Intersection recognition method, device, equipment and storage medium

Abstract

The application discloses a kind of intersection identification method, device, equipment and storage medium.The method comprises: according to the data generated by vehicle-mounted sensor collection logical map, road boundary line information, lane center line information and vehicle trajectory information are contained in logical map;According to road boundary line information, lane center line information and vehicle trajectory information, the intersection corresponding to the current position of vehicle is identified.Through the above technical solution, the problem that the intersection identification method in the prior art uses deep learning to identify the intersection is solved, the implementation process is complex and the cost is high, the complexity of the implementation process is effectively reduced while the cost is considered, and the accuracy of the intersection identification is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intersection recognition method, apparatus, device and storage medium. Background Technology

[0002] With the popularization of autonomous parking technology and users' increasingly higher demands for intelligence, it is necessary to continuously improve the accuracy of various algorithms and functions in intelligent vehicles. The output and accuracy of strong semantic information in maps are crucial to improving the accuracy of downstream algorithm modules. A key aspect of this is the accurate and efficient identification of various road conditions and intersections. Currently, deep learning can be used for intersection recognition, but it requires a large amount of prior knowledge and extensive training on large amounts of image or LiDAR data, making the process complex and costly. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and storage medium for intersection recognition to improve recognition efficiency.

[0004] According to one aspect of the present invention, an intersection recognition method is provided, comprising:

[0005] A logical map is generated based on data collected by onboard sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information.

[0006] The intersection corresponding to the vehicle's current location is identified based on road boundary line information, lane center line information, and vehicle trajectory information.

[0007] According to another aspect of the present invention, an intersection recognition device is provided, comprising:

[0008] The generation module is used to generate a logical map based on the data collected by the vehicle sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information.

[0009] The recognition module is used to identify the intersection corresponding to the current position of a vehicle based on road boundary line information, lane center line information, and vehicle trajectory information.

[0010] According to another aspect of the present invention, an intersection recognition device is provided, the intersection recognition device comprising:

[0011] At least one processor; and

[0012] A memory that is communicatively connected to at least one processor; wherein,

[0013] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the intersection recognition method of any embodiment of the present invention.

[0014] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the intersection recognition method of any embodiment of the present invention.

[0015] The technical solution provided by this invention generates a logical map based on data collected by vehicle-mounted sensors. This logical map includes road boundary line information, lane center line information, and vehicle trajectory information. The intersection corresponding to the vehicle's current position is identified based on the road boundary line information, lane center line information, and vehicle trajectory information. By employing this technical solution, vehicle-mounted sensors acquire vehicle-related data information, and a logical map is constructed based on this information. The logical relationships between information within the logical map enable intersection identification. This solves the problem of complex and costly implementation processes in existing intersection identification methods that use deep learning for intersection identification. It achieves the beneficial effect of effectively reducing the complexity of the implementation process while improving the accuracy of intersection identification, all while considering cost.

[0016] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an intersection recognition method provided in Embodiment 1 of the present invention;

[0019] Figure 2 This is a flowchart of an intersection recognition method provided in Embodiment 2 of the present invention;

[0020] Figure 3 This is a flowchart of an intersection recognition method provided in Embodiment 3 of the present invention;

[0021] Figure 4 This is a schematic diagram of identifying a suspected intersection according to Embodiment 3 of the present invention;

[0022] Figure 5 This is a schematic diagram of the structure of an intersection recognition device provided in Embodiment 4 of the present invention;

[0023] Figure 6 This is a schematic diagram of the structure of an intersection recognition device provided in Embodiment 5 of the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] Example 1

[0027] Figure 1 This is a flowchart of an intersection recognition method provided in Embodiment 1 of the present invention. This embodiment is applicable to intersection recognition scenarios. The method can be executed by an intersection recognition device, which can be implemented in hardware and / or software. The intersection recognition device can be configured in an intersection recognition equipment, such as an in-vehicle device or vehicle. Figure 1 As shown, the method includes:

[0028] S110. Generate a logical map based on the data collected by the vehicle-mounted sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information.

