Vehicle tailgating control method and device, vehicle and storage medium

Abstract

The invention relates to the technical field of intelligent driving, and discloses a vehicle tailgating control method and device, a vehicle and a storage medium, and the method comprises the steps: determining an obstacle to which each reference point in a preset length road section in front of the vehicle belongs; based on the obstacle to which each reference point belongs, using a recursive space segmentation algorithm to sequentially calculate a residual passing space corresponding to each reference point; if any reference point does not have the residual passing space, determining a borrowing lane based on the lane information, and expanding the current lane in combination with the borrowing lane to form an expanded road section; in the extended road section, the obstacle to which each reference point belongs is determined again, and the residual passing space of each reference point is calculated by adopting a recursive space segmentation algorithm; if the remaining passing spaces corresponding to all the reference points are larger than or equal to the preset passing width threshold value, the vehicle is controlled to pass by means of tailgating. By means of the method, the tailgating judgment accuracy can be improved, and then safe passing under the complex road condition is achieved.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and in particular to a vehicle lane-changing control method, device, vehicle, and storage medium. Background Technology

[0002] In autonomous driving technology, lane-changing is a crucial strategy for vehicles to cope with obstacles or abnormal road conditions. Existing methods for determining lane-changing space are typically based on static rules or simple geometric models, considering only the relationship between obstacles and lane boundaries. They fail to adequately consider the relative positions and spatial distribution of obstacles, leading to inaccurate lane-changing space assessments. For example, in the presence of multiple parallel obstacles, existing methods may misjudge the existence of available lane-changing space when in reality the area is completely occupied by obstacles, resulting in lane-changing failure or safety hazards. Summary of the Invention

[0003] In view of this, embodiments of this application provide a vehicle lane-changing control method, device, vehicle, and storage medium to achieve safe and accurate lane-changing judgment.

[0004] In a first aspect, embodiments of this application provide a vehicle lane-changing control method, including: Determine the obstacles associated with each reference point within a pre-defined road segment ahead of the vehicle; Based on the obstacles associated with each reference point, a recursive spatial segmentation algorithm is used to calculate the remaining passage space corresponding to each reference point in turn; wherein the width of the remaining passage space is greater than or equal to a preset passage width threshold. If no remaining passage space exists at any of the reference points, a borrowing lane is determined based on the lane information, and the current lane is expanded in conjunction with the borrowing lane to form an extended road segment. Within the extended road segment, the obstacles belonging to each reference point are re-determined, and the remaining passage space for each reference point is calculated using a recursive spatial partitioning algorithm; If the remaining passage space corresponding to all the reference points is greater than or equal to the preset passage width threshold, then the vehicle is controlled to pass through the other lane.

[0005] In some embodiments, before determining the obstacle belonging to each reference point within a preset length of road ahead of the vehicle, the method further includes: Obtain lane information of the current lane in which the vehicle is traveling; wherein, the lane information includes lane type and lane direction; Along the reference line of the current lane, determine reference points and lane boundary information of the lane position of each reference point at preset intervals; wherein the preset interval is less than the vehicle length.

[0006] In some embodiments, determining the obstacle belonging to each reference point within a preset length of road ahead of the vehicle includes: Based on the vehicle's position coordinates, the preset length of road segment in front of the vehicle, the lane boundary information of the lane position of each reference point in the current lane, and the coordinate information of each obstacle within the range collected by the vehicle perception system, a set of candidate obstacles is determined. Based on the coordinate information of each reference point, obstacles located within a preset range of the current reference point are selected from the candidate obstacle set and designated as the obstacles belonging to the current reference point.

[0007] In some embodiments, a recursive space partitioning algorithm is used to calculate the remaining passage space corresponding to the reference point, including: Based on the lane boundary information of the lane where the current reference point is located, the initial passage space of the current reference point is determined, and the initial passage space is used as a space in the candidate passage space set; Based on each obstacle in the obstacle set belonging to the current reference point, each candidate passage space in the candidate passage space set is segmented sequentially; After each segmentation, the passage space with a width greater than or equal to the preset passage width threshold is retained as a new candidate passage space. Based on the next obstacle, each candidate passage space in the candidate passage space set is segmented until all obstacles belonging to the current reference point are traversed, and the remaining passage space is obtained.

[0008] In some embodiments, the step of sequentially segmenting each candidate passage space in the candidate passage space set based on each obstacle in the current reference point to obtain the remaining passage space includes: Each candidate passage space in the current candidate passage space set is segmented sequentially based on the left and right boundaries of each obstacle in the obstacles to which the current reference point belongs, so as to obtain the left remaining space, obstacle area and right remaining space of the corresponding candidate passage space, and the segmented candidate passage space is deleted from the candidate passage space set; If the width of the remaining space on the left is greater than or equal to the preset passage width threshold, then the remaining space on the left is added as a new candidate passage space to the candidate passage space set; If the width of the remaining space on the right is greater than or equal to the preset passage width threshold, then the remaining space on the right is added as a new candidate passage space to the candidate passage space set. After all obstacles in the obstacles belonging to the current reference point have been traversed, the remaining candidate passage space in the candidate passage space set is taken as the remaining passage space of the current reference point.

