Intersection lane scene recognition method and related equipment

Abstract

The invention discloses an intersection lane scene recognition method and related equipment, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining scene data, traversing a frame with separated lane elements or combined lane elements in the scene data, taking the traversed frame as a current frame, determining a corresponding perception map based on the current frame, and carrying out the recognition of an intersection lane scene. And determining a start frame and an end frame corresponding to the lane center line set in the sensing map, and determining an intersection lane scene corresponding to a first lane center line and a second lane center line in the lane center line set based on the start frame, the end frame and the sensing map. According to the method, the sensing map is determined based on the scene data corresponding to the frame with the separated lane elements or the combined lane elements, then the specific type of the intersection lane scene corresponding to the two lane lines with the overlapped areas in the sensing map is judged, and the intersection lane scene is accurately determined from a large amount of scene data.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method for identifying lane scenes at intersections and related equipment. Background Art

[0002] With the evolution of intelligent transportation systems, refined data collection and analysis of intersection and lane scenes have become a core requirement for improving traffic management efficiency. Therefore, how to accurately identify intersection and lane scenes is crucial.

[0003] However, with the explosive growth of data scale, the amount of data to be processed has increased, making it very difficult to accurately identify intersection lane scenes from historical driving data.

[0004] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is related technology. Summary of the Invention

[0005] The main purpose of this application is to provide a method for identifying lane intersection scenes, aiming to solve the technical problem of how to accurately identify lane intersection scenes.

[0006] To achieve the above objectives, the present application proposes a method for identifying lane scenes at intersections, the method comprising: Get scene data; Traversing frames in the scene data that contain separated lane elements or merged lane elements, taking the traversed frames as current frames, and determining a corresponding perception map based on the current frames; Determine a start frame and an end frame corresponding to a lane centerline set in the perception map, and determine, based on the start frame, the end frame, and the perception map, an intersection lane scene corresponding to a first lane centerline and a second lane centerline in the lane centerline set, wherein the intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set includes two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

[0007] In one embodiment, the step of determining, based on the start frame, the end frame, and the perception map, the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set includes: Calculating, based on the start frame, the end frame, and the perception map, a first projection distance from the vehicle to the centerline of the first lane to obtain the first projection distance corresponding to each frame from the start frame to the end frame, and calculating a second projection distance from the vehicle to the centerline of the second lane to obtain the second projection distance corresponding to each frame from the start frame to the end frame; Calculating the sum of the first projection distance and the second projection distance corresponding to the same frame to obtain multiple sets of third projection distances; Calculating an average value of the plurality of sets of third projection distances, determining a difference between the third projection distance corresponding to the start frame and the average value, and determining a difference between the third projection distance corresponding to the end frame and the average value; If the third projection distance corresponding to the starting frame is greater than the average value, and the third projection distance corresponding to the ending frame is less than the average value, it is determined that the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a merging lane scene; if the third projection distance corresponding to the starting frame is less than the average value, and the third projection distance corresponding to the ending frame is greater than the average value, then the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a separating lane scene.

[0008] In one embodiment, the step of determining the corresponding perception map based on the current frame includes: Determining a plurality of lane centerline coordinate sequences based on the scene data; A global grid is constructed, each coordinate in the coordinate sequence is mapped to the global grid, and a perception map corresponding to the current frame is determined.

[0009] In one embodiment, the steps of constructing a global grid, mapping each coordinate in the coordinate sequence to the global grid, and determining a perception map corresponding to the current frame include: Determining a lane centerline boundary value based on the coordinate sequence, and constructing a global grid based on the boundary value; Mapping each coordinate in the coordinate sequence to the global grid to obtain a target grid; Based on the target grid, searching for the centerline of the third lane and the centerline of the fourth lane where there is an overlapping area, determining the number of overlaps between the centerline of the third lane and the centerline of the fourth lane, and determining whether the number of overlaps is greater than or equal to a preset overlap number threshold; If the number of overlaps is greater than or equal to a preset overlap threshold, the third lane centerline and the fourth lane centerline are added to the same lane centerline set; or, if the number of overlaps is less than the preset overlap threshold, the third lane centerline and the fourth lane centerline are added to different lane centerline sets respectively. Based on the set of all lane centerlines, determine the perception map corresponding to the current frame.

[0010] In one embodiment, the step of mapping each coordinate in the coordinate sequence to the global grid to obtain a target grid further includes any one of the following: Mapping each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, and performing a first thinning operation on the initial grid to obtain a target grid; Each coordinate in the coordinate sequence is mapped to a corresponding position in the grid to obtain an initial grid, a first thinning operation is performed on the initial grid, and a second thinning operation is performed on the grid after the first thinning operation to obtain a target grid, wherein the corresponding thinning operation includes a floor operation.

