Vehicle positioning method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a vehicle positioning method and device, a storage medium and electronic equipment, and the method comprises the steps: recognizing a lane line in a road image, and obtaining a lane line sensing result; the lane line sensing result comprises a first left-right topological structure of a plurality of sensing lane lines sensed by the vehicle at the current moment and a first virtual-real attribute of each sensing lane line; under the condition that the lane where the vehicle is located at the previous moment of the current moment is not determined, real lane line information of the vehicle is obtained from a map; the real lane line information comprises a second left-right topological structure of a plurality of real lane lines in the domain range of the current position where the vehicle is located, and a second virtual-real attribute of each real lane line; determining the current lane of the vehicle under the condition that the lane line sensing result is successfully matched with the real lane line information; according to the distance error of the lane where the vehicle is located currently, the current pose of the vehicle is adjusted, and the problem of vehicle positioning is solved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of autonomous driving technology, and more specifically, to a vehicle positioning method, device, storage medium, and electronic device. Background Art

[0002] With the rapid development of artificial intelligence technology, autonomous driving systems have become an important research direction in the automotive industry. The navigation and positioning system of autonomous driving vehicles is one of its core components and must meet strict requirements such as high precision, high resolution, high reliability, low cost, low power consumption, and miniaturization to ensure the safe operation of vehicles in complex environments.

[0003] At present, the GNSS / INS combined navigation system is the most widely used positioning method in the field of autonomous driving. It provides stable and reliable positioning services by integrating the global coverage, all-weather operation and high positioning accuracy of the satellite navigation system (GNSS) and the short-term high accuracy and immediacy of the inertial navigation system (INS). However, in environments where satellite signals are blocked, such as urban canyons, tunnels, and underground garages, the positioning accuracy will drop significantly or even fail completely, and high-precision positioning cannot be obtained. Summary of the invention

[0004] The embodiments of the present application provide a vehicle positioning method, device, storage medium and electronic device to at least solve the problem of low positioning accuracy of autonomous driving vehicles in complex environments due to obstruction of GNSS / INS signals in related technologies.

[0005] According to one aspect of an embodiment of the present application, a vehicle positioning method is provided, comprising: acquiring a road image collected by the vehicle at a current moment, and identifying lane lines in the road image to obtain a lane line perception result; the lane line perception result includes a first left-right topological structure of multiple perceived lane lines perceived by the vehicle at the current moment, and a first virtual-real attribute of each perceived lane line in the multiple perceived lane lines; when the lane where the vehicle was located at the previous moment at the current moment is not determined, obtaining real lane line information within the domain of the current position of the vehicle from a map; the real lane line information includes the real lane line information within the domain of the current position of the vehicle a second left-right topological structure of multiple real lane lines of the vehicle, and a second virtual-real attribute of each of the multiple real lane lines; matching the lane line perception result with the real lane line information to obtain a matching result; when the matching result indicates that the lane line perception result and the real lane line information are successfully matched, determining the lane in which the vehicle is currently located according to the second left-right topological structure; obtaining the current posture of the vehicle, and adjusting the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

[0006] According to another aspect of an embodiment of the present application, a vehicle positioning device is provided, comprising: an image acquisition module, for acquiring a road image collected by the vehicle at a current moment, and identifying lane lines in the road image to obtain a lane line perception result; the lane line perception result includes a first left-right topological structure of multiple perceived lane lines perceived by the vehicle at the current moment, and a first virtual-real attribute of each perceived lane line in the multiple perceived lane lines; a map acquisition module, for acquiring real lane line information within the domain of the current position of the vehicle from a map when the lane in which the vehicle was located at a moment before the current moment is not determined; the real lane line information includes multiple real lane line information within the domain of the current position of the vehicle. A second left-right topological structure of a real lane line, and a second virtual-real attribute of each of the multiple real lane lines; a matching module, used to match the lane line perception result with the real lane line information to obtain a matching result; a lane determination module, used to determine the lane in which the vehicle is currently located according to the second left-right topological structure when the matching result indicates that the lane line perception result and the real lane line information are successfully matched; a posture adjustment module, used to obtain the current posture of the vehicle, and adjust the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

[0007] According to another aspect of the embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned vehicle positioning method when running.

[0008] According to another aspect of an embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the vehicle positioning method through the computer program.

[0009] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in any of the above vehicle positioning method embodiments are implemented.

[0010] Through the embodiment of the present application, the vehicle is used to collect road images in real time and identify the lane lines therein, and the lane line perception results including the "first left-right topological structure" and the "first virtual-real attribute" are obtained. This step provides a visual positioning reference even when the satellite signal is unreliable, ensuring the continuity of the vehicle's positioning in a complex environment; then, the real lane line information of the "second left-right topological structure" and the "second virtual-real attribute" within the current position of the vehicle is obtained from the high-precision map, and the lane line perception results are matched with the real lane line information. When the match is successful, that is, the "first left-right topological structure" is consistent with the "second left-right topological structure" and the "first virtual-real attribute" is consistent with the "second virtual-real attribute", the lane where the vehicle is located is determined, and the positioning accuracy is significantly improved; finally, based on the matching results, by calculating the distance error between the perceived lane line and the real lane line of the lane where the vehicle is currently located, the current posture of the vehicle is adjusted to minimize the positioning error. This dynamic adjustment step not only further improves the positioning accuracy, but also ensures the driving safety and stability of the vehicle during the automatic driving process. In summary, through the coordinated work of the above steps, it is possible to provide stable and accurate vehicle positioning services in complex environments where satellite signals are blocked, solving the problem of low positioning accuracy of autonomous driving vehicles in complex environments due to blocked GNSS / INS signals in related technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a hardware structure block diagram of an optional vehicle positioning method in an embodiment of the present application;

[0012] Figure 2 is a flow chart of an optional vehicle positioning method according to an embodiment of the present application;

[0013] Figure 3is a schematic diagram of an optional first left-right topology structure according to an embodiment of the present application;

[0014] Figure 4 is a flow chart of an optional vehicle positioning according to an embodiment of the present application;

[0015] Figure 5 is an optional schematic diagram of a vehicle lane change according to an embodiment of the present application;

[0016] Figure 6 is a structural block diagram of an optional vehicle positioning device according to an embodiment of the present application;

[0017] Figure 7 It is a structural block diagram of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0019] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0020] According to one aspect of an embodiment of the present application, a vehicle positioning method is provided. Optionally, in this embodiment, the vehicle positioning method can be applied to Figure 1 In the hardware environment composed of the vehicle terminal 102 and the server 104 shown in FIG. Figure 1As shown, the server 104 is connected to the vehicle terminal 102 via a network, and can be used to provide services (such as application services, etc.) for the terminal or a client installed on the terminal. A database can be set on the server or independently of the server to provide data storage services for the server 104. Cloud computing and / or edge computing services can be configured on the server or independently of the server to provide data computing services for the server 104.

