Vehicle steering scenario recognition method and related device

Abstract

The application discloses a vehicle turning scene recognition method and a related device thereof, and relates to the technical field of vehicles. The vehicle turning scene recognition method comprises the following steps: acquiring a historical scene segment; determining a scene recognition result based on each frame of data in the historical scene segment, so as to avoid mistaking the data of a non-key road section as a turning scene (for example, slight turning of a vehicle when the vehicle is driving straight), thereby improving the accuracy of the scene recognition result; the scene recognition result can be that there is a vehicle left-turning scene in the historical scene segment, and / or there is a vehicle right-turning scene in the historical scene segment, and / or there is no vehicle turning scene in the historical scene segment; the application only needs to acquire the historical scene segment, and accurately obtains the scene recognition result about vehicle turning in the historical scene segment through the above method, so that the data about the vehicle turning scene is simply acquired.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a method for recognizing vehicle steering scenes and related equipment. Background Technology

[0002] With the development of intelligent driving technology, the demand for data from real traffic scenarios is increasing. This data from real traffic scenarios is not only used to train and validate autonomous driving algorithms, but also to improve vehicle perception, decision-making, and control algorithms.

[0003] High-quality data from real-world traffic scenarios can significantly improve the accuracy of autonomous driving algorithms, vehicle perception, decision-making, and control algorithms. However, due to technical limitations in the data collection process, as well as the relatively rare and complex nature of steering scenarios, acquiring data on vehicle steering scenarios is quite challenging. Summary of the Invention

[0004] The main purpose of this application is to provide a vehicle turning scene recognition method and related equipment, aiming to solve the technical problem of the high difficulty in acquiring data on vehicle turning scenes.

[0005] To achieve the above objectives, this application proposes a vehicle turning scene recognition method, which includes:

[0006] Obtain historical scene fragments;

[0007] When an intersection scene exists in the historical scene segment, the scene recognition result is determined based on the data of each frame in the historical scene segment;

[0008] The scene recognition results include the presence of vehicle left-turning scenes, vehicle right-turning scenes, and / or the absence of vehicle turning scenes in the historical scene segments.

[0009] In one embodiment, the step of determining the scene recognition result based on the frame data in the historical scene segment includes:

[0010] Determine whether there is a possibility of the vehicle turning between each frame from the target frame to the preset frame, wherein the target frame includes the starting frame, and the preset frame is spaced apart from the target frame by a preset frame length;

[0011] If it exists, then based on the vehicle heading angles corresponding to the target frame and the preset frame respectively, determine the change in the vehicle's driving direction in the scene sub-segment from the target frame to the preset frame;

[0012] The scene recognition result is determined based on the changes in the vehicle's driving direction.

[0013] In one embodiment, the step of determining the change in vehicle travel direction in the corresponding scene sub-segment from the target frame to the preset frame based on the vehicle heading angles corresponding to the target frame and the preset frame respectively includes:

[0014] Calculate the angle difference between the vehicle heading angles corresponding to the target frame and the preset frame, respectively;

[0015] Based on the angle difference, the change in the vehicle's driving direction in the corresponding scene sub-segment from the target frame to the preset frame is determined.

[0016] In one embodiment, after determining the change in vehicle driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference, the method further includes:

[0017] Update the target frame to a frame that slides backward a preset number of frames, return to the step of determining whether there is a possibility of the vehicle turning in each frame between the target frame and the preset frame, until the preset frame is the end frame, and obtain the changes in the vehicle's driving direction in multiple scene sub-segments corresponding to the start frame to the end frame.

[0018] In one embodiment, the step of determining the scene recognition result based on the change in the vehicle's driving direction includes:

[0019] Based on the changes in the vehicle's driving direction in multiple scene sub-segments corresponding to the start frame to the end frame, the scene recognition results of the multiple scene sub-segments are determined;

[0020] After determining the scene recognition result based on the change in the vehicle's driving direction, the method further includes:

[0021] The timing of vehicle turning is determined based on a target scene sub-segment where vehicle turning occurs.

[0022] In one embodiment, the step of determining the change in vehicle driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference includes:

[0023] The angle difference is normalized.

[0024] If the normalized angle difference is within the first angle range, then the change in the vehicle's driving direction is determined to be a left turn;

[0025] If the normalized angle difference is within the second angle range, then the change in the vehicle's driving direction is determined to be a right turn;

[0026] If the normalized angle difference is within the third angle range, then the change in the vehicle's driving direction is determined to be that no steering has occurred;

[0027] There is no overlap between the first angle interval, the second angle interval, and the third angle interval.

[0028] In one embodiment, the step of determining whether there is a possibility of the vehicle turning in each frame between the target frame and the preset frame includes:

[0029] Determine whether the vehicle speed corresponding to the target frame is less than a preset speed threshold;

[0030] If the vehicle speed is greater than or equal to a preset speed threshold, the vehicle travel distance from the target frame to the preset frame is calculated based on the vehicle position information corresponding to each frame.

[0031] Determine whether the distance traveled by the vehicle is less than a first distance threshold;

[0032] If the vehicle travels a distance greater than or equal to a first distance threshold, it is determined that there is a possibility that the vehicle in each frame between the target frame and the preset frame may be turning.

[0033] In one embodiment, after the step of acquiring historical scene fragments, the method further includes:

[0034] Determine whether intersection signage information exists in each frame of data;

[0035] If it exists, the target distance between the intersection and the vehicle is calculated based on the intersection signage information;

[0036] When the target distance is less than the second distance threshold, it is determined that there is an intersection scene in the historical scene segment.

[0037] In one embodiment, the intersection signage information includes traffic light information, the traffic light information includes first location information of at least one traffic light, and the step of calculating the target distance between the intersection and the vehicle based on the intersection signage information includes:

[0038] Based on the first location information and the vehicle location information corresponding to each frame of data, the Euclidean distance between the at least one traffic light and the vehicle, and the first relative direction between the at least one traffic light and the vehicle are calculated.