[0011] In one embodiment, the step of determining a plurality of lane centerline coordinate sequences based on the scene data further includes: Determining vehicle position information based on the scene data, wherein the vehicle position information includes vehicle coordinates and a vehicle heading angle; constructing a vehicle coordinate system based on the vehicle position information; Obtaining a plurality of coordinates corresponding to the centerline of each lane, and determining a first distance from each coordinate to the vehicle based on the vehicle coordinate system; Each coordinate is saved in a coordinate sequence in ascending order based on the first distance, to obtain multiple sets of lane centerline coordinate sequences.

[0012] In one embodiment, the step of constructing a vehicle coordinate system based on the vehicle position information further includes: Extracting vehicle coordinates and vehicle heading angle from the vehicle position information; The vehicle coordinates are set as the origin, the direction corresponding to the vehicle heading angle is set as the positive direction of the y-axis, and based on the origin and the positive direction of the y-axis, a vehicle coordinate system corresponding to each set of local perception type map data is constructed.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a road intersection lane scene recognition device, the road intersection lane scene recognition device comprising: An acquisition module, configured to acquire scene data; a first determining module configured to traverse frames containing separated lane elements or merged lane elements in the scene data, use the traversed frames as current frames, and determine a corresponding perception map based on the current frames; A second determination module is used to determine the starting frame and the ending frame corresponding to the lane centerline set in the perception map, and determine the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set based on the starting frame, the ending frame and the perception map, wherein the intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set contains two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

[0014] In one embodiment, the second determining module includes: a first calculation unit, configured to calculate, based on the start frame, the end frame, and the perception map, a first projection distance from the vehicle to the center line of the first lane, and obtain the first projection distance corresponding to each frame from the start frame to the end frame; and calculate a second projection distance from the vehicle to the center line of the second lane, and obtain the second projection distance corresponding to each frame from the start frame to the end frame; a second calculation unit, configured to calculate the sum of the first projection distance and the second projection distance corresponding to the same frame to obtain multiple sets of third projection distances; a third calculating unit, configured to calculate an average value of the plurality of sets of third projection distances, determine a difference between the third projection distance corresponding to the start frame and the average value, and determine a difference between the third projection distance corresponding to the end frame and the average value; The first determination unit is used to determine that the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a merging lane scene if the third projection distance corresponding to the starting frame is greater than the average value and the third projection distance corresponding to the ending frame is less than the average value; if the third projection distance corresponding to the starting frame is less than the average value and the third projection distance corresponding to the ending frame is greater than the average value, then the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a separating lane scene.

[0015] In one embodiment, the first determining module includes: a second determining unit, configured to determine a plurality of sets of lane centerline coordinate sequences based on the scene data; The first construction unit is configured to construct a global grid, map each coordinate in the coordinate sequence to the global grid, and determine a perception map corresponding to the current frame.

[0016] In one embodiment, the first determining module further includes: a third determining unit, configured to determine a boundary value of a lane centerline based on the coordinate sequence, and construct a global grid based on the boundary value; A first mapping unit is configured to map each coordinate in the coordinate sequence to the global grid to obtain a target grid; a search unit configured to search, based on the target grid, for a third lane centerline and a fourth lane centerline having an overlapping area, determine a number of overlaps between the third lane centerline and the fourth lane centerline, and determine whether the number of overlaps is greater than or equal to a preset overlap threshold; an adding unit, configured to add the third lane centerline and the fourth lane centerline to the same lane centerline set if the number of overlaps is greater than or equal to a preset overlap number threshold; or to add the third lane centerline and the fourth lane centerline to different lane centerline sets if the number of overlaps is less than the preset overlap number threshold; The fourth determining unit is configured to determine a perception map corresponding to the current frame based on a set of all lane centerlines.

[0017] In one embodiment, the first determining module further includes: a second mapping unit, configured to map each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, and perform a first thinning operation on the initial grid to obtain a target grid; The third mapping unit is used to map each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, perform a first thinning operation on the initial grid, and perform a second thinning operation on the grid after the first thinning operation to obtain a target grid, wherein the corresponding thinning operation includes a rounding-down operation.

[0018] In one embodiment, the first determining module further includes: a fifth determining unit, configured to determine vehicle position information based on the scene data, wherein the vehicle position information includes vehicle coordinates and a vehicle heading angle; A second construction unit is configured to construct a vehicle coordinate system based on the vehicle position information; a sixth determining unit, configured to obtain a plurality of coordinates corresponding to a centerline of each lane, and determine a first distance from each coordinate to the vehicle based on the vehicle coordinate system; The storage unit is used to save each coordinate in a coordinate sequence in ascending order based on the first distance to obtain multiple sets of lane centerline coordinate sequences.

[0019] In one embodiment, the first determining module further includes: an extraction unit, configured to extract the vehicle coordinates and the vehicle heading angle from the vehicle position information; A setting unit is used to set the vehicle coordinates as the origin, set the direction corresponding to the vehicle heading angle as the positive direction of the y-axis, and construct a vehicle coordinate system corresponding to each set of local perception type map data based on the origin and the positive direction of the y-axis.

