Vehicle state change scene recognition method and related equipment

Abstract

The invention discloses a vehicle state change scene recognition method and related equipment, and relates to the technical field of scene recognition, and the vehicle state change scene recognition method comprises the steps: obtaining a scene data segment, and extracting scene related information of each frame from the scene data segment; based on the scene related information, identifying whether an obstacle associated with the vehicle exists in the scene data segment; if the obstacle exists, a vehicle state change scene recognition result is obtained through judgment of the relative position of the obstacle and the vehicle, and the vehicle state change scene recognition result comprises a braking scene and a vehicle following starting scene. The situation that the braking scene and the car-following starting scene cannot be accurately recognized in the scene without a high-precision map is avoided.

Description

Technical Field

[0001] The present application relates to the field of scene recognition technology, and in particular to a method for recognizing vehicle state change scenes and related equipment. Background Art

[0002] Vehicle state change scenarios refer to the various state transitions that a vehicle experiences in different situations, such as braking scenarios and following vehicle starting scenarios.

[0003] In related technologies, detailed static road information provided in high-precision maps, such as lane lines, traffic signs, signal lights, speed bumps and speed limit zones, is used in combination with real-time sensor data to identify vehicle state change scenarios such as braking and following a vehicle. However, in scenarios without high-precision maps, braking and following vehicle starting scenarios cannot be accurately identified. Summary of the Invention

[0004] The main purpose of this application is to provide a vehicle state change scene recognition method and related equipment, aiming to solve the technical problem that braking scenes and following vehicle starting scenes cannot be accurately identified in scenarios without high-precision maps.

[0005] To achieve the above objectives, the present application proposes a vehicle state change scene recognition method, which includes:

[0006] Acquire scene data segments, and extract scene-related information of each frame from the scene data segments;

[0007] Based on the scene-related information, identifying whether there is an obstacle associated with the ego-vehicle in the scene data segment;

[0008] If the obstacle exists, the vehicle state change scene recognition result is obtained by judging the relative position of the obstacle and the vehicle, wherein the vehicle state change scene recognition result includes a braking scene and a following vehicle starting scene.

[0009] In one embodiment, the identifying, based on the scene-related information, whether there is an obstacle associated with the vehicle in the scene data segment includes:

[0010] Based on the scene-related information, determining whether the vehicle has a start-stop moment in the scene data segment;

[0011] If the vehicle has the start-stop moment, it is identified whether there is an obstacle associated with the vehicle in the frame associated with the start-stop moment.

[0012] In one embodiment, the step of determining whether the vehicle has a start / stop moment in the scene data segment based on the scene-related information includes:

[0013] generating a speed sequence based on the speed information and a timing of occurrence of frames corresponding to the speed information;

[0014] When the number of elements in the speed sequence is greater than or equal to a preset threshold, it is determined whether the vehicle has a start-stop moment in the scene data segment based on the speed corresponding to each element in the speed sequence and the acceleration between each element and the corresponding adjacent element.

[0015] In one embodiment, the step of determining whether the vehicle has a start-stop moment in the scene data segment based on the speed corresponding to each element in the speed sequence and the acceleration between adjacent elements corresponding to each element includes:

[0016] Determining the acceleration between each element and its corresponding adjacent element based on the velocity corresponding to each element in the velocity sequence and the occurrence time sequence;

[0017] generating an initial acceleration sequence based on the occurrence timing and the acceleration;

[0018] Processing the initial acceleration sequence to obtain a target acceleration sequence having the same length as the velocity sequence;

[0019] If there is a target element in the target acceleration sequence whose acceleration is greater than a preset acceleration threshold, determining the sequence number of the target element in the target acceleration sequence, and determining the target speed corresponding to the sequence number in the speed sequence;

[0020] Based on the target speed, it is determined whether the vehicle has a start-stop moment in the scene data segment.

[0021] In one embodiment, the step of determining whether the vehicle has a start-stop moment in the scene data segment based on the target speed includes:

[0022] Determine a frame position of a frame corresponding to the target speed in the scene data segment;

[0023] If the frame position is the first frame, determining whether the vehicle has a start-stop moment in the first frame based on the target speed corresponding to the first frame and a first speed set corresponding to a preset number of frames after the first frame;

[0024] If the frame position is the last frame, determining whether the vehicle has a start-stop moment in the last frame based on the target speed corresponding to the last frame and a second speed set corresponding to a preset number of frames after the last frame;

[0025] If the frame position is an intermediate frame, it is determined whether the vehicle has a start / stop moment in the intermediate frame based on the target speed corresponding to the intermediate frame.

[0026] In one embodiment, if the frame position is an intermediate frame, then the step of determining whether the vehicle has a start-stop moment in the intermediate frame based on the target speed corresponding to the intermediate frame includes:

[0027] If the frame position is an intermediate frame, determining a sequence number of a target speed corresponding to the intermediate frame in the speed sequence, and determining an acceleration direction corresponding to a target acceleration associated with the sequence number of the target speed;

[0028] If the acceleration direction corresponding to the target acceleration is a first direction, the target speed is greater than a first threshold, the speed corresponding to the frame before the frame position is less than or equal to the first threshold, and the speed after the preset number of frames at the frame position is greater than or equal to a second threshold, then determining the moment corresponding to the target speed as the start moment;

[0029] If the acceleration direction corresponding to the target acceleration is the second direction, the target speed is less than or equal to the first threshold, the speed corresponding to the frame before the frame position is greater than the first threshold, and the speed before the preset number of frames at the frame position is greater than or equal to the second threshold, then the moment corresponding to the target speed is determined to be the stop moment, wherein the first direction is opposite to the second direction.

[0030] In one embodiment, if the vehicle has the start-stop moment, the step of identifying whether there is an obstacle in the scene data frame associated with the start-stop moment includes:

[0031] If the vehicle has the start-stop moment, determining whether there is an obstacle in each frame of the scene data segment;

[0032] If the obstacle exists, an obstacle is identified, wherein the category corresponding to the obstacle is consistent with the preset category of obstacles to be considered, the speed of the obstacle is greater than or equal to the start threshold, and the average speed of the obstacle is greater than a preset average speed threshold.

