Intelligent driving data generation method and apparatus

Abstract

Provided in the present application are an intelligent driving data generation method, an apparatus and a vehicle, which can be applied to the field of intelligent driving. The method comprises: acquiring a drivable area of a vehicle; on the basis of the drivable area, a driving trajectory of the vehicle, and a movement trajectory of a dynamic object around the vehicle, determining an insertable area; and inserting a foreground object in the insertable area. In this way, human participation is not required in the process of generating intelligent driving data, helping to improve the efficiency of generating the intelligent driving data and also to reduce the cost of generating the intelligent driving data.

Description

Intelligent driving data generation method and device

[0001] The present application claims priority to the Chinese patent application No. 202411010047.6, filed on July 25, 2024, with the State Intellectual Property Office of China, and the Chinese patent application No. 202411010047.6 has the title of "Intelligent driving data generation method and device", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of intelligent driving, and more particularly, to an intelligent driving data generation method and device. BACKGROUND

[0003] With the development and wide application of intelligent driving technology, the demand for data by intelligent driving perception technology is becoming more and more diversified. The current intelligent driving simulation data generation method is mainly generated by simulation software. The data generated by the simulation software has poor authenticity, and the way of adding foreground to the real background is to paste the map. This method needs more human participation. This will result in low efficiency and high cost of simulation data generation. SUMMARY

[0004] The present application provides an intelligent driving data generation method and device, which does not need human participation in the process of intelligent driving data generation, helps to improve the efficiency of intelligent driving data generation, and also helps to reduce the cost of intelligent driving data generation.

[0005] In a first aspect, an intelligent driving data generation method is provided, which includes: obtaining a drivable area of a vehicle; determining an insertable area according to the drivable area, a driving trajectory of the vehicle, and a moving trajectory of a dynamic target around the vehicle; and inserting a foreground target in the insertable area.

[0006] Based on the above technical solution, by combining the driving trajectory of the vehicle and the moving trajectory of the dynamic target around the vehicle (or, it can be called traffic flow information), the insertable area can be automatically generated and the foreground target can be inserted in the insertable area. In this way, without human participation, it helps to improve the efficiency of intelligent driving data generation, and also helps to reduce the cost of intelligent driving data generation. At the same time, by introducing the traffic flow information, the collision between the inserted foreground target and the vehicle or the dynamic target is avoided, which can realize accurate intelligent driving data generation and improve the quality and availability of intelligent driving data.

[0007] In some possible implementation manners, the driving trajectory of the vehicle can be a driving trajectory of the vehicle in a future period of time.

[0008] In some possible implementation manners, the moving track of the dynamic target can be a moving track of the dynamic target in a future period of time. Taking the dynamic target as a dynamic vehicle for example, the moving track of the dynamic vehicle can be a driving track of the dynamic vehicle in a future period of time.

[0009] In some possible implementation manners, the dynamic target includes, but is not limited to, a vehicle, a pedestrian, a motorcycle, a non-motor vehicle, and the like.

[0010] In some possible implementation manners, the intelligent driving data can also be referred to as intelligent driving simulation data.

[0011] With reference to the first aspect, in some implementation manners of the first aspect, the inserting the foreground target in the insertable region includes: obtaining a first sample point set, the first sample point set including a plurality of sample points; performing screening on the first sample point set according to an occlusion relationship between the foreground target and the background at the plurality of sample points to obtain a second sample point set; and inserting the foreground target at each sample point in the second sample point set.

[0012] Based on the technical solution described above, the sample points can be screened according to the occlusion relationship between the foreground target and the background, so that the sample points that are occluded by the background can be screened out in the process of automatically inserting the sample points. For example, when a sample point is located at a gate, the foreground target inserted at the sample point can be occluded by a roadside tree or a fence, and the sample point can be excluded at this time. In this way, the sample points inserted at the gate but not on the road can be excluded, which helps to improve the availability of the intelligent driving data.