[0009] In some embodiments, the lane type includes dashed lines and solid lines, and the lane direction includes forward and reverse directions; The process of determining the borrowing lane based on lane information and expanding the current lane in conjunction with the borrowing lane to form an extended road segment includes: If the lane line between the current lane and the adjacent lane in the forward direction is the dashed line, then the adjacent lane in the forward direction is designated as a lane that can be used for merging, and the current lane and the lane that can be used for merging are merged into a merged lane. The pre-defined length of road segment in front of the vehicles in the merged lane is used as the extended road segment.

[0010] In some embodiments, the method further includes: if the lane line between the current lane and the adjacent lane in the forward direction is the solid line or the current extended road segment is impassable, and the lane line between the current lane and the adjacent lane in the reverse direction is the dashed line, and the safe lane-borrowing conditions are met, then the adjacent lane in the reverse direction is designated as the lane that can be borrowed. The current lane is merged with the available lane to form the merged lane.

[0011] Secondly, embodiments of this application provide a vehicle lane-changing control device, comprising: The calculation module is used to determine the obstacles associated with each reference point within a preset length of road ahead of the vehicle; The calculation module is also used to calculate the remaining passage space corresponding to each reference point in turn using a recursive space segmentation algorithm based on the obstacles to which each reference point belongs; wherein the width of the remaining passage space is greater than or equal to a preset passage width threshold. An extension module is used to determine a borrowing lane based on lane information if any of the reference points does not have the remaining passage space, and to extend the current lane in combination with the borrowing lane to form an extended road segment; The calculation module is also used to redetermine the obstacles belonging to each of the reference points within the extended road segment, and to calculate the remaining passage space of each of the reference points using a recursive spatial segmentation algorithm; The determination module is used to control vehicles to pass through lanes if the remaining passage space corresponding to all the reference points is greater than or equal to the preset passage width threshold.

[0012] Thirdly, embodiments of this application provide a vehicle, the vehicle including a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the above-described vehicle lane-changing control method.

[0013] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed on a processor, implements the aforementioned vehicle lane-changing control method.

[0014] The embodiments of this application have the following beneficial effects: By determining the obstacles belonging to each reference point within a preset length of road ahead of the vehicle, and calculating the corresponding remaining passage space based on the obstacles belonging to each reference point using a recursive spatial segmentation algorithm, this application can achieve refined determination of the vehicle's passable area in complex road environments. When there is an impassable area in the current lane, by combining lane information to determine a detour lane and expanding the current lane, the remaining passage space within the expanded road segment is recalculated, thereby achieving dynamic judgment and safety assessment of the detour path. In this way, this application can improve the accuracy of detour judgment, thus enabling vehicles to pass safely and reliably in scenarios with obstacles or road restrictions. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A first flowchart of the vehicle lane-changing control method according to an embodiment of this application is shown; Figure 2 This paper shows a second flowchart of the vehicle lane-changing control method according to an embodiment of the present application; Figure 3 A third flowchart of the vehicle lane-changing control method according to an embodiment of this application is shown; Figure 4 A fourth flowchart of the vehicle lane-changing control method according to an embodiment of this application is shown; Figure 5 This illustration shows a schematic diagram of the process for obtaining the remaining passage space according to an embodiment of this application; Figure 6 The fifth flowchart of the vehicle lane-changing control method according to an embodiment of this application is shown; Figure 7 This illustration shows a vehicle continuing to travel along its current lane without using another lane, according to an embodiment of this application. Figure 8 The sixth flowchart of the vehicle lane-changing control method according to an embodiment of this application is shown; Figure 9 A schematic diagram of a forward bypass embodiment of this application is shown; Figure 10 A schematic diagram of reverse bypassing according to an embodiment of this application is shown; Figure 11 A schematic diagram of a vehicle lane-changing control device according to an embodiment of this application is shown. Detailed Implementation

[0017] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0018] The components of the embodiments of this application described and illustrated in the accompanying drawings can be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0019] In the following text, the terms "comprising," "having," and their cognates, which may be used in various embodiments of this application, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more combinations thereof. Furthermore, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0020] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be construed as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.

[0021] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0022] In this embodiment, the judgment of vehicle lane-changing behavior is based on the Frenet coordinate system to improve the modeling accuracy and computational efficiency of traffic space in complex road environments.

[0023] The Frenet coordinate system is a local coordinate system commonly used to describe the position and motion state of objects moving along a curved path. This coordinate system is particularly suitable for fields such as autonomous driving, path planning, and vehicle control, providing a more natural and intuitive description of the lateral and longitudinal positional relationships of a vehicle on a road curve.

[0024] In the Frenet coordinate system, the position of a point is represented by two parameters, s and l. s (arc length) represents the length of the curve along a reference line (such as the center line of a road) from the starting point to the current position; l (lateral offset) represents the lateral distance relative to the reference path, i.e., the offset perpendicular to the path direction.

[0025] By projecting the positional information of vehicles and surrounding obstacles into the Frenet coordinate system, the passable space of vehicles in the current lane and potential lanes that can be bypassed can be analyzed more accurately. Specifically: using the s-coordinate, the distribution range of obstacles along the road direction can be determined; using the l-coordinate, the relative positional relationship between obstacles and lane boundaries or adjacent lanes can be determined; through the joint analysis of s and l, a passable space model at each reference point can be constructed, and a recursive space segmentation algorithm can be used to determine whether the conditions for bypassing lanes are met.

