Intersection traffic methods, devices, electronic equipment, media, products and vehicles

Abstract

This application relates to the field of assisted driving technology, specifically to a method, device, electronic device, medium, product, and vehicle for intersection passage. The method includes: corresponding to a scenario where a target vehicle is determined to enter an intersection, determining the current location of the target vehicle, at least one exit lane corresponding to the target vehicle's current direction of travel, and at least one virtual lane connecting the target vehicle's current location to each exit lane based on acquired first perception data; determining at least one drivable lane from the at least one virtual lane based on acquired second perception data; and, corresponding to the at least one drivable lane including multiple drivable lanes, determining the target drivable lane from the multiple drivable lanes based on acquired third perception data. This achieves stable and accurate selection of a safe lane for passage through the intersection using a purely visual approach.

Description

Technical Field

[0001] This application relates to the field of assisted driving technology, specifically to a method, device, electronic equipment, medium, product, and vehicle for crossing intersections. Background Technology

[0002] Current driver assistance technologies typically rely on navigation information provided by high-definition maps (HD maps) and standard-definition maps (SD maps) to solve lane planning problems (e.g., which lane a vehicle can choose to go straight after entering an intersection). This allows the vehicle to continuously determine the lane it needs to exit the intersection before or immediately upon entering, thus providing lane-level technical support for safer path planning in driver assistance systems.

[0003] However, HD Maps are costly to update and maintain frequently, and their updates lag in dynamic and complex scenarios, such as sudden increases in traffic flow or unexpected road repairs, making it difficult to provide accurate support for driver assistance systems. Furthermore, HD Maps may experience map gaps (e.g., missing map information for remote areas) or be unavailable (e.g., map service interruptions or disconnections of in-vehicle terminal wireless connections).

[0004] However, SD Map cannot provide detailed parameters such as lane information, and its real-time update speed for dynamic environments (such as traffic flow) is slow, which means that using SD Map cannot provide accurate lane planning for driver assistance systems.

[0005] If the aforementioned limited HD Map and SD Map methods are not used, lane planning assistance systems can rely solely on a pure vision-based approach to perceive the environmental features around the target vehicle. For example, photoelectric sensors and cameras can be used to acquire these features. However, this pure vision-based approach may be limited by the shooting range, shooting accuracy, and environmental interference (such as obstacle occlusion), making it difficult to achieve stable and accurate lane planning.

[0006] Therefore, how to stably and accurately select safe lanes using only a visual solution without relying on maps (including HD Maps and SD Maps) has become a problem that needs to be solved. Summary of the Invention

[0007] This application provides a method, device, electronic device, medium, product, and vehicle for traffic crossing at intersections. The method is used to solve the problem of how to stably and accurately select a safe lane using only a visual solution without relying on maps (including HD Maps and SD Maps).

[0008] In a first aspect, embodiments of this application propose an intersection passage method applied to an in-vehicle terminal. The method includes: corresponding to a scenario where a target vehicle is determined to enter an intersection, determining the current location of the target vehicle, at least one exit lane corresponding to the current direction of travel of the target vehicle, and at least one virtual lane for connecting the current location of the target vehicle with each exit lane based on acquired first perception data; determining at least one drivable lane from the at least one virtual lane based on acquired second perception data; and corresponding to the at least one drivable lane including multiple drivable lanes, determining a target drivable lane from the multiple drivable lanes based on acquired third perception data.

[0009] It is understandable that after a target vehicle determines the scenario of entering the intersection, it can collect and identify perception data in the environment through a vision solution, and then construct at least one virtual lane based on the perception data. From all virtual lanes, a lane that can be entered is pre-screened. Corresponding to at least one lane that can be entered includes multiple lanes, the target lane is selected from the multiple lanes based on third perception data. This achieves the stable and accurate selection of a lane that can safely pass through the intersection through a pure vision solution.

[0010] In some possible implementations of the first aspect above, determining a target entry lane from multiple drivable lanes based on the acquired third perception data includes: performing a drivability assessment on each drivable lane in the multiple drivable lanes based on the acquired third perception data to obtain a first score for each drivable lane; and determining the drivable lane in each drivable lane whose first score satisfies a first score condition as the target entry lane.

[0011] It is understandable that by using third-party perception data to conduct multi-dimensional accessibility assessments of each accessible lane to select the better or optimal lane, the accessibility of the lane can be evaluated from multiple dimensions, enabling a stable and accurate selection of lanes that can safely pass through intersections using a purely visual approach.

[0012] In some possible implementations of the first aspect above, corresponding to at least one accessible lane includes one accessible lane, and one accessible lane is used as the target accessible lane.

[0013] It is understandable that, when there is only one lane available for entry, the onboard terminal can directly target the vehicle and enter the lane.

[0014] In some possible implementations of the first aspect mentioned above, the first perception data includes intersection recognition features and a drivable centerline for the vehicle. The intersection recognition features include an intersection entrance line, an intersection exit line, an internal guide line, and an exit lane line. Furthermore, based on the acquired first perception data, the current location of the target vehicle, at least one exit lane corresponding to the target vehicle's current travel direction, and at least one virtual lane connecting the target vehicle's current location with each exit lane are determined. This includes determining the target vehicle's current relative position to the intersection based on the intersection entrance line, intersection exit line, internal guide line, and exit lane line, thus determining the target vehicle's current location; and using the target vehicle's current location as the origin, determining the intersection entrance / exit lane line. The system calculates the coordinates of the intersection line, the intersection exit line, the intersection guide lines, the exit lane lines, and the trajectory of the vehicle's drivable centerline. Based on the coordinates of the exit lane lines, it determines the centerline of the exit lanes. Based on the trajectory of the vehicle's drivable centerline, the coordinates of the intersection entrance line, the coordinates of the intersection exit line, and the geometric relationship with the exit lane centerline, it determines the unique entrance point and the set of exit points. Based on the unique entrance point and the set of exit points, it determines the intersection entrance angle and the intersection exit angle corresponding to each exit lane. Based on the intersection entrance angle, the intersection exit angle corresponding to each exit lane, the position parameters of the unique entrance point, and the position parameters of the exit point set, it determines at least one virtual lane.

[0015] It is understandable that the vehicle terminal can construct at least one curved trajectory within the intersection based on the intersection entrance angle corresponding to each exit lane, the intersection exit angle corresponding to each exit lane, the location parameters of the unique entrance point, and the location parameters of the set of exit points. This at least one curved trajectory is the virtual lane.

[0016] In some embodiments, the vehicle terminal can construct a cubic Bézier curve based on the intersection entrance angle corresponding to each exit lane, the intersection exit angle corresponding to each exit lane, the location parameters of the unique entrance point, and the location parameters of the set of exit points.

[0017] In some possible implementations of the first aspect described above, the second sensing data includes intersection recognition features and a drivable area. The intersection recognition features include an intersection entrance line, an intersection exit line, an internal guide line, and an exit lane line. Furthermore, determining at least one drivable lane from at least one virtual lane based on the acquired second sensing data includes: determining a roadside line based on the intersection recognition features; identifying a first type of non-drivable lane from at least one virtual lane based on the roadside line, wherein the first type of non-drivable lane includes auxiliary roads, oncoming lanes, and non-motorized vehicle lanes; identifying a second type of non-drivable lane from at least one virtual lane based on the lane centerline, drivable area, intersection entrance line, intersection exit line, internal guide line, and exit lane line of at least one virtual lane, wherein the second type of non-drivable lane includes non-drivable lanes intersecting with the roadside, guide line, and lane line; and removing the first and second type of non-drivable lanes from at least one virtual lane to obtain at least one drivable lane.

[0018] Thus, the vehicle terminal can pre-select the first and second categories of inaccessible lanes from all virtual lanes to obtain at least one accessible lane, thereby reducing the amount of data processing required for subsequent lane accessibility evaluation.

[0019] In some possible implementations of the first aspect mentioned above, the third perception data includes intersection recognition features, traffic flow features, drivable areas, and vehicle drivable centerlines. The intersection recognition features include intersection entrance lines, intersection exit lines, intersection guide lines, and exit lane lines. Furthermore, based on the acquired third perception data, the drivability of each of the multiple drivable lanes is assessed to obtain a first score for each drivable lane. This includes: if it is determined that the target vehicle has not entered the intersection, determining the heading angle of the target vehicle entering each drivable lane based on the target vehicle's current location and the lane entrance positions of each drivable lane, and determining a first dimension score for each drivable lane based on the heading angle. The first dimension score is used to characterize each drivable lane. The system assesses the smoothness of steering angles; it acquires coupling trend parameters between the guide lines within the intersection and each accessible lane, and determines a second-dimensional score for each accessible lane based on these parameters. The second-dimensional score characterizes the degree of coupling between each accessible lane and the guide lines within the intersection. It also determines obstacle characteristics on each accessible lane based on traffic flow characteristics, and determines collision parameters between the target vehicle and obstacles on each accessible lane based on these obstacle characteristics and the target vehicle. A third-dimensional score is then determined based on these collision parameters, characterizing the probability of the target vehicle colliding with an obstacle while driving in each accessible lane. Finally, a first score is determined based on the first, second, and third-dimensional scores.

