Self-vehicle afflux scene recognition method, device and equipment and storage medium

Abstract

The invention discloses a self-vehicle afflux scene recognition method, device and equipment and a storage medium, and relates to the technical field of image recognition, and the method comprises the steps: screening an effective fragment of a to-be-interacted vehicle from the obtained historical driving data during the afflux of a self-vehicle; judging whether the first lateral displacement of the vehicle and the to-be-interacted vehicle in a current frame image exceeds a preset afflux distance or not, wherein the current frame image is a frame image which is being processed in the effective segment; if it is determined that the first transverse displacement exceeds the preset afflux distance, whether the to-be-interacted vehicle is a target vehicle interacted with the vehicle or not is judged based on the first transverse displacement and second transverse displacement and longitudinal displacement of the vehicle and the to-be-interacted vehicle in the last frame of image, and the last frame of image is the last frame of image in the effective segment; and if it is determined that the to-be-interacted vehicle is the target vehicle, determining that the historical driving data is an afflux scene. The situation that the lane changing scene of the own vehicle is recognized as the own vehicle afflux scene is avoided, so that the recognition accuracy of the own vehicle afflux scene is improved.

Description

Technical Field

[0001] This application relates to the field of image recognition technology, and in particular to a method, apparatus, device, and storage medium for recognizing vehicles merging into a scene. Background Technology

[0002] To optimize the vehicle merging scenario, it is necessary to analyze the vehicle merging scenario in the vehicle's historical driving data. Since the historical driving data contains a large number of scenario data fragments, it is necessary to identify the vehicle merging scenario from the historical driving data.

[0003] Currently, identifying merging scenarios from historical driving data is based on the relative lateral and longitudinal distances between the vehicle and other vehicles, as well as the vehicle's speed and the speeds of other vehicles. However, since identifying lane-changing scenarios from historical driving data also relies on these same factors, lane-changing scenarios may be mistakenly identified as merging scenarios, thus affecting the accuracy of identifying whether a scenario in historical driving data is indeed a merging scenario.

[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 method for recognizing merging scenarios of vehicles, aiming to solve the technical problem of misjudging lane changes of vehicles as merging scenarios of vehicles, thereby affecting the accuracy of recognizing whether historical driving data is a merging scenario of vehicles.

[0006] To achieve the above objectives, this application proposes a method for recognizing merging vehicles, the method comprising:

[0007] Filter out valid segments of vehicles that are waiting to be interacted with when the vehicle merges in from the acquired historical driving data;

[0008] Determine whether the first lateral displacement of the autonomous vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance, wherein the current frame image is the frame image being processed in the effective segment;

[0009] If it is determined that the first lateral displacement exceeds the preset merging distance, then based on the first lateral displacement, and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image, it is determined whether the vehicle to be interacted with is the target vehicle of the self-vehicle interaction, and the last frame image is the last frame image in the effective segment.

[0010] If the vehicle to be interacted with is determined to be the target vehicle, then the historical driving data is determined to be a scenario where the vehicle merges in.

[0011] In one embodiment, the step of determining whether the first lateral displacement of the self-vehicle and the vehicle to be interacted with in the current frame image exceeds a preset merging distance includes:

[0012] Based on the valid fragment, determine the vehicle trajectory of the autonomous vehicle and the object trajectory of the vehicle to be interacted with;

[0013] Based on the vehicle trajectory and the object trajectory, determine whether the directional change angle of the vehicle in the effective segment meets the preset merging angle;

[0014] If the angle of change of direction meets the preset merging angle, then it is determined whether the first lateral displacement exceeds the preset merging distance.

[0015] In one embodiment, the step of determining whether the directional change angle of the vehicle in the effective segment satisfies the preset merging angle based on the vehicle trajectory and the object trajectory includes:

[0016] Extract the trajectory length corresponding to the shorter trajectory from the vehicle trajectory and the object trajectory;

[0017] Based on the trajectory length, a valid vehicle trajectory is extracted from the vehicle trajectory, and the position point of the vehicle in the current frame trajectory is determined as the starting trajectory point of the valid vehicle trajectory.

[0018] Determine the starting direction angle of the starting trajectory point and the ending direction angle of the ending trajectory point in the effective vehicle trajectory;

[0019] Based on the starting direction angle and the ending direction angle, the directional change angle of the vehicle in the effective segment is determined;

[0020] Determine whether the angle of change of direction meets the preset merging angle.

[0021] In one embodiment, the step of determining the starting direction angle of the starting trajectory point and the ending direction angle of the ending trajectory point in the valid vehicle trajectory includes:

[0022] Identify the end point of the valid vehicle trajectory;

[0023] The starting trajectory point is vectorized to determine the starting direction vector of the starting trajectory point, and the ending trajectory point is vectorized to determine the ending direction vector of the ending trajectory point;

[0024] Determine whether the length of the first vector of the starting direction vector is less than a preset length, and determine whether the length of the second vector of the ending direction vector is less than a preset length;

[0025] If the lengths of the first vector and the second vector are both not less than the preset length, then the starting direction angle of the starting trajectory point is determined based on the starting direction vector, and the ending direction angle of the ending trajectory point is determined based on the ending direction vector.

[0026] In one embodiment, the step of determining the directional change angle of the vehicle in the effective segment based on the starting direction angle and the ending direction angle includes:

[0027] Calculate the angle difference between the starting direction angle and the ending direction angle;

[0028] The angle difference is normalized to determine the directional change angle of the vehicle in the effective segment.