[0026] Compared to traditional Cartesian coordinate systems (such as xy coordinate systems), this method has stronger spatial modeling capabilities and computational stability in handling curved roads, lanes with varying widths, and scenarios with multiple obstacles, which helps to improve the accuracy and safety of lane borrowing judgments.

[0027] In this embodiment, "borrowing a lane" refers to the behavior of a vehicle temporarily entering an adjacent lane or the edge area of ​​a lane to bypass obstacles or avoid other road users during its journey. This behavior is a path adjustment strategy that is widely used in path planning and behavioral decision-making processes in complex traffic environments.

[0028] The act of using a passageway is typically triggered in the following scenarios: 1. Obstacle Avoidance: When there are fixed or moving obstacles (such as pedestrians, parked vehicles, construction areas, etc.) in front of the vehicle, making it impossible to pass normally in the current lane, the system can determine whether the conditions for borrowing a lane are met, and enter the adjacent lane to avoid the obstacle and pass through, provided that the safety requirements are met.

[0029] 2. Avoid other road users: When encountering emergency vehicles (such as ambulances and fire trucks), vehicles changing lanes abnormally, or pedestrians crossing the road suddenly, vehicles can actively adjust their driving path by using other vehicles to avoid potential collision risks and improve driving safety.

[0030] 3. Responding to road traffic restrictions: Under certain special road conditions (such as temporary road parking, road narrowing, traffic control, etc.), vehicles may need to use adjacent lanes for short-term passage in order to maintain the continuity of traffic flow and traffic efficiency.

[0031] In this embodiment, the judgment and execution of the above-mentioned lane-borrowing behavior is based on spatial modeling and obstacle analysis technology in the Frenet coordinate system. By constructing a traffic space model of the vehicle in the current lane and the lanes that can be borrowed, and combining information such as lane type (e.g., dashed line / solid line) and lane direction (e.g., forward / reverse), it is determined whether the conditions for safe lane-borrowing are met, and a corresponding path adjustment strategy is generated accordingly.

[0032] The following describes the vehicle lane-changing control method using specific embodiments.

[0033] Figure 1 A schematic flowchart of a vehicle lane-changing control method according to an embodiment of this application is shown. Exemplarily, the vehicle lane-changing control method includes the following steps: Step S100: Determine the obstacle at each reference point within a preset length of road ahead of the vehicle.

[0034] The preset length of road segment ahead of the vehicle is defined as a segment extending forward along the lane reference line from the arc length coordinate (s-coordinate) of the vehicle's rear axle center in the Frenet coordinate system. This preset length defines the area ahead that needs to be analyzed during the lane-changing judgment process, and its value can be set according to actual application requirements. In this embodiment, the preset length can be 20 meters, but it is not limited to 20 meters; for example, it can also be 15 meters, 18 meters, 22 meters, 25 meters, etc. Typically, this preset length is not set too long to avoid introducing too many distant obstacles, thereby improving the accuracy and real-time performance of lane-changing space judgment.

[0035] In some implementations, such as Figure 2 As shown, steps S110-S120 are included before step S100: Step S110: Obtain lane information of the current lane in which the vehicle is traveling.

[0036] In this step, the lane information of the vehicle's current lane can be identified and obtained based on the high-precision map data, vehicle-mounted perception system (such as cameras and radar), or V2X communication module.

[0037] Lane information includes, but is not limited to, lane type and lane direction. Lane type refers to the lane markings between the current lane and adjacent lanes, including dashed lines and solid lines; lane direction refers to the direction of travel in the current lane, relative to the direction of vehicle travel, including forward lanes (same as the direction of vehicle travel) and reverse lanes (opposite to the direction of vehicle travel).

[0038] In this embodiment, lane type identification can be performed using lane line information predefined in a high-precision map or lane line features identified in real time by a camera; lane direction is determined by matching the vehicle's current driving direction with the lane's defined direction in the high-precision map.

[0039] The purpose of obtaining lane information is to provide basic data support for subsequent lane-borrowing decisions, such as determining whether it is permissible to borrow from an adjacent lane, and whether the adjacent lane is a solid line lane, thereby improving the accuracy and safety of lane-borrowing strategies.

[0040] Step S120: Determine the reference points and lane boundary information of the lane position of each reference point at preset intervals along the reference line of the current lane.

[0041] The reference line is the centerline of the current lane, i.e., the path reference line defined in the Frenet coordinate system. Reference points are determined starting from the s-coordinate of the vehicle's rear axle center in the Frenet coordinate system, and are sequentially set at preset intervals along the reference line, forming a sampling sequence along the road direction. Each reference point corresponds to an s-coordinate, used to identify its longitudinal position on the reference line. Simultaneously, the system acquires the lane boundary information of the lane where the reference point is located, including the lateral offset coordinates (l-values) of the left and right lane boundaries in the Frenet coordinate system, thus providing spatial boundary basis for subsequent obstacle selection and traffic space modeling.

[0042] The preset interval is less than the vehicle length and can be any value between 0.5 and 2.0 meters (e.g., 1 meter). By setting a sampling interval less than the vehicle length, it is possible to ensure that each key location on the path ahead is finely modeled during vehicle movement, thereby improving the accuracy and safety of lane-changing judgments.