[0020] It is understandable that, if it is determined that the target vehicle has not entered the intersection, the vehicle terminal can estimate the static risks during the process of the target vehicle entering the intersection after entering each of the available lanes.

[0021] If it is determined that the target vehicle has not entered the intersection, the vehicle terminal can focus on assessing structural risk, such as the smoothness of the steering angle of each accessible lane (corresponding to the calculation of steering smoothness based on the minishot algorithm for the intersection entrance and exit angles), the coupling trend parameters with the guide lines within the intersection (corresponding to lane-level guidance), and the probability of the target vehicle colliding with obstacles while driving in each accessible lane (corresponding to collision risk).

[0022] In some embodiments, the vehicle terminal can determine the first score of each lane that can be entered based on the first dimension score, the second dimension score, and the third dimension score. For example, the vehicle terminal can sum the first dimension score, the second dimension score, and the third dimension score to obtain the first score, or the vehicle terminal can weighted sum the first dimension score, the second dimension score, and the third dimension score to obtain the first score.

[0023] The weighting values ​​for the weighted summation can be freely set according to the user's needs, and no specific restrictions are imposed here.

[0024] In some possible implementations of the first aspect mentioned above, the third perception data includes intersection recognition features, traffic flow features, drivable areas, and drivable centerlines for vehicles. The intersection recognition features include intersection entrance lines, intersection exit lines, internal guide lines, and exit lane lines. Furthermore, based on the acquired third perception data, the drivability of each of the multiple drivable lanes is assessed to obtain a first score for each drivable lane. This includes: when it is determined that the target vehicle has entered the intersection, acquiring historical target entry lanes, determining the first lateral offset between each drivable lane and the historical target entry lane, and determining a fourth-dimensional score for each drivable lane based on the first lateral offset. The fourth-dimensional score characterizes the lateral offset of the target vehicle from the historical target entry lane into each drivable lane. Obstacle features on each drivable lane are determined based on traffic flow features. Collision parameters between the target vehicle and obstacles in each drivable lane are determined based on the obstacle features and the target vehicle. Finally, the collision parameters are used to determine the... The fifth dimension score for each accessible lane represents the probability of a target vehicle colliding with an obstacle while traveling in each accessible lane. The sixth dimension score for each accessible lane is determined based on the second lateral offset of the vehicle's centerline relative to each accessible lane, representing the degree of vehicle sway while traveling in each lane. A first steering parameter from the intersection entrance to the target vehicle's current location and a second steering parameter from the target vehicle's current location to the entrance of each accessible lane are determined based on the target vehicle's current location, the intersection entrance location, and the entrance locations of each accessible lane. A seventh dimension score for each accessible lane is determined based on the first steering parameter and the second steering parameter, representing the target vehicle's right-of-way while traveling in each accessible lane. Finally, a first score is determined based on the fourth, fifth, sixth, and seventh dimension scores.

[0025] Once it is determined that the target vehicle has entered the intersection, the on-board terminal can focus on assessing dynamic risk, such as the steering comfort of each drivable lane and the historical target drivable lane (e.g., lateral acceleration can be determined based on the lateral offset between the two, corresponding to latitude), the probability of collision between the target vehicle and obstacles in each drivable lane (corresponding to collision risk), the degree of vehicle body sway of the target vehicle while driving in each drivable lane (corresponding to vehicle body shake), and the right of way of the target vehicle while driving in each drivable lane.

[0026] In some embodiments, the vehicle terminal can determine the first score for each lane that can be entered based on the fourth-dimensional score, the fifth-dimensional score, the sixth-dimensional score, and the seventh-dimensional score. For example, the vehicle terminal can sum the fourth-dimensional score, the fifth-dimensional score, the sixth-dimensional score, and the seventh-dimensional score to obtain the first score, or the vehicle terminal can obtain the first score by weighted summation of the fourth-dimensional score, the fifth-dimensional score, the sixth-dimensional score, and the seventh-dimensional score.

[0027] In some possible implementations of the first aspect above, determining the drivable lane in each drivable lane that satisfies the first score condition as the target drivable lane includes: determining the drivable lane with the highest first score in each drivable lane as the target drivable lane.

[0028] In some possible implementations of the first aspect mentioned above, determining the drivable lane in each drivable lane that satisfies the first score condition as the target drivable lane further includes: if the right-of-way of the target vehicle in each drivable lane conflicts with the right-of-way of surrounding vehicles, and if the vehicle body volume of the surrounding vehicles is larger than that of the target vehicle, then reducing the first score of the drivable lane in which the surrounding vehicles are about to travel; and determining the drivable lane in each drivable lane with the highest first score as the target drivable lane.

[0029] It is understandable that after selecting the lane with the highest score, the vehicle terminal can further perform a secondary safety check on the size of surrounding vehicles and the size of the vehicle itself, and select the optimal lane as the target lane to ensure the driving safety of the target vehicle as much as possible.

[0030] In some possible implementations of the first aspect described above, the method further includes: capturing one or more images of the target vehicle's surroundings, obtaining the target vehicle's driving parameters corresponding to each of the multiple images of the target vehicle's surroundings, and determining first perception data, second perception data, and third perception data from the one or more images of the target vehicle's surroundings based on the target vehicle's driving parameters corresponding to each image of the target vehicle's surroundings.

[0031] Secondly, embodiments of this application also provide an intersection passage device applied to a target vehicle. The device includes a virtual lane construction module, a lane pre-screening module, and a data processing module. The virtual lane construction module is used to determine, based on acquired first perception data, the current location of the target vehicle, at least one exit lane corresponding to the current travel direction of the target vehicle, and at least one virtual lane connecting the current location of the target vehicle with each exit lane, corresponding to a scenario where the target vehicle is determined to enter the intersection. The lane pre-screening module is used to determine at least one drivable lane from the at least one virtual lane based on acquired second perception data. The data processing module is used to determine the target drivable lane from the multiple drivable lanes based on acquired third perception data, corresponding to the at least one drivable lane including multiple drivable lanes.

[0032] Thirdly, embodiments of this application also provide an electronic device, including: one or more processors; one or more memories; the one or more memories storing one or more programs, which, when executed by one or more processors, cause the electronic device to perform the intersection passage method proposed in the first aspect and various implementations of the first aspect.

[0033] Fourthly, embodiments of this application also provide a computer-readable medium storing instructions that, when executed on a machine, cause the machine to perform the intersection passage method proposed in the first aspect and various implementations thereof.

[0034] Fifthly, embodiments of this application also provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the intersection passage method proposed in the first aspect and various implementations of the first aspect.

[0035] Sixthly, embodiments of this application also provide a vehicle, which includes the electronic equipment proposed in the third aspect above.

[0036] It is understood that the beneficial effects of the second to sixth aspects mentioned above can be referred to the first aspect and the beneficial effects of various implementations of the first aspect, which will not be elaborated here.

[0037] The technical solution provided in this application has at least the following beneficial effects:

[0038] Once the target vehicle determines the scenario of entering the intersection, it can collect and identify perception data in the environment through a vision solution, and then construct at least one virtual lane based on the perception data. From all virtual lanes, it can pre-screen the lanes that can be entered, and when multiple lanes that can be entered are determined, it can select the better or optimal lane based on the perception data, so as to achieve stable and accurate selection of lanes that can safely pass through the intersection through a pure vision solution. Attached Figure Description

[0039] Figure 1 This paper presents a schematic diagram of a traffic scenario at an intersection, as proposed in an embodiment of this application.

[0040] Figure 2 A schematic diagram of a framework structure for lane selection based on perception data, as proposed in an embodiment of this application, is shown.

