Cut-in scene identification method and device, electronic equipment and storage medium
By acquiring and fusing multi-frame multi-sensor data, determining the entry scene of the target vehicle, the problem of inaccurate identification of sensor field angle boundaries is solved, and the safety and accuracy of autonomous driving vehicles are improved.
Patent Information
- Application Number
- CN202310700364.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-13
- Publication Date
- 2024-12-13
AI Technical Summary
In the prior art, the target vehicle cannot accurately identify its intention to enter when it is at the boundary of the field angle of the sensor, making it difficult for autonomous vehicles to predict collision risks and make correct decisions.
By acquiring the driving state parameters of the target vehicle at least 2 frames, integrating the data of multiple on-board sensors, forming the integrated driving state parameters, and determining the cutting mark position based on the preset cutting conditions, and finally determining the cutting scene of the target vehicle.
It improves the accuracy of the identification of target vehicles for the target vehicles to enter the scene, enhances the reliability of avoidance measures, and improves the safety of autonomous vehicles.
Smart Images
Figure CN119116994A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of autonomous driving technology, and in particular, relates to a method, device, electronic device and storage medium for identifying a cut-in scene. Background Art
[0002] At present, autonomous driving has gradually become an important research area in the automotive industry. As the level of autonomous driving increases, the precision of the environment perception required of autonomous driving vehicles becomes higher. In order to avoid collision accidents, it is necessary to accurately identify the movement scene of the target vehicle in order to predict the collision risk and facilitate correct decision-making and planning.
[0003] In the prior art, if the target vehicle is at the boundary of the field of view (FOV) of the sensor, the target vehicle cannot be completely identified, resulting in the inability to accurately identify the target vehicle's intention to cut in.
[0004] The prior art has the problem that when the target vehicle is at the boundary of the field of view of the sensor, the target vehicle's cutting-in intention cannot be accurately identified. Summary of the invention
[0005] The embodiments of the present application provide a method, device, electronic device and storage medium for identifying a cut-in scene, which can solve the problem that the cut-in intention of a target vehicle cannot be accurately identified when the target vehicle is at the boundary of the field of view of the sensor.
[0006] In a first aspect, an embodiment of the present application provides a method for identifying a cut-in scene, comprising:
[0007] Sequentially acquiring driving state parameters of at least two frames of a target vehicle, wherein the target vehicle is a vehicle in a preset field of view area around the autonomous driving vehicle, and the driving state parameters of any frame include multiple driving state parameters of the target vehicle acquired by at least two vehicle-mounted sensors;
[0008] Fusing the driving state parameters of the target vehicle in any frame to form a fused driving state parameter;
[0009] Determine the cut-in mark position of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition;
[0010] Based on the cut-in flag, an cut-in scene for the target vehicle to cut in to the autonomous driving vehicle is determined.
[0011] In one embodiment, the fused driving state parameters of any frame include longitudinal speed, vehicle type, movement direction and position information;
[0012] After the step of fusing the driving state parameters of the target vehicle of any frame to form a fused driving state parameter, the method further includes:
[0013] If the longitudinal speed, the vehicle type, the movement direction and the position information of any frame of the target vehicle meet the preset trigger conditions, the driving state parameters of the target vehicle in any frame are fused by weighted filtering to form a fused driving state parameter.
[0014] In one of the embodiments, the cut-in flag includes a first cut-in flag and a second cut-in flag;
[0015] The step of confirming the cut-in flag of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition includes:
[0016] Determining the first cut-in mark position of the target vehicle based on the fused driving state parameters of each frame of at least one of the vehicle-mounted sensors;
[0017] If the fused driving state parameters of each frame of the target vehicle meet the preset cut-in condition, the second cut-in flag position of the target vehicle is determined.