[0043] In some implementations, such as Figure 3 As shown, step S100 specifically includes steps S130-S140: Step S130: Based on the vehicle's position coordinates, the preset length of road segment in front of the vehicle, the lane boundary information of each reference point in the current lane, and the coordinate information of each obstacle within the range collected by the vehicle perception system, determine the candidate obstacle set.

[0044] The vehicle's position coordinates are the arc length coordinate (s) and lateral offset coordinate (l) of the rear axle center in the Frenet coordinate system. The coordinate information of each obstacle within the field of view is based on the vehicle's current position. The coordinate information of each obstacle within the range of the perception system is the obstacle coordinate information obtained by the perception system (such as cameras, radar, lidar, etc.) and transformed in the Frenet coordinate system, based on the vehicle's current position. The transformed obstacle coordinate information includes: the minimum arc length coordinate (min_s) of the obstacle along the road direction (longitudinal) in the Frenet coordinate system, representing the obstacle's starting position on the road reference line, i.e., the obstacle's... The obstacles are defined as follows: the longitudinal position of the end closest to the vehicle's starting point, the maximum arc length coordinate of the obstacle along the road direction in the Frenet coordinate system (max_s, representing the end position of the obstacle on the road reference line, i.e., the longitudinal position of the end furthest from the vehicle's starting point), the minimum lateral offset of the obstacle in the Frenet coordinate system (min_l, representing the leftmost position of the obstacle in the lateral direction (perpendicular to the road direction), i.e., the distance of the left boundary of the obstacle relative to the reference line), and the maximum lateral offset of the obstacle in the Frenet coordinate system (max_l, representing the rightmost position of the obstacle in the lateral direction, i.e., the distance of the right boundary of the obstacle relative to the reference line).

[0045] When determining the candidate obstacle set based on the above information, obstacles with min_s or max_s greater than the arc length coordinate of the rear axle center of the vehicle and min_s or max_s less than the arc length coordinate of the rear axle center of the vehicle plus a preset length can be selected as the initial candidate obstacle set. Then, obstacles with min_l and max_l within the lane boundary of the lane position of the corresponding reference point (i.e., the range from the left lateral offset coordinate to the right lateral offset coordinate of the lane position of the current reference point) are selected from the initial obstacle set, and the selected obstacles are used as the candidate obstacle set.

[0046] Step S140: Based on the coordinate information of each reference point, select obstacles located within a preset range of the current reference point from the candidate obstacle set and use them as the obstacles belonging to the current reference point.

[0047] The preset range is a longitudinal range centered on the reference point in the Frenet coordinate system, extending forward and backward without exceeding the length of the vehicle. After determining this preset range, obstacles whose min_s or max_s fall within this range are selected from the candidate obstacle set determined in step S130. These are the obstacles belonging to the current reference point.

[0048] This screening process ensures that each reference point interacts with obstacles only in its spatial vicinity, thereby improving the computational efficiency and accuracy of subsequent recursive spatial segmentation algorithms, while avoiding the introduction of distant and irrelevant obstacle information that could interfere with the detour decision.

[0049] Step S200: Based on the obstacles to which each reference point belongs, a recursive space partitioning algorithm is used to calculate the remaining passage space corresponding to each reference point in turn.

[0050] The remaining passage space is the lateral space range that meets the passage requirements in the current lane or extended lane, and its width is greater than or equal to the preset passage width threshold.

[0051] In this embodiment, a recursive spatial segmentation algorithm can be used to segment the lateral passage space at each reference point layer by layer, remove the area occupied by obstacles, and retain the space that meets the minimum passage width, thereby determining whether the current reference point has the conditions for passage.

[0052] In some implementations, such as Figure 4 As shown, a recursive space partitioning algorithm is used to calculate the remaining passage space corresponding to each reference point in sequence, specifically including sub-steps S210-S230: Step S210: Based on the lane boundary information of the lane where the current reference point is located, determine the initial passage space of the current reference point, and use the initial passage space as a space in the candidate passage space set.

[0053] The initial passable space refers to the lateral passable range of the lane where the current reference point is located in the Frenet coordinate system, that is, the interval between the lateral offset corresponding to the left boundary of the lane and the lateral offset corresponding to the right boundary. Figure 5 As shown, the initial passage space is the space between the two black dots. This interval can be represented as a closed interval [left_start, right_end], which represents the maximum lateral space that a vehicle can theoretically pass in the current lane at this reference point. For example, if the left boundary of the lane where the current reference point is located has a lateral offset of -1.8m in the Frenet coordinate system (relative to the reference line) and the right boundary has a lateral offset of +1.8m, then the initial passage space is [-1.8m, +1.8m].

[0054] In this embodiment, the left and right boundary information of the lane can be determined based on high-precision map data or lane line information obtained in real time by the perception system.

[0055] The aforementioned initial passage space [left_start, right_end] can be used as the initial input for the subsequent recursive space segmentation algorithm. It is added to the candidate passage space set. By introducing obstacles one by one and segmenting the space, the passable area is gradually reduced, and finally the remaining passage space that meets the passage width requirement is obtained.

[0056] Step S220: Based on each obstacle in the obstacle set belonging to the current reference point, divide each candidate passage space in the candidate passage space set in turn.

[0057] Step S230: After each segmentation, retain the passage space with a width greater than or equal to the preset passage width threshold as a new candidate passage space, and continue to segment each candidate passage space in the candidate passage space set based on the next obstacle until all obstacles belonging to the current reference point have been traversed, and then the remaining passage space is obtained.