[0041] Figure 3 A schematic flowchart of a method for traffic passage at an intersection according to an embodiment of this application is shown;

[0042] Figure 4 A schematic diagram of a scenario showing an intersection entrance angle and an intersection exit angle according to some embodiments of this application is shown;

[0043] Figure 5 A schematic diagram of a virtual lane scene at an intersection is shown according to some embodiments of this application;

[0044] Figure 6 A schematic diagram illustrating a strategy selection for an evaluation dimension according to some embodiments of this application is shown;

[0045] Figure 7 The illustration shows a scene of yielding according to some embodiments of this application;

[0046] Figure 8 A schematic diagram of a scenario where the vehicle does not yield to another vehicle is shown in some embodiments of this application;

[0047] Figure 9 The illustration shows a scenario of yielding to a bus according to some embodiments of this application;

[0048] Figure 10 A complete flowchart of the secondary security verification proposed according to some embodiments of this application is shown;

[0049] Figure 11 This illustration shows a schematic diagram of another target entry lane selection principle proposed according to some embodiments of this application;

[0050] Figure 12 A schematic diagram of the frame structure of an intersection passage device 1200 according to an embodiment of this application is shown;

[0051] Figure 13 This diagram illustrates the structure of an electronic device according to some embodiments of the present application. Detailed Implementation

[0052] This application provides a method, apparatus, electronic device, medium, product, and vehicle for traffic passage at an intersection, wherein the vehicle refers to the description of the target vehicle in this application. To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0053] It is understood that, in order to solve the problem of how to stably and accurately select a safe lane using only a purely visual solution without relying on maps (including HD Maps and SD Maps), this application proposes an intersection passage method applied to an in-vehicle terminal. The method includes: corresponding to the scenario of determining that a target vehicle is entering the intersection, determining the current location of the target vehicle, at least one exit lane corresponding to the current direction of travel of the target vehicle, and at least one virtual lane for connecting the current location of the target vehicle with each exit lane based on the acquired first perception data; determining at least one drivable lane from the at least one virtual lane based on the acquired second perception data; corresponding to the at least one drivable lane including multiple drivable lanes, determining the target drivable lane from the multiple drivable lanes based on the acquired third perception data.

[0054] The technical solution provided in this application has at least the following beneficial effects:

[0055] Once the target vehicle determines the scenario of entering the intersection, it can collect and identify perception data in the environment through a vision solution, and then construct at least one virtual lane based on the perception data. From all virtual lanes, it can pre-screen the lanes that can be entered, and when multiple lanes that can be entered are determined, it can select the better or optimal lane based on the perception data, so as to achieve stable and accurate selection of lanes that can safely pass through the intersection through a pure vision solution.

[0056] It is understood that the perception data can be dynamic or static information about the target vehicle’s surroundings identified using a vision scheme. For example, the perception data includes intersection recognition features, traffic flow features, drivable area (free space), and ego-driveline. The intersection recognition features include intersection entrance lines, intersection exit lines, intersection guide lines, and exit lane lines.

[0057] Figure 1 A schematic diagram of a traffic scenario at an intersection, as proposed in an embodiment of this application, is shown.

[0058] Figure 2 A schematic diagram of a framework structure for lane selection based on perception data, as proposed in an embodiment of this application, is shown.

[0059] refer to Figure 1The target vehicle is about to enter intersection P1, which includes the intersection entrance line L1, the intersection exit line L2, and the intersection guide line L3. The target vehicle can determine multiple virtual lanes that can pass through intersection P1 based on perception data, such as the first virtual lane A1, the second virtual lane A2, and the third virtual lane A3.

[0060] Therefore, for reference Figure 2 When the vehicle-mounted terminal identifies an intersection based on the aforementioned perception data, it collects various static and discrete elements within the intersection to identify the intersection entrance line, exit line, guide lines, and exit lane lines. Furthermore, it identifies traffic flow characteristics, the drivable area (free space), and the vehicle's drivable centerline. Based on one or more features from the aforementioned perception data, it selects the target lane. Static elements can include lane lines, curb lines, guide lines, zebra crossings, double yellow lines, stop lines, and / or water-filled traffic cones, among other static traffic signs. Dynamic elements can include dynamic traffic participants, such as surrounding vehicles, pedestrians, and / or cyclists.

[0061] Traffic flow characteristics are a set of macroscopic parameters and statistical laws that describe the collective movement of groups of vehicles (rather than individual vehicles) on a road. They can describe the overall operational status of traffic flow on a road over time and space. For example, volume / flow indicates the number of vehicles passing through a cross section per unit time; speed indicates the space-mean speed of all vehicles in a certain road segment; and density indicates the number of vehicles per unit length of road.

[0062] The drivable area (free space) refers to the "safe and unobstructed usable road surface area" around the target vehicle. Without considering traffic rules (such as traffic lights and lane markings), it is a "flat, unobstructed road surface" within the target vehicle's current height range that allows free passage. It excludes all dynamic and static obstacles (surrounding vehicles, pedestrians, curbs, unknown objects, etc.) and provides a "safe and feasible" boundary for path planning. In some embodiments, the boundary of the drivable area (e.g., polar coordinate point arrays or polygons) can be generated in real time using LiDAR / vision / fusion algorithms.

[0063] The ego-driveline is a local path centerline calculated within the current drivable area in the vehicle's coordinate system. Its starting point can be the rear axle center of the vehicle, extending forward a predetermined distance, ranging from 30 to 100 meters, influenced by vehicle speed and sensor field of view. In some embodiments, the perceived image (e.g., an image of the target vehicle's surrounding environment captured by a camera, or a point cloud image of the target vehicle's surrounding environment scanned by radar) can be sliced ​​within the drivable area at a predetermined longitudinal resolution (e.g., 0.5m). The midpoint of each cross-section is calculated to obtain a discrete center point sequence. This discrete center point sequence is then smoothed (e.g., using splines, Bezier equations, piecewise cubic polynomials, etc.) to output a continuous, differentiable, and curvature-continuous centerline. This centerline is then transformed to the origin of the vehicle's coordinate system to obtain the ego-driveline. In some embodiments, the update frequency of this ego-driveline can be 10 Hz.

[0064] The intersection entrance line is the section of the lane center line or edge line that precedes the stop line (such as a solid white line or a yellow stop line) before entering the intersection.

[0065] The exit line at an intersection is the lane line after the target vehicle leaves the intersection area.

[0066] Guide lines within an intersection are white dashed or dotted lines placed inside the intersection and spanning the intersection area. They are used to indicate "how target vehicles should turn / go straight within the intersection, and must not cross the lines or go against traffic."

[0067] Exit lane lines are lane dividing lines (such as white dashed or solid lines) that separate lanes after exiting the intersection.

[0068] For example, the aforementioned perception data can be obtained based on perception information. This perception information can be environmental information within a preset range around the vehicle body collected by the vehicle-mounted terminal from perception components integrated within the target vehicle (such as multi-view cameras, LiDAR, millimeter-wave radar, or infrared / thermal imaging night vision). Examples include one or more frames of road images around the vehicle taken by a camera, or three-dimensional radar point cloud data around the vehicle body collected by various radars. Furthermore, the vehicle-mounted terminal can identify static and dynamic road features based on the characteristics carried in the environmental information. These static road features include, but are not limited to, lane edges, curb positions, road width, lane width, road surface width, traffic markings, and static obstacles. The dynamic road features include, but are not limited to, surrounding vehicles encountered during driving (i.e., dynamic obstacles).

[0069] In some embodiments of this application, the above method further includes: capturing one or more images of the target vehicle's surroundings, obtaining the target vehicle's driving parameters corresponding to each of the multiple images of the target vehicle's surroundings, and determining first perception data, second perception data, and third perception data from one or more images of the target vehicle's surroundings based on the target vehicle's driving parameters corresponding to each image of the target vehicle's surroundings.

[0070] The following is combined Figure 3 The specific implementation process of the intersection passage method proposed in the embodiments of this application will be described in detail.

[0071] Figure 3 A schematic flowchart of an intersection passage method according to an embodiment of this application is shown.

[0072] It is understandable that the execution subject of this method can be the vehicle terminal, which will not be elaborated here.

[0073] Specifically, the method includes the following steps:

[0074] S301, corresponding to the scenario of determining that a target vehicle is entering the intersection, determines the current location of the target vehicle, at least one exit lane corresponding to the current direction of travel of the target vehicle, and at least one virtual lane for connecting the current location of the target vehicle with each exit lane based on the acquired first perception data.

[0075] For example, the vehicle-mounted terminal can determine the scenario of entering an intersection based on perception data. For instance, based on the acquired first perception data, it can identify one or more of the intersection entrance line, intersection exit line, and exit lane line, thus determining that the target vehicle has entered the intersection. Furthermore, the vehicle-mounted terminal can determine the location of the target vehicle (e.g., the target vehicle is located at the point about to enter the intersection, or the target vehicle is already inside the intersection) and at least one exit lane corresponding to the target vehicle's current direction of travel (e.g., the vehicle-mounted terminal can determine at least one exit lane based on the exit lane line) based on the acquired first perception data. It can also construct at least one virtual lane connecting the target vehicle's current location with each exit lane based on the first perception data.