[0021] The above network may include but is not limited to at least one of the following: wired network, wireless network. The above wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the above wireless network may include but is not limited to at least one of the following: WIFI, Bluetooth.

[0022] The vehicle positioning method of the embodiment of the present application can be executed by the vehicle terminal 102, or can be executed by the server 104 and the vehicle terminal 102 together. Taking the vehicle positioning method of the embodiment executed by the vehicle terminal 104 as an example, Figure 2 is a flow chart of an optional vehicle positioning method according to an embodiment of the present application, such as Figure 2 As shown, the process of the method may include the following steps:

[0023] Step S202, obtain the road image collected by the vehicle at the current moment, and identify the lane lines in the road image to obtain the lane line perception result; the lane line perception result includes the first left and right topological structure of multiple perceived lane lines perceived by the vehicle at the current moment, and the first virtual and real attributes of each perceived lane line in the multiple perceived lane lines.

[0024] The road image refers to the image data collected by the camera or other imaging sensors on the vehicle on the current driving road. The road image includes all visible information such as the physical structure of the road, lane lines, obstacles, traffic signs, etc.

[0025] Lane lines refer to the markings on the road used to divide lanes. They are usually white or yellow and can be solid or dotted. They are used to guide vehicles and keep them safe in the lane. Perceived lane lines refer to lane lines that are identified after analyzing the road image through the vehicle's visual sensors (such as cameras).

[0026] The first left-right topology refers to the lane layout perceived by the vehicle through its visual sensor (such as a camera) at the current moment. It is used to describe the left-right relative relationship between multiple perceived lanes within the vehicle's current field of view, that is, the layout structure of the lanes relative to the vehicle's current position. Through the first left-right topology, it is possible to determine which lane the vehicle is traveling on and the relative positions between the lanes.

[0027] For example, Figure 3FIG. 1 is a schematic diagram of a first left-right topological structure in an embodiment. Figure 3 As shown in the figure, assume that the vehicle is driving on a three-lane road and is located in the middle lane. The vehicle's camera is used to capture the image of the current road, and the following information is obtained through lane line recognition technology: a solid line is sensed on the left side of the vehicle, and this lane line is marked as L1, and it is determined to be a solid line (Solid); on the left side of L1, the vehicle also senses a solid line, marked as L2, which is also a solid line (Solid); a dashed line is sensed on the right side of the vehicle, and this lane line is marked as R1, and it is determined to be a dashed line (Dashed); on the right side of R1, the vehicle also senses a solid line, marked as R2, which is also a solid line (Solid). Based on this information, the first left-right topology can be constructed:

[0028] L1 (solid lane line on the left): The first lane line on the left side of the vehicle, which is on the left relative to the current position of the vehicle and is a solid line.

[0029] L2 (solid lane line on the left): The lane line on the left side of the L1 lane line is on the left relative to the current position of the vehicle. The attribute is a solid line, which indicates that there are two lane lines on the left side of the vehicle.

[0030] R1 (right dashed lane line): The first lane line on the right side of the vehicle, which is on the right relative to the current position of the vehicle and is a dashed line.

[0031] R2 (solid lane line on the right): The lane line on the right side of the R1 lane line is on the right relative to the current position of the vehicle. The attribute is a solid line, which indicates that there are two lane lines on the right side of the vehicle.

[0032] In some embodiments, the first topological structure can be represented as a sequence or matrix, where each element represents a lane line and its attributes and relative position. For example, it can be simplified as: left (L1) - solid line, left-left (L2) - solid line, right (R1) - dotted line, right-right (R2) - solid line.

[0033] Among them, the first left-right topology helps the autonomous driving system understand and determine the position and direction of the vehicle on the road, especially in an environment with no or weak GNSS signals, where visual information becomes the key basis for positioning. By comparing the first left-right topology at different times, it can also be determined whether the vehicle has changed lanes or adjusted its driving direction.

[0034] The first virtual-real attribute refers to the characteristics of the perceived lane line, distinguishing whether the lane line is a solid line or a dotted line. Among them, the first virtual-real attribute is usually actively combined with the road image captured by the visual sensor (such as a camera) on the vehicle, and is realized through image processing and machine learning algorithms. For example, a lane line detection algorithm such as Hough transform, Canny edge detection, template matching or a deep learning model (such as a convolutional neural network CNN) is used to identify the position of the lane line from the preprocessed road image, wherein these algorithms can detect the boundary of the lane line, including the shape of a solid line or a dotted line; for the detected lane line, its attributes are further analyzed by the following method to determine whether the lane line is a solid line or a dotted line:

[0035] 1) Classification based on deep learning: Train a deep learning model, such as using a classifier or semantic segmentation model, to distinguish between solid and dashed lines by learning the visual features of lane lines, such as line continuity, line segment length, and line density.

[0036] 2) Classification based on geometric features: Analyze the geometric features of lane lines, such as the length and continuity of line segments. Solid lines are usually represented as continuous line segments, while dashed lines are represented as discontinuous line segments.

[0037] 3) Based on time series analysis: The attributes of lane lines are judged by their frequency and duration in consecutive frames. Solid lines should persist in consecutive frames, while dashed lines will disappear and appear intermittently.

[0038] Optionally, the vehicle obtains the road image collected by the vehicle at the current moment through a camera installed on the vehicle, and transmits the road image to the vehicle terminal. The vehicle terminal uses image processing and machine learning technology, such as deep learning convolutional neural network (CNN) or semantic segmentation model, to analyze the road image, identify the left-right relative relationship between multiple perceived lane lines within the vehicle's current field of view and the first virtual-real attribute of each perceived lane line, and arrange the perceived multiple perceived lane lines in order from left to right or from right to left, to obtain the first left-right topological structure of the multiple perceived lane lines perceived by the vehicle at the current moment.

[0039] Step S204, when the lane that the vehicle was in at the previous moment before the current moment is not determined, obtain the real lane line information within the scope of the current position of the vehicle from the map; the real lane line information includes a second left-right topological structure of multiple real lane lines within the scope of the current position of the vehicle, and a second virtual-real attribute of each of the multiple real lane lines.

[0040] Among them, the current position of the vehicle refers to the coordinate position of the vehicle in geographic space, which is usually obtained through a GPS or GNSS system. However, in an environment with poor signals, other technologies may be needed to assist in determining the position.

[0041] The scope of the vehicle's current position refers to an area surrounding the vehicle's current position, which usually covers all parts of the road near the vehicle that can affect its driving decisions. It is used to obtain detailed road information and lane line data.

[0042] Real lane line information refers to lane line information obtained from map data, including the precise location, topological structure, and virtual and real attributes of lane lines. Real lane line information is used to compare and verify with the information captured in real time by vehicle sensors (i.e., lane line perception results).

[0043] The second left-right topology structure refers to the left-right layout relationship of the real lane lines described in the map database relative to the current position of the vehicle, which provides a reference coordinate system for the autonomous driving system. The representation and example of the second left-right topology structure are similar to those of the first left-right topology structure and will not be repeated here.

