Cuten behavior identification method, device and equipment and storage medium

Abstract

The application discloses a cutin behavior recognition method and device, equipment and a storage medium, relates to the technical field of vehicle behavior recognition, and discloses a cutin behavior recognition method, which comprises the following steps: screening effective segments under a no-map LCC scene from acquired historical driving data; screening a to-cut-in vehicle that wants to cut into a lane of an ego vehicle from a current frame of the effective segments; identifying a bound lane where the to-cut-in vehicle is located in a reference frame of the effective segments based on a preset global map, wherein the preset global map comprises lanes where each vehicle is located at different frame moments; and judging whether the to-cut-in vehicle has a cutin behavior based on the bound lane where the to-cut-in vehicle is located in different reference frames. That is, the lane where the to-cut-in vehicle is located in different frame data is accurately determined by using the global map, and whether the vehicle in the historical driving data has a cutin behavior is accurately recognized by the lane where the to-cut-in vehicle is located at different moments.

Description

Technical Field

[0001] This application relates to the field of vehicle behavior recognition technology, and in particular to cutin behavior recognition methods, devices, equipment and storage media. Background Technology

[0002] Cutin is the behavior of a vehicle cutting into its own lane. Because cutting in can be dangerous, it is necessary to accurately identify cutin behavior from the vehicle's historical driving data in order to improve vehicle optimization.

[0003] Currently, to ensure accurate identification of cut-in behavior from historical driving data, high-precision images of surrounding vehicles are typically obtained first. These images are then used to extract information such as the position, speed, and acceleration of the surrounding vehicles, and corresponding algorithms are applied to determine whether a cut-in behavior has occurred. However, in map-less LCC scenarios, each frame of map information in the historical driving data is generated in real-time by perception, and the same lane centerline appears differently in each frame. Therefore, accurate identification of vehicle cut-in behavior is impossible in map-less LCC scenarios.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a cutin behavior recognition method, which aims to solve the technical problem of not being able to recognize vehicle cutin behavior in mapless LCC scenarios.

[0006] To achieve the above objectives, this application proposes a cutin behavior recognition method, which includes:

[0007] Filter valid segments from the acquired historical driving data for LCC scenarios without maps;

[0008] Select vehicles to be switched into the vehicle lane from the current frame of the effective segment;

[0009] Based on a preset global map, the bound lane of the vehicle to be cut into is identified in the reference frame of the effective segment. The preset global map includes the lanes in which each vehicle is located at different frame times.

[0010] Based on the bound lane where the vehicle to be switched in is located in different reference frames, it is determined whether the vehicle to be switched in has performed a cutin action.

[0011] In one embodiment, the step of identifying the bound lane where the vehicle to be switched in is located in the reference frame of the effective segment based on a preset global map includes:

[0012] Find the nearest valid frame preceding the current frame from the valid segments;

[0013] Based on the preset global map, the first bound lane where the vehicle to be cut into is located in the current frame, and the second bound lane where it is located in the most recent valid frame are obtained;

[0014] Determine whether the first bound lane and the second bound lane are the same lane;

[0015] If the first bound lane and the second bound lane are not the same lane, then based on the preset global map, the bound lanes in each reference frame after the current frame in the effective segment are obtained, and the bound lanes in each reference frame before the most recent effective frame in the effective segment are also obtained.

[0016] In one embodiment, the step of determining whether the vehicle to be switched in has performed a cutin action based on the bound lane where the vehicle to be switched in is located in different reference frames includes:

[0017] Based on the bound lane of the vehicle to be switched into in different reference frames, the first percentage of all frames in the reference frames after the current frame whose bound lane is the first bound lane is calculated.

[0018] And, the second percentage of all frames in the reference frames preceding the most recent valid frame in which the bound lane is the second bound lane;

[0019] Determine whether both the first percentage and the second percentage satisfy the preset percentage;

[0020] If all the preset ratios are met, then the vehicle to be switched in is determined to have performed a cutin action.

[0021] In one embodiment, the step of filtering vehicles to be switched into the vehicle lane from the current frame of the valid segment includes:

[0022] Extract candidate vehicles for cutin from the current frame of the valid segment;

[0023] Obtain the azimuth angle of the vehicle in the current frame and the position coordinates of the vehicle mapped in the global map in the current frame;

[0024] Based on the location information of the current cutin candidate vehicle and the location coordinates, the relative lateral displacement and relative longitudinal displacement between the current cutin candidate vehicle and the vehicle are determined, and the current cutin candidate vehicle is the cutin candidate vehicle that is currently in the current position.

[0025] Determine whether the relative longitudinal displacement is less than the currently recorded minimum longitudinal displacement, and determine whether the relative lateral displacement meets a preset threshold.

[0026] If the relative lateral displacement is less than the minimum longitudinal displacement, and the relative longitudinal displacement meets a preset threshold, then the current cutin candidate vehicle is used to update the currently recorded cutin vehicle until all cutin candidate vehicles are judged, and the cutin candidate vehicle with the smallest relative lateral displacement is determined as the cutin vehicle.

