A yaw positioning adaptive correction method and system

Abstract

The application provides a yaw adaptive correction method and system, which comprises the following steps: sensing the vehicle position, the drivable area and the lane line information of a preset map; extracting the guide strip boundary information based on the drivable area and the lane line information; determining the target guide strip based on the guide strip boundary information, and determining the positional relationship between the vehicle position and the target guide strip; performing yaw detection based on the positional relationship and the preset map; and correcting the position of the vehicle in the preset map based on the yaw detection result. The application can effectively perform adaptive correction positioning when the road-level positioning is abnormal or yaw.

Description

Technical Field

[0001] This application belongs to the field of intelligent driving technology, specifically relating to a positioning yaw adaptive correction method and system. Background Technology

[0002] With the rapid development of intelligent driving and robotic environmental perception technologies, high-precision road-level positioning has become a crucial foundation for the safe and reliable navigation of L2 / L3 level autonomous vehicles, unmanned delivery vehicles, unmanned trucks, and mobile robots. Currently, mainstream road-level positioning schemes based on Standard Definition Maps (SD) typically rely on an Inertial Navigation System (INS), prior information from the SD map, and post-processing results from perception data from sensors such as vision or LiDAR (e.g., lane detection, traffic light recognition). These information are then fused using specific road-level positioning algorithms to ultimately output the vehicle's or robot's specific location and lane assignment within the road network. This positioning result then provides critical input to downstream planning and control modules, supporting vehicle operation even without detailed high-precision maps.

[0003] However, in practical applications, road-level positioning algorithms may suffer from insufficient robustness, abnormal environmental perception, and SD map mapping biases, leading to positioning errors or deviations. Current technologies lack adaptive correction algorithms for these errors. For example, due to differences in visual characteristics between the guide strip area and traditional lane lines, and the potential for ambiguity in intersection topology, existing systems are highly prone to road-level positioning errors in such scenarios, resulting in vehicles being incorrectly positioned on adjacent or unexpected roads. This causes downstream planning and control modules to generate driving trajectories and instructions based on erroneous location information, potentially leading to vehicles deviating from their intended paths, missing critical turning points, or even posing a collision risk with other road users, thus creating serious safety hazards. Summary of the Invention

[0004] To address the aforementioned technical problems, this application proposes a positioning yaw adaptive correction method and system capable of adaptively correcting positioning when road-level positioning algorithms experience positioning anomalies or yaw.

[0005] Specifically, this application proposes a positioning yaw adaptive correction method, comprising: sensing the current vehicle position, drivable area, and lane line information; extracting guide strip boundary information based on the drivable area and lane line information; determining a target guide strip based on the guide strip boundary information and determining the positional relationship between the vehicle position and the target guide strip; performing positioning yaw detection based on the positional relationship and a preset map; and correcting the vehicle's position in the preset map based on the result of the positioning yaw detection.

[0006] In the above technical solution, by extracting the boundary information of the guide strip from the drivable area and lane line information, determining the target guide strip, and establishing the positional relationship between the vehicle position and the target guide strip, positioning yaw detection is performed based on the positional relationship. This achieves accurate identification of the guide strip intersection position. The vehicle position in the preset map is corrected based on the positioning yaw detection results, realizing adaptive correction of positioning anomalies or yaws, and improving the robustness of the system. After correcting the vehicle position, the correct positioning result is output, effectively preventing the downstream planning and control module from generating driving trajectories and instructions based on incorrect position information. This avoids vehicles deviating from the expected path, missing key turning points, or even causing collisions with other road users, effectively improving vehicle driving safety.

[0007] In one implementation, the guide strip boundary information includes at least guide strip boundary point information and guide line information; extracting the guide strip boundary information based on the drivable area and lane line information includes: clipping the drivable area based on a preset clipping range, traversing all data points in the clipped drivable area, and performing type segmentation on all data points to obtain the guide strip boundary point information; filtering lane lines of type guide line from the lane line information; and clipping the lane lines of type guide line based on the preset clipping range to obtain guide line information.

[0008] By cropping the drivable area and segmenting all data points by type, the system effectively focuses on the key decision-making area near the current vehicle, reducing the amount of data, improving data processing efficiency, and avoiding data errors caused by inaccurate perception at a distance or map deviations. This also increases the system's adaptability to various complex scenarios.

[0009] Furthermore, based on the boundary information of the guide strip, the target guide strip is determined. The guide strip boundary information is then subjected to clustering processing, including: marking all data points of the guide strip boundary point information and guide line information as unvisited, and constructing an accessed state list; sequentially traversing all data points, calculating the data volume of other data points within a preset range centered on each data point; when the data volume is greater than or equal to a preset data volume threshold, marking the data point at the center as a core point; constructing a data cluster and adding all core points and other data points within their preset range to the data cluster, and marking data points not written into the data cluster as noise points for removal.