[0039] Based on the angle between the first relative direction and the target direction of the vehicle's driving speed corresponding to each frame of data, it is determined whether there is a target traffic light in front of the vehicle.

[0040] If it exists, the minimum distance in the Euclidean distance corresponding to the target traffic light is taken as the target distance between the intersection and the vehicle.

[0041] In one embodiment, the intersection signage information further includes zebra crossing information, the zebra crossing information including second position information corresponding to any corner point of at least one zebra crossing corresponding to a rectangle, and the step of calculating the target distance between the intersection and the vehicle based on the intersection signage information includes:

[0042] Based on the second location information and the vehicle location information corresponding to each frame of data, the projected distance between the vehicle and the at least one zebra crossing, and the second relative direction between the at least one zebra crossing and the vehicle are calculated.

[0043] Based on the angle between the second relative direction and the target direction of the vehicle's driving speed corresponding to each frame of data, it is determined whether there is a target zebra crossing in front of the vehicle.

[0044] If it exists, the minimum distance among the projected distances corresponding to the target zebra crossing will be taken as the target distance between the intersection and the vehicle.

[0045] Furthermore, to achieve the above objectives, this application also proposes a vehicle turning scene recognition device, which includes:

[0046] The acquisition module is used to acquire historical scene fragments;

[0047] The recognition module is used to determine the scene recognition result based on the data of each frame in the historical scene segment, wherein the scene recognition result includes the presence of a vehicle turning left scene, a vehicle turning right scene, and / or the absence of a vehicle turning scene in the historical scene segment.

[0048] In addition, to achieve the above objectives, this application also proposes a vehicle turning scene recognition device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the vehicle turning scene recognition method described above.

[0049] 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 turning scene recognition method described above.

[0050] 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 steering scene recognition method described above.

[0051] One or more technical solutions proposed in this application have at least the following technical effects:

[0052] This application obtains historical scene fragments; when an intersection scene exists in the historical scene fragment, it determines the scene recognition result based on the data of each frame in the historical scene fragment, avoiding misidentifying data of non-critical road sections as turning scenes (e.g., slight turning of a vehicle while traveling straight), thereby improving the accuracy of the scene recognition result; the scene recognition result may be that a vehicle turns left in the historical scene fragment, and / or a vehicle turns right in the historical scene fragment, and / or no vehicle turns in the historical scene fragment; this application only needs to obtain historical scene fragments and accurately obtain the scene recognition result of vehicle turning in the historical scene fragments through the above method, thereby easily obtaining data about vehicle turning scenes. Attached Figure Description

[0053] 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.

[0054] 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.

[0055] Figure 1 This is a flowchart illustrating an embodiment of the vehicle turning scene recognition method of this application.

[0056] Figure 2 A simplified flowchart illustrating the vehicle turning scene recognition method provided in Embodiment 1 of this application;

[0057] Figure 3 This is a flowchart illustrating Embodiment 2 of the vehicle turning scene recognition method of this application;

[0058] Figure 4 This is a flowchart illustrating Embodiment 3 of the vehicle turning scene recognition method of this application;

[0059] Figure 5 This is a schematic diagram of the module structure of the vehicle steering scene recognition device according to an embodiment of this application;

[0060] Figure 6 This is a schematic diagram of the device structure of the hardware operating environment involved in the vehicle turning scene recognition method in this application embodiment.

[0061] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0062] 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.

[0063] 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.

[0064] 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 vehicle steering scene recognition device capable of performing the above functions. The following description uses a vehicle steering scene recognition device as an example to illustrate this embodiment and the subsequent embodiments.

[0065] Based on this, embodiments of this application provide a vehicle turning scene recognition method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the vehicle turning scene recognition method of this application.

[0066] In this embodiment, the vehicle turning scene recognition method includes steps S10~S20:

[0067] Step S10: Obtain historical scene fragments;

[0068] It should be noted that with the development of intelligent driving technology, the demand for data from real-world traffic scenarios is increasing. This data is not only used to train and validate autonomous driving algorithms, but also to improve vehicle perception, decision-making, and control algorithms. High-quality real-world traffic scenario data can significantly improve the accuracy of autonomous driving algorithms, vehicle perception, decision-making, and control algorithms; however, due to technical limitations involved in data collection, and the relatively rare and complex nature of steering scenarios, obtaining data on vehicle steering scenarios is quite difficult.

[0069] To address the aforementioned issues, this embodiment acquires historical scene fragments to identify offline whether vehicle turning scenarios exist within these fragments. This allows the historical scene fragments with scene recognition results to serve as effective sample data for training and validating autonomous driving algorithms, vehicle perception algorithms, driving decision-making algorithms, and vehicle control algorithms.

[0070] Specifically, historical scene fragments can be historically collected vehicle driving records or vehicle driving information generated through simulation. Historical scene fragments include multiple frames of data, and the scene information corresponding to each frame of data can include dynamic perception information collected by sensors such as cameras and radar, or static map information obtained by matching historically collected vehicle latitude and longitude with high-precision maps. Among them, dynamic perception information includes historically collected road images, timestamps, vehicle position information (e.g., vehicle coordinates in the global coordinate system), vehicle speed and acceleration, vehicle direction and steering angle, vehicle travel direction, vehicle steering wheel angle, driver operation (e.g., braking, accelerator force, gear shifting, etc.), etc., and static map information includes traffic sign information on the road, the position and behavior of other vehicles, pedestrians, obstacles, etc.

[0071] Step S20: When there is an intersection scene in the historical scene segment, the scene recognition result is determined based on the data of each frame in the historical scene segment. The scene recognition result includes the presence of a vehicle turning left scene, a vehicle turning right scene, and / or the absence of a vehicle turning scene in the historical scene segment.

[0072] Since vehicle turning operations usually occur at intersections, in order to improve the recognition efficiency of vehicle turning scenarios for historical scene segments, this embodiment only continues to recognize vehicle turning scenarios for historical scene segments when it is determined that there is an intersection scenario in the historical scene segment; if there is no intersection scenario in the historical scene segment, vehicle turning scenario recognition is not performed on the historical scene segment, and the historical scene segment is abandoned as training data.

