve the accuracy and robustness of the matching result.

[0179] Exemplarily, the vehicle-mounted terminal can determine the second pose corresponding to each pair of matching relationships in the first matching pair according to the following formula (16).

[0180] F(P) = ∑ i weight(i) × Distance(X i , Y i ), i ∈ (1, n) (16)

[0181] Where X i is the i-th target road structure information in each pair of matching relationships, Y i is the second road structure information corresponding to X i in each pair of matching relationships, and Distance(X i , Yi ) is the distance between the i-th target road structure information and the i-th second road structure information in each pair of matching relationships, and F(P) is the second pose of the vehicle on the target map in each pair of matching relationships.

[0182] 605. The in-vehicle terminal adopts a preset iteration rule to determine the target pose according to the first pose and the second pose.

[0183] Exemplarily, according to the second pose corresponding to each pair of matching relationships in the first matching pair, the following formulas (17) and (18) can be used to determine the target pose corresponding to the second pose.

[0184] P L+1 = P L - (J T × J + γ × I) × J T × F(P L ), L ∈ (1, n - 1) (17)

[0185] |P L+1 - P L | < W or K > U (18)

[0186] Among them, L is the number of iterations, P L is the first pose after L iterations, P L+1 is the target pose corresponding to the second pose, F(P L ) is the second pose after L iterations, J is the transformation matrix, γ is a constant coefficient, I is the identity matrix, U is the iteration number threshold, and W is the difference threshold. The iteration number threshold and the difference threshold can be set according to the actual situation.

[0187] Optionally, the in-vehicle terminal can obtain the above transformation matrix J by taking the derivative of formula (16). The transformation matrix J can be expressed by formula (19).

[0188]

[0189] Among them, represents the transpose matrix of the matrix obtained by taking the derivative of the second pose F(P) corresponding to the n-th pair of matching relationships.

[0190] Optionally, in the transformation matrix J can be calculated by formula (20).

[0191]

[0192] Among them, e x = [1, 0, 0] T , e y = [0, 1, 0]T , e z = [0, 0, 1] T , is the partial derivative of the rotation matrix R with respect to θ in the optimized increment corresponding to the second pose x , θ y , θ z .

[0193] Exemplarily, the values of can be respectively:

[0194]

[0195]

[0196]

[0197] The above mainly introduces the solution provided by the embodiments of the present invention from the perspective of the device. It can be understood that in order for the device to implement the above functions, it includes the corresponding hardware structure and / or software module for executing each function. Those skilled in the art should easily realize that, combining the algorithm steps of each example described in the embodiments disclosed in this article, the present invention can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the way of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described function for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0198] Figure 7 shows a possible composition schematic diagram of the vehicle positioning device 700 involved in the above embodiments, as Figure 7 shown, the vehicle positioning device 700 may include: an acquisition unit 701 and a determination unit 702.

[0199] Among them, the obtaining unit 701 is configured to obtain first road structure information and the corresponding recognition accuracy through a vision sensor. The determining unit 702 is configured to determine the first pose when the vision sensor obtains the first road structure information. The obtaining unit 701 is further configured to obtain a target map according to the first pose, where the target map is a local map corresponding to the first pose, and the target map includes second road structure information. The determining unit 702 is further configured to perform visual environment modeling on the reference information to determine third road structure information, where the reference information includes the first road structure information and the second road structure information, match the second road structure information and the third road structure information to determine a first matching pair, the first matching pair includes the successfully matched second road structure information and the target road structure information, and determine the target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structure information.

[0200] Optionally, the obtaining unit 701 is further configured to obtain a historical image according to the first pose, where the historical image includes an image obtained at the position corresponding to the first pose, and the historical image includes fourth road structure information.

[0201] Optionally, the reference information further includes fourth road structure information.

[0202] Optionally, the determining unit 702 is specifically configured to: determine a first coordinate set of the target road structure information in the coordinate system of the vision sensor, where the first coordinate set includes the coordinates of a plurality of first discrete points; determine a second coordinate set of the second road structure information in the first matching pair in the coordinate system of the vision sensor, where the second coordinate set includes the coordinates of a plurality of second discrete points; determine the target distance between the target road structure information and the corresponding second road structure information according to the first coordinate set and the second coordinate set; determine the second pose according to the recognition accuracy and the target distance; and determine the target pose according to the first pose and the second pose by using a preset iterative rule.