[0029] In one embodiment, the step of determining whether the vehicle to be interacted with is the target vehicle for interaction by the self-vehicle based on the first lateral displacement and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image includes:

[0030] Determine whether the longitudinal displacement of the autonomous vehicle and the vehicle to be interacted with in the last frame of the image satisfies a preset longitudinal condition;

[0031] And, whether the second lateral displacement of the self-vehicle and the vehicle to be interacted with in the last frame of the image satisfies the preset merging distance;

[0032] If it is determined that the longitudinal displacement satisfies the longitudinal condition and the second lateral displacement satisfies the preset merging distance, then it is determined whether the lateral displacement difference between the first lateral displacement and the second lateral displacement is less than or equal to the preset lateral condition.

[0033] If it is determined that the longitudinal displacement satisfies the longitudinal condition, and the lateral displacement difference is less than or equal to the preset lateral condition, then the vehicle to be interacted with is determined to be the target vehicle for the autonomous vehicle interaction.

[0034] In one embodiment, the step of determining the historical driving data as a scenario where the vehicle to be interacted with is determined to be the target vehicle includes:

[0035] If the vehicle to be interacted with is the target vehicle, then update the reference number of the target vehicle recorded.

[0036] Determine whether the reference quantity meets the preset quantity condition;

[0037] If the preset quantity condition is met, the historical driving data is determined to be a scenario where a vehicle merges in.

[0038] Furthermore, to achieve the above objectives, this application also proposes a vehicle merging scene recognition device, which includes:

[0039] The acquisition module is used to filter valid segments of vehicles that exist for interaction when the vehicle merges in from the acquired historical driving data.

[0040] The first judgment module is used to determine whether the first lateral displacement of the self-vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance, wherein the current frame image is the frame image being processed in the effective segment;

[0041] The second judgment module is used to determine whether the vehicle to be interacted with is the target vehicle of the self-vehicle if it is determined that the first lateral displacement exceeds the preset merging distance, based on the first lateral displacement and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image, and the last frame image is the last frame image in the effective segment.

[0042] The determination module is used to determine the historical driving data as a scenario where the vehicle to be interacted with is determined to be the target vehicle.

[0043] In addition, to achieve the above objectives, this application also proposes a vehicle merging scene recognition device, the device comprising: 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 vehicle merging scene recognition method described above.

[0044] 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 vehicle merging scene recognition method described above.

[0045] 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 vehicle merging scene recognition method described above.

[0046] One or more technical solutions proposed in this application have at least the following technical effects:

[0047] First, valid segments containing the vehicle to be interacted with are selected from historical driving data when the vehicle merges in. Then, the first lateral displacement of the vehicle and the vehicle to be interacted with in the current frame image is determined. The current frame image is the frame image being processed among the valid segments. It is then determined whether the first lateral displacement exceeds a preset merging distance. This preset merging distance is used to determine whether the vehicle exceeds the safe merging distance (preset merging distance) in the current frame image of the lane where the vehicle to be interacted with is located. If the first lateral displacement exceeds the preset merging distance, it is determined that the vehicle meets the conditions of the merging scenario, that is, it meets the requirements for safe merging. Then, the vehicle and the vehicle to be interacted with are used to determine whether the vehicle meets the requirements for safe merging in the current frame image. The vehicle's first lateral displacement in the current frame image, and its second lateral and longitudinal displacements in the last frame of the valid segment, are used to further determine whether the vehicle will interact with the vehicle to be interacted with. If the vehicle interacts with the vehicle to be interacted with, then the vehicle to be interacted with is determined to be the target vehicle of the vehicle. Since the vehicle will not interact with the vehicle to be interacted with in the scenario of changing lanes, the vehicle's movement process can be analyzed by analyzing the positional relationship between the vehicle and the vehicle to be interacted with in different frames of the valid segment. This can avoid misidentifying the scenario of the vehicle changing lanes as the scenario of the vehicle merging into the scene, thereby improving the accuracy of identifying the scenario of the vehicle merging into the scene. Attached Figure Description

[0048] 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.

[0049] 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.

[0050] Figure 1 This is a flowchart illustrating an embodiment of the vehicle merging scene recognition method of this application.

[0051] Figure 2 This is a flowchart illustrating Embodiment 2 of the vehicle merging scene recognition method of this application;

[0052] Figure 3 This is a flowchart illustrating Embodiment 3 of the vehicle merging scene recognition method of this application;

[0053] Figure 4 This is a schematic diagram of the module structure of the vehicle merging scene recognition device according to an embodiment of this application;

[0054] Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the vehicle merging scene recognition method in this application embodiment.

[0055] 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

[0056] 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.

[0057] 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.

[0058] The main solution of this application embodiment is as follows: The data analysis device filters valid segments from the acquired historical driving data where there is a vehicle to be interacted with when the vehicle merges in; it determines whether the first lateral displacement of the vehicle and the vehicle to be interacted with in the current frame image exceeds a preset merging distance, where the current frame image is the frame image being processed in the valid segments; if it is determined that the first lateral displacement exceeds the preset merging distance, then based on the first lateral displacement and the second lateral and longitudinal displacements of the vehicle and the vehicle to be interacted with in the last frame image, it determines whether the vehicle to be interacted with is the target vehicle for the vehicle interaction; if it is determined that the vehicle to be interacted with is the target vehicle, then the historical driving data is determined to be a vehicle merging scenario.

[0059] In this embodiment, for ease of description, the following description uses a data analysis device as the execution subject.