[0027] In one embodiment, before the step of filtering vehicles to be switched into the vehicle lane from the current frame of the valid segment, the method further includes:

[0028] The coordinates of the lane centerline in each frame of data are mapped to the global grid to obtain an initial global map, and the data in each frame constitutes the effective segment;

[0029] The overlapping lanes selected from the initial global map are merged to obtain a global grid map;

[0030] Map each object in the effective segment to the global grid map to obtain the index position of each object in the global grid map in each frame of data;

[0031] Based on the index position, each object in each frame of data is bound to the global lane centerline in the global grid map to obtain a preset global map.

[0032] In one embodiment, the step of binding each object in each frame of data to the global lane centerline in the global grid map based on the index position to obtain a preset global map includes:

[0033] Based on the index location, determine the target lane corresponding to the object in the global grid map;

[0034] Determine the number of objects present in the target lane in each frame of data;

[0035] If the number of objects is not one, then determine the projected distance of the object on the global lane centerline of the target lane;

[0036] By binding the objects in each frame of data with the global lane centerline and projection distance of the corresponding target lane, a preset global map is obtained.

[0037] In one embodiment, the step of merging overlapping lanes selected from the initial global map to obtain a global grid map includes:

[0038] Filter overlapping lanes with overlapping areas from the initial global map;

[0039] The overlapping lanes are merged to obtain the global lane centerline of each lane in the effective segment in the initial global map;

[0040] The centerline coordinates of the global lane centerline are thinned at least once to obtain the thinned centerline coordinates.

[0041] Based on the thinned centerline coordinates, the global lane centerlines are gridded to obtain a global grid map.

[0042] Furthermore, to achieve the above objectives, this application also proposes a cutin behavior recognition device, which includes:

[0043] The acquisition module is used to filter valid segments in mapless LCC scenarios from the acquired historical driving data;

[0044] The filtering module is used to filter vehicles to be cut into the vehicle lane from the current frame of the effective segment;

[0045] The identification module is used to identify the bound lane of the vehicle to be cut into in the reference frame of the effective segment based on a preset global map, wherein the preset global map includes the lanes in which each vehicle is located at different frame times.

[0046] The judgment module is used to determine whether the vehicle to be cut in has performed a cutin action based on the bound lane in different reference frames.

[0047] In addition, to achieve the above objectives, this application also proposes a Cutin behavior recognition device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the Cutin behavior recognition method as described above.

[0048] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the cutin behavior recognition method described above.

[0049] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the cutin behavior recognition method described above.

[0050] One or more technical solutions proposed in this application have at least the following technical effects:

[0051] Because historical driving data contains a large number of HD format frame fragments, it is necessary to first filter out valid fragments for LCC scenarios without images from the historical driving data. Furthermore, since there may be multiple vehicles near the vehicle, it is necessary to filter out the vehicles to be cut into the vehicle's lane from the current frame of the valid fragments. Then, using a pre-created global map (preset global map), the bound lanes of the vehicles to be cut into in different frame data are determined. Since the global map maps the lane information of each vehicle at different frame times, the lanes of the vehicles to be cut into in different frame data can be accurately determined using the global map. Finally, by using the lanes of the vehicles to be cut into at different times, it is possible to accurately identify whether vehicles in the historical driving data have performed cutin behavior. Attached Figure Description

[0052] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0053] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a flowchart illustrating an embodiment of the cutin behavior recognition method of this application.

[0055] Figure 2 This is a flowchart illustrating Embodiment 2 of the cutin behavior recognition method of this application;

[0056] Figure 3 A flowchart is provided for implementation three of the cutin behavior recognition method in this application;

[0057] Figure 4 This is a schematic diagram of the module structure of the cutin behavior recognition device according to an embodiment of this application;

[0058] Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the cutin behavior recognition method in this application embodiment.

[0059] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0060] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0061] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0062] The main solution of this application embodiment is as follows: the data analysis device filters valid segments in mapless LCC scenarios from the acquired historical driving data; filters vehicles to be cut into the lane from the current frame of the valid segments; based on a preset global map, identifies the bound lane of the vehicle to be cut into in the reference frame of the valid segment, the preset global map including the lanes of each vehicle at different frame times; and determines whether the vehicle to be cut into has performed a cutin behavior based on the bound lane of the vehicle to be cut into in different reference frames.

[0063] In this embodiment, for ease of description, the following description uses a data analysis device as the execution subject.

[0064] Currently, to ensure accurate identification of cut-in behavior from historical driving data, high-precision images of surrounding vehicles are typically obtained first. These images are then used to extract information such as the position, speed, and acceleration of these vehicles, and corresponding algorithms are applied to determine whether a cut-in has occurred. However, in map-less LCC scenarios, each frame of map information in the historical driving data is generated in real-time by perception, and the same lane centerline appears differently in each frame. Therefore, in map-less LCC scenarios, accurate identification of vehicle cut-in behavior is impossible.

[0065] This application provides a solution whereby, due to the large number of HD format frame fragments in historical driving data, it is necessary to first filter out valid fragments for mapless LCC scenarios from the historical driving data. Furthermore, since multiple vehicles may be near the vehicle, it is necessary to filter out vehicles to be cut into the vehicle's lane from the current frame of the valid fragments. Then, using a pre-created global map (preset global map), the bound lanes of the vehicles to be cut into in different frame data are determined. Since the global map maps the lane information of each vehicle at different frame times, the lanes of the vehicles to be cut into in different frame data can be accurately determined using the global map. Finally, by using the lanes of the vehicles to be cut into at different times, it is possible to accurately identify whether vehicles in the historical driving data have performed cutin behavior.