[0203] Optionally, the determining unit 702 is specifically configured to: determine a third coordinate set of the target road structure information in the coordinate system of the vision sensor. If the third coordinate set includes the coordinates of a plurality of third discrete points, use the coordinates of the plurality of third discrete points as the coordinates of the first discrete points; if the third coordinate set includes the coordinates of line information, perform discrete processing on the line information to obtain the coordinates of the first discrete points corresponding to the line information.

[0204] Optionally, the line information is any one of lane boundary information, road boundary information, or lamp post information.

[0205] Optionally, the determining unit 702 is specifically configured to: determine a fourth coordinate set of the second road structure information in the first matching pair in the coordinate system of the target map; and convert the fourth coordinate set into the second coordinate set in the coordinate system of the vision sensor.

[0206] Optionally, the determining unit 702 is specifically configured to: determine the weight corresponding to each target road structure information according to the recognition accuracy of each target road structure information and the sum of the recognition accuracies of all target road structure information; and determine the second pose according to the weight corresponding to each target road structure information and the target distance between each target road structure information and the corresponding second road structure information.

[0207] Optionally, the obtaining unit 701 is specifically configured to: obtain a captured image through a vision sensor; perform image recognition processing on the captured image through the vision sensor to obtain at least one first road structure information included in the captured image and the recognition accuracy corresponding to each first road structure information.

[0208] Optionally, the determining unit 702 is specifically configured to: determine the first pose according to the third pose, the first moment, the second moment, and the motion information between the first moment and the second moment, where the first moment is the moment when the vehicle is in the third pose, and the second moment is the moment when the vision sensor obtains the captured image.

[0209] Optionally, the second road structure information includes M first lane lines, and the third road structure information includes N second lane lines, where N and M are positive integers.

[0210] Optionally, the determining unit 702 is specifically configured to: determine the first lane information included in the second road structure information according to the first pose and the positioning accuracy, where the lane information is used to indicate the lane where the vehicle is located; and determine the successfully matched second matching pair according to the first lane information and the second lane information included in the third road structure information.

[0211] Certainly, the vehicle positioning device 700 provided in the embodiments of the present invention includes but is not limited to the above modules.

[0212] In actual implementation, the obtaining unit 701 and the determining unit 702 may be Figure 2 implemented by the processor 21 shown in Figures 3 to 6 calling the program code in the memory 22. The specific execution process may refer to the description of the vehicle positioning method part shown in

[0213] Another embodiment of the present invention further provides a vehicle, which includes a processor and a memory. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the vehicle executes the vehicle positioning method shown in the above method embodiment.

[0214] Another embodiment of the present invention further provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions run on the vehicle positioning device 700, the vehicle positioning device 700 executes each step performed by the vehicle positioning device in the method flow shown in the above method embodiment. Or, when the computer instructions run on the vehicle, the vehicle executes each step in the method flow shown in the above method embodiment.

[0215] Another embodiment of the present invention further provides a chip system, which is applied to the vehicle positioning device 700. The chip system includes one or more interface circuits and one or more processors 21. The interface circuits and the processor 21 are interconnected by lines. The interface circuit is used to receive a signal from the memory 22 of the vehicle positioning device 700 and send the signal to the processor 21, and the signal includes the computer instructions stored in the memory 22. When the processor 21 executes the computer instructions, the vehicle positioning device 700 executes each step performed by the vehicle positioning device 700 in the method flow shown in the above method embodiment.

[0216] In another embodiment of the present invention, a computer program product is further provided, and the computer program product includes instructions. When the instructions run on the vehicle positioning device 700, the vehicle positioning device 700 executes each step performed by the vehicle positioning device 700 in the method flow shown in the above method embodiment.

[0217] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer execution instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that contains one or more integrated media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0218] As described above, the above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A vehicle positioning method, characterized in that, Including: Obtaining first road structured information and corresponding recognition accuracy through a vision sensor; Determining a first pose of the vehicle when the vision sensor obtains the first road structured information; Obtaining a target map according to the first pose, where the target map is a local map corresponding to the first pose, and the target map includes second road structured information; Performing visual environment modeling on reference information to determine third road structured information, where the reference information includes the first road structured information and the second road structured information; Matching the second road structured information and the third road structured information to determine a first matching pair, where the first matching pair includes the second road structured information with successful matching and target road structured information; Determining a target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structured information.