[0020] In addition, to achieve the above-mentioned purpose, the present application also proposes a road intersection lane scene recognition device, which includes: a memory, a processor, and a computer program stored on the memory and runnable on the processor, and the computer program is configured to implement the steps of the road intersection lane scene recognition method as described above.

[0021] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the intersection lane scene recognition method described above are implemented.

[0022] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the intersection lane scene recognition method as described above.

[0023] One or more technical solutions proposed in this application have at least the following technical effects: The present application proposes a method and related equipment for identifying lane scenes at an intersection, which relate to the field of autonomous driving technology. Compared with related technologies, in the context of explosive growth in data scale, the amount of data to be processed increases, making it very difficult to accurately identify lane scenes at an intersection from historical driving data. In the present application, scene data is first acquired. Then, frames containing separated lane elements or merged lane elements in the scene data are traversed, the traversed frame is used as the current frame, and a corresponding perception map is determined based on the current frame. Furthermore, a start frame and an end frame corresponding to a lane centerline set in the perception map are determined. Based on the start frame, the end frame, and the perception map, an intersection lane scene corresponding to a first lane centerline and a second lane centerline in the lane centerline set is determined. The intersection lane scene includes a separated lane scene and a merged lane scene. The lane centerline set includes two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

[0024] It can be understood that this application determines the perception map based on the scene data corresponding to the frames with separated lane elements or merged lane elements, and then judges the specific type of intersection lane scene corresponding to the two lane lines with overlapping areas in the perception map, and accurately identifies the intersection lane scene from a large amount of scene data. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0026] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0027] Figure 1 A flowchart of the first embodiment of the lane intersection scene recognition method provided in this application; Figure 2 A flowchart of the second embodiment of the lane intersection scene recognition method provided in this application; Figure 3 A flowchart of the third embodiment of the lane scene recognition method for intersections provided in this application; Figure 4 This is a schematic diagram of the module structure of the lane intersection scene recognition device according to an embodiment of the present application; Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the intersection lane scene recognition method in the embodiment of this application.

[0028] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0029] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0030] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0031] The main solutions of the embodiments of this application are: In this embodiment, for ease of description, the following description is made with the intersection lane scene recognition device as the execution entity.

[0032] Due to existing technologies: in the context of explosive growth in data scale, the amount of data to be processed increases, making it very difficult to accurately identify intersection lane scenes from historical driving data.

[0033] The present application provides a solution that: obtains scene data, traverses frames in the scene data where separated lane elements or merged lane elements exist, uses the traversed frame as the current frame, and determines a corresponding perception map based on the current frame, determines the starting frame and ending frame corresponding to the lane centerline set in the perception map, and determines the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set based on the starting frame, the ending frame, and the perception map. The present application determines the perception map based on scene data corresponding to frames where separated lane elements or merged lane elements exist, and then determines the specific type of intersection lane scene corresponding to two lane lines with overlapping areas in the perception map, accurately determining the intersection lane scene from a large amount of scene data.

[0034] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication, and program execution capabilities, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device capable of implementing the aforementioned functions, such as an intersection lane scene recognition device. This embodiment and the following embodiments will be described below using the intersection lane scene recognition device as an example.

[0035] Based on this, the embodiment of the present application provides a method for identifying lane intersection scenes, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the intersection lane scene recognition method of the present application.

[0036] In this embodiment, the intersection lane scene recognition method includes steps S100 to S300: Step S100, acquiring scene data; It should be noted that scene data is data collected by the vehicle's sensors during the vehicle's driving process. The scene data includes lane data, pedestrian data on the roadside, etc. during the vehicle's driving process. Among them, the vehicle's sensors include visual sensors, radars, etc.

[0037] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, grid communication, and program execution capabilities, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device capable of implementing the above functions, such as an intersection lane scene recognition device. The following uses the intersection lane scene recognition device as an example to illustrate this embodiment and the following embodiments.

[0038] In this application, specific application scenarios may be: Vehicles traveling on the road use sensors such as lidar, radar, and cameras to collect road data, vehicle driving data, and traffic environment data in real time. Furthermore, intersection lane scene recognition equipment receives the road data, vehicle driving data, and traffic environment data collected by the sensors. Among them, the scene data is the road data, vehicle driving data, and traffic environment data collected by the sensors in real time.

[0039] Step S200, traversing frames in the scene data that contain separated lane elements or merged lane elements, taking the traversed frame as the current frame, and determining a corresponding perception map based on the current frame; It should be noted that if the scene data corresponding to a certain frame contains at least two lane centerlines (the first lane centerline and the second lane centerline), where the first lane centerline and the second lane centerline have an overlapping area, and neither the first lane centerline nor the second lane centerline is an intersection lane guide line, then it is considered that the scene data corresponding to the frame contains a separated lane element or a merged lane element, and the frame is set as the current frame.