[0033] In one embodiment, if the obstacle exists, the step of determining the relative position of the obstacle and the vehicle to obtain a vehicle state change scene recognition result includes:

[0034] Normalizing the navigation angle of the ego vehicle in each frame of the scene data segment to obtain a normalized heading angle within a preset range, and converting the heading angle into a first degree;

[0035] Calculating a second degree corresponding to the normalized heading angle of the obstacle, and determining a direction angle difference between the vehicle and the obstacle based on the first degree and the second degree;

[0036] Calculating the local coordinate difference of the obstacle relative to the vehicle;

[0037] A vehicle state change scene recognition result is determined based on the direction angle difference and the local coordinate difference.

[0038] In one embodiment, the step of determining a vehicle state change scene recognition result based on the direction angle difference and the local coordinate difference includes:

[0039] If the vertical coordinate difference is less than or equal to the preset vertical coordinate difference, the horizontal coordinate difference is within the preset horizontal coordinate difference range, and the direction angle difference is less than or equal to the preset angle, then the vehicle state change scenario corresponding to the starting moment of the vehicle is the following starting scenario, and the vehicle state change scenario corresponding to the stopping moment of the vehicle is the braking scenario.

[0040] In addition, to achieve the above-mentioned purpose, the present application also proposes a vehicle state change scene recognition device, the vehicle state change scene recognition device comprising:

[0041] An acquisition module, configured to acquire scene data segments and extract scene-related information of each frame from the scene data segments;

[0042] an identification module, configured to identify whether there is an obstacle associated with the vehicle in the scene data segment based on the scene-related information;

[0043] The judgment module is used to obtain a vehicle state change scene recognition result by judging the relative position of the obstacle and the vehicle if the obstacle exists, wherein the vehicle state change scene recognition result includes a braking scene and a following vehicle starting scene.

[0044] In addition, to achieve the above-mentioned purpose, the present application also proposes a vehicle state change scene recognition device, which includes: a memory, a processor, and a computer program stored on the memory and runnable on the processor, and the computer program is configured to implement the steps of the vehicle state change scene recognition method as described above.

[0045] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the vehicle state change scene recognition method described above are implemented.

[0046] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the vehicle state change scene recognition method as described above.

[0047] One or more technical solutions proposed in this application have at least the following technical effects:

[0048] Compared with the related art, the detailed static road information provided in the high-precision map, such as lane lines, traffic signs, signal lights, speed bumps and speed limit zones, is combined with real-time sensor data to identify the vehicle state change scenarios of braking and following the vehicle. However, in the scenario without high-precision maps, the braking scene and the following the vehicle start scene cannot be accurately identified. In this application, the scene data fragment is obtained, and the scene related information of each frame is extracted from the scene data fragment; based on the scene related information, it is identified whether there is an obstacle associated with the vehicle in the scene data fragment; if the obstacle exists, the vehicle state change is obtained by judging the relative position of the obstacle and the vehicle. Scene recognition results, wherein the vehicle state change scene recognition results include braking scenes and following vehicle starting scenes. It can be understood that after extracting scene-related information from the vehicle's scene data segment, this application will determine whether there is an obstacle associated with the vehicle based on the scene-related information. If so, the vehicle state change scene recognition results including braking scenes and following vehicle starting scenes will be obtained based on the relative position of the obstacle and the vehicle. Detailed static road information of the high-precision map is not required. The vehicle state change scene recognition results can be obtained only through the relative position between the vehicle and the obstacle, avoiding the inability to accurately identify braking scenes and following vehicle starting scenes in scenarios without high-precision maps. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0051] Figure 1 A flowchart of the first embodiment of the vehicle state change scene recognition method provided in this application;

[0052] Figure 2 A flow chart illustrating a second embodiment of the vehicle state change scene recognition method of the present application;

[0053] Figure 3 A flow chart illustrating a second embodiment of the vehicle state change scene recognition method of the present application;

[0054] Figure 4This is a schematic diagram of the module structure of the vehicle state change scene recognition device according to an embodiment of the present application;

[0055] Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the vehicle state change scene recognition method in the embodiment of the present application.

[0056] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0058] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0059] The main solution of the embodiment of the present application is: obtaining scene data fragments and extracting scene-related information of each frame from the scene data fragments; based on the scene-related information, identifying whether there is an obstacle associated with the vehicle in the scene data fragments; if the obstacle exists, obtaining the vehicle state change scene recognition result by judging the relative position of the obstacle and the vehicle, wherein the vehicle state change scene recognition result includes braking scene and following vehicle starting scene.

[0060] In related technologies, detailed static road information provided in high-precision maps, such as lane lines, traffic signs, signal lights, speed bumps and speed limit zones, is used in combination with real-time sensor data to identify vehicle state change scenarios such as braking and following a vehicle. However, in scenarios without high-precision maps, braking and following vehicle starting scenarios cannot be accurately identified.

[0061] After extracting scene-related information from the vehicle's scene data fragment, this application will determine whether there is an obstacle associated with the vehicle based on the scene-related information. If so, it will obtain vehicle state change scene recognition results including braking scenes and following vehicle starting scenes based on the relative position of the obstacle and the vehicle. It does not require detailed static road information from a high-precision map, but only needs to use the relative position between the vehicle and the obstacle to obtain the vehicle state change scene recognition results, avoiding the inability to accurately identify braking scenes and following vehicle starting scenes in scenarios without high-precision maps.

[0062] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication, and program execution capabilities, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device capable of implementing the aforementioned functions. This embodiment and the following embodiments will be described below using a vehicle state change scene recognition device as an example.

[0063] Based on this, the embodiment of the present application provides a method for identifying vehicle state change scenarios, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the vehicle state change scene recognition method of the present application.

[0064] In this embodiment, the vehicle state change scene recognition method includes steps S100 to S300:

[0065] Step S100, obtaining scene data segments, and extracting scene-related information of each frame from the scene data segments;

[0066] It should be noted that the execution entity of this embodiment is a vehicle state change scene recognition device. The vehicle state change scene recognition device is equipped with a sensor that is used to collect data about the vehicle's surrounding environment in real time and generate continuous scene data segments. The scene data segments include necessary relevant information about the vehicle, including the vehicle's speed, frame information, and heading angle information. After obtaining the scene data segments, the vehicle state change scene recognition device extracts scene-related information from the data of each frame of the scene data segment.

[0067] Step S200, based on the scene-related information, identifying whether there is an obstacle associated with the vehicle in the scene data segment;

[0068] It is understood that obstacles associated with the ego vehicle are obstacles that may affect the ego vehicle's driving state. After extracting scene-related information about the ego vehicle, the vehicle state change scene recognition device analyzes the scene-related information to identify whether there are obstacles in the scene data segment that may affect the ego vehicle's driving state.