[0066] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or data analysis device capable of performing the above functions. The following description uses a data analysis device as an example to illustrate this embodiment and the subsequent embodiments.

[0067] Based on this, embodiments of this application provide a cutin behavior recognition method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the cutin behavior recognition method of this application.

[0068] In this embodiment, the cutin behavior recognition method includes steps S10 to S40:

[0069] Step S10: Filter valid segments in the mapless LCC scenario from the acquired historical driving data;

[0070] It should be noted that historical driving data refers to fragments of environmental information surrounding the vehicle obtained from the vehicle's previously recorded driving data, and this historical driving data consists of multiple frames of images. Since each frame of data is generated in real time by perception in the imageless LCC scenario, even the same lane centerline will appear differently in each frame.

[0071] Understandably, since HD format images are different from the format of each frame of data in the effective segment, there will be differences between the information of each vehicle. Furthermore, mapping the image data of the two formats to the global grid requires a complex conversion process, which will reduce the efficiency of cutin behavior recognition. Therefore, HD format images should be avoided when searching for the most recent effective frame.

[0072] Step S20: Select vehicles to be switched into the vehicle lane from the current frame of the effective segment;

[0073] It should be noted that the current frame is the frame data being processed within the valid segment. The "own vehicle lane" is the lane that the vehicle is currently traveling in. The vehicle to be cut in is the vehicle that needs to change lanes into the own vehicle lane and is adjacent to the own vehicle changing lanes. Among them, "own vehicle" refers to the vehicle that provides historical driving data.

[0074] Understandably, since there is more than one vehicle in the current frame that needs to change lanes to the driver's lane, it is necessary to select the vehicle most likely to cut into the driver's lane from all the vehicles in the current frame in order to avoid performing cut-in confirmation detection on too many vehicles, thereby improving the efficiency of cutin behavior recognition.

[0075] Furthermore, step S20 also includes:

[0076] Extract candidate vehicles for cutin from the current frame of the valid segment;

[0077] Obtain the azimuth angle of the vehicle in the current frame and the position coordinates of the vehicle mapped in the global map in the current frame;

[0078] Based on the location information of the current cutin candidate vehicle and the location coordinates, the relative lateral displacement and relative longitudinal displacement between the current cutin candidate vehicle and the vehicle are determined, and the current cutin candidate vehicle is the cutin candidate vehicle that is currently in the current position.

[0079] Determine whether the relative longitudinal displacement is less than the currently recorded minimum longitudinal displacement, and determine whether the relative lateral displacement meets a preset threshold.

[0080] If the relative lateral displacement is less than the minimum longitudinal displacement, and the relative longitudinal displacement meets a preset threshold, then the current cutin candidate vehicle is used to update the currently recorded cutin vehicle until all cutin candidate vehicles are judged, and the cutin candidate vehicle with the smallest relative lateral displacement is determined as the cutin vehicle.

[0081] It should be noted that a cutin candidate vehicle is a vehicle in the current frame that may cut into the lane of the vehicle being used from the current lane or other lanes. The azimuth angle is the vehicle's orientation angle in the current frame. It can be the angle value relative to a fixed direction (such as due north) based on the vehicle's direction of travel. The position coordinates are the vehicle's current position mapped in the global map, represented in two-dimensional or three-dimensional coordinates, thus reflecting the vehicle's specific position in space. The position information is the current position of the current cutin candidate vehicle mapped in the global map. The relative lateral displacement is the lateral distance between the current cutin candidate vehicle and the vehicle being used, reflecting the candidate vehicle's relative position in the lateral direction. The relative longitudinal displacement is the longitudinal distance between the current cutin candidate vehicle and the vehicle being used, reflecting the candidate vehicle's relative position in the longitudinal direction. The minimum longitudinal displacement is the value with the smallest relative longitudinal displacement among all cutin candidate vehicles. The preset threshold is the maximum lateral distance left between the current cutin candidate vehicle and the vehicle being used when cutting into the vehicle's lane; that is, the maximum lateral distance for determining whether the current cutin candidate vehicle has cut into the vehicle's lane.

[0082] Understandably, by accurately calculating the relative lateral and longitudinal displacements between vehicles, it is possible to more accurately identify the vehicle most likely to cut into the driving lane. Since the relative position and dynamic changes between vehicles are taken into account, the accuracy of cutin behavior recognition is improved.

[0083] In the specific implementation, the cutin candidate vehicle index `cutin_candidate` in the current frame is calculated. The position information of each cutin candidate vehicle is compared to determine which cutin candidate vehicle is most likely to complete the cut-in action. Specifically, a temporary minimum value (`local_x_temp`) of the longitudinal displacement of the object in the ego coordinate system can be defined and initialized to a large value of 10000; a threshold (`local_y_threshold`) for determining whether the object is inside the lane can be defined and set to 2.8 meters; the object index (`cutin_candidate`) of the current cutin candidate vehicle most likely to cut into the lane can be defined and initialized to -1. After completing the above preparatory work, the azimuth angle `ego_theta` of the ego vehicle in the current frame, the lane index `ego_laneid`, and the position coordinates `ego_x` and `ego_y` in the global map of the scene are obtained; the object index list `obj_id_list` of the current frame is traversed, and each `obj_id_` is determined... If `obj_id` in the list is equal to 0, and 0 is the index of the vehicle, then skip the current object. If `obj_id` is not equal to 0, retrieve the lane bound to the current cutin candidate vehicle's `obj_id` in the global map. If `obj_laneid` is not equal to `ego_laneid`, then it is determined that the two are not in the same lane, so the current cutin candidate vehicle needs to be skipped. The next cutin candidate vehicle is then selected as the current vehicle. If `obj_laneid` equals `ego_laneid`, then obtain the current object's position coordinates `obj_x` and `obj_y`, and further calculate the position (`local_x`, `local_y`) of the current cutin candidate vehicle's `obj_id` relative to the vehicle in the current frame, where `curr_x` is the longitudinal displacement and `curr_y` is the lateral displacement. The specific calculation process is as follows:

