Road topology determination methods, vehicle control methods, equipment, media and products

Abstract

This application provides a road topology determination method, vehicle control method, device, medium, and product, which can be used in the field of autonomous driving technology. The method includes: acquiring road topology information; acquiring target navigation information corresponding to a target intersection; matching multiple second lanes with multiple first lanes based on the number of first lanes and the number of second lanes; adding the steering attributes corresponding to each first lane to the intersection information to obtain updated road topology information; the steering attribute corresponding to the first lane is the steering attribute corresponding to the second lane matched with the first lane. The lane steering attributes determined by this application have high accuracy, enabling vehicles to select the correct lane when approaching a target intersection, reducing traffic accidents or violations caused by steering errors, and at a lower cost.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a road topology determination method, vehicle control method, device, medium and product. Background Technology

[0002] When vehicles travel on public transport roads, they are obligated to obey traffic rules, such as following road signs and choosing the appropriate lane.

[0003] In related technologies, during vehicle operation, a camera perception system or a third-party map is used to obtain the lane's steering attributes (such as left turn, straight, right turn, etc.), and the appropriate lane is selected for travel based on the lane's steering attributes.

[0004] However, camera-based perception systems, which rely on neural networks, inherently suffer from limitations in long-tail scenarios. Furthermore, in heavy traffic, road markings are easily obscured by vehicles, affecting the accuracy of camera perception. Third-party maps, on the other hand, require significant resources for data collection, processing, and updating, typically incurring fees and resulting in higher costs, as well as issues with untimely updates. Summary of the Invention

[0005] This application provides a road topology determination method, vehicle control method, device, medium, and product. The lane steering attributes determined are highly accurate, enabling vehicles to select the correct lane when approaching the target intersection, reducing traffic accidents or violations caused by steering errors, and at a lower cost.

[0006] In a first aspect, this application provides a method for determining road topology, including:

[0007] Obtain road topology information; the road topology information includes intersection information corresponding to the target intersection, the intersection information includes the number of first lanes, and the number of first lanes is the total number of multiple first lanes at the target intersection;

[0008] Obtain target navigation information corresponding to the target intersection; the target navigation information includes road information corresponding to the road where the vehicle is located, the road information includes the number of second lanes and the turning attributes corresponding to the multiple second lanes respectively; the number of second lanes is the total number of lanes included in the road where the vehicle is located;

[0009] Based on the number of the first lanes and the number of the second lanes, multiple second lanes are matched with multiple first lanes;

[0010] The steering attributes corresponding to each of the first lanes are added to the intersection information to obtain updated road topology information; the steering attributes corresponding to the first lane are the steering attributes corresponding to the second lane that matches the first lane.

[0011] In some embodiments, obtaining the target navigation information corresponding to the target intersection includes:

[0012] Acquire multi-frame real-time navigation information; the multi-frame real-time navigation information includes real-time navigation information of the current time and multiple consecutive historical moments before the current time, and the real-time navigation information includes navigation timestamps and road information of the road where the vehicle is located;

[0013] The real-time navigation information with the same road information in the multi-frame real-time navigation information is merged into a single reference navigation information to obtain at least one reference navigation information; the reference navigation information includes the road information, a start timestamp, and an end timestamp.

[0014] From the at least one set of reference navigation information, determine the target navigation information corresponding to the target intersection.

[0015] In some embodiments, the intersection information further includes the intersection location, and the reference navigation information further includes the vehicle location corresponding to the start timestamp and the vehicle location corresponding to the end timestamp;

[0016] Determining the target navigation information corresponding to the target intersection from the at least one reference navigation information includes:

[0017] Reference navigation information whose distance between the vehicle location corresponding to the termination timestamp and the intersection location meets a preset distance condition is determined as target navigation information corresponding to the target intersection; the preset distance condition is that the distance is equal to or less than a preset distance.

[0018] In some embodiments, after acquiring multi-frame real-time navigation information, the method further includes:

[0019] For each of the aforementioned real-time navigation information, the following operations are performed:

[0020] Obtain vehicle location information whose sampling timestamps meet preset time conditions; the preset time conditions are that the time interval between the sampling timestamp and the navigation timestamp in the real-time navigation information is equal to or less than a preset time interval, and the vehicle location information includes the sampling timestamp and the vehicle location collected at the time represented by the sampling timestamp;

[0021] The vehicle location information is added to the real-time navigation information to obtain updated real-time navigation information.

[0022] In some embodiments, it also includes:

[0023] If no target navigation information corresponding to the target intersection is determined from the at least one reference navigation information, a preset acquisition strategy is used to obtain the steering attributes corresponding to each of the first lanes, and the steering attributes corresponding to each of the first lanes are added to the intersection information to obtain updated road topology information.

[0024] In some embodiments, matching a plurality of second lanes with a plurality of first lanes based on the first number of lanes and the second number of lanes includes:

[0025] If the number of the first lanes is equal to the number of the second lanes, then the second lanes that are matched with each of the first lanes are determined in a preset order.

[0026] If the number of first lanes is less than the number of second lanes, and the road topology information includes road edge information, then based on the road edge information, multiple second lanes are matched with multiple first lanes; the road edge information is used to indicate whether a road edge is observed and the type of the observed road edge, wherein the type is left edge or right edge.

[0027] In some embodiments, matching the plurality of second lanes with the plurality of first lanes based on the road edge information includes:

[0028] If the road edge information indicates that the observed road edge and the type of the observed road edge are left edges, then starting from the leftmost lane, the second lanes that match each of the first lanes are determined sequentially until all of the first lanes are matched.

[0029] If the road edge information indicates that the observed road edge and the type of the observed road edge are right edges, then starting from the rightmost lane, the second lanes that match each of the first lanes are determined sequentially until all of the first lanes are matched.

[0030] In some embodiments, matching a plurality of second lanes with a plurality of first lanes based on the first number of lanes and the second number of lanes includes:

[0031] If the number of lanes in the first lane is greater than the number of lanes in the second lane, then the matching is determined to have failed;

[0032] If a matching failure is determined, a preset acquisition strategy is used to obtain the steering attributes corresponding to each of the first lanes, and the steering attributes corresponding to each of the first lanes are added to the intersection information to obtain updated road topology information.

[0033] In some embodiments, the target navigation information further includes the vehicle turning direction; adding the turning attributes corresponding to each of the first lanes to the intersection information to obtain updated road topology information includes:

[0034] The steering attributes corresponding to each of the first lanes are added to the intersection information, and the vehicle steering direction is added to the intersection information to obtain updated road topology information.

[0035] Secondly, this application provides a vehicle control method, including:

[0036] Based on road topology information, at least one first drivable lane is selected from multiple first lanes at the target intersection; the road topology information includes the steering attributes corresponding to the multiple first lanes and the vehicle steering direction corresponding to the target intersection; the road topology information is determined based on target navigation information corresponding to the target intersection, the target navigation information includes road information corresponding to the road where the vehicle is located, the road information includes the number of second lanes and the steering attributes corresponding to the multiple second lanes; the number of second lanes is the total number of lanes included in the road where the vehicle is located;

[0037] Determine the next section of road after passing the target intersection from the preset driving route, and select at least one second drivable lane from multiple lanes of the next section of road;

[0038] Generate at least one connection path from the at least one first drivable lane to the at least one second drivable lane;

[0039] According to a preset decision-making strategy, a target connection path is selected from the at least one connection path, and the vehicle is controlled to travel along the target connection path.

[0040] Thirdly, this application provides an electronic device, including: a processor and a memory communicatively connected to the processor;

[0041] The memory stores computer-executed instructions;

[0042] The processor executes computer execution instructions stored in the memory to implement the road topology determination method as described in any of the first aspects or the vehicle control method as described in the second aspect.

[0043] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the road topology determination method of any one of the first aspects or the vehicle control method of the second aspect.

[0044] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the road topology determination method described in any of the first aspects or the vehicle control method described in the second aspect.