[0060] Currently, identifying merging scenarios from historical driving data relies on the relative lateral and longitudinal distances between the vehicle and other vehicles, as well as the vehicle's speed and the speeds of other vehicles. Similarly, identifying lane-changing scenarios from historical driving data also relies on these same factors. Therefore, this process can lead to misidentifying lane-changing scenarios as merging scenarios, thus affecting the accuracy of identifying merging scenarios from historical driving data.

[0061] This application provides a solution that first filters valid segments from historical driving data where the vehicle merges and there is a vehicle to be interacted with. Then, it determines the first lateral displacement between the vehicle and the vehicle to be interacted with in the current frame image, where the current frame image is the frame image being processed from the valid segments. It then determines whether the first lateral displacement exceeds a preset merging distance. This preset merging distance is used to determine whether the vehicle exceeds the safe merging distance (preset merging distance) in the current frame image from the lane where the vehicle to be interacted with is located. When the first lateral displacement exceeds the preset merging distance, it is determined that the vehicle meets the conditions for the merging scenario, that is, it meets the requirements for safe merging. Then, it utilizes the... The first lateral displacement of the vehicle and the vehicle to be interacted with in the current frame image, and the second lateral displacement and longitudinal displacement in the last frame image of the effective segment, are used to further determine whether the vehicle will interact with the vehicle to be interacted with. If the vehicle interacts with the vehicle to be interacted with, then the vehicle to be interacted with is determined to be the target vehicle of the vehicle. Since the vehicle will not interact with the vehicle to be interacted with in the scenario of the vehicle changing lanes, the movement process of the vehicle can be analyzed by analyzing the positional relationship between the vehicle and the vehicle to be interacted with in different frames of the effective segment, so as to avoid misidentifying the scenario of the vehicle changing lanes as the scenario of the vehicle merging into the scene, thereby improving the accuracy of identifying the scenario of the vehicle merging into the scene.

[0062] 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.

[0063] Based on this, this application provides a method for recognizing merging vehicle scenes, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the vehicle merging scene recognition method of this application.

[0064] In this embodiment, the vehicle merging scene recognition method includes steps S10 to S40:

[0065] Step S10: Filter out valid segments of vehicles that need to be interacted with when the self-vehicle merges from the acquired historical driving data;

[0066] 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 frame images. The "self vehicle" is the vehicle that provides the historical driving data. The "vehicle to be interacted with" is the vehicle that may be interacted with when the self vehicle merges in.

[0067] Step S20: Determine whether the first lateral displacement of the self-vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance. The current frame image is the frame image being processed in the effective segment.

[0068] It should be noted that the first lateral displacement is the lateral distance between the self-vehicle and the vehicle to be interacted with in the current frame image. The current frame image is the frame image being processed at the current moment in the valid segment. The preset merging distance is the minimum distance required to determine whether the self-vehicle can safely merge.

[0069] It is understandable that, since it is necessary to ensure the safety of the vehicle and other vehicles in the lane when the vehicle merges, a certain lateral distance is usually left as a safe merging distance. Therefore, when it is necessary to identify the merging scenario, it is possible to first check whether the lateral distance between the vehicle and the vehicle to be interacted with meets the preset merging distance.

[0070] Understandably, since the merging of a vehicle is a dynamic process, it is possible to determine whether the first lateral displacement of the vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance in order to determine whether the vehicle meets the conditions for safe merging. If the conditions for safe merging are met, further judgment can be made to improve the efficiency of recognizing the merging scene of the vehicle.

[0071] Step S30: If it is determined that the first lateral displacement exceeds the preset merging distance, then based on the first lateral displacement and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image, it is determined whether the vehicle to be interacted with is the target vehicle of the self-vehicle interaction.

[0072] It should be noted that the last frame image is the image of the last frame in the valid segment.

[0073] Understandably, since vehicle merging is a dynamic process, if the lateral distance between the vehicle and the vehicle to be interacted with exceeds the preset merging distance in the current frame image, then the vehicle is determined to meet the conditions for safe merging. Therefore, in order to accurately identify whether the vehicle is in the merging scene, it is necessary to further determine whether the vehicle is in the merging scene by examining the image of the last frame in the effective segment.

[0074] In the specific implementation, the position (curr_x, curr_y) of the vehicle to be interacted with relative to the vehicle itself in the current frame image is first obtained. Here, curr_x is the longitudinal displacement and curr_y is the lateral displacement. If the absolute value of curr_y is less than 5 meters (the preset merging distance), the vehicle to be interacted with is considered to be too close to the vehicle laterally, and is not considered a lane-merging action. If it exceeds 5 meters, the position (future_x, future_y) of the vehicle to be interacted with relative to the vehicle itself at the corresponding moment in the last frame image is obtained. If the longitudinal distance future_x is between 0 and 30 meters, and the absolute value of the lateral distance future_y is less than 5 meters, the vehicle to be interacted with is considered to be approaching the vehicle, meaning that the vehicle to be interacted with is the interaction target when the vehicle merges in.

[0075] It should be noted that the second horizontal displacement is the lateral distance between the self-vehicle and the vehicle to be interacted with in the last frame of the image. The vertical displacement is the longitudinal distance between the self-vehicle and the vehicle to be interacted with in the last frame of the image. The target vehicle is the vehicle that the self-vehicle will interact with when it merges in.