[0040] Step S300: Determine the starting frame and ending frame corresponding to the lane centerline set in the perception map, and determine the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set based on the starting frame, the ending frame, and the perception map, wherein the intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set includes two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

[0041] It should be noted that the lane centerline is a characteristic line formed by connecting the center points of the lane in sequence. The lane centerline can reflect the plane position and curvature of the road.

[0042] Specifically, the step of determining the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set based on the start frame, the end frame, and the perception map includes steps S310 to S340: Step S310: Calculating a first projection distance from the vehicle to the centerline of the first lane based on the start frame, the end frame, and the perception map to obtain the first projection distance corresponding to each frame from the start frame to the end frame, and calculating a second projection distance from the vehicle to the centerline of the second lane to obtain the second projection distance corresponding to each frame from the start frame to the end frame; It should be noted that, in actual scenarios, it is only necessary to determine the first projection distance and the second projection distance once every n frames, which reduces the amount of calculation and improves calculation efficiency.

[0043] Step S320, calculating the sum of the first projection distance and the second projection distance corresponding to the same frame to obtain multiple sets of third projection distances; Step S330, calculating an average value of the plurality of sets of third projection distances, determining a difference between the third projection distance corresponding to the start frame and the average value, and determining a difference between the third projection distance corresponding to the end frame and the average value; Step S340: If the third projection distance corresponding to the starting frame is greater than the average value, and the third projection distance corresponding to the ending frame is less than the average value, it is determined that the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a merged lane scene; if the third projection distance corresponding to the starting frame is less than the average value, and the third projection distance corresponding to the ending frame is greater than the average value, then the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a separated lane scene.

[0044] It is understood that if the third projected distance corresponding to the starting frame is greater than the average value, and the third projected distance corresponding to the ending frame is less than the average value, it indicates that the distance between the two lanes is gradually decreasing, and the intersection scene corresponding to the two lanes is therefore considered to be a merging lane scene. Similarly, if the third projected distance corresponding to the starting frame is less than the average value, and the third projected distance corresponding to the ending frame is greater than the average value, it indicates that the distance between the two lanes is gradually increasing, and the intersection scene corresponding to the two lanes is therefore considered to be a separating lane scene.

[0045] One or more technical solutions proposed in this application have at least the following technical effects: The present application proposes a method and related equipment for identifying lane scenes at an intersection, which relate to the field of autonomous driving technology. Compared with related technologies, the amount of data to be processed increases under the background of explosive growth in data scale, making it very difficult to accurately identify lane scenes at an intersection from historical driving data. In the present application, first, scene data is acquired. Then, frames containing separated lane elements or merged lane elements in the scene data are traversed, the traversed frame is used as the current frame, and a corresponding perception map is determined based on the current frame. Furthermore, a start frame and an end frame corresponding to a lane centerline set in the perception map are determined. Based on the start frame, the end frame, and the perception map, an intersection lane scene corresponding to a first lane centerline and a second lane centerline in the lane centerline set is determined. The intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set includes two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold. Finally, based on the intersection scene, a map scene corresponding to the perception map corresponding to the current frame is determined.

[0046] It can be understood that the present application determines the perception map based on the scene data corresponding to the frames with separated lane elements or merged lane elements, and then judges the specific type of intersection lane scene corresponding to the two lane lines with overlapping areas in the perception map. Based on the intersection lane scene in the perception map, the intersection lane scene is accurately determined from a large amount of scene data.

[0047] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above embodiment 1 can be referred to the above introduction and will not be described in detail later. Figure 2 The step of determining the corresponding perception map based on the current frame further includes steps A100 to A200: Step A100, determining a coordinate sequence of multiple lane centerlines based on the scene data; It should be noted that, specifically, the step of determining the coordinate sequences of multiple lane centerlines based on the scene data further includes steps A110 to A140: Step A110: determining vehicle position information based on the scene data, wherein the vehicle position information includes vehicle coordinates and vehicle heading angle; It should be noted that the vehicle position information generally includes the vehicle's coordinates (x, y) and the vehicle's heading angle. Specifically, based on the scene data, the steps of determining the vehicle position information include: First, the intersection lane scene recognition device collects all information about the vehicle's position from the scene data, including the vehicle's GPS coordinates, the vehicle's IMU (inertial measurement unit) data, the vehicle's wheel speed information, the vehicle's steering angle, etc.

[0048] It is understandable that since a single data source may not be accurate enough, intersection lane scene recognition equipment uses data fusion technology to improve the accuracy of location information. Common data fusion technologies include Kalman filtering, particle filtering, etc.

[0049] Furthermore, the intersection lane scene recognition device uses GPS data to obtain the vehicle's global coordinates (longitude, latitude, altitude).

[0050] The vehicle's heading angle can be determined using magnetometer data from the IMU, or estimated using the vehicle's steering angle and trajectory. If the scene data includes IMU data, the intersection and lane scene recognition device can directly use the magnetometer data to determine the heading angle. If IMU data is not available, the intersection and lane scene recognition device may need to use other methods (such as visual odometry, wheel speedometer, etc.) to estimate the heading angle.