2. The vehicle positioning method according to claim 1, wherein The vehicle positioning method further includes: Obtaining a historical image according to the first pose, where the historical image includes an image obtained at a position corresponding to the first pose, and the historical image includes fourth road structured information; The reference information further includes the fourth road structured information.

3. The vehicle positioning method according to claim 1 or 2, characterized in that, The determining the target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structured information includes: Determining a first coordinate set of the target road structured information in the coordinate system of the vision sensor, where the first coordinate set includes coordinates of a plurality of first discrete points; Determining a second coordinate set of the second road structured information in the first matching pair in the coordinate system of the vision sensor, where the second coordinate set includes coordinates of a plurality of second discrete points; Determining a target distance between the target road structured information and the corresponding second road structured information according to the first coordinate set and the second coordinate set; Determining a second pose according to the recognition accuracy and the target distance; Adopting a preset iteration rule to determine the target pose according to the first pose and the second pose.

4. The vehicle positioning method according to claim 3, wherein The determining the first coordinate set of the target road structured information in the coordinate system of the vision sensor includes: Determining a third coordinate set of the target road structured information in the coordinate system of the vision sensor; If the third coordinate set includes coordinates of a plurality of third discrete points, using the coordinates of the plurality of third discrete points as the coordinates of the first discrete points; If the third coordinate set includes coordinates of line information, performing discrete processing on the line information to obtain coordinates of first discrete points corresponding to the line information; The line information is any one of lane boundary information, road boundary information, or lamp post information.

5. The vehicle positioning method according to claim 3, wherein The determining the second coordinate set of the second road structured information in the first matching pair in the coordinate system of the vision sensor includes: Determining a fourth coordinate set of the second road structured information in the first matching pair in the coordinate system of the target map; Converting the fourth coordinate set into the second coordinate set in the coordinate system of the vision sensor.

6. The vehicle positioning method according to claim 3, wherein, Determining the second pose according to the recognition accuracy and the target distance includes: Determining the weight corresponding to each piece of target road structure information according to the recognition accuracy of each piece of target road structure information and the sum of the recognition accuracies of all pieces of target road structure information; Determining the second pose according to the weight corresponding to each piece of target road structure information and the target distance between each piece of target road structure information and the corresponding second road structure information.

7. The vehicle positioning method according to claim 1 or 2, characterized in that, Obtaining the first road structure information and the corresponding recognition accuracy through a vision sensor includes: Obtaining a captured image through the vision sensor; Performing image recognition processing on the captured image through the vision sensor to obtain at least one piece of the first road structure information included in the captured image and the recognition accuracy corresponding to each piece of the first road structure information.

8. The vehicle positioning method according to claim 1 or 2, characterized in that, Determining the first pose of the vehicle when the vision sensor obtains the first road structure information includes: Determining the first pose according to the third pose, the first moment, the second moment, and the motion information between the first moment and the second moment, where the first moment is the moment when the vehicle is in the third pose, and the second moment is the moment when the vision sensor obtains the first road structure information.

9. The vehicle positioning method according to claim 1 or 2, characterized in that, The second road structure information includes M first lane lines, and the third road structure information includes N second lane lines, where N and M are positive integers; Matching the second road structure information and the third road structure information to determine a first matching pair includes: Determining the first lane information included in the second road structure information according to the first pose, the positioning accuracy, and the second road structure information, where the lane information is used to indicate the lane where the vehicle is located; Determining a successfully matched second matching pair according to the first lane information and the second lane information included in the third road structure information, where the second matching pair is used to indicate the matching relationship between the second road structure information and the third road structure information; Deleting abnormal data in the second matching pair to determine the first matching pair.

10. A vehicle positioning device, characterized in that, Includes: An acquisition unit for obtaining the first road structure information and the corresponding recognition accuracy through a vision sensor; A determination unit for determining the first pose when the vision sensor obtains the first road structure information; The acquisition unit is further configured to obtain a target map according to the first pose, where the target map is a local map corresponding to the first pose and includes second road structure information; The determining unit is further configured to perform visual environment modeling on the reference information to determine third road structure information. The reference information includes the first road structure information and the second road structure information. Match the second road structure information and the third road structure information to determine a first matching pair. The first matching pair includes the successfully matched second road structure information and the target road structure information. And determine the target pose of the vehicle in the target map according to the first pose, the first matching pair, and the recognition accuracy of the target road structure information.