[0010] This clustering process adds all core points and corresponding data points from the guide zone boundary point information and guide line information into data clusters. Data points not written into the data clusters are marked as noise points and removed. This effectively distinguishes between genuine guide zone boundary signals and interference noise such as false alarms, ground debris reflections, and irrelevant road markings. It significantly improves the system's tolerance to sensing anomalies and real-world interference, avoiding misjudgments of the entire guide zone morphology due to noise points.

[0011] Furthermore, determining the target guide zone based on the guide zone boundary information also includes: calculating the center point, foremost point, last point, leftmost point, and rightmost point of each guide zone based on all core points and other data points in the data cluster; comparing the foremost point and last point with preset maximum vehicle front distance thresholds and preset maximum vehicle rear distance thresholds, and retaining only the guide zones where the foremost point and last point are within the preset maximum vehicle front distance thresholds and preset maximum vehicle rear distance thresholds.

[0012] By calculating the center point, foremost point, last point, leftmost point, and rightmost point of each guide strip, the spatial range of the guide strip is determined. The spatial range is then compared with the preset maximum front and rear vehicle clustering thresholds, thereby enabling the determination of the guide strip boundary range and performing a spatial filter on the guide strip, reducing the amount of subsequent computational data and decision complexity.

[0013] Furthermore, the lane line information also includes edge lane line information; determining the target guide strip based on the guide strip boundary information further includes: retaining the guide strip when the center point of the guide strip is located to the left of the leftmost edge lane line or to the right of the rightmost edge lane line in the edge lane line information; otherwise, not retaining the guide strip; retaining the guide strip when there is a lane line exceeding a preset length threshold to the left and right of the center point of the guide strip; otherwise, determining the guide strip to be invalid.

[0014] By retaining guide strips whose center point is located to the left of the leftmost edge lane line or to the right of the rightmost edge lane line in the edge lane line information, the validity of the roadside was determined, effectively eliminating false guide strip targets caused by perception errors or point cloud clustering anomalies. By retaining guide strips with one lane line exceeding a preset length threshold to the left and one to the right of their center point, non-bifurcation interference can be reliably eliminated.

[0015] Furthermore, determining the target guide zone based on the guide zone boundary information further includes: dividing the guide zone into regions of a first preset numerical block, retaining only the target guide zone region of the second preset numerical block; the target guide zone region may include at least a first target guide zone region, a second target guide zone region, and a third target guide zone region; in a coordinate system with the rear axle center of the vehicle as the origin, obtaining the corresponding rightmost lateral coordinate and leftmost lateral coordinate in each target guide zone region; obtaining the first maximum lateral region, the second maximum lateral region, and the third maximum lateral region of the first target guide zone region, the second target guide zone region, and the third target guide zone region based on the rightmost lateral coordinate and the leftmost lateral coordinate; when the first maximum lateral region is successively greater than the second maximum lateral region and the third maximum lateral region, the guide zone is determined to be valid and is taken as the target guide zone; otherwise, the guide zone is determined to be invalid.

[0016] By intelligently dividing the guide strip into several continuous regions along its direction and analyzing the maximum lateral area of ​​each region, the detection of key morphological features is achieved by comparing the maximum lateral area of ​​each target guide strip. This ensures the reliability of the target guide strip that is ultimately retained and used for decision-making, minimizing the possibility of misidentification.

[0017] Furthermore, determining the positional relationship between the vehicle position and the target guide strip includes: forming a first triangular bounding box based on the foremost, leftmost, and rightmost points of the target guide strip; forming a second triangular bounding box based on the last, leftmost, and rightmost points of the target guide strip; if the vehicle position is within either the first or second triangular bounding box, the vehicle position is determined to be within the bifurcation; otherwise, if the vehicle position is not within the bifurcation of the guide strip, it is further determined whether the lateral coordinate of the center point of the target guide strip is positive; if positive, the vehicle position is determined to be on the left side of the bifurcation; if negative, the vehicle position is determined to be on the right side of the bifurcation; otherwise, the vehicle position is determined to be another position on the target guide strip; when the positional relationship between the vehicle position and the target guide strip is the same in consecutive preset value data frames, the positional relationship is taken as the final positional relationship.

[0018] By determining the relationship between the vehicle's position and the bifurcation point of the guide strip, correction can be triggered directly and efficiently when the vehicle's position in the preset map is inconsistent with the current position, ensuring both accuracy and efficiency in the correction.