[0051] Finally, the intersection lane scene recognition device obtains the vehicle position information including the vehicle coordinates and heading angle.

[0052] Step A120: constructing a vehicle coordinate system based on the vehicle position information; It's important to note that constructing a vehicle coordinate system (VCS) is a crucial step in autonomous driving and vehicle navigation. It helps understand and control the vehicle's position and motion in space. A VCS is typically a two-dimensional coordinate system with the center of the vehicle's rear axle as its origin, the x-axis pointing forward, and the y-axis pointing to the left. The following are the steps for constructing a VCS: It is understandable that since the vehicle is moving, the vehicle coordinate system needs to be updated in real time to reflect the vehicle's latest position and orientation, which is usually achieved through the vehicle's sensor data (such as GPS, IMU, wheel speedometer, etc.).

[0053] Specifically, the step of constructing a vehicle coordinate system based on the vehicle position information further includes steps A121 to A122: Step A121, extracting the vehicle coordinates and vehicle heading angle from the vehicle position information; Step A122: Set the vehicle coordinates as the origin, set the direction corresponding to the vehicle heading angle as the positive direction of the y-axis, and construct a vehicle coordinate system corresponding to each set of local perception type map data based on the origin and the positive direction of the y-axis.

[0054] It is understandable that the lane centerline may be a curved curve. In this case, a reference coordinate is required to save the coordinates of the lane centerline in sequence in the coordinate sequence. The vehicle travels in the same direction in the lane over time. Using the vehicle as the origin to construct a coordinate system can provide a basis for saving the coordinates of the lane centerline in sequence in the coordinate sequence in the subsequent steps.

[0055] Step A130: obtaining multiple coordinates corresponding to the centerline of each lane, and determining a first distance from each coordinate to the vehicle based on the vehicle coordinate system; It should be noted that the multiple coordinates corresponding to each lane centerline are not obtained based on the vehicle coordinate system.

[0056] In step A140 , each coordinate is saved in a coordinate sequence in ascending order based on the first distance, to obtain multiple sets of lane centerline coordinate sequences.

[0057] Step A200: construct a global grid, map each coordinate in the coordinate sequence to the global grid, and obtain a global perception map of preset accuracy, so as to determine the vehicle coordinates of the vehicle in the future frame based on the global perception map.

[0058] Specifically, the step of constructing a global grid, mapping each coordinate in the coordinate sequence to the global grid, and obtaining a global perception map of preset accuracy includes steps A210 to A250: Step A210: determining a lane centerline boundary value based on the coordinate sequence, and constructing a global grid based on the boundary value; It should be noted that the size of the global grid is associated with the boundary value of the lane centerline. For example, if the horizontal coordinate range of the boundary value is [0, 5] and the vertical coordinate range of the boundary value is [0, 5], then the size of the global grid is 5x5.

[0059] Step A220, mapping each coordinate in the coordinate sequence to the global grid to obtain a target grid; Step A230: Based on the target grid, searching for the first lane centerline and the second lane centerline in an overlapping area, determining the number of overlaps between the first lane centerline and the second lane centerline, and determining whether the number of overlaps is greater than or equal to a preset overlap threshold. Step A240: If the number of overlaps is greater than or equal to a preset overlap threshold, the first lane centerline and the second lane centerline are added to the same lane centerline set; or, if the number of overlaps is less than the preset overlap threshold, the first lane centerline and the second lane centerline are added to different lane centerline sets. It should be noted that in the perception map, for the same lane line, the perception results of the same sensor at different times are different. For example, at the first moment, the perception result corresponding to lane line 1 is lane line 1, while at the second moment, the perception result corresponding to lane line becomes lane line 2, which reduces the accuracy of the perception animation.

[0060] It can be understood that if the number of overlaps between the center line of the first lane and the center line of the second lane is greater than or equal to the preset overlap number threshold, it indicates that the two lane lines may be the same lane line, and then, the first lane center line and the second lane center line are placed in the same lane center line set.

[0061] Step A250: Determine the perception map corresponding to the current frame based on the set of all lane centerlines.

[0062] It is understandable that by putting lane centerlines with a high degree of overlap into the same lane centerline set, a more accurate perception map can be obtained.

[0063] Based on the first and second embodiments of the present application, in the third embodiment of the present application, the same or similar contents as those in the first and second embodiments can be referred to above and will not be described in detail. Figure 3 The step of mapping each coordinate in the coordinate sequence to the global grid to obtain a target grid further includes any one of steps A221 to A222: Step A221, mapping each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, performing a first thinning operation on the initial grid to obtain a target grid; Step A222: Map each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, perform a first thinning operation on the initial grid, and perform a second thinning operation on the grid after the first thinning operation to obtain a target grid, wherein the corresponding thinning operation includes a floor operation and a rounding operation.