[0044] The second virtual-real attribute refers to the real lane line attribute recorded in the map database, including whether the lane line is a solid line or a dotted line, which is used to correct or confirm the attributes of the lane line identified by the vehicle's visual sensor. Unlike the first virtual-real attribute, the second virtual-real attribute is obtained through data entry and is used to reflect the real virtual-real attribute of the real lane line.

[0045] Optionally, when the lane in which the vehicle was located at the previous moment before the current moment is not determined, the vehicle terminal extracts lane line information in the current position area from the high-precision map database, including the precise coordinates, topological structure and virtual-real attributes of the lane line, and based on the extracted lane line information, generates a second left-right topological structure of multiple real lane lines within the area of ​​the current position of the vehicle, as well as a second virtual-real attribute of each of the multiple real lane lines.

[0046] Step S206, matching the lane line perception result with the real lane line information to obtain a matching result;

[0047] Among them, the matching result refers to the result obtained after comparing and analyzing the lane line perception result with the real lane line information, which is used to determine whether the lane line perception result of the vehicle at the current moment is consistent with the real lane line information in the map. The matching result is mainly based on the comparison of two key aspects: the topological structure of the lane line and the virtual-real attribute of the lane line. Among them, if the first left-right topological structure in the lane line perception result is consistent with the second left-right topological structure in the real lane line information in the map, that is, the lane line layout in the vehicle's field of view is the same as the lane line layout at that location on the map, then the topological structure of the lane line is successfully matched; if the first virtual-real attribute of each perceived lane line matches the second virtual-real attribute of the corresponding real lane line in the map, then the virtual-real attribute of the lane line is successfully matched. When the lane line layout in the vehicle's field of view (first left-right topological structure) is consistent with the lane line layout at that location in the map (second left-right topological structure), and the first virtual-real attribute of each lane line identified by the vehicle is consistent with the second virtual-real attribute of the corresponding lane line on the map, the matching result is characterized as "matching success".

[0048] Optionally, the vehicle terminal performs structured processing on the lane line perception results and the real lane line information obtained in step S202 and step S204 to form a comparable model, which includes the topological structure and virtual and real attributes of the lane line. By comparing the topological structure (left-right relative relationship) and virtual and real attributes of the lane line perception results and the real lane line information, a match between the two is achieved to obtain a matching result.

[0049] Step S208, when the matching result indicates that the lane line perception result and the actual lane line information are successfully matched, the lane in which the vehicle is currently located is determined according to the second left-right topological structure.

[0050] Optionally, when the matching result represents that the lane line perception result and the real lane line information match successfully, the vehicle terminal determines the lane the vehicle is currently in based on the second left-right topological structure within the domain of the current position of the vehicle obtained from the map. When the matching result represents that the lane line perception result and the real lane line information fail to match, the vehicle terminal determines that it may be caused by the misdetection of the first virtual-real attribute of the perceived lane line. Therefore, in this case, the vehicle terminal does not process it.

[0051] Step S210, obtaining the current posture of the vehicle, and adjusting the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located, to obtain an adjusted current posture of the vehicle.

[0052] The current posture of the vehicle includes information such as the position, direction and posture of the vehicle. Generally, the current posture of the vehicle is obtained by interpolating the original posture output by the combined inertial navigation according to the timestamp of the lane line perception result obtained at the current moment.

[0053] Distance error refers to the position deviation between the perceived lane line and the actual lane line. Distance error is a key indicator for evaluating vehicle positioning accuracy. The smaller the distance error, the more accurate the positioning.

[0054] Optionally, the vehicle terminal obtains the current posture of the vehicle, compares the position difference between the perceived lane line of the vehicle in the current lane and the actual lane line of the vehicle in the current lane in the map, calculates the distance error between the two, and uses nonlinear least squares method or other optimization algorithm to iteratively update the current posture of the vehicle according to the distance error to obtain the final adjusted posture, which will be used for vehicle navigation and control.

[0055] In some embodiments, obtaining the current posture of the vehicle includes the following steps: when the combined inertial navigation (such as GPS or GNSS system) signal is poor, obtaining the original posture most recently output by the combined inertial navigation; determining the time range between the current moment of the vehicle and the timestamp of the original posture, and interpolating the posture corresponding to each timestamp within the time range according to the original posture to obtain the current posture of the vehicle; when the combined inertial navigation (such as GPS or GNSS system) signal is good, obtaining the current posture of the vehicle through the combined inertial navigation.

[0056] Through the above steps S202 to S210, the vehicle is used to collect road images in real time and identify the lane lines therein, and a lane line perception result including a "first left-right topological structure" and a "first virtual-real attribute" is obtained. This step provides a positioning reference based on vision even when the satellite signal is unreliable, ensuring the continuity of the vehicle's positioning in a complex environment; then, the real lane line information of the "second left-right topological structure" and the "second virtual-real attribute" within the current position of the vehicle is obtained from the high-precision map, and the lane line perception result is matched with the real lane line information. When the match is successful, that is, the "first left-right topological structure" is consistent with the "second left-right topological structure" and the "first virtual-real attribute" is consistent with the "second virtual-real attribute", the lane where the vehicle is located is determined, and the positioning accuracy is significantly improved; finally, based on the matching result, by calculating the distance error between the perceived lane line and the real lane line of the lane where the vehicle is currently located, the current posture of the vehicle is adjusted to minimize the positioning error. This dynamic adjustment step not only further improves the positioning accuracy, but also ensures the driving safety and stability of the vehicle during the automatic driving process. In summary, through the coordinated work of the above steps, the embodiment of the present application can provide stable and accurate vehicle positioning services in complex environments where satellite signals are blocked, solving the problem of low positioning accuracy of autonomous driving vehicles in complex environments due to blocked GNSS / INS signals in related technologies.

[0057] In an exemplary embodiment, the first left-right topological structure includes a left-right relative relationship between the current position of the vehicle and each perceived lane line, and the second left-right topological structure includes a left-right relative relationship between the current position of the vehicle and each real lane line.

[0058] The left-right relative relationship in the first left-right topological structure is based on the lane line information perceived by the vehicle through the camera at the current moment. The left-right relative relationship in the first left-right topological structure describes the left-right position of each perceived lane line relative to the center line of the vehicle. For example, the lane line on the left side of the vehicle may be identified as the first lane line on the left, and the lane line on the right side of the vehicle may be identified as the first lane line on the right.

[0059] The left-right relative relationship in the second left-right topology structure is a description of the real lane line information within the current position of the vehicle obtained from the map data. The left-right relative relationship in the second left-right topology structure also describes the left-right position of each real lane line relative to the current position of the vehicle. For example, the map information may indicate that there are two lanes on the left side of the vehicle and one lane on the right side.