[0013] With reference to the first aspect, in some implementation manners of the first aspect, a line connecting the position of the vehicle and each sample point in the second sample point set does not intersect with the contour of the drivable region.

[0014] Based on the technical solution described above, whether the line connecting the position of the vehicle and the sample point intersects with the contour of the drivable region can be used to determine whether the foreground target inserted at the sample point is occluded by the background. In this way, by excluding the sample points that are occluded by the background, the availability of the data can be improved.

[0015] With reference to the first aspect, in some implementation manners of the first aspect, the performing screening on the first sample point set according to the occlusion relationship between the plurality of sample points and the background to obtain the second sample point set includes: performing screening on the first sample point set according to the occlusion relationship between the plurality of sample points and the background to obtain a third sample point set; and clustering the sample points in the third sample point set to obtain the second sample point set.

[0016] Based on the technical solution, the position of the insertable region and the insertable point can be automatically generated through the occlusion relationship judgment and clustering analysis of the foreground target and the background, accurate simulation data generation can be achieved, and the simulation data quality is improved. At the same time, it helps to reduce the artificial verification cost and improve the data production efficiency.

[0017] At the same time, the clustering algorithm can be used to filter out sampling points far apart, which helps to improve the diversity of intelligent driving data.

[0018] In combination with the first aspect, in some implementations of the first aspect, inserting the foreground target in the insertable region includes: obtaining a first sampling point; when there is occlusion between the foreground target and the background at the first sampling point and the foreground target and the background at the first sampling point can be segmented, inserting the foreground target at the first sampling point.

[0019] Based on the technical solution, when the foreground target is occluded by the background and the foreground target and the background can be segmented, the foreground target can also be inserted at the sampling point, which helps to improve the selection range of the sampling point while ensuring data availability, and helps to improve the diversity of intelligent driving data.

[0020] In combination with the first aspect, in some implementations of the first aspect, inserting a plurality of sampling points in the insertable region includes: determining a first region according to the driving trajectory of the vehicle; and inserting the plurality of sampling points in a region overlapping the first region and the insertable region.

[0021] Based on the technical solution, by combining the first region determined by the driving trajectory of the vehicle and the insertable region, the overlapping region can be obtained, so that the sampling points can be inserted in the overlapping region.

[0022] In some possible implementations, the first region is a region where the middle and rear sections of the driving trajectory of the vehicle are located.

[0023] In combination with the first aspect, in some implementations of the first aspect, the inserting the foreground target in the insertable region includes: in the insertable region, obtaining information of a road where a second sampling point is located; determining a pose of the foreground target at the second sampling point according to the information of the road; and inserting the foreground target at the second sampling point according to the pose.

[0024] Based on the technical solution, the pose of the foreground target can be automatically generated based on the information of the road. In this way, the foreground target inserted at the insertion point is more realistic, which helps to ensure the authenticity and accuracy of the simulation data.

[0025] For example, taking the construction sign as the foreground target, by combining the information of the road, one side of the construction sign with pictures or texts at the sampling point can face the vehicle.

[0026] With reference to the first aspect, in some implementations of the first aspect, the foreground object includes one or more of a rockery, a water barrier, a road sign, a manhole cover, a construction sign, a road arrow, a lane line, a traffic light, and a road stop line.

[0027] With reference to the first aspect, in some implementations of the first aspect, the obtaining the drivable area of the vehicle includes: obtaining data collected by one or more cameras outside a cabin of the vehicle; and inputting the data into a prediction model to obtain the drivable area.

[0028] Based on the above technical solution, the drivable area of the vehicle can be obtained through the data collected by the camera. In this way, the high-precision map can not be relied on in the process of generating the simulation data, and the amount of manual participation in the intelligent driving data generation process can be greatly reduced.

[0029] In some possible implementations, the inputting the data into the prediction model to obtain the drivable area includes: inputting the data into the prediction model to obtain a three-dimensional semantic map, the three-dimensional semantic map including information of the drivable area.

