Steering line generation method and vehicle navigation driving strategy determination method and device

Abstract

The invention provides a steering line generation method and a vehicle navigation driving strategy determination method and device, and relates to the technical field of computers, in particular to the technical field of artificial intelligence such as big data, intelligent traffic and automatic driving. According to the specific scheme, the method comprises the steps of firstly obtaining a first track data set associated with a target driving direction in a target intersection, then determining candidate track center lines corresponding to the target driving direction in a preset time period based on a plurality of first track sequences, then determining a direction change point sequence based on a direction angle of each track point in each candidate track center line, and obtaining a direction change point sequence; and finally, based on the direction change point sequence, generating a target steering line of the target intersection in the current target driving direction.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, in particular to the field of artificial intelligence technology such as big data, intelligent transportation, and autonomous driving, and specifically to a method for generating a turning line, and a method and device for determining a vehicle navigation driving strategy. Background Art

[0002] With the development of autonomous driving technology, road trajectory data has become an important part of the autonomous driving system. By collecting, analyzing and updating the road trajectory data of vehicles, more accurate and real-time road information can be provided to support the safety and efficiency of the autonomous driving system. In addition, due to the different sizes and complex structures of urban intersections, there is a large amount of information interaction during the autonomous driving process. The lack of lane-level map intersection turning information or inaccurate accuracy will become a bottleneck restricting the navigation and path planning of intelligent driving vehicles, leading to safety issues such as collisions during intelligent driving.

[0003] In the prior art, map data is usually collected by professional equipment and the road conditions are updated periodically. However, this method is costly and has low accuracy and real-time performance. Summary of the invention

[0004] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0005] The first embodiment of the present disclosure provides a method for generating a turning line, comprising:

[0006] Acquire a first trajectory data set associated with a target driving direction at a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period;

[0007] Based on the multiple first trajectory sequences, determining a candidate trajectory centerline corresponding to the target driving direction within the preset time period;

[0008] Determine a sequence of direction change points based on the direction angles of each trajectory point in the center line of each candidate trajectory;

[0009] Based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated.

[0010] The second aspect of the present disclosure provides a method for determining a vehicle navigation driving strategy, including:

[0011] Acquire a target turning line currently output by a turning line generation system, wherein the turning line generation system generates the target turning line by executing the method according to the first aspect of the present disclosure;

[0012] Determining a current navigation driving strategy of the vehicle based on the target turning line;

[0013] The third aspect of the present disclosure provides a device for generating a turning line, including:

[0014] A first acquisition module is used to acquire a first trajectory data set associated with a target driving direction in a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period;

[0015] A first determination module, configured to determine, based on the plurality of first trajectory sequences, a candidate trajectory centerline corresponding to the target driving direction within the preset time period;

[0016] A second determination module is used to determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory;

[0017] A generation module is used to generate a target turning line of the target intersection currently in the target driving direction based on the direction change point sequence.

[0018] The fourth aspect of the present disclosure provides a device for determining a vehicle navigation driving strategy, including:

[0019] A third acquisition module, used for acquiring a target turning line currently output by a turning line generation system, wherein the turning line generation system generates the target turning line by executing the method according to the first aspect of the present disclosure;

[0020] A third determination module is used to determine the current navigation driving strategy of the vehicle based on the target turning line;

[0021] The fifth aspect embodiment of the present disclosure proposes a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the method for generating a turning line as proposed in the first aspect embodiment of the present disclosure, or the method for determining a vehicle navigation driving strategy as proposed in the second aspect embodiment of the present disclosure.

[0022] The sixth aspect embodiment of the present disclosure proposes a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the method for generating a turning line as proposed in the first aspect embodiment of the present disclosure, or the method for determining a vehicle navigation driving strategy as proposed in the second aspect embodiment of the present disclosure.

[0023] The seventh aspect embodiment of the present disclosure proposes a computer program product, including a computer program, which, when executed by a processor, implements the method for generating a turning line as proposed in the first aspect embodiment of the present disclosure, or the method for determining a vehicle navigation driving strategy as proposed in the second aspect embodiment of the present disclosure.

[0024] The method for generating a turning line, the method for determining a vehicle navigation driving strategy and the device provided by the present disclosure have the following beneficial effects:

[0025] In the disclosed embodiment, firstly, a first trajectory data set associated with the target driving direction in the target intersection is obtained, then based on a plurality of first trajectory sequences, a candidate trajectory centerline corresponding to the target driving direction within a preset period is determined, then based on the direction angle of each trajectory point in each candidate trajectory centerline, a direction change point sequence is determined, and finally based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated. Thus, by determining the candidate trajectory centerline of the target driving direction within a preset period based on a plurality of trajectory sequences corresponding to the target driving direction in the target intersection, and generating the target turning line of the target intersection currently in the target driving direction based on the direction change point sequence of each candidate trajectory centerline, the accuracy and real-time performance of the generated turning line are improved on the basis of reducing the cost of generating the turning line.

[0026] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0028] Figure 1 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure;

[0029] Figure 2 A schematic diagram of a process for determining a candidate trajectory centerline provided in an embodiment of the present disclosure;

[0030] Figure 3 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure;

[0031] Figure 4 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure;

[0032] Figure 5 A schematic diagram of the distribution of trajectory point aggregation density provided by an embodiment of the present disclosure;

[0033] Figure 6 A schematic diagram of the distribution of trajectory point aggregation density after removing drifting trajectory points provided in an embodiment of the present disclosure;

[0034] Figure 7 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure;

[0035] Figure 8 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure;

[0036] Fig. 9 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure;

[0037] Fig.10 A schematic diagram of a target intersection obstacle provided by an embodiment of the present disclosure;

[0038] Fig.11 A flowchart of a method for determining a vehicle navigation driving strategy provided by an embodiment of the present disclosure;

[0039] Fig.12 A schematic diagram of the structure of a turning line generating device provided in an embodiment of the present disclosure;

[0040] Fig.13 A schematic diagram of the structure of a device for determining a vehicle navigation driving strategy provided by an embodiment of the present disclosure;

[0041] Fig.14 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0042] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0043] The present disclosure relates to the fields of artificial intelligence technologies such as big data and intelligent transportation.

[0044] Artificial Intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence.

[0045] Big data refers to a collection of data that cannot be captured, managed, and processed by conventional software tools within a certain time frame. It is a massive, high-growth, and diverse information asset that requires new processing models to have stronger decision-making power, insight discovery, and process optimization capabilities.

[0046] Intelligent Traffic is the effective and integrated application of advanced science and technology (information technology, computer technology, data communication technology, sensor technology, electronic control technology, automatic control theory, operations research, artificial intelligence, etc.) to transportation, service control and vehicle manufacturing, strengthening the connection between vehicles, roads and users, thus forming a comprehensive transportation that ensures safety, improves efficiency, improves the environment and saves energy.

[0047] Autonomous driving, also known as unmanned driving, computer driving or wheeled mobile robot, is a cutting-edge technology that relies on computers and artificial intelligence technology to achieve complete, safe and efficient driving without human control.

[0048] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0049] The following describes a method for generating a turning line and a method and device for determining a vehicle navigation driving strategy according to an embodiment of the present disclosure with reference to the accompanying drawings.

[0050] Figure 1 A schematic flow chart of a method for generating a turning line provided in an embodiment of the present disclosure.

[0051] like Figure 1 As shown, the method for generating the turning line may include the following steps:

[0052] Step 101: Acquire a first trajectory data set associated with a target driving direction at a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period.

[0053] The target driving direction may be any direction, for example, the target driving direction may be a straight direction, or a left turn direction, a right turn direction, etc., which is not limited in the present disclosure.

[0054] Among them, the first trajectory data set may be a multi-source trajectory data set corresponding to the target driving direction in the target intersection, which may be obtained through satellite positioning technology and / or position sensing equipment, etc., and the present disclosure does not limit this.

[0055] It should be noted that the duration of the preset time period can be pre-set, and the present disclosure does not limit this.

[0056] The first trajectory sequence may be a sequence consisting of a plurality of trajectory data in the target driving direction, which is not limited in the present disclosure.

[0057] In the present disclosure, a first trajectory data set associated with a target driving direction in a target intersection is obtained to provide a data basis for generating a turning line of the target intersection, thereby ensuring the accuracy of the generated turning line.

[0058] Step 102: determining a candidate trajectory centerline corresponding to a target driving direction within a preset time period based on a plurality of first trajectory sequences.

[0059] It should be noted that, based on multiple first trajectory sequences, the number of candidate trajectory center lines corresponding to the determined target driving direction within a preset time period may be multiple, and the present disclosure does not limit this.