[0060] In some embodiments, the lane line perception result and the real lane line information are matched to obtain a matching result, including:

[0061] 1. When the lane line perception result and the real lane line information also include a road boundary label, a target perceived lane line marked with a road boundary label is extracted from multiple perceived lane lines, and a target real lane line marked with a road boundary label is extracted from multiple real lane lines.

[0062] Among them, the road boundary label is special information marked for the edge of the road or the boundary lane line, which is used to distinguish the edge of the road from the lane line inside the road. In the lane line perception results perceived by the vehicle, the road boundary label is usually stored together with the coordinates of the perceived lane line, the first left and right topological structure, and the first virtual and real attributes. In the high-precision map, the road boundary label is usually stored together with the coordinates of the real lane line, the second left and right topological structure, and the second virtual and real attributes, and provided to the autonomous driving system. The road boundary is a strong constraint relationship for the lane line. According to the ID of the lane line where the road boundary is located, the current lane of the vehicle can be reversely calculated. For example, if "right right - curb" is perceived, and the rightmost lane line in the map is also a curb, then it can be inferred that the current vehicle is located in the second lane line from the right, and the lane line ID can also be directly determined.

[0063] It should be noted that road boundaries, as fixed reference points in the road structure, have obvious features and are easily captured by cameras. Using target lane lines marked with road boundary labels can significantly enhance the matching accuracy between lane line perception results and real lane line information in map data. This is because the recognition of road boundaries is usually more stable than that of lane lines and is not easily affected by environmental factors such as light and weather. Therefore, as a strong constraint, it can reduce uncertainty in the matching process and improve the reliability of matching results.

[0064] Target-perceived lane lines refer to lane lines that are closest to the vehicle's current position or related to the road boundary, selected from multiple perceived lane lines in the vehicle's current field of view through road boundary labels. Target-perceived lane lines are key objects for lane line information matching.

[0065] The target real lane line refers to the real lane line information extracted from the map database, which corresponds to the position of the target perceived lane line and has the road boundary label. It contains the precise coordinates, topological relationship, virtual and real attributes of the target lane line, and is used for comparison and verification with the perceived lane line information.

[0066] The first left-right relative relationship describes the left-right orientation relationship between the vehicle's current position and the target perceived lane line, that is, the target perceived lane line is relative to the left or right side of the vehicle. This information helps determine the relative position of the vehicle on the road.

[0067] The second left-right relative relationship describes the left-right azimuth relationship between the vehicle's current position and the target real lane line. It is also used to determine the relative position of the vehicle on the map and is an important reference for verifying positioning accuracy.

[0068] Optionally, Figure 4 FIG. 1 is a flow chart of vehicle positioning in one embodiment, as shown in FIG. Figure 4 As shown, when the lane line perception results and the real lane line information also include road boundary labels, the vehicle terminal detects lane lines in the road image through an image processing algorithm to identify the target perceived lane lines marked as road boundaries, and searches and extracts the target real lane lines marked as road boundaries from the high-precision map data according to the current position of the vehicle.

[0069] 2. Compare the first left-right relative relationship between the current position of the vehicle and the target perceived lane line with the second left-right relative relationship between the current position of the vehicle and the target real lane line, and compare the first virtual-real attribute of the target perceived lane line with the second virtual-real attribute of the target real lane line.

[0070] Alternatively, if Figure 4 As shown, the vehicle terminal compares the first left-right relative relationship between the vehicle's current position and the target perceived lane line with the second left-right relative relationship between the vehicle's current position and the target real lane line to check whether the two are consistent. At the same time, the first virtual-real attribute of the target perceived lane line is compared with the second virtual-real attribute of the target real lane line to confirm whether the two match.

[0071] 3. When the first left-right relative relationship matches the second left-right relative relationship, and the first virtual-real attribute of the target perceived lane line matches the second virtual-real attribute of the target real lane line, it is determined that the matching result represents a successful match between the lane line perception result and the real lane line information.

[0072] Alternatively, if Figure 4 As shown, if the first left-right relative relationship and the second left-right relative relationship match, and the first virtual-real attribute of the target perceived lane line matches the second virtual-real attribute of the target real lane line, then the vehicle terminal considers that the lane line perception result matches the real lane line information successfully. If the first left-right relative relationship and the second left-right relative relationship do not match, and / or the first virtual-real attribute of the target perceived lane line does not match the second virtual-real attribute of the target real lane line, then the vehicle terminal does not process it.

[0073] Through this embodiment, the road boundary label is used as a strong constraint condition to extract the target perceived lane line and the target real lane line from the perceived lane line and the real lane line for matching. Since the road boundary has stable characteristics and a clear position, it is a strong constraint information in the road structure. By using the road boundary label, it can ensure that the recognition of the target lane line and the corresponding relationship in the map data are more accurate, thereby improving the accuracy of lane line matching. This matching strategy based on strong constraints effectively reduces the interference of environmental factors on lane line recognition and enhances the reliability of the matching process. Even under complex conditions such as poor lighting and blurred lane lines, lane matching can be performed stably. Furthermore, only the left-right relative relationship and virtual-real attributes between the target perceived lane line marked by the road boundary label and the target real lane line are compared, rather than full-map matching, which significantly reduces the amount of calculation and reduces the demand for computing resources.

[0074] In an exemplary embodiment, matching the lane line perception result with the real lane line information to obtain a matching result further includes:

[0075] 1. When at least one of the lane line perception result and the real lane line information does not include a road boundary label, an augmented matrix is ​​constructed based on the lane line perception result and the real lane line information; each row of the augmented matrix represents each perceived lane line, each column of the augmented matrix represents each real lane line, and an element of the augmented matrix represents a matching degree between a first virtual-real attribute of the perceived lane line represented by the corresponding row and a second virtual-real attribute of the real lane line represented by the corresponding column.

[0076] Among them, if the lane line perception result or the actual lane line information does not contain the road boundary label, it means that the vehicle terminal cannot directly simplify the matching process through the boundary information, but needs to adopt a more complex matching strategy.

[0077] The calculation of the matching degree can be based on a variety of factors. For example, if the lane line attributes are exactly the same, the matching degree is high, and if they are different, the matching degree is low. Specifically, it can be represented by a real number between 0 and 1, where 1 represents a complete match and 0 represents a complete mismatch.

[0078] Alternatively, if Figure 4As shown, when at least one of the lane line perception result and the real lane line information does not include a road boundary label, the vehicle terminal determines each perceived lane line in the lane line perception result and each real lane line in the real lane line information, and each perceived lane line and real lane line will be assigned a unique row and column index; the vehicle terminal calculates the matching degree between them according to the first virtual-real attribute of the perceived lane line (such as a real line or a dotted line) and the second virtual-real attribute of the real lane line. The vehicle terminal constructs an augmented matrix, in which each element represents the matching degree of the corresponding perceived lane line and the real lane line, the row of the matrix represents the perceived lane line, the column represents the real lane line, and the value of the matrix element represents the matching degree.

[0079] Second, an optimization algorithm is used to solve the augmented matrix. When there is only one set of solutions, the matching result is determined to represent the successful match between the lane line perception result and the actual lane line information.