[0019] Furthermore, based on the positional relationship and the preset map, positioning deviation detection is performed, including: when the positional relationship is that the vehicle position is on the left or right side of the fork, searching the preset map information to obtain the relationship between the fork node and the vehicle position in the preset map information; when the relationship between the fork node and the vehicle position in the preset map information is inconsistent with the positional relationship, determining that the vehicle position is deviated.

[0020] Among them, by sensing whether the vehicle's position is consistent with the relationship in the preset map information on the left or right of the fork, the positioning deviation trigger condition is clarified, which effectively reduces the probability of false triggering and missed triggering.

[0021] Furthermore, based on the results of the positioning yaw detection, the vehicle's position in the preset map is corrected, including: when the vehicle's position in the preset map is to the right of a fork node, but the positional relationship is that the vehicle's position is to the left of the fork in the target guide zone, the vehicle's position is determined to be yawed, and the permissions of the driving path to the right of the fork are reconfigured based on a preset weight coefficient, correcting the vehicle's position to the path to the left of the fork; when the vehicle's position in the preset map is to the left of a fork node, but the positional relationship is that the vehicle's position is to the right of the fork in the target guide zone, the vehicle's position is determined to be yawed, and the permissions of the driving path to the left of the fork are reconfigured based on a preset weight coefficient, correcting the vehicle's position to the path to the right of the fork.

[0022] By reconfiguring permissions for the driving path, the vehicle's position is corrected to the correct fork in the road, fundamentally eliminating erroneous decisions caused by incorrect positioning information and enabling the system to adaptively correct positioning yaw. The complexity of positioning yaw correction is effectively reduced through permission configuration.

[0023] Based on the same inventive concept, this application also proposes a positioning yaw adaptive correction system, the system comprising at least an electronic device; the electronic device comprising a sensing device for sensing the current drivable area and vehicle position, including an environmental image containing lane line information; a human-computer interaction module for displaying a navigation map; and a memory for storing computer programs for multiple functional modules, the memory being connected to a processor via a bus, the processor including an executable program for executing any of the computer programs stored in the memory; wherein the functional modules include at least: an information extraction module comprising a computer program for extracting guide zone boundary information based on the drivable area and lane line information; a position determination module comprising a computer program for determining a target guide zone based on the guide zone boundary information and determining the positional relationship between the vehicle position and the target guide zone; a yaw detection module comprising a computer program for performing positioning yaw detection based on the positional relationship and the preset map; and a yaw correction module comprising a computer program for correcting the vehicle's position on the preset map based on the result of the positioning yaw detection.

[0024] Compared with the prior art, this application has at least the following beneficial effects: This application extracts the boundary information of the guide strip from the drivable area and lane line information. After determining the positional relationship between the vehicle and the target guide strip, it performs positioning yaw detection based on the positional relationship, thereby achieving accurate identification of the guide strip intersection position. The vehicle position in the preset map is corrected by the yaw detection results, realizing adaptive correction of positioning anomalies or yaw, and improving the robustness of the system. After correcting the vehicle position, the correct positioning result is output, which effectively avoids the downstream planning and control module from generating driving trajectories and instructions based on incorrect position information. This prevents the vehicle from deviating from the expected path, missing key turning points, or even causing collisions with other road users, effectively improving the safety of vehicle driving. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the adaptive positioning and yaw correction method in an embodiment of this application.

[0026] Figure 2 This is a schematic diagram illustrating a car driving out of the guide strip, as shown in an embodiment of this application.

[0027] Figure 3 This is a schematic diagram illustrating the division of the flow guidance area in an embodiment of this application.

[0028] Figure 4 This is a schematic diagram of the triangular bounding box of the target guide strip shown in the embodiments of this application. Figure 5 This is a schematic diagram of the bifurcation of the guide strip shown in an embodiment of this application.

[0029] Figure 6 This is a schematic diagram of the positioning and yaw adaptive correction system shown in the embodiments of this application. Detailed Implementation

[0030] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices. Example 1:

[0032] Please refer to Figure 1 The positioning yaw adaptive correction method mainly includes steps S100 to S500.

[0033] Step S100 includes: perceiving the current vehicle position, drivable area, and lane line information. This perception can primarily be achieved through sensing devices such as visual sensors, radar sensors, Global Navigation Satellite System (GNSS), and Inertial Measurement Unit (IMU). The visual sensor can be a monocular camera, a binocular stereo camera, or a surround-view camera. Deep learning algorithms can be used to identify lane lines from the image data acquired by the visual sensor, thereby obtaining lane line information. These deep learning algorithms can include CNN (Convolutional Neural Network), YOLO (You Only Look Once), etc., without limitation. The radar sensor can be a lidar or millimeter-wave radar. The radar sensor can generate a 3D point cloud of the current vehicle's surrounding environment. The image data acquired by the visual sensor and the point cloud data acquired by the radar sensor are fused, and the fused data is then used with a deep learning algorithm to effectively identify the current drivable area. Global Navigation Satellite System (GNSS) can be used, such as but not limited to Russia's GLONASS, the European Union's Galileo, and China's BeiDou (BDS), in conjunction with IMU (Inertial Measurement Unit) to obtain the vehicle's position.