[0045] In a sixth aspect, this application provides a mobile platform, including the electronic device described in the third aspect.

[0046] The road topology determination method, vehicle control method, device, medium, and product provided in this application combine road topology information with target navigation information corresponding to the target intersection. By matching lanes, the correspondence between the first lane at the target intersection and the second lane on the road where the vehicle is located is determined. The steering attribute of the second lane is then considered as the steering attribute of the matched first lane, and the steering attribute of the first lane is added to the intersection information corresponding to the target intersection. Due to the high accuracy of the navigation information, the accuracy of the determined lane steering attribute is also high, enabling the vehicle to choose the correct lane when approaching the target intersection, reducing traffic accidents or violations caused by incorrect steering. Compared to methods relying on camera perception systems, this application does not require complex neural network processing, reducing hardware costs and computational resource requirements. Furthermore, camera perception systems may have unstable performance in some special scenarios (such as changes in lighting, inclement weather, occlusion, etc.), leading to low accuracy. This application, by combining road topology information and navigation information, reduces the impact of these scenarios and improves accuracy. Compared to methods relying on third-party maps, this application uses navigation information, which reduces the cost of purchasing and updating map data, and the navigation information is updated quickly, improving the accuracy of road topology information. Furthermore, this application is applicable to various types of roads and traffic conditions, and is not limited to specific cities or regions, thus having broad applicability. Attached Figure Description

[0047] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0048] Figure 1 This is a schematic diagram illustrating an application scenario according to an exemplary embodiment;

[0049] Figure 2 This is a flowchart illustrating a road topology determination method according to an exemplary embodiment;

[0050] Figure 3 This is a flowchart illustrating a road topology determination method according to another exemplary embodiment;

[0051] Figure 4This is a schematic diagram illustrating a navigation information merging process according to an exemplary embodiment;

[0052] Figure 5 This is a flowchart illustrating a road topology determination method according to yet another exemplary embodiment;

[0053] Figure 6 This is a schematic flowchart illustrating a vehicle control method according to an exemplary embodiment;

[0054] Figure 7 This is a schematic diagram of the structure of a road topology determination device according to an exemplary embodiment;

[0055] Figure 8 This is a schematic diagram of the structure of a vehicle control device according to an exemplary embodiment;

[0056] Figure 9 This is a schematic diagram of the structure of an electronic device according to an exemplary embodiment.

[0057] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0058] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0059] The terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. In the following descriptions of embodiments, "a plurality of" means two or more, unless otherwise explicitly defined.

[0060] When vehicles travel on public transport roads, they are obligated to obey traffic rules in order to improve road safety and traffic efficiency. In intersection scenarios, vehicles need to obey intersection signs and road turning marks, and choose the appropriate lane to travel.

[0061] Human drivers can understand the semantics of each lane by observing the turn signs and intersection signs on the road, and accurately obtain the turn attributes of each lane, which can be left turn, straight or right turn.

[0062] For autonomous or assisted driving vehicles, in intersection scenarios, the first step is to acquire the steering attributes of each lane. Then, these lane steering attributes are correlated with a lane map to select the appropriate lane and driving behavior. This can be achieved either through camera perception systems using technologies such as Bird's Eye View Transformer (BEVFormer) and Optical Character Recognition (OCR) to obtain the semantic information of road signs and thus the steering attributes of each lane, or by leveraging third-party maps. Examples of third-party maps include Standard Definition Maps (SDMaps) and High Definition Maps (HDMaps).

[0063] Real-world road scenarios are diverse, and the aforementioned technologies, whether using camera-based perception systems to understand scene semantics or acquiring data through third-party maps, all have some limitations:

[0064] 1. Camera perception systems are based on neural networks, which require data training and iteration. They have inherent limitations in long-tail scenarios, and different cities have different road construction styles. For uncommon scenarios, camera perception systems have poor understanding capabilities.

[0065] 2. In heavy traffic scenarios, road signs are easily obscured by vehicles, which can affect the accuracy of the camera's perception system.

[0066] 3. For special types of lanes such as tidal lanes and reversible lanes, the turning marks on the road surface are not obvious and need to be understood in conjunction with intersection signs, which places high demands on the computing power of the camera perception system.

[0067] 4. The drawback of obtaining data through third-party maps is that:

[0068] There is a risk that the map data is not compliant;

[0069] The freshness of high-precision map data is difficult to guarantee: because the collection, processing and updating of map data requires a lot of manpower, material resources and financial resources, the data collection cycle is long, the data update cost is high and the technical difficulty is high;

[0070] High usage cost: Because the collection, processing and updating of map data require a lot of resources, third-party maps usually charge a certain fee.

[0071] To address the aforementioned technical problems, this application provides a road topology determination method, vehicle control method, device, medium, and product, aiming to provide a low-cost solution for effectively and accurately obtaining lane-level steering attributes.

[0072] Figure 1 This is a schematic diagram illustrating an application scenario according to an exemplary embodiment. For example... Figure 1 As shown, this application scenario includes: electronic device 1. Electronic device 1 is used to control vehicle movement. Exemplarily, electronic device 1 is installed on the vehicle, or electronic device 1 is installed independently, and electronic device 1 is capable of communicating with the vehicle. Exemplarily, electronic device 1 is a vehicle control terminal or other device capable of controlling vehicle movement, such as a vehicle domain controller.

[0073] In one scenario, when the vehicle is about to reach the target intersection, electronic device 1 obtains road topology information and target navigation information corresponding to the target intersection. It matches multiple first lanes at the target intersection with multiple second lanes on the road where the vehicle is located, and treats the steering attribute of the second lane as the steering attribute of the matched first lane. It then adds this information to the intersection information corresponding to the target intersection to obtain updated road topology information. Based on the updated road topology information, it controls the vehicle's movement.

[0074] The road topology determination method provided in this application is executed by a road topology determination device, which can be integrated into electronic device 1. The vehicle control method provided in this application is executed by a vehicle control device, which can be integrated into electronic device 1.

[0075] The technical solution of this application and how it solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0076] Figure 2 This is a flowchart illustrating a road topology determination method according to an exemplary embodiment. Figure 2 As shown, the road topology determination method provided in this embodiment includes the following steps:

[0077] Step S101: Obtain road topology information; the road topology information includes the intersection information corresponding to the target intersection, the intersection information includes the number of first lanes, and the number of first lanes is the total number of multiple first lanes at the target intersection.

[0078] The target intersection can be any intersection encountered during the vehicle's journey. An intersection refers to a road junction, that is, the point where two or more roads intersect. The lane at the target intersection where the vehicle is located can be called the first lane, and this first lane is located at the intersection of the vehicle's road and the target intersection.

[0079] Optionally, the intersection information corresponding to the target intersection also includes lane markings for each of the multiple first lanes. The lane markings for each first lane are used to uniquely identify that first lane, facilitating the differentiation between different lanes. Optionally, the representation of the lane markings can be set according to actual needs, and this application does not limit this. For example, if there are three first lanes at the target intersection, the lane markings for these three first lanes are: Lane 1, Lane 2, and Lane 3, respectively.

[0080] Optionally, the intersection information corresponding to the target intersection may also include the positional relationships between multiple first lanes. These positional relationships can be front-to-back, left-to-right, or adjacent. For example, lane 1 and lane 2 are left-to-right adjacent, with lane 1 located to the left of lane 2.

[0081] Optionally, the intersection information corresponding to the target intersection also includes the intersection location. The intersection location indicates the location of the target intersection. Optionally, the representation of the intersection location can be set according to actual needs, and this application does not limit this, for example, it can be represented by latitude and longitude coordinates.

[0082] Optionally, the road topology information also includes the intersection identifier corresponding to the target intersection, and the intersection identifier corresponding to the target intersection exists in the road topology information in correspondence with the intersection information. The intersection identifier is used to uniquely identify the target intersection to distinguish different intersections. Optionally, the representation of the intersection identifier can be set according to actual needs, and this application does not limit it; for example, the intersection identifier could be "Intersection A".