[0064] Taking rounding down as an example, first, use the int (rounding down) operator to convert the coordinate point to the corresponding position (temp_x, temp_y) in the grid, where temp_x = int(temp_x) - int(x_min), temp_y = int(temp_y) - int(y_min), where (temp_x, temp_y) represents the grid position corresponding to the previous coordinate point.

[0065] Construct a list polyline to store the new lane centerline coordinates after thinning. If the list polyline is empty, add point to the list polyline and set the (temp_x, temp_y) position in the grid to 1. No further judgment is made on the coordinate point, where point represents the coordinate point on the lane centerline. If the list polyline is not empty, further judgment is required on the coordinate point. The specific judgment steps are as follows: Check the grid position (temp_x, temp_y) corresponding to the current coordinate point. If it is 0, it means that the coordinate point appears at this grid position (temp_x, temp_y) for the first time, and the position is set to 1 to mark it. If it is 1, it means that the coordinate point has already been marked and it is skipped. If the coordinate point appears at this grid position (temp_x, temp_y) for the first time, further determination is required to determine whether the coordinate point should be retained.

[0066] Furthermore, if the distance between the current coordinate point and the last point added to the polyline list is less than 1, the two points are considered too close and the current point can be ignored, thus achieving thinning. Otherwise, the current coordinate point is added to the polyline list and marked in the grid.

[0067] It is understandable that in path planning, overly dense path points will not only increase the burden of calculation and storage, but may also lead to path redundancy and reduced efficiency. Therefore, thinning processing is required while retaining the shape characteristics and trajectory information of the path to ensure the accuracy and real-time performance of path planning.

[0068] In this embodiment, the amount of calculation is reduced and the calculation efficiency is improved by thinning, thereby ensuring real-time performance.

[0069] It should be noted that the above examples are only used to understand this application and do not constitute a limitation on the intersection lane scene recognition method of this application. More simple transformations based on this technical concept are all within the scope of protection of this application.

[0070] This application also provides a road lane scene recognition device, please refer to Figure 4 , the intersection lane scene recognition device includes: An acquisition module 10, configured to acquire scene data; a first determining module 20 configured to traverse frames containing separated lane elements or merged lane elements in the scene data, use the traversed frames as current frames, and determine a corresponding perception map based on the current frames; A second determination module 30 is used to determine the starting frame and the ending frame corresponding to the lane centerline set in the perception map, and determine the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set based on the starting frame, the ending frame and the perception map, wherein the intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set contains two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

[0071] In one embodiment, the second determining module includes: a first calculation unit, configured to calculate, based on the start frame, the end frame, and the perception map, a first projection distance from the vehicle to the center line of the first lane, and obtain the first projection distance corresponding to each frame from the start frame to the end frame; and calculate a second projection distance from the vehicle to the center line of the second lane, and obtain the second projection distance corresponding to each frame from the start frame to the end frame; a second calculation unit, configured to calculate the sum of the first projection distance and the second projection distance corresponding to the same frame to obtain multiple sets of third projection distances; a third calculating unit, configured to calculate an average value of the plurality of sets of third projection distances, determine a difference between the third projection distance corresponding to the start frame and the average value, and determine a difference between the third projection distance corresponding to the end frame and the average value; The first determination unit is used to determine that the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a merging lane scene if the third projection distance corresponding to the starting frame is greater than the average value and the third projection distance corresponding to the ending frame is less than the average value; if the third projection distance corresponding to the starting frame is less than the average value and the third projection distance corresponding to the ending frame is greater than the average value, then the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a separating lane scene.

[0072] In one embodiment, the first determining module includes: a second determining unit, configured to determine a plurality of sets of lane centerline coordinate sequences based on the scene data; The first construction unit is configured to construct a global grid, map each coordinate in the coordinate sequence to the global grid, and determine a perception map corresponding to the current frame.

[0073] In one embodiment, the first determining module further includes: a third determining unit, configured to determine a boundary value of a lane centerline based on the coordinate sequence, and construct a global grid based on the boundary value; A first mapping unit is configured to map each coordinate in the coordinate sequence to the global grid to obtain a target grid; a search unit configured to search, based on the target grid, for a third lane centerline and a fourth lane centerline having an overlapping area, determine a number of overlaps between the third lane centerline and the fourth lane centerline, and determine whether the number of overlaps is greater than or equal to a preset overlap threshold; an adding unit, configured to add the third lane centerline and the fourth lane centerline to the same lane centerline set if the number of overlaps is greater than or equal to a preset overlap number threshold; or to add the third lane centerline and the fourth lane centerline to different lane centerline sets if the number of overlaps is less than the preset overlap number threshold; The fourth determining unit is configured to determine a perception map corresponding to the current frame based on a set of all lane centerlines.

[0074] In one embodiment, the first determining module further includes: a second mapping unit, configured to map each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, and perform a first thinning operation on the initial grid to obtain a target grid; The third mapping unit is used to map each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, perform a first thinning operation on the initial grid, and perform a second thinning operation on the grid after the first thinning operation to obtain a target grid, wherein the corresponding thinning operation includes a rounding-down operation.