[0060] In the present disclosure, after obtaining a first trajectory data set associated with a target driving direction in a target intersection, a candidate trajectory center line corresponding to the target driving direction within a preset time period can be determined based on multiple first trajectory sequences in the first trajectory data set, thereby providing conditions for determining a turning line in the target driving direction of the target intersection.

[0061] It should be noted that the method for determining the candidate trajectory center line corresponding to the target driving direction within the preset time period based on multiple first trajectory sequences can be any pre-set method. Figure 2 For example, Figure 2 A schematic diagram of a process for determining a candidate trajectory centerline according to an embodiment of the present disclosure is shown in FIG. Figure 2 As shown, based on the spatial distribution information of multiple first trajectory sequences, the overall distribution of the trajectory can be highlighted by Gaussian blurring to obtain a trajectory imaging diagram, and then a centerline extraction algorithm (such as a Hessian matrix algorithm) is used to obtain a trajectory peak diagram, and the trajectory peak diagram is regarded as a three-dimensional surface (such as a peak). Finally, the candidate trajectory centerline is determined according to the local curvature (as shown by the ridge in the trajectory peak diagram), and the present disclosure does not limit this.

[0062] Step 103: Determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory.

[0063] The direction change point sequence may be a sequence of trajectory points whose directions change in the center line of the candidate trajectory.

[0064] In the present disclosure, after determining the candidate trajectory center lines corresponding to the target driving direction within a preset time period based on multiple first trajectory sequences, the direction angles of each trajectory point in each candidate trajectory center line can be first determined, and then the direction change point sequence of each candidate trajectory center line can be determined through the changes in the direction angles of each trajectory point in each candidate trajectory center line, thereby improving the accuracy of the generated turning line.

[0065] Optionally, after obtaining the candidate trajectory centerline corresponding to the target driving direction within a preset time period, the direction angle of each trajectory point in each candidate trajectory centerline can be first determined, and then, when the direction angle corresponding to the trajectory point in the candidate trajectory centerline is greater than a direction angle threshold, it can be determined that the direction of the trajectory point has changed compared to the directions of other trajectory points in the candidate trajectory centerline. At this time, the trajectory point can be determined as a direction change point, and finally all the direction change points in the same trajectory centerline are determined as a direction change point sequence, thereby improving the accuracy of the determined direction change point sequence of the candidate trajectory centerline.

[0066] The direction angle threshold may be a critical value of the direction angle used to determine whether a trajectory point in the candidate trajectory centerline is a direction change point, which may be preset and is not limited in the present disclosure.

[0067] Step 104 , generating a target turning line at the target intersection in the target driving direction based on the direction change point sequence.

[0068] The target turning line may be the final turning line of the target intersection in the target driving direction.

[0069] In the present disclosure, after determining the direction change point sequence of each candidate trajectory centerline, a target turning line of the target intersection in the target driving direction can be generated based on the direction change point sequence, thereby improving the accuracy and real-time performance of the generated target turning line.

[0070] In the disclosed embodiment, firstly, a first trajectory data set associated with the target driving direction in the target intersection is obtained, then based on a plurality of first trajectory sequences, a candidate trajectory centerline corresponding to the target driving direction within a preset period is determined, then based on the direction angle of each trajectory point in each candidate trajectory centerline, a direction change point sequence is determined, and finally based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated. Thus, by determining the candidate trajectory centerline of the target driving direction within a preset period based on a plurality of trajectory sequences corresponding to the target driving direction in the target intersection, and generating the target turning line of the target intersection currently in the target driving direction based on the direction change point sequence of each candidate trajectory centerline, the accuracy and real-time performance of the generated turning line are improved on the basis of reducing the cost of generating the turning line.

[0071] Figure 3 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure.

[0072] like Figure 3 As shown, the method for generating the turning line may include the following steps:

[0073] Step 301, based on the first position information of each trajectory point in the multi-source trajectory data obtained within a preset time period and the second position information of the target intersection, a second trajectory data set associated with the target intersection is obtained from the multi-source trajectory data.

[0074] Among them, the first location information and the second location information are respectively the longitude and latitude coordinates of the trajectory point in the multi-source trajectory data in the geographic coordinate system, and the longitude and latitude coordinates of the target intersection in the geographic coordinate system, which is not limited in the present disclosure.

[0075] The second trajectory data set may be a data set including trajectories in any driving direction of the target intersection.

[0076] In the present disclosure, firstly, crowd-source trajectory data within a preset time period is obtained, and then based on the first position information of each trajectory point in the crowd-source trajectory data and the second position information of the target intersection, the second trajectory data set associated with the target intersection is determined from the crowd-source trajectory data by any method. For example, after obtaining the first position information of each trajectory point in the crowd-source trajectory data and the second position information of the target intersection, the trajectory point data whose corresponding first position information is within a certain distance range (such as 50 meters) of the second position information can be determined as the trajectory point data in the second trajectory data set associated with the target intersection, and the present disclosure does not limit this.

[0077] It should be noted that the trajectory points in the acquired second trajectory data set may include not only the position information of the trajectory points, but also the timestamp, linear speed, direction angle and other information of the trajectory points, which is not limited in the present disclosure.

[0078] Optionally, after obtaining the crowd-source trajectory data of the target intersection, due to possible problems such as navigation positioning errors, coordinate system conversion errors, map accuracy errors, etc., the trajectory points in the crowd-source trajectory data deviate from the vehicle's driving road. In order to determine the exact position of the trajectory point in the road network topology structure, the crowd-source trajectory data can be matched to the actual road network topology structure of the target intersection, and a hidden Markov Model (HMM) model can be used for trajectory matching. Since the trajectory points are observation sequences and the roads actually traveled by the trajectories are hidden sequences, at this time, the Viterbi algorithm can be used to match the trajectory points to the roads traveled by the trajectories, thereby improving the accuracy of the crowd-source trajectory data. The present disclosure does not limit this.

[0079] Optionally, after obtaining the crowd-source trajectory data of the target intersection, since a large amount of repeated trajectory point data (such as stop points or too close trajectory points) may be generated in a small space when the vehicle is stopped or driving at a low speed, resulting in an excessive density of trajectory points, at this time, the present disclosure can perform thinning and smoothing operations on the crowd-source trajectory data to address the above problem. For point sets with repeated or too close timestamps and / or positions in the trajectory data, the first trajectory point is retained and other redundant trajectory point data are deleted, thereby improving the accuracy of the crowd-source trajectory data. The present disclosure is not limited to this.

[0080] Optionally, after obtaining the crowd-source trajectory data of the target intersection, since the crowd-source trajectory data may include non-motor vehicle data (such as turning on navigation during riding), it may also include trajectory data of reverse illegal driving. At this time, the present disclosure may, according to relevant regulations, such as the speed of the non-motor vehicle is less than or equal to the speed threshold (which can be pre-set), check the maximum speed, average speed, minimum speed, probability of the trajectory being a motor vehicle, trajectory confidence and other characteristic parameters in the crowd-source trajectory data during its life cycle. Trajectory data that meets the corresponding preset threshold can be retained, otherwise it can be determined as non-motor vehicle trajectory data and deleted, thereby improving the accuracy of the crowd-source trajectory data. The present disclosure does not limit this.

[0081] It should be noted that after obtaining the multi-source trajectory data of the target intersection, the multi-source trajectory data can also be filtered based on multi-dimensional feature parameters such as the trajectory shape (such as the shape is a ball), the number of trajectory points, the number of trajectory points with large angle changes, etc. in the multi-source trajectory data to improve the quality of the trajectory data. The present disclosure does not limit this.

[0082] Step 302: acquiring candidate trajectory data associated with the target driving direction from the second trajectory dataset according to the direction angle corresponding to each trajectory point in the second trajectory dataset.

[0083] In the present disclosure, after obtaining the second trajectory data set associated with the target intersection from the crowd-source trajectory data, the candidate trajectory data associated with the target driving direction can be obtained from the second trajectory data set by any method based on the direction angle corresponding to each trajectory point in the second trajectory data set. For example, the candidate trajectory data associated with the target driving direction can be determined based on the angle between the direction angle corresponding to each trajectory point and the target driving direction, or the direction angle range can be determined based on the target driving direction, and all trajectory points within the direction angle range are determined as candidate trajectory data, etc., so as to improve the reliability of the determined candidate trajectory data associated with the target driving direction, and the present disclosure does not limit this.

[0084] Step 303 : Segment the candidate trajectory data into a plurality of trajectory segments according to a preset distance value along the target driving direction.

[0085] The distance value may be a distance value used to segment the candidate trajectory data, and may be any pre-set distance value. For example, the distance value may be 10 meters, etc., which is not limited in the present disclosure.