[0083] Optionally, the road topology information also includes road edge information. This road edge information indicates whether a road edge has been observed and the type of observed road edge, which is either a left edge or a right edge. Optionally, the representation of the road edge information can be set according to actual needs, and this application does not limit this. For example, it can be represented by two bits: the first bit indicates whether a road edge has been observed (e.g., 0 indicates no observation, 1 indicates observation), and the second bit indicates the type of observed road edge (e.g., 0 indicates a left edge, 1 indicates a right edge).

[0084] Regarding the methods for obtaining road topology information:

[0085] In some embodiments, intersection information corresponding to multiple intersections can be pre-stored, for example, these multiple intersections are intersections within a preset address range. The preset address range can be set according to actual needs, and this application does not limit it. For example, the preset address range can be the vehicle's frequently used address range, such as city A and city B. For autonomous vehicles, the driving route is determined when starting to drive, so the intersections that may be passed during the driving process are determined. Accordingly, the intersections that the current driving route may pass through can be filtered from the stored multiple intersections, and the intersection information corresponding to the filtered intersections can be obtained, thereby obtaining road topology information. Accordingly, the road topology information includes the intersection information corresponding to each intersection that the current driving route may pass through. The filtered intersections include the aforementioned target intersection. In practical applications, considering the limited local storage space, the intersection information corresponding to multiple intersections can be stored in the cloud, so that the road topology information can be obtained from the cloud when the vehicle starts to drive. Thus, when the vehicle is about to reach the target intersection, the road topology information of the target intersection can be obtained from the filtered intersections.

[0086] In other embodiments, for each intersection that the current driving route may pass through, when the vehicle is about to reach any intersection, taking the target intersection as an example, a camera perception system is used to obtain the intersection information corresponding to the target intersection, thereby obtaining road topology information.

[0087] Step S102: Obtain target navigation information corresponding to the target intersection; the target navigation information includes road information corresponding to the road where the vehicle is located, the road information includes the number of second lanes and the turning attributes corresponding to the multiple second lanes respectively; the number of second lanes is the total number of lanes included in the road where the vehicle is located.

[0088] In the target navigation information, the lane on the road where the vehicle is located can be called the second lane, and the road where the vehicle is located includes multiple second lanes. The number of second lanes is also the total number of these multiple second lanes.

[0089] The lane's steering attribute is used to determine whether the lane allows vehicles to make a U-turn, turn left, turn right, or go straight. Accordingly, the steering attribute is at least one of U-turn, left turn, go straight, or right turn.

[0090] Optionally, the road information also includes lane markings corresponding to multiple second lanes. The lane markings for each second lane are used to uniquely identify that second lane, facilitating differentiation between different lanes. Optionally, the way the lane markings are represented can be set according to actual needs; this application does not limit this. For example, if there are three second lanes on the road where the vehicle is located, the lane markings for these three second lanes are: Lane 1, Lane 2, and Lane 3, respectively.

[0091] Optionally, the target navigation information may also include the positional relationships between multiple second lanes. These positional relationships can be front-to-back, left-to-right, or adjacent. For example, lane 1 and lane 2 may be adjacent to each other, with lane 1 located to the left of lane 2.

[0092] Optionally, the target navigation information also includes drivable lane information. This drivable lane information indicates which lanes among multiple second lanes are drivable. "Drivable" means drivable under the premise of complying with traffic rules and following the navigation route.

[0093] Optionally, the target navigation information also includes information about the next intersection. This information pertains to the intersection the vehicle is about to reach. It includes the intersection sign, the distance between the vehicle and the next intersection, and the vehicle's turning direction at the next intersection. The turning direction is determined by the vehicle's route and can be one of a U-turn, left turn, straight ahead, or right turn. For example, a left turn indicates the vehicle needs to turn left at the next intersection, while a straight ahead indicates the vehicle needs to go straight at the next intersection.

[0094] In some embodiments, the target navigation information can be obtained by using local navigation software or third-party navigation software to obtain real-time navigation information of the target intersection when the vehicle is about to arrive at the target intersection. For example, the application programming interface (API) of the third-party navigation software can be called to obtain the real-time navigation information of the target intersection.

[0095] It should be noted that this application does not limit the execution order of steps S101 and S102. For example, step S101 can be executed first, followed by step S102, that is, when the vehicle is about to reach the target intersection, the road topology information is obtained first, and then the target navigation information corresponding to the target intersection is obtained; or step S102 can be executed first, followed by step S101, that is, when the vehicle is about to reach the target intersection, the target navigation information corresponding to the target intersection is obtained first, and then the road topology information is obtained; or steps S101 and S102 can be executed simultaneously. For ease of description, this embodiment uses the example of executing step S101 first, followed by step S102, for illustration. Other cases are similar and will not be described in detail in this embodiment.

[0096] Step S103: Based on the number of first lanes and the number of second lanes, match multiple second lanes with multiple first lanes.

[0097] In this embodiment, the correspondence between the first lane and the second lane can be determined by matching the first lane with the second lane. Optionally, after matching multiple second lanes with multiple first lanes, the lane identifiers corresponding to the successfully matched first lanes and the lane identifiers corresponding to the second lanes can be stored to realize the storage of the correspondence between the first lane and the second lane.

[0098] Step S104: Add the steering attributes corresponding to each first lane to the intersection information to obtain updated road topology information; the steering attribute corresponding to the first lane is the steering attribute corresponding to the second lane that matches the first lane.

[0099] In this embodiment, the road information includes the turning attributes corresponding to each second lane. After matching multiple second lanes with multiple first lanes, the correspondence between the first lanes and the second lanes is obtained. Accordingly, for any first lane, the turning attribute corresponding to the second lane matched with the first lane can be regarded as the turning attribute corresponding to the first lane. Thus, the turning attributes corresponding to each first lane can be added to the intersection information to obtain the updated intersection information, and then the updated road topology information is obtained.

[0100] Optionally, the intersection information corresponding to the target intersection also includes lane markings corresponding to multiple first lanes, and in the intersection information, for each first lane, the lane marking and turning attribute of that first lane exist.

[0101] In some embodiments, the updated road topology information has more complete intersection information, so that when a vehicle travels to a target intersection, the vehicle's travel can be controlled according to the updated road topology information.

[0102] This embodiment combines road topology information with target navigation information corresponding to the target intersection. By matching lanes, it determines the correspondence between the first lane at the target intersection and the second lane on the road where the vehicle is located. The steering attribute of the second lane is then considered as the steering attribute of the matched first lane. This steering attribute of the first lane is then added to the intersection information corresponding to the target intersection, enabling the vehicle to choose the correct lane when approaching the target intersection, reducing traffic accidents or violations caused by incorrect steering. Compared to methods relying on camera perception systems, this application does not require complex neural network processing, reducing hardware costs and computational resource requirements. Furthermore, camera perception systems may experience performance instability in certain scenarios (such as changes in lighting, inclement weather, occlusion, etc.), leading to low accuracy. This application, by combining road topology information and navigation information, reduces the impact of these scenarios and improves accuracy. Compared to methods relying on third-party maps, this application uses navigation information, reducing the cost of purchasing and updating map data. Navigation information is also updated quickly, improving the accuracy of road topology information. In addition, this application is applicable to various types of roads and traffic conditions, not limited to specific cities or regions, and has broad applicability.

[0103] Figure 3 This is a flowchart illustrating a road topology determination method according to another exemplary embodiment, such as... Figure 3 As shown, the road topology determination method provided in this embodiment is based on the road topology determination method provided in the previous embodiment of this application, and further refines step S102. The road topology determination method provided in this embodiment includes the following steps:

[0104] Step S201: When the vehicle is about to reach the target intersection, obtain road topology information.

[0105] In some embodiments, step S201 is implemented in the same way as step S101 in the previous embodiment, and will not be described again here.

[0106] In this embodiment, after step S201, step S102 is executed, that is, the target navigation information corresponding to the target intersection is obtained. In some embodiments, step S102 is implemented by including the following steps S202-S204.