[0073] Specifically, the specific implementation method for determining the scene recognition result of the historical scene segment based on the data of each frame in the historical scene segment can be: determining the scene recognition result of the historical scene segment based on the dynamic perception information and static map information corresponding to each frame in the historical scene segment.

[0074] The scene recognition results of the historical scene segments include whether there is a vehicle turning left, a vehicle turning right, or no vehicle turning scene in the historical scene segments. Therefore, historical scene segments with scene recognition results can be used as training data.

[0075] Since historical scene segments include multiple frames of data, the scene recognition results of historical scene segments can also be that each frame of data contains a scene of vehicles turning left, a scene of vehicles turning right, and / or no scene of vehicles turning. In order to increase the amount of data, each frame of data with scene recognition results can be used as training data.

[0076] Specifically, please refer to Figure 2Step S20 includes steps A1 to A3:

[0077] Step A1: Determine whether there is a possibility of the vehicle turning between each frame from the target frame to the preset frame. The target frame includes the starting frame, and the preset frame is spaced apart from the target frame by a preset frame length.

[0078] Since historical scene segments include multiple frames of data, in order to improve data processing efficiency, we can first determine whether there is a possibility of the vehicle turning between each frame from the target frame to the preset frame. If there is no possibility of turning, we can skip further identification of the scene sub-segments from the target frame to the preset frame, thereby saving computing resources and time.

[0079] In this context, assuming the total number of frames in the historical scene segment is N, the preset frame length (frame_size) is determined based on the time required for the vehicle to complete one turning operation. Specifically, the preset frame length can be 50 frames or 100 frames, etc. The target frame (frame_i) includes the starting frame and other frames, and the preset frame is spaced apart from the target frame by a preset frame length. It can be understood that when the target frame is frame 0, the preset frame can be frame 100, and when the target frame is frame 20, the preset frame can be frame 120.

[0080] That is, to determine whether there is a possibility of the vehicle turning between the target frame and the preset frame.

[0081] Specifically, a feasible implementation of determining whether there is a possibility of the vehicle turning in each frame between the target frame and the preset frame could be:

[0082] Determine whether the vehicle speed corresponding to the target frame is less than a preset speed threshold; if the vehicle speed is greater than or equal to the preset speed threshold, calculate the vehicle travel distance from the target frame to the preset frame based on the vehicle position information corresponding to each frame; determine whether the vehicle travel distance is less than a first distance threshold; if the vehicle travel distance is greater than or equal to the first distance threshold, determine that there is a possibility that the vehicles corresponding to each frame between the target frame and the preset frame are turning.

[0083] It should be noted that since a vehicle can only perform a steering operation when its speed meets certain conditions, this embodiment determines whether the vehicle speed corresponding to the target frame is less than a preset speed threshold (vel_static) to exclude the case where the vehicle is stationary, thereby avoiding further steering judgments and reducing the waste of computing resources.

[0084] The preset speed threshold (distance_static) can be 1.0m / s or 1.5m / s, etc. If the vehicle speed corresponding to the 0th frame is less than 1.0m / s, it is considered that the vehicle corresponding to the 0th frame is stationary and no further scene recognition is performed on that frame.

[0085] Furthermore, if the vehicle speed corresponding to the target frame is greater than or equal to a preset speed threshold, the vehicle travel distance from the target frame to the preset frame is calculated based on the vehicle position information corresponding to each frame. It is then determined whether the vehicle travel distance is less than a first distance threshold to exclude the case of low-speed vehicle travel, avoid further steering judgment, and reduce the waste of computing resources.

[0086] The first distance threshold (distance_static) can be 5.0m or 6m, etc. If the vehicle travel distance from frame 0 to frame 100 is less than 5.0m, it is considered that the vehicle is in a low-speed driving condition from frame 0 to frame 100, and no further scene recognition is performed on that frame.

[0087] Specifically, the vehicle position information includes vehicle coordinate information, which can be the vehicle coordinates coordinate_start(start_x, start_y) in the target frame (frame_i) and the vehicle coordinates coordinate_end(end_x, end_y) in the preset frame (frame_i+100). Based on the vehicle position information corresponding to each frame, the specific implementation of calculating the vehicle travel distance from the target frame to the preset frame can be: calculating the straight-line distance dist between coordinate_start and coordinate_end. If dist is less than distance_static, it is considered that the vehicle is in a low-speed driving condition from the target frame to the preset frame.

[0088] Furthermore, if the vehicle travel distance is greater than or equal to the first distance threshold, it is determined that there is a possibility that the vehicle is turning in each frame between the target frame and the preset frame, and then the presence of a vehicle turning scene in each frame between the target frame and the preset frame is identified.

[0089] It is understood that by setting a preset speed threshold and a first distance threshold, this embodiment can effectively filter out false steering signals caused by sensor noise or slight vibration, ensuring that only genuine steering behavior is identified.

[0090] Step A2: If it exists, then based on the vehicle heading angles corresponding to the target frame and the preset frame respectively, determine the change in the vehicle's driving direction in the scene sub-segment from the target frame to the preset frame.

[0091] It should be noted that the vehicle heading angle can be obtained through the vehicle box or the vehicle speed. Specifically, the vehicle bounding box is usually detected by visual or distance sensors such as cameras and radar, so the vehicle heading angle can be calculated based on the vehicle bounding box. The vehicle heading angle can also be derived by analyzing the vehicle's velocity vector (including velocity magnitude and direction).

[0092] Since the vehicle heading angle can reflect the change in the vehicle's driving direction, if it is determined that there is a possibility that the vehicle is turning between the target frame and the preset frame, then based on the vehicle heading angles corresponding to the target frame and the preset frame, the change in the vehicle's driving direction in the scene sub-segment from the target frame to the preset frame is determined.

[0093] Step A3: Determine the scene recognition result based on the change in the vehicle's driving direction.

[0094] Furthermore, by analyzing changes in the vehicle's direction of travel, it is possible to determine whether a vehicle is in a left-turn scenario, a right-turn scenario, and / or whether a turning scenario is not present, thus obtaining the scene recognition result.