[0076] Understandably, as the lateral and longitudinal distances between the vehicle and the vehicle to be interacted with gradually decrease during merging, the lateral distance changes between the vehicle and the vehicle to be interacted with are judged by referring to the first and second lateral displacements to avoid identifying a lane change as a merging. Since the longitudinal displacement between the vehicle and the vehicle to be interacted with remains within a certain range during merging to ensure the safety of the vehicle and to ensure a smooth merging, avoiding interference from vehicles in the merging lane, the longitudinal displacement between the vehicle and the vehicle to be interacted with can be used to more accurately determine whether the vehicle to be identified is the target vehicle to be interacted with during the merging.

[0077] Step S40: If the vehicle to be interacted with is determined to be the target vehicle, then the historical driving data is determined to be a scenario where the vehicle merges in.

[0078] It is understandable that when a vehicle merges, it needs to identify the vehicle it is interacting with. Also, when a vehicle is simply changing lanes, there may not be any vehicles in the lane that needs to be changed. Therefore, when the vehicle to be interacted with is identified as the target vehicle, it can be determined that the vehicle is merging, thus avoiding misidentifying a vehicle changing lanes as the vehicle merging and improving the accuracy of identifying the vehicle merging scenario.

[0079] In the specific implementation, if the vehicle to be interacted with in a frame is determined to be the interaction target when an egovehicle merges in, the counter `mergein_car_num` is incremented by 1. The number of interactive object vehicles in that frame, `mergein_car_num`, is stored in the dictionary `mergein_car_num_dict` for the corresponding frame. Based on the dictionary `mergein_car_num_dict`, it can be further determined whether it is an egovehicle merging scenario. For example, one method is to count the number of frames where `mergein_car_num > 0`, `mergein_car_frame_count`. Furthermore, the proportion of `mergein_car_frame_count` to the total number of data frames can be set to characterize whether a data segment is an egovehicle merging scenario.

[0080] Optionally, step S40 may also include:

[0081] If the vehicle to be interacted with is the target vehicle, then update the reference number of the target vehicle recorded.

[0082] Determine whether the reference quantity meets the preset quantity condition;

[0083] If the preset quantity condition is met, the historical driving data is determined to be a scenario where a vehicle merges in.

[0084] Understandably, to avoid the randomness of historical driving segments, the accuracy of recognition can be improved by judging the reference number of target vehicles and determining that the historical driving data is the scene of the vehicle entering the scene when the reference number meets the preset number conditions.

[0085] This embodiment provides a method for recognizing merging scenarios of autonomous vehicles. First, it filters valid segments from historical driving data where an interactive vehicle exists during the merging process. Then, it determines the first lateral displacement of the autonomous vehicle and the interactive vehicle within the current frame image, where the current frame image is the frame image being processed from the valid segments. It also determines whether the first lateral displacement exceeds a preset merging distance. This preset merging distance is used to determine whether the autonomous vehicle exceeds the safe merging distance (preset merging distance) within the current frame image from the lane where the interactive vehicle is located. If the first lateral displacement exceeds the preset merging distance, it is determined that the autonomous vehicle meets the conditions for a merging scenario, that is, it meets the requirements for safe merging. By utilizing the first lateral displacement of the vehicle and the vehicle to be interacted with in the current frame image, and the second lateral and longitudinal displacements in the last frame image of the effective segment, it is further determined whether the vehicle will interact with the vehicle to be interacted with. If the vehicle interacts with the vehicle to be interacted with, then the vehicle to be interacted with is determined to be the target vehicle of the vehicle. Since the vehicle will not interact with the vehicle to be interacted with in the scenario of the vehicle changing lanes, the movement process of the vehicle can be analyzed by analyzing the positional relationship between the vehicle and the vehicle to be interacted with in different frames of the effective segment, so as to avoid misidentifying the scenario of the vehicle changing lanes as the scenario of the vehicle merging into the scene, thereby improving the accuracy of identifying the scenario of the vehicle merging into the scene.

[0086] 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 Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 Step S10 includes steps S11 to S13:

[0087] Step S11: Based on the valid fragment, determine the vehicle trajectory of the self-driving vehicle and the object trajectory of the vehicle to be interacted with;

[0088] Step S12: Based on the vehicle trajectory and the object trajectory, determine whether the directional change angle of the vehicle in the effective segment meets the preset merging angle;

[0089] Step S13: If the direction change angle satisfies the preset merging angle, then determine whether the first lateral displacement exceeds the preset merging distance.

[0090] It should be noted that the vehicle trajectory is the trajectory formed by the position points of the vehicle in each frame of the valid segment. The object trajectory is the trajectory formed by the position points of the vehicle to be interacted with in each frame of the valid segment. The direction change angle is the angle by which the vehicle's driving direction changes. The preset merging angle is the minimum angle by which the driving direction needs to change when the vehicle needs to merge.

[0091] Understandably, by aligning the trajectory of the vehicle with the trajectory of the vehicle to be interacted with, the trajectory points of the vehicle and the vehicle to be interacted with are synchronized in time. This ensures that when determining whether the vehicle to be interacted with is the vehicle to be interacted with, they meet at the same time, thus avoiding misidentification due to inconsistency in timing and improving the accuracy of identifying vehicles merging into the scene.

[0092] In the specific implementation, the future trajectory of the vehicle to be interacted with, `car_traj`, and the future trajectory of the vehicle itself, `ego_traj`, are obtained under the current frame image index `frame_idx`. Based on the valid fragment `data_vector`, the current frame image index `frame_idx`, and the object's `obj_id` (or the vehicle's own ID), the trajectory points of the object within a future time period (relative to the current frame image) are obtained. The acquisition of `car_traj` and `ego_traj` is similar, specifically as follows:

[0093] 1. Initialize an empty list traj to store the trajectory points of the object.