[0107] Step S202: Obtain multi-frame real-time navigation information; the multi-frame real-time navigation information includes the current time and real-time navigation information of multiple consecutive historical moments before the current time, and the real-time navigation information includes navigation timestamps and road information of the road where the vehicle is located.

[0108] The navigation timestamp in the real-time navigation information indicates the time when the real-time navigation information was generated, and the road where the vehicle is located is the road where the vehicle is located at that time.

[0109] The road information for the road where the vehicle is located includes the number of third lanes and the turning attributes corresponding to each of the third lanes. The number of third lanes is the total number of lanes included in the road where the vehicle is located. The turning attribute of a lane refers to whether the lane allows vehicles to make a U-turn, left turn, right turn, or go straight. Accordingly, the turning attribute is at least one of U-turn, left turn, straight, or right turn.

[0110] Optionally, the road information also includes lane markings corresponding to multiple third lanes. The lane markings for each third lane are used to uniquely identify that third lane, facilitating differentiation between different lanes. Optionally, the way the lane markings are represented can be set according to actual needs; this application does not limit this. For example, if there are three third lanes on the road where the vehicle is located, the lane markings for these three third lanes are: Lane 1, Lane 2, and Lane 3, respectively.

[0111] Optionally, the road information also includes the positional relationships between multiple third lanes. These positional relationships can be front-to-back, left-to-right, or adjacent. For example, lane 1 and lane 2 are adjacent to each other, with lane 1 located to the left of lane 2.

[0112] Optionally, the real-time navigation information also includes drivable lane information. This drivable lane information is used to indicate which lanes are drivable for vehicles among multiple third lanes.

[0113] Optionally, the real-time navigation information also includes information about the next intersection. This information pertains to the intersection the vehicle is about to reach. It includes the intersection sign, the distance between the vehicle and the intersection, and the vehicle's turning direction at the intersection. The turning direction is determined by the vehicle's route and can be one of a U-turn, left turn, straight ahead, or right turn.

[0114] Optionally, the method for obtaining multi-frame real-time navigation information is as follows: using local navigation software or third-party navigation software to obtain multi-frame real-time navigation information. For example, calling the API interface of the local navigation software or third-party navigation software to obtain real-time navigation information. For example, real-time navigation information can be obtained periodically; for instance, a preset number of real-time navigation information frames can be obtained and stored each time, so that when the vehicle is about to reach the target intersection, the stored multi-frame real-time navigation information can be obtained. The preset number can be one frame or multiple frames, and can be set according to actual needs; this application does not limit this.

[0115] Optionally, the total number of frames for multi-frame real-time navigation information can be set according to actual needs, and this application does not limit it. For example, the total number of frames can be 4, 5, or 6.

[0116] In some scenarios, real-time navigation information also includes vehicle location, which is the vehicle's position at the time indicated by the navigation timestamp. Optionally, the method of representing the vehicle location can be set according to actual needs, and this application does not limit it, for example, it can be represented by latitude and longitude coordinates.

[0117] In other scenarios, the real-time navigation information does not include vehicle location. In such cases, it is necessary to additionally bind the corresponding vehicle location to the real-time navigation information. Optionally, after step S202, the road topology determination method provided in this embodiment further includes: performing the following operations for each piece of real-time navigation information, including:

[0118] Obtain vehicle location information whose sampling timestamps meet preset time conditions; the preset time conditions are that the time interval between the sampling timestamp and the navigation timestamp in the real-time navigation information is equal to or less than a preset time interval, and the vehicle location information includes the sampling timestamp and the vehicle location collected at the time represented by the sampling timestamp.

[0119] Add vehicle location information to the real-time navigation information to obtain updated real-time navigation information.

[0120] For example, during vehicle operation, the vehicle's position is periodically collected, for instance, through the vehicle's positioning system. This vehicle position information is then stored. For any given real-time navigation information, vehicle position information whose sampling timestamp matches a preset time condition is retrieved from the stored vehicle position information. If the sampling timestamp matches the preset time condition, it means that the sampling timestamp is the same as or close to the navigation timestamp. Therefore, the vehicle position collected at the sampling timestamp can be approximated as the vehicle position collected at the navigation timestamp, and this vehicle position information can be added to the real-time navigation information to obtain updated real-time navigation information. The preset time interval can be set according to actual needs, and this application does not limit it. For example, the preset time interval could be 0.1 seconds, 0.2 seconds, etc.

[0121] The above implementation method adds vehicle location information with sampling timestamps that meet preset time conditions to the real-time navigation information, which can enrich the content of the real-time navigation information and provide a basis for vehicle navigation.

[0122] For example, see Figure 4In the left image, the lower road has four lanes, from left to right: Lane 1, Lane 2, Lane 3, and Lane 4. The upper road has three lanes, from left to right: Lane 5, Lane 6, and Lane 7. Vehicles travel along Lane 3 in the lower road, then move to Lane 6 in the upper road. Solid and hollow dots represent the vehicle's position at the time indicated by the sampling timestamp, and each dot has a corresponding box to its right. Solid dots are bound to real-time navigation information, and their corresponding boxes contain the turning attributes of each lane provided by the real-time navigation information. For example, the turning attributes for each lane in the lower road are U-turn, left turn, straight, and straight + right turn; the turning attributes for each lane in the upper road are left turn, left turn + straight, and straight. Hollow dots are not bound to real-time navigation information, and their corresponding boxes are empty, indicating that there is no corresponding real-time navigation information. Figure 4 In the image, the solid dots representing the vehicle's position and the squares representing the real-time navigation information together form a frame of updated real-time navigation information. This real-time navigation information also includes the total number of lanes on the road where the vehicle is located and the navigation timestamp. Figure 4 (Not shown).

[0123] It should be noted that the above implementation method adds vehicle location information to real-time navigation information. Optionally, only the vehicle location information can be added to real-time navigation information to reduce information redundancy.

[0124] Step S203: Merge real-time navigation information with the same road information from multiple frames of real-time navigation information into a single reference navigation information to obtain at least one reference navigation information; the reference navigation information includes the road information, a start timestamp, and an end timestamp.

[0125] In practical applications, the navigation timestamps of two consecutive frames of real-time navigation information are relatively close, and the road information they contain is likely to be the same. Therefore, multiple frames of real-time navigation information can be preprocessed to merge real-time navigation information with the same road information to obtain reference navigation information. This reference navigation information can be regarded as a navigation information group, and the information it contains can represent the merged multiple frames of real-time navigation information.

[0126] The start timestamp of the reference navigation information is the earliest navigation timestamp among multiple frames of real-time navigation information with the same road information, and the end timestamp is the latest navigation timestamp among multiple frames of real-time navigation information with the same road information.

[0127] Optionally, if each frame of real-time navigation information also includes vehicle position, then the reference navigation information also includes the vehicle position from each frame of merged real-time navigation information. Furthermore, the reference navigation information may also include only the vehicle position corresponding to the start timestamp and the vehicle position corresponding to the end timestamp to reduce information redundancy.

[0128] Optionally, if each frame of real-time navigation information also includes drivable lane information, then the reference navigation information also includes the drivable lane information from each frame of merged real-time navigation information. Furthermore, if the drivable lane information in each frame of merged real-time navigation information is the same, the reference navigation information may include only one frame of drivable lane information to reduce information redundancy.

[0129] Optionally, if each frame of real-time navigation information also includes information related to the next intersection, then the reference navigation information also includes the information related to the next intersection from each frame of merged real-time navigation information. Furthermore, if the information related to the next intersection is the same in each frame of merged real-time navigation information, the reference navigation information may include only one frame of information related to the next intersection to reduce information redundancy.

[0130] For example, see Figure 4 The right image shows that multiple frames of real-time navigation information are merged to obtain two reference navigation information.

[0131] Step S204: Determine the target navigation information corresponding to the target intersection from at least one reference navigation information.