[0086] In the present disclosure, after the candidate trajectory data associated with the target driving direction is obtained from the second trajectory data set, the trajectory data is a point sample data presenting the distribution of road surface characteristics. The candidate trajectory data can be divided into multiple trajectory segments along the target driving direction at a preset distance value, such as every 10 meters. The present disclosure does not limit this.

[0087] It should be noted that when the candidate trajectory data is divided into multiple trajectory segments, due to possible equipment failure or signal interruption when acquiring the multi-source trajectory data, the vehicle position information cannot be obtained, which may cause the trajectory point data in the candidate trajectory data to be missing, making the distance between adjacent trajectory points of the trajectory far, or the local trajectory vector and the driving direction angle of the entire trajectory differ greatly, presenting a discontinuous driving state. At this time, the trajectory can be cut and segmented. In addition, the sampling time difference threshold (which can be pre-set) and the sampling distance threshold (which can be pre-set) between adjacent trajectory points in the acquired trajectory can also be pre-set. When the time difference between adjacent trajectory points of the trajectory is greater than the sampling time difference threshold, or the distance between adjacent trajectory points is greater than the sampling distance threshold, it can be determined that the current driving state of the vehicle has changed. At this time, the trajectory can be cut and segmented, and the present disclosure does not limit this.

[0088] Step 304, determining the drift trajectory points and lane change trajectory points in each trajectory segment.

[0089] In the present disclosure, after the candidate trajectory data is segmented into a plurality of trajectory segments, the drift trajectory points and the lane change trajectory points within each trajectory segment may be determined, thereby improving the accuracy of the generated turning line.

[0090] It should be noted that the drift trajectory points and lane change trajectory points contained in each trajectory segment are different. Among them, each trajectory point may contain a drift trajectory point, or may also contain a lane change trajectory point, or may also contain a drift trajectory point and a lane change trajectory point, and the present disclosure does not limit this.

[0091] Optionally, after the candidate trajectory data is divided into multiple trajectory segments, the clustering density corresponding to each trajectory point in each trajectory segment can be first determined based on the position information of the trajectory point in the trajectory segment, wherein the clustering density corresponding to each trajectory point is determined by the number of trajectory points whose distance therefrom is less than a distance threshold, that is, the clustering density corresponding to each trajectory point can also be considered as the number of trajectory points around the trajectory point (within the range where the distance from the trajectory point is less than the distance threshold), and then the trajectory points in the trajectory segment whose corresponding clustering density is less than the density threshold are determined as drift trajectory points, thereby improving the accuracy of the determined drift trajectory points.

[0092] The distance threshold may be a distance critical value used to determine the clustering density corresponding to each trajectory point, which may be preset and is not limited in the present disclosure.

[0093] Among them, the density threshold can be a critical value of the aggregation density used to determine whether a trajectory point in a trajectory segment is a drift trajectory point. It can be pre-set, or it can be determined in real time by a clustering algorithm based on the trajectory data currently associated with the target intersection. The present disclosure does not limit this.

[0094] It should be noted that different target intersections may have different corresponding density thresholds, or may be the same, and the present disclosure does not limit this.

[0095] Step 305 , remove the drift trajectory points and lane change trajectory points in the candidate trajectory data to obtain a first trajectory data set associated with the target driving direction.

[0096] In the present disclosure, after determining the drift trajectory points and lane change trajectory points in each trajectory segment, the drift trajectory points and lane change trajectory points in the candidate trajectory data are eliminated to obtain a first trajectory data set associated with the target driving direction, thereby improving the reliability of the obtained first trajectory data set and providing conditions for improving the accuracy and efficiency of turning line generation.

[0097] It should be noted that after determining the drift trajectory points and lane change trajectory points in the candidate trajectory data, the drift trajectory points and lane change trajectory points can be refined and eliminated based on the DBSCAN clustering model, and the present disclosure does not limit this.

[0098] Among them, DBSCAN is the abbreviation of Density-Based Spatial Clustering of Applications with Noise.

[0099] Step 306 : determining a candidate trajectory centerline corresponding to the target driving direction within a preset time period based on the plurality of first trajectory sequences included in the first trajectory data set.

[0100] Step 307: Determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory.

[0101] Step 308: Generate a target turning line at the target intersection in the target driving direction based on the direction change point sequence.

[0102] The specific implementation forms of steps 306 to 308 may refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0103] In the disclosed embodiment, firstly, based on the first position information of each trajectory point in the crowd-source trajectory data obtained within a preset time period and the second position information of the target intersection, a second trajectory data set associated with the target intersection is obtained from the crowd-source trajectory data, and according to the direction angle corresponding to each trajectory point in the second trajectory data set, candidate trajectory data associated with the target driving direction is obtained from the second trajectory data set, and then along the target driving direction, the candidate trajectory data is divided into a plurality of trajectory segments with a preset distance value, and then the drift trajectory points and lane change trajectory points in each trajectory segment are determined, and the drift trajectory points and lane change trajectory points in the candidate trajectory data are removed to obtain the first trajectory data set associated with the target driving direction, and based on the plurality of first trajectory sequences included in the first trajectory data set, the candidate trajectory center line corresponding to the target driving direction within the preset time period is determined, and finally based on the direction angle of each trajectory point in each candidate trajectory center line, a direction change point sequence is determined, and based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated. Therefore, a second trajectory data set associated with the target intersection is determined based on the position information of each trajectory point in the multi-source trajectory data within a preset time period and the position information of the target intersection, and candidate trajectory data associated with the target driving direction is determined according to the direction angle corresponding to each trajectory point in the second trajectory data set, and the drift trajectory points and lane change trajectory points in the candidate trajectory data are eliminated to obtain a plurality of first trajectory sequences associated with the target driving direction, and based on the plurality of first trajectory sequences, the candidate trajectory center line corresponding to the target driving direction within the preset time period is determined, and based on the direction change point sequence corresponding to the candidate trajectory center line, the target turning line of the target intersection currently in the target driving direction is generated, thereby reducing the cost of generating the turning line and improving the accuracy and efficiency of the generated turning line.

[0104] Figure 4 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure.

[0105] like Figure 4 As shown, the method for generating the turning line may include the following steps:

[0106] Step 401, based on the first position information of each trajectory point in the multi-source trajectory data obtained within a preset time period and the second position information of the target intersection, a second trajectory data set associated with the target intersection is obtained from the multi-source trajectory data.

[0107] Step 402 : acquiring candidate trajectory data associated with the target driving direction from the second trajectory dataset according to the direction angle corresponding to each trajectory point in the second trajectory dataset.

[0108] Step 403 : segment the candidate trajectory data into a plurality of trajectory segments according to a preset distance value along the target driving direction.

[0109] The specific implementation forms of steps 401 to 403 may refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0110] Step 404, based on the position information of the trajectory points in each trajectory segment, determine the road width occupied by all the trajectory points in the trajectory segment and the clustering density corresponding to each trajectory point, wherein the clustering density corresponding to each trajectory point is determined by the number of trajectory points whose distance to it is less than a distance threshold.

[0111] In the present disclosure, after the candidate trajectory data is divided into multiple trajectory segments, the road width occupied by all trajectory points in the trajectory segment and the clustering density corresponding to each trajectory point can be determined based on the position information of the trajectory points in each trajectory segment, thereby providing a data basis for determining the drift trajectory points in each trajectory segment.

[0112] It should be noted that, for different trajectory segments, the road width occupied by all trajectory points in a determined trajectory segment and the clustering density corresponding to each trajectory point may be different, or may be the same, and the present disclosure does not limit this.

[0113] Step 405, determining the clustering density distribution of the trajectory points within the road surface width according to the clustering density corresponding to all the trajectory points in the trajectory segment.

[0114] In the present disclosure, after determining the road width occupied by all trajectory points in each trajectory segment and the clustering density corresponding to each trajectory point, the clustering density distribution of the trajectory points within the road width can be determined based on the clustering density corresponding to all trajectory points in the trajectory segment, wherein the determined clustering density distribution of the trajectory points within the road width can include the clustering density distribution of the trajectory points within the road surface, and can also include the clustering density distribution of the trajectory points at the road boundary, which is not limited in the present disclosure.

[0115] Step 406, determining a track width threshold in the target driving direction according to the number of lanes at the target intersection in the target driving direction.

[0116] The track width threshold may be a track width critical value used to determine a non-drifting track point within a track segment.

[0117] It should be noted that the more lanes there are in the target driving direction, the larger the trajectory width threshold of the determined target driving direction may be, and the present disclosure does not limit this.