[0094] 2. Define the maximum number of steps as max_step as 60, that is, it can judge the data of up to 60 frames.

[0095] 3. Use a loop to iterate through future frame data starting from the current frame, with a step size of 5, meaning data is read every 5 frames. For each frame, check if its frame index exceeds the frame length of the valid segment data_vector. If it does, exit the loop early. Then check if the current frame contains the specified interactive vehicle obj_id. If the interactive vehicle does not exist, exit the loop early. If the interactive vehicle obj_id exists, obtain its position information and add it to the trajectory list traj. Then, based on the vehicle trajectory and object trajectory in the trajectory list, determine if the vehicle's directional change angle in the valid segment meets the preset merging angle. Further determine if the first lateral displacement of the vehicle and the interactive vehicle in the current frame exceeds the preset merging distance.

[0096] Further, step S13 includes:

[0097] Extract the trajectory length corresponding to the shorter trajectory from the vehicle trajectory and the object trajectory;

[0098] Based on the trajectory length, a valid vehicle trajectory is extracted from the vehicle trajectory, and the position point of the vehicle in the current frame trajectory is determined as the starting trajectory point of the valid vehicle trajectory.

[0099] Determine the starting direction angle of the starting trajectory point and the ending direction angle of the ending trajectory point in the effective vehicle trajectory;

[0100] Based on the starting direction angle and the ending direction angle, the directional change angle of the vehicle in the effective segment is determined;

[0101] Determine whether the angle of change of direction meets the preset merging angle.

[0102] It should be noted that the shorter trajectory is the trajectory shorter between the vehicle trajectory and the object trajectory. For example, if the object trajectory is shorter than the vehicle trajectory, the object trajectory is the shorter trajectory; if the vehicle trajectory is shorter than the object trajectory, the vehicle trajectory is the shorter trajectory; if the object trajectory and the supporting trajectory have the same length, the shorter trajectory can be either of them. The trajectory length is the number of trajectory points contained in the trajectory; that is, if the trajectory consists of two trajectory points, the trajectory length is 2. The effective vehicle trajectory is a portion or the entire trajectory extracted from the vehicle trajectory. It is a segment of the trajectory extending from the vehicle's position point in the current frame image towards the past time direction. The length of the effective vehicle trajectory is determined by the length of the shorter trajectory. The starting trajectory point is the starting point of the effective vehicle trajectory, i.e., the vehicle's position point in the current frame image. The ending trajectory point is the ending point of the effective vehicle trajectory. The starting direction angle is the angle between the vehicle's direction of movement at the starting trajectory point and the horizontal axis; that is, the vector angle formed by the starting trajectory point and its adjacent trajectory points. The ending direction angle is the angle between the direction of the vehicle's movement at the ending trajectory point and the horizontal axis, that is, the vector angle formed by the ending trajectory point and the adjacent trajectory points.

[0103] Understandably, in order to ensure that the time of each trajectory point between the vehicle trajectory and the object trajectory is equal, it is necessary to extract the trajectory length of the shorter trajectory from the vehicle trajectory and the object trajectory. This ensures that the trajectory segments with the same time are extracted from the vehicle trajectory as valid vehicle trajectories, thereby ensuring that subsequent comparisons are performed at the same time points.

[0104] In the specific implementation, the vehicle trajectory and the object trajectory are processed synchronously. That is, the shorter trajectory length (number of trajectory points) is taken as the valid vehicle trajectory length to ensure that subsequent comparisons are performed at the same time point. The valid trajectory size valid_size is determined by taking the minimum value of the length of the elements in the car_traj and ego_traj lists. The first valid_size trajectory points of the vehicle object trajectory car_traj are taken as the new vehicle object trajectory car_traj, and the first valid_size trajectory points of the vehicle trajectory ego_traj are taken as the new vehicle trajectory ego_traj. Based on the new vehicle object trajectory car_traj and the new vehicle trajectory ego_traj, the orientation start_heading and end_heading at the starting and ending trajectory points are calculated respectively. If the angle of change of direction ego_turn_angle is greater than 30 degrees (preset merging angle), it is considered that the vehicle is performing a large turning action.

[0105] Further, the steps of determining the starting direction angle of the starting trajectory point and the ending direction angle of the ending trajectory point in the valid vehicle trajectory include:

[0106] Identify the end point of the valid vehicle trajectory;

[0107] The starting trajectory point is vectorized to determine the starting direction vector of the starting trajectory point, and the ending trajectory point is vectorized to determine the ending direction vector of the ending trajectory point;

[0108] Determine whether the length of the first vector of the starting direction vector is less than a preset length, and determine whether the length of the second vector of the ending direction vector is less than a preset length;

[0109] If the lengths of the first vector and the second vector are both not less than the preset length, then the starting direction angle of the starting trajectory point is determined based on the starting direction vector, and the ending direction angle of the ending trajectory point is determined based on the ending direction vector.

[0110] It should be noted that the starting direction vector is the vector formed by the starting trajectory point and its corresponding adjacent trajectory points. The ending direction vector is the vector formed by the ending trajectory point and its corresponding connected trajectory points. The preset length is the minimum length of a vector that has reference value.