[0084] x_diff=global_x-ego_x

[0085] y_diff=global_y-ego_y

[0086] angle_cos = math.cos(ego_theta)

[0087] angle_sin = math.sin(ego_theta)

[0088] local_x=x_diff*angle_cos-y_diff*angle_sin

[0089] local_y=x_diff*angle_sin+y_diff*angle_cos.

[0090] After determining the position of the current cutin candidate vehicle relative to its own vehicle (relative lateral displacement (local_y) and relative longitudinal displacement (local_x)), the next step is to assess the likelihood of the current cutin candidate vehicle cutting in. Specifically, this involves checking if local_x is greater than 0 and less than the currently recorded minimum longitudinal displacement local_x_temp. Simultaneously, it checks if the absolute value of local_y is less than the set threshold local_y_threshold. If both conditions are met, the current cutin candidate vehicle is considered closer to completing the cutin action than previous candidates. Therefore, cutin_candidate needs to be updated to the current cutin candidate vehicle index obj_id, and local_x_temp needs to be updated to the current value of local_x.

[0091] Step S30: Based on a preset global map, identify the bound lane where the vehicle to be switched is located in the reference frame of the effective segment. The preset global map includes the lanes where each vehicle is located at different frame times.

[0092] It should be noted that the preset global map is obtained by mapping the centerline coordinates of each lane in the effective segment to the positions of each object in the effective segment onto the global grid. Therefore, the preset global map includes the lanes where each vehicle is located at different frame times. The reference frame is frame data used to determine whether the vehicle to be cut in will perform a cutin action. The bound lanes are the lanes mapped to each vehicle in the global network and bound to the corresponding vehicle.

[0093] Understandably, in mapless LCC scenarios, sensors can typically only perceive the current behavior information of surrounding vehicles. The behavior information around the vehicle in the effective frame segment is the behavior information perceived by the vehicle at that moment. Therefore, it is necessary to create a global map that can accurately represent the position and lane of each vehicle. By using the global map and detailed effective segments, the bound lane of the vehicle to be cut in can be accurately identified, thereby improving the accuracy of cutin behavior recognition.

[0094] Step S40: Based on the bound lane where the vehicle to be cut in is located in different reference frames, determine whether the vehicle to be cut in has performed a cutin action.

[0095] Understandably, by identifying the bound lanes of the vehicle to be cut into in different reference frames, it is possible to determine whether the vehicle to be cut into changes lanes to its own lane in the effective segment, thereby accurately determining whether the vehicle to be cut into has engaged in cutin behavior. This enables accurate identification of cutin behavior even in the effective segment of a mapless LCC scenario.

[0096] This embodiment provides a cutin behavior recognition method. Since historical driving data contains a large number of HD format frame segments, it is necessary to first filter out valid segments in mapless LCC scenarios from the historical driving data. Furthermore, since multiple vehicles may be near the vehicle, it is necessary to filter out vehicles to be cut into the vehicle's lane from the current frame of the valid segments. Then, using a pre-created global map (preset global map), the bound lane of the vehicle to be cut into in different frame data is determined. Because the global map maps the lane information of each vehicle at different frame times, the lane of the vehicle to be cut into in different frame data can be accurately determined using the global map. Finally, by using the lane of the vehicle to be cut into at different times, it is possible to accurately identify whether a vehicle in the historical driving data has performed a cutin behavior.

[0097] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 Step S30 also includes steps S31 to S34:

[0098] Step S31: Find the nearest valid frame preceding the current frame from the valid segments;

[0099] Step S32: Based on the preset global map, obtain the first bound lane where the vehicle to be cut into is located in the current frame, and the second bound lane where it is located in the most recent valid frame;

[0100] Step S33: Determine whether the first bound lane and the second bound lane are the same lane;

[0101] Step S34: If the first bound lane and the second bound lane are not the same lane, then based on the preset global map, obtain the bound lanes in each reference frame after the current frame in the effective segment, and obtain the bound lanes in each reference frame before the most recent effective frame in the effective segment.

[0102] It is understandable that, due to the possibility of errors in the information collected by the vehicle when gathering information about surrounding vehicles, and the possibility that the vehicle may abandon the cut-in process, a small number of frames in the valid segment may show a different lane than the lane where the vehicle to be cut in is located. To avoid misidentifying this situation as a cut-in behavior, when it is determined that the first bound lane of the vehicle to be cut in the current frame is different from the second bound lane in the valid frame, it is necessary to further determine whether the bound lane of the vehicle to be cut in the reference frames after the current frame is the first bound lane, and whether the bound lane of the vehicle to be cut in the reference frames before the most recent valid frame is the second bound lane, in order to improve the accuracy of cut-in behavior identification.