[0132] Steps S202-S204 are one way to obtain target navigation information corresponding to the target intersection. By obtaining multiple frames of real-time navigation information and merging real-time navigation information with the same road information into a reference navigation information, the current road conditions can be effectively reflected and redundant data can be reduced. Based on this, at least one reference navigation information obtained by merging may contain target navigation information corresponding to the target intersection. Therefore, the target navigation information can be determined, providing a basis for subsequent lane matching.

[0133] In some embodiments, the intersection information corresponding to the target intersection also includes the intersection location, and the reference navigation information also includes the vehicle location corresponding to the start timestamp and the vehicle location corresponding to the end timestamp. Target navigation information can then be filtered based on the vehicle location and the intersection location. Accordingly, step S204 is implemented by: determining the reference navigation information whose distance between the vehicle location corresponding to the end timestamp and the intersection location meets a preset distance condition as the target navigation information; the preset distance condition is that the distance is equal to or less than a preset distance.

[0134] The fact that the distance between the vehicle location corresponding to the termination timestamp and the intersection location meets the preset distance condition indicates that the vehicle location corresponding to the termination timestamp and the intersection location are relatively close. Therefore, the closer the road information in the reference navigation information to which the termination timestamp belongs is to the intersection information corresponding to the target intersection, the more likely the reference navigation information can be regarded as the target navigation information corresponding to the target intersection. The preset distance can be set according to actual needs, and this application does not limit it. For example, the preset distance can be 5 meters, 4 meters, 3 meters, etc.

[0135] By using the distance between the vehicle's location and the intersection as a filtering criterion, navigation information related to the target intersection can be determined more accurately. This method can effectively avoid using irrelevant navigation information as a reference and improve the accuracy of the target navigation information.

[0136] In other embodiments, target navigation information can be filtered based on timestamps. Accordingly, step S204 is implemented by determining the reference navigation information to which the end timestamp closest to the current time belongs as the target navigation information. Specifically, if the vehicle is about to reach the target intersection, and the closer the end timestamp of a reference navigation information is to the current time, the closer the road information in that reference navigation information is to the intersection information corresponding to the target intersection; therefore, that reference navigation information can be considered as the target navigation information corresponding to the target intersection. For example, if the time difference between the end timestamp and the current time is smaller, it indicates that the end timestamp and the current time are closer; conversely, if the time difference between the end timestamp and the current time is larger, it indicates that the end timestamp and the current time are farther apart.

[0137] This method uses the time difference between the end timestamp and the current time as a filtering condition. It can filter target navigation information using only the timestamp without any other additional information. Furthermore, by selecting the reference navigation information to which the end timestamp closest to the current time belongs, it can ensure that the navigation information used is up-to-date and can reflect the current road conditions to the greatest extent.

[0138] In some embodiments, the road topology information also includes the intersection sign corresponding to the target intersection, and the reference navigation information also includes information related to the next intersection, which includes the intersection sign of the next intersection. Accordingly, target navigation information can be filtered based on the intersection sign. Accordingly, step S204 is implemented by determining the reference navigation information whose intersection sign of the next intersection is the same as the intersection sign of the target intersection as the target navigation information. Wherein, if the intersection sign of the next intersection is the same as the intersection sign of the target intersection, it means that the next intersection and the target intersection are the same intersection. Then, the closer the road information in the reference navigation information to which the intersection sign of the next intersection belongs is to the intersection information corresponding to the target intersection, the more likely the reference navigation information can be regarded as the target navigation information corresponding to the target intersection.

[0139] By matching the intersection sign of the next intersection with the intersection sign of the target intersection, it can be ensured that the selected reference navigation information is highly consistent with the actual situation of the target intersection. This method can effectively avoid using irrelevant navigation information as a reference and improve the accuracy of the target navigation information.

[0140] In practical applications, navigation information exists only in certain locations. At least one set of reference navigation information may or may not contain target navigation information corresponding to the target intersection. In some embodiments, if target navigation information corresponding to the target intersection is determined from at least one set of reference navigation information, steps S206-S207 are executed to update the road topology information; if target navigation information corresponding to the target intersection is not determined from at least one set of reference navigation information, step S205 is executed to update the road topology information.

[0141] Step S205: Use a preset acquisition strategy to obtain the steering attributes corresponding to each first lane, and add the steering attributes corresponding to each first lane to the intersection information to obtain updated road topology information.

[0142] The preset acquisition strategy can be to use a camera perception system to obtain the steering attributes corresponding to each first lane. Specifically, a camera is used to acquire images of the target intersection, the acquired images are preprocessed (such as noise reduction, contrast enhancement, color correction, etc. to improve image quality), and computer vision algorithms are used to detect lane lines in the images. Machine learning or deep learning models are used to identify the detected lane lines to identify lane markings (such as arrows, text, etc.), thereby obtaining the steering attributes corresponding to each lane.

[0143] This embodiment provides an alternative solution that, in the absence of target navigation information, can use a preset acquisition strategy to obtain the steering attributes corresponding to each first lane, thereby updating the road topology information and improving flexibility and reliability.

[0144] In some embodiments, step S205 may be executed only if the target navigation information corresponding to the target intersection has not been determined. Alternatively, when obtaining the road topology information in step S201, a preset acquisition strategy may be used to obtain the steering attributes corresponding to each first lane and add the steering attributes corresponding to each first lane to the intersection information for subsequent use.

[0145] In this embodiment, step S208 is executed after step S205.

[0146] Step S206: Based on the number of first lanes and the number of second lanes, match multiple second lanes with multiple first lanes.

[0147] In some embodiments, step S206 can be implemented in several ways, including the following:

[0148] In the first case, if the number of first lanes is equal to the number of second lanes, then the second lanes that are matched with each of the first lanes are determined in a preset order.

[0149] Optionally, the preset order can be any of the following: from the leftmost lane to the rightmost lane, from the rightmost lane to the leftmost lane, from the middle lane to both sides, or other suitable matching order.

[0150] The number of lanes in the first lane can be considered as the number of lanes observed, and the number of lanes in the second lane can be considered as the actual number of lanes. If the number of lanes in the first lane is the same as the number of lanes in the second lane, it means that the observation is relatively accurate, and they can be matched sequentially according to the preset order.

[0151] For example, if the total number of lanes at the target intersection, i.e. the number of first lanes, is 4, and the four first lanes from left to right are lane 1, lane 2, lane 3, and lane 4, and the total number of lanes on the road where the vehicle is located, i.e. the number of second lanes, is 4, and the four second lanes from left to right are lane 5, lane 6, lane 7, and lane 8, then lane 1 matches lane 5, lane 2 matches lane 6, lane 3 matches lane 7, and lane 4 matches lane 8.

[0152] In the second case, if the number of first lanes is less than the number of second lanes, and the road topology information includes road edge information, then multiple second lanes are matched with multiple first lanes based on the road edge information; the road edge information is used to indicate whether the road edge is observed and the type of the observed road edge, which is either the left edge or the right edge.

[0153] If the number of lanes in the first lane is less than the number of lanes in the second lane, it indicates that the number of observed lanes is less than the actual number of lanes, suggesting inaccurate observation. This could be due to obstructions during observation, or the observation range not covering all lanes, such as a large number of lanes and the vehicle's lane being at the edge, or the vehicle's camera having a small field of view. Considering that road topology information may also include road edge information, which indicates whether road edges were observed and the type of road edge indicates its location, observing the left edge suggests that the overall position of multiple first lanes is more biased towards the left edge, and observing the right edge suggests that the overall position of multiple first lanes is more biased towards the right edge. Therefore, road edge information can be combined to match lanes, thus ensuring matching accuracy.

[0154] Optionally, the method of matching multiple second lanes with multiple first lanes based on road edge information includes:

[0155] If the road edge information indicates that the observed road edge and the type of the observed road edge are left edges, then starting from the leftmost lane, the second lanes that match each first lane are determined sequentially until all first lanes are matched.

[0156] If the road edge information indicates that the observed road edge and the type of the observed road edge are right edges, then starting from the rightmost lane, the second lanes that match each of the first lanes are determined sequentially until all the first lanes are matched.