[0111] Understandably, since the distance between two trajectory points is too small, the vector determined by these two trajectory points cannot accurately represent the direction of the vector. Therefore, it is necessary to filter the starting or ending direction vector by a preset length to determine whether the starting and ending direction vectors have reference significance, thereby ensuring the accuracy of the final recognition.

[0112] In the specific implementation, it checks whether the input trajectory point list `car_traj` contains at least two points. If the number of trajectory points is less than two, the direction angle cannot be calculated. In this case, `start_heading = None` and `end_heading = None`. If the trajectory point list `car_traj` contains more than two points, the direction vectors of the starting and ending trajectory points of the vehicle's trajectory are calculated by subtracting the coordinates of two consecutive points. For example, the starting direction vector is obtained by subtracting the first point from the second point, and the ending direction vector is obtained by subtracting the second-to-last point from the last point. To ensure that the direction vectors are not zero vectors (i.e., the length is less than the preset length (1.0)), it checks whether the length of each direction vector is greater than or equal to 1.0. If the length of any direction vector is less than 1.0, it indicates that the distance between the trajectory points is too close to reliably calculate the direction angle. In this case, the function sets two `None` values: `start_heading = None` and `end_heading = None`. Finally, the direction angle of each direction vector is calculated using the `math.atan2` function. The `math.atan2(y, x)` function takes the y-components and x-components of a two-dimensional vector as arguments and returns the angle (in radians) between that vector and the positive x-axis. Thus, we can obtain the starting and ending direction angles of the trajectory.

[0113] Further, the step of determining the directional change angle of the vehicle in the effective segment based on the starting direction angle and the ending direction angle includes:

[0114] Calculate the angle difference between the starting direction angle and the ending direction angle;

[0115] The angle difference is normalized to determine the directional change angle of the vehicle in the effective segment.

[0116] It should be noted that the normalization process uses 360 degrees as a circle and normalizes the angle values ​​to the range (-π, π].

[0117] Understandably, since the angle difference between the starting and ending direction angles can characterize the change in the vehicle's direction angle, it is necessary to obtain the angle difference between the starting and ending direction angles and normalize this angle difference to determine the angle of change in the vehicle's direction within the valid segment. Since both the starting and ending direction angles are the direction angles of two trajectory points in the vehicle's trajectory, the angle of change in direction determined by these starting and ending direction angles can accurately represent the vehicle's direction change, thereby improving the accuracy of identifying whether the vehicle is in a merging scene.

[0118] In the specific implementation, the direction change angle is determined by ego_turn_angle = abs(normal_theta(ego_end_heading-ego_start_heading)), where normal_theta is the normalization of ego_end_heading-ego_start_heading to the interval (-π, π].

[0119] 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 Step S30 includes steps S01 to S02:

[0120] Step S01: Determine whether the longitudinal displacement of the self-vehicle and the vehicle to be interacted with in the last frame image satisfies the preset longitudinal condition;

[0121] Step S02, and whether the second lateral displacement of the self-vehicle and the vehicle to be interacted with in the last frame image satisfies the preset merging distance;

[0122] Step S03: If it is determined that the longitudinal displacement satisfies the longitudinal condition and the second lateral displacement satisfies the preset merging distance, then determine whether the lateral displacement difference between the first lateral displacement and the second lateral displacement is less than or equal to the preset lateral condition.

[0123] Step S04: If it is determined that the longitudinal displacement satisfies the longitudinal condition and the lateral displacement difference is less than or equal to the preset lateral condition, then the vehicle to be interacted with is determined to be the target vehicle for the autonomous vehicle interaction.

[0124] It should be noted that the preset longitudinal condition is a pre-set longitudinal displacement threshold used to determine whether the relative positional change between the self-vehicle and the vehicle to be interacted with in the longitudinal direction meets specific requirements. For example, if the longitudinal displacement exceeds a certain threshold, the relative positional change of the self-vehicle in the longitudinal direction is considered large. The preset lateral condition refers to a pre-set lateral displacement difference threshold used to determine whether the trend of the relative positional change between the self-vehicle and the vehicle to be interacted with in the lateral direction meets specific requirements. For example, if the lateral displacement difference is less than a certain threshold, the relative positional change of the self-vehicle in the lateral direction is considered small.

[0125] Understandably, in order to more accurately identify whether a vehicle is merging or not changing lanes, a threshold for lateral distance change can be set. That is, a preset lateral condition is set, and the vehicle to be interacted with is considered to be the target vehicle of the vehicle interaction only when the difference between the first lateral displacement and the second lateral displacement is less than or equal to the preset lateral condition. This further avoids identifying a vehicle changing lanes as a vehicle merging.

[0126] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the vehicle merging scene recognition method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0127] This application also provides a vehicle merging scene recognition device; please refer to [reference needed]. Figure 4 The vehicle merging scene recognition device includes:

[0128] The acquisition module 10 is used to filter valid segments of vehicles that need to be interacted with when the vehicle merges in from the acquired historical driving data;

[0129] The first judgment module 20 is used to determine whether the first lateral displacement of the self-vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance, wherein the current frame image is the frame image being processed in the effective segment;

[0130] The second judgment module 30 is used to determine whether the vehicle to be interacted with is the target vehicle of the self-vehicle if it is determined that the first lateral displacement exceeds the preset merging distance, based on the first lateral displacement and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image, and the last frame image is the last frame image in the effective segment.

[0131] The determination module 40 is used to determine the historical driving data as a self-entry scenario if the vehicle to be interacted with is determined to be the target vehicle.