[0018] In one embodiment, the fused driving state parameters of any frame include lateral speed, lateral distance, vehicle head orientation angle and position information, the vehicle head orientation angle is the angle between the vehicle head orientation of the target vehicle and the vehicle head orientation of the autonomous driving vehicle, and the preset cut-in condition includes a first preset cut-in condition, a second preset cut-in condition, a third preset cut-in condition and a fourth preset cut-in condition;
[0019] The step of determining the second cut-in flag position of the target vehicle if the fused driving state parameters of each frame of the target vehicle meet the preset cut-in condition comprises:
[0020] If the difference direction of the lateral speed of each frame satisfies the first preset entry condition, the difference direction of the lateral distance of each frame satisfies the second preset entry condition, and the difference direction of the vehicle head direction angle of each frame satisfies the third preset entry condition, and the connecting line formed by the position coordinate points of the position information of each frame satisfies the fourth preset entry condition, the second entry mark position of the target vehicle is determined.
[0021] In one embodiment, the step of determining the entry scenario of the target vehicle entering the autonomous driving vehicle based on the entry flag includes:
[0022] Based on the first cut-in flag position and the second cut-in flag position, an cut-in scene of the target vehicle cutting into the autonomous driving vehicle is determined.
[0023] In one of the embodiments, the cut-in flag bit further includes a third cut-in flag bit;
[0024] The step of confirming the cut-in mark of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition further includes:
[0025] Deep learning is performed based on the fused driving state parameters of each frame of the target vehicle to determine the third cut-in mark position of the target vehicle.
[0026] In one of the embodiments, the step of determining the entry scenario of the target vehicle entering the autonomous driving vehicle based on the entry flag position further includes:
[0027] Based on the first cut-in flag, the second cut-in flag and the third cut-in flag, an cut-in scene for the target vehicle to cut into the autonomous driving vehicle is determined.
[0028] In a second aspect, an embodiment of the present application provides a device for identifying a cut-in scene, including:
[0029] an acquisition module, configured to sequentially acquire driving state parameters of at least two frames of a target vehicle, wherein the target vehicle is a vehicle in a preset field of view area around the autonomous driving vehicle, and the driving state parameters of any frame include a plurality of driving state parameters of the target vehicle acquired by at least two vehicle-mounted sensors;
[0030] A fusion module, used for fusing the driving state parameters of the target vehicle in any frame to form a fused driving state parameter;
[0031] A first determination module is used to determine the cut-in mark position of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition;
[0032] The second determination module is used to determine an entry scene of the target vehicle entering the automatic driving vehicle based on the entry flag.
[0033] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements a method as described in any one of the contents of the first aspect above.
[0034] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method as described in any one of the contents of the first aspect above is implemented.
[0035] In a fifth aspect, an embodiment of the present application provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any of the methods described in the first aspect above.
[0036] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.
[0037] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0038] The present application sequentially obtains driving state parameters of at least 2 frames of a target vehicle, where the target vehicle is a vehicle in a preset field of view around the autonomous driving vehicle, and the driving state parameters of any frame include multiple driving state parameters of the target vehicle obtained by at least 2 on-board sensors; the driving state parameters of the target vehicle in any frame are fused to form fused driving state parameters; based on the fused driving state parameters of the target vehicle in each frame and preset entry conditions, a cut-in flag of the target vehicle is determined; based on the cut-in flag, an entry scene of the target vehicle entering the autonomous driving vehicle is determined; since the fused driving state parameters of multiple frames and multiple sensors are fused to obtain the fused driving state parameters, and the cut-in flag is determined by judging whether the preset entry conditions are met, and then the cut-in scene of the target vehicle is determined, the accuracy of the autonomous driving vehicle in identifying the cut-in scene of the target vehicle is improved, which facilitates the autonomous driving vehicle to take avoidance measures to avoid collision with the target vehicle, thereby improving the safety of the autonomous driving vehicle. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0040] Figure 1 It is a flowchart of a method for identifying an entry scene provided by an embodiment of the present application;
[0041] Figure 2 is a schematic diagram of a preset viewing angle area provided in an embodiment of the present application;
[0042] Figure 3 It is a flow chart of confirming the cut-in mark position of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in conditions provided by an embodiment of the present application;
[0043] Figure 4 is a schematic diagram of a fourth preset cut-in condition provided in an embodiment of the present application;
[0044] Figure 5 It is a structural diagram of a scene-cutting recognition device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0045] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0046] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.