[0029] In this embodiment, the vehicle-mounted sensors can be understood as devices that acquire vehicle-related information, and may include four surround-view fisheye cameras, an inertial measurement unit (IMU), and wheel speed sensors. The four surround-view fisheye cameras are used for close-range vehicle perception; by stitching together the data from the four fisheye cameras, a 360-degree surround-view view of the vehicle can be formed. The data from the IMU and wheel speed sensors, through fusion, can provide the vehicle's trajectory information. The logical map can be used to reflect basic road logic information, providing a road reference map for vehicle control; it is a 3D map by default, but is backward compatible with 2D maps. Road boundary line information can be understood as the boundary line information on both sides of the road in the logical map, used to determine whether the boundaries of the two roads are the boundaries of the same road. Lane centerline information can be used to determine the direction of travel of vehicles in the lane.

[0030] Specifically, the vehicle's surrounding environment and trajectory information are collected by onboard sensors. The collected data is then input into a mapping algorithm to generate a logical map. This logical map mainly includes road boundary line information, lane center line information, vehicle trajectory and their distribution relationship. By reading the logical map, road boundary line information and lane center line information can be obtained, and this information can then be used to provide road references for vehicle control.

[0031] S120: Identify the intersection corresponding to the current location of the vehicle based on road boundary line information, lane center line information, and vehicle trajectory information.

[0032] For example, a preliminary assessment of intersections is made using lane centerline information and vehicle trajectory information to identify potential intersections. Then, road centerline information is obtained using road boundary line information. Road centerline information is used to distinguish the direction of travel for vehicles on the road. Based on the curvature of the road centerline information, suspicious intersections can be identified and the true intersection can be determined. The information of this intersection is then obtained and saved in a corresponding file for use by downstream modules. The preliminary assessment of potential intersections can be achieved by projecting a segment of the vehicle trajectory from the current moment onto the road boundary (including the boundaries on both sides of the road). If no road boundary line is found in the projection, it indicates that the vehicle's current position may correspond to an intersection. Alternatively, a segment of the lane centerline from the vehicle's current position can be projected onto the road boundary (including the boundaries on both sides of the road). If no road boundary line is found in the projection, it indicates that the vehicle's current position may correspond to an intersection.

[0033] The technical solution provided by this invention generates a logical map based on data collected by vehicle-mounted sensors. This logical map includes road boundary line information, lane center line information, and vehicle trajectory information. The intersection corresponding to the vehicle's current position is identified based on the road boundary line information, lane center line information, and vehicle trajectory information. By employing this technical solution, vehicle-mounted sensors acquire vehicle-related data information, and a logical map is constructed based on this information. The logical relationships between information within the logical map enable intersection identification. This solves the problem of complex and costly implementation processes in existing intersection identification methods that use deep learning for intersection identification. It achieves the beneficial effect of effectively reducing the complexity of the implementation process while improving the accuracy of intersection identification, all while considering cost.

[0034] In some embodiments, road boundary information, lane centerline information, and vehicle trajectory information are stored in a point cloud format. This technical solution enables efficient data compression and rapid data retrieval, improving data storage and management efficiency.

[0035] Specifically, to improve data storage and management efficiency, road boundary line information, lane centerline information, and vehicle trajectory information are stored in the logical map as point clouds. Each road segment includes both road boundary lines, corresponding to a unique ID, and the IDs for consecutive road segments are continuous.

[0036] Example 2

[0037] Figure 2 This is a flowchart of an intersection recognition method provided in Embodiment 2 of the present invention. This embodiment optimizes and extends the above-mentioned optional embodiments. This embodiment further optimizes S120: after generating a logical map, relevant information about roads and vehicles is obtained using the logical map; based on this relevant information, suspected intersections are identified; and the authenticity of the suspected intersections is determined based on the road centerline, thereby confirming the intersection. Figure 2 As shown, the method includes:

[0038] S210. Generate a logical map based on the data collected by the vehicle-mounted sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information.