[0118] In the present disclosure, after determining the clustering density distribution of trajectory points within the road width, the trajectory width threshold of the target driving direction can be determined based on the number of lanes at the target intersection in the target driving direction, thereby improving the accuracy of the determined drift trajectory points.

[0119] Step 407, determining the position range of the non-drifting trajectory points according to the trajectory width threshold and the clustering density distribution.

[0120] In the present disclosure, after determining the trajectory width threshold of the target driving direction, the location range of the non-drifting trajectory points can be determined based on the trajectory width threshold of the target driving direction and the clustering density distribution of all trajectory points in each trajectory segment, thereby providing a data basis for determining the drifting trajectory points in the trajectory segment.

[0121] It should be noted that when the track width threshold is less than the road width, the position range of the track points with a larger clustering density distribution within the road width and located within the track width threshold range can be determined as the position range of the non-drifting track points. For example, when the track width threshold is 3.5 meters and the road width is 4 meters, since the road width is greater than the track width threshold, at this time, the range occupied by the track points with a larger clustering density distribution within the road width range in the track segment can be first determined, and then the track points within the first 3.5 meters of the range can be determined as non-drifting track points, and the range they occupy can be determined as the position range of the non-drifting track points, and the present disclosure does not limit this.

[0122] It should be noted that after determining the clustering density distribution of the track points within the road width, the corresponding clustering density distribution map can be generated according to the clustering density distribution of the track points within the road width, and then the position range of the non-drifting track points can be determined according to the track width threshold and the clustering density distribution map. For example, Figure 5 As shown, Figure 5 A schematic diagram of the distribution of trajectory point aggregation density provided by an embodiment of the present disclosure, comprising Figure 5 Available, Figure 5The clustering density of the light-colored trajectory points in the image is larger than that of the dark-colored trajectory points. At this time, the light-colored trajectory points can be determined as non-drifting trajectory points, and their location range is the location range of the non-drifting trajectory points, thereby realizing a visual display of the distribution of trajectory point clustering density and improving the efficiency of determining the location range of the non-drifting trajectory points. The present disclosure does not limit this.

[0123] Step 408: determine the trajectory points in the trajectory segment that are outside the position range as drift trajectory points.

[0124] In the present disclosure, after determining the position range of the non-drifting trajectory points of a trajectory segment, in order to improve the quality of the trajectory points in the trajectory segment, the trajectory points in the trajectory segment that are outside the position range can be determined as drifting trajectory points, for example, Figure 5 The clustering density of the dark-colored trajectory points in the image is smaller than that of the light-colored trajectory points. At this time, the dark-colored trajectory points can be determined as drift trajectory points, so that the drift trajectory points in the trajectory segment can be removed to reduce the noise trajectory points in the trajectory segment.

[0125] Step 409, determining the lane change trajectory point in each trajectory segment.

[0126] Step 410 , remove the drift trajectory points and lane change trajectory points in the candidate trajectory data to obtain a first trajectory data set associated with the target driving direction.

[0127] In the present disclosure, after removing the drifting trajectory points and lane changing trajectory points in the trajectory segment, the noise in the candidate trajectory data can be reduced and the trajectory data quality can be improved. Figure 6 As shown, Figure 6 A schematic diagram of the distribution of trajectory point aggregation density after removing drift trajectory points provided by an embodiment of the present disclosure, Figure 6 It can be seen that after removing the drift trajectory points, the quality of the candidate trajectory data is greatly improved, and the present disclosure does not limit this.

[0128] Step 411 : determining a candidate trajectory centerline corresponding to the target driving direction within a preset time period based on a plurality of first trajectory sequences included in the first trajectory data set.

[0129] Step 412: Determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory.

[0130] Step 413, based on the direction change point sequence, generate a target turning line at the target intersection in the current target driving direction.

[0131] The specific implementation forms of steps 409 to 413 may refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0132] In the embodiment of the present disclosure, firstly, based on the first position information of each trajectory point in the crowd-source trajectory data obtained within a preset time period and the second position information of the target intersection, a second trajectory data set associated with the target intersection is obtained from the crowd-source trajectory data, and according to the direction angle corresponding to each trajectory point in the second trajectory data set, candidate trajectory data associated with the target driving direction is obtained from the second trajectory data set, and then along the target driving direction, the candidate trajectory data is divided into a plurality of trajectory segments with a preset distance value, and according to the position information of the trajectory points in each trajectory segment, the road width occupied by all the trajectory points in the trajectory segment and the clustering density corresponding to each trajectory point are determined, and then according to the clustering density corresponding to all the trajectory points in the trajectory segment, the clustering density distribution of the trajectory points within the road width is determined, and then according to the target intersection The number of lanes in the target driving direction is used to determine the track width threshold of the target driving direction. According to the track width threshold and the clustering density distribution, the position range of the non-drifting track points is determined, and the track points outside the position range in the track segment are determined as drifting track points. The lane-changing track points in each track segment are determined, and the drifting track points and the lane-changing track points in the candidate track data are eliminated to obtain a first track data set associated with the target driving direction. Finally, based on a plurality of first track sequences included in the first track data set, the candidate track center line corresponding to the target driving direction within a preset time period is determined, and based on the direction angle of each track point in each candidate track center line, a direction change point sequence is determined. Based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated. Therefore, after determining multiple trajectory segments corresponding to candidate trajectory data associated with the target driving direction based on multi-source trajectory data within a preset time period, the drift trajectory points in the trajectory segment are determined according to the clustering density distribution of each trajectory point within the road width occupied by all trajectory points in each trajectory segment, and the trajectory width threshold of the target driving direction. The drift trajectory points and lane change trajectory points in the trajectory segment are eliminated to obtain a target trajectory data set associated with the target driving direction, and based on the trajectory data set, a target turning line in the target driving direction is generated, thereby improving the quality of the trajectory data and improving the quality and reliability of the generated turning line.

[0133] Figure 7 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure.

[0134] like Figure 7 As shown, the method for generating the turning line may include the following steps:

[0135] Step 701, based on the first position information of each trajectory point in the multi-source trajectory data obtained within a preset time period and the second position information of the target intersection, a second trajectory data set associated with the target intersection is obtained from the multi-source trajectory data.

[0136] Step 702: Acquire candidate trajectory data associated with the target driving direction from the second trajectory dataset according to the direction angle corresponding to each trajectory point in the second trajectory dataset.

[0137] Step 703 : Segment the candidate trajectory data into a plurality of trajectory segments along the target driving direction with a preset distance value.

[0138] Step 704, determine the drift trajectory points within each trajectory segment.

[0139] The specific implementation forms of steps 701 to 704 can refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0140] Step 705: determine the direction difference threshold of the trajectory segment.

[0141] Among them, the direction difference threshold can be a direction difference critical value used to determine whether a trajectory point in a trajectory segment is a lane change trajectory point. It can be pre-set or determined according to actual conditions, and the present disclosure does not limit this.

[0142] It should be noted that different trajectory segments may have different corresponding direction difference thresholds, or may be the same, and the present disclosure does not limit this.

[0143] Optionally, after obtaining multiple trajectory segments corresponding to the candidate trajectory data, in order to improve the accuracy of determining the lane changing trajectory point within the trajectory segment, the direction difference threshold of the trajectory segment can be determined. When determining the direction difference threshold, the first angle difference and the second angle difference corresponding to each trajectory point can be first determined based on the direction angle corresponding to each trajectory point in the trajectory segment, wherein the first angle difference is the direction angle difference between the trajectory point and the adjacent previous trajectory point, and the second angle difference is the direction angle difference between the trajectory point and the adjacent subsequent trajectory point. Then, based on the first angle difference and the second angle difference corresponding to all the trajectory points in the trajectory segment, the direction difference threshold of the trajectory segment is determined, thereby improving the accuracy of the determined direction difference threshold and providing conditions for determining the lane changing trajectory point within the trajectory segment.

[0144] It should be noted that, when determining the direction difference threshold of the trajectory segment based on the first angle difference and the second angle difference corresponding to all the trajectory points in the trajectory segment, the difference between the first angle difference and the second angle difference of each trajectory point in the trajectory segment can be first determined, and then the average of all the differences can be determined based on the difference between the first angle difference and the second angle difference of each trajectory point. Finally, based on the average, the preset direction difference threshold is corrected to obtain the current direction difference threshold of the trajectory segment. The present disclosure does not limit this.

[0145] It should be noted that, when determining the direction difference threshold of the trajectory segment based on the first angle difference and the second angle difference corresponding to all the trajectory points in the trajectory segment, the difference between the first angle difference and the second angle difference of each trajectory point in the trajectory segment can be first determined, and then the mean of all the differences can be determined based on the difference between the first angle difference and the second angle difference of each trajectory point. In the case where different means correspond to different correction angles, the preset direction difference threshold can also be corrected based on the correction angle corresponding to the mean to obtain the current direction difference threshold of the trajectory segment. The present disclosure does not limit this.