[0047] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0048] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0049] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
[0050] At present, autonomous driving has gradually become an important research area in the automotive industry. As the level of autonomous driving increases, the precision of the environment perception required of autonomous driving vehicles becomes higher. In order to avoid collision accidents, it is necessary to accurately identify the movement scene of the target vehicle in order to predict the collision risk and facilitate correct safety decision-making planning.
[0051] In the prior art, each sensor has its own field of view. If the target vehicle is at the boundary of the field of view (FOV) of any one or more sensors, the target vehicle cannot be fully identified, resulting in the inability to accurately identify the target vehicle's intention to cut in. For example, the target vehicle is moving forward and parallel to the autonomous driving vehicle, but the sensor identifies the direction of movement as backward, which is opposite to the actual direction of movement of the target vehicle; for another example, the target vehicle has not crossed the right lane line of the lane where the autonomous driving vehicle is located, but the sensor identifies the target vehicle has pressed the right lane line of the lane where the autonomous driving vehicle is located; for another example, the lateral speed of the target vehicle is small, but the lateral speed of the target vehicle identified by the sensor is greater than the actual lateral speed, resulting in inaccurate identification of the target vehicle's intention to cut in.
[0052] The prior art has the problem that when the target vehicle is at the boundary of the field of view of the sensor, the target vehicle's cutting-in intention cannot be accurately identified.
[0053] The present application sequentially obtains driving state parameters of at least 2 frames of a target vehicle, where the target vehicle is a vehicle in a preset field of view around the autonomous driving vehicle, and the driving state parameters of any frame include multiple driving state parameters of the target vehicle obtained by at least 2 on-board sensors; the driving state parameters of the target vehicle in any frame are fused to form fused driving state parameters; based on the fused driving state parameters of the target vehicle in each frame and preset entry conditions, a cut-in flag of the target vehicle is determined; based on the cut-in flag, an entry scene of the target vehicle entering the autonomous driving vehicle is determined; since the fused driving state parameters of multiple frames and multiple sensors are fused to obtain the fused driving state parameters, and the cut-in flag is determined by judging whether the preset entry conditions are met, and then the cut-in scene of the target vehicle is determined, the accuracy of the autonomous driving vehicle in identifying the cut-in scene of the target vehicle is improved, which facilitates the autonomous driving vehicle to take avoidance measures to avoid collision with the target vehicle, thereby improving the safety of the autonomous driving vehicle.
[0054] The technical solution of the present application is described below through specific embodiments.
[0055] First, as Figure 1 As shown, this embodiment provides a method for identifying a cut-in scene, including:
[0056] S100, sequentially obtain driving state parameters of at least 2 frames of the target vehicle, where the target vehicle is a vehicle in a preset field of view area around the autonomous driving vehicle, and the driving state parameters of any frame include multiple driving state parameters of the target vehicle obtained by at least 2 vehicle-mounted sensors.
[0057] In one embodiment, driving state parameters of at least 2 frames of a target vehicle are acquired in sequence. The target vehicle is a vehicle in a preset field of view angle area around the autonomous driving vehicle. Limiting the target vehicle to the preset field of view angle area is beneficial for filtering out irrelevant target vehicles, reducing the amount of calculation of the autonomous driving vehicle, increasing the reaction speed, and improving the safety of the autonomous driving vehicle.
[0058] In one embodiment, the vehicle-mounted sensor includes any one of a lidar sensor, a camera, a millimeter-wave radar sensor, or an ultrasonic radar sensor; for example, the multiple sensors include any two of a forward lidar sensor, a corner lidar sensor, a surround camera, a forward millimeter-wave radar sensor, a corner millimeter-wave radar sensor, and an ultrasonic radar sensor.