[0039] S220. Determine whether the area where the vehicle is currently located is a suspected intersection based on road boundary line information, lane center line information, and vehicle trajectory information.

[0040] The road boundary line includes the first direction road boundary line and the second direction road boundary line.

[0041] In this embodiment, the first direction road boundary line can be understood as the road boundary line on the left side of the vehicle when the vehicle is traveling along the road in the positive or diagonal direction of the logical map. The second direction road boundary line can be understood as the road boundary line on the right side of the vehicle when the vehicle is traveling along the road in the positive or diagonal direction of the logical map.

[0042] For example, road boundary information and vehicle trajectory information are obtained from a logical map. Based on the vehicle trajectory information, the current position of the vehicle can be determined. Then, trajectory points within a certain range of the vehicle's trajectory, including its current position, are obtained in chronological order. The projection formula from a point to a line is used to record the projections of the points onto the road boundary lines on both sides. If the boundary line can be observed in the projections formed by all trajectory points within this range, the area where the vehicle is currently located is considered not an intersection. If the boundary line cannot be observed in the projections formed by all points within this range, the area where the vehicle is currently located is determined to be a suspected intersection. In this embodiment, the projection formula is not limited; different projection formulas are used depending on the environment. For example, the projection formula can be a point-to-line projection formula, a point-to-curve projection formula, an orthogonal projection formula for a curved surface, a point-to-free curve projection formula, etc.

[0043] S230. If it is a suspected intersection, the road centerline is determined based on the road boundary line information, and the authenticity of the suspected intersection is determined based on the road centerline.

[0044] In this embodiment, the road centerline is used to distinguish lanes in different directions. It can be obtained by finding the midpoint of the road boundary line information on both sides. Each road centerline segment has an ID, and the centerlines of consecutive road segments are continuous, as are their IDs. The ID of the road centerline can be the same as the ID of the corresponding road boundary line.

[0045] For example, a suspected intersection may not be a real intersection and requires further confirmation. First, information about the road boundary lines on both sides is obtained. This is done by segmenting the road boundary lines on both sides and connecting the midpoints of the lines, then drawing a line from these midpoints to form the road centerline. Since intersections in the logical map mainly include straight-ahead intersections and turning intersections, the curvature of the road centerline can be detected to determine whether the current vehicle is on a straight or turning road. A corresponding range for the vehicle's trajectory at a straight-ahead or turning intersection is defined. By determining whether the boundary line can be observed in the projection of the trajectory points within the corresponding range of the vehicle's trajectory at a straight-ahead or turning intersection, the authenticity of the suspected intersection is determined, thus identifying the real intersection.

[0046] The intersection recognition technology solution provided in this invention generates a logical map, then uses the logical map to obtain relevant information about roads and vehicles. Based on this information, it identifies potential intersections and determines their authenticity based on the road centerline, ultimately confirming them as intersections. By repeatedly identifying and confirming intersections on the logical map using this technology, the accuracy of intersection recognition is effectively improved.

[0047] As an optional embodiment, determining the road centerline based on road boundary line information includes: dividing the first-direction road boundary line and the second-direction road boundary line into corresponding segments based on the road boundary line information; for each segment, taking the midpoint of the first-direction road boundary line and the second-direction road boundary line; and using the line connecting the midpoints of each segment as the road centerline. This technical solution can accurately determine the road centerline.

[0048] Specifically, considering that the road centerline in the logical map may be a straight line or a curve, it needs to be calculated using road boundary line information. First, the road boundary lines corresponding to the vehicle's current location in the first and second directions are segmented. The midpoint of the first and second direction road boundary lines in each segment is then connected. Finally, the midpoints of the obtained midpoints of the entire road boundary lines on both sides are connected to form the road centerline. The road centerline is used to distinguish the travel direction of vehicles on both sides of the road.