[0095] In this embodiment, historical scene segments are acquired; based on the data of each frame in the historical scene segments, scene recognition results are determined to avoid mistaking data from non-critical road sections for steering scenarios (e.g., slight steering of a vehicle while driving straight). By setting a preset speed threshold and a first distance threshold, false steering signals caused by sensor noise or slight vibrations can be effectively filtered out, ensuring that only genuine steering behavior is recognized. This embodiment only needs to acquire historical scene segments and accurately obtain the scene recognition results of vehicle steering in the historical scene segments through the above method, thereby easily obtaining data about vehicle steering scenarios.

[0096] 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 3 Step A2 includes steps B1 to B2:

[0097] Step B1: Calculate the angle difference between the vehicle heading angles corresponding to the target frame and the preset frame respectively;

[0098] It should be noted that when it is determined that there is a possibility of vehicle turning between the target frame and the preset frame, it is still impossible to determine whether the vehicle turning has actually occurred or not. This embodiment accurately captures the change in vehicle driving direction by calculating the angle difference between the vehicle heading angles corresponding to the target frame and the preset frame respectively.

[0099] For example, calculate the angle difference (theta_end - theta_start) between the vehicle heading angle (theta_start) corresponding to the target frame (frame_i) and the vehicle heading angle (theta_end) corresponding to the preset frame (frame_i+100).

[0100] Step B2: Based on the angle difference, determine the change in the vehicle's driving direction in the corresponding scene sub-segment from the target frame to the preset frame;

[0101] Since the angle difference can reflect the change in the vehicle's driving direction from the target frame to the preset frame, this embodiment can determine the change in the vehicle's driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference.

[0102] For example, if the angle difference is within a certain range, the vehicle can be considered to have turned; if it is outside that range, the vehicle can be considered not to have turned.

[0103] Alternatively, the specific implementation method for determining the change in vehicle driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference can also be:

[0104] The angle difference is normalized; if the normalized angle difference is within the first angle interval, the change in the vehicle's driving direction is determined to be a left turn; if the normalized angle difference is within the second angle interval, the change in the vehicle's driving direction is determined to be a right turn; if the normalized angle difference is within the third angle interval, the change in the vehicle's driving direction is determined to be no turn; wherein, there is no intersection between the first angle interval, the second angle interval, and the third angle interval.

[0105] It should be noted that, in scenarios where a vehicle turns left or right, the angle difference between the vehicle heading angles corresponding to the target frame and the preset frame may be - 2 3 6 In order to facilitate the subsequent determination of the steering type corresponding to the angle difference, this embodiment normalizes the angle difference to make the determination of the change in vehicle driving direction more intuitive, simple and accurate.

[0106] The normalization of the angle difference can be achieved by normalizing the angle difference to the same quantization standard, for example, by normalizing the angle difference to [0, 2]. )or(- , ) etc.; specifically, the angle difference is normalized to [0, 2 The calculation formula within the range can be: Normalized angle difference (theta_diff) = (theta_end - theta_start + 2) )% (2 ).

[0107] Correspondingly, a first angle interval, a second angle interval, and a third angle interval can be set to represent the intervals containing the corresponding angle differences in the scene. There is no overlap between the first angle interval, the second angle interval, and the third angle interval. It can be understood that the first angle interval can be... The second angular interval can be The third angle interval can be and That is, there is no intersection between the first angle interval, the second angle interval, and the third angle interval.

[0108] It is understandable that, since the angle difference between the vehicle heading angles corresponding to the target frame and the preset frame is normalized to the same quantization standard, the change in the vehicle's driving direction can be determined simply by comparing the angles, without causing - 6 In situations where the same standard cannot be directly used for comparison at the same angle, this improves data processing efficiency.

[0109] Specifically, if the normalized angle difference is within the first angle range, the change in the vehicle's driving direction is determined to be a left turn; if the normalized angle difference is within the second angle range, the change in the vehicle's driving direction is determined to be a right turn; if the normalized angle difference is within the third angle range, the change in the vehicle's driving direction is determined to be no turn.

[0110] Furthermore, after determining the change in vehicle driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference, the following steps can be taken: update the target frame to a frame that slides backward a preset number of frames, return to the step of determining whether there is a possibility of vehicle turning in each frame between the target frame and the preset frame, until the preset frame is the end frame, thereby obtaining the change in vehicle driving direction in multiple scene sub-segments from the start frame to the end frame.

[0111] Since the above only determines whether there is a possibility of vehicle turning between each frame from the target frame to the preset frame, and the change of vehicle driving direction in the scene sub-segment between the target frame and the preset frame, in order to determine the change of vehicle driving direction in multiple scene sub-segments from the start frame to the end frame, the target frame can be updated to a frame that slides backward a preset number of frames, and the step of determining whether there is a possibility of vehicle turning between each frame from the target frame to the preset frame can be returned until the preset frame is the end frame.

[0112] The preset frame number can be less than the frame number corresponding to the preset frame length, thereby increasing the number of scene sub-segments identified without missing the identification of each frame of data in the historical scene segments; the preset frame number can also be equal to the frame number corresponding to the preset frame length, reducing repeated scenes and improving scene recognition efficiency; the preset frame number can also be greater than the frame number corresponding to the preset frame length but less than (N-100 frames), thereby reducing the number of scene sub-segments to be identified and improving data processing efficiency.

[0113] In order to increase the number of identified scene sub-segments without missing the identification of each frame of data in the historical scene segments, this embodiment can return to the step of judging whether there is a possibility of the vehicle turning between each frame from the target frame to the preset frame every 20 frames or every 50 frames, until the preset frame is the end frame (that is, until all frames have been traversed).

[0114] Furthermore, the specific implementation of determining the scene recognition result based on the change in the vehicle's driving direction can be: determining the scene recognition result of the multiple scene sub-segments based on the change in the vehicle's driving direction in the multiple scene sub-segments corresponding to the start frame to the end frame.