[0034] Step S200 includes: extracting the boundary information of the guide strip based on the drivable area and lane line information. Note that since positioning yaw generally only occurs in scenarios where the guide strip merges, please refer to... Figure 2 This is a schematic diagram of a traffic duct that creates a flow path for vehicles. The road in the diagram conforms to the shape of a triangle.

[0035] Step S300 includes: determining the target guide zone based on the guide zone boundary information, and determining the positional relationship between the vehicle position and the target guide zone. A clustering algorithm can be used to cluster the guide zone boundary information, and the target guide zone is determined based on the clustering results. This clustering algorithm can primarily be DBSCAN (Density-Based Spatial Clustering of Applications with Noise). DBSCAN is a density-based clustering algorithm that does not require pre-specifying the number of clusters; instead, it identifies the cluster structure based on the spatial distribution density of data and effectively handles noise points. After obtaining the clustering results, a guide zone is represented by a data cluster. The obtained guide zone is then subjected to boundary range judgment, guide zone roadside validity judgment, guide zone lane line validity judgment, and merging guide zone box detection to obtain the target guide zone. Based on this target guide zone, the positional relationship between the vehicle position and the target guide zone is determined.

[0036] Step S400 includes: performing positioning deviation detection based on the positional relationship and a preset map. Specifically, it determines whether the positional relationship matches the positional relationship between the vehicle and the bifurcation point of the guide strip in the preset map. If they match, it proves that the current positioning has no deviation; otherwise, it determines that the current positioning has a deviation. The preset map can be a raw Standard Definition Map (SD) map received by a map engine.

[0037] And step S500 includes: correcting the vehicle's position in the preset map based on the result of the positioning yaw detection.

[0038] In some embodiments, the guide strip boundary information includes at least guide strip boundary point information and guide line information; extracting the guide strip boundary information based on the drivable area and lane line information includes: clipping the drivable area based on a preset clipping range, traversing all data points in the clipped drivable area, and performing type segmentation on all data points to obtain the guide strip boundary point information; filtering lane lines of type guide line from the lane line information; and clipping the lane lines of type guide line based on the preset clipping range to obtain guide line information.

[0039] Since the data volume of the entire drivable area is large, extracting data points from the entire drivable area is computationally intensive. Therefore, a coordinate system is defined with the center of the rear axle of the vehicle as the origin, with the X-axis pointing forward from the rear of the vehicle (positive direction) and the Y-axis pointing to the left and right (positive direction to the right). The entire drivable area can be approximately 300m x 300m. The preset clipping range is approximately 20m behind and 50m in front of the origin of the coordinate system, with 10m to the left and right. By clipping the drivable area, it is not necessary to traverse the entire drivable area, thus improving data processing efficiency. The main approach involves defining data points as follows: type=1 for drivable areas, type=2 for guide strip areas, and type=3 for non-drivable areas. By traversing all data points within the cropped drivable area, the guide strip area data points are obtained. Points within this guide strip area where the preceding point has a type other than 2, and the following point has a type of 2, or vice versa, are retained as the extracted guide strip boundary points. The lane line information includes the type information of each identified lane line, which can include guide strip type, edge lane line type, and middle lane line type.

[0040] Optionally, determining the target guide zone based on the guide zone boundary information includes clustering the guide zone boundary information, including: marking all data points of the guide zone boundary point information and guide line information as unvisited, and constructing an accessed state list; sequentially traversing all data points, calculating the data volume of other data points within a preset range centered on each data point; when the data volume is greater than or equal to a preset data volume threshold, marking the data point at the center as a core point; constructing a data cluster and adding all core points and other data points within their preset range to the data cluster, and marking data points not written into the data cluster as noise points for removal.

[0041] The preset range mainly refers to a circular region centered at point p with radius ε. Point p and radius ε can be set according to actual needs and are not limited. The preset data volume threshold can be a pre-set minimum number of points. Each data cluster obtained after processing by this preset clustering algorithm contains a core point, other data points, and a set of noise points.

[0042] Optionally, determining the target guide zone based on the guide zone boundary information further includes: calculating the center point, foremost point, last point, leftmost point, and rightmost point of each guide zone based on all core points and other data points in the data cluster; comparing the foremost point and last point with preset maximum vehicle front distance thresholds and preset maximum vehicle rear distance thresholds, and retaining only the guide zones where the foremost point and last point are within the preset maximum vehicle front distance thresholds and preset maximum vehicle rear distance thresholds.