[0049] As another optional embodiment, determining the authenticity of a suspected intersection based on the road centerline includes: determining the type of the suspected intersection based on the curvature of the road centerline; searching for an observable road boundary line in the projections of trajectory points within the corresponding range onto the road boundary lines in the first and second directions; if no observable road boundary line is found in the projections of trajectory points within the corresponding range, the suspected intersection is determined to be a true intersection. This technical solution effectively improves the accuracy of intersection recognition.

[0050] For example, intersections in the logical map include straight-ahead intersections and non-straight-ahead intersections. Based on the curvature (or bend) of the road centerline, the type of a suspected intersection can be determined as either a straight-ahead or non-straight-ahead road. Then, an observable road boundary line is searched in the projection within a corresponding range to confirm whether the intersection is a straight-ahead or non-straight-ahead intersection. Specifically, if it is a straight-ahead intersection, the corresponding range can be 5m. If no boundary line is observed in the projections of a vehicle trajectory longer than 5m to both sides, it is considered a straight-ahead intersection; otherwise, it is considered a false positive. If it is a non-straight-ahead intersection, the corresponding range can be 2m. If no boundary line is observed in the projections of a vehicle trajectory longer than 2m to both sides, it is considered a non-straight-ahead intersection; otherwise, it is considered a false positive. This embodiment does not limit the length of the corresponding range for straight-ahead and non-straight-ahead intersections; the optimal range length can be selected based on the environment.

[0051] As another optional embodiment, the type includes straight-ahead intersections or non-straight-ahead intersections; the corresponding range for straight-ahead intersections is larger than that for non-straight-ahead intersections. Through the above technical solution, setting the corresponding range based on the intersection type lays the foundation for improving the accuracy of intersection recognition.

[0052] Specifically, the logical map contains two types of roads: straight and non-straight. Therefore, the corresponding intersection types are also either straight or non-straight. In practical applications, straight intersections are generally longer than non-straight intersections. Based on testing experience, the corresponding range for straight intersections is greater than 5m, and the corresponding range for non-straight intersections is greater than 2m. This embodiment does not limit the length of the corresponding ranges for straight roads and non-straight intersections; the optimal range length can be selected based on the environment.

[0053] It should be noted that finding observable road boundary lines by projecting trajectory points within the corresponding range onto the first and second direction road boundary lines can be understood as a process of verifying suspected intersections initially identified using trajectory points within a set range. In this verification process, the corresponding range for straight-ahead intersections is larger than the corresponding range for non-straight-ahead intersections. Preferably, the corresponding range for non-straight-ahead intersections is greater than or equal to the set range used in the initial identification of suspected intersections. For example, a 1.5m trajectory point can be used for initial identification of a suspected intersection, a 5m trajectory point can be used for identification to determine if the suspected intersection is a true straight-ahead intersection, and a 2m trajectory point can be used for identification to determine if the suspected intersection is a true non-straight-ahead intersection.

[0054] Optionally, the vehicle trajectory can be predicted and extended based on the vehicle trajectory and the current driving direction. When identifying intersections using a set range or corresponding range, the predicted and extended trajectory points can be used to identify intersections earlier and predict when a vehicle is about to enter the intersection.

[0055] For example, to predict a vehicle's trajectory, the trajectory is extended by a predetermined length (e.g., 2m). The trajectory points between the vehicle's current position and the point where the trajectory is extended by 2m are projected onto the road boundary lines on both sides. If no boundary line is observed in either projection, the vehicle's current position is considered a potential intersection. Similarly, the trajectory points from 1m before the vehicle's current position to the point where the trajectory is extended by 1m (a total of 2m) are projected onto the road boundary lines on both sides. If no boundary line is observed in either projection, the vehicle's current position is considered a potential intersection. In other words, the trajectory points used for projection can include points from the vehicle's actual trajectory, as well as points from the predicted trajectory or the extended trajectory. Preferably, the trajectory length traversed by the projected trajectory points only needs to reach a predetermined range.