[0080] Among them, the optimization algorithm is used to solve the augmented matrix and find the optimal matching solution between the perceived lane line and the real lane line, so that the total matching degree or similarity of the pairing is optimal. Common optimization algorithms include the Hungarian algorithm (applicable to solving the maximum matching problem of bipartite graphs), linear programming, dynamic programming, genetic algorithm, etc. In the process of solving the optimization algorithm, if the augmented matrix has only one set of solutions (that is, there is a unique way to make the matching degree of all lane line perception results and the real lane line information optimal), then this set of solutions indicates that the lane line perception results and the real lane line information are matched successfully. In this case, the match between the lane line perception results and the map data is the most accurate, and there is no uncertainty caused by multiple solutions. Once the optimization algorithm confirms that there is only one set of solutions, the matching result between the lane line perception results and the real lane line information is determined to represent a successful match. This means that each perceived lane line can find a corresponding real lane line in the map, and the matching degree of this set of correspondences is the highest. Through this process, even in the absence of road boundary labels, the virtual and real attributes and augmented matrix can be used to find the optimal match between the perceived lane line and the real lane line, thereby ensuring the accuracy and reliability of vehicle positioning.

[0081] Alternatively, if Figure 4 As shown, the vehicle terminal selects a suitable optimization algorithm, such as the Hungarian algorithm or the linear programming algorithm, and finds all possible matching combinations in the augmented matrix through the optimization algorithm, evaluates the total matching degree, and when there is only one matching combination with the largest total matching degree, determines the matching combination with the largest total matching degree as the optimal matching solution, and determines that the matching result represents the successful match between the lane line perception result and the actual lane line information.

[0082] Through this embodiment, the process of solving the augmented matrix using an optimization algorithm is intended to find the best matching solution between the perceived lane line and the real lane line through a mathematical optimization method. In the case of only one set of solutions, it can be determined that the match between the lane line perception result and the real lane line information is successful, thereby ensuring the accuracy and reliability of vehicle positioning. Even in an environment without road boundary labels, high-precision vehicle positioning can be achieved through the comparison of virtual and real attributes.

[0083] In an exemplary embodiment, the lane line perception result also includes a first identifier for each perceived lane line, and the real lane line information also includes a second identifier for each real lane line. The first identifier of each perceived lane line is used to mark the perceived lane line within the current field of view of the vehicle; the second identifier of each real lane line is used to mark each real lane line.

[0084] The first identifier refers to a unique number or mark generated in real time by the vehicle based on the road image collected at the current moment, which is used to identify the lane line perceived within the vehicle's field of view. The first identifier helps the vehicle track the changes in the lane line, thereby determining whether the vehicle changes lanes or stays in the same lane.

[0085] The second identifier refers to the unique number or mark assigned to each real lane line in the high-precision map, which is used to distinguish and identify different lane lines in the map database. The second identifier is static, reflecting the position and attributes of the real lane line in the map data, and is not affected by changes in the external environment. The second identifier is included in the real lane line information, which helps the system determine the precise lane position of the vehicle on the map.

[0086] The perceived lane lines within the current field of view of the vehicle refer to the lane line information identified and extracted from the current road image collected by the visual sensor (such as a camera) carried by the vehicle. The perceived lane lines within the current field of view of the vehicle are directly observed by the vehicle, including their positions, first virtual and real attributes, and first identifiers.

[0087] In some embodiments, the vehicle positioning method further includes:

[0088] 1. When the lane that the vehicle is in at the last moment before the current moment has been determined, the first identifier of the lane line perceived by the vehicle at the current moment is compared with the first identifier of the lane line perceived by the vehicle at the last moment before the current moment.

[0089] Alternatively, if Figure 4As shown, when the lane in which the vehicle is located at the previous moment has been determined, the vehicle terminal identifies the lane line from the road image collected at the current moment, obtains the lane line perception result including the first identifier, and compares the first identifier of the lane line perceived at the current moment with the first identifier recorded by the vehicle at the previous moment.

[0090] 2. When the first identifier of the lane line perceived by the vehicle at the current moment is different from the first identifier of the lane line perceived by the vehicle at the previous moment before the current moment, the lane change direction of the vehicle is determined according to the first left-right topological structure.

[0091] Among them, if the first identifier of the lane line perceived at the current moment is consistent with the first identifier recorded by the vehicle at the previous moment before the current moment, it is determined that the vehicle may remain in the same lane; if the first identifier of the lane line perceived at the current moment is inconsistent with the first identifier recorded by the vehicle at the previous moment before the current moment, it is determined that the vehicle has changed lanes.

[0092] For example, Figure 5 FIG. 1 is a schematic diagram of a vehicle changing lanes in an embodiment. Figure 5 As shown in the figure, suppose an autonomous vehicle is driving in the middle lane of a city street. There are three lanes on the street. The camera in front of the vehicle can collect road images and identify lane lines in real time. At a certain point in time, the lane line recognition system of the vehicle recognizes three lane lines, namely the left lane line, the right lane line and the rightmost lane line, and assigns them different first identifiers. For example, the left lane line recognized by the system is marked as L1, the right lane line is marked as R1, and the rightmost lane line is marked as R2. When the vehicle changes lanes from the middle lane to the right lane, R1, which was originally marked as the right lane line, will become the first identifier L1 of the left lane line perceived by the vehicle, and R2, which was originally marked as the rightmost lane line, will become the first identifier R1 of the perceived right lane line. By comparing the first identifiers of the vehicle before and after the lane change, it can be determined whether the vehicle has changed lanes and the lane change direction is to the right.

[0093] Optionally, when the first identifier of the lane line perceived by the vehicle at the current moment is different from the first identifier of the lane line perceived by the vehicle at the moment before the current moment, the vehicle terminal determines the position change of the lane line relative to the vehicle (such as from the left lane line to the right lane line) based on the first left-right topology structure collected at the current moment, thereby determining the vehicle's lane change direction (changing lanes to the left or to the right).

[0094] 3. Determine the current lane of the vehicle based on the second left-right topological structure and the lane change direction.

[0095] Optionally, the vehicle terminal determines the current lane of the vehicle based on the determined lane change direction and the second left-right topological structure in the real lane line information corresponding to the current moment.

[0096] 4. Obtain the current posture of the vehicle, and adjust the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the actual lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

[0097] Optionally, the vehicle terminal obtains the current posture of the vehicle, compares the position difference between the perceived lane line of the vehicle in the current lane and the actual lane line of the vehicle in the current lane in the map, calculates the distance error between the two, and uses nonlinear least squares method or other optimization algorithm to iteratively update the current posture of the vehicle according to the distance error to obtain the final adjusted posture, which will be used for vehicle navigation and control.