[0146] In the present disclosure, after obtaining the candidate trajectory data associated with the target driving direction and its corresponding multiple trajectory segments, after determining the drift trajectory points within each trajectory segment, in order to further improve the trajectory data quality within each trajectory segment, the lane changing trajectory points within each trajectory segment can also be determined. At this time, the direction difference threshold of each trajectory segment can be first determined, thereby providing conditions for determining the lane changing trajectory points within the trajectory segment.

[0147] Step 706: The trajectory point whose corresponding direction angle in the trajectory segment is greater than the direction difference threshold is determined as a lane change trajectory point.

[0148] In the present disclosure, after determining the direction difference threshold of the trajectory segment, when the direction angle of a trajectory point within the trajectory segment is greater than the direction difference threshold, it can be determined that the corresponding trajectory point has changed lanes within the trajectory segment. At this time, the trajectory point can be determined as a lane changing trajectory point, thereby improving the accuracy of the determined lane changing trajectory point.

[0149] Step 707 , remove the drift trajectory points and lane change trajectory points in the candidate trajectory data to obtain a first trajectory data set associated with the target driving direction.

[0150] Step 708 : determining a candidate trajectory centerline corresponding to the target driving direction within a preset time period based on the plurality of first trajectory sequences included in the first trajectory data set.

[0151] Step 709: Determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory.

[0152] Step 710, based on the direction change point sequence, generate a target turning line at the target intersection in the current target driving direction.

[0153] The specific implementation forms of steps 707 to 710 may refer to the detailed descriptions in other embodiments of the present disclosure and will not be described in detail here.

[0154] In the disclosed embodiment, firstly, based on the first position information of each trajectory point in the crowd-source trajectory data obtained within a preset time period and the second position information of the target intersection, a second trajectory data set associated with the target intersection is obtained from the crowd-source trajectory data, and according to the direction angle corresponding to each trajectory point in the second trajectory data set, candidate trajectory data associated with the target driving direction is obtained from the second trajectory data set, and along the target driving direction, the candidate trajectory data is divided into a plurality of trajectory segments with a preset distance value, and the drift trajectory points in each trajectory segment are determined, and then the direction difference threshold of the trajectory segment is determined, and the trajectory points whose corresponding direction angles in the trajectory segment are greater than the direction difference threshold are determined as lane change trajectory points, and then the drift trajectory points and lane change trajectory points in the candidate trajectory data are eliminated to obtain a first trajectory data set associated with the target driving direction, and based on the plurality of first trajectory sequences included in the first trajectory data set, the candidate trajectory center line corresponding to the target driving direction within the preset time period is determined, and finally based on the direction angles of each trajectory point in the center line of each candidate trajectory, a direction change point sequence is determined, and based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated. Therefore, after determining multiple trajectory segments corresponding to candidate trajectory data associated with the target driving direction and drift trajectory points in each trajectory segment based on multi-source trajectory data within a preset time period, the trajectory points in each trajectory segment whose corresponding direction angles are greater than the direction difference threshold of the trajectory segment are determined as lane change trajectory points, and the drift trajectory points and lane change trajectory points are extracted from the trajectory segments to obtain a target trajectory data set associated with the target driving direction, and based on the target trajectory data set, a target turning line in the target driving direction is generated, thereby improving the quality of the trajectory data and improving the quality and reliability of the generated turning line.

[0155] Figure 8 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure.

[0156] like Figure 8 As shown, the method for generating the turning line may include the following steps:

[0157] Step 801: Acquire a first trajectory data set associated with a target driving direction at a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period.

[0158] Step 802: Determine the candidate trajectory centerline corresponding to the target driving direction within a preset time period based on the multiple first trajectory sequences.

[0159] Step 803: Determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory.

[0160] The specific implementation forms of steps 801 to 803 can refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0161] Step 804, determining the length and curvature of the target intersection and the attributes of the road section to which the target intersection belongs.

[0162] The attributes of the road section may include any attributes of the target intersection, which may be preset. For example, the attributes of the road section to which the target intersection belongs may include attributes such as the road grade of the target intersection, the intersection scene (such as a waiting area, an elevated road, etc.), etc., which are not limited in this disclosure.

[0163] It should be noted that, for different target intersections, the length, curvature, and attributes of the road section to which the target intersection belongs may be different, and the present disclosure does not limit this.

[0164] In the present disclosure, after determining the sequence of direction change points of the center line of each candidate trajectory, in order to improve the accuracy of the generated turning line of the target intersection, the length, curvature, and attributes of the road end to which the target intersection belongs can be first determined, thereby providing a data basis for determining the turning line of the target intersection.

[0165] Step 805, using each trajectory point in the direction change point sequence as a control point, based on the length, curvature and attributes of the target intersection, a candidate turning line of the target intersection in the target driving direction is generated.

[0166] In the present disclosure, after determining the length, curvature, and attributes of the road section to which the target intersection belongs, in order to make the generated turning line of the target intersection fit the actual situation of the target intersection, after obtaining the direction change point sequence corresponding to each candidate trajectory center line of the target intersection in the current driving direction, the trajectory points in each direction change point sequence can be used as control points, and based on the length, curvature and attributes of the target intersection, a candidate turning line of the target intersection currently in the target driving direction can be generated by any preset method. For example, the candidate turning line corresponding to each candidate trajectory center line can be generated by a B-spline curve algorithm, and the present disclosure does not limit this.

[0167] It should be noted that the control points in each direction change point sequence are different, and the generated candidate turning lines may be different, which is not limited in the present disclosure.

[0168] Step 806, determining a constraint parameter value corresponding to each candidate turning line, wherein the constraint parameter is at least one of the following parameters: curvature, angle with the associated lane, and distance with the associated lane.

[0169] The constraint parameter values ​​may be determined by a motion planning algorithm based on dynamic constraints (such as the KinoDynamicRRT* algorithm), or may be pre-set, which is not limited in the present disclosure.

[0170] Among them, KinoDynamic is a compound word of kinematic and dynamics.

[0171] Among them, RRT is the abbreviation of Rapidly-exploring Random Tree (RRT).

[0172] In the present disclosure, after generating multiple candidate turning lines at the target intersection in the target driving direction, in order to improve the accuracy of the target turning line determined from the candidate turning lines, the constraint parameter value corresponding to each candidate turning line can be first determined before determining the target turning line.

[0173] Step 807 , based on a preset constraint parameter threshold and a constraint parameter value corresponding to each candidate turning line, a target turning line is determined from a plurality of candidate turning lines.

[0174] Among them, the constraint parameter threshold value can be a constraint parameter critical value used to determine the target turning line from multiple candidate turning lines, which can be at least one of the following parameter thresholds: curvature threshold, angle threshold with associated lanes, and distance threshold with associated lanes, which can be any pre-set value. For example, when the constraint parameter threshold is the curvature threshold, it can be 0.169, when the constraint parameter threshold is the angle threshold with associated lanes, it can be 5 degrees, when the constraint parameter threshold is the distance threshold with associated lanes, it can be 0.4 meters, etc., and the present disclosure does not limit this.

[0175] In the present disclosure, after determining the constraint parameter value corresponding to each candidate turn line, a target turn line can be determined from multiple candidate turn lines based on a preset constraint parameter threshold and the constraint parameter value corresponding to each candidate turn line, so that the determined target turn line is more consistent with the actual situation of the target intersection, and the accuracy of the determined target turn line is improved. For example, when the constraint parameter values ​​corresponding to each candidate turn line include curvature, angle with the associated lane, and distance with the associated lane, and the corresponding constraint parameter thresholds include a curvature threshold of 0.169, an angle threshold of 5 degrees with the associated lane, and a distance threshold of 0.4 meters with the associated lane, the candidate turn line with a corresponding curvature less than 0.169, a corresponding angle with the associated lane less than 5 degrees, and a corresponding distance with the associated lane less than 0.4 meters can be determined as the target turn line, and the present disclosure does not limit this.