[0056] Example 3

[0057] Figure 3 This is a flowchart of an intersection recognition method provided in Embodiment 3 of the present invention. This embodiment optimizes and extends the above-mentioned optional embodiments. This embodiment further optimizes S220: after generating a logical map, the current position and driving direction of the vehicle are determined using the logical map. Based on this information, points on a specified trajectory range are projected onto the boundary lines of the lanes on both sides. Suspected intersections are identified by determining whether the boundary lines can be observed in the projection. Then, the intersection is determined by judging the authenticity of the suspected intersection. Figure 3 As shown, the method includes:

[0058] S310. Generate a logical map based on the data collected by the vehicle-mounted sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information.

[0059] S320: Determine the vehicle's current position and direction of travel based on lane centerline information and vehicle trajectory information.

[0060] Specifically, in the logical map, the lane center line is used to indicate the direction that vehicles should travel on that road, so the current driving direction of the vehicle can be determined, and then the current position of the vehicle can be obtained based on the vehicle trajectory information in the logical map.

[0061] S330. Based on the vehicle's current position and driving direction, project the trajectory points within a set range in the vehicle trajectory information onto the road boundary lines in the first and second directions, respectively.

[0062] In this embodiment, the set range includes the vehicle trajectory within a certain range encompassing the vehicle's current location. During initial intersection identification, to avoid missed identifications, the set range can be the corresponding range of a non-straight-ahead intersection or a range smaller than that of a non-straight-ahead intersection. For example, the corresponding range of a non-straight-ahead intersection is typically 1.5-2.5m. Taking a non-straight-ahead intersection with a corresponding range of 2m as an example, the set range can be 1.5m, 1.8m, or 2m. In this embodiment, the corresponding range is an optimal threshold obtained based on testing experience. The optimal threshold may differ in different environments, so this embodiment does not limit the size of the corresponding range.

[0063] Specifically, after determining the vehicle's current position and driving direction, the vehicle trajectory within a certain range, including the vehicle's current position, is defined as the set range. For example, the corresponding range for a non-straight intersection is 2 meters. Therefore, the vehicle trajectory within 2 meters, including the vehicle's current position, is defined as the set range. The trajectory points within this range are then obtained, and the projection of the points onto the road boundary lines on both sides is recorded using a point-to-line projection formula. In this embodiment, the projection formula is not limited; different projection formulas are used depending on the environment. For example, the projection formula can be a point-to-line projection formula, a point-to-curve projection formula, an orthogonal projection formula for a curved surface, a point-to-free curve projection formula, etc.

[0064] S340. If there are no observable road boundary lines in the projection of the trajectory points within the set range, then the area where the vehicle is currently located is identified as a suspected intersection.

[0065] Specifically, if there is an observable road boundary line in the projection of the trajectory points within the set range, then there is no intersection in the area where the vehicle is currently located. If there is no observable road boundary line on either side, then there is a suspected intersection in the area where the vehicle is currently located.

[0066] Figure 4 This is a schematic diagram illustrating how to determine a suspected intersection according to Embodiment 3 of the present invention. Figure 4 As shown, the road centerline is 31, the lane centerline is 32, the vehicle trajectory is 33, and the set range is from point A to point B (which can be 2m). When the vehicle travels to point B, there is no observable road boundary line in the projection of the trajectory points within the set range, so it is a suspected intersection.

[0067] S350. If it is a suspected intersection, the road centerline is determined based on the road boundary line information, and the authenticity of the suspected intersection is determined based on the road centerline.

[0068] Specifically, if the curvature of the road centerline is large, a smaller corresponding range (e.g., 2m) can be used to check whether an observable road boundary line exists in the projection of the trajectory point. If it does not exist, the suspected intersection can be confirmed as a true turning intersection.

[0069] The intersection recognition scheme provided in this invention, after generating a logical map, uses relevant information from the logical map to determine the current position and direction of vehicle travel. Based on this information, points within a specified trajectory range are projected onto the lane boundaries on both sides. Suspected intersections are identified by determining whether the boundary lines can be observed in the projection. Furthermore, the intersection is confirmed by judging whether the suspected intersection is genuine or not. This technical solution identifies all possible intersections, effectively reducing the omission rate and increasing the intersection recognition rate.