[0115] It can be understood that if the vehicle's direction of travel changes to left in the scene sub-segment from frame 0 to frame 100, and no turning occurs in the scene sub-segment from frame 20 to frame 120, ..., and the vehicle's direction of travel changes to right in the scene sub-segment from frame (N-100) to frame N; then the corresponding scene recognition results from frame 0 to frame 100 are that there is a vehicle turning left, from frame 20 to frame 120 are that there is no vehicle turning, ..., and from frame (N-100) to frame N are that there is a vehicle turning right.

[0116] Furthermore, after determining the scene recognition result based on the change in the vehicle's driving direction, it is also possible to: determine the vehicle turning time based on the target scene sub-segment where a vehicle turning scene exists.

[0117] Since the changes in the vehicle's driving direction in multiple scene sub-segments from the start frame to the end frame are obtained through the above method, the turning time corresponding to the target scene sub-segment where the vehicle turns can be determined.

[0118] For example, the target scene sub-segments containing vehicle turning scenarios include the scene sub-segments corresponding to frames 0 to 100, frames 80 to 180, ..., and the scene sub-segments corresponding to frames (N-100) to N. In order to obtain high-quality training data, the vehicle turning time (including the vehicle left turn time and the vehicle right turn time) can be determined based on the above target scene sub-segments containing vehicle turning scenarios. For example, the vehicle left turn time is from frame 0 to 100, and the vehicle right turn time is from frame (N-100) to N.

[0119] This embodiment can quickly locate vehicle left-turn and vehicle right-turn scenarios in historical scene segments using the above method, and can accurately locate the time of vehicle left-turn and vehicle right-turn; thereby obtaining more detailed data about vehicle turning scenarios.

[0120] Based on the first and second embodiments of this application, in the third 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 4 After step S10, steps C1 to C3 are also included:

[0121] Step C1: Determine whether intersection signage information exists in each frame of data;

[0122] To improve the efficiency and accuracy of intersection scene recognition, this embodiment first determines whether there is intersection signage information in each frame of data; if it exists, the probability that there is an intersection in the scene corresponding to that frame is higher, and if it does not exist, the probability that there is an intersection in the scene corresponding to that frame is lower.

[0123] The specific implementation method for determining whether there is intersection sign information in each frame of data can be: based on the road image in the dynamic perception information, identify whether there is intersection sign information in each frame of data through a preset target recognition model; or it can be to determine whether there is intersection sign information in the traffic sign information in the static map information corresponding to each frame of data.

[0124] The intersection signage information can include speed bumps, intersection warning signs, traffic lights (red and green lights), zebra crossings, etc.

[0125] Step C2: If it exists, calculate the target distance between the intersection and the vehicle based on the intersection sign information;

[0126] It is understandable that since vehicles usually turn when they are about to reach an intersection, in order to effectively filter out frames that are unlikely to turn, this embodiment calculates the target distance between the intersection and the vehicle based on the intersection sign information.

[0127] Specifically, for each frame of data, there may be traffic light information and / or zebra crossing information. If a frame does not have traffic light information traffic_lights_infos, then a large positive value (such as 10000) is assigned to the distance from the traffic light to the vehicle to indicate that there is no intersection ahead.

[0128] If traffic light information exists in a certain frame, then the specific implementation method for calculating the target distance between the intersection and the vehicle based on the intersection sign information can be:

[0129] Based on the first location information and the vehicle location information corresponding to each frame of data, calculate the Euclidean distance between the at least one traffic light and the vehicle, and the first relative direction between the at least one traffic light and the vehicle; based on the angle between the first relative direction and the target direction of the vehicle's driving speed corresponding to each frame of data, determine whether there is a target traffic light in front of the vehicle; if there is, take the minimum distance in the Euclidean distance corresponding to the target traffic light as the target distance between the intersection and the vehicle.

[0130] It is understandable that, for each frame of data, in this embodiment, when there is traffic light information in the road perception information, since there may be multiple sets of traffic light information at the same time, this embodiment can take the minimum distance from the vehicle to each traffic light as the target distance.

[0131] It should be noted that the traffic light information includes the first position information of at least one traffic light, which may be the first coordinate of at least one traffic light in the global coordinate system.

[0132] For at least one traffic light in each frame of data, the Euclidean distance between the at least one traffic light and the vehicle can be calculated based on the first location information and the vehicle location information corresponding to each frame of data. At the same time, the first relative direction between the at least one traffic light and the vehicle can be calculated based on the first location information and the vehicle location information corresponding to each frame of data.

[0133] Specifically, the Euclidean distance between the first coordinate and the vehicle coordinate corresponding to at least one traffic light can be calculated, and the direction from at least one traffic light to the vehicle can be taken as the first relative direction (Traffic_light_direction_vector, i.e., represented by a vector composed of the first coordinate and the vehicle coordinate).

[0134] Based on the angle (vector angle) between the first relative direction and the target direction (V, i.e., represented by a vector with velocity direction) of the vehicle driving speed corresponding to each frame of data, it is determined whether there is a target traffic light in front of the vehicle.

[0135] It can be understood that if the absolute value of the angle between the vectors corresponding to the traffic light is less than 90 degrees, the traffic light is considered to be in front of the vehicle; if the absolute value of the angle between the vectors corresponding to the traffic light is greater than or equal to 90 degrees, the traffic light is considered to be behind the vehicle.

[0136] To improve judgment efficiency, this embodiment uses the minimum distance (ego_to_traffic_dist) in the Euclidean distance corresponding to the target traffic light in front of the vehicle as the target distance between the intersection and the vehicle.

[0137] If a frame does not contain crosswalk information (crosswalks_infos), the distance from the traffic light to the vehicle (ego_to_crosswalks_dist) is assigned a large positive value (e.g., 10000) to indicate that there is no intersection ahead.