[0103] Furthermore, step S40 also includes:

[0104] Based on the bound lane of the vehicle to be switched into in different reference frames, the first percentage of all frames in the reference frames after the current frame whose bound lane is the first bound lane is calculated.

[0105] And, the second percentage of all frames in the reference frames preceding the most recent valid frame in which the bound lane is the second bound lane;

[0106] Determine whether both the first percentage and the second percentage satisfy the preset percentage;

[0107] If all the preset ratios are met, then the vehicle to be switched in is determined to have performed a cutin action.

[0108] It should be noted that the first percentage is the proportion of frames in the reference frames following the current frame where the lane to be entered is in the first bound lane out of the total number of reference frames. For example, if the reference frame is 15 frames after the current frame, and 14 of those 15 frames contain a lane in the first bound lane, then the first percentage is 94%. The second percentage is the proportion of frames in the reference frames preceding the most recent valid frame where the lane to be entered is in the second bound lane out of the total number of reference frames. For example, if the reference frame is 15 frames after the current frame, and 15 of those 15 frames contain a lane in the first bound lane, then the second percentage is 100%. The preset percentage is the threshold percentage for determining whether the lane to be entered in the reference frame is in the first or second bound lane when the vehicle performs a cutin action.

[0109] Understandably, in order to accurately identify whether a vehicle will engage in cutin behavior and will not exit cutin behavior when preparing to engage or during the engagement process, after obtaining the first proportion of all frames in the reference frames after the current frame that are bound to the first bound lane, and the second proportion of all frames in the reference frames before the most recent valid frame that are bound to the second bound lane, it is necessary to further determine whether the first proportion and the second proportion both meet the preset proportion. Only when both the first proportion and the second proportion meet the preset proportion is it determined that the vehicle will engage in cutin behavior, so as to improve the accuracy of cutin behavior identification.

[0110] In the specific implementation, it determines whether the most likely cutin candidate object (cutin_candidate) in the most recent valid frame has actually performed a cutin action. That is, it checks whether the current cutin candidate vehicle index (obj_id) is in the current cutin candidate vehicle index list (obj_id_list_pre) of the most recent valid previous frame. If not, it skips the current loop. It then obtains the information (obstacle_item) of the current cutin candidate vehicle index (obj_id) in the most recent frame, and based on the global map, obtains the bound lane (obj_laneid) of the vehicle to be cut in on the global map. Retrieve the object information `obstacle_item_pre` from the most recent valid previous frame for the object `obj_id`. Obtain the lane bound to the current object `obj_id` in the global map at that frame time: `obj_laneid_pre`. If `obj_laneid == -1` or `obj_laneid == obj_laneid_pre`, skip this loop. If `obj_laneid` and `obj_laneid_pre` are different, calculate the number of frames (`count_pre`) in which the object is bound to the lane index `obj_laneid_pre` in the previous frame and several frames prior to it (usually 15 frames). Calculate the number of frames (`count_next`) in which the object is bound to the lane index `obj_laneid` in the current frame and several frames following it (usually 15 frames). If both `count_pre` and `count_next` are greater than 9 (a preset threshold (90%)), a lane change is considered to have occurred. If `utin_candidate == obj_id`, the vehicle waiting to enter has performed a cutin action.

[0111] Based on the first and second embodiments of this application, the same or similar content as the above embodiments in the third embodiment of this application can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 Before step S20, the cutin behavior recognition method further includes steps S01 to S04:

[0112] Step S01: Map the coordinates of the lane centerline on each frame of data to the global grid to obtain an initial global map. The data from each frame constitutes the effective segment.

[0113] Step S02: Merge the overlapping lanes selected from the initial global map to obtain a global grid map;

[0114] Step S03: Map each object in the effective segment to the global grid map to obtain the index position of each object in the global grid map in each frame of data;

[0115] Step S04: Based on the index position, bind each object in each frame of data to the global lane centerline in the global grid map to obtain a preset global map.

[0116] It's important to note that the lane centerline is a virtual line on the road used to indicate the direction of vehicle travel, typically located in the center of the lane. The global grid is a two-dimensional or three-dimensional coordinate system used to map the actual road environment into a unified, standardized space. This grid system usually has a fixed resolution and size and can cover the entire area where autonomous vehicles might travel. The initial global map is a preliminary map obtained by mapping the lane centerline coordinates from each frame of data to the global grid. This map contains the basic shape and direction of the road but may contain overlapping or inconsistent lane information. Overlapping lanes occur in the initial global map due to differences or errors between different frames of data, resulting in multiple lane centerlines appearing at the same or similar locations. The global grid map is obtained by merging overlapping lanes from the initial global map and performing further processing and optimization. The index location is the coordinate or position information corresponding to each object in the global grid map.

[0117] Understandably, mapping the coordinates of lane centerlines from each frame of data to a unified global grid forms an initial global map, ensuring spatial consistency of lane information. In the initial global map, due to overlap or errors between valid segments, multiple lane centerlines may appear in the same or similar positions. By filtering and merging these overlapping lanes, redundancy and contradictions can be eliminated, resulting in a more accurate and consistent global grid map. Mapping objects (such as vehicles, pedestrians, traffic signs, etc.) in valid segments to the global grid map allows us to obtain the index position of each object in each frame of data within the global grid map, and corrects the lanes in the global grid map, ensuring the correctness of each lane. Finally, based on the index position of the object in the global grid map, the objects in each frame of data are bound to the global lane centerlines.