[0132] Optionally, the first judgment module 20 is further configured to determine the vehicle trajectory of the vehicle and the object trajectory of the vehicle to be interacted with based on the effective segment; determine whether the directional change angle of the vehicle in the effective segment meets the preset merging angle based on the vehicle trajectory and the object trajectory; and determine whether the first lateral displacement exceeds the preset merging distance if the directional change angle meets the preset merging angle.

[0133] Optionally, the first judgment module 20 is further configured to extract the trajectory length corresponding to the shorter trajectory from the vehicle trajectory and the object trajectory; based on the trajectory length, extract a valid vehicle trajectory from the vehicle trajectory and determine the position point of the vehicle in the current frame trajectory as the starting trajectory point of the valid vehicle trajectory; determine the starting direction angle of the starting trajectory point and the ending direction angle of the ending trajectory point in the valid vehicle trajectory; based on the starting direction angle and the ending direction angle, determine the direction change angle of the vehicle in the valid segment; and determine whether the direction change angle satisfies the preset merging angle.

[0134] Optionally, the first judgment module 20 is further configured to calculate the angle difference between the starting direction angle and the ending direction angle; normalize the angle difference to determine the direction change angle of the vehicle in the effective segment.

[0135] Optionally, the second judgment module 30 is further configured to determine whether the longitudinal displacement of the self-vehicle and the vehicle to be interacted with in the last frame image satisfies a preset longitudinal condition; and whether the second lateral displacement of the self-vehicle and the vehicle to be interacted with in the last frame image satisfies the preset merging distance; if it is determined that the longitudinal displacement satisfies the longitudinal condition and the second lateral displacement satisfies the preset merging distance, then it is determined whether the lateral displacement difference between the first lateral displacement and the second lateral displacement is less than or equal to the preset lateral condition; if it is determined that the longitudinal displacement satisfies the longitudinal condition and the lateral displacement difference is less than or equal to the preset lateral condition, then the vehicle to be interacted with is determined to be the target vehicle for the self-vehicle interaction.

[0136] Optionally, the determining module 40 is configured to update the reference number of the target vehicle if the vehicle to be interacted with is the target vehicle; determine whether the reference number meets a preset number condition; and if the preset number condition is met, determine that the historical driving data is a scenario where a vehicle merges in.

[0137] The vehicle merging scene recognition device provided in this application, employing the vehicle merging scene recognition method in the above embodiments, can solve the technical problem of misjudging lane changes by vehicles as merging scenes, thereby affecting the accuracy of identifying whether historical driving data constitutes a vehicle merging scene. Compared with the prior art, the beneficial effects of the vehicle merging scene recognition device provided in this application are the same as those of the vehicle merging scene recognition method provided in the above embodiments, and other technical features in the vehicle merging scene recognition device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0138] This application provides a vehicle merging 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 executable 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 perform the vehicle merging scene recognition method in the above embodiment 1.

[0139] The following is for reference. Figure 5 This document illustrates a structural schematic diagram of a vehicle merging scene recognition device suitable for implementing embodiments of this application. The vehicle merging scene recognition device in this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio 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 vehicle merging scene 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.

[0140] like Figure 5As shown, the vehicle merging scene 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 vehicle merging scene 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 I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the vehicle merging scene recognition device to exchange data wirelessly or via wired communication with other devices. Although a vehicle merging scene recognition device with various systems is shown in the figure, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0141] 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.

[0142] The vehicle merging scene recognition device provided in this application, employing the vehicle merging scene recognition method in the above embodiments, can solve the technical problem of misjudging lane changes by vehicles as merging scenes, thereby affecting the accuracy of identifying whether historical driving data constitutes a vehicle merging scene. Compared with the prior art, the beneficial effects of the vehicle merging scene recognition device provided in this application are the same as those of the vehicle merging scene recognition method provided in the above embodiments, and other technical features of this vehicle merging scene recognition device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0143] 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.

[0144] 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.

[0145] 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 vehicle merging scene recognition method in the above embodiments.

[0146] 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.

[0147] The aforementioned computer-readable storage medium may be included in the vehicle merging scene recognition device; or it may exist independently and not be installed in the vehicle merging scene recognition device.

[0148] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the vehicle merging scene recognition device, the vehicle merging scene recognition device: determines whether the first lateral displacement of the vehicle and the vehicle to be interacted with in the current frame image exceeds a preset merging distance, wherein the current frame image is the frame image being processed in the effective segment; if it is determined that the first lateral displacement exceeds the preset merging distance, then based on the first lateral displacement and the second lateral and longitudinal displacements of the vehicle and the vehicle to be interacted with in the last frame image, it determines whether the vehicle to be interacted with is the target vehicle for the vehicle interaction; if it is determined that the vehicle to be interacted with is the target vehicle, then it determines that the historical driving data is a vehicle merging scene.

[0149] 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).

[0150] 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.

[0151] 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.

[0152] 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 vehicle merging scene recognition method. This solves the technical problem of misjudging lane changes by vehicles as vehicle merging scenes, thereby affecting the accuracy of identifying whether historical driving data represents a vehicle merging scene. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the vehicle merging scene recognition method provided in the above embodiments, and will not be repeated here.

[0153] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the vehicle merging scene recognition method described above.

[0154] The computer program product provided in this application can solve the technical problem of misjudging a lane change by a vehicle as a merging scenario, thereby affecting the accuracy of identifying whether a historical driving data contains a merging scenario. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the merging scenario identification method provided in the above embodiments, and will not be repeated here.