[0138] If zebra crossing information exists in a certain frame, then the specific implementation method for calculating the target distance between the intersection and the vehicle based on the intersection sign information can be:

[0139] Based on the second location information and the vehicle location information corresponding to each frame of data, the projected distance between the vehicle and the at least one zebra crossing, and the second relative direction between the at least one zebra crossing and the vehicle are calculated; based on the angle between the second relative direction and the target direction of the vehicle's driving speed corresponding to each frame of data, it is determined whether there is a target zebra crossing in front of the vehicle; if there is, the minimum distance among the projected distances corresponding to the target zebra crossing is taken as the target distance between the intersection and the vehicle.

[0140] It is understandable that for each frame of data, in this embodiment, when there is zebra crossing information in the road perception information, since there may be multiple sets of zebra crossing information at the same time, this embodiment can take the minimum distance from the vehicle to each zebra crossing as the target distance.

[0141] It should be noted that the zebra crossing includes at least one second position information corresponding to any corner point of the zebra crossing's rectangular frame. The second position information can be the second coordinate of any corner point in the global coordinate system.

[0142] For any corner point of the rectangle corresponding to at least one zebra crossing in each frame of data, the projected distance between the vehicle and the at least one zebra crossing can be calculated based on the second position information and the vehicle position information corresponding to each frame of data. At the same time, the second relative direction between at least one zebra crossing and the vehicle (crosswalks_direction_vector, i.e., a vector composed of the second coordinates and the vehicle coordinates) can be calculated based on the second position information and the vehicle position information corresponding to each frame of data.

[0143] Specifically, based on the angle (vector angle crosswalks_angle) between the second relative direction and the target direction (V, i.e., represented by a vector with velocity direction) of the vehicle driving speed corresponding to each frame of data, it is determined whether there is a target zebra crossing in front of the vehicle.

[0144] It is understandable that if the absolute value of the crosswalks_angle, the vector corresponding to the zebra crossing, is less than 90 degrees, then the zebra crossing is considered to be in front of the vehicle; if the absolute value of the crosswalks_angle, the vector corresponding to the zebra crossing, is greater than or equal to 90 degrees, then the zebra crossing is considered to be behind the vehicle.

[0145] To improve judgment efficiency, this embodiment uses the minimum distance (ego_to_crosswalks_dist) among the projected distances of the target zebra crossing in front of the vehicle as the target distance between the intersection and the vehicle.

[0146] If a frame contains both traffic light information and zebra crossing information, the minimum distance in the projected distance corresponding to the target zebra crossing can be used as the target distance between the intersection and the vehicle, or the minimum distance in the Euclidean distance corresponding to the target traffic light can be used as the target distance between the intersection and the vehicle, or the smaller distance between the minimum distance in the projected distance and the minimum distance in the Euclidean distance can be used as the target distance.

[0147] Step C3: When the target distance is less than the second distance threshold, it is determined that there is an intersection scene in the historical scene segment.

[0148] It is understood that if the minimum distance in the projected distance corresponding to the target zebra crossing or the minimum distance in the Euclidean distance corresponding to the target traffic light is less than the second distance threshold (valid_dist_threshold, for example, 1 meter or 2 meters), then the vehicle is considered to be close enough to the intersection, and the intersection is considered to be a valid intersection.

[0149] That is, if ego_to_traffic_dist is less than valid_dist_threshold, or ego_to_crosswalks_dist is less than valid_dist_threshold, then the frame is considered to have a valid intersection; otherwise, the frame does not have a valid intersection.

[0150] If no valid intersections exist in any frame of a historical scene segment, then no further vehicle turning scene recognition will be performed on that historical scene segment.

[0151] This embodiment accurately identifies whether there are valid intersections in historical scene segments through the above method, thereby improving the accuracy of data acquisition regarding vehicle turning scenarios.

[0152] This application also provides a vehicle turning scene recognition device, please refer to... Figure 5 The vehicle turning scene recognition device includes:

[0153] Module 10 is used to acquire historical scene fragments;

[0154] The recognition module 20 is used to determine the scene recognition result based on the data of each frame in the historical scene segment, wherein the scene recognition result includes the presence of a vehicle turning left scene, a vehicle turning right scene, and / or the absence of a vehicle turning scene in the historical scene segment.

[0155] Optionally, the identification module 20 includes:

[0156] The first judgment submodule is used to determine whether there is a possibility of the vehicle turning between each frame from the target frame to the preset frame. The target frame includes the starting frame, and the preset frame is spaced apart from the target frame by a preset frame length.

[0157] The determination submodule is used, if it exists, to determine the change in the vehicle's driving direction in the scene sub-segment from the target frame to the preset frame based on the vehicle heading angles corresponding to the target frame and the preset frame respectively;

[0158] The recognition submodule is used to determine the scene recognition result based on the changes in the vehicle's driving direction.

[0159] Optionally, the determining submodule includes:

[0160] The first calculation unit is used to calculate the angle difference between the vehicle heading angles corresponding to the target frame and the preset frame, respectively.

[0161] The determining unit is used to determine the change in the vehicle's driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference.

[0162] Optionally, after the step of determining the change in vehicle driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference, the determining sub-module further includes:

[0163] The return unit is used to update the target frame to a frame that slides backward a preset number of frames, and return to the step of determining whether there is a possibility of the vehicle turning in each frame between the target frame and the preset frame, until the preset frame is the end frame, so as to obtain the change of the vehicle's driving direction in multiple scene sub-segments corresponding to the start frame to the end frame.

[0164] Optionally, the identification submodule includes:

[0165] The recognition unit is used to determine the scene recognition result of the multiple scene sub-segments based on the changes in the vehicle's driving direction in the multiple scene sub-segments corresponding to the start frame to the end frame;

[0166] The step of determining the scene recognition result based on the change in the vehicle's driving direction further includes:

[0167] The first determining module is used to determine the vehicle turning time based on a target scene sub-segment where a vehicle turning scenario exists.

[0168] Optionally, the determining unit includes:

[0169] A normalization subunit is used to normalize the angle difference;

[0170] The first determining subunit is used to determine that the change in the vehicle's driving direction is a left turn if the normalized angle difference is within the first angle range.