[0069] Step S300: If the obstacle exists, a vehicle state change scene recognition result is obtained by judging the relative position of the obstacle and the vehicle, wherein the vehicle state change scene recognition result includes a braking scene and a following vehicle start scene.

[0070] It should be noted that if the vehicle state change scene recognition device identifies that there is an obstacle in the scene data segment that may affect the driving state of the vehicle, it will judge the relative position of the obstacle and the vehicle to obtain the vehicle's braking scene and following vehicle starting scene.

[0071] In a feasible implementation manner, step S200 includes the following steps:

[0072] Based on the scene-related information, determining whether the vehicle has a start-stop moment in the scene data segment;

[0073] It is understood that the start-stop time includes the start time and stop time during the vehicle's driving process. The vehicle state change scene recognition device will determine whether there is a start time and a stop time during the vehicle's driving process based on the extracted scene-related information.

[0074] If the vehicle has the start-stop moment, it is identified whether there is an obstacle associated with the vehicle in the frame associated with the start-stop moment.

[0075] It should be noted that after the vehicle state change scene recognition device determines that there are start-stop moments during the vehicle's driving process, it will identify whether there are obstacles that may affect the vehicle's driving state in the frames associated with the start-stop moments.

[0076] In a feasible implementation, if the vehicle has the start-stop moment, the step of identifying whether there is an obstacle associated with the vehicle in the frame associated with the start-stop moment includes the following steps:

[0077] If the vehicle has the start-stop moment, determining whether there is an obstacle in each frame of the scene data segment;

[0078] It is understandable that after determining that the vehicle has a start or stop moment, the vehicle state change scene recognition device will detect each frame in the scene data segment to detect whether there is an obstacle in each frame.

[0079] If the obstacle exists, an obstacle is identified, wherein the category corresponding to the obstacle is consistent with the preset category of obstacles to be considered, the speed of the obstacle is greater than or equal to the start threshold, and the average speed of the obstacle is greater than a preset average speed threshold.

[0080] It should be noted that the default obstacle categories considered in the data segment include cars, buses, trucks, construction vehicles, general obstacles, and unknown obstacles. The vehicle state change scene recognition device pre-stores a list of object types to consider, which includes the preset obstacle categories. These preset obstacle categories include object types that require special attention, typically dynamic objects that may affect the vehicle's movement.

[0081] It is understandable that the vehicle state change scene recognition device identifies, for each object in each frame, an object whose type is in a preset obstacle category to be considered, whose object speed is greater than or equal to 0, and whose average speed is greater than a preset average speed threshold as an obstacle.

[0082] Specifically, the vehicle state change scene recognition device pre-sets a speed mean threshold v_mean_threshold by default (0.35 m / s) and sets a storage list:

[0083] need_obj_type defines the list of object types that need to be considered. By default, cars, buses, trucks, engineering vehicles, general obstacles and unknown obstacles appearing in the data segment are considered.

[0084] The object_info empty dictionary is used to record the speed history of each object. Each object has a unique ID as a key, and the corresponding value is a list containing the speed of the object in each frame.

[0085] The object_mean_v empty dictionary is used to store the average speed of each object in the entire data segment. Similarly, the ID of each object is used as the key and its average speed is used as the value;

[0086] The consider_obj_id_lis empty list is used to store the object IDs whose average speed exceeds the preset threshold;

[0087] Furthermore, for each object (obj_info) in each frame, the vehicle state change scene recognition device first checks whether its type is in the need_obj_type list. If so, it proceeds to the next step; if not, it ignores the object and then checks whether its velocity, obj_v, is greater than or equal to 0. Only objects that are moving or at least not moving backward are considered. If the object's velocity meets the criteria, it is added to the list of the corresponding object ID in the object_info dictionary. If the object ID does not yet appear in object_info, a new empty list is created for it and the current velocity is added to it.

[0088] In particular, if the vehicle state change scene recognition device detects that the speed of an object is less than 0, or the object type is not in the need_obj_type list, the object is skipped and its information is not added to the storage structure.

[0089] Furthermore, the vehicle state change scene recognition device calculates the average value of the corresponding speed list for each object ID in the object_info dictionary. If the average speed of an object exceeds the preset threshold v_mean_threshold, the object is considered to be active and may affect the driving of the vehicle. The ID of the object is then added to the consider_obj_id_list list for further analysis. Through the above steps, the vehicle state change scene recognition device can effectively screen out objects that are associated with the vehicle and may affect its driving path or behavior. These objects can then be used for more in-depth analysis, such as determining whether there is a braking scenario or a following vehicle starting scenario, thereby improving the safety and reliability of the autonomous driving system.

[0090] In a feasible implementation manner, the step of determining whether the vehicle has a start / stop moment in the scene data segment based on the scene related information includes the following steps:

[0091] generating a speed sequence based on the speed information and a timing of occurrence of frames corresponding to the speed information;

[0092] It is understood that scene-related information includes the vehicle's speed information. The acquisition sequence is the acquisition sequence. The speed sequence is the ego_v list. The vehicle state change scene recognition device stores the speed information in the ego_v list based on the acquisition sequence of the corresponding frames, thereby generating a speed sequence containing the speed information.

[0093] When the number of elements in the speed sequence is greater than or equal to a preset threshold, it is determined whether the vehicle has a start-stop moment in the scene data segment based on the speed corresponding to each element in the speed sequence and the acceleration between each element and the corresponding adjacent element.

[0094] It should be noted that the preset number threshold can be set to 2, as at least two velocity values ​​are required to calculate acceleration. The vehicle state change scene recognition device first checks the length of the velocity sequence. If the velocity sequence is at least 2, it then determines whether the vehicle in the scene data segment has started or stopped by comparing the velocity of each element and the acceleration between each element and its adjacent elements.

[0095] In particular, if the vehicle state change scene recognition device detects that the speed sequence has less than 2 elements, it will directly return and will not continue to recognize the following vehicle starting scene and braking scene for this scene.