[0076] In some embodiments of this application, the aforementioned first perception data may include intersection recognition features and a drivable centerline for the vehicle. The intersection recognition features include an intersection entrance line, an intersection exit line, an inner guide line, and an exit lane line. Determining the current location of the target vehicle, at least one exit lane corresponding to the target vehicle's current travel direction, and at least one virtual lane connecting the target vehicle's current location to each exit lane based on the acquired first perception data includes: determining the target vehicle's current relative position to the intersection based on the intersection entrance line, intersection exit line, inner guide line, and exit lane line, thereby determining the target vehicle's current location; and determining the coordinates of the intersection entrance line, intersection exit line, inner guide line, and exit lane line, and the drivable centerline for the vehicle, using the target vehicle's current location as the origin. The trajectory of the driving centerline is determined; the centerline of the exit lane is determined based on the coordinates of the exit lane lines; based on the trajectory of the driving centerline of the vehicle, the coordinates of the intersection entrance line, the coordinates of the intersection exit line, and the geometric relationship with the centerline of the exit lane, the unique entrance point and the set of exit points of the intersection are determined; based on the unique entrance point and the set of exit points, the intersection entrance angle and the intersection exit angle corresponding to each exit lane are determined; based on the intersection entrance angle, the intersection exit angle corresponding to each exit lane, the position parameters of the unique entrance point, and the position parameters of the set of exit points, at least one virtual lane is determined.

[0077] It is understandable that the vehicle-mounted terminal can determine the current relative position of a target vehicle with respect to the intersection based on the intersection entrance line, intersection exit line, intersection guide line, and exit lane line. For example, it can determine whether the target vehicle is outside the intersection, or, if the target vehicle is inside the intersection, where it is located within the intersection. In some embodiments, the vehicle-mounted terminal does not need to directly locate the specific latitude and longitude coordinates of the target vehicle. Instead, it constructs a vehicle coordinate system with the target vehicle itself as the origin, and determines the coordinates of the intersection entrance line, intersection exit line, intersection guide line, exit lane line, and the trajectory of the vehicle's drivable centerline based on the first perception data, with the target vehicle's current location as the origin.

[0078] Furthermore, the vehicle terminal can determine at least one exit lane centerline of the intersection based on the coordinates of the exit lane line and the current travel direction of the target vehicle. Based on the trajectory of the vehicle's drivable centerline, the coordinates of the intersection entrance line, the coordinates of the intersection exit line, and the geometric relationship with the exit lane centerline, it can determine the unique entrance point of the intersection and the set of multiple exit points of the intersection through which the target vehicle enters the intersection. The unique entrance point can be the intersection of the vehicle's drivable centerline and the intersection entrance line, and the set of multiple exit points of the intersection can be the intersection of the vehicle's drivable centerline and the exit lane centerline, and the intersection exit line.

[0079] In some embodiments, the target vehicle’s current direction of travel may include going straight, turning left, or turning right.

[0080] In other embodiments, the target vehicle's current travel direction includes only going straight.

[0081] Next, the vehicle terminal can connect to each exit point in the set of unique entry and exit points, thereby determining the intersection entry angle corresponding to each exit lane and the intersection exit angle corresponding to each exit lane.

[0082] Figure 4 The illustration shows a scenario of an intersection entrance angle and an intersection exit angle according to some embodiments of this application.

[0083] In this system, the target vehicle is located in the starting lane outside the intersection. The intersection of this starting lane and the intersection entrance line is the unique entrance point Q1. The exit point corresponding to the exit lane is marked as the first exit point Q2. Connecting the unique entrance point Q1 and the first exit point Q2 yields the vector Q1→Q2, which represents the turning angle of the target vehicle within the intersection. The on-board terminal determines the intersection entrance angle (also known as the "entry angle") based on the angle between the target vehicle's current heading and the vector Q1→Q2. Based on the centerline of the exit lane, it determines the target heading extending from the first exit point Q2. The angle between this target heading and the vector Q1→Q2 then determines the intersection exit angle (also known as the "exit angle").

[0084] Furthermore, the vehicle-mounted terminal can construct at least one curved trajectory within the intersection based on the intersection entrance angle corresponding to each exit lane, the intersection exit angle corresponding to each exit lane, the location parameters of the unique entrance point, and the location parameters of the set of exit points. This at least one curved trajectory is the virtual lane. In some embodiments, the vehicle-mounted terminal can construct a cubic Bézier curve based on the intersection entrance angle corresponding to each exit lane, the intersection exit angle corresponding to each exit lane, the location parameters of the unique entrance point, and the location parameters of the set of exit points.

[0085] Figure 5 A schematic diagram of a virtual lane scene at an intersection is shown according to some embodiments of this application.

[0086] refer to Figure 5 The vehicle-mounted terminal can construct the first virtual lane A4, the second virtual lane A5, the third virtual lane A6, and the fourth virtual lane A7 within the intersection.

[0087] S302, based on the acquired second perception data, determine at least one accessible lane from at least one virtual lane.

[0088] It is understandable that the second perception data can be the perception data acquired by the target vehicle, which can be used to filter out drivable lanes from at least one virtual lane. For example, it can determine the guide lines within the intersection, or the lane type of the exit lane indicated by the exit lane line. Thus, the on-board terminal can pre-select inaccessible lanes from all virtual lanes, reducing the amount of data processing required for subsequent lane accessibility evaluation.

[0089] For example, the second perception data includes intersection identification features and drivable areas. The intersection identification features include intersection entrance lines, intersection exit lines, intersection guide lines, and exit lane lines.

[0090] In some embodiments of this application, determining at least one drivable lane from at least one virtual lane based on acquired second perception data includes: determining a roadside line based on intersection recognition features; identifying a first type of non-drivable lane from at least one virtual lane based on the roadside line, wherein the first type of non-drivable lane includes auxiliary roads, oncoming lanes, and non-motorized vehicle lanes; identifying a second type of non-drivable lane from at least one virtual lane based on the lane centerline, drivable area, intersection entrance line, intersection exit line, intersection guide line, and exit lane line of at least one virtual lane, wherein the second type of non-drivable lane includes non-drivable lanes intersecting with the roadside, guide line, and lane line; and removing the first type of non-drivable lane and the second type of non-drivable lane from at least one virtual lane to obtain at least one drivable lane.

[0091] It is understandable that the vehicle terminal can identify non-drivable roadside lines (such as road shoulders, median strip barriers, and other non-drivable obstacles) based on intersection recognition features. Then, the vehicle terminal can identify auxiliary roads, reverse lanes, and non-motorized vehicle lanes from at least one virtual lane based on the roadside lines, so as to prevent the target vehicle from entering the aforementioned infeasible exit lanes.

[0092] Furthermore, the vehicle-mounted terminal can also eliminate inaccessible lanes that intersect with curbs, guide lines, and lane lines to prevent target vehicles from colliding with inaccessible curbs and to prevent target vehicles from violating traffic regulations by driving over guide lines or lane lines, thus ensuring the safe driving of target vehicles within intersections.

[0093] For example, continue to refer to Figure 1 Of the first virtual lane A1, the second virtual lane A2, and the third virtual lane A3, only the second virtual lane A2 does not intersect with the guide line. Therefore, the first virtual lane A1 and the third virtual lane A3, which are not allowed to drive, are eliminated, and the second virtual lane A2 is selected as the allowed lane.

[0094] Thus, the vehicle terminal can pre-select the first and second categories of inaccessible lanes from all virtual lanes to obtain at least one accessible lane, thereby reducing the amount of data processing required for subsequent lane accessibility evaluation.

[0095] S303, corresponding to at least one accessible lane including multiple accessible lanes, determines the target access lane from the multiple accessible lanes based on the acquired third perception data.

[0096] It is understandable that the vehicle terminal can evaluate multiple drivable lanes based on the acquired third-party perception data and select the target drivable lane.

[0097] For example, corresponding to at least one accessible lane includes one accessible lane, and one accessible lane is used as the target accessible lane.

[0098] It is understandable that when there is only one accessible lane, the vehicle terminal can use that accessible lane as the target lane to enter.

[0099] In other embodiments, determining a target entry lane from multiple drivable lanes based on the acquired third perception data includes: assessing the drivability of each drivable lane from the multiple drivable lanes based on the acquired third perception data to obtain a first score for each drivable lane; and determining the drivable lane whose first score satisfies a first score condition as the target entry lane.

[0100] It is understandable that the vehicle terminal can conduct multi-dimensional accessibility assessments of each of the multiple accessible lanes based on the acquired third-party perception data. For example, it can comprehensively assess the collision risk, driving comfort, right-of-way, and guidance accuracy of the target vehicle in each accessible lane, so as to select the lane that meets the score conditions as the target access lane.

[0101] In some embodiments of this application, when it is determined that the target vehicle has not entered the intersection, the vehicle terminal can estimate the static risks of the target vehicle entering the intersection after entering each accessible lane; while when it is determined that the target vehicle has entered the intersection, the vehicle terminal can estimate the dynamic risks from surrounding vehicles during the process of the target vehicle entering each accessible lane.