[0059] In one embodiment, Figure 2 As shown, the preset field of view angle area is set to an area where the angle between the first side and the second side in the front direction of the autonomous driving vehicle is less than or equal to 150°, along the direction from the front to the rear of the autonomous driving vehicle, the central axis of the width of the autonomous driving vehicle is the center line, the first side is a line where the center line is rotated clockwise at an angle less than or equal to 75°, and the second side is a line where the center line is rotated counterclockwise at an angle less than or equal to 75°.
[0060] S200, various driving state parameters of the target vehicle of any frame are fused to form a fused driving state parameter.
[0061] In one embodiment, the driving state parameters of the target vehicle in any frame are fused to form a fused driving state parameter, which can process multiple frames of target vehicle information output by multiple on-board sensors, and fuse the data of driving state parameters with the same attributes to obtain better detection performance than a single sensor, thereby improving the detection performance of the autonomous driving vehicle. The fusion method can also enhance the system function or increase the system safety features. For example, the millimeter wave radar and laser radar sensors simultaneously detect the target vehicle in front of the autonomous driving vehicle, and fuse the driving state parameters of the target vehicle, and output the fused driving state parameters of the fused target vehicle, thereby improving the detection accuracy, and the autonomous driving vehicle can still operate in the event of failure of any sensor, thus meeting the functional safety requirements of failure-operational capability.
[0062] In one embodiment, the fused driving state parameters of any frame include longitudinal speed, vehicle type, movement direction and position information. The longitudinal speed is the speed of the target vehicle along the longitudinal driving direction of the autonomous driving vehicle. Vehicle types include passenger cars, trucks, special vehicles, motorcycles and electric vehicles. The movement direction is the driving direction of the autonomous driving vehicle or the target vehicle. The position information includes the position coordinate point of the target vehicle in each frame.
[0063] Due to unexpected driving situations of the target vehicle, the driving state parameters may jump. In order to avoid distortion of the fused driving state parameters of previous and next frames or multiple frames, it is necessary to filter the fused driving state parameters that trigger the preset trigger conditions to screen out interference and improve the accuracy of the fused driving state parameters.
[0064] In one embodiment, after the step of fusing the driving state parameters of the target vehicle of any frame to form a fused driving state parameter, the method further includes:
[0065] If the longitudinal speed, vehicle type, movement direction and position information of the target vehicle in any frame meet the preset trigger conditions, the driving state parameters of the target vehicle in any frame are fused by weighted filtering to form a fused driving state parameter.
[0066] In one embodiment, the preset trigger condition includes at least one of the following:
[0067] The longitudinal speed of the target vehicle in any frame is greater than or equal to the preset longitudinal speed;
[0068] The vehicle type of any frame of the target vehicle is the preset vehicle type;
[0069] The moving direction of the target vehicle in any frame meets the preset moving direction;
[0070] The number of historical position information of any frame of position information of the target vehicle is greater than or equal to a preset number, and the position information of any frame of the target vehicle is within a preset filtering area.
[0071] The preset longitudinal speed range is greater than or equal to 60 km / h;
[0072] Preset vehicle types include passenger cars and trucks;
[0073] The preset movement direction includes that the driving directions of the autonomous driving vehicle and the target vehicle are both in the same direction;
[0074] The range of the preset number is greater than or equal to 5;
[0075] The preset filtering area is an area adjacent to the lane where the autonomous driving vehicle is located in the preset field of view angle area, and the straight-line distance to the autonomous driving vehicle is less than or equal to 20m.
[0076] In one embodiment, the fused driving state parameter includes a first fused driving state parameter and a second fused driving state parameter.