[0157] If the left edge is observed, it means that the overall position of multiple first lanes is more biased towards the left edge. In this case, matching can start from the leftmost lane and proceed sequentially. For example, if there are 3 first lanes, and the 3 first lanes from left to right are lane 1, lane 2, and lane 3, and the left edge is observed, and there are 4 second lanes, and the 4 second lanes from left to right are lane 5, lane 6, lane 7, and lane 8, then lane 1 matches lane 5, lane 2 matches lane 6, and lane 3 matches lane 7.

[0158] If the right edge is observed, it means that the overall position of multiple first lanes is more biased towards the right edge. In this case, matching can start from the rightmost lane and proceed sequentially. For example, if there are 3 first lanes, and the 3 first lanes from left to right are lane 1, lane 2, and lane 3, and the right edge is observed, and there are 4 second lanes, and the 4 second lanes from left to right are lane 5, lane 6, lane 7, and lane 8, then lane 3 matches lane 8, lane 2 matches lane 7, and lane 1 matches lane 6.

[0159] By using clear road edge information to guide the matching process, the risk of being misled due to inaccurate lane numbers can be effectively avoided. Especially in multi-lane or complex road conditions, using road edge information for matching can significantly reduce matching errors, enabling more accurate lane matching and improving accuracy.

[0160] In the third case, if the number of lanes in the first lane is greater than the number of lanes in the second lane, then the matching is determined to be unsuccessful; accordingly, if the matching is determined to be unsuccessful, then step S205 is executed.

[0161] If the number of first lanes is greater than the number of second lanes, it means that the number of observed lanes is greater than the actual number of lanes. This situation may be caused by an observation error or an error in the navigation information. If lane matching is still performed, the matching result is likely to be incorrect and has little reference value. Therefore, it can be directly determined that the matching has failed, and step S205 is executed, that is, the preset acquisition strategy is used to obtain the turning attributes corresponding to each first lane, and the turning attributes corresponding to each first lane are added to the intersection information to obtain the updated road topology information.

[0162] This embodiment provides an alternative solution so that when the number of first lanes is greater than the number of second lanes, a preset acquisition strategy can be used to obtain the steering attributes corresponding to each first lane, thereby updating the road topology information and improving flexibility and reliability.

[0163] Step S207: Add the steering attributes corresponding to each first lane to the intersection information to obtain updated road topology information; the steering attribute corresponding to the first lane is the steering attribute corresponding to the second lane that matches the first lane.

[0164] In some embodiments, step S207 is implemented in the same way as step S104 in the previous embodiment, and will not be described again here.

[0165] In some embodiments, the target navigation information also includes the vehicle turning direction, for example, the vehicle turning direction may be included in the information related to the next intersection, that is, the turning direction of the vehicle at the next intersection. Accordingly, step S207 is implemented by adding the turning attributes corresponding to each first lane to the intersection information and adding the vehicle turning direction to the intersection information to obtain updated road topology information.

[0166] By adding the vehicle's turning direction to the intersection information, a basis can be provided for selecting the appropriate lane based on the vehicle's turning direction.

[0167] This application proposes a method to improve the accuracy of lane turning attributes using real-time navigation information. Benefiting from the large user base and widespread application of navigation software, its errors can be corrected promptly, making the accuracy of real-time navigation information higher than that of camera-based perception systems. This, combined with the updated road topology information from real-time navigation, results in more accurate road topology information, enabling vehicle control based on this information. This effectively improves the driving experience of autonomous driving systems and reduces the risk of traffic violations. Furthermore, real-time navigation information is relatively inexpensive and carries no risk of data non-compliance, thus possessing broad applicability across various scenarios.

[0168] In some embodiments, local navigation software or third-party navigation software can directly provide the turning attributes corresponding to each lane at the intersection. For ease of description and distinction, this lane will be referred to as the fourth lane below. Unlike the first lane, which is the lane at the target intersection included in the road topology information, the fourth lane is the lane at the target intersection provided by the local navigation software or third-party navigation software. Accordingly, Figure 5 This is a flowchart illustrating a road topology determination method according to yet another exemplary embodiment, such as... Figure 5 As shown, the method includes the following steps:

[0169] Step S301: Obtain road topology information; the road topology information includes the intersection information corresponding to the target intersection, the intersection information includes the number of first lanes, and the number of first lanes is the total number of multiple first lanes at the target intersection.

[0170] In some embodiments, the implementation of step S301 is the same as that of step S101 in the above embodiments, and will not be described again here.

[0171] Step S302: Use local navigation software or third-party navigation software to obtain the number of fourth lanes and the turning attributes corresponding to each fourth lane at the target intersection.

[0172] The first lane number is the total number of lanes at the target intersection included in the road topology information. Unlike the first lane number, the fourth lane number is the total number of lanes at the target intersection provided by local navigation software or third-party navigation software.

[0173] Step S303: Based on the number of first lanes and the number of fourth lanes, match multiple first lanes with multiple fourth lanes.

[0174] In some embodiments, step S303 is implemented in the same way as step S206 in the above embodiments, and will not be described again here.

[0175] Step S304: Add the steering attributes corresponding to each first lane to the intersection information to obtain updated road topology information. The steering attribute corresponding to the first lane is the steering attribute corresponding to the fourth lane that matches the first lane.

[0176] In some embodiments, step S304 is implemented in the same way as step S207 in the above embodiments, and will not be described again here.

[0177] In this embodiment, local navigation software or third-party navigation software can directly provide the turning attributes corresponding to each lane at the intersection, thereby updating the road topology information without using navigation information to update the road topology information, which is more efficient and has higher real-time performance.

[0178] Figure 6 This is a schematic flowchart illustrating a vehicle control method according to an exemplary embodiment. Figure 6 As shown, the vehicle control method provided in this embodiment includes the following steps:

[0179] Step S401: Based on road topology information, select at least one first drivable lane from multiple first lanes at the target intersection; the road topology information includes the steering attributes corresponding to the multiple first lanes and the vehicle steering direction corresponding to the target intersection; the road topology information is determined based on the target navigation information corresponding to the target intersection, the target navigation information includes the road information corresponding to the road where the vehicle is located, the road information includes the number of second lanes and the steering attributes corresponding to the multiple second lanes; the number of second lanes is the total number of lanes included in the road where the vehicle is located.

[0180] In some embodiments, the road topology information determination process is described above. Figure 2 , Figure 3 or Figure 5 The embodiments shown are not described in detail here.

[0181] For example, if the lane turning direction is right, then the first lane with the right-turn attribute is designated as the first drivable lane; if the lane turning direction is left, then the first lane with the left-turn attribute is designated as the first drivable lane; and if the lane turning direction is straight, then the first lane with the straight-go attribute is designated as the first drivable lane.

[0182] Step S402: Determine the next section of road after passing the target intersection from the preset driving route, and select at least one second drivable lane from the multiple lanes of the next section of road.

[0183] The preset driving route is determined when the vehicle starts driving and can be updated during the driving process. Optionally, the preset driving route is a route selected by the user in the navigation software or a route automatically planned by the vehicle's autonomous driving software.

[0184] Optionally, the preset driving route clearly marks the next section of road after each intersection, and the next section of road after passing the target intersection can be directly determined from the preset driving route.

[0185] Optionally, selecting at least one second drivable lane from multiple lanes in the next road segment is specifically implemented by: obtaining road information of the next road segment; and selecting at least one second drivable lane based on the obtained road information.

[0186] The road information includes the total number of lanes in the next road segment and the turning attributes of each lane. Optionally, the road information may also include at least one of the following: lane markings for each lane, the positional relationships between multiple lanes, and information on drivable lanes.

[0187] Optionally, at least one second drivable lane can be selected based on the obtained road information. The criteria for selecting a lane can be set according to specific needs. For example, priority can be given to straight lanes: if a vehicle needs to go straight, then a straight lane can be selected first; or, the lane can be selected based on traffic conditions: select a lane with better traffic conditions to avoid congestion; or, the lane can be selected based on lane type: select an appropriate lane based on vehicle type and driving needs (e.g., buses can select bus lanes).