[0030] In some possible implementations, the three-dimensional semantic map further includes information of a lane line, a road edge, or a road sign.

[0031] With reference to the first aspect, in some implementations of the first aspect, the method further includes: fusing the foreground object and the background to obtain the intelligent driving data.

[0032] The second aspect provides an intelligent driving data generation device. The device includes: an obtaining unit configured to obtain a drivable area of a vehicle; a determining unit configured to determine an insertable area according to the drivable area, a driving trajectory of the vehicle, and a moving trajectory of a dynamic target around the vehicle; and a foreground object insertion unit configured to insert a foreground object into the insertable area.

[0033] With reference to the second aspect, in some implementations of the second aspect, the obtaining unit is further configured to obtain a first sample point set, the first sample point set including a plurality of sample points; a screening unit is configured to screen the first sample point set according to an occlusion relationship between the foreground object and the background at the plurality of sample points to obtain a second sample point set; and the foreground object insertion unit is specifically configured to insert a foreground object at each sample point in the second sample point set.

[0034] With reference to the second aspect, in some implementations of the second aspect, a line connecting the position of the vehicle and each sample point in the second sample point set does not intersect with an outline of the drivable area.

[0035] With reference to the second aspect, in some implementations of the second aspect, the screening unit is specifically configured to: screen the first set of sampling points according to the occlusion relationship between the sampling points and the background, to obtain a third set of sampling points; and cluster the sampling points in the third set of sampling points to obtain the second set of sampling points.

[0036] With reference to the second aspect, in some implementations of the second aspect, the obtaining unit is further configured to obtain a first sampling point; and the foreground object insertion unit is specifically configured to: insert a foreground object at the first sampling point when the foreground object and the background at the first sampling point exist occlusion and the foreground object and the background at the first sampling point can be segmented.

[0037] With reference to the second aspect, in some implementations of the second aspect, the determining unit is further configured to determine a first region according to the driving trajectory of the vehicle; and the foreground object insertion unit is specifically configured to: insert the plurality of sampling points in a region where the first region and the insertable region overlap.

[0038] With reference to the second aspect, in some implementations of the second aspect, the obtaining unit is further configured to obtain information of a road where a second sampling point is located in the insertable region; the determining unit is further configured to determine a pose of a foreground object at the second sampling point according to the information of the road; and the foreground object insertion unit is specifically configured to: insert a foreground object at the second sampling point according to the pose.

[0039] With reference to the second aspect, in some implementations of the second aspect, the foreground object includes one or more of a boulder, a water barrier, a road surface mark, a manhole cover, a construction mark, a road surface arrow, a lane line, a traffic indicator light, and a road stop line.

[0040] With reference to the second aspect, in some implementations of the second aspect, the obtaining unit is specifically configured to: obtain data collected by one or more camera devices outside the vehicle cabin; and input the data into a prediction model to obtain the drivable region.

[0041] With reference to the second aspect, in some implementations of the second aspect, the device further includes a fusion unit configured to fuse the foreground object and the background to obtain the intelligent driving data.

[0042] In a third aspect, the present application provides an intelligent driving data generation device, which includes a memory and a processor, the memory is configured to store a computer program, and the processor is configured to execute the computer program in the memory, so that the intelligent driving device can implement the method in the first aspect and any possible implementation manner thereof.

[0043] In a fourth aspect, the present application provides an intelligent driving data generation system, which comprises a perception system and the device of the second aspect or the third aspect.

[0044] For example, the perception system can be located in a vehicle, and the device can be located in a server or a data processing device.

[0045] In a fifth aspect, the present application provides a data processing device, which comprises the device of the second aspect or the third aspect, or the system of the fourth aspect.

[0046] The vehicle in the present application is a vehicle in a broad sense, which can be a traffic tool (such as a commercial vehicle, a passenger vehicle, a motorcycle, a flying vehicle, a train, etc.), an industrial vehicle (such as a forklift, a trailer, a tractor, etc.), an engineering vehicle (such as an excavator, a bulldozer, a crane, etc.), an agricultural device (such as a mower, a harvester, etc.), a recreational device, a toy vehicle, etc. The type of the vehicle is not limited in the embodiments of the present application.