[0075] In one embodiment, the first determining module further includes: a fifth determining unit, configured to determine vehicle position information based on the scene data, wherein the vehicle position information includes vehicle coordinates and a vehicle heading angle; A second construction unit is configured to construct a vehicle coordinate system based on the vehicle position information; a sixth determining unit, configured to obtain a plurality of coordinates corresponding to a centerline of each lane, and determine a first distance from each coordinate to the vehicle based on the vehicle coordinate system; The storage unit is used to save each coordinate in a coordinate sequence in ascending order based on the first distance to obtain multiple sets of lane centerline coordinate sequences.

[0076] In one embodiment, the first determining module further includes: an extraction unit, configured to extract the vehicle coordinates and the vehicle heading angle from the vehicle position information; A setting unit is used to set the vehicle coordinates as the origin, set the direction corresponding to the vehicle heading angle as the positive direction of the y-axis, and construct a vehicle coordinate system corresponding to each set of local perception type map data based on the origin and the positive direction of the y-axis.

[0077] The lane intersection scene recognition device provided in this application utilizes the lane intersection scene recognition method described in the aforementioned embodiment to address the technical challenges of lane intersection scene recognition. Compared to the prior art, the lane intersection scene recognition device provided in this application achieves the same beneficial effects as the lane intersection scene recognition method described in the aforementioned embodiment. Other technical features of the lane intersection scene recognition device are the same as those disclosed in the aforementioned embodiment and are not further elaborated here.

[0078] The present application provides a lane intersection scene recognition device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the lane intersection scene recognition method in the above-mentioned embodiment one.

[0079] Reference below Figure 5 , which shows a schematic diagram of the structure of a lane intersection scene recognition device suitable for implementing embodiments of the present application. The lane intersection scene recognition device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The intersection lane scene recognition device shown is merely an example and should not impose any limitations on the functions and scope of use of the embodiments of the present application.

[0080] like Figure 5As shown, the intersection lane scene recognition device may include a processing device 1001 (e.g., a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the intersection lane scene recognition device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to I / O interface 1006: input devices 1007, such as a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008, such as a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003, such as a magnetic tape or hard disk; and communication device 1009. The communication device 1009 can allow the intersection and lane scene recognition device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an intersection and lane scene recognition device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented or have alternatively.

[0081] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0082] The lane intersection scene recognition device provided in this application utilizes the lane intersection scene recognition method described in the aforementioned embodiment to address the technical challenges of lane intersection scene recognition. Compared to the prior art, the lane intersection scene recognition device provided in this application achieves the same beneficial effects as the lane intersection scene recognition method described in the aforementioned embodiment. Other technical features of the lane intersection scene recognition device are the same as those disclosed in the aforementioned embodiment and are not further elaborated here.

[0083] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0084] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0085] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, a computer program) stored thereon, and the computer-readable program instructions are used to execute the intersection lane scene recognition method in the above-mentioned embodiment.

[0086] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0087] The computer-readable storage medium may be included in the intersection and lane scene recognition device; or it may exist independently without being incorporated into the intersection and lane scene recognition device.

[0088] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the intersection and lane scene recognition device, the intersection and lane scene recognition device: Get scene data; Traversing frames in the scene data that contain separated lane elements or merged lane elements, taking the traversed frames as current frames, and determining a corresponding perception map based on the current frames; Determine a start frame and an end frame corresponding to a lane centerline set in the perception map, and determine, based on the start frame, the end frame, and the perception map, an intersection lane scene corresponding to a first lane centerline and a second lane centerline in the lane centerline set, wherein the intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set includes two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

[0089] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0090] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0091] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0092] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned intersection lane scene recognition method, thereby resolving the technical issues surrounding intersection lane scene recognition. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided in this application are similar to those of the intersection lane scene recognition method provided in the aforementioned embodiments, and are not further elaborated here.

[0093] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned intersection lane scene recognition method when executed by a processor.

[0094] The computer program product provided in this application can solve the technical problem of lane scene recognition at intersections. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the lane scene recognition method provided in the above embodiment, and will not be repeated here.

[0095] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for identifying lane scenes at intersections, characterized in that: The intersection lane scene recognition method includes: Get scene data; Traversing frames in the scene data that contain separated lane elements or merged lane elements, taking the traversed frames as current frames, and determining a corresponding perception map based on the current frames; Determine a start frame and an end frame corresponding to a lane centerline set in the perception map, and determine, based on the start frame, the end frame, and the perception map, an intersection lane scene corresponding to a first lane centerline and a second lane centerline in the lane centerline set, wherein the intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set includes two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