[0043] The center point of the guide strip can be determined by summing the X and Y coordinates of all points within the guide strip, dividing the sums by the total number of points, and obtaining the arithmetic mean of the X and Y coordinates. This arithmetic mean of the X and Y coordinates is then used as the center point of the guide strip. The preset maximum front and rear vehicle distance thresholds are typically 5m. For example, the maximum front vehicle distance threshold is Vehx,max = 5m, and the maximum rear vehicle distance is Vehx,min = -5m. Assuming the foremost point of the guide strip... Finally Filter out or The guide strip retains only the foremost and rearmost points within the preset maximum front distance threshold and preset maximum rear distance threshold.

[0044] Optionally, the lane line information further includes edge lane line information; determining the target guide strip based on the guide strip boundary information further includes: retaining the guide strip when the center point of the guide strip is located to the left of the leftmost edge lane line or to the right of the rightmost edge lane line in the edge lane line information; otherwise, not retaining the guide strip; retaining the guide strip when there is a lane line exceeding a preset length threshold to the left and right of the center point of the guide strip; otherwise, determining the guide strip to be invalid.

[0045] The preset length threshold can be 3m, but is not limited to this. The effectiveness of the guide strip is determined by whether its center is located to the left of the leftmost lane line or to the right of the rightmost lane line. The effectiveness of the lane lines is determined by whether there is a lane line exceeding the preset length threshold to the left and right of the center point of the guide strip.

[0046] Optionally, determining the target guide zone based on the guide zone boundary information further includes: dividing the guide zone into regions of a first preset numerical block, retaining only the target guide zone region of the second preset numerical block; the target guide zone region may include at least a first target guide zone region, a second target guide zone region, and a third target guide zone region; obtaining the corresponding rightmost lateral coordinate and leftmost lateral coordinate in each target guide zone region in a coordinate system with the rear axle center of the vehicle as the origin; obtaining the first maximum lateral region, the second maximum lateral region, and the third maximum lateral region of the first target guide zone region, the second target guide zone region, and the third target guide zone region based on the rightmost lateral coordinate and the leftmost lateral coordinate; when the first maximum lateral region is successively greater than the second maximum lateral region and the third maximum lateral region, the guide zone is determined to be valid and is taken as the target guide zone; otherwise, the guide zone is determined to be invalid.

[0047] The first preset value block can primarily consist of 5 blocks, and the second preset value block can primarily consist of 3 blocks, but is not limited to these. Please refer to [link / reference needed]. Figure 3 The flow guide zone is divided into 5 regions, but only three of them—l2-l3, l3-l4, and l4-l5—are retained. L2-l3 is designated as the first target flow guide region, l3-l4 as the second target flow guide region, and l4-l5 as the third target flow guide region. The rightmost horizontal coordinate of the first target flow guide region is obtained as follows: Leftmost horizontal coordinate Obtain the maximum horizontal area of ​​the first target diversion area, that is, the first maximum horizontal area. Similarly, obtain the second largest horizontal region. and the third largest horizontal region .when When the time is right, the guide strip is deemed valid and is taken as the target guide strip.

[0048] Optionally, determining the positional relationship between the vehicle position and the target guide strip includes: forming a first triangular bounding box based on the foremost, leftmost, and rightmost points in the target guide strip; forming a second triangular bounding box based on the last, leftmost, and rightmost points in the target guide strip; if the vehicle position is within either the first or second triangular bounding box, the vehicle position is determined to be within a bifurcation; otherwise, if the vehicle position is not within a bifurcation, it is further determined whether the lateral coordinate of the center point of the target guide strip is positive; if positive, the vehicle position is determined to be on the left side of the bifurcation; if negative, the vehicle position is determined to be on the right side of the bifurcation; otherwise, the vehicle position is determined to be another position on the target guide strip; when the positional relationship between the vehicle position and the target guide strip is the same in consecutive preset value data frames, the positional relationship is taken as the final positional relationship.

[0049] The positional relationship between the guide strip and the vehicle is categorized into four types: the vehicle is located to the left of the guide strip's bifurcation, the vehicle is located to the right of the guide strip's bifurcation, the vehicle is located within the guide strip's bifurcation, and the vehicle does not fall into any of these three positional relationships. This continuous preset value can primarily be 10; that is, when the vehicle's position matches the target guide strip's positional relationship for 10 consecutive data frames, this positional relationship is determined as the final positional relationship. Please refer to... Figure 4 As shown in the left-middle image, since some airflow guides are not perfectly triangular, two triangular bounding boxes are constructed using four points for judgment. According to the right-middle image, when the airflow guide conforms to a standard triangle, the foremost point will coincide with either the rightmost or leftmost point.