[0058] Exemplarily, steps S220 to S230 are used to segment each candidate space in the candidate passage space set layer by layer based on the obstacles corresponding to the current reference point, in order to determine the remaining passage space at the reference point. In this process, this embodiment sequentially traverses the obstacles belonging to the current reference point, and for each obstacle, segments each candidate space in the candidate passage space set based on its lateral boundary information (i.e., min_l and max_l) in the Frenet coordinate system. After each segmentation, the system retains only passage spaces with a width greater than or equal to a preset passage width threshold (such as 1.2 or 1.5 times the vehicle width) as new candidate spaces to ensure that only areas with actual passage capacity are retained. This process continues until all obstacles corresponding to the current reference point have been processed, and the space retained in the final candidate passage space set is the remaining passage space corresponding to the reference point. If the space is not empty, it indicates that the reference point has passable conditions; if it is empty, it indicates that the reference point is impassable, and a detour judgment process needs to be triggered.

[0059] In some implementations, such as Figure 6 As shown, based on each obstacle in the current reference point's obstacle set, each candidate passage space in the candidate passage space set is segmented to obtain the remaining passage space, including steps S231-S234: Step S231: Based on the left and right boundaries of each obstacle in the current reference point, each candidate passage space in the current candidate passage space set is segmented in turn to obtain the left remaining space, obstacle area and right remaining space of the corresponding candidate passage space, and the segmented candidate passage space is deleted from the candidate passage space set.

[0060] Step S232: If the width of the remaining space on the left is greater than or equal to the preset passage width threshold, then the remaining space on the left is added to the candidate passage space set as a new candidate passage space.

[0061] Step S233: If the width of the remaining space on the right is greater than or equal to the preset passage width threshold, then the remaining space on the right is added to the candidate passage space set as a new candidate passage space.

[0062] Step S234: After traversing all obstacles in the obstacles belonging to the current reference point, the remaining candidate passage space in the candidate passage space set is taken as the remaining passage space of the current reference point.

[0063] Exemplary, when executing the recursive spatial segmentation algorithm, if the obstacles belonging to the current reference point include multiple obstacles, these obstacles can be numbered first so that the space can be segmented sequentially according to the numbering order. Alternatively, they can be left unnumbered and the space segmentation can be performed directly according to the order of the obstacles in the lateral (l-direction) relative to the vehicle from left to right or from right to left. In other words, the processing order of multiple obstacles can be set according to actual needs, and this embodiment does not impose specific limitations on this.

[0064] It should be noted that, in this embodiment, "left" and "right" refer to directions defined based on the vehicle's frontal orientation. Specifically, the left side of the vehicle refers to the left side of the vehicle, and the right side refers to the right side of the vehicle. This directional definition is used to determine the lateral position (i.e., l-value) of the obstacle in the Frenet coordinate system and its relationship with the candidate passage space.

[0065] In step S210 above, the initial passage space has been determined based on the boundary information of the lane where the current reference point is located, and it has been added to the candidate passage space set as the first candidate passage space. Therefore, the initial state of the candidate passage space set is not empty, and subsequent spatial segmentation processing of obstacles can continue.

[0066] The spatial segmentation process is illustrated with an example: We begin processing from the first obstacle. Based on the obstacle's lateral boundary information—its minimum and maximum lateral offsets (min_l and max_l) in the Frenet coordinate system—each candidate space in the current candidate passable space set is segmented. Since the initial candidate passable space set contains only one initial passable space [left_start, right_end], segmentation is only based on this space. Specifically, the remaining space on the left is formed by the left boundary (left_start) of the current candidate space to the left boundary (min_l) of the obstacle; the obstacle region is formed by the left boundary (min_l) of the obstacle to its right boundary (max_l); and the remaining space on the right is formed by the right boundary (max_l) of the obstacle to the right boundary (right_end) of the current candidate space.

[0067] After completing the above segmentation, the original candidate space [left_start, right_end] is removed from the candidate passage space set, and the left or right remaining space that meets the preset passage width threshold is added back to the candidate passage space set as a new candidate space for further segmentation of obstacles.

[0068] If multiple candidate spaces exist in the current candidate space set (e.g., multiple subspaces were generated in the previous segmentation), each candidate space needs to be segmented with the current obstacle separately to ensure that each candidate space is fully processed. This process is repeated until all obstacles belonging to the current reference point have been traversed. The candidate spaces retained in the final candidate passage space set are the remaining passage spaces corresponding to that reference point. This embodiment, through the above recursive segmentation mechanism, can effectively eliminate areas occupied by obstacles and retain lateral space that meets the passage width requirements, thereby providing an accurate spatial judgment basis for subsequent detour decisions.

[0069] For ease of understanding, the following will use... Figure 5 The process of obtaining the remaining passage space is illustrated by example. Figure 5 The two black dots at either end of the dashed line represent the left and right boundaries of the lane, and the gray box between the two black dots represents the obstacle to which the current reference point belongs. The "mark" indicates that the space does not meet the preset passage width threshold (i.e., vehicles cannot pass through), while the space within the gray dashed box indicates that it meets the preset passage width threshold. Figure 5 of (a) Figure 5 (b) and Figure 5 In (c), the candidate passage space is divided according to the obstacles in a left-to-right order.