[0098] Through this embodiment, when the lane in which the vehicle is located at the previous moment before the current moment has been determined, the first identifier of the perceived lane line perceived at the adjacent moment is compared in real time, so it can be immediately identified whether the vehicle has changed lanes. The introduction of the first identifier of the perceived lane line can track the vehicle's driving trajectory more accurately, ensuring that the vehicle can accurately identify the new lane when changing lanes, avoiding safety hazards caused by positioning deviations.

[0099] In an exemplary embodiment, the vehicle positioning method further includes:

[0100] 1. When the first identifier of the lane line perceived by the vehicle at the current moment is the same as the first identifier of the lane line perceived by the vehicle at the moment before the current moment, the lane in which the vehicle was located at the moment before the current moment is determined as the lane in which the vehicle is currently located.

[0101] Optionally, when the first identifier of the lane line perceived by the vehicle at the current moment is the same as the first identifier of the lane line perceived by the vehicle at the previous moment, the vehicle terminal determines that the vehicle has not changed lanes and directly determines the lane in which the vehicle was at the previous moment as the lane in which the vehicle is currently located.

[0102] 2. Obtain the current posture of the vehicle, and adjust the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the actual lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

[0103] Optionally, the vehicle terminal obtains the current posture of the vehicle, compares the position difference between the perceived lane line of the vehicle in the current lane and the actual lane line of the vehicle in the current lane in the map, calculates the distance error between the two, and uses nonlinear least squares method or other optimization algorithm to iteratively update the current posture of the vehicle according to the distance error to obtain the final adjusted posture, which will be used for vehicle navigation and control.

[0104] Through this embodiment, when the first identifier of the lane line sensed by the vehicle at the current moment is the same as the first identifier of the lane line sensed by the vehicle at the previous moment before the current moment, it is determined that the vehicle has not changed lanes. At this time, the lane in which the vehicle was located at the previous moment before the current moment is determined as the lane in which the vehicle is currently located, which can avoid executing the complete lane matching process, thereby significantly reducing the consumption of computing resources and improving the operating efficiency of the system.

[0105] In an exemplary embodiment, identifying lane lines in a road image and obtaining lane line perception results includes:

[0106] Lane lines in a road image are identified to obtain a plurality of perceived lane lines; each perceived lane line includes a first virtual-real attribute; each perceived lane line is sorted according to a left-right relative relationship between a current position of the vehicle and each perceived lane line to obtain a sorting result; when the sorting result indicates that there are missed lane lines, the missed lane lines are determined according to the sorting result, and the missed lane lines and each perceived lane line are re-sorted to obtain a first left-right topological structure.

[0107] Among them, missed lane lines refer to the failure to identify all actually existing lane lines during the perception process of lane lines due to limitations in image quality, lighting conditions, vehicle field of view or other factors. After sorting each perceived lane line, the embodiment of the present application determines whether the detected lane lines have been missed by judging whether the left and right lane lines of each lane exist. For example, the lane line perception results include: {"left-left - solid line", "right - solid line", "right-right - curb"}, where the left lane line is missing, then the left lane line is the missed lane line, and the missed left lane line is completed, and the first virtual-real attribute of the missed left lane line is marked as unknown.

[0108] Optionally, the vehicle terminal uses image processing and machine learning techniques, such as a deep learning convolutional neural network (CNN) or a semantic segmentation model, to analyze road images, identify the left-right relative relationship between multiple perceived lane lines within the vehicle's current field of view and the first virtual-real attribute of each perceived lane line, and arrange the multiple perceived lane lines in order from left to right or from right to left to obtain a sorting result; when there is a lane in the sorting result that lacks one of the left and right lane lines, it is determined that the sorting result indicates the existence of a missed lane line. At this time, the missed lane line is determined based on the sorting result, and the missed lane line and each perceived lane line are re-sorted to obtain a first left-right topological structure.

[0109] Through this embodiment, in the case where there are missed lane lines, the missed lane lines are determined according to the sorting results, and the missed lane lines are supplemented, so as to ensure that all lane lines within the vehicle's field of view are accurately acquired, thereby improving the integrity of lane line recognition and thus improving the accuracy of vehicle positioning.

[0110] In an exemplary embodiment, adjusting the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located to obtain the adjusted current posture of the vehicle includes:

[0111] The current posture of the vehicle is iteratively updated using the nonlinear least squares method to minimize the objective function; the objective function is the residual sum of squares of the distance error between the perceived lane line of the lane in which the vehicle is currently located and the actual lane line of the lane in which the vehicle is currently located; the posture that minimizes the objective function is determined as the adjusted current posture of the vehicle.

[0112] Among them, the nonlinear least squares method is an iterative optimization algorithm used to adjust the vehicle posture to minimize the distance error between the perceived lane line and the actual lane line, thereby improving the accuracy of vehicle positioning.

[0113] Minimizing the objective function means that the adjusted vehicle posture is closest to the actual lane line, which improves the accuracy of positioning.

[0114] Optionally, the vehicle terminal extracts the boundary points of the real lane line of the current lane in the map according to the current position of the vehicle and the current lane, and extracts the boundary points of the perceived lane line of the current lane in the perceived lane line, and determines the correspondence between the boundary points of the real lane line of the current lane of the vehicle and the boundary points of the perceived lane line of the current lane; constructs an objective function according to the distance error between the perceived lane line and the real lane line, and the objective function is expressed in the form of residual sum of squares, that is, the squares of the distance errors of all corresponding boundary points between the perceived lane line and the real lane line are summed; starting from the current posture of the vehicle, the iterative optimization process is started using the nonlinear least squares method. In each iteration, the nonlinear least squares method predicts the position of the lane line based on the current posture parameters, and then calculates the residual between the predicted position and the position of the real lane line. Through gradient descent or other optimization strategies, the algorithm adjusts the posture parameters to reduce the value of the objective function (residual sum of squares) until the objective function reaches the set convergence condition or reaches the maximum number of iterations, and the posture that minimizes the objective function is determined as the adjusted current posture of the vehicle.

[0115] Through this embodiment, the nonlinear least squares method continuously adjusts the current posture of the vehicle through iterative optimization to minimize the distance error between the perceived lane line and the actual lane line on the map, and can gradually correct the posture estimation until the best matching state is reached, thereby significantly improving the accuracy of vehicle positioning.

[0116] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0117] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware server, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0118] According to another aspect of the embodiment of the present application, a vehicle positioning device is also provided, and the device is used to implement the vehicle positioning method provided in the above-mentioned embodiment, and the description has been made no further. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented with software, the implementation of hardware, or the combination of software and hardware is also possible and conceived.