[0102] The following section, in conjunction with relevant accompanying diagrams, provides a detailed explanation of how to assess static risks when the target vehicle has not entered the intersection.

[0103] Figure 6 A schematic diagram illustrating a strategy selection for an evaluation dimension proposed according to some embodiments of this application is shown.

[0104] refer to Figure 6If it is determined that the target vehicle has not entered the intersection, the vehicle terminal can focus on assessing structural risk, such as the smoothness of the steering angle of each accessible lane (corresponding to the steering smoothness calculated based on the minishot algorithm for the intersection entrance and exit angles), the coupling trend parameters with the guide lines within the intersection (corresponding to lane-level guidance), and the probability of the target vehicle colliding with obstacles while driving in each accessible lane (corresponding to collision risk).

[0105] For example, the aforementioned third-sensing data includes intersection recognition features, traffic flow features, drivable areas, and drivable centerlines for vehicles. Intersection recognition features include intersection entrance lines, intersection exit lines, guide lines within the intersection, and exit lane lines. Furthermore, based on the acquired third-sensing data, the drivability of each of the multiple drivable lanes is assessed to obtain a first score for each drivable lane. This includes: if it is determined that the target vehicle has not entered the intersection, determining the heading angle of the target vehicle entering each drivable lane based on the target vehicle's current location and the lane entrance positions of each drivable lane, and determining a first-dimensional score for each drivable lane based on the heading angle. The first-dimensional score characterizes the steering angle smoothness of each drivable lane. The coupling trend parameters between the guide lines within the intersection and each drivable lane are obtained, and the first-dimensional score for each drivable lane is determined based on the coupling trend parameters. The second dimension score of the drivable lanes is used to characterize the coupling degree between each drivable lane and the guide lines within the intersection; the obstacle characteristics on each drivable lane are determined based on traffic flow characteristics; the collision parameters between the target vehicle and the obstacles on each drivable lane are determined based on the obstacle characteristics on each drivable lane and the target vehicle; the third dimension score of each drivable lane is determined based on the collision parameters; the third dimension score is used to characterize the probability of the target vehicle colliding with the obstacle while driving in each drivable lane; the first score is determined based on the first dimension score, the second dimension score, and the third dimension score.

[0106] For example, the vehicle-mounted terminal can determine the heading angle of the target vehicle entering each accessible lane based on the target vehicle's current location and the lane entrance positions of each accessible lane (corresponding to the intersection exit lines), and determine the intersection entrance angle and intersection exit angle based on the heading angle. The heading angle can include the angle between the target vehicle's current heading and vector Q1→Q2 (corresponding to the intersection entrance angle) and the angle between the target heading and vector Q1→Q2 (corresponding to the intersection exit angle). The specific implementation process for determining the heading angle of each accessible lane can be found above. Figure 4 The specific implementation process for obtaining the intersection entrance angle and intersection exit angle as illustrated in the example will not be elaborated here.

[0107] Furthermore, the onboard terminal can evaluate the smoothness of the steering angle of each accessible lane by summing the absolute value of the difference between the intersection entrance angle and the intersection exit angle with the absolute value of the sum of the intersection entrance angle and the intersection exit angle. Therefore, the more similar the intersection entrance angle and the intersection exit angle are, the more uniform and smooth the steering process of the target vehicle in that accessible lane will be, resulting in a higher first-dimensional score.

[0108] For example, continue referring to the example above. Figure 5 Since the opposing lanes are all drivable lanes, the third virtual lane A6 can be selected as the optimal lane (corresponding to the target drivable lane) based on the smoothness of the steering angle.

[0109] For example, the vehicle terminal can evaluate the coupling trend parameters between each drivable lane and the guide line in the intersection by the mean and / or variance of the distance values ​​of multiple trajectory points corresponding to the lane lines of each drivable lane and the guide lines in the intersection on the lateral y-axis in the vehicle coordinate system.

[0110] Therefore, the smaller the mean and / or the smaller the variance, the higher the coupling between the lane and the guide line within the intersection, and the more closely it follows the guide line within the intersection, the higher the score of the second dimension.

[0111] For example, the vehicle-mounted terminal can assess the probability of a target vehicle colliding with an obstacle in each accessible lane by summing the intersection area of ​​the vehicle's outline and the obstacle's outline over a first preset time period. For instance, the first preset time period can be 3 seconds, in which case the vehicle-mounted terminal can extrapolate the intersection area of ​​the vehicle's outline and the obstacle's outline over the next three seconds to assess the collision risk. For example, the vehicle-mounted terminal can calculate the intersection area of ​​the vehicle's outline and the obstacle's outline sequentially in three consecutive collision probability predictions at times t1, 2, and 3, and then sum the three intersection areas to determine the dynamic collision parameters for colliding with obstacles in each accessible lane. The smaller the collision parameters, the higher the third-dimensional score.

[0112] Next, the vehicle terminal can determine the first score for each lane that can be entered based on the first dimension score, the second dimension score, and the third dimension score. For example, the vehicle terminal can sum the first dimension score, the second dimension score, and the third dimension score to obtain the first score, or the vehicle terminal can weighted sum the first dimension score, the second dimension score, and the third dimension score to obtain the first score.

[0113] The weighting values ​​for the weighted summation can be freely set according to the user's needs, and no specific restrictions are imposed here.

[0114] Continue to refer to Figure 6Once it is determined that the target vehicle has entered the intersection, the on-board terminal can focus on assessing dynamic risk, such as the steering comfort of each drivable lane and the historical target drivable lane (e.g., lateral acceleration can be determined based on the lateral offset between the two, corresponding to latitude), the probability of collision between the target vehicle and obstacles in each drivable lane (corresponding to collision risk), the degree of vehicle body sway of the target vehicle while driving in each drivable lane (corresponding to vehicle body shake), and the right of way of the target vehicle while driving in each drivable lane.

[0115] For example, the third perception data includes intersection recognition features, traffic flow features, drivable areas, and drivable centerlines for vehicles. Intersection recognition features include intersection entrance lines, intersection exit lines, internal guide lines, and exit lane lines. Furthermore, based on the acquired third perception data, the drivability of each of the multiple drivable lanes is assessed to obtain a first score for each drivable lane. This includes: if it is determined that a target vehicle has already entered the intersection, acquiring historical target entry lanes, determining the first lateral offset between each drivable lane and the historical target entry lane, and determining a fourth-dimensional score for each drivable lane based on the first lateral offset. The fourth-dimensional score characterizes the lateral offset of the target vehicle from the historical target entry lane into each drivable lane. Obstacle features on each drivable lane are determined based on traffic flow features. Collision parameters between the target vehicle and obstacles on each drivable lane are determined based on the obstacle features and the target vehicle. The collision parameters are then used to determine the drivable lane's... The fifth dimension score characterizes the probability of the target vehicle colliding with obstacles while traveling in each accessible lane. A sixth dimension score is determined for each accessible lane based on the second lateral offset of the vehicle's centerline relative to each lane, characterizing the degree of vehicle sway while traveling in each lane. A first steering parameter is determined from the target vehicle's current location, the intersection's entrance location, and the entrance locations of each accessible lane, along with a second steering parameter from the target vehicle's current location to the entrance locations of each lane. A seventh dimension score is determined based on the first steering parameter and the second steering parameter of each lane, characterizing the target vehicle's right-of-way while traveling in each lane. Finally, a first score is determined based on the fourth, fifth, sixth, and seventh dimension scores.

[0116] It is understandable that the historical target entry lane can be the target entry lane selected by past target vehicles, so that the on-board terminal can determine the entry lane with higher lane changing comfort based on the historical target entry lane and the newly determined entry lanes.

[0117] For example, the vehicle terminal can determine the first lateral offset between each drivable lane and the historical target drivable lane based on the maximum value of the absolute value of the difference between the trajectory points of each drivable lane and the trajectory points of the historical target drivable lane in the lateral direction. The smaller the first lateral offset, the larger the fourth dimension score.

[0118] For example, the vehicle terminal can use the sum of the intersection area of ​​the vehicle outline and the obstacle outline over a second preset time period as a collision parameter to evaluate the probability of the target vehicle colliding with the obstacle while driving in each accessible lane. The smaller the collision parameter obtained, the larger the fifth dimension score.

[0119] In some embodiments, the second preset duration can be 3 seconds. The process of accumulating the intersection area of ​​the vehicle contour and the obstacle contour can be referred to the specific implementation process of accumulating the intersection area in the third dimension scoring example above, and will not be repeated here.