[0070] like Figure 5As shown in (a), there are two obstacles. The first obstacle divides the initial candidate passage space to obtain the left remaining space L1 and the right remaining space R1. The left remaining space L1 does not meet the passage conditions, while the right remaining space R1 does. Therefore, the right remaining space R1 is added to the candidate passage space set as a candidate passage space. Then, the newly added right remaining space R1 is divided by the second obstacle to obtain the left remaining space L2 and the right remaining space R2 corresponding to the current candidate remaining space. According to the judgment, the left remaining space L2 is greater than or equal to the preset passage width threshold, so it meets the passage conditions. However, the right remaining space R2 does not meet the passage conditions. Therefore, the right remaining space R1 is deleted from the candidate passage space set, and the right remaining space R2 is added to the candidate passage space set. Since there are no subsequent obstacles, only the right remaining space R2 remains in the candidate passage space set. R2 is the remaining passage space corresponding to the current reference point.

[0071] Similarly, Figure 5 (b) sequentially passes through two obstacles to divide the candidate passage space. Finally, only the remaining space R3 on the right side remains in the candidate passage space set. R3 is the remaining passage space corresponding to the current reference point. Figure 5 If (c) divides the candidate passage space by passing through two obstacles in sequence, and none of the divided spaces meet the passage conditions, then there is no remaining passage space for the reference point.

[0072] It is understood that the recursive space segmentation algorithm provided in this embodiment can eliminate impassable areas layer by layer according to the distribution of obstacles in the horizontal (l direction) and retain the space that meets the passage conditions, thereby realizing the refined modeling and dynamic judgment of the passage space, and thus improving the accuracy and safety of the passage strategy.

[0073] In step S300, if there is no remaining passage space at any reference point, a borrowing lane is determined based on the lane information, and the current lane is expanded in combination with the borrowing lane to form an extended road segment.

[0074] Exemplarily, during the execution of the recursive spatial segmentation algorithm to obtain the remaining passage space corresponding to each reference point, if any reference point is determined to have no remaining passage space satisfying the preset passage width threshold, it indicates that the current lane is not passable at that location. In this case, there is no need to continue spatial segmentation calculations for subsequent reference points; the current lane can be immediately determined to be impassable, and a lane-borrowing process can be triggered. Further, if all reference points have remaining passage space satisfying the passage conditions, the current lane is considered to have normal passage capacity, and vehicles can continue to travel along the current lane without borrowing a lane. Figure 7As shown. When it is determined that the current lane is not passable, the system will determine whether there is an adjacent lane that can be borrowed based on the obtained lane information (such as lane line type, lane direction, etc.), and determine the target lane to borrow. The current lane will be merged with the target lane to form an extended road segment.

[0075] In some implementations, such as Figure 8 As shown, the process involves determining a borrowing lane based on lane information and expanding the current lane using the borrowing lane to form an extended road segment, including steps S310-S320: Step S310: If the lane line between the current lane and the adjacent lane in the forward direction is a dashed line, then the adjacent lane in the forward direction is taken as a lane that can be borrowed, and the current lane and the lane that can be borrowed are merged into a merged lane.

[0076] Step S320: The pre-set length of road segment in front of the vehicle in the merged lane is used as the extended road segment.

[0077] As an example, when determining whether a lane-changing maneuver is permissible, if high-precision map data or an onboard perception system identifies the lane line between the current lane and the adjacent lane in the forward direction as a dashed line (e.g., ... Figure 9 If the current lane is adjacent to the adjacent lane in the forward direction (as shown in the image), it indicates that there is a possibility of lane sharing between the current lane and the adjacent lane in the forward direction. In this case, the system treats the adjacent lane as a lane that can be shared, and based on lane boundary information and reference line information, logically merges the current lane and the adjacent lane to form an expanded merged lane. Then, a preset length of road segment ahead of the vehicle is determined within the merged lane and used as the expanded road segment.

[0078] Furthermore, if the lane line between the current lane and the adjacent lane in the forward direction is a solid line or the current extended section is impassable, and the lane line between the current lane and the adjacent lane in the reverse direction is a dashed line, and the conditions for safe lane borrowing are met, then the adjacent lane in the reverse direction will be used as a lane that can be borrowed; the current lane and the lane that can be borrowed will be merged as a merged lane.

[0079] It is understandable that if the lane line between the current lane and the adjacent lane in the forward direction is a solid line, or if the current extended road segment still does not meet the conditions for passage, the system will further determine whether the lane line between the current lane and the adjacent lane in the reverse direction is a dashed line, and combine this with traffic environment information to determine whether the conditions for safe lane-changing are met. If the lane line between the current lane and the adjacent lane in the reverse direction is a dashed line, and the conditions for safe lane-changing are met (e.g., no high-speed vehicles approaching from the opposite lane, sufficient space for lane-changing, no pedestrians crossing), then the system will designate the adjacent lane in the reverse direction as a lane that can be used (e.g., ...). Figure 10 As shown in the diagram, based on lane boundary information and reference line information, the current lane and the adjacent lane in the opposite direction are logically merged to form an expanded merged lane. Figure 9 and Figure 10 The dark gray rectangular frame represents an obstacle.

[0080] In step S400, within the extended road segment, the obstacles belonging to each reference point are re-determined, and the remaining passage space for each reference point is calculated using a recursive spatial partitioning algorithm.