[0155] 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 recognizing merging vehicles, characterized in that, The method includes: Filter out valid segments of vehicles that are waiting to be interacted with when the vehicle merges in from the acquired historical driving data; Determine whether the first lateral displacement of the autonomous vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance, wherein the current frame image is the frame image being processed in the effective segment; If it is determined that the first lateral displacement exceeds the preset merging distance, then based on the first lateral displacement, and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image, it is determined whether the vehicle to be interacted with is the target vehicle of the self-vehicle interaction, and the last frame image is the last frame image in the effective segment. If the vehicle to be interacted with is determined to be the target vehicle, then the historical driving data is determined to be a scenario where the vehicle merges in.

2. The method as described in claim 1, characterized in that, Before the step of determining whether the first lateral displacement of the self-vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance, the method further includes: Based on the valid fragment, determine the vehicle trajectory of the autonomous vehicle and the object trajectory of the vehicle to be interacted with; Based on the vehicle trajectory and the object trajectory, determine whether the directional change angle of the vehicle in the effective segment meets the preset merging angle; If the angle of change of direction meets the preset merging angle, then it is determined whether the first lateral displacement exceeds the preset merging distance.

3. The method as described in claim 2, characterized in that, The step of determining whether the directional change angle of the vehicle in the effective segment satisfies the preset merging angle based on the vehicle trajectory and the object trajectory includes: Extract the trajectory length corresponding to the shorter trajectory from the vehicle trajectory and the object trajectory; Based on the trajectory length, a valid vehicle trajectory is extracted from the vehicle trajectory, and the position point of the vehicle in the current frame trajectory is determined as the starting trajectory point of the valid vehicle trajectory. Determine the starting direction angle of the starting trajectory point and the ending direction angle of the ending trajectory point in the effective vehicle trajectory; Based on the starting direction angle and the ending direction angle, the directional change angle of the vehicle in the effective segment is determined; Determine whether the angle of change of direction meets the preset merging angle.

4. The method as described in claim 3, characterized in that, The steps of determining the starting direction angle of the starting trajectory point and the ending direction angle of the ending trajectory point in the valid vehicle trajectory include: Identify the end point of the valid vehicle trajectory; The starting trajectory point is vectorized to determine the starting direction vector of the starting trajectory point, and the ending trajectory point is vectorized to determine the ending direction vector of the ending trajectory point; Determine whether the length of the first vector of the starting direction vector is less than a preset length, and determine whether the length of the second vector of the ending direction vector is less than a preset length; If the lengths of the first vector and the second vector are both not less than the preset length, then the starting direction angle of the starting trajectory point is determined based on the starting direction vector, and the ending direction angle of the ending trajectory point is determined based on the ending direction vector.

5. The method as described in claim 3, characterized in that, The step of determining the directional change angle of the vehicle in the effective segment based on the starting direction angle and the ending direction angle includes: Calculate the angle difference between the starting direction angle and the ending direction angle; The angle difference is normalized to determine the directional change angle of the vehicle in the effective segment.

6. The method as described in claim 1, characterized in that, The step of determining whether the vehicle to be interacted with is the target vehicle for interaction by the self-vehicle based on the first lateral displacement and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image includes: Determine whether the longitudinal displacement of the autonomous vehicle and the vehicle to be interacted with in the last frame of the image satisfies a preset longitudinal condition; And, whether the second lateral displacement of the self-vehicle and the vehicle to be interacted with in the last frame of the image satisfies the preset merging distance; If it is determined that the longitudinal displacement satisfies the longitudinal condition and the second lateral displacement satisfies the preset merging distance, then it is determined whether the lateral displacement difference between the first lateral displacement and the second lateral displacement is less than or equal to the preset lateral condition. If it is determined that the longitudinal displacement satisfies the longitudinal condition, and the lateral displacement difference is less than or equal to the preset lateral condition, then the vehicle to be interacted with is determined to be the target vehicle for the autonomous vehicle interaction.

7. The method as described in claim 1, characterized in that, The step of determining the historical driving data as a scenario where the vehicle to be interacted with is determined to be the target vehicle includes: If the vehicle to be interacted with is the target vehicle, then update the reference number of the target vehicle recorded. Determine whether the reference quantity meets the preset quantity condition; If the preset quantity condition is met, the historical driving data is determined to be a scenario where a vehicle merges in.

8. A vehicle merging scene recognition device, characterized in that, The device includes: The acquisition module is used to filter valid segments of vehicles that exist for interaction when the vehicle merges in from the acquired historical driving data. The first judgment module is used to determine whether the first lateral displacement of the self-vehicle and the vehicle to be interacted with in the current frame image exceeds the preset merging distance, wherein the current frame image is the frame image being processed in the effective segment; The second judgment module is used to determine whether the vehicle to be interacted with is the target vehicle of the self-vehicle if it is determined that the first lateral displacement exceeds the preset merging distance, based on the first lateral displacement and the second lateral and longitudinal displacements of the self-vehicle and the vehicle to be interacted with in the last frame image, and the last frame image is the last frame image in the effective segment. The determination module is used to determine the historical driving data as a scenario where the vehicle to be interacted with is determined to be the target vehicle.

9. A vehicle merging 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, the computer program being configured to implement the steps of the vehicle merging scene recognition method as described in any one of claims 1 to 7.

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, it implements the steps of the vehicle merging scene recognition method as described in any one of claims 1 to 7.

Images