[0120] For example, the in-vehicle terminal can assess the degree of vehicle sway in each drivable lane based on the second lateral offset of the vehicle's trajectory points (corresponding to the vehicle's drivable centerline) relative to reference points on each drivable lane within a third preset time period on the y-axis of the vehicle's coordinate system. For instance, the third preset time period can be 1.5 seconds. The smaller the second lateral offset, the lower the degree of sway of the target vehicle in the drivable lane, the higher the comfort level of the target vehicle's occupants, and the higher the obtained sixth-dimensional score.

[0121] For example, the vehicle terminal can assess the right-of-way of a target vehicle in each accessible lane using a first steering parameter from the entrance position of the intersection to the target vehicle's current position and a second steering parameter from the target vehicle's current position to the entrance positions of each accessible lane. The first steering parameter can be the absolute value of the difference between the intersection entrance angle and the vehicle's steering angle, and the sum of the absolute values ​​of the intersection entrance angle and the vehicle's steering angle. The second steering parameter can be the absolute value of the difference between the vehicle's steering angle and the intersection exit angle, and the sum of the absolute values ​​of the vehicle's steering angle and the intersection exit angle.

[0122] Therefore, the sum of the first and second steering parameters can be used as a dynamic assessment of right-of-way. The first steering parameter characterizes the smoothness of steering from the intersection entrance to the target vehicle's current location, while the second steering parameter characterizes the smoothness of steering from the target vehicle's current location to the intersection exit. Thus, the smaller the sum of the first and second steering parameters, the higher the right-of-way of the drivable lane relative to other drivable lanes, and the higher the resulting seventh-dimensional score.

[0123] Next, the vehicle terminal can determine the first score for each lane that can be entered based on the fourth, fifth, sixth, and seventh dimension scores. For example, the vehicle terminal can sum the fourth, fifth, sixth, and seventh dimension scores to obtain the first score, or the vehicle terminal can weighted sum the fourth, fifth, sixth, and seventh dimension scores to obtain the first score.

[0124] The weighting values ​​for the weighted summation can be freely set according to the user's needs, and no specific restrictions are imposed here.

[0125] In some embodiments, the first score condition may include: determining the drivable lane with a first score greater than or equal to a preset score as the target drivable lane; or, determining the drivable lane with the highest first score among the drivable lanes as the target drivable lane.

[0126] It's understandable that the lane with the highest score is the optimal lane.

[0127] In some embodiments, the above-mentioned preset score can be freely set according to user needs, and there are no restrictions here.

[0128] In other embodiments of this application, determining the drivable lane in each drivable lane that satisfies the first score condition as the target drivable lane further includes: if the right-of-way of the target vehicle in each drivable lane conflicts with the right-of-way of surrounding vehicles, and if the vehicle body volume of the surrounding vehicles is larger than that of the target vehicle, then reducing the first score of the drivable lane in which the surrounding vehicles are about to travel; and determining the drivable lane with the highest first score among all drivable lanes as the target drivable lane.

[0129] It is understandable that if there is a right-of-way conflict between your vehicle and surrounding vehicles, you need to confirm whether to change lanes to give way to the surrounding vehicles.

[0130] The following section provides a detailed explanation of the yielding schemes for the aforementioned right-of-way conflicts, using examples from the accompanying diagrams.

[0131] Figure 7A schematic diagram of a yielding scenario according to some embodiments of this application is shown.

[0132] Figure 8 A schematic diagram of a scenario where the vehicle does not yield to another vehicle is shown in some embodiments of this application.

[0133] Figure 9 The illustration shows a scenario of yielding to a bus according to some embodiments of this application.

[0134] Figure 10 A complete flowchart of the secondary security verification proposed according to some embodiments of this application is shown.

[0135] refer to Figure 7 When the target vehicle and surrounding cars (corresponding to surrounding vehicles) choose the same lane, if the surrounding cars have higher right-of-way than the target vehicle, the target vehicle needs to give way. For example, the vehicle terminal controls the target vehicle to switch to the right lane to exit the road.

[0136] refer to Figure 8 When the target vehicle and surrounding vehicles are in the same lane, since the target vehicle is going straight and the surrounding vehicles are turning, the target vehicle has a higher right-of-way than the surrounding vehicles, so the on-board terminal does not need to control the target vehicle to change lanes. In some embodiments, if the surrounding vehicles turning are traveling at a higher speed, the target vehicle can be controlled to slow down, but there is no need to control the target vehicle to change lanes.

[0137] refer to Figure 9 When the target vehicle and the surrounding buses choose the same lane, because the buses are large vehicles, meaning the size of the surrounding vehicles is larger than that of the target vehicle, for safety reasons, even if the target vehicle has a greater right-of-way than the turning bus when it is going straight, the on-board terminal will still control the target vehicle to change lanes to the right lane to give way, so as to ensure the driving safety of the target vehicle.

[0138] refer to Figure 10 The vehicle terminal can select drivable lanes from the virtual lanes through pre-screening, evaluate the scores of drivable lanes based on multiple dimensions, and after selecting the drivable lane with the highest score, further perform a secondary safety check on the volume of surrounding vehicles and the vehicle's own volume to select the optimal lane as the target drivable lane, so as to ensure the driving safety of the target vehicle as much as possible.

[0139] Figure 11 A schematic diagram illustrating another target entry lane selection principle proposed according to some embodiments of this application is shown.

[0140] In other embodiments of this application, reference is made to Figure 11After the vehicle terminal obtains the target entry lane, it can update the historical target entry lane. Each time a new target entry lane is selected, the historical target entry lane can be merged with the currently available entry lanes obtained in real time based on the first perception data, the second perception data, and the third perception data to form a filtered available entry lane. The historical target entry lane and the currently obtained available entry lanes are evaluated simultaneously to ensure the stability of the selected target entry lane and avoid frequent lane changes.

[0141] Therefore, through the above steps S301 to S303, the intersection passage method proposed in this application embodiment allows the vehicle terminal to collect and identify perception data in the environment through a visual solution after the target vehicle determines the scenario of entering the intersection. Then, based on different types of perception data, at least one virtual lane is constructed. From all virtual lanes, drivable lanes are pre-screened. In the case of multiple drivable lanes, the target drivable lane is determined from the multiple drivable lanes. For example, the drivable lane can be combined with historical target drivable lanes for multi-dimensional drivability assessment, and secondary safety verification can be performed by combining right-of-way and the volume of surrounding vehicles, thereby obtaining the optimal lane as the target drivable lane. Thus, a stable and accurate selection of a lane that can safely pass through the intersection can be achieved through a pure visual solution.

[0142] Based on the intersection passage method proposed in the embodiments of this application, this application also provides an intersection passage device 1200.

[0143] Figure 12 A schematic diagram of the frame structure of an intersection passage device 1200 according to an embodiment of this application is shown.

[0144] refer to Figure 12 The intersection traffic control device 1200 includes a virtual lane construction module 1201, a lane pre-screening module 1202, and a data processing module 1203. The virtual lane construction module 1201, corresponding to a scenario where a target vehicle is determined to enter the intersection, determines, based on acquired first perception data, the current location of the target vehicle, at least one exit lane corresponding to the target vehicle's current travel direction, and at least one virtual lane connecting the target vehicle's current location to each exit lane. The lane pre-screening module 1202, based on acquired second perception data, determines at least one accessible lane from the at least one virtual lane. The data processing module 1203, corresponding to at least one accessible lane including multiple accessible lanes, determines the target entry lane from the multiple accessible lanes based on acquired third perception data.

[0145] The specific implementation of the virtual lane construction module 1201 can be referred to the specific implementation process of step S301 above, and will not be repeated here.

[0146] The specific implementation of the lane pre-screening module 1202 can be referred to the specific implementation process of step S302 above, and will not be repeated here.

[0147] The specific implementation of the data processing module 1203 can be referred to the specific implementation process of step S303 above, and will not be repeated here.

[0148] According to the intersection passage method provided in the embodiments of this application, this application also provides an electronic device, which includes: one or more processors; one or more memories; the one or more memories storing one or more programs, which, when executed by one or more processors, cause the electronic device to execute the intersection passage method in any of the above embodiments.

[0149] In some embodiments, the vehicle terminal exemplified above may be this electronic device.

[0150] According to the intersection passage method provided in the embodiments of this application, this application also provides a vehicle, which includes the electronic equipment in any of the above embodiments.

[0151] In some embodiments, the vehicle may be the target vehicle mentioned above. Therefore, the vehicle further includes a perception module, which includes an imaging component and a sensor. The imaging component is used to capture one or more images of the target vehicle's surroundings. The sensor is used to acquire the target vehicle's driving parameters corresponding to each of the multiple images of the target vehicle's surroundings.