[0070] Example 4

[0071] Figure 5 This is a schematic diagram of the structure of an intersection recognition device provided in Embodiment 4 of the present invention. Figure 5 As shown, the device includes: a generation module 41 and an identification module 42. Wherein:

[0072] The generation module 41 is used to generate a logical map based on the data collected by the vehicle-mounted sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information. The recognition module 42 is used to identify the intersection corresponding to the current position of the vehicle based on the road boundary line information, lane center line information, and vehicle trajectory information.

[0073] The technical solution provided by the embodiments of the present invention solves the problem that the existing intersection recognition methods using deep learning for intersection recognition are complex and costly in terms of implementation process. It achieves the beneficial effect of effectively reducing the complexity of the implementation process and improving the accuracy of intersection recognition while taking into account cost.

[0074] Optionally, the recognition module 42 includes:

[0075] The suspected intersection identification unit is used to determine whether the area where the vehicle is currently located is a suspected intersection based on road boundary line information, lane center line information, and vehicle trajectory information.

[0076] The true / false attribute determination unit is used to determine the road centerline based on the road boundary line information, and to determine the true / false attributes of the suspected intersection based on the road centerline.

[0077] The road boundary line includes the first direction road boundary line and the second direction road boundary line.

[0078] Optional, the suspected intersection identification unit includes:

[0079] The first determining subunit is used to determine the current position and direction of travel of the vehicle based on the lane centerline information and vehicle trajectory information.

[0080] The boundary line projection subunit is used to project trajectory points within a set range in the vehicle trajectory information onto the first direction road boundary line and the second direction road boundary line, based on the vehicle's current position and driving direction.

[0081] The suspected intersection determination subunit is used to determine the area where the vehicle is currently located as a suspected intersection if there is no observable road boundary line in the projection of trajectory points within a set range.

[0082] Optionally, the road centerline can be determined based on the road boundary line information, including:

[0083] Based on the road boundary line information, the road boundary lines in the first direction and the road boundary lines in the second direction are divided into corresponding segments;

[0084] For each segment, take the midpoint between the first-direction road boundary line and the second-direction road boundary line;

[0085] The line connecting the midpoints of each segment is taken as the center line of the road.

[0086] Optionally, the authenticity of a suspected intersection can be determined based on the road centerline, including:

[0087] Determine the type of suspected intersection based on the curvature of the road centerline;

[0088] Based on the type, find the observable road boundary line in the projection of the trajectory points within the corresponding range onto the road boundary line in the first direction and the road boundary line in the second direction;

[0089] If no observable road boundary line is found in the projection of the trajectory points within the corresponding range, the suspected intersection is determined to be a real intersection.

[0090] Optional, the type includes straight-ahead intersections or non-straight-ahead intersections;

[0091] The corresponding range of straight-ahead intersections is larger than that of non-straight-ahead intersections.

[0092] Optionally, road boundary information, lane centerline information, and vehicle trajectory information can be stored in point cloud format.

[0093] The intersection recognition device provided in the embodiments of the present invention can execute the intersection recognition method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0094] Example 5

[0095] Figure 6 This is a schematic diagram of the structure of an intersection recognition device according to Embodiment 5 of the present invention. The intersection recognition device can be an electronic device, 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 processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0096] like Figure 6As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0097] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0098] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as intersection recognition methods.

[0099] In some embodiments, the intersection recognition method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded into and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the intersection recognition method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the intersection recognition method by any other suitable means (e.g., by means of firmware).