[0171] The second determining subunit is used to determine that the change in the vehicle's driving direction is a right turn if the normalized angle difference is within the second angle range.

[0172] The third determining subunit is used to determine that the change in the vehicle's driving direction is no turning if the normalized angle difference is within the third angle interval.

[0173] There is no overlap between the first angle interval, the second angle interval, and the third angle interval.

[0174] Optionally, the first determination submodule includes:

[0175] The first judgment unit is used to determine whether the vehicle speed corresponding to the target frame is less than a preset speed threshold.

[0176] The second calculation unit is used to calculate the vehicle travel distance from the target frame to the preset frame based on the vehicle position information corresponding to each frame if the vehicle speed is greater than or equal to a preset speed threshold.

[0177] The second judgment unit is used to determine whether the vehicle's travel distance is less than the first distance threshold.

[0178] The third calculation unit is used to determine the possibility that the vehicle may turn between the target frame and the preset frame if the vehicle's travel distance is greater than or equal to the first distance threshold.

[0179] Optionally, after the step of acquiring historical scene fragments, the method further includes:

[0180] The judgment module is used to determine whether intersection sign information exists in each frame of data;

[0181] The calculation module is used to calculate the target distance between the intersection and the vehicle based on the intersection signage information, if such a distance exists.

[0182] The second determining module is used to determine that an intersection scene exists in the historical scene segment when the target distance is less than a second distance threshold.

[0183] Optionally, the intersection signage information includes traffic light information, the traffic light information includes first location information of at least one traffic light, and the calculation module includes:

[0184] The first calculation submodule is used to calculate the Euclidean distance between the at least one traffic light and the vehicle, and the first relative direction between the at least one traffic light and the vehicle, based on the first location information and the vehicle location information corresponding to each frame of data.

[0185] The second judgment submodule is used to determine whether there is a target traffic light in front of the vehicle based on the angle between the first relative direction and the target direction of the vehicle driving speed corresponding to each frame of data.

[0186] The first setting submodule is used to, if present, take the minimum distance in the Euclidean distance corresponding to the target traffic light as the target distance between the intersection and the vehicle.

[0187] Optionally, the intersection signage information further includes zebra crossing information, which includes second position information corresponding to any corner point of at least one zebra crossing-corresponding rectangle. The calculation module includes:

[0188] The second calculation submodule is used to calculate the projected distance between the vehicle and the at least one zebra crossing, and the second relative direction between the at least one zebra crossing and the vehicle, based on the second location information and the vehicle location information corresponding to each frame of data.

[0189] The third judgment submodule is used to determine whether there is a target zebra crossing in front of the vehicle based on the angle between the second relative direction and the target direction of the vehicle driving speed corresponding to each frame of data.

[0190] The second setting submodule is used to, if it exists, take the minimum distance among the projected distances corresponding to the target zebra crossing as the target distance between the intersection and the vehicle.

[0191] The vehicle steering scene recognition device provided in this application, employing the vehicle steering scene recognition method in the above embodiments, can solve the technical problem of the high difficulty in acquiring data related to vehicle steering scenes. Compared with the prior art, the beneficial effects of the vehicle steering scene recognition device provided in this application are the same as those of the vehicle steering scene recognition method provided in the above embodiments, and other technical features in the vehicle steering scene recognition device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0192] This application provides a vehicle steering 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 steering scene recognition method in the above embodiment 1.

[0193] The following is for reference. Figure 6 The diagram illustrates a structural schematic of a vehicle turning scene recognition device suitable for implementing embodiments of this application. The vehicle turning scene recognition device in this application may include, but is not limited to, mobile terminals such as mobile phones, tablets, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 6 The vehicle turning 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.

[0194] like Figure 6As shown, the vehicle steering 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 steering 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 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. The communication device 1009 allows the vehicle steering scene recognition device to communicate wirelessly or wiredly with other devices to exchange data. Although the figures show vehicle steering scene 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.

[0195] 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.

[0196] The vehicle turning scene recognition device provided in this application, employing the vehicle turning scene recognition method in the above embodiments, can solve the technical problem of the high difficulty in acquiring data related to vehicle turning scenes. Compared with the prior art, the beneficial effects of the vehicle turning scene recognition device provided in this application are the same as those of the vehicle turning scene recognition method provided in the above embodiments, and other technical features in this vehicle turning scene recognition device are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0197] 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.

[0198] 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 technical scope 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 protection of the claims.

[0199] 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 steering scene recognition method in the above embodiments.

[0200] 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.

[0201] The aforementioned computer-readable storage medium may be included in the vehicle steering scene recognition device; or it may exist independently and not be assembled into the vehicle steering scene recognition device.

[0202] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the vehicle steering scene recognition device, cause the vehicle steering scene recognition device to perform the aforementioned vehicle steering scene recognition method.

[0203] 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).

[0204] 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.

[0205] 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.

[0206] 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 steering scene recognition method, which can solve the technical problem of the high difficulty in obtaining data on vehicle steering scenes. 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 vehicle steering scene recognition method provided in the above embodiments, and will not be repeated here.

[0207] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the vehicle steering scene recognition method described above.

[0208] The computer program product provided in this application can solve the technical problem of the high difficulty in acquiring data on vehicle steering scenarios. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the vehicle steering scenario recognition method provided in the above embodiments, and will not be repeated here.

[0209] 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 content of this application specification and drawings, 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 vehicle turning scenes, characterized in that, The vehicle turning scene recognition method includes: Obtain historical scene fragments; When there is an intersection scene in the historical scene segment, it is determined whether the vehicle speed corresponding to the target frame is less than a preset speed threshold. The target frame includes the starting frame, and the preset frame is spaced apart from the target frame by a preset frame length. If the vehicle speed is greater than or equal to a preset speed threshold, the vehicle travel distance from the target frame to the preset frame is calculated based on the vehicle position information corresponding to each frame. Determine whether the distance traveled by the vehicle is less than a first distance threshold; If the vehicle travels a distance greater than or equal to a first distance threshold, it is determined that there is a possibility that the vehicle in each frame between the target frame and the preset frame may be turning. If there is no possibility of turning, then skip the identification of the scene sub-segment corresponding to the target frame to the preset frame; If it exists, then based on the vehicle heading angles corresponding to the target frame and the preset frame respectively, determine the change in the vehicle's driving direction in the scene sub-segment from the target frame to the preset frame; Based on the changes in the vehicle's driving direction, a scene recognition result is determined, wherein the scene recognition result includes the presence of a vehicle turning left scene, a vehicle turning right scene, and / or the absence of a vehicle turning scene in the historical scene segment.