[0077] In one embodiment, each driving state parameter of the target vehicle of any frame is fused by weighted filtering to form a fused driving state parameter, including:
[0078] Obtaining various driving state parameters of a target vehicle in any current frame and various driving state parameters of a target vehicle in a previous frame of the current frame;
[0079] Obtaining a first weight coefficient of each driving state parameter of a current frame and a second weight coefficient of each driving state parameter of a previous frame;
[0080] Determine the first fused driving state parameter based on the product of the value of each driving state parameter of the current frame multiplied by the first weight coefficient, and determine the second fused driving state parameter based on the product of the value of each driving state parameter of the previous frame multiplied by the second weight coefficient;
[0081] The fused driving state parameter is determined based on a sum of the first fused driving state parameter and the second fused driving state parameter.
[0082] It can be understood that if the second weight coefficient of the previous frame is greater than the first weight coefficient of the current frame, the change speed of the lateral distance, lateral speed, and front angle of the fused driving state parameters of the target vehicle will become slower, if the second weight coefficient of the previous frame is less than the first weight coefficient of the current frame, the change speed of the lateral distance, lateral speed, and front angle of the fused driving state parameters of the target vehicle will become faster, if the second weight coefficient of the previous frame is equal to the first weight coefficient of the current frame, the change speed of the lateral distance, lateral speed, and front angle of the fused driving state parameters of the target vehicle is between the above two. In this embodiment, the autonomous driving vehicle sets the weight coefficient according to the specific situation of the target vehicle to adapt to various scenarios.
[0083] In another embodiment, the fused driving state parameter further includes a third fused driving state parameter.
[0084] In another embodiment, each driving state parameter of the target vehicle of any frame is fused by weighted filtering to form a fused driving state parameter, further comprising:
[0085] Sequentially obtain the driving state parameters of the target vehicle in any current frame, the previous frame of the current frame, and the previous previous frame of the previous frame;
[0086] Obtaining a first weight coefficient of each driving state parameter of a current frame, a second weight coefficient of each driving state parameter of a previous frame, and a third weight coefficient of each driving state parameter of a previous previous frame;
[0087] Determine the first fused driving state parameter based on the product of the value of each driving state parameter of the current frame multiplied by the first weight coefficient, determine the second fused driving state parameter based on the product of the value of each driving state parameter of the previous frame multiplied by the second weight coefficient, and determine the third fused driving state parameter based on the product of the value of each driving state parameter of the previous frame multiplied by the third weight coefficient;
[0088] The fused driving state parameter is determined based on a sum of the first fused driving state parameter, the second fused driving state parameter, and the third fused driving state parameter.
[0089] In one embodiment, if the vehicle type is a passenger car, the first weight coefficient of the current frame, the second weight coefficient of the previous frame, and the third weight coefficient of the previous frame are set to 0.5, 0.3, and 0.2, respectively; if the vehicle type is a truck, the first weight coefficient of the current frame, the second weight coefficient of the previous frame, and the third weight coefficient of the previous frame are set to 0.4, 0.3, and 0.3, respectively, so as to improve the accuracy of the fused driving state parameters and facilitate the autonomous driving vehicle to identify the driving scene.
[0090] S300, determining a cut-in mark position of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition.
[0091] In one embodiment, the cut-in flag includes a first cut-in flag and a second cut-in flag.
[0092] In one embodiment, Figure 3 As shown, based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in conditions, the cut-in flag position of the target vehicle is confirmed, including:
[0093] S310, determining a first cut-in mark position of the target vehicle based on fused driving state parameters of each frame of at least one vehicle-mounted sensor.
[0094] In one embodiment, the vehicle-mounted sensor is a camera, and image processing is performed based on the fused driving state parameters of each frame of the camera, and the first entry mark position of the target vehicle is determined according to a preset entry mark position condition.
[0095] S320: If the fused driving state parameters of each frame of the target vehicle meet the preset cut-in condition, determine a second cut-in flag position of the target vehicle.
[0096] In one embodiment, if the fused driving state parameters of each frame of the target vehicle meet the preset cut-in conditions, determining the second cut-in mark of the target vehicle can make up for the deficiency of the cut-in mark of the camera sensor to avoid collision caused by misidentification of the camera cut-in mark, thereby improving the accuracy of the autonomous driving vehicle in identifying the cut-in mark.