[0096] In a feasible implementation, the step of determining whether the vehicle has a start-stop moment in the scene data segment based on the speed corresponding to each element in the speed sequence and the acceleration between adjacent elements corresponding to each element includes the following steps:

[0097] Determining the acceleration between each element and its corresponding adjacent element based on the velocity corresponding to each element in the velocity sequence and the occurrence time sequence;

[0098] It can be understood that the vehicle state change scene recognition device calculates the acceleration between each element and the corresponding adjacent element through the occurrence time sequence.

[0099] generating an initial acceleration sequence based on the occurrence timing and the acceleration;

[0100] It should be noted that the initial acceleration sequence is an acceleration list, which is used to store accelerations. The vehicle state change scene recognition device saves the accelerations to the acceleration list according to the time sequence of the accelerations, thereby generating the initial acceleration sequence.

[0101] Processing the initial acceleration sequence to obtain a target acceleration sequence having the same length as the velocity sequence;

[0102] It is understandable that since the acceleration calculation is the speed difference between two adjacent frames, the initial acceleration sequence has one less element than the speed sequence. Therefore, the vehicle state change scene recognition device copies the first element of the initial acceleration sequence to the beginning of the sequence to obtain the target acceleration sequence. Furthermore, the length of the target acceleration sequence is consistent with the length of the speed sequence.

[0103] If there is a target element in the target acceleration sequence whose acceleration is greater than a preset acceleration threshold, determining the sequence number of the target element in the target acceleration sequence, and determining the target speed corresponding to the sequence number in the speed sequence;

[0104] It should be noted that the target speed is the speed in the frames that are initially screened for possible start / stop moments. The target element is the element corresponding to the acceleration in the frames that are likely to have start / stop moments. To prevent the vehicle from moving slowly and affecting the judgment constraint, and to prevent braking or starting after a vehicle from causing speed changes exceeding the movement threshold, a preset acceleration threshold is set for preliminary acceleration screening.

[0105] It is understandable that the vehicle state change scene recognition device performs a preliminary screening of the elements in the target acceleration sequence, obtains the target elements in the target acceleration sequence whose acceleration exceeds the preset acceleration threshold, and determines the serial number in the target acceleration sequence. Since the length of the target acceleration sequence is consistent with that of the speed sequence, the elements corresponding to the same serial number in different sequences are all associated data in the same frame, so the speed in the frame where the start-stop moment may exist can be determined based on the acceleration in the frame where the start-stop moment may exist.

[0106] Based on the target speed, it is determined whether the vehicle has a start-stop moment in the scene data segment.

[0107] It is understandable that the vehicle state change scene recognition device can determine whether the vehicle has a start-stop moment in the scene data segment based on the speed in the frame where the start-stop moment may exist.

[0108] In this embodiment, the vehicle state change scene recognition device can determine the start and stop time of the vehicle by the speed and acceleration of the vehicle at different times in the scenario without high-precision maps. By judging the relative position between the vehicle and the obstacle in the frame corresponding to the start and stop time, the vehicle state change scene recognition result can be obtained, avoiding the problem of not being able to accurately identify braking scenes and following vehicle starting scenes in the scenario without high-precision maps.

[0109] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above embodiment 1 can be referred to the above introduction and will not be described in detail later. Figure 2 Based on the target speed, determining whether the vehicle has a start-stop moment in the scene data segment, the vehicle state change scene recognition method further includes steps A10 to A40:

[0110] Step A10, determining the frame position of the frame corresponding to the target speed in the scene data segment;

[0111] It should be noted that the frame position refers to the position of a specific frame in the scene data segment relative to the entire sequence. Frames can be divided into three categories based on their relative position in the sequence: first frame, last frame, and middle frame. The vehicle state change scene recognition device will correspond to the target speed frame's position in the entire sequence.

[0112] Step A20: If the frame position is the first frame, based on the target speed corresponding to the first frame and the first speed set corresponding to a preset number of frames after the first frame, determine whether the vehicle has a start-stop moment in the first frame;

[0113] It is understandable that the preset number of frames can be set to 3. The first speed set includes the speed of the frame after the current frame, the speed of the next two frames, and the speed after n frames of the current frame, wherein n frames can be set to 50 frames, but is not specifically limited. If the number of subsequent frames is less than n, then the speed is taken until the last frame. If the vehicle state change scene recognition device determines that the frame position is the first frame, it reads the target speed corresponding to the first frame, and starting from the first frame, takes the speed values ​​of 3 preset position frames backward to form a first speed set, and judges whether there is a start-stop moment of the vehicle in the first frame through the target speed and the first speed set.

[0114] Specifically, the vehicle state change scene recognition device checks whether the speed of the first frame is less than or equal to the threshold (default 0.5m / s), and checks whether the speeds of two consecutive frames in the first speed set meet the acceleration condition, that is, the speed of the frame before the current frame is less than or equal to the threshold, and the speed of the current frame is greater than the threshold. It also checks whether there is a speed value greater than or equal to threshold_n (default 3.0m / s) in the frame 50 frames after the current frame in the first speed set to confirm that the vehicle has indeed started a significant acceleration process. If all the above conditions are met, it is considered that the first frame has a start moment;

[0115] In addition, the vehicle state change scene recognition device checks whether the speed of the first frame is greater than the threshold threshold (default 0.5m / s), and checks whether there are two consecutive frames in the first speed set that meet the acceleration condition, that is, the speed of the previous frame of the current frame is greater than the threshold, and the speed of the current frame is less than or equal to the threshold, and checks whether there is a speed value greater than or equal to threshold_n (default 3.0m / s) in the frame 50 frames after the current frame in the first speed set, to confirm that the vehicle has indeed started a significant acceleration process. If all the above conditions are met, it is considered that there is a stop moment in the first frame.

[0116] Step A30: If the frame position is the last frame, determining whether the vehicle has a start-stop moment in the last frame based on the target speed corresponding to the last frame and a second speed set corresponding to a preset number of frames after the last frame;

[0117] It should be noted that the second speed set includes the speed of the frame preceding the current frame, the speed of the frames preceding the current frame, and the speed of the frame n frames preceding the current frame. The number n frames can be set to 50, but is not specifically limited to n. If the number of frames preceding the current frame is less than n, the speed is taken up to the first frame. If the vehicle state change scene recognition device determines that the frame position is the last frame, it reads the target speed corresponding to the last frame and, starting from the last frame, takes the speed values ​​of the three preset position frames backward to form a second speed set. The target speed and the second speed set are used to determine whether the vehicle has started or stopped in the last frame.