[0152] According to the intersection passage method provided in the embodiments of this application, this application also provides a computer program product, which includes: computer program code, which, when run on a computer, causes the computer to implement the steps executed by the electronic device in any of the above embodiments.

[0153] According to the intersection passage method provided in the embodiments of this application, this application also provides a computer-readable medium storing program code, which, when run on a computer, causes the computer to implement the steps executed by the electronic device in any of the above embodiments.

[0154] In some embodiments, the above-described electronic device may be electronic device 1500 as exemplified below.

[0155] The specific structure of the electronic device 1500 will be described in detail below with reference to the accompanying drawings.

[0156] Figure 13 This diagram illustrates the structure of an electronic device 1500 according to some embodiments of the present application.

[0157] like Figure 13 As shown, the electronic device 1500 includes one or more processors 1501, system memory 1502, non-volatile memory (NVM) 1503, communication interface 1504, input / output (I / O) device 1505, and system control logic 1506 for coupling the processor 1501, system memory 1502, non-volatile memory 1503, communication interface 1504, and input / output (I / O) device 1505. Wherein:

[0158] Processor 1501 may include one or more processing units, such as data processing units or processing circuits that may include a central processing unit (CPU), graphics processing unit (GPU), digital signal processor (DSP), micro-programmed control unit (MCU), artificial intelligence (AI) processor, field programmable gate array (FPGA), neural network processing unit (NPU), etc., and may include one or more single-core or multi-core processors. In some embodiments, processor 1501 may be used to execute instructions to implement the above-described intersection passage method.

[0159] System memory 1502 is volatile memory, such as random-access memory (RAM), double data rate synchronous dynamic random access memory (DDR SDRAM), etc. System memory 1502 is used for temporary storage of data and / or instructions. For example, in some embodiments, system memory 1502 can be used to store instructions, or it can be used to store original data objects and modified data objects.

[0160] The non-volatile memory 1503 may include one or more tangible, non-transitory computer-readable media for storing data and / or instructions. In some embodiments, the non-volatile memory 1503 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as a hard disk drive (HDD), compact disc (CD), digital versatile disc (DVD), solid-state drive (SSD), etc. In some embodiments, the non-volatile memory 1503 may also be a removable storage medium, such as a secure digital (SD) memory card. In other embodiments, the non-volatile memory 1503 may be used to store instructions, or to store original data objects and modified data objects.

[0161] In some embodiments, system memory 1502 and non-volatile memory 1503 may each include a temporary copy and a permanent copy of instruction 1507. Instruction 1507 may include, when executed by at least one of processors 1501, causing electronic device 1500 to implement the intersection passage method provided in the embodiments of this application.

[0162] Communication interface 1504 may include a transceiver for providing a wired or wireless communication interface for electronic device 1500, thereby enabling communication with any other suitable device via one or more networks. In some embodiments, communication interface 1504 may be integrated into other components of electronic device 1500, for example, communication interface 1504 may be integrated into processor 1501. In some embodiments, electronic device 1500 may communicate with other devices through communication interface 1504. For example, electronic device 1500 may establish a communication connection with other devices through communication interface 1504 to send data change requests, obtain original data objects, and send changed data objects to other devices through the communication connection.

[0163] Input / output (I / O) device 1505 may include input devices such as keyboards and mice, and output devices such as monitors. Users can interact with electronic devices 1500 through input / output (I / O) device 1505. For example, business personnel can input / select the content to be changed through input / output (I / O) device 1505.

[0164] System control logic 1506 may include any suitable interface controller to provide any suitable interface to other modules of electronic device 1500. For example, in some embodiments, system control logic 1506 may include one or more memory controllers to provide an interface to system memory 1502 and non-volatile memory 1503.

[0165] In some embodiments, at least one of the processors 1501 may be packaged together with the logic of one or more controllers for the system control logic unit 1506 to form a system in package (SiP). In other embodiments, at least one of the processors 1501 may also be integrated on the same chip with the logic of one or more controllers for the system control logic unit 1506 to form a system-on-chip (SoC).

[0166] Understandable. Figure 13 The structure of the electronic device 1500 shown is merely an example. In other embodiments, the electronic device 1500 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0167] Various embodiments of the mechanisms disclosed in this application can be implemented in hardware, software, firmware, or a combination of these implementation methods. Embodiments of this application can be implemented as computer modules or module code executing on a programmable system, the programmable system including at least one processor, a storage system (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device.

[0168] Module code can be applied to input instructions to execute the functions described in this application and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, the processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application-specific integrated circuit (ASIC), or a microprocessor.

[0169] Module code can be implemented using a high-level modular language or an object-oriented programming language to communicate with the processing system. Assembly language or machine language can also be used to implement module code when needed. In fact, the mechanisms described in this application are not limited to any particular programming language. In either case, the language can be a compiled language or an interpreted language.

[0170] In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored thereon on one or more temporary or non-temporary machine-readable (e.g., computer-readable) storage media, which may be read and executed by one or more processors. For example, the instructions may be distributed via a network or through other computer-readable media. Therefore, machine-readable media may include any mechanism for storing or transmitting information in a machine-readable (e.g., computer-readable) form, including but not limited to floppy disks, optical disks, CD-ROMs, magneto-optical disks, read-only memory (ROM), random access memory (RAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic cards or optical cards, flash memory, or tangible machine-readable storage for transmitting information (e.g., carrier waves, infrared signals, digital signals, etc.) using the Internet in the form of electrical, optical, acoustic, or other forms of propagated signals. Therefore, machine-readable media include any type of machine-readable medium suitable for storing or transmitting electronic instructions or information in a machine-readable (e.g., computer-readable) form.

[0171] In the accompanying drawings, some structural or methodological features may be shown in a specific arrangement and / or order. However, it should be understood that such a specific arrangement and / or order may not be necessary. Rather, in some embodiments, these features may be arranged in a manner and / or order different from that shown in the illustrative drawings. Furthermore, the inclusion of structural or methodological features in a particular figure does not imply that such features are required in all embodiments, and in some embodiments, these features may be omitted or may be combined with other features.

[0172] Various embodiments of the mechanisms disclosed in this application can be implemented in hardware, software, firmware, or combinations of these implementation methods. Embodiments of this application can be implemented as computer programs or program code executable on a programmable system, the programmable system including at least one processor, a storage system (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device.

[0173] Program code can be applied to input instructions to execute the functions described in this application and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, the processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application-specific integrated circuit (ASIC), or a microprocessor.

[0174] The program code can be implemented using a high-level procedural language or an object-oriented programming language to communicate with the processing system. Assembly language or machine language can also be used when needed. In fact, the mechanisms described in this application are not limited to any particular programming language. In either case, the language can be a compiled language or an interpreted language.

[0175] It should be noted that all units / modules mentioned in the device embodiments of this application are logical units / modules. Physically, a logical unit / module can be a physical unit / module, a part of a physical unit / module, or a combination of multiple physical units / modules. The physical implementation of these logical units / modules themselves is not the most important factor; the combination of functions implemented by these logical units / modules is the key to solving the technical problems proposed in this application. Furthermore, to highlight the innovative aspects of this application, the above-described device embodiments of this application have not introduced units / modules that are not closely related to solving the technical problems proposed in this application. This does not mean that the above-described device embodiments do not contain other units / modules.

[0176] It should be noted that in the examples and description of this application, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0177] In this specification, references to "some embodiments" or "embodiments" mean that a specific feature, structure, or characteristic described in connection with an embodiment is included in at least one exemplary implementation or technology disclosed according to embodiments of this application. The phrase "in some embodiments" appearing in various places in the specification does not necessarily refer to the same embodiment.

[0178] Furthermore, the language used in this specification has been primarily chosen for readability and instructional purposes and may not have been chosen to depict or limit the disclosed subject matter. Therefore, the embodiments disclosed herein are intended to illustrate, and not limit, the scope of the concepts discussed herein.

Claims

1. A method for traffic passage at an intersection, applied to an on-board terminal, characterized in that, The method includes: Corresponding to the scenario of determining that a target vehicle is entering the intersection, based on the acquired first perception data, the current location of the target vehicle, at least one exit lane corresponding to the current direction of travel of the target vehicle, and at least one virtual lane for connecting the current location of the target vehicle with each exit lane are determined. Based on the acquired second perception data, at least one accessible lane is determined from the at least one virtual lane; The at least one accessible lane includes multiple accessible lanes. The target access lane is determined from the multiple accessible lanes based on the acquired third perception data. The first perception data, the second perception data, and the third perception data are obtained based on one or more images of the area surrounding the target vehicle.

2. The intersection passage method according to claim 1, characterized in that, The determination of the target entry lane from the plurality of accessible lanes based on the acquired third perception data includes: Based on the acquired third perception data, the drivability of each of the multiple drivable lanes is assessed to obtain a first score for each drivable lane. The drivable lanes that satisfy the first score condition among the drivable lanes are determined as the target drivable lanes.