[0050] In the specific implementation, the front_point, left_point, and right_point of the guide strip i form a triangular bounding box 1, and the back_point, left_point, and right_point form a triangular bounding box 2. When the vehicle's coordinate point (0,0) is within either triangular bounding box 1 or triangular bounding box 2, it is determined that the current vehicle is located within the bifurcation of the guide strip, Kin. If the vehicle is not located within the bifurcation of the guide strip, it is determined whether the current vehicle is located to the left or right of the bifurcation. If the y-coordinate of the center point of the target guide strip is positive, it means that the bifurcation is to the right of the vehicle, and the result is output as the current vehicle is located at the left (Kleft) of the bifurcation. If the y-coordinate of the center point is negative, it means that the bifurcation is to the left of the vehicle, and the result is output as the current vehicle is located at the right (Kright) of the bifurcation. If none of the above three results are found, the vehicle position is output as not belonging to any of the above three relationships. If the output of the guide strip and vehicle position determination results in the same outcome for 10 consecutive frames, that outcome is output as the final positional relationship. For example, if the result for 10 consecutive frames is that the current vehicle is located at the left (Kleft) of the guide strip's bifurcation, then the final output result is that the current vehicle is located at the left (Kleft) of the guide strip's bifurcation. If there are no 10 consecutive frames with the same outcome, then the final output result is that the vehicle position does not belong to any of the above three relationships.

[0051] Optionally, positioning deviation detection is performed based on the positional relationship and the preset map, including: when the positional relationship is that the vehicle position is on the left or right side of the fork, searching the preset map information to obtain the relationship between the fork node and the vehicle position in the preset map information; when the relationship between the fork node and the vehicle position in the preset map information is inconsistent with the positional relationship, determining that the vehicle position is deviated.

[0052] Please refer to Figure 5Location 4 is a bifurcation node on the map. Assuming the positional relationship is that the vehicle is located to the left of the bifurcation, that is, at... Figure 5 The positions of nodes 9, 10, and 11 are shown, but in the default map, the vehicle's position is located to the right of the bifurcation node, i.e., at... Figure 5 When the positions of nodes 5, 6, 7, and 8 are determined to be inconsistent, it is determined that the vehicle's position is off course.

[0053] Optionally, based on the result of the positioning yaw detection, correcting the vehicle's position in the preset map includes: when the vehicle's position in the preset map is to the right of a fork node, but the vehicle's position is to the left of a fork in the target guide zone, determining that the vehicle's position is yawed, reconfiguring the permissions of the driving path to the right of the fork based on a preset weight coefficient, and correcting the vehicle's position to the path to the left of the fork; when the vehicle's position in the preset map is to the left of a fork node, but the vehicle's position in the positional relationship is to the right of a fork in the target guide zone, determining that the vehicle's position is yawed, reconfiguring the permissions of the driving path to the left of the fork based on a preset weight coefficient, and correcting the vehicle's position to the path to the right of the fork.

[0054] For example, if the vehicle is currently located to the left of a map fork (within the range 4-9, 9-10, 10-11), but the vehicle's position is to the right of the fork, then the vehicle's positioning is detected as off-target. A off-target flag is reported downstream, and the prior probability of the path 4-9, 9-10, 10-11 is multiplied by a penalty weight coefficient of 0.1 to reconfigure permissions, thereby adaptively moving the positioning to the right of the map (within the range 4-5, 5-6, 6-7, 7-8). Conversely, if the vehicle is located to the right of a map fork (within the range 4-5, 6-7, 7-8), but the vehicle's position is to the left of the fork, then the vehicle's positioning is detected as off-target. A off-target flag is reported downstream, and the prior probability of the path 4-5, 5-6, 6-7, 7-8 is multiplied by a penalty weight coefficient of 0.1, thereby moving the positioning to the left of the map (within the range 4-9, 9-10, 10-11). The prior probability refers to the likelihood that a vehicle will appear on a specific road without any historical information about vehicle movement. Multiplying this prior probability by a penalty weighting coefficient of 0.1 reduces the prior probability to one-tenth of its original value. Example 2:

[0055] Please refer to Figure 6This application also proposes a system employing the positioning yaw adaptive correction method described in Embodiment 1. The system mainly includes an electronic device; the electronic device includes a sensing device for sensing the current drivable area and vehicle position, including environmental images with lane line information; a human-computer interaction module for displaying a navigation map; and a memory for storing computer programs for multiple functional modules. The memory is connected to a processor via a bus, and the processor includes an executable program for executing the computer program of any functional module stored in the memory.