[0188] Step S403: Generate at least one connection path from at least one first drivable lane to at least one second drivable lane.

[0189] Optionally, for each of the at least one first drivable lanes, a connection path is generated between the first drivable lane and each of the at least one second drivable lane, thereby obtaining at least one connection path corresponding to each first drivable lane.

[0190] For example, at least one first drivable lane includes lane 1 and lane 2, and at least one second drivable lane includes lane 3 and lane 4. For lane 1, a connection path is generated between lane 1 and lane 3 and between lane 1 and lane 4. For lane 2, a connection path is generated between lane 2 and lane 3 and between lane 2 and lane 4, thus obtaining a total of 4 connection paths.

[0191] Step S404: Select a target connection path from at least one connection path according to a preset decision strategy, and control the vehicle to travel along the target connection path.

[0192] The preset decision-making strategy can be provided by the vehicle's decision-making and planning function module. Optionally, the preset decision-making strategy can be to randomly select a connection path as the target connection path from at least one connection path, or to prioritize the selection of a connection path that includes the vehicle's current lane as the target connection path, or to prioritize the selection of a connection path that matches the user's preferences as the target connection path.

[0193] This embodiment selects a first drivable lane based on the vehicle's turning direction and lane turning attributes in the road topology information, and selects a second drivable lane from a preset driving route. It then generates a connecting path from the first drivable lane to the second drivable lane and selects the optimal path according to a preset decision-making strategy, ensuring the accuracy of path selection. This avoids vehicles taking the wrong route at complex intersections and improves navigation accuracy. By clearly defining the turning attributes and vehicle turning direction to guide path selection, the risk of being misled by inaccurate or incomplete lane information can be effectively avoided. Especially at complex or multi-lane intersections, clear path information can effectively reduce the possibility of incorrect navigation.

[0194] Figure 7 This is a schematic diagram illustrating the structure of a road topology determination device according to an exemplary embodiment, such as... Figure 7 As shown, in this embodiment, the road topology determination device 50 can be installed in an electronic device, and the road topology determination device 50 includes:

[0195] The first acquisition module 501 is used to acquire road topology information; the road topology information includes the intersection information corresponding to the target intersection, and the intersection information includes the number of first lanes, which is the total number of multiple first lanes at the target intersection;

[0196] The second acquisition module 502 is used to acquire target navigation information corresponding to the target intersection; the target navigation information includes road information corresponding to the road where the vehicle is located, the road information includes the number of second lanes and the turning attributes corresponding to the multiple second lanes respectively; the number of second lanes is the total number of lanes included in the road where the vehicle is located;

[0197] The matching module 503 is used to match multiple second lanes with multiple first lanes based on the number of first lanes and the number of second lanes;

[0198] The update module 504 is used to add the steering attributes corresponding to each first lane to the intersection information to obtain the updated road topology information; the steering attribute corresponding to the first lane is the steering attribute corresponding to the second lane that matches the first lane.

[0199] In some embodiments, the second acquisition module 502 is configured to:

[0200] Acquire multi-frame real-time navigation information; multi-frame real-time navigation information includes real-time navigation information at the current time and multiple consecutive historical moments before the current time, and real-time navigation information includes navigation timestamps and road information of the road where the vehicle is located;

[0201] Real-time navigation information with the same road information in multiple frames of real-time navigation information is merged into a single reference navigation information to obtain at least one reference navigation information; the reference navigation information includes road information, start timestamp, and end timestamp.

[0202] Determine the target navigation information corresponding to the target intersection from at least one reference navigation information.

[0203] In some embodiments, the intersection information also includes the intersection location, and the reference navigation information also includes the vehicle location corresponding to the start timestamp and the vehicle location corresponding to the end timestamp;

[0204] The second acquisition module 502, when determining the target navigation information corresponding to the target intersection from at least one reference navigation information, is used to:

[0205] Reference navigation information whose distance between the vehicle location corresponding to the termination timestamp and the intersection location meets the preset distance conditions is determined as the target navigation information corresponding to the target intersection; the preset distance conditions are that the distance is equal to or less than the preset distance.

[0206] In some embodiments, after acquiring multiple frames of real-time navigation information, the second acquisition module 502 is further configured to:

[0207] For each piece of real-time navigation information, perform the following operations:

[0208] Obtain vehicle location information whose sampling timestamps meet preset time conditions; the preset time conditions are that the time interval between the sampling timestamp and the navigation timestamp in the real-time navigation information is equal to or less than the preset time interval, and the vehicle location information includes the sampling timestamp and the vehicle location collected at the time represented by the sampling timestamp.

[0209] Add vehicle location information to the real-time navigation information to obtain updated real-time navigation information.

[0210] In some embodiments, the update module 504 is further configured to:

[0211] If no target navigation information corresponding to the target intersection is determined from at least one reference navigation information, a preset acquisition strategy is used to obtain the steering attributes corresponding to each first lane, and the steering attributes corresponding to each first lane are added to the intersection information to obtain updated road topology information.

[0212] In some embodiments, the matching module 503 is configured to:

[0213] If the number of first lanes is equal to the number of second lanes, then the second lanes that match each first lane are determined in a preset order; the preset order is from the leftmost lane to the rightmost lane or from the rightmost lane to the leftmost lane.

[0214] If the number of first lanes is less than the number of second lanes, and the road topology information includes road edge information, then multiple second lanes are matched with multiple first lanes based on the road edge information; the road edge information is used to indicate whether a road edge is observed and the type of the observed road edge, which is either the left edge or the right edge.

[0215] In some embodiments, the matching module 503 is configured to:

[0216] If the road edge information indicates that the observed road edge and the type of the observed road edge are left edges, then starting from the leftmost lane, the second lanes that match each first lane are determined sequentially until all first lanes are matched.

[0217] If the road edge information indicates that the observed road edge and the type of the observed road edge are right edges, then starting from the rightmost lane, the second lanes that match each of the first lanes are determined sequentially until all the first lanes are matched.

[0218] In some embodiments, the matching module 503 is further configured to: determine that the matching has failed if the number of first lanes is greater than the number of second lanes;

[0219] The update module 504 is also used to: if it is determined that the matching fails, use a preset acquisition strategy to obtain the steering attributes corresponding to each first lane, and add the steering attributes corresponding to each first lane to the intersection information to obtain the updated road topology information.

[0220] In some embodiments, the target navigation information further includes the vehicle turning direction; the update module 504 is configured to:

[0221] Add the steering attributes corresponding to each first lane to the intersection information, and add the vehicle steering direction to the intersection information to obtain updated road topology information.

[0222] The road topology determination device 50 provided in this embodiment can execute the technical solution of the corresponding method embodiment. Its implementation principle and technical effect are similar to those of the corresponding method embodiment, and will not be described in detail here.

[0223] Figure 8 This is a schematic diagram illustrating the structure of a vehicle control device according to an exemplary embodiment, such as... Figure 8As shown, in this embodiment, the vehicle control device 60 can be housed in an electronic device, and the vehicle control device 60 includes:

[0224] The first selection module 601 is used to select at least one first drivable lane from multiple first lanes at a target intersection based on road topology information. The road topology information includes the steering attributes corresponding to the multiple first lanes and the vehicle steering direction corresponding to the target intersection. The road topology information is determined based on target navigation information corresponding to the target intersection. The target navigation information includes road information corresponding to the road where the vehicle is located, and the road information includes the number of second lanes and the steering attributes corresponding to the multiple second lanes. The number of second lanes is the total number of lanes included in the road where the vehicle is located.

[0225] The second selection module 602 is used to determine the next segment of the road after passing the target intersection from the preset driving route, and select at least one second drivable lane from multiple lanes of the next segment of the road;

[0226] The generation module 603 is used to generate at least one connection path from at least one first drivable lane to at least one second drivable lane;

[0227] The control module 604 is used to select a target connection path from at least one connection path according to a preset decision strategy, and control the vehicle to travel along the target connection path.