[0097] In one embodiment, the fused driving state parameters of any frame also include lateral speed, lateral distance, front heading angle and position information, the front heading angle is the angle between the front heading of the target vehicle and the front heading of the autonomous driving vehicle, and the preset entry conditions include a first preset entry condition, a second preset entry condition, a third preset entry condition and a fourth preset entry condition.
[0098] In one embodiment, if the fused driving state parameters of each frame of the target vehicle meet the preset cut-in condition, the step of determining the second cut-in flag position of the target vehicle includes:
[0099] If the difference direction of the lateral speed of each frame meets the first preset entry condition, the difference direction of the lateral distance of each frame meets the second preset entry condition, and the difference direction of the vehicle head direction angle of each frame meets the third preset entry condition, and the connecting line formed by the position coordinate points of the position information of each frame meets the fourth preset entry condition, the second entry mark position of the target vehicle is determined.
[0100] In one embodiment, the first preset entry condition is that the difference direction of the lateral speed of each frame is toward the autonomous driving vehicle, and the difference of the lateral speed of each frame is greater than or equal to 0.1m / s; the second preset entry condition is that the difference direction of the lateral distance of each frame is toward the autonomous driving vehicle, and the difference of the lateral distance of each frame is greater than or equal to 0.2m; the third preset entry condition is that the difference direction of the front angle of each frame is toward the autonomous driving vehicle, and the difference of the front angle of each frame is greater than or equal to 1°; the fourth preset entry condition is that the angle formed by the line connecting the position coordinate points of the position information of each frame and the center line of the autonomous driving vehicle is greater than or equal to 10°, and the center line of the autonomous driving vehicle is the central axis of the width of the autonomous driving vehicle along the direction from the front to the rear of the autonomous driving vehicle, such as Figure 4 shown.
[0101] In another embodiment, the cut-in flag further includes a third cut-in flag.
[0102] In another embodiment, based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition, confirming the cut-in mark of the target vehicle further includes:
[0103] Deep learning is performed based on the fused driving state parameters of each frame of the target vehicle to determine the third entry mark position of the target vehicle.
[0104] In one embodiment, deep learning is performed based on the fused driving state parameters of each frame of the target vehicle to determine the third cut-in mark position of the target vehicle, including:
[0105] Based on the fused driving state parameters of each frame of the target vehicle, a manually labeled first data set and a second data set, as well as an unlabeled third data set are obtained, wherein the first data set represents a training set without a cut-in mark, the second data set represents a training set with a cut-in mark, and the third data set represents fused driving state parameters of any continuous multiple frames;
[0106] Based on the first data set and the second data set, a pre-constructed cut-in recognition model is used for deep learning training until a preset accuracy rate is met to obtain a trained pre-constructed cut-in recognition model;
[0107] The third cut-in flag is obtained by predicting the third data set based on the trained pre-built cut-in recognition model.
[0108] It can be understood that after obtaining the first entry mark and the second entry mark, the third entry mark is determined, which further improves the accuracy of the autonomous driving vehicle in identifying the entry mark.
[0109] S400: Determine an entry scenario for the target vehicle to enter the autonomous driving vehicle based on the entry flag.
[0110] In one embodiment, based on the cut-in flag, the step of determining the cut-in scenario of the target vehicle cutting into the autonomous driving vehicle includes:
[0111] Based on the first cut-in flag and the second cut-in flag, an cut-in scenario for the target vehicle to cut in to the autonomous driving vehicle is determined.
[0112] In one embodiment, based on the first entry mark and the second entry mark, the entry scene of the target vehicle into the autonomous driving vehicle is determined, which can make up for the deficiency of the first entry mark of the camera sensor to avoid collision caused by misidentification of the camera entry mark, thereby improving the accuracy of the autonomous driving vehicle in identifying the entry mark.