[0118] Understandably, by integrating lane centerline information from multiple frames of data and filtering and merging it, redundancy and contradictions between data are effectively eliminated. At the same time, by binding objects to lane centerlines, the accuracy and reliability of the map are further improved.

[0119] Furthermore, step S04 also includes:

[0120] Based on the index location, determine the target lane corresponding to the object in the global grid map;

[0121] Determine the number of objects present in the target lane in each frame of data;

[0122] If the number of objects is not one, then determine the projected distance of the object on the global lane centerline of the target lane;

[0123] By binding the objects in each frame of data with the global lane centerline and projection distance of the corresponding target lane, a preset global map is obtained.

[0124] It's important to note that the target lane is the lane where the object is located, determined within the global grid map based on its index position. The object count is the total number of objects present in a specific target lane within a given frame of data. Since the target lane has only one object, it means there's only one vehicle in that lane, so the target lane's coordinates can be assigned to that object. The projection distance is the vertical projection distance of the object onto the global lane centerline of the target lane; it's the shortest distance from the object's position to the lane centerline. The global lane centerline is a virtual line in the global grid map used to represent lane direction and position, located in the center of the lane.

[0125] Understandably, by constructing a pre-defined global map based on index location and target lane, there is no need to perform a global scan and reconstruction of the entire road environment, which reduces the complexity and computational load of data processing. Furthermore, by calculating the projected distance of an object on the global lane centerline of the target lane, the relative position of the object in the lane can be accurately described, thereby more accurately determining the object's driving trajectory and possible behavior, thus improving the accuracy of identifying cutin behavior.

[0126] Understandably, by being able to handle situations where there are multiple objects in the target lane, and by judging the number of objects and calculating the projection distance, the autonomous driving system provides rich environmental information, enabling it to better adapt to complex traffic scenarios such as congestion and intersections, thereby improving the system's adaptability and robustness.

[0127] Furthermore, step S02 also includes:

[0128] Filter overlapping lanes with overlapping areas from the initial global map;

[0129] The overlapping lanes are merged to obtain the global lane centerline of each lane in the effective segment in the initial global map;

[0130] The centerline coordinates of the global lane centerline are thinned at least once to obtain the thinned centerline coordinates.

[0131] Based on the thinned centerline coordinates, the global lane centerlines are gridded to obtain a global grid map.

[0132] It should be noted that thinning refers to simplifying the centerline coordinates of the global lane centerline to reduce the number of data points while maintaining the overall shape and direction of the lane centerline. Thinned centerline coordinates refer to the coordinates of the remaining data points on the global lane centerline after thinning. Gridding refers to the process of dividing the global lane centerline into multiple grid cells according to certain rules.

[0133] Understandably, by filtering and merging overlapping lanes, the overlap in the initial global map can be eliminated, making the lane centerlines clearer and more accurate. This improves the accuracy and readability of the global map. Furthermore, the thinning process significantly reduces the number of data points for the global lane centerlines, thereby reducing the complexity and computational load of data processing, simplifying the map building process, and improving the efficiency of global map generation. When using the global map, it also avoids wasting too much time searching for the lane centerline of the vehicle to cut in due to too many centerline coordinates, thus improving the efficiency of identifying cutin behavior.

[0134] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the cutin behavior recognition method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0135] This application also provides a cutin behavior recognition device, please refer to... Figure 4 The cutin behavior recognition device includes:

[0136] The acquisition module 10 is used to filter valid segments in the mapless LCC scenario from the acquired historical driving data;

[0137] The filtering module 20 is used to filter vehicles to be cut into the vehicle lane from the current frame of the effective segment;

[0138] The identification module 30 is used to identify the bound lane of the vehicle to be cut into in the reference frame of the effective segment based on a preset global map, wherein the preset global map includes the lanes in which each vehicle is located at different frame times.

[0139] The judgment module 40 is used to determine whether the vehicle to be cut in has performed a cutin action based on the bound lane in which the vehicle to be cut in is located in different reference frames.

[0140] Optionally, the identification module 30 is further configured to: find the most recent valid frame preceding the current frame from the valid segments; obtain, based on the preset global map, the first bound lane where the vehicle to be switched in is located in the current frame, and the second bound lane where it is located in the most recent valid frame; determine whether the first bound lane and the second bound lane are the same lane; if the first bound lane and the second bound lane are not the same lane, then, based on the preset global map, obtain the bound lanes in each reference frame following the current frame in the valid segments, and obtain the bound lanes in each reference frame preceding the most recent valid frame in the valid segments.

[0141] Optionally, the identification module 30 is further configured to, based on the bound lane where the vehicle to be cut in is located in different reference frames, calculate a first percentage of all frames in the reference frames after the current frame whose bound lane is the first bound lane; and a second percentage of all frames in the reference frames before the most recent valid frame whose bound lane is the second bound lane; determine whether the first percentage and the second percentage both satisfy a preset percentage; if both satisfy the preset percentage, then determine that the vehicle to be cut in has performed a cutin action.