3. The intersection passage method according to claim 1, characterized in that, The at least one accessible lane includes one accessible lane, and the one accessible lane is used as the target accessible lane.

4. The intersection passage method according to any one of claims 1 to 3, characterized in that, The first perception data includes intersection recognition features and a drivable centerline for the vehicle. The intersection recognition features include intersection entrance lines, intersection exit lines, intersection guide lines, and exit lane lines. The step of determining the current location of the target vehicle, at least one exit lane corresponding to the current travel direction of the target vehicle, and at least one virtual lane for connecting the current location of the target vehicle with each exit lane based on the acquired first perception data includes: The current relative position of the target vehicle with respect to the intersection is determined based on the intersection entrance line, intersection exit line, intersection guide line and exit lane line, so as to determine the current location of the target vehicle. Using the current location of the target vehicle as the origin, determine the coordinates of the intersection entrance line, the intersection exit line, the intersection guide line, the exit lane line, and the trajectory of the vehicle's drivable center line. The center line of the exit lane at the intersection is determined based on the coordinates of the exit lane lines. Based on the trajectory of the drivable centerline of the vehicle, the coordinates of the intersection entrance line, the coordinates of the intersection exit line, and the geometric relationship with the centerline of the exit lane, the unique entrance point of the intersection and the set of exit points of the intersection are determined. The intersection entrance angle and the intersection exit angle corresponding to each exit lane are determined based on the unique entrance point and the set of exit points. The at least one virtual lane is determined based on the intersection entrance angle corresponding to each exit lane, the intersection exit angle corresponding to each exit lane, the location parameters of the unique entrance point, and the location parameters of the set of exit points.

5. The intersection passage method according to any one of claims 1 to 3, characterized in that, The second sensing data includes intersection recognition features and drivable areas. The intersection recognition features include intersection entrance lines, intersection exit lines, intersection guide lines, and exit lane lines. The determination of at least one accessible lane from the at least one virtual lane based on the acquired second perception data includes: Determine the exit route based on the intersection identification features; Based on the roadside line, a first type of non-traveling lane is identified from the at least one virtual lane, wherein the first type of non-traveling lane includes auxiliary roads, oncoming lanes, and non-motorized vehicle lanes; Based on the lane centerline of the at least one virtual lane, the drivable area, the intersection entrance line, the intersection exit line, the guide line within the intersection, and the exit lane line, a second type of non-drivable lane is identified from the at least one virtual lane. The second type of non-drivable lane includes non-drivable lanes that intersect with the curb, guide line, and lane line. The first type of non-drivable lane and the second type of non-drivable lane are removed from the at least one virtual lane to obtain the at least one drivable lane.

6. The intersection passage method according to claim 2, characterized in that, The third sensing data includes intersection recognition features, traffic flow features, drivable areas, and vehicle drivable centerlines. The intersection recognition features include intersection entrance lines, intersection exit lines, intersection guide lines, and exit lane lines. The accessibility of each of the multiple accessible lanes is assessed based on the acquired third-sensory data to obtain a first score for each accessible lane, including: If it is determined that the target vehicle has not entered the intersection, the heading angle of the target vehicle entering each drivable lane is determined based on the current location of the target vehicle and the lane entrance location of each drivable lane, and a first dimension score of each drivable lane is determined based on the heading angle, wherein the first dimension score is used to characterize the steering angle smoothness of each drivable lane. The coupling trend parameters between the guide line within the intersection and each drivable lane are obtained, and a second dimension score for each drivable lane is determined based on the coupling trend parameters. The second dimension score is used to characterize the degree of coupling between each drivable lane and the guide line within the intersection. Based on the traffic flow characteristics, the obstacle characteristics on each accessible lane are determined. Based on the obstacle characteristics on each accessible lane and the target vehicle, the collision parameters between the target vehicle and the obstacles on each accessible lane are determined. Based on the collision parameters, a third-dimensional score for each accessible lane is determined. The third-dimensional score is used to characterize the probability that the target vehicle collides with the obstacle while driving in each accessible lane. The first score is determined based on the first dimension score, the second dimension score, and the third dimension score.

7. The intersection passage method according to claim 2, characterized in that, The third sensing data includes intersection recognition features, traffic flow features, drivable areas, and vehicle drivable centerlines. The intersection recognition features include intersection entrance lines, intersection exit lines, intersection guide lines, and exit lane lines. The accessibility of each of the multiple accessible lanes is assessed based on the acquired third-sensory data to obtain a first score for each accessible lane, including: If it is determined that the target vehicle has entered the intersection, the historical target entry lane is obtained, the first lateral offset between each drivable lane and the historical target entry lane is determined, and the fourth dimension score of each drivable lane is determined based on the first lateral offset, wherein the fourth dimension score is used to characterize the lateral offset of the target vehicle from the historical target entry lane into each drivable lane. Based on the traffic flow characteristics, the obstacle characteristics on each accessible lane are determined. Based on the obstacle characteristics on each accessible lane and the target vehicle, the collision parameters between the target vehicle and the obstacles on each accessible lane are determined. Based on the collision parameters, a fifth dimension score for each accessible lane is determined. The fifth dimension score is used to characterize the probability that the target vehicle collides with the obstacle while driving in each accessible lane. Based on the second lateral offset of the vehicle's drivable centerline relative to each drivable lane, a sixth dimension score is determined for each drivable lane, wherein the sixth dimension score is used to characterize the degree of vehicle body sway of the target vehicle when driving in each drivable lane. Based on the current location of the target vehicle, the entrance location of the intersection, and the entrance locations of each accessible lane, a first steering parameter from the entrance location of the intersection to the current location of the target vehicle and a second steering parameter from the current location of the target vehicle to the entrance locations of each accessible lane are determined. Based on the first steering parameter and the second steering parameter of each accessible lane, a seventh dimension score for each accessible lane is determined, wherein the seventh dimension score is used to characterize the priority right of passage of the target vehicle when traveling in each accessible lane. The first score is determined based on the fourth dimension score, the fifth dimension score, the sixth dimension score, and the seventh dimension score.

8. The intersection passage method according to claim 2, characterized in that, The step of determining the drivable lanes among the drivable lanes whose first score meets the first score condition as the target drivable lanes includes: The lane with the highest score in the first rating among all the drive-in lanes is determined as the target drive-in lane.

9. The intersection passage method according to claim 8, characterized in that, The step of determining the drivable lanes among the drivable lanes whose first score satisfies the first score condition as the target drivable lanes further includes: If the right-of-way of the target vehicle in each of the accessible lanes conflicts with the right-of-way of surrounding vehicles, and if the vehicle body size of the surrounding vehicles is larger than that of the target vehicle, then the first score of the accessible lane that the surrounding vehicles are about to enter will be reduced. The lane with the highest score in the first rating among all the drive-in lanes is determined as the target drive-in lane.

10. The intersection passage method according to claim 1, characterized in that, The method further includes: Take one or more images of the target vehicle and its surroundings, and obtain the target vehicle's driving parameters corresponding to each of the multiple images of the target vehicle's surroundings. Based on the target vehicle driving parameters corresponding to each target vehicle surrounding image, the first perception data, the second perception data, and the third perception data are determined from one or more target vehicle surrounding images.

11. A traffic control device for intersections, applied to target vehicles, characterized in that, The device includes a virtual lane construction module, a lane pre-screening module, and a data processing module, wherein... The virtual lane construction module is used to determine the current location of the target vehicle, at least one exit lane corresponding to the current direction of travel of the target vehicle, and at least one virtual lane connecting the current location of the target vehicle with each exit lane, based on the first perception data obtained, in the context of determining the scenario where the target vehicle enters the intersection. The lane pre-screening module is used to determine at least one accessible lane from the at least one virtual lane based on the acquired second perception data; The data processing module is used to determine the target entry lane from the multiple drivable lanes based on the acquired third perception data, corresponding to the at least one drivable lane including multiple drivable lanes. The first perception data, the second perception data, and the third perception data are obtained by identification based on one or more images around the target vehicle.

12. An electronic device, characterized in that, include: One or more processors; One or more memories; the one or more memories storing one or more programs, which, when executed by the one or more processors, cause the electronic device to perform the intersection passage method according to any one of claims 1-10.

13. A computer-readable medium, characterized in that, The computer-readable medium stores instructions that, when executed on a machine, cause the machine to perform the intersection passage method according to any one of claims 1-10.

14. A computer program product, characterized in that, It includes a computer program / instruction that, when executed by a processor, implements the intersection passage method according to any one of claims 1-10.

15. A vehicle, characterized in that, The vehicle includes the electronic equipment as described in claim 12.

Images