[0113] In another embodiment, the step of determining the cut-in scenario of the target vehicle cutting into the autonomous driving vehicle based on the cut-in flag position further includes:
[0114] Based on the first cut-in flag, the second cut-in flag and the third cut-in flag, an cut-in scenario for the target vehicle to cut in to the autonomous driving vehicle is determined.
[0115] In another embodiment, after obtaining the first cut-in flag and the second cut-in flag, the third cut-in flag is added. Since the fused driving state parameters of each frame are predicted in advance using a deep learning model, the amount of calculation is reduced, and the deficiencies of the first cut-in flag of the camera sensor and the second cut-in flag determined based on preset cut-in conditions are further compensated, so as to avoid collisions caused by misidentification of the first cut-in flag of the camera and the second cut-in flag determined based on preset cut-in conditions, thereby further improving the safety of the autonomous driving vehicle.
[0116] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0117] Compared with the prior art, the present embodiment has the following beneficial effects:
[0118] This embodiment sequentially obtains driving state parameters of at least 2 frames of a target vehicle, where the target vehicle is a vehicle in a preset field of view around the autonomous driving vehicle, and the driving state parameters of any frame include multiple driving state parameters of the target vehicle obtained by at least 2 on-board sensors; the driving state parameters of the target vehicle in any frame are fused to form fused driving state parameters; based on the fused driving state parameters of the target vehicle in each frame and preset entry conditions, a cut-in flag of the target vehicle is determined; based on the cut-in flag, a cut-in scene of the target vehicle cutting into the autonomous driving vehicle is determined; since the fused driving state parameters of multiple frames and multiple sensors are fused to obtain the fused driving state parameters, the cut-in flag is determined by judging whether the preset entry conditions are met, and then the cut-in scene of the target vehicle is determined, the accuracy of the autonomous driving vehicle in identifying the cut-in scene of the target vehicle is improved, the autonomous driving vehicle is facilitated to take avoidance measures to avoid collision with the target vehicle, and the safety of the autonomous driving vehicle is improved.
[0119] Second, as Figure 5 As shown, an embodiment of the present application provides a device for identifying a cut-in scene, including:
[0120] The acquisition module 100 is used to sequentially acquire driving state parameters of at least two frames of a target vehicle, where the target vehicle is a vehicle in a preset field of view area around the autonomous driving vehicle, and the driving state parameters of any frame include multiple driving state parameters of the target vehicle acquired by at least two vehicle-mounted sensors.
[0121] The fusion module 200 is used to fuse various driving state parameters of the target vehicle in any frame to form a fused driving state parameter.
[0122] The first determination module 300 is used to determine the cut-in mark of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition.
[0123] The second determination module 400 is used to determine the entry scenario of the target vehicle into the autonomous driving vehicle based on the entry flag.
[0124] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / modules are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.
[0125] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0126] In a third aspect, this embodiment provides a terminal device, including:
[0127] A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described in any one of the above-mentioned first aspects is implemented.
[0128] In a fourth aspect, this embodiment further provides a computer-readable storage medium, including:
[0129] The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the contents of the first aspect is implemented.
[0130] In a fifth aspect, an embodiment of the present application provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any of the methods described in the first aspect above.
[0131] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.
[0132] The method for identifying an entry scene provided in an embodiment of the present application can be applied to terminal devices such as mobile phones, tablet computers, wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, laptop computers, ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (PDA), etc. The embodiment of the present application does not impose any restrictions on the specific type of the terminal device.
[0133] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by a processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc.
[0134] The computer-readable medium may include at least: any entity or device capable of carrying computer program codes to a camera / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electric carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disk. In some jurisdictions, according to legislation and patent practice, a computer-readable medium cannot be an electric carrier signal or a telecommunication signal.