[0228] In some embodiments, the road topology information is determined by the road topology determination device 50 in the above embodiments. The specific implementation process is described in the above embodiments and will not be repeated here.

[0229] The vehicle control device 60 provided in this embodiment can execute the technical solution of the corresponding method embodiment. Its implementation principle and technical effect are similar to those of the corresponding method embodiment, and will not be described in detail here.

[0230] This application also provides an electronic device. Figure 9 This is a schematic diagram illustrating the structure of an electronic device according to an exemplary embodiment. For example... Figure 9 As shown, the electronic device 70 includes a processor 701 and a memory 702 communicatively connected to the processor 701.

[0231] The memory 702 stores computer execution instructions; the processor 701 executes the computer execution instructions stored in the memory 702 to implement the road topology determination method or vehicle control method provided in this application.

[0232] In this embodiment, the memory 702 and the processor 701 are connected via a bus. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be categorized as an address bus, a data bus, a control bus, etc.

[0233] The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein. The various components are interconnected via different buses and can be mounted on a common motherboard or otherwise as required.

[0234] In an exemplary embodiment, a computer-readable storage medium is also provided, which stores computer-executable instructions that, when executed by a processor, are used to implement the road topology determination method or vehicle control method provided in this application.

[0235] In an exemplary embodiment, a computer program product is also provided, including a computer program, which, when executed by a processor, is used to implement the road topology determination method or vehicle control method provided in this application.

[0236] In an exemplary embodiment, a mobile platform is also provided, including an electronic device that can serve as a client, such as a vehicle domain controller, capable of implementing the road topology determination method or vehicle control method provided in this application. The mobile platform can be a vehicle, drone, mobile phone, computer, or other device.

[0237] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0238] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0239] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0240] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0241] When an integrated unit / module is implemented in hardware, the hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic component. Unless otherwise specified, memory can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as USB flash drives, random-access memory (RAM), static random-access memory (SRAM), dynamic random-access memory (DRAM), enhanced dynamic random-access memory (EDRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), resistive random access memory (RRAM), high-bandwidth memory (HBM), and hybrid memory cube (HMC). Cube, magnetic storage, flash storage, disk, optical disk, portable hard drive or magnetic disk, and other media that can store program code.

[0242] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause an electronic device to execute all or part of the steps of the methods of the various embodiments of this application.

[0243] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0244] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0245] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for determining road topology, characterized in that, include: Obtain road topology information; the road topology information includes intersection information corresponding to the target intersection, the intersection information includes the number of first lanes, and the number of first lanes is the total number of multiple first lanes at the target intersection; Obtain target navigation information corresponding to the target intersection; the target navigation information includes road information corresponding to the road where the vehicle is located, the road information includes the number of second lanes and the turning attributes corresponding to the multiple second lanes respectively; the number of second lanes is the total number of lanes included in the road where the vehicle is located; Based on the number of the first lanes and the number of the second lanes, multiple second lanes are matched with multiple first lanes; The steering attributes corresponding to each of the first lanes are added to the intersection information to obtain updated road topology information; the steering attributes corresponding to the first lane are the steering attributes corresponding to the second lane that matches the first lane.

2. The method according to claim 1, characterized in that, The step of obtaining the target navigation information corresponding to the target intersection includes: Acquire multi-frame real-time navigation information; the multi-frame real-time navigation information includes real-time navigation information of the current time and multiple consecutive historical moments before the current time, and the real-time navigation information includes navigation timestamps and road information of the road where the vehicle is located; The real-time navigation information with the same road information in the multi-frame real-time navigation information is merged into a single reference navigation information to obtain at least one reference navigation information; the reference navigation information includes the road information, a start timestamp, and an end timestamp. From the at least one set of reference navigation information, determine the target navigation information corresponding to the target intersection.

3. The method according to claim 2, characterized in that, The intersection information also includes the intersection location, and the reference navigation information also includes the vehicle location corresponding to the start timestamp and the vehicle location corresponding to the end timestamp. Determining the target navigation information corresponding to the target intersection from the at least one reference navigation information includes: The reference navigation information whose distance between the vehicle location corresponding to the termination timestamp and the intersection location meets the preset distance conditions is determined as the target navigation information corresponding to the target intersection. The preset distance condition is that the distance is equal to or less than a preset distance.

4. The method according to claim 2 or 3, characterized in that, After acquiring multi-frame real-time navigation information, the process further includes: For each of the aforementioned real-time navigation information, the following operations are performed: Obtain vehicle location information whose sampling timestamps meet preset time conditions; the preset time conditions are that the time interval between the sampling timestamp and the navigation timestamp in the real-time navigation information is equal to or less than a preset time interval, and the vehicle location information includes the sampling timestamp and the vehicle location collected at the time represented by the sampling timestamp; The vehicle location information is added to the real-time navigation information to obtain updated real-time navigation information.

5. The method according to claim 2, characterized in that, Also includes: If no target navigation information corresponding to the target intersection is determined from the at least one reference navigation information, a preset acquisition strategy is used to obtain the steering attributes corresponding to each of the first lanes, and the steering attributes corresponding to each of the first lanes are added to the intersection information to obtain updated road topology information.

6. The method according to claim 1, characterized in that, The step of matching multiple second lanes with multiple first lanes based on the first lane number and the second lane number includes: If the number of the first lanes is equal to the number of the second lanes, then the second lanes that are matched with each of the first lanes are determined in a preset order. If the number of first lanes is less than the number of second lanes, and the road topology information includes road edge information, then based on the road edge information, multiple second lanes are matched with multiple first lanes; the road edge information is used to indicate whether a road edge is observed and the type of the observed road edge, wherein the type is left edge or right edge.

7. The method according to claim 6, characterized in that, The step of matching multiple second lanes with multiple first lanes based on the road edge information includes: If the road edge information indicates that the observed road edge and the type of the observed road edge are left edges, then starting from the leftmost lane, the second lanes that match each of the first lanes are determined sequentially until all of the first lanes are matched. If the road edge information indicates that the observed road edge and the type of the observed road edge are right edges, then starting from the rightmost lane, the second lanes that match each of the first lanes are determined sequentially until all of the first lanes are matched.

8. The method according to claim 1 or 6, characterized in that, The step of matching multiple second lanes with multiple first lanes based on the first lane number and the second lane number includes: If the number of lanes in the first lane is greater than the number of lanes in the second lane, then the matching is determined to have failed; If a matching failure is determined, a preset acquisition strategy is used to obtain the steering attributes corresponding to each of the first lanes, and the steering attributes corresponding to each of the first lanes are added to the intersection information to obtain updated road topology information.

9. The method according to claim 1, characterized in that, The target navigation information also includes the vehicle turning direction; adding the turning attributes corresponding to each of the first lanes to the intersection information to obtain updated road topology information includes: The steering attributes corresponding to each of the first lanes are added to the intersection information, and the vehicle steering direction is added to the intersection information to obtain updated road topology information.

10. The method according to claim 1, characterized in that, The method further includes: Based on road topology information, at least one first drivable lane is selected from multiple first lanes at the target intersection; the road topology information includes the steering attributes corresponding to the multiple first lanes and the vehicle turning direction corresponding to the target intersection. Determine the next section of road after passing the target intersection from the preset driving route, and select at least one second drivable lane from multiple lanes of the next section of road; Generate at least one connection path from the at least one first drivable lane to the at least one second drivable lane; According to a preset decision-making strategy, a target connection path is selected from the at least one connection path, and the vehicle is controlled to travel along the target connection path.

11. An electronic device, characterized in that, include: A processor and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the road topology determination method as described in any one of claims 1 to 10.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the road topology determination method as described in any one of claims 1 to 10.

13. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the road topology determination method as described in any one of claims 1 to 10.

14. A mobile platform, characterized in that, include: The electronic device as claimed in claim 11.

Images