[0081] In step S500, if the remaining passage space corresponding to all reference points is greater than or equal to the preset passage width threshold, then control the vehicle to pass through the other lane.

[0082] Exemplary, in the above process, after determining that the current lane is impassable and triggering the lane-borrowing procedure, an extended road segment is constructed based on the merging result of the current lane and the available lanes (such as forward or reverse dashed lanes). This extended road segment includes the merged space of the current lane and the lane-borrowing lane. Using the original reference point determination method, the position of each reference point and its corresponding lane boundary information are redefined within this extended space. Subsequently, based on the updated lane boundary information, the obstacles belonging to each reference point are re-determined. After updating the obstacles belonging to each reference point, the same recursive spatial segmentation algorithm as used for determining the current lane is employed to sequentially segment the space of the obstacles belonging to each reference point and calculate its corresponding remaining passage space. The determination of reference points, the determination of the obstacles belonging to each reference point, and the acquisition of the remaining passage space are all the same as described above, and therefore will not be repeated here. This embodiment, by reconstructing the obstacle set and passage space model of each reference point in the extended road segment, enables the system to accurately assess the actual passage capacity under lane-borrowing conditions, thereby providing a basis for subsequent lane-borrowing decisions.

[0083] After determining the remaining passage space at each reference point in the extended road segment, it is sequentially checked whether the remaining passage space at each reference point meets the preset passage width threshold. If the remaining passage space at all reference points is greater than or equal to the threshold, it means that the vehicle can safely pass through the entire preset length of the road segment on the current bypass path. In this case, it is determined that the current bypass passage condition is met, and a corresponding path planning instruction is generated to guide the vehicle into the bypass path. If the remaining passage space at any reference point is less than the threshold, it means that there are still impassable areas on the bypass path. In this case, it will be decided whether to try other bypass paths (such as trying other adjacent lanes) or trigger other obstacle avoidance strategies (such as deceleration, stopping and waiting) based on the actual situation.

[0084] Furthermore, after the detour route is determined, traffic rules and sensory information can be combined to determine whether there is a potential conflict with other traffic participants, such as oncoming vehicles or pedestrians, during the detour process. If a potential conflict is identified, the conflict risk level is further assessed, and the detour is carried out only if safety conditions are met; otherwise, the detour decision is delayed or canceled.

[0085] This embodiment maps vehicles, road environments, and surrounding obstacles into the Frenet coordinate system. Combining lane boundary information with the spatial distribution of obstacles, a recursive spatial segmentation algorithm is used to segment and filter the lateral traffic space at each reference point layer by layer, retaining spatial areas that meet a preset traffic width threshold, thereby accurately identifying passable paths. When the current lane cannot meet the traffic conditions, the system dynamically expands the lane boundary based on lane line type (dashed / solid line) and lane direction (forward / backward), reconstructs the expanded road segment, and performs secondary spatial modeling and lane-borrowing judgment to ensure the safety and feasibility of lane-borrowing behavior. Furthermore, by setting a reference point sampling interval smaller than the vehicle length and combining it with an obstacle filtering strategy, the real-time performance of lane-borrowing judgment and the accuracy of spatial modeling are further improved, making it particularly suitable for complex traffic scenarios such as curved roads, lane width variations, and multiple obstacles.

[0086] Figure 11 A schematic diagram of a vehicle lane-changing control device according to an embodiment of this application is shown. Exemplarily, the lane-changing device includes: The calculation module 100 is used to determine the obstacles associated with each reference point within a preset length of road ahead of the vehicle.

[0087] The calculation module 100 is also used to calculate the remaining passage space corresponding to each reference point in turn using a recursive space segmentation algorithm based on the obstacle to which each reference point belongs; wherein the width of the remaining passage space is greater than or equal to a preset passage width threshold.

[0088] The extension module 200 is used to determine a borrowing lane based on lane information if there is no remaining passage space at any reference point, and to extend the current lane in combination with the borrowing lane to form an extended road segment.

[0089] The calculation module 100 is also used to redetermine the obstacles belonging to each reference point within the extended road segment, and to calculate the remaining passage space of each reference point using a recursive spatial segmentation algorithm.

[0090] The determination module 300 is used to determine whether to use the bypass if the remaining passage space corresponding to all reference points is greater than or equal to the preset passage width threshold.

[0091] It is understood that the device in this embodiment corresponds to the vehicle lane-changing control method in the above embodiments, and the options in the above embodiments are also applicable to this embodiment, so they will not be described again here.

[0092] This application also provides a vehicle, exemplary in that the vehicle includes a processor and a memory, wherein the memory stores a computer program, and the processor, by running the computer program, causes the vehicle to perform the functions of the various modules in the above-described vehicle lane-changing control method or the above-described vehicle lane-changing control device.

[0093] The processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Network Processor (NP), Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0094] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory is used to store computer programs, and the processor can execute the computer programs accordingly after receiving execution instructions.