[0100] Various embodiments 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-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0101] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0102] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0103] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device 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 provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0104] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0105] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0106] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0107] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for intersection recognition, characterized in that, include: A logical map is generated based on data collected by vehicle sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information. The intersection corresponding to the current position of the vehicle is identified based on the road boundary line information, the lane center line information, and the vehicle trajectory information. Identifying the intersection corresponding to the vehicle's current position based on the road boundary line information, the lane center line information, and the vehicle trajectory information includes: Based on the road boundary line information, the lane center line information, and the vehicle trajectory information, determine whether the area where the vehicle is currently located is a suspected intersection; If it is a suspected intersection, the road centerline is determined based on the road boundary line information, and the authenticity of the suspected intersection is determined based on the road centerline. The road boundary line includes a first-direction road boundary line and a second-direction road boundary line; Determining the authenticity of the suspected intersection based on the road centerline includes: The type of the suspected intersection is determined based on the curvature of the road centerline; the type includes straight-ahead intersections or non-straight-ahead intersections, the corresponding range of the straight-ahead intersection is a first threshold, the corresponding range of the non-straight-ahead intersection is a second threshold, and the first threshold is greater than the second threshold; both the first threshold and the second threshold are optimal range lengths selected based on the environment; According to the type, find the observable road boundary line in the projection of the trajectory points within the corresponding range onto the road boundary line in the first direction and the road boundary line in the second direction; If no observable road boundary line is found in the projection of the trajectory points within the corresponding range, the suspected intersection is determined to be a real intersection.

2. The method according to claim 1, characterized in that, Determining whether the area where the vehicle is currently located is a suspected intersection based on the road boundary line information, the lane center line information, and the vehicle trajectory information includes: The current position and direction of travel of the vehicle are determined based on the lane centerline information and the vehicle trajectory information. Based on the vehicle's current position and the vehicle's driving direction, the trajectory points within a set range in the vehicle trajectory information are projected onto the road boundary lines in the first and second directions, respectively. If no observable road boundary line exists in the projection of the trajectory points within the set range, the area where the vehicle is currently located is identified as a suspected intersection.

3. The method according to claim 1, characterized in that, Determining the road centerline based on the road boundary line information includes: Based on the road boundary line information, the road boundary line in the first direction and the road boundary line in the second direction are divided into corresponding segments; For each segment, take the midpoint between the boundary line of the road in the first direction and the boundary line of the road in the second direction; The line connecting the midpoints of each segment is taken as the centerline of the road.

4. The method according to any one of claims 1-3, characterized in that, The road boundary information, the lane centerline information, and the vehicle trajectory information are stored in a point cloud format.

5. An intersection recognition device, characterized in that, include: The generation module is used to generate a logical map based on data collected by vehicle sensors. The logical map includes road boundary line information, lane center line information, and vehicle trajectory information. The identification module is used to identify the intersection corresponding to the current position of the vehicle based on the road boundary line information, the lane center line information, and the vehicle trajectory information. The identification module includes: The suspected intersection determination unit is used to determine whether the area where the vehicle is currently located is a suspected intersection based on the road boundary line information, the lane center line information, and the vehicle trajectory information. The true / false attribute determination unit is used to determine the road centerline based on the road boundary line information if it is a suspected intersection, and to determine the true / false attribute of the suspected intersection based on the road centerline. The road boundary line includes a first-direction road boundary line and a second-direction road boundary line; The true / false attribute determination unit is further used for: The type of the suspected intersection is determined based on the curvature of the road centerline; the type includes straight-ahead intersections or non-straight-ahead intersections, the corresponding range of the straight-ahead intersection is a first threshold, the corresponding range of the non-straight-ahead intersection is a second threshold, and the first threshold is greater than the second threshold; both the first threshold and the second threshold are optimal range lengths selected based on the environment; According to the type, find the observable road boundary line in the projection of the trajectory points within the corresponding range onto the road boundary line in the first direction and the road boundary line in the second direction; If no observable road boundary line is found in the projection of the trajectory points within the corresponding range, the suspected intersection is determined to be a real intersection.

6. An intersection recognition device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the intersection recognition method as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the intersection recognition method as described in any one of claims 1-4.

Images