[0118] Specifically, the vehicle state change scene recognition device checks whether the tail frame speed is less than or equal to the threshold threshold, and checks whether there are two consecutive frames in the second speed set whose speeds meet the deceleration condition, that is, the previous frame speed is greater than the threshold, and the current frame speed is less than or equal to the threshold, and checks whether there is at least one speed value greater than or equal to threshold_n among the speeds 50 frames before the previous frame in the second speed set, to confirm that the vehicle has indeed experienced a significant deceleration process. If all the above conditions are met, it is considered that there is a stop moment in the tail frame.

[0119] In addition, the vehicle state change scene recognition device checks whether the speed of the first frame is greater than the threshold threshold (default 0.5m / s), and checks whether there are two consecutive frames in the first speed set that meet the acceleration condition, that is, the speed of the previous frame of the current frame is less than or equal to the threshold, and the speed of the current frame is greater than the threshold, and checks whether there is a speed value greater than or equal to threshold_n (default 3.0m / s) in the frame 50 frames after the current frame in the first speed set to confirm that the vehicle has indeed started a significant acceleration process. If all the above conditions are met, it is considered that there is a start moment in the first frame.

[0120] Step A40: If the frame position is an intermediate frame, based on the target speed corresponding to the intermediate frame, it is determined whether the vehicle has a start-stop moment in the intermediate frame.

[0121] It is understandable that if the vehicle state change scene recognition device determines that the frame position is an intermediate frame, it determines whether there is a start-stop moment of the vehicle in the intermediate frame based on the target speed corresponding to the intermediate frame.

[0122] In a feasible implementation manner, step A40 includes the following steps:

[0123] If the frame position is an intermediate frame, determining a sequence number of a target speed corresponding to the intermediate frame in the speed sequence, and determining an acceleration direction corresponding to a target acceleration associated with the sequence number of the target speed;

[0124] It should be noted that due to the sign difference between accelerations obtained during start and stop, the acceleration directions corresponding to start and stop times are also different. When the vehicle state change scene recognition device determines that the frame position is an intermediate frame, it uses the sequence number of the target velocity corresponding to the intermediate frame in the velocity sequence to determine the acceleration direction corresponding to the target acceleration associated with the sequence number.

[0125] If the acceleration direction corresponding to the target acceleration is a first direction, the target speed is greater than a first threshold, the speed corresponding to the frame before the frame position is less than or equal to the first threshold, and the speed after the preset number of frames at the frame position is greater than or equal to a second threshold, then determining the moment corresponding to the target speed as the start moment;

[0126] It is understandable that the first threshold is threshold (current frame speed threshold, default is 0.5m / s). The second threshold is threshold_n, which represents the speed threshold of n frames before (if any) or n frames after (if any) the current frame (default is 50 frames). The preset number of frames defaults to 50 frames and can be adjusted according to specific circumstances. The first direction is the positive direction, which corresponds to the acceleration direction. If the vehicle state change scene recognition device determines that the target acceleration direction is the first direction (acceleration), the target speed is greater than the threshold (current frame speed threshold, default is 0.5m / s), the speed corresponding to the previous frame of the frame position is less than or equal to the threshold (current frame speed threshold, default is 0.5m / s), and the speed 50 frames after the current frame is greater than or equal to threshold_n, then the moment corresponding to the target speed is considered to be the start moment.

[0127] Specifically, the vehicle state change scene recognition device predefines start_indices and stop_indices lists, which represent the start and stop time point indexes respectively. During initialization, both start_indices and stop_indices are empty lists.

[0128] Furthermore, the vehicle state change scene recognition device determines its direction based on the calculated acceleration value:

[0129] If acceleration>0, the acceleration direction is the first direction (acceleration);

[0130] If acceleration<0, the acceleration direction is the second direction (deceleration);

[0131] If the target acceleration direction is the first direction (acceleration) and the following conditions are met, then the moment is determined to be the start moment:

[0132] The target speed is greater than the first threshold: ego_v[target_frame_id]>threshold;

[0133] The previous frame speed is less than or equal to the first threshold: ego_v[target_frame_id-1]<=threshold;

[0134] The speed after the preset number of frames is greater than or equal to the second threshold: check the next N frames (eg, N=50) starting from target_frame_id, and the speed of at least one frame is greater than or equal to threshold_n.

[0135] If the acceleration direction corresponding to the target acceleration is the second direction, the target speed is less than or equal to the first threshold, the speed corresponding to the frame before the frame position is greater than the first threshold, and the speed before the preset number of frames at the frame position is greater than or equal to the second threshold, then the moment corresponding to the target speed is determined to be the stop moment, wherein the first direction is opposite to the second direction.

[0136] It should be noted that the second direction is the negative direction, i.e., the direction corresponding to deceleration. If the vehicle state change scene recognition device determines that the target acceleration direction is the first direction (acceleration), the target speed is less than or equal to the threshold (the current frame speed threshold, which defaults to 0.5 m / s), the speed corresponding to the previous frame position is greater than the threshold (the current frame speed threshold, which defaults to 0.5 m / s), and the speed 50 frames before the current frame is greater than or equal to threshold_n, then the moment corresponding to the target speed is considered to be the stop moment.

[0137] Specifically, the vehicle state change scene recognition device determines that the moment is the stop moment if the target acceleration direction is the second direction (deceleration) and the following conditions are met:

[0138] The target speed is less than or equal to the first threshold: ego_v[target_frame_id]<=threshold;

[0139] The speed of the previous frame is greater than the first threshold: ego_v[target_frame_id-1]>threshold;

[0140] The speed before the preset number of frames is greater than or equal to the second threshold: check that the speed of the first N frames (eg, N=50) starting from target_frame_id is greater than or equal to threshold_n.

[0141] In this embodiment, the vehicle state change scene recognition device can accurately determine whether the target speed corresponding to the intermediate frame marks the start or stop moment. It not only considers the speed information of a single frame, but also combines the acceleration direction and the speed change trend within a preset number of frames before and after, thereby improving the accuracy of the judgment. By introducing the judgment of the acceleration direction, the start and stop events can be distinguished more accurately, avoiding misjudgment caused by instantaneous speed fluctuations.