[0056] The processor can be any suitable processing device or set of processing devices, such as, but not limited to, a microprocessor, a microcontroller-based platform, an integrated circuit, one or more field-programmable gate arrays (FPGAs) and / or one or more application-specific integrated circuits (ASICs).

[0057] Memory can be volatile memory (e.g., RAM including non-volatile RAM, magnetic RAM, ferroelectric RAM, etc.), non-volatile memory (e.g., disk storage, flash memory, EPROM, EEPROM, memristor-based non-volatile solid-state memory, etc.), immutable memory (e.g., EPROM), read-only memory, and / or high-capacity storage devices (e.g., hard disk drives, solid-state drives, etc.). In some examples, memory includes multiple types of memory, particularly volatile and non-volatile memory. Memory is a computer-readable medium on which one or more computer programs (such as software for operating the methods of this disclosure) can be embedded. The computer program can embody one or more of the methods or logic described herein. For example, the computer program resides wholly or at least partially within any one or more of memory, computer-readable media, and / or within a processor during execution.

[0058] The human-machine interface module may include a display, an audio module, etc. The display is configured to present an interface and / or other visual information to vehicle occupants (e.g., operators, passengers). In some examples, the display includes a head-up display, a liquid crystal display (LCD), an organic light-emitting diode (OLED) display, a flat panel display, a solid-state display, and / or any other type of display configured to present an interface and / or other visual information to vehicle occupants.

[0059] The functional modules include at least: an information extraction module comprising a computer program for extracting guide strip boundary information based on the drivable area and lane line information; a position determination module comprising a computer program for determining a target guide strip based on the guide strip boundary information and determining the positional relationship between the vehicle position and the target guide strip; a yaw detection module comprising a computer program for performing positioning yaw detection based on the positional relationship and a preset map; and a yaw correction module comprising a computer program for correcting the vehicle's position in the preset map based on the results of the positioning yaw detection.

[0060] In summary, this application addresses the technical problem of the lack of adaptive correction algorithms in existing technologies for road-level positioning algorithms when positioning is abnormal or deviates from the target guide strip. By extracting the boundary information of the guide strip from the drivable area and lane line information, and determining the positional relationship between the vehicle position and the target guide strip, positioning deviation detection is performed based on the positional relationship. This achieves accurate identification of the guide strip intersection position. The vehicle position in the preset map is corrected based on the deviation detection results, realizing adaptive correction of positioning abnormalities or deviations and improving the robustness of the system. After correcting the vehicle position, the correct positioning result is output, effectively preventing the downstream planning and control module from generating driving trajectories and instructions based on incorrect position information. This avoids the vehicle deviating from the expected path, missing key turning points, and even causing collisions with other road users, effectively improving vehicle driving safety.

[0061] In the several embodiments provided in this application, it will be understood that each block in the flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.

[0062] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0063] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application for those skilled in the art.

Claims

1. A positioning yaw adaptive correction method, characterized in that, include: It can perceive the current vehicle location, drivable area, and lane information. Extract the guide strip boundary information based on the drivable area and lane line information; The target guide strip is determined based on the boundary information of the guide strip, and the positional relationship between the vehicle position and the target guide strip is determined. Based on the aforementioned positional relationships and a preset map, positioning and yaw detection are performed. Based on the results of the positioning yaw detection, the vehicle's position in the preset map is corrected.

2. The positioning yaw adaptive correction method according to claim 1, characterized in that, The guide strip boundary information includes at least guide strip boundary point information and guide line information; Based on the drivable area and lane line information, the boundary information of the guide strip is extracted, including: The drivable area is clipped based on a preset clipping range. All data points in the clipped drivable area are traversed, and all data points are segmented by type to obtain the boundary point information of the guide strip. Filter out lane lines of type guide line from the lane line information; Based on the preset trimming range, the lane lines of the type of guide line are trimmed to obtain guide line information.

3. The positioning yaw adaptive correction method according to claim 2, characterized in that, Determining the target guide zone based on the guide zone boundary information includes clustering the guide zone boundary information, including: Mark all data points of the guide strip boundary point information and guide line information as unaccessed, and construct an access status list; Iterate through all data points in sequence and calculate the amount of data for other data points within a preset range centered on each data point. When the amount of data is greater than or equal to a preset data amount threshold, the data point at the center is marked as a core point; A data cluster is constructed and all core points and other data points within a preset range are added to the data cluster. Data points not written into the data cluster are marked as noise points and removed.

4. The positioning yaw adaptive correction method according to claim 3, characterized in that, Determining the target guide zone based on the guide zone boundary information also includes: Based on all core points and other data points in the data cluster, calculate the center point, foremost point, last point, leftmost point, and rightmost point of each guide strip; Based on the comparison between the foremost and last points and the preset maximum front distance threshold and the preset maximum rear distance threshold, only the guide strips of the foremost and last points within the preset maximum front distance threshold and the preset maximum rear distance threshold are retained.