[0176] In the disclosed embodiment, a first trajectory data set associated with a target driving direction in a target intersection is first acquired, and based on a plurality of first trajectory sequences, a candidate trajectory centerline corresponding to the target driving direction within a preset time period is determined, and then a direction change point sequence is determined based on the direction angle of each trajectory point in each candidate trajectory centerline, and then the length, curvature, and attributes of the road section to which the target intersection belongs are determined, and the trajectory points in each direction change point sequence are respectively used as control points, and based on the length, curvature and attributes of the target intersection, a candidate turning line of the target intersection currently in the target driving direction is generated, and finally, a constraint parameter value corresponding to each candidate turning line is determined, and based on a preset constraint parameter threshold and the constraint parameter value corresponding to each candidate turning line, a target turning line is determined from a plurality of candidate turning lines. Therefore, after determining the direction change point sequence of each candidate trajectory center line associated with the target intersection in the target driving direction, the trajectory points in each square change point sequence are used as control points, and the corresponding candidate turning lines are generated based on the length, curvature and attributes of the target intersection. Based on the relationship between the constraint parameter value and the constraint parameter threshold of each candidate turning line, the target turning line is determined from the candidate turning lines, thereby improving the accuracy of the generated target turning line while ensuring that the determined target turning line is consistent with the actual situation of the target intersection.

[0177] Fig. 9 A schematic flow chart of a method for generating a turning line provided in one embodiment of the present disclosure.

[0178] like Fig. 9 As shown, the method for generating the turning line may include the following steps:

[0179] Step 901: Acquire a first trajectory data set associated with a target driving direction at a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period.

[0180] Step 902: Determine the candidate trajectory centerline corresponding to the target driving direction within a preset time period based on the multiple first trajectory sequences.

[0181] Step 903: Determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory.

[0182] Step 904, determining the length and curvature of the target intersection, and the attributes of the road section to which the target intersection belongs.

[0183] Step 905, using each trajectory point in the direction change point sequence as a control point, based on the length, curvature and attributes of the target intersection, a candidate turning line of the target intersection in the target driving direction is generated.

[0184] Step 906, determining a constraint parameter value corresponding to each candidate turning line, wherein the constraint parameter is at least one of the following parameters: curvature, angle with the associated lane, and distance with the associated lane.

[0185] Step 907 , based on a preset constraint parameter threshold and a constraint parameter value corresponding to each candidate turning line, determine a target turning line from a plurality of candidate turning lines.

[0186] Among them, the specific implementation forms of steps 901 to 907 can refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0187] Step 908, obtaining the image data currently corresponding to the target intersection.

[0188] It should be noted that the image data currently corresponding to the target intersection may be image data including the current real-time road conditions of the target intersection, and the present disclosure does not limit this.

[0189] In the present disclosure, after determining the target turning line of the target intersection, in order to ensure the real-time and reliability of the determined target turning line, the image data currently corresponding to the target intersection may be first acquired.

[0190] Step 909: perform obstacle recognition on the image data to determine the obstacle information currently corresponding to the target intersection, wherein the obstacle information includes the position and shape of the obstacle.

[0191] In the present disclosure, after obtaining the image data corresponding to the target intersection, in order to improve the reliability and real-time performance of the target turning line, the image data may be firstly subjected to obstacle recognition to determine the obstacle information corresponding to the target intersection, such as Fig.10 As shown, Fig.10 A schematic diagram of a target intersection obstacle provided by an embodiment of the present disclosure, Fig.10 In the figure, the light-colored blocks are obstacles at the target intersection, and the dotted line is the target turning line at the target intersection. Fig.10 The content is only an example and is not limited to this disclosure.

[0192] Step 910: Update the target turning line according to the obstacle information.

[0193] In the present disclosure, after determining the obstacle information currently corresponding to the target intersection, the passable area within the target intersection can be determined based on the obstacle information and the position of the obstacle relative to the target intersection (located in the front, middle, and rear of the target intersection), and the target turning line can be updated, so that the target turning line is consistent with the real-time road conditions of the target intersection, effectively avoiding collisions between vehicles and obstacles at the target intersection, and improving the reliability and real-time performance of the target turning line.

[0194] Optionally, after updating the target turning line according to the obstacle information of the target intersection, the line distance difference result between the target turning line and the existing turning line in the same driving direction of the target intersection can also be determined. When the differential result is greater than the differential result threshold, the turning line in the same driving direction of the target intersection is updated based on the target turning line. The differential result threshold can be a differential result critical value used to determine whether to update the existing turning line of the target intersection. It can be pre-set, for example, it can be 0.3 meters, etc., and the present disclosure is not limited to this.

[0195] In the disclosed embodiment, a first trajectory data set associated with a target driving direction in a target intersection is first obtained, and based on a plurality of first trajectory sequences, a candidate trajectory centerline corresponding to the target driving direction within a preset time period is determined, and based on the direction angle of each trajectory point in each candidate trajectory centerline, a direction change point sequence is determined, and then the length, curvature, and attributes of the road section to which the target intersection belongs are determined, and the trajectory points in each direction change point sequence are respectively used as control points, and based on the length, curvature and attributes of the target intersection, a candidate turning line of the target intersection currently in the target driving direction is generated, and then the constraint parameter value corresponding to each candidate turning line is determined, and based on a preset constraint parameter threshold and the constraint parameter value corresponding to each candidate turning line, a target turning line is determined from a plurality of candidate turning lines, and finally image data currently corresponding to the target intersection is obtained, and obstacle recognition is performed on the image data to determine obstacle information currently corresponding to the target intersection, and the target turning line is updated according to the obstacle information. Therefore, after generating the target turning line at the target intersection in the target driving direction, obstacle recognition is performed on the current image data of the target intersection to determine the obstacle information, and the target turning line is updated based on the obstacle information, thereby improving the real-time and reliability of the generated target turning line.

[0196] Fig.11 A flowchart of a method for determining a vehicle navigation driving strategy provided by an embodiment of the present disclosure is provided.

[0197] like Fig.11 As shown, the method for determining the vehicle navigation driving strategy may include the following steps:

[0198] Step 1101, obtaining the target turning line currently output by the turning line generation system.

[0199] The turning line generation system may generate a target turning line by executing the method described in the above embodiment.

[0200] In the present disclosure, when determining the current navigation driving strategy of the vehicle, the target turning line currently output by the turning line generation system can be first obtained, thereby providing conditions for determining the navigation driving strategy.

[0201] Step 1102, determining the current navigation driving strategy of the vehicle based on the target turning line.

[0202] In the present disclosure, after obtaining the target turning line, the vehicle's current navigation driving strategy can be determined based on the target turning line, thereby improving the accuracy and reliability of the determined navigation driving strategy, effectively solving the risk of the vehicle colliding with obstacles at the target intersection, and improving vehicle driving safety.

[0203] In the disclosed embodiment, the target turning line currently output by the turning line generation system is first obtained, and then based on the target turning line, the current navigation driving strategy of the vehicle is finally determined. Thus, by determining the current navigation driving strategy of the vehicle based on the current target turning line of the target intersection, the real-time and reliability of the determined current navigation driving strategy of the vehicle are ensured, thereby improving the driving safety of the vehicle.

[0204] In order to implement the above embodiment, the present disclosure also proposes a turning line generating device.

[0205] Fig.12 A schematic diagram of the structure of a turning line generating device provided in an embodiment of the present disclosure.

[0206] like Fig.12 As shown, the turning line generating device 1200 includes: a first acquiring module 1201 , a first determining module 1202 , a second determining module 1203 , and a generating module 1204 .

[0207] A first acquisition module 1201 is used to acquire a first trajectory data set associated with a target driving direction at a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period;

[0208] A first determination module 1202, configured to determine, based on a plurality of first trajectory sequences, a candidate trajectory centerline corresponding to a target driving direction within a preset time period;

[0209] A second determination module 1203 is used to determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory;

[0210] The generating module 1204 is used to generate a target turning line at the target intersection in the target driving direction based on the direction change point sequence.

[0211] In a possible implementation of the present disclosure, the first acquisition module 1201 is specifically configured to:

[0212] Based on the first position information of each trajectory point in the multi-source trajectory data obtained within a preset time period and the second position information of the target intersection, obtaining a second trajectory data set associated with the target intersection from the multi-source trajectory data;

[0213] Acquire candidate trajectory data associated with the target driving direction from the second trajectory dataset according to the direction angle corresponding to each trajectory point in the second trajectory dataset;

[0214] Segment the candidate trajectory data into multiple trajectory segments according to preset distance values ​​along the target driving direction;

[0215] Determine drift trajectory points and lane change trajectory points within each trajectory segment;

[0216] The drift trajectory points and lane change trajectory points in the candidate trajectory data are removed to obtain a first trajectory data set associated with the target driving direction.

[0217] In a possible implementation of the present disclosure, the first acquisition module 1201 is further configured to:

[0218] Determine the clustering density corresponding to each trajectory point in each trajectory segment according to the position information of the trajectory point in the trajectory segment, wherein the clustering density corresponding to each trajectory point is determined by the number of trajectory points whose distance to the trajectory point is less than a distance threshold;

[0219] The trajectory points whose corresponding clustering density in the trajectory segment is less than the density threshold are determined as drift trajectory points.