[0119] Figure 6 is a structural block diagram of an optional vehicle positioning device according to an embodiment of the present application, such as Figure 6 As shown, the device may include:

[0120] The image acquisition module 601 is used to acquire the road image collected by the vehicle at the current moment, and identify the lane lines in the road image to obtain the lane line perception result; the lane line perception result includes the first left-right topological structure of multiple perceived lane lines perceived by the vehicle at the current moment, and the first virtual-real attribute of each of the multiple perceived lane lines;

[0121] A map acquisition module 602 is used to acquire real lane line information within the scope of the current position of the vehicle from a map when the lane of the vehicle at the previous moment before the current moment is not determined; the real lane line information includes a second left-right topological structure of multiple real lane lines within the scope of the current position of the vehicle, and a second virtual-real attribute of each of the multiple real lane lines;

[0122] A matching module 603 is used to match the lane line perception result with the real lane line information to obtain a matching result;

[0123] A lane determination module 604 is used to determine the lane in which the vehicle is currently located according to the second left-right topological structure when the matching result indicates that the lane line perception result and the real lane line information are successfully matched;

[0124] The posture adjustment module 605 is used to obtain the current posture of the vehicle and adjust the current posture of the vehicle according to the distance error between the perceived lane line of the lane where the vehicle is currently located and the real lane line of the lane where the vehicle is currently located, so as to obtain the adjusted current posture of the vehicle.

[0125] In an exemplary embodiment, the first left-right topological structure includes the left-right relative relationship between the current position of the vehicle and each perceived lane line, and the second left-right topological structure includes the left-right relative relationship between the current position of the vehicle and each real lane line; the matching module 603 is also used to extract the target perceived lane line marked with a road boundary label from multiple perceived lane lines, and to extract the target real lane line marked with a road boundary label from multiple real lane lines when the lane line perception result and the real lane line information also include a road boundary label; compare the first left-right relative relationship between the current position of the vehicle and the target perceived lane line with the second left-right relative relationship between the current position of the vehicle and the target real lane line, and compare the first virtual-real attribute of the target perceived lane line with the second virtual-real attribute of the target real lane line; when the first left-right relative relationship matches the second left-right relative relationship, and the first virtual-real attribute of the target perceived lane line matches the second virtual-real attribute of the target real lane line, determine that the matching result indicates that the match between the lane line perception result and the real lane line information is successful.

[0126] In an exemplary embodiment, the matching module 603 is also used to construct an augmented matrix based on the lane line perception results and the real lane line information when at least one of the lane line perception results and the real lane line information does not include a road boundary label; each row of the augmented matrix represents each perceived lane line, each column of the augmented matrix represents each real lane line, and the elements of the augmented matrix represent the degree of match between the first virtual-real attribute of the perceived lane line represented by the corresponding row and the second virtual-real attribute of the real lane line represented by the corresponding column; the augmented matrix is ​​solved by an optimization algorithm, and when there is only one set of solutions, it is determined that the matching result represents a successful match between the lane line perception results and the real lane line information.

[0127] In an exemplary embodiment, the lane line perception result also includes a first identifier of each perceived lane line, and the real lane line information also includes a second identifier of each real lane line, and the first identifier of each perceived lane line is used to mark the perceived lane line within the current field of view of the vehicle; the second identifier of each real lane line is used to mark each real lane line; the map acquisition module 602 is also used to compare the first identifier of the lane line perceived by the vehicle at the current moment with the first identifier of the lane line perceived by the vehicle at the previous moment when the lane in which the vehicle is located at the previous moment has been determined; when the first identifier of the lane line perceived by the vehicle at the current moment is different from the first identifier of the lane line perceived by the vehicle at the previous moment, determine the lane change direction of the vehicle according to the first left-right topological structure; determine the lane in which the vehicle is currently located according to the second left-right topological structure and the lane change direction; obtain the current posture of the vehicle, and adjust the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

[0128] In an exemplary embodiment, the map acquisition module 602 is also used to determine the lane in which the vehicle was located at the previous moment before the current moment as the lane in which the vehicle is currently located when the first identifier of the lane line perceived by the vehicle at the current moment is the same as the first identifier of the lane line perceived by the vehicle at the previous moment before the current moment; obtain the current posture of the vehicle, and adjust the current posture of the vehicle according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the actual lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

[0129] In an exemplary embodiment, the image acquisition module 601 is used to identify lane lines in a road image to obtain multiple perceived lane lines; each perceived lane line includes a first virtual-real attribute; each perceived lane line is sorted according to the left-right relative relationship between the current position of the vehicle and each perceived lane line to obtain a sorting result; when the sorting result indicates that there are missed lane lines, the missed lane lines are determined according to the sorting result, and the missed lane lines and each perceived lane line are re-sorted to obtain a first left-right topological structure.

[0130] In an exemplary embodiment, the posture adjustment module 605 is also used to iteratively update the current posture of the vehicle using a nonlinear least squares method to minimize an objective function; the objective function is the residual sum of squares of the distance error between the perceived lane line of the lane in which the vehicle is currently located and the actual lane line of the lane in which the vehicle is currently located; the posture that minimizes the objective function is determined as the adjusted current posture of the vehicle.

[0131] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiments. It should be noted that the above modules as part of the device can be run in Figure 1 In the hardware environment shown, it can be implemented by software or by hardware, wherein the hardware environment includes a network environment.

[0132] According to another aspect of the embodiments of the present application, a storage medium is also provided. Optionally, in this embodiment, the storage medium can be used to execute the program code of any of the vehicle positioning methods in the embodiments of the present application.

[0133] Optionally, in this embodiment, the storage medium may be located on at least one network device among a plurality of network devices in the network shown in the above embodiment.

[0134] Optionally, in this embodiment, the storage medium is configured to store a program code for executing any one of the above-mentioned vehicle positioning methods.

[0135] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, which will not be described in detail in this embodiment.

[0136] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a ROM, a RAM, a mobile hard disk, a magnetic disk, or an optical disk.

[0137] According to another aspect of the embodiment of the present application, an electronic device for implementing the above-mentioned vehicle positioning method is also provided. The electronic device may be a smart device, a server, a terminal, or a combination thereof.

[0138] Figure 7 is a structural block diagram of an optional electronic device according to an embodiment of the present application, such as Figure 7 As shown, it includes a processor 702, a communication interface 704, a memory 706 and a communication bus 708, wherein the processor 702, the communication interface 704 and the memory 706 communicate with each other through the communication bus 708, wherein,

[0139] Memory 706, used to store computer programs;

[0140] The processor 702 is used to implement the program code of any of the above-mentioned vehicle positioning methods when executing the computer program stored in the memory 706.

[0141] Optionally, the communication bus may be a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 The communication interface is used for communication between the above electronic device and other devices.

[0142] The memory may include RAM, or may include non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0143] As an example, the memory 706 may include but is not limited to each module in the vehicle positioning device. In addition, it may also include but is not limited to other module units in the vehicle positioning device, which will not be repeated in this example.

[0144] The above-mentioned processor can be a general-purpose processor, which can include but not be limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; it can also be DSP (Digital Signal Processing), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0145] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment will not be described in detail here.

[0146] It can be understood by those skilled in the art that Figure 7 The structure shown is for illustration only. The device for implementing the above-mentioned vehicle positioning method may be a terminal device, which may be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, and other terminal devices. Figure 7 The structure of the electronic device is not limited. Figure 7 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 7 Different configurations shown.