5. The positioning yaw adaptive correction method according to claim 4, characterized in that, The lane line information also includes edge lane line information; the step of determining the target guide strip based on the guide strip boundary information further includes: The guide strip is retained when its center point is located to the left of the leftmost edge lane line or to the right of the rightmost edge lane line in the edge lane line information; otherwise, the guide strip is not retained. If there is a lane line exceeding a preset length threshold on both the left and right sides of the center point of the guide strip, the guide strip is retained; otherwise, the guide strip is deemed invalid.

6. The positioning yaw adaptive correction method according to claim 5, characterized in that, Determining the target guide zone based on the guide zone boundary information also includes: The flow guide is divided into regions of a first preset value block, and only the target flow guide region of the second preset value block is retained; the target flow guide region may include at least a first target flow guide region, a second target flow guide region, and a third target flow guide region; In a coordinate system with the center of the rear axle of the vehicle as the origin, the rightmost and leftmost lateral coordinates are obtained in each target guide area; Based on the rightmost and leftmost horizontal coordinates, obtain the first maximum horizontal region, the second maximum horizontal region, and the third maximum horizontal region of the first target guidance region, the second target guidance region, and the third target guidance region; When the first maximum lateral region is successively larger than the second maximum lateral region and the third maximum lateral region, the guide strip is determined to be valid and is used as the target guide strip; otherwise, the guide strip is determined to be invalid.

7. The positioning yaw adaptive correction method according to claim 6, characterized in that, Determining the positional relationship between the vehicle's location and the target guide strip includes: A triangular bounding box is formed based on the foremost point, leftmost point, and rightmost point in the target guide strip; The second triangular bounding box is formed by the last point, the leftmost point, and the rightmost point in the target guide strip; If the vehicle is located within triangle bounding box one or triangle bounding box two, the vehicle is determined to be located within the bifurcation point; otherwise, the vehicle is determined not to be located within the bifurcation point, and then it is further determined whether the lateral coordinate of the center point of the target guide strip is positive. If it is positive, the vehicle is determined to be located to the left of the bifurcation point; if it is negative, the vehicle is determined to be located to the right of the bifurcation point. Otherwise, the vehicle's location is determined to be another location within the target guide zone; When the vehicle position and the target guide strip have the same positional relationship in consecutive preset value data frames, the positional relationship is taken as the final positional relationship.

8. The positioning yaw adaptive correction method according to claim 7, characterized in that, Based on the aforementioned positional relationships and a preset map, positioning and yaw detection are performed, including: When the positional relationship is that the vehicle is located to the left or right of the fork, the preset map information is searched to obtain the relationship between the fork node and the vehicle position in the preset map information. When the relationship between the fork node and the vehicle position in the preset map information is inconsistent with the positional relationship, it is determined that the vehicle position is off course.

9. The positioning yaw adaptive correction method according to claim 8, characterized in that, Based on the results of the positioning yaw detection, the vehicle's position in the preset map is corrected, including: When the vehicle position in the preset map is to the right of the fork node, but the positional relationship is that the vehicle position is to the left of the fork in the target guide strip, it is determined that the vehicle position is off course. Based on the preset weight coefficient, the right driving path of the fork is reconfigured to correct the vehicle position to the left path of the fork. When the vehicle position in the preset map is to the left of the fork node, but the positional relationship is that the vehicle position is to the right of the fork in the target guide strip, the vehicle position is determined to be off course. Based on the preset weight coefficient, the permissions of the driving path to the left of the fork are reconfigured to correct the vehicle position to the path to the right of the fork.

10. A system employing the positioning yaw adaptive correction method according to any one of claims 1-9, characterized in that, The system is deployed on an electronic device; the electronic device includes sensing devices for sensing the current drivable area and vehicle position, including environmental images with lane line information; A human-computer interaction module for displaying navigation maps; A memory for storing computer programs of multiple functional modules, the memory being connected to a processor via a bus, the processor including an executable program for executing any of the computer programs of the functional modules stored in the memory; The functional modules include at least the following: The information extraction module includes a computer program for extracting guide strip boundary information based on the drivable area and lane line information; The location determination module includes a computer program for determining a target guide strip based on the guide strip boundary information and determining the positional relationship between the vehicle position and the target guide strip; The yaw detection module includes a computer program for performing positioning yaw detection based on the positional relationship and the preset map; Furthermore, the yaw correction module includes a computer program for correcting the vehicle's position in the preset map based on the results of the positioning yaw detection.

Images