[0142] Based on the first and second embodiments of this application, in the third embodiment of this application, the same or similar contents as those in the first and second embodiments above can be referred to above and will not be described in detail later. Figure 3 The vehicle state change scene recognition method step S300 further includes steps S310 to S340:

[0143] Step S310, normalizing the navigation angle of the vehicle in each frame of the scene data segment to obtain a normalized heading angle within a preset range, and converting the heading angle into a first degree;

[0144] It is understood that the first degree corresponds to the ego vehicle's body heading angle. The vehicle state change scene recognition device obtains the ego vehicle's body heading angle for each frame in the scene data segment, normalizes the ego vehicle's heading angle to ensure that the ego vehicle's heading angle value is within a specific range, such as [-π, π], to avoid problems caused by angle values ​​that cross 0 degrees or 360 degrees, and converts the normalized ego vehicle heading angle into degrees.

[0145] Specifically, the vehicle state change scene recognition device normalizes the vehicle heading angle obj_theta_origin in each frame of data to a preset range, usually [-π,π]. This is achieved through the following normalization rules:

[0146] If obj_theta_origin<-π, then theta=obj_theta_origin+2π;

[0147] If obj_theta_origin>π, then theta=obj_theta_origin-2π;

[0148] Otherwise, theta = obj_theta_origin;

[0149] The normalized heading angle theta is the standard heading angle of the ego vehicle in this frame. The normalized heading angle theta is converted from radians to degrees, which is the first degree ego_heading_degree, and is stored in the ego_heading list.

[0150] Step S320, calculating a second degree corresponding to the normalized heading angle of the obstacle, and determining a direction angle difference between the vehicle and the obstacle based on the first degree and the second degree;

[0151] It should be noted that the azimuth angle difference refers to the difference in heading angle (i.e., orientation angle) between the ego vehicle and the obstacle. The second degree is the degree corresponding to the speed and heading angle of the obstacle. Since the ego vehicle does not fully recognize the position and posture of the obstacle, it is analyzed through the speed and heading angle of the obstacle. The vehicle state change scene recognition device extracts the speed and heading angle of each obstacle and normalizes it according to the same rules as the ego vehicle to ensure that its value also falls within the range of [-π,π]. The normalized speed and heading angle is converted to the second degree, and the difference between the first and second degrees is calculated to obtain the azimuth angle difference.

[0152] Step S330, calculating the local coordinate difference of the obstacle relative to the vehicle;

[0153] It is understood that the local coordinate difference refers to the relative position of the obstacle with respect to the ego vehicle and is used to describe the specific position of the obstacle in the ego vehicle's coordinate system. The vehicle state change scene recognition device calculates the local coordinate difference of the obstacle with respect to the ego vehicle based on the ego vehicle's position coordinates.

[0154] Specifically, the vehicle state change scene recognition device calculates the local coordinate difference of the obstacle relative to the vehicle, local_x, local_y, using the formula:

[0155] local_x = obj_x - ego_obj_x;

[0156] local_y=obj_y-ego_obj_y.

[0157] Step S340 : determining a vehicle state change scene recognition result based on the direction angle difference and the local coordinate difference.

[0158] It should be noted that the vehicle state change scene recognition device determines the vehicle state change scene recognition result based on the direction angle difference and the local coordinate difference.

[0159] In a feasible implementation, step S340 includes the following steps:

[0160] If the vertical coordinate difference is less than or equal to the preset vertical coordinate difference, the horizontal coordinate difference is within the preset horizontal coordinate difference range, and the direction angle difference is less than or equal to the preset angle, then the vehicle state change scenario corresponding to the starting moment of the vehicle is the following starting scenario, and the vehicle state change scenario corresponding to the stopping moment of the vehicle is the braking scenario.

[0161] It is understood that the preset vertical coordinate difference is 1.5m. The horizontal coordinate difference range is 0 to 20m. The preset angle is 45 degrees. If the vehicle state change scenario recognition device detects that the vertical coordinate difference of the local coordinate difference is less than or equal to 1.5m, the horizontal coordinate difference is between 0 and 20m, and the direction angle difference is not greater than 45 degrees, then the vehicle state change scenario corresponding to the start time is the following start scenario, and the vehicle state change scenario corresponding to the stop time is the braking scenario.

[0162] In this embodiment, the vehicle state change scene recognition device normalizes the heading angle of the ego vehicle in each frame in the scene data segment, and calculates the angular difference and local coordinate difference between the ego vehicle and the obstacle. Based on this information, the braking scene and the following vehicle starting scene can be accurately identified, thereby improving the perception ability and decision-making accuracy of the autonomous driving system.

[0163] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the vehicle state change scene recognition method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0164] This application also provides a vehicle state change scene recognition device, please refer to Figure 4 , the vehicle state change scene recognition device includes:

[0165] An acquisition module 10 is configured to acquire scene data segments and extract scene-related information of each frame from the scene data segments;

[0166] an identification module 20 for identifying whether there is an obstacle associated with the vehicle in the scene data segment based on the scene-related information;

[0167] The judgment module 30 is used to obtain a vehicle state change scene recognition result by judging the relative position of the obstacle and the vehicle if the obstacle exists, wherein the vehicle state change scene recognition result includes a braking scene and a following vehicle starting scene.

[0168] Optionally, the identification module includes:

[0169] The start-stop moment judgment submodule is used to judge whether the vehicle has a start-stop moment in the scene data segment based on the scene-related information; if the vehicle has the start-stop moment, identify whether there is an obstacle associated with the vehicle in the frame associated with the start-stop moment.

[0170] Optionally, the start / stop time determination submodule includes:

[0171] A start-stop moment judgment unit is configured to generate a speed sequence based on the speed information and the occurrence timing of frames corresponding to the speed information; and when the number of elements in the speed sequence is greater than or equal to a preset number threshold, determine whether the vehicle has a start-stop moment in the scene data segment based on the speed corresponding to each element in the speed sequence and the acceleration between each element and the corresponding adjacent element.

[0172] Optionally, the start / stop time determination unit includes:

[0173] A sequence number determination subunit is configured to determine, based on the speed corresponding to each element in the speed sequence and the occurrence sequence, the acceleration between each element and its corresponding adjacent element; generate an initial acceleration sequence based on the occurrence sequence and the acceleration; process the initial acceleration sequence to obtain a target acceleration sequence having a length consistent with the speed sequence; if a target element having an acceleration greater than a preset acceleration threshold exists in the target acceleration sequence, determine the sequence number of the target element in the target acceleration sequence, and determine a target speed corresponding to the sequence number in the speed sequence; and based on the target speed, determine whether the vehicle has a start-stop moment in the scene data segment.