[0147] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0148] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0149] The embodiments of the present application also provide a computer program, which includes computer instructions, which are stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in any one of the above method embodiments.

[0150] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which can include: a flash drive, ROM, RAM, a magnetic disk or an optical disk, etc.

[0151] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0152] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.

[0153] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0154] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0155] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in this embodiment.

[0156] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist physically separately, or at least two units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0157] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A vehicle positioning method, characterized in that: include: Acquire a road image collected by the vehicle at the current moment, and identify lane lines in the road image to obtain a lane line perception result; The lane line perception result includes a first left-right topological structure of a plurality of perceived lane lines perceived by the vehicle at a current moment, and a first virtual-real attribute of each of the plurality of perceived lane lines; When the lane of the vehicle at the last moment before the current moment is not determined, obtaining real lane line information within the range of the current position of the vehicle from a map; The real lane line information includes a second left-right topological structure of a plurality of real lane lines within a range of the current position of the vehicle, and a second virtual-real attribute of each of the plurality of real lane lines; Matching the lane line perception result with the real lane line information to obtain a matching result; When the matching result indicates that the lane line perception result and the real lane line information are matched successfully, determining the lane in which the vehicle is currently located according to the second left-right topological structure; The current posture of the vehicle is obtained, and the current posture of the vehicle is adjusted according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

2. The method according to claim 1, characterized in that The first left-right topological structure includes a left-right relative relationship between the current position of the vehicle and each of the perceived lane lines, and the second left-right topological structure includes a left-right relative relationship between the current position of the vehicle and each of the real lane lines; The matching of the lane line perception result with the real lane line information to obtain a matching result includes: In a case where the lane line perception result and the real lane line information also include a road boundary label, extracting a target perceived lane line marked by the road boundary label from the multiple perceived lane lines, and extracting a target real lane line marked by the road boundary label from the multiple real lane lines; Comparing a first left-right relative relationship between the current position of the vehicle and the target perceived lane line with a second left-right relative relationship between the current position of the vehicle and the target real lane line, and comparing a first virtual-real attribute of the target perceived lane line with a second virtual-real attribute of the target real lane line; When the first left-right relative relationship matches the second left-right relative relationship, and the first virtual-real attribute of the target perceived lane line matches the second virtual-real attribute of the target real lane line, determining the matching result indicates that the match between the lane line perception result and the real lane line information is successful.

3. The method according to claim 2, characterized in that The matching the lane line perception result with the real lane line information to obtain a matching result further includes: In a case where at least one of the lane line perception result and the real lane line information does not include the road boundary label, constructing an augmented matrix based on the lane line perception result and the real lane line information; each row of the augmented matrix represents each perceived lane line, each column of the augmented matrix represents each real lane line, and an element of the augmented matrix represents a matching degree between a first virtual-real attribute of the perceived lane line represented by the corresponding row and a second virtual-real attribute of the real lane line represented by the corresponding column; An optimization algorithm is used to solve the augmented matrix, and when there is only one set of solutions, it is determined that the matching result represents a successful match between the lane line perception result and the real lane line information.

4. The method according to claim 1, characterized in that: The lane line perception result further includes a first identifier of each perceived lane line, and the real lane line information further includes a second identifier of each real lane line, and the first identifier of each perceived lane line is used to mark the perceived lane line within the current field of view of the vehicle; The second identifier of each real lane line is used to mark each real lane line; The method further comprises: When the lane in which the vehicle is located at a moment before the current moment has been determined, comparing the first identifier of the lane line sensed by the vehicle at the current moment with the first identifier of the lane line sensed by the vehicle at the moment before the current moment; When the first identifier of the lane line sensed by the vehicle at the current moment is different from the first identifier of the lane line sensed by the vehicle at a moment before the current moment, determining the lane change direction of the vehicle according to the first left-right topological structure; Determining a lane in which the vehicle is currently located according to the second left-right topological structure and the lane change direction; The current posture of the vehicle is obtained, and the current posture of the vehicle is adjusted according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

5. The method according to claim 4, characterized in that The method further comprises: When the first identifier of the lane line sensed by the vehicle at the current moment is the same as the first identifier of the lane line sensed by the vehicle at the moment before the current moment, the lane in which the vehicle was located at the moment before the current moment is determined as the lane in which the vehicle is currently located; The current posture of the vehicle is obtained, and the current posture of the vehicle is adjusted according to the distance error between the perceived lane line of the lane in which the vehicle is currently located and the real lane line of the lane in which the vehicle is currently located, to obtain the adjusted current posture of the vehicle.

6. The method according to claim 1, characterized in that The identifying the lane line in the road image to obtain the lane line perception result includes: Identify lane lines in the road image to obtain the plurality of perceived lane lines; each of the perceived lane lines comprises a first virtual-real attribute; Sorting each of the sensed lane lines according to a left-right relative relationship between the current position of the vehicle and each of the sensed lane lines to obtain a sorting result; In the case where the sorting result indicates that there is a missed lane line, the missed lane line is determined according to the sorting result, and the missed lane line and each sensed lane line are re-sorted to obtain the first left-right topological structure.

7. The method according to any one of claims 1 to 6, characterized in that The adjusting the current posture of the vehicle according to the distance error between the perceived lane line of the lane currently located by the vehicle and the real lane line of the lane currently located by the vehicle to obtain the adjusted current posture of the vehicle includes: Iteratively updating the current position of the vehicle using a nonlinear least squares method to minimize an objective function; the objective function is the residual sum of squares of the distance error between the perceived lane line of the lane where the vehicle is currently located and the real lane line of the lane where the vehicle is currently located; The posture that minimizes the objective function is determined as the adjusted current posture of the vehicle.

8. A vehicle positioning device, characterized in that: include: An image acquisition module is used to acquire a road image collected by the vehicle at the current moment, identify lane lines in the road image, and obtain a lane line perception result; The lane line perception result includes a first left-right topological structure of a plurality of perceived lane lines perceived by the vehicle at a current moment, and a first virtual-real attribute of each of the plurality of perceived lane lines; A map acquisition module, used to acquire real lane line information within the range of the current position of the vehicle from a map when the lane of the vehicle at the previous moment at the current moment is not determined; The real lane line information includes a second left-right topological structure of a plurality of real lane lines within a range of the current position of the vehicle, and a second virtual-real attribute of each of the plurality of real lane lines; A matching module, used to match the lane line perception result with the real lane line information to obtain a matching result; a lane determination module, configured to determine the lane in which the vehicle is currently located according to the second left-right topological structure when the matching result indicates that the lane line perception result and the real lane line information are successfully matched; The posture adjustment module is used to obtain the current posture of the vehicle and adjust the current posture of the vehicle according to the distance error between the perceived lane line of the lane where the vehicle is currently located and the real lane line of the lane where the vehicle is currently located, so as to obtain the adjusted current posture of the vehicle.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of any one of the methods of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images