[0095] This application also provides a computer-readable storage medium for storing the computer program used in the aforementioned vehicle. For example, the computer-readable storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0096] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that, in alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0097] In addition, the functional modules or units in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0098] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a smartphone, personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0099] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A vehicle lane-changing control method, characterized in that, include: Determine the obstacles associated with each reference point within a pre-defined road segment ahead of the vehicle; Based on the obstacles associated with each reference point, a recursive spatial segmentation algorithm is used to calculate the remaining passage space corresponding to each reference point in turn; wherein the width of the remaining passage space is greater than or equal to a preset passage width threshold. If no remaining passage space exists at any of the reference points, a borrowing lane is determined based on the lane information, and the current lane is expanded in conjunction with the borrowing lane to form an extended road segment. Within the extended road segment, the obstacles belonging to each reference point are re-determined, and the remaining passage space for each reference point is calculated using a recursive spatial partitioning algorithm; If the remaining passage space corresponding to all the reference points is greater than or equal to the preset passage width threshold, then the vehicle is controlled to pass through the other lane.

2. The vehicle lane-changing control method according to claim 1, characterized in that, Before determining the obstacle at each reference point within a preset length of road ahead of the vehicle, the method further includes: Obtain lane information of the current lane in which the vehicle is traveling; wherein, the lane information includes lane type and lane direction; Along the reference line of the current lane, determine reference points and lane boundary information of the lane position of each reference point at preset intervals; wherein the preset interval is less than the length of the vehicle.

3. The vehicle lane-changing control method according to claim 2, characterized in that, The process of determining the obstacles associated with each reference point within a preset length of road ahead of the vehicle includes: Based on the vehicle's position coordinates, the preset length of road segment in front of the vehicle, the lane boundary information of the lane position of each reference point in the current lane, and the coordinate information of each obstacle within the range collected by the vehicle perception system, a set of candidate obstacles is determined. Based on the coordinate information of each reference point, obstacles located within a preset range of the current reference point are selected from the candidate obstacle set and designated as the obstacles belonging to the current reference point.

4. The vehicle lane-changing control method according to claim 1, characterized in that, The remaining passage space corresponding to the reference point is calculated using a recursive space partitioning algorithm, including: Based on the lane boundary information of the lane where the current reference point is located, the initial passage space of the current reference point is determined, and the initial passage space is used as a space in the candidate passage space set; Based on each obstacle in the obstacle set belonging to the current reference point, each candidate passage space in the candidate passage space set is segmented sequentially; After each segmentation, the passage space with a width greater than or equal to the preset passage width threshold is retained as a new candidate passage space. Based on the next obstacle, each candidate passage space in the candidate passage space set is segmented until all obstacles belonging to the current reference point are traversed, and the remaining passage space is obtained.

5. The vehicle lane-changing control method according to claim 4, characterized in that, The step of sequentially segmenting each candidate passage space in the candidate passage space set based on each obstacle in the current reference point to obtain the remaining passage space includes: Each candidate passage space in the current candidate passage space set is segmented sequentially based on the left and right boundaries of each obstacle in the obstacles to which the current reference point belongs, so as to obtain the left remaining space, obstacle area and right remaining space of the corresponding candidate passage space, and the segmented candidate passage space is deleted from the candidate passage space set; If the width of the remaining space on the left is greater than or equal to the preset passage width threshold, then the remaining space on the left is added as a new candidate passage space to the candidate passage space set; If the width of the remaining space on the right is greater than or equal to the preset passage width threshold, then the remaining space on the right is added as a new candidate passage space to the candidate passage space set. After all obstacles in the obstacles belonging to the current reference point have been traversed, the remaining candidate passage space in the candidate passage space set is taken as the remaining passage space of the current reference point.

6. The vehicle lane-changing control method according to claim 2, characterized in that, The lane types include dashed lines and solid lines, and the lane directions include forward and reverse directions; The process of determining the borrowing lane based on lane information and expanding the current lane in conjunction with the borrowing lane to form an extended road segment includes: If the lane line between the current lane and the adjacent lane in the forward direction is the dashed line, then the adjacent lane in the forward direction is designated as a lane that can be used for merging, and the current lane and the lane that can be used for merging are merged into a merged lane. The pre-defined length of road segment in front of the vehicles in the merged lane is used as the extended road segment.

7. The vehicle lane-changing control method according to claim 6, characterized in that, Also includes: If the lane line between the current lane and the adjacent lane in the forward direction is a solid line or the current extended road section is impassable, and the lane line between the current lane and the adjacent lane in the reverse direction is a dashed line, and the safe lane-borrowing conditions are met, then the adjacent lane in the reverse direction will be designated as the lane that can be borrowed. The current lane is merged with the available lane to form the merged lane.

8. A vehicle lane-changing control device, characterized in that, include: The calculation module is used to determine the obstacles associated with each reference point within a preset length of road ahead of the vehicle; The calculation module is also used to calculate the remaining passage space corresponding to each reference point in turn using a recursive space segmentation algorithm based on the obstacles to which each reference point belongs; wherein the width of the remaining passage space is greater than or equal to a preset passage width threshold. An extension module is used to determine a borrowing lane based on lane information if any of the reference points does not have the remaining passage space, and to extend the current lane in combination with the borrowing lane to form an extended road segment; The calculation module is also used to redetermine the obstacles belonging to each of the reference points within the extended road segment, and to calculate the remaining passage space of each of the reference points using a recursive spatial segmentation algorithm; The determination module is used to control vehicles to pass through lanes if the remaining passage space corresponding to all the reference points is greater than or equal to the preset passage width threshold.

9. A vehicle, characterized in that, The vehicle includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the vehicle lane-changing control method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed on a processor, implements the vehicle lane-changing control method according to any one of claims 1-7.

Images