2. The method for identifying lane intersection scenes according to claim 1, wherein: The step of determining, based on the start frame, the end frame, and the perception map, the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set, includes: Calculating, based on the start frame, the end frame, and the perception map, a first projection distance from the vehicle to the centerline of the first lane to obtain the first projection distance corresponding to each frame from the start frame to the end frame, and calculating a second projection distance from the vehicle to the centerline of the second lane to obtain the second projection distance corresponding to each frame from the start frame to the end frame; Calculating the sum of the first projection distance and the second projection distance corresponding to the same frame to obtain multiple sets of third projection distances; Calculating an average value of the plurality of sets of third projection distances, determining a difference between the third projection distance corresponding to the start frame and the average value, and determining a difference between the third projection distance corresponding to the end frame and the average value; If the third projection distance corresponding to the starting frame is greater than the average value, and the third projection distance corresponding to the ending frame is less than the average value, it is determined that the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a merging lane scene; if the third projection distance corresponding to the starting frame is less than the average value, and the third projection distance corresponding to the ending frame is greater than the average value, then the intersection lane scene corresponding to the first lane centerline and the second lane centerline is a separating lane scene.

3. The method for identifying lane intersection scenes according to claim 1, wherein: The step of determining the corresponding perception map based on the current frame includes: Determining a plurality of lane centerline coordinate sequences based on the scene data; A global grid is constructed, each coordinate in the coordinate sequence is mapped to the global grid, and a perception map corresponding to the current frame is determined.

4. The method for identifying lane intersection scenes according to claim 3, wherein: The steps of constructing a global grid, mapping each coordinate in the coordinate sequence to the global grid, and determining a perception map corresponding to the current frame include: Determining a lane centerline boundary value based on the coordinate sequence, and constructing a global grid based on the boundary value; Mapping each coordinate in the coordinate sequence to the global grid to obtain a target grid; Based on the target grid, searching for the centerline of the third lane and the centerline of the fourth lane where there is an overlapping area, determining the number of overlaps between the centerline of the third lane and the centerline of the fourth lane, and determining whether the number of overlaps is greater than or equal to a preset overlap number threshold; If the number of overlaps is greater than or equal to a preset overlap threshold, the third lane centerline and the fourth lane centerline are added to the same lane centerline set; or, if the number of overlaps is less than the preset overlap threshold, the third lane centerline and the fourth lane centerline are added to different lane centerline sets respectively. Based on the set of all lane centerlines, determine the perception map corresponding to the current frame.

5. The method for identifying lane intersection scenes according to claim 4, wherein: The step of mapping each coordinate in the coordinate sequence to the global grid to obtain a target grid further includes any one of the following: Mapping each coordinate in the coordinate sequence to a corresponding position in the grid to obtain an initial grid, and performing a first thinning operation on the initial grid to obtain a target grid; Each coordinate in the coordinate sequence is mapped to a corresponding position in the grid to obtain an initial grid, a first thinning operation is performed on the initial grid, and a second thinning operation is performed on the grid after the first thinning operation to obtain a target grid, wherein the corresponding thinning operation includes a floor operation.

6. The method for identifying lane intersection scenes according to claim 3, wherein: The step of determining a plurality of lane centerline coordinate sequences based on the scene data further includes: Determining vehicle position information based on the scene data, wherein the vehicle position information includes vehicle coordinates and a vehicle heading angle; constructing a vehicle coordinate system based on the vehicle position information; Obtaining a plurality of coordinates corresponding to the centerline of each lane, and determining a first distance from each coordinate to the vehicle based on the vehicle coordinate system; Each coordinate is saved in a coordinate sequence in ascending order based on the first distance, to obtain multiple sets of lane centerline coordinate sequences.

7. The method for identifying lane intersection scenes according to claim 6, wherein: The step of constructing a vehicle coordinate system based on the vehicle position information further includes: Extracting vehicle coordinates and vehicle heading angle from the vehicle position information; The vehicle coordinates are set as the origin, the direction corresponding to the vehicle heading angle is set as the positive direction of the y-axis, and based on the origin and the positive direction of the y-axis, a vehicle coordinate system corresponding to each set of local perception type map data is constructed.

8. A road intersection lane scene recognition device, characterized in that: The intersection lane scene recognition device includes: An acquisition module, configured to acquire scene data; a first determining module configured to traverse frames containing separated lane elements or merged lane elements in the scene data, use the traversed frames as current frames, and determine a corresponding perception map based on the current frames; A second determination module is used to determine the starting frame and the ending frame corresponding to the lane centerline set in the perception map, and determine the intersection lane scene corresponding to the first lane centerline and the second lane centerline in the lane centerline set based on the starting frame, the ending frame and the perception map, wherein the intersection lane scene includes a separated lane scene and a merged lane scene, the lane centerline set contains two or more groups of lane centerlines with overlapping areas, and the number of overlaps between the first lane centerline and the second lane centerline is greater than or equal to a preset overlap number threshold.

9. A road intersection lane scene recognition device, characterized in that: The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the intersection lane scene recognition method according to any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the intersection lane scene recognition method according to any one of claims 1 to 8 are implemented.

11. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the intersection lane scene recognition method according to any one of claims 1 to 8 are implemented.

Images