2. The vehicle turning scene recognition method as described in claim 1, characterized in that, The step of determining the change in vehicle travel direction in the corresponding scene sub-segment from the target frame to the preset frame based on the vehicle heading angles corresponding to the target frame and the preset frame respectively includes: Calculate the angle difference between the vehicle heading angles corresponding to the target frame and the preset frame, respectively; Based on the angle difference, the change in the vehicle's driving direction in the corresponding scene sub-segment from the target frame to the preset frame is determined.

3. The vehicle turning scene recognition method as described in claim 2, characterized in that, After determining the change in vehicle driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference, the method further includes: Update the target frame to a frame that slides backward a preset number of frames, return to the step of determining whether there is a possibility of the vehicle turning in each frame between the target frame and the preset frame, until the preset frame is the end frame, and obtain the changes in the vehicle's driving direction in multiple scene sub-segments corresponding to the start frame to the end frame.

4. The vehicle turning scene recognition method as described in claim 3, characterized in that, The step of determining the scene recognition result based on the change in the vehicle's driving direction includes: Based on the changes in the vehicle's driving direction in multiple scene sub-segments corresponding to the start frame to the end frame, the scene recognition results of the multiple scene sub-segments are determined; After determining the scene recognition result based on the change in the vehicle's driving direction, the method further includes: The timing of vehicle turning is determined based on a target scene sub-segment where vehicle turning occurs.

5. The vehicle turning scene recognition method as described in claim 2, characterized in that, The step of determining the change in vehicle driving direction in the corresponding scene sub-segment from the target frame to the preset frame based on the angle difference includes: The angle difference is normalized. If the normalized angle difference is within the first angle range, then the change in the vehicle's driving direction is determined to be a left turn; If the normalized angle difference is within the second angle range, then the change in the vehicle's driving direction is determined to be a right turn; If the normalized angle difference is within the third angle range, then the change in the vehicle's driving direction is determined to be that no steering has occurred; There is no overlap between the first angle interval, the second angle interval, and the third angle interval.

6. The vehicle turning scene recognition method as described in claim 1, characterized in that, Following the step of acquiring historical scene fragments, the method further includes: Determine whether intersection signage information exists in each frame of data; If it exists, the target distance between the intersection and the vehicle is calculated based on the intersection sign information; When the target distance is less than the second distance threshold, it is determined that there is an intersection scene in the historical scene segment.

7. The vehicle turning scene recognition method as described in claim 6, characterized in that, The intersection signage information includes traffic light information, and the traffic light information includes first position information of at least one traffic light. The step of calculating the target distance between the intersection and the vehicle based on the intersection signage information includes: Based on the first location information and the vehicle location information corresponding to each frame of data, the Euclidean distance between the at least one traffic light and the vehicle, and the first relative direction between the at least one traffic light and the vehicle are calculated. Based on the angle between the first relative direction and the target direction of the vehicle's driving speed corresponding to each frame of data, it is determined whether there is a target traffic light in front of the vehicle. If it exists, the minimum distance in the Euclidean distance corresponding to the target traffic light is taken as the target distance between the intersection and the vehicle.

8. The vehicle turning scene recognition method as described in claim 6, characterized in that, The intersection signage information also includes zebra crossing information, which includes second position information corresponding to any corner point of at least one zebra crossing rectangle. The step of calculating the target distance between the intersection and the vehicle based on the intersection signage information includes: Based on the second location information and the vehicle location information corresponding to each frame of data, the projected distance between the vehicle and the at least one zebra crossing, and the second relative direction between the at least one zebra crossing and the vehicle are calculated. Based on the angle between the second relative direction and the target direction of the vehicle's driving speed corresponding to each frame of data, it is determined whether there is a target zebra crossing in front of the vehicle. If it exists, the minimum distance among the projected distances corresponding to the target zebra crossing will be taken as the target distance between the intersection and the vehicle.

9. A vehicle turning scene recognition device, characterized in that, The vehicle turning scene recognition device includes: The acquisition module is used to acquire historical scene fragments; The identification module includes: The first judgment submodule includes: The first judgment unit is used to determine whether the vehicle speed corresponding to the target frame is less than a preset speed threshold when there is an intersection scene in the historical scene segment. The target frame includes a start frame and the preset frame is spaced apart from the target frame by a preset frame length. The second calculation unit is used to calculate the vehicle travel distance from the target frame to the preset frame based on the vehicle position information corresponding to each frame if the vehicle speed is greater than or equal to a preset speed threshold. The second judgment unit is used to determine whether the vehicle's travel distance is less than the first distance threshold. The third calculation unit is used to determine the possibility that the vehicle is turning between the target frame and the preset frame if the vehicle's travel distance is greater than or equal to the first distance threshold. The determination submodule is used to skip the identification of the scene sub-segment from the target frame to the preset frame if there is no possibility of turning, and if there is, determine the change of the vehicle's driving direction in the scene sub-segment from the target frame to the preset frame based on the vehicle heading angles corresponding to the target frame and the preset frame respectively. The recognition submodule is used to determine the scene recognition result based on the change in the vehicle's driving direction, wherein the scene recognition result includes the presence of a vehicle turning left scene, a vehicle turning right scene, and / or the absence of a vehicle turning scene in the historical scene segment.

10. A vehicle turning scene recognition device, characterized in that, The vehicle turning scene 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 vehicle turning scene recognition method as described in any one of claims 1 to 8.

11. 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 turning scene recognition method as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the vehicle steering scene recognition method as described in any one of claims 1 to 8.

Images