[0142] Optionally, the filtering module 20 is further configured to extract cutin candidate vehicles from the current frame of the effective segment; obtain the azimuth angle of the vehicle in the current frame and the position coordinates of the vehicle mapped in the global map in the current frame; determine the relative lateral displacement and relative longitudinal displacement between the current cutin candidate vehicle and the vehicle based on the position information of the current cutin candidate vehicle and the position coordinates, wherein the current cutin candidate vehicle is the currently active cutin candidate vehicle; determine whether the relative longitudinal displacement is less than the currently recorded minimum longitudinal displacement, and determine whether the relative lateral displacement meets a preset threshold; if the relative lateral displacement is less than the minimum longitudinal displacement and the relative longitudinal displacement meets the preset threshold, then the current cutin candidate vehicle is used to update the currently recorded vehicles to be cut in, until all cutin candidate vehicles are judged, and the cutin candidate vehicle with the smallest relative lateral displacement is determined as the vehicle to be cut in.

[0143] Optionally, the filtering module 20 is further configured to map the coordinates of the lane centerlines on each frame of data to a global grid to obtain an initial global map, wherein the frames of data constitute the effective segments; merge overlapping lanes filtered from the initial global map to obtain a global grid map; map each object in the effective segments to the global grid map to obtain the index position of each object in each frame of data in the global grid map; and bind each object in each frame of data to the global lane centerlines in the global grid map based on the index positions to obtain a preset global map.

[0144] Optionally, the filtering module 20 is further configured to: determine the target lane corresponding to the object in the global grid map based on the index position; determine the number of objects present in each frame of data for the target lane; if the number of objects is not one, determine the projection distance of the object on the global lane centerline of the target lane; and bind the objects in each frame of data with the global lane centerline and projection distance of the corresponding target lane to obtain a preset global map.

[0145] Optionally, the filtering module 20 is further configured to: filter overlapping lanes with overlapping areas from the initial global map; merge the overlapping lanes to obtain the global lane centerline of each lane in the effective segment in the initial global map; perform at least one thinning process on the centerline coordinates of the global lane centerline to obtain thinned centerline coordinates; and grid the global lane centerline based on the thinned centerline coordinates to obtain a global grid map.

[0146] The cutin behavior recognition device provided in this application, employing the cutin behavior recognition method in the above embodiments, can solve the technical problem of being unable to recognize vehicle cutin behavior in mapless LCC scenarios. Compared with the prior art, the beneficial effects of the cutin behavior recognition device provided in this application are the same as those of the cutin behavior recognition method provided in the above embodiments, and other technical features in the cutin behavior recognition device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0147] This application provides a cutin behavior recognition device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the cutin behavior recognition method in the above embodiment 1.

[0148] The following is for reference. Figure 5The diagram illustrates a structural schematic of a cutin behavior recognition device suitable for implementing embodiments of this application. The cutin behavior recognition device in this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The cutin behavior recognition device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0149] like Figure 5 As shown, the Cutin behavior recognition device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the Cutin behavior recognition device. The processing unit 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 can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the Cutin behavior recognition device to communicate wirelessly or wiredly with other devices to exchange data. While the figures show Cutin behavior recognition devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0150] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0151] The cutin behavior recognition device provided in this application, employing the cutin behavior recognition method in the above embodiments, can solve the technical problem of being unable to recognize vehicle cutin behavior in LCC scenarios without a map. Compared with the prior art, the beneficial effects of the cutin behavior recognition device provided in this application are the same as those of the cutin behavior recognition method provided in the above embodiments, and other technical features in this cutin behavior recognition device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0152] 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 suitable manner in one or more embodiments or examples.

[0153] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0154] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the cutin behavior recognition method in the above embodiments.

[0155] 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, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, 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.

[0156] The aforementioned computer-readable storage medium may be included in the cutin behavior recognition device; or it may exist independently and not assembled into the cutin behavior recognition device.

[0157] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the cutin behavior recognition device, the cutin behavior recognition device causes the following: to filter valid segments in mapless LCC scenarios from the acquired historical driving data; to filter vehicles to be cut into the lane from the current frame of the valid segments; to identify the bound lane of the vehicle to be cut into in the reference frame of the valid segments based on a preset global map, wherein the preset global map includes the lanes where each vehicle is located at different frame times; and to determine whether the vehicle to be cut into has performed a cutin behavior based on the bound lanes where the vehicle to be cut into is located in different reference frames.

[0158] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can 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 can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0159] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0160] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0161] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described cutin behavior recognition method, which can solve the technical problem of not being able to recognize vehicle cutin behavior in mapless LCC scenarios. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the cutin behavior recognition method provided in the above embodiments, and will not be repeated here.

[0162] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the cutin behavior recognition method described above.

[0163] The computer program product provided in this application can solve the technical problem of being unable to identify vehicle cut-in behavior in LCC scenarios without a map. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the cut-in behavior recognition method provided in the above embodiments, and will not be repeated here.