[0174] Optionally, the sequence number determination subunit includes:

[0175] a frame position determination component for determining the frame position of the frame corresponding to the target speed in the scene data segment; if the frame position is the first frame, then based on the target speed corresponding to the first frame and a first speed set corresponding to a preset number of frames after the first frame, determining whether the vehicle has a start or stop moment in the first frame; if the frame position is the last frame, then based on the target speed corresponding to the last frame and a second speed set corresponding to a preset number of frames after the last frame, determining whether the vehicle has a start or stop moment in the last frame; if the frame position is an intermediate frame, then based on the target speed corresponding to the intermediate frame, determining whether the vehicle has a start or stop moment in the intermediate frame.

[0176] Optionally, the frame position determination component is also used to, if the frame position is an intermediate frame, determine the serial number of the target speed corresponding to the intermediate frame in the speed sequence, and determine the acceleration direction corresponding to the target acceleration associated with the serial number of the target speed; if the acceleration direction corresponding to the target acceleration is a first direction, the target speed is greater than a first threshold, the speed corresponding to the previous frame of the frame position is less than or equal to the first threshold, and the speed after the preset number of frames at the frame position is greater than or equal to a second threshold, then determine that the moment corresponding to the target speed is the start moment; if the acceleration direction corresponding to the target acceleration is a second direction, the target speed is less than or equal to the first threshold, the speed corresponding to the previous frame of the frame position is greater than the first threshold, and the speed before the preset number of frames at the frame position is greater than or equal to the second threshold, then determine that the moment corresponding to the target speed is the stop moment, wherein the first direction is opposite to the second direction.

[0177] Optionally, the start / stop time determination submodule includes:

[0178] The obstacle judgment unit is configured to, if the vehicle has the start-stop moment, judge whether an obstacle exists in each frame of the scene data segment; if the obstacle exists, identify an obstacle whose category is consistent with a preset obstacle category to be considered, whose speed is greater than or equal to a start threshold, and whose average speed is greater than a preset average speed threshold.

[0179] Optionally, the judgment module includes:

[0180] The normalization submodule is configured to normalize the heading angle of the ego vehicle for each frame in the scene data segment to obtain a normalized heading angle within a preset range, and convert the heading angle into a first degree; calculate a second degree corresponding to the normalized heading angle of the obstacle; determine a direction angle difference between the ego vehicle and the obstacle based on the first degree and the second degree; calculate a local coordinate difference of the obstacle relative to the ego vehicle; and determine a vehicle state change scene recognition result based on the direction angle difference and the local coordinate difference.

[0181] Optionally, the normalization submodule includes:

[0182] The scene recognition unit is used to determine that if the vertical coordinate difference is less than or equal to a preset vertical coordinate difference, the horizontal coordinate difference is within a preset horizontal coordinate difference range, and the direction angle difference is less than or equal to a preset angle, then the vehicle state change scene corresponding to the starting moment of the own vehicle is a following starting scene, and the vehicle state change scene corresponding to the stopping moment of the own vehicle is a braking scene.

[0183] The vehicle state change scene recognition device provided in this application utilizes the vehicle state change scene recognition method of the aforementioned embodiment to solve the technical problem of vehicle state change scene recognition. Compared with the prior art, the beneficial effects of the vehicle state change scene recognition device provided in this application are the same as those of the vehicle state change scene recognition method provided in the aforementioned embodiment. The other technical features of the vehicle state change scene recognition device are the same as those disclosed in the aforementioned embodiment method and are not further described here.

[0184] The present application provides a vehicle state change scene recognition device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the vehicle state change scene recognition method in the above-mentioned embodiment one.

[0185] Reference below Figure 5, which shows a schematic structural diagram of a vehicle state change scene recognition device suitable for implementing an embodiment of the present application. The vehicle state change scene recognition device in the embodiment of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, tablet computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The vehicle state change scene recognition device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0186] like Figure 5 As shown, the vehicle state change scene recognition device may include a processing device 1001 (such as 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 to a random access memory (RAM) 1004. Various programs and data required for the operation of the vehicle state change scene recognition device are also stored in RAM 1004. The processing device 1001, ROM 1002, and RAM 1004 are connected to each other 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: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the vehicle state change scene recognition device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a vehicle state change scene recognition device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have instead.

[0187] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0188] The vehicle state change scene recognition device provided in this application utilizes the vehicle state change scene recognition method of the above-mentioned embodiment to solve the technical problem of vehicle state change scene recognition. Compared with the prior art, the beneficial effects of the vehicle state change scene recognition device provided in this application are the same as those of the vehicle state change scene recognition method provided in the above-mentioned embodiment. The other technical features of the vehicle state change scene recognition device are the same as those disclosed in the above-mentioned embodiment and are not further described here.

[0189] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0190] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0191] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, a computer program) stored thereon, and the computer-readable program instructions are used to execute the vehicle state change scene recognition method in the above-mentioned embodiment.

[0192] 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, systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0193] The computer-readable storage medium may be included in the vehicle state change scene recognition device; or may exist independently without being assembled into the vehicle state change scene recognition device.

[0194] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the vehicle state change scene recognition device, the vehicle state change scene recognition device: obtains a scene data segment and extracts scene-related information of each frame from the scene data segment; based on the scene-related information, identifies whether there is an obstacle associated with the vehicle in the scene data segment; if the obstacle exists, obtains a vehicle state change scene recognition result by judging the relative position of the obstacle and the vehicle, wherein the vehicle state change scene recognition result includes a braking scene and a following vehicle starting scene.

[0195] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0196] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0197] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0198] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned vehicle state change scene recognition method, thereby resolving the technical problem of vehicle state change scene recognition. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the vehicle state change scene recognition method provided in the aforementioned embodiment, and are not further elaborated here.

[0199] The present application also provides a computer program product, comprising a computer program, which implements the steps of the above-mentioned vehicle state change scene recognition method when executed by a processor.

[0200] The computer program product provided in this application can solve the technical problem of identifying vehicle state change 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 state change scenario identification method provided in the above embodiment, and will not be repeated here.

[0201] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for identifying vehicle state change scenarios, characterized in that: The vehicle state change scene recognition method includes: Acquire scene data segments, and extract scene-related information of each frame from the scene data segments; Based on the scene-related information, identifying whether there is an obstacle associated with the ego-vehicle in the scene data segment; If the obstacle exists, the vehicle state change scene recognition result is obtained by judging the relative position of the obstacle and the vehicle, wherein the vehicle state change scene recognition result includes a braking scene and a following vehicle starting scene.