11. The vehicle positioning device according to claim 10, characterized in that, The obtaining unit is further configured to obtain a historical image according to the first pose. The historical image includes an image obtained at the position corresponding to the first pose, and the historical image includes fourth road structure information. The reference information further includes the fourth road structure information.

12. The vehicle positioning device according to claim 10 or 11, characterized in that, The determining unit is specifically configured to: Determine a first coordinate set of the target road structure information in the coordinate system of the visual sensor. The first coordinate set includes the coordinates of a plurality of first discrete points. Determine a second coordinate set of the second road structure information in the first matching pair in the coordinate system of the visual sensor. The second coordinate set includes the coordinates of a plurality of second discrete points. Determine a target distance between the target road structure information and the corresponding second road structure information according to the first coordinate set and the second coordinate set. Determine a second pose according to the recognition accuracy and the target distance. Adopt a preset iterative rule to determine the target pose according to the first pose and the second pose.

13. The vehicle positioning device according to claim 12, wherein The determining unit is specifically configured to: The determining the first coordinate set of the target road structure information in the coordinate system of the visual sensor includes: Determine a third coordinate set of the target road structure information in the coordinate system of the visual sensor. If the third coordinate set includes the coordinates of a plurality of third discrete points, use the coordinates of the plurality of third discrete points as the coordinates of the first discrete points. If the third coordinate set includes the coordinates of line information, perform discrete processing on the line information to obtain the coordinates of the first discrete points corresponding to the line information. The line information is any one of lane boundary information, road boundary information, or lamp post information.

14. The vehicle positioning device according to claim 13, characterized in that, The determining unit is specifically configured to: Determine a fourth coordinate set of the second road structure information in the first matching pair in the coordinate system of the target map. Convert the fourth coordinate set into the second coordinate set in the coordinate system of the visual sensor.

15. The vehicle positioning device according to claim 12, wherein The determining unit is specifically configured to: Determine the weight corresponding to each target road structure information according to the recognition accuracy of each target road structure information and the sum of the recognition accuracies of all target road structure information. Determine the second pose according to the weight corresponding to each target road structure information and the target distance between each target road structure information and the corresponding second road structure information.

16. The vehicle positioning device according to claim 10 or 11, characterized in that, The obtaining unit is specifically configured to: Obtain a captured image through the visual sensor. Performing image recognition processing on the captured image through the visual sensor to obtain at least one piece of the first road structure information included in the captured image and the recognition accuracy corresponding to each piece of the first road structure information.

17. The vehicle positioning device according to claim 10 or 11, characterized in that, The determining unit is specifically configured to: Determine the first pose according to the third pose, the first moment, the second moment, and the motion information between the first moment and the second moment, where the first moment is the moment when the vehicle is in the third pose, and the second moment is the moment when the visual sensor acquires the first road structure information.

18. The vehicle positioning device according to claim 10 or 11, characterized in that, The second road structure information includes M first lane lines, and the third road structure information includes N second lane lines, where N and M are positive integers; The determining unit is specifically configured to: Determine the first lane information included in the second road structure information according to the first pose, the positioning accuracy, and the second road structure information, where the lane information is used to indicate the lane where the vehicle is located; Determine a second matching pair that matches successfully according to the first lane information and the second lane information included in the third road structure information, where the second matching pair is used to indicate the matching relationship between the second road structure information and the third road structure information; Delete abnormal data in the second matching pair and determine the first matching pair.

19. A vehicle positioning device, characterized in that, The vehicle positioning device includes: a processor and a memory; the memory is used to store computer program code, and the computer program code includes computer instructions; when the processor executes the computer instructions, the vehicle positioning device executes the vehicle positioning method according to any one of claims 1-9.

20. A vehicle, characterized in that, The vehicle includes: a processor and a memory; the memory is used to store computer program code, and the computer program code includes computer instructions; when the processor executes the computer instructions, the vehicle executes the vehicle positioning method according to any one of claims 1-9.

21. A computer-readable storage medium, characterized in that, Including computer instructions, when the computer instructions run on the vehicle positioning device, enabling the vehicle positioning device to execute the vehicle positioning method according to any one of claims 1-9; or when the computer instructions run on the vehicle, enabling the vehicle to execute the vehicle positioning method according to any one of claims 1-9.

Images