[0164] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for identifying cutin behavior, characterized in that, The cutin behavior recognition method includes: The coordinates of the lane centerline in each frame of data are mapped to the global grid to obtain an initial global map, and the data in each frame constitutes a valid segment; The overlapping lanes selected from the initial global map are merged to obtain a global grid map; Each object in the effective segment is mapped to the global grid map to obtain the index position of each object in each frame of data in the global grid map, and the lanes in the global grid map are corrected. The index position includes the coordinates corresponding to each object in the global grid map. Based on the index position, each object in each frame of data is bound to the global lane centerline in the global grid map to obtain a preset global map; Filter valid segments from the acquired historical driving data for LCC scenarios without maps; Select vehicles to be switched into the vehicle lane from the current frame of the effective segment; Based on a preset global map, the bound lane of the vehicle to be cut into is identified in the reference frame of the effective segment. The preset global map includes the lanes in which each vehicle is located at different frame times. Based on the bound lane where the vehicle to be switched in is located in different reference frames, it is determined whether the vehicle to be switched in has performed a cutin action.

2. The cutin behavior recognition method as described in claim 1, characterized in that, The step of identifying the bound lane where the vehicle to be switched in is located in the reference frame of the effective segment based on a preset global map includes: Find the nearest valid frame preceding the current frame from the valid segments; Based on the preset global map, the first bound lane where the vehicle to be cut into is located in the current frame, and the second bound lane where it is located in the most recent valid frame are obtained; Determine whether the first bound lane and the second bound lane are the same lane; If the first bound lane and the second bound lane are not the same lane, then based on the preset global map, the bound lanes in each reference frame after the current frame in the effective segment are obtained, and the bound lanes in each reference frame before the most recent effective frame in the effective segment are also obtained.

3. The cutin behavior recognition method as described in claim 2, characterized in that, The step of determining whether the vehicle to be switched in has engaged in a cutin behavior based on the lane it is bound to in different reference frames includes: Based on the bound lane of the vehicle to be switched into in different reference frames, the first percentage of all frames in the reference frames after the current frame whose bound lane is the first bound lane is calculated. And, the second percentage of all frames in the reference frames preceding the most recent valid frame in which the bound lane is the second bound lane; Determine whether both the first percentage and the second percentage satisfy the preset percentage; If all the preset ratios are met, then the vehicle to be switched in is determined to have performed a cutin action.

4. The cutin behavior recognition method as described in claim 1, characterized in that, The step of selecting vehicles to switch into the vehicle lane from the current frame of the valid segment includes: Extract candidate vehicles for cutin from the current frame of the valid segment; Obtain the azimuth angle of the vehicle in the current frame and the position coordinates of the vehicle mapped in the global map in the current frame; Based on the location information of the current cutin candidate vehicle and the location coordinates, the relative lateral displacement and relative longitudinal displacement between the current cutin candidate vehicle and the vehicle are determined, and the current cutin candidate vehicle is the cutin candidate vehicle that is currently in the current position. Determine whether the relative longitudinal displacement is less than the currently recorded minimum longitudinal displacement, and determine whether the relative lateral displacement meets a preset threshold. If the relative lateral displacement is less than the minimum longitudinal displacement, and the relative longitudinal displacement meets a preset threshold, then the current cutin candidate vehicle is used to update the currently recorded cutin vehicle until all cutin candidate vehicles are judged, and the cutin candidate vehicle with the smallest relative lateral displacement is determined as the cutin vehicle.

5. The cutin behavior recognition method as described in claim 1, characterized in that, The step of binding each object in each frame of data to the global lane centerline in the global grid map based on the index position to obtain the preset global map includes: Based on the index location, determine the target lane corresponding to the object in the global grid map; Determine the number of objects present in the target lane in each frame of data; If the number of objects is not one, then determine the projected distance of the object on the global lane centerline of the target lane; By binding the objects in each frame of data with the global lane centerline and projection distance of the corresponding target lane, a preset global map is obtained.

6. The cutin behavior recognition method as described in claim 1, characterized in that, The step of merging overlapping lanes selected from the initial global map to obtain a global grid map includes: Filter overlapping lanes with overlapping areas from the initial global map; The overlapping lanes are merged to obtain the global lane centerline of each lane in the effective segment in the initial global map; The centerline coordinates of the global lane centerline are thinned at least once to obtain the thinned centerline coordinates. Based on the thinned centerline coordinates, the global lane centerlines are gridded to obtain a global grid map.

7. A cutin behavior recognition device, characterized in that, The cutin behavior recognition device includes: The acquisition module is used to filter valid segments in mapless LCC scenarios from the acquired historical driving data; The filtering module is used to filter vehicles to be cut into the vehicle lane from the current frame of the effective segment; The identification module is used to identify the bound lane of the vehicle to be cut into in the reference frame of the effective segment based on a preset global map, wherein the preset global map includes the lanes in which each vehicle is located at different frame times. The judgment module is used to determine whether the vehicle to be cut in has performed a cutin action based on the bound lane where the vehicle to be cut in is located in different reference frames; The filtering module is further configured to: map the coordinates of the lane centerlines on each frame of data to a global grid to obtain an initial global map; the frames of data constitute the effective segments; merge overlapping lanes filtered from the initial global map to obtain a global grid map; map each object in the effective segments to the global grid map to obtain the index position of each object in each frame of data in the global grid map; and correct the lanes in the global grid map, wherein the index position includes the coordinates corresponding to each object in the global grid map; and bind each object in each frame of data to the global lane centerline in the global grid map based on the index position to obtain a preset global map.

8. A cutin behavior recognition device, characterized in that, The Cutin behavior recognition device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the Cutin behavior recognition method as described in any one of claims 1 to 6.

9. 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, it implements the steps of the cutin behavior recognition method as described in any one of claims 1 to 6.

Images