2. The vehicle state change scene recognition method according to claim 1, characterized in that: The scene-related information includes speed information, and identifying whether there is an obstacle associated with the vehicle in the scene data segment based on the scene-related information includes: Based on the scene-related information, determining whether the vehicle has a start-stop moment in the scene data segment; If the vehicle has the start-stop moment, it is identified whether there is an obstacle associated with the vehicle in the frame associated with the start-stop moment.

3. The vehicle state change scene recognition method according to claim 2, characterized in that: The scene-related information includes speed information, and the step of determining whether the vehicle has a start-stop moment in the scene data segment based on the scene-related information includes: generating a speed sequence based on the speed information and a timing of occurrence of frames corresponding to the speed information; When the number of elements in the speed sequence is greater than or equal to a preset threshold, it is determined whether the vehicle has a start-stop moment in the scene data segment based on the speed corresponding to each element in the speed sequence and the acceleration between each element and the corresponding adjacent element.

4. The vehicle state change scene recognition method according to claim 3, characterized in that: The step of determining whether the vehicle has a start-stop moment in the scene data segment based on the speed corresponding to each element in the speed sequence and the acceleration between adjacent elements corresponding to each element includes: Determining the acceleration between each element and its corresponding adjacent element based on the velocity corresponding to each element in the velocity sequence and the occurrence time sequence; generating an initial acceleration sequence based on the occurrence timing and the acceleration; Processing the initial acceleration sequence to obtain a target acceleration sequence having the same length as the velocity sequence; If there is a target element in the target acceleration sequence whose acceleration is greater than a preset acceleration threshold, determining the sequence number of the target element in the target acceleration sequence, and determining the target speed corresponding to the sequence number in the speed sequence; Based on the target speed, it is determined whether the vehicle has a start-stop moment in the scene data segment.

5. The vehicle state change scene recognition method according to claim 4, characterized in that: The step of determining whether the vehicle has a start-stop moment in the scene data segment based on the target speed includes: Determine a frame position of a frame corresponding to the target speed in the scene data segment; If the frame position is the first frame, determining whether the vehicle has a start-stop moment in the first frame based on the target speed corresponding to the first frame and a first speed set corresponding to a preset number of frames after the first frame; If the frame position is the last frame, determining whether the vehicle has a start-stop moment in the last frame based on the target speed corresponding to the last frame and a second speed set corresponding to a preset number of frames after the last frame; If the frame position is an intermediate frame, it is determined whether the vehicle has a start / stop moment in the intermediate frame based on the target speed corresponding to the intermediate frame.

6. The vehicle state change scene recognition method according to claim 5, characterized in that: If the frame position is an intermediate frame, then based on the target speed corresponding to the intermediate frame, the step of determining whether the vehicle has a start-stop moment in the intermediate frame includes: If the frame position is an intermediate frame, determining a sequence number of a target speed corresponding to the intermediate frame in the speed sequence, and determining an acceleration direction corresponding to a target acceleration associated with the sequence number of the target speed; If the acceleration direction corresponding to the target acceleration is a first direction, the target speed is greater than a first threshold, the speed corresponding to the frame before the frame position is less than or equal to the first threshold, and the speed after the preset number of frames at the frame position is greater than or equal to a second threshold, then determining the moment corresponding to the target speed as the start moment; If the acceleration direction corresponding to the target acceleration is the second direction, the target speed is less than or equal to the first threshold, the speed corresponding to the frame before the frame position is greater than the first threshold, and the speed before the preset number of frames at the frame position is greater than or equal to the second threshold, then the moment corresponding to the target speed is determined to be the stop moment, wherein the first direction is opposite to the second direction.

7. The vehicle state change scene recognition method according to claim 2, characterized in that: If the vehicle has the start-stop moment, the step of identifying whether there is an obstacle in the scene data frame associated with the start-stop moment includes: If the vehicle has the start-stop moment, determining whether there is an obstacle in each frame of the scene data segment; If the obstacle exists, an obstacle is identified, wherein the category corresponding to the obstacle is consistent with the preset category of obstacles to be considered, the speed of the obstacle is greater than or equal to the start threshold, and the average speed of the obstacle is greater than a preset average speed threshold.

8. The vehicle state change scene recognition method according to claim 6, characterized in that: If the obstacle exists, the step of determining the relative position between the obstacle and the vehicle to obtain a vehicle state change scene recognition result includes: Normalizing the navigation angle of the ego vehicle in each frame of the scene data segment to obtain a normalized heading angle within a preset range, and converting the heading angle into a first degree; Calculating a second degree corresponding to the normalized heading angle of the obstacle, and determining a direction angle difference between the vehicle and the obstacle based on the first degree and the second degree; Calculating the local coordinate difference of the obstacle relative to the vehicle; A vehicle state change scene recognition result is determined based on the direction angle difference and the local coordinate difference.

9. The vehicle state change scene recognition method according to claim 8, characterized in that: The local coordinate difference includes a vertical coordinate difference and a horizontal coordinate difference. The step of determining a vehicle state change scene recognition result based on the direction angle difference and the local coordinate difference includes: If the vertical coordinate difference is less than or equal to the preset vertical coordinate difference, the horizontal coordinate difference is within the preset horizontal coordinate difference range, and the direction angle difference is less than or equal to the preset angle, then the vehicle state change scenario corresponding to the starting moment of the vehicle is the following starting scenario, and the vehicle state change scenario corresponding to the stopping moment of the vehicle is the braking scenario.

10. A vehicle state change scene recognition device, characterized in that: The device comprises: An acquisition module, configured to acquire scene data segments and extract scene-related information of each frame from the scene data segments; an identification module, configured to identify whether there is an obstacle associated with the vehicle in the scene data segment based on the scene-related information; The judgment module is used to obtain a vehicle state change scene recognition result by judging the relative position of the obstacle and the vehicle if the obstacle exists, wherein the vehicle state change scene recognition result includes a braking scene and a following vehicle starting scene.

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

12. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the vehicle state change scene recognition method according to any one of claims 1 to 9 are implemented.

13. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the vehicle state change scene recognition method according to any one of claims 1 to 9 are implemented.

Images