[0220] In a possible implementation of the present disclosure, the first acquisition module 1201 is further configured to:

[0221] According to the position information of the track points in each track segment, determine the road width occupied by all the track points in the track segment and the clustering density corresponding to each track point, wherein the clustering density corresponding to each track point is determined by the number of track points whose distance to it is less than a distance threshold;

[0222] Determine the distribution of the clustering density of the track points within the road width according to the clustering density corresponding to all the track points within the track segment;

[0223] According to the number of lanes at the target intersection in the target driving direction, a track width threshold in the target driving direction is determined;

[0224] According to the track width threshold and the clustering density distribution, the location range of the non-drifting track points is determined;

[0225] The trajectory points in the trajectory segment that are outside the position range are determined as drift trajectory points.

[0226] In a possible implementation of the present disclosure, the first acquisition module 1201 is further configured to:

[0227] Determine the direction difference threshold of the trajectory segment;

[0228] A trajectory point in the trajectory segment whose corresponding direction angle is greater than the direction difference threshold is determined as a lane change trajectory point.

[0229] In a possible implementation of the present disclosure, the first acquisition module 1201 is further configured to:

[0230] According to the direction angle corresponding to each track point in the track segment, determine the first angle difference and the second angle difference corresponding to each track point, wherein the first angle difference is the direction angle difference between the track point and the adjacent previous track point, and the second angle difference is the direction angle difference between the track point and the adjacent next track point;

[0231] Based on the first angle differences and the second angle differences respectively corresponding to all the trajectory points in the trajectory segment, a direction difference threshold of the trajectory segment is determined.

[0232] In a possible implementation of the present disclosure, the second determining module 1203 is specifically configured to:

[0233] Determine each trajectory point in the candidate trajectory centerline whose corresponding direction angle is greater than the direction angle threshold as a direction change point;

[0234] All direction change points in the center line of the same trajectory are determined as a direction change point sequence.

[0235] In a possible implementation of the present disclosure, the generating module 1204 is specifically configured to:

[0236] Determine the length and curvature of the target intersection, and the properties of the road segment to which the target intersection belongs;

[0237] Taking the trajectory points in each direction change point sequence as control points, a candidate turning line of the target intersection in the target driving direction is generated based on the length, curvature and attributes of the target intersection;

[0238] Determine a constraint parameter value corresponding to each candidate turning line, wherein the constraint parameter is at least one of the following parameters: curvature, angle with an associated lane, and distance with an associated lane;

[0239] Based on a preset constraint parameter threshold and a constraint parameter value corresponding to each candidate turning line, a target turning line is determined from a plurality of candidate turning lines.

[0240] In a possible implementation of the present disclosure, after determining a target turning line from a plurality of candidate turning lines based on a preset constraint parameter threshold and a constraint parameter value corresponding to each candidate turning line, the method further includes:

[0241] A second acquisition module (not shown in the figure) is used to acquire the image data currently corresponding to the target intersection;

[0242] An identification module (not shown in the figure) is used to identify obstacles on the image data and determine the obstacle information currently corresponding to the target intersection, wherein the obstacle information includes the location and shape of the obstacle;

[0243] The updating module (not shown in the figure) is used to update the target turning line according to the obstacle information.

[0244] The functions and specific implementation principles of the above modules in the embodiments of the present disclosure can be referred to the above method embodiments, and will not be repeated here.

[0245] In the disclosed embodiment, firstly, a first trajectory data set associated with the target driving direction in the target intersection is obtained, then based on a plurality of first trajectory sequences, a candidate trajectory centerline corresponding to the target driving direction within a preset period is determined, then based on the direction angle of each trajectory point in each candidate trajectory centerline, a direction change point sequence is determined, and finally based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated. Thus, by determining the candidate trajectory centerline of the target driving direction within a preset period based on a plurality of trajectory sequences corresponding to the target driving direction in the target intersection, and generating the target turning line of the target intersection currently in the target driving direction based on the direction change point sequence of each candidate trajectory centerline, the accuracy and real-time performance of the generated turning line are improved on the basis of reducing the cost of generating the turning line.

[0246] In order to implement the above embodiment, the present disclosure also proposes a device for determining a vehicle navigation driving strategy.

[0247] Fig.13 A schematic diagram of the structure of a device for determining a vehicle navigation driving strategy provided in an embodiment of the present disclosure.

[0248] like Fig.13 As shown, the device 1300 for determining the vehicle navigation driving strategy includes: a third acquisition module 1301 and a third determination module 1302 .

[0249] A third acquisition module 1301 is used to acquire a target turning line currently output by the turning line generation system, wherein the turning line generation system can generate the target turning line by configuring the turning line generation device in the above embodiment of the present disclosure;

[0250] The third determination module 1302 is used to determine the current navigation driving strategy of the vehicle based on the target turning line;

[0251] The functions and specific implementation principles of the above modules in the embodiments of the present disclosure can be referred to the above method embodiments, and will not be repeated here.

[0252] In the embodiment of the present disclosure, the target turning line currently output by the turning line generation system is first obtained, and then based on the target turning line, the current navigation driving strategy of the vehicle is finally determined. Thus, by determining the current navigation driving strategy of the vehicle based on the current target turning line of the target intersection, the real-time and reliability of the determined current navigation driving strategy of the vehicle are ensured, thereby improving the driving safety of the vehicle.

[0253] Fig.14 A schematic block diagram of an example electronic device 1400 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0254] like Fig.14 As shown, the device 1400 includes a computing unit 1401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1402 or a computer program loaded from a storage unit 1408 into a random access memory (RAM) 1403. In the RAM 1403, various programs and data required for the operation of the device 1400 can also be stored. The computing unit 1401, the ROM 1402, and the RAM 1403 are connected to each other via a bus 1404. An input / output (I / O) interface 1405 is also connected to the bus 1404.

[0255] A number of components in the device 1400 are connected to the I / O interface 1405, including: an input unit 1406, such as a keyboard, a mouse, etc.; an output unit 1407, such as various types of displays, speakers, etc.; a storage unit 1408, such as a disk, an optical disk, etc.; and a communication unit 1409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1409 allows the device 1400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0256] The computing unit 1401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1401 performs the various methods and processes described above, such as the method for generating a turning line. For example, in some embodiments, the method for generating a turning line may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1408. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1400 via the ROM 1402 and / or the communication unit 1409. When the computer program is loaded into the RAM 1403 and executed by the computing unit 1401, one or more steps of the method for generating a turning line described above may be performed. Alternatively, in other embodiments, the computing unit 1401 may be configured to execute the turning line generation method in any other appropriate manner (eg, by means of firmware).

[0257] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0258] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a general-purpose computer, a special-purpose computer, or other programmable processor or controller, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0259] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0260] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0261] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0262] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0263] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0264] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. In the description of the present disclosure, the words "if" and "if" used can be interpreted as "at the time of" or "when" or "in response to determination" or "under the circumstances of".

[0265] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for generating a turning line, comprising: Acquire a first trajectory data set associated with a target driving direction at a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period; Based on the multiple first trajectory sequences, determining a candidate trajectory centerline corresponding to the target driving direction within the preset time period; Determine a sequence of direction change points based on the direction angles of each trajectory point in the center line of each candidate trajectory; Based on the direction change point sequence, a target turning line of the target intersection currently in the target driving direction is generated.

2. The method of claim 1, wherein: The step of obtaining a first trajectory data set associated with a target driving direction at a target intersection includes: Based on the first position information of each trajectory point in the multi-source trajectory data acquired within the preset time period and the second position information of the target intersection, acquiring a second trajectory data set associated with the target intersection from the multi-source trajectory data; Acquire candidate trajectory data associated with the target driving direction from the second trajectory dataset according to the direction angle corresponding to each trajectory point in the second trajectory dataset; Segmenting the candidate trajectory data into a plurality of trajectory segments according to a preset distance value along the target driving direction; Determine drift trajectory points and lane change trajectory points within each trajectory segment; The drift trajectory points and the lane change trajectory points in the candidate trajectory data are eliminated to obtain a first trajectory data set associated with the target driving direction.

3. The method of claim 2, wherein: Determining the drift trajectory point in each trajectory segment includes: Determine, according to the position information of the trajectory points in each of the trajectory segments, the clustering density corresponding to each trajectory point in the trajectory segment, wherein the clustering density corresponding to each trajectory point is determined by the number of trajectory points whose distance thereto is less than a distance threshold; The trajectory points whose corresponding clustering density in the trajectory segment is less than the density threshold are determined as drift trajectory points.