[0135] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0136] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0137] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0138] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0139] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for identifying a cut-in scene, characterized in that: include: Sequentially acquiring driving state parameters of at least two frames of a target vehicle, wherein the target vehicle is a vehicle in a preset field of view area around the autonomous driving vehicle, and the driving state parameters of any frame include multiple driving state parameters of the target vehicle acquired by at least two vehicle-mounted sensors; Fusing the driving state parameters of the target vehicle in any frame to form a fused driving state parameter; Determine the cut-in mark position of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition; Based on the cut-in flag, an cut-in scene for the target vehicle to cut in to the autonomous driving vehicle is determined.
2. The method according to claim 1, characterized in that The fused driving state parameters of any frame include longitudinal speed, vehicle type, movement direction and position information; After the step of fusing the driving state parameters of the target vehicle of any frame to form a fused driving state parameter, the method further includes: If the longitudinal speed, the vehicle type, the movement direction and the position information of any frame of the target vehicle meet the preset trigger conditions, the driving state parameters of the target vehicle in any frame are fused by weighted filtering to form a fused driving state parameter.
3. The method according to claim 1, characterized in that The cut-in flag position includes a first cut-in flag position and a second cut-in flag position; The step of confirming the cut-in flag of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition includes: Determining the first cut-in mark position of the target vehicle based on the fused driving state parameters of each frame of at least one of the vehicle-mounted sensors; If the fused driving state parameters of each frame of the target vehicle meet the preset cut-in condition, the second cut-in flag position of the target vehicle is determined.
4. The method according to claim 3, characterized in that The fused driving state parameters of any frame further include lateral speed, lateral distance, vehicle head orientation angle and position information, the vehicle head orientation angle is the angle between the vehicle head orientation of the target vehicle and the vehicle head orientation of the autonomous driving vehicle, and the preset cut-in condition includes a first preset cut-in condition, a second preset cut-in condition, a third preset cut-in condition and a fourth preset cut-in condition; The step of determining the second cut-in flag position of the target vehicle if the fused driving state parameters of each frame of the target vehicle meet the preset cut-in condition comprises: If the difference direction of the lateral speed of each frame satisfies the first preset entry condition, the difference direction of the lateral distance of each frame satisfies the second preset entry condition, and the difference direction of the vehicle head direction angle of each frame satisfies the third preset entry condition, and the connecting line formed by the position coordinate points of the position information of each frame satisfies the fourth preset entry condition, the second entry mark position of the target vehicle is determined.
5. The method according to claim 4, characterized in that The step of determining the cut-in scenario of the target vehicle cutting into the automatic driving vehicle based on the cut-in flag position includes: Based on the first cut-in flag position and the second cut-in flag position, an cut-in scene of the target vehicle cutting into the autonomous driving vehicle is determined.
6. The method according to claim 3, characterized in that The cut-in flag bit also includes a third cut-in flag bit; The step of confirming the cut-in mark of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition further includes: Deep learning is performed based on the fused driving state parameters of each frame of the target vehicle to determine the third cut-in mark position of the target vehicle.
7. The method according to claim 6, characterized in that The step of determining an entry scenario for the target vehicle to enter the autonomous driving vehicle based on the entry flag position further includes: Based on the first cut-in flag, the second cut-in flag and the third cut-in flag, an cut-in scene for the target vehicle to cut into the autonomous driving vehicle is determined.
8. A device for identifying a cut-in scene, characterized in that: include: an acquisition module, configured to sequentially acquire driving state parameters of at least two frames of a target vehicle, wherein the target vehicle is a vehicle in a preset field of view area around the autonomous driving vehicle, and the driving state parameters of any frame include a plurality of driving state parameters of the target vehicle acquired by at least two vehicle-mounted sensors; A fusion module, used for fusing the driving state parameters of the target vehicle in any frame to form a fused driving state parameter; A first determination module is used to determine the cut-in mark position of the target vehicle based on the fused driving state parameters of the target vehicle in each frame and the preset cut-in condition; The second determination module is used to determine an entry scene of the target vehicle entering the automatic driving vehicle based on the entry flag.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.