4. The method of claim 2, wherein: Determining the drift trajectory point in each trajectory segment includes: According to the position information of the track points in each of the track segments, determine the road width occupied by all the track points in the track segment and the clustering density corresponding to each track point, wherein the clustering density corresponding to each track point is determined by the number of track points whose distance to the track point is less than a distance threshold; Determining the distribution of the clustering density of the track points within the road width according to the clustering density corresponding to all the track points within the track segment; Determining a track width threshold in the target driving direction according to the number of lanes of the target intersection in the target driving direction; Determining a position range of non-drifting trajectory points according to the trajectory width threshold and the clustering density distribution; The trajectory points in the trajectory segment that are outside the position range are determined as drift trajectory points.

5. The method according to claim 2 or 3, wherein: The method for determining the lane change trajectory point includes: Determine the direction difference threshold of the trajectory segment; A trajectory point in the trajectory segment whose corresponding direction angle is greater than the direction difference threshold is determined as the lane change trajectory point.

6. The method of claim 5, wherein: Determining the direction difference threshold of the trajectory segment includes: According to the direction angle corresponding to each track point in the track segment, determine the first angle difference and the second angle difference corresponding to each track point, wherein the first angle difference is the direction angle difference between the track point and the adjacent previous track point, and the second angle difference is the direction angle difference between the track point and the adjacent next track point; Based on the first angle differences and the second angle differences respectively corresponding to all the trajectory points in the trajectory segment, a direction difference threshold of the trajectory segment is determined.

7. The method of claim 1, wherein: The step of determining a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory includes: Determine each trajectory point in the candidate trajectory centerline whose corresponding direction angle is greater than the direction angle threshold as a direction change point; All direction change points in the center line of the same trajectory are determined as a direction change point sequence.

8. The method according to any one of claims 1 to 7, wherein: The step of generating a target turning line of the target intersection currently in the target driving direction based on the direction change point sequence comprises: Determine the length and curvature of the target intersection and the attributes of the road section to which the target intersection belongs; Taking each trajectory point in the direction change point sequence as a control point, based on the length, curvature and attributes of the target intersection, a candidate turning line of the target intersection currently in the target driving direction is generated; Determine a constraint parameter value corresponding to each candidate turning line, wherein the constraint parameter is at least one of the following parameters: curvature, angle with an associated lane, and distance with an associated lane; Based on a preset constraint parameter threshold and a constraint parameter value corresponding to each of the candidate turning lines, a target turning line is determined from the plurality of candidate turning lines.

9. The method of claim 8, wherein: After determining the target turning line from the plurality of candidate turning lines based on the preset constraint parameter threshold and the constraint parameter value corresponding to each candidate turning line, the method further includes: Acquire image data currently corresponding to the target intersection; Perform obstacle recognition on the image data to determine obstacle information currently corresponding to the target intersection, wherein the obstacle information includes the position and shape of the obstacle; The target turning line is updated according to the obstacle information.

10. A method for determining a vehicle navigation driving strategy, wherein: include: Acquire a target turning line currently output by a turning line generation system, wherein the turning line generation system generates the target turning line by executing the method according to any one of claims 1 to 9; Based on the target turning line, the current navigation driving strategy of the vehicle is determined.

11. A device for generating a turning line, wherein: The device comprises: A first acquisition module is used to acquire a first trajectory data set associated with a target driving direction in a target intersection, wherein the first trajectory data set includes a plurality of first trajectory sequences traveling along the target driving direction within a preset time period; A first determination module, configured to determine, based on the plurality of first trajectory sequences, a candidate trajectory centerline corresponding to the target driving direction within the preset time period; A second determination module is used to determine a direction change point sequence based on the direction angle of each trajectory point in the center line of each candidate trajectory; A generation module is used to generate a target turning line of the target intersection currently in the target driving direction based on the direction change point sequence.

12. The device of claim 11, wherein: The first acquisition module is specifically used to: Based on the first position information of each trajectory point in the multi-source trajectory data acquired within the preset time period and the second position information of the target intersection, acquiring a second trajectory data set associated with the target intersection from the multi-source trajectory data; Acquire candidate trajectory data associated with the target driving direction from the second trajectory dataset according to the direction angle corresponding to each trajectory point in the second trajectory dataset; Segmenting the candidate trajectory data into a plurality of trajectory segments according to a preset distance value along the target driving direction; Determine drift trajectory points and lane change trajectory points within each trajectory segment; The drift trajectory points and the lane change trajectory points in the candidate trajectory data are eliminated to obtain a first trajectory data set associated with the target driving direction.

13. The device of claim 12, wherein: The first acquisition module is further used for: Determine, according to the position information of the trajectory points in each of the trajectory segments, the clustering density corresponding to each trajectory point in the trajectory segment, wherein the clustering density corresponding to each trajectory point is determined by the number of trajectory points whose distance thereto is less than a distance threshold; The trajectory points whose corresponding clustering density in the trajectory segment is less than the density threshold are determined as drift trajectory points.

14. The device of claim 12, wherein: The first acquisition module is further used for: According to the position information of the track points in each of the track segments, determine the road width occupied by all the track points in the track segment and the clustering density corresponding to each track point, wherein the clustering density corresponding to each track point is determined by the number of track points whose distance to the track point is less than a distance threshold; Determining the distribution of the clustering density of the track points within the road width according to the clustering density corresponding to all the track points within the track segment; Determining a track width threshold in the target driving direction according to the number of lanes of the target intersection in the target driving direction; Determining a position range of non-drifting trajectory points according to the trajectory width threshold and the clustering density distribution; The trajectory points in the trajectory segment that are outside the position range are determined as drift trajectory points.

15. The device according to claim 12 or 13, wherein: The first acquisition module is further used for: Determine the direction difference threshold of the trajectory segment; A trajectory point in the trajectory segment whose corresponding direction angle is greater than the direction difference threshold is determined as the lane change trajectory point.

16. The device of claim 15, wherein: The first acquisition module is further used for: According to the direction angle corresponding to each track point in the track segment, determine the first angle difference and the second angle difference corresponding to each track point, wherein the first angle difference is the direction angle difference between the track point and the adjacent previous track point, and the second angle difference is the direction angle difference between the track point and the adjacent next track point; Based on the first angle differences and the second angle differences respectively corresponding to all the trajectory points in the trajectory segment, a direction difference threshold of the trajectory segment is determined.

17. The device of claim 11, wherein: The second determining module is specifically used to: Determine each trajectory point in the candidate trajectory centerline whose corresponding direction angle is greater than the direction angle threshold as a direction change point; All direction change points in the center line of the same trajectory are determined as a direction change point sequence.

18. The device according to any one of claims 11 to 17, wherein: The generating module is specifically used for: Determine the length and curvature of the target intersection and the attributes of the road section to which the target intersection belongs; Taking each trajectory point in the direction change point sequence as a control point, based on the length, curvature and attributes of the target intersection, a candidate turning line of the target intersection currently in the target driving direction is generated; Determine a constraint parameter value corresponding to each candidate turning line, wherein the constraint parameter is at least one of the following parameters: curvature, angle with an associated lane, and distance with an associated lane; Based on a preset constraint parameter threshold and a constraint parameter value corresponding to each of the candidate turning lines, a target turning line is determined from the plurality of candidate turning lines.

19. The device of claim 18, wherein: After determining the target turning line from the plurality of candidate turning lines based on the preset constraint parameter threshold and the constraint parameter value corresponding to each candidate turning line, the method further includes: A second acquisition module is used to acquire the image data currently corresponding to the target intersection; An identification module, used to perform obstacle identification on the image data and determine obstacle information currently corresponding to the target intersection, wherein the obstacle information includes the position and shape of the obstacle; An updating module is used to update the target turning line according to the obstacle information.

20. A device for determining a vehicle navigation driving strategy, wherein: The device comprises: A third acquisition module, used for acquiring a target turning line currently output by a turning line generation system, wherein the turning line generation system generates the target turning line by executing the method according to any one of claims 1 to 9; The third determination module is used to determine the current navigation driving strategy of the vehicle based on the target turning line.

21. An electronic device, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, the memory stores instructions that may be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for generating a turning line as described in any one of claims 1-9, or the method for determining the vehicle navigation driving strategy as described in claim 10.

22. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to enable the computer to execute the method for generating a turning line according to any one of claims 1 to 9, or the method for determining a vehicle navigation driving strategy according to claim 10.

23. A computer program product, comprising a computer program, which, when executed by a processor, implements the method for generating a turning line according to any one of claims 1 to 9, or the method for determining a vehicle navigation driving strategy according to claim 10.

Images