[0047] In a sixth aspect, the present application provides a computer program product, which comprises computer program code, when the computer program code is run on a computer, the computer program code causes the computer to execute the method in any possible implementation manner of the first aspect.

[0048] In a seventh aspect, the present application provides a computer readable storage medium, which stores a computer program, when the computer program is run on a computer, the computer program causes the computer to execute the method in any possible implementation manner of the first aspect.

[0049] In an eighth aspect, the present application provides a chip, which comprises a circuit for executing the method in any possible implementation manner of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0050] FIG. 1 is a functional block diagram provided by an embodiment of the present application.

[0051] FIG. 2 is a schematic block diagram of an intelligent driving system provided by an embodiment of the present application.

[0052] FIG. 3 is a schematic flowchart of an intelligent driving data generation method provided by an embodiment of the present application.

[0053] FIG. 4 is a schematic diagram of an intelligent driving scene provided by an embodiment of the present application.

[0054] FIG. 5 is another schematic diagram of an intelligent driving scene provided by an embodiment of the present application.

[0055] FIG. 6 is another schematic diagram of an intelligent driving scene provided by an embodiment of the present application.

[0056] FIG. 7 is another schematic diagram of the intelligent driving scene according to an embodiment of the present application.

[0057] FIG. 8 is a schematic diagram of foreground object insertion and background separation according to an embodiment of the present application.

[0058] FIG. 9 is another schematic diagram of the intelligent driving scene according to an embodiment of the present application.

[0059] FIG. 10 is another schematic flowchart of the method for generating intelligent driving data according to an embodiment of the present application.

[0060] FIG. 11 is a schematic block diagram of the apparatus for generating intelligent driving data according to an embodiment of the present application. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings. In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; in this document, "and / or" is merely a description of the association relationship between the associated objects, which means that there can be three relationships, for example, A and / or B, which means that there can be three relationships, for example, A and B, and B alone. "At least one" means one or more. For example, "at least one of A and B" is similar to "A and / or B", which describes the association relationship between the associated objects, which means that there can be three relationships, for example, A and B, and B alone.

[0062] In the embodiments of the present application, the prefix words such as "first", "second" are used only to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of ordinal words such as prefixes in the embodiments of the present application does not limit the described objects, and the description of the described objects in the claims or embodiments should not be construed as redundant restrictions. In addition, in the description of the embodiments, unless otherwise specified, the meaning of "multiple" is two or more.

[0063] FIG. 1 is a functional block diagram of a vehicle 100 according to an embodiment of the present application. The vehicle 100 can include a perception system 110, a computing platform 120, and a display device 130. The perception system 110 can include one or more sensors that sense information about the environment surrounding the vehicle 100. For example, the perception system 110 can include a positioning system, which can be a global positioning system (GPS), a Beidou system, or another positioning system. For another example, the perception system 110 can include one or more of an inertial measurement unit (IMU), an acceleration sensor, a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera.

[0064] Some or all functions of the vehicle 100 can be controlled by the computing platform 120. The computing platform 120 can include one or more processors, such as processors 121 through 12n (n is a positive integer), which are circuits having a processing capability of signals. In one implementation, the processors can be circuits having an instruction reading and running capability, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a kind of microprocessor), a digital signal processor (DSP), or the like. In another implementation, the processors can be circuits having a certain function implemented by a logic relationship of hardware circuits, which is fixed or reconfigurable. For example, the processors can be hardware circuits implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field programmable gate array (FPGA). In the reconfigurable hardware circuit, the processor loads a configuration document to implement the hardware circuit configuration. It can be understood that the processor loads instructions to implement the functions of the above units. In addition, the processors can also be hardware circuits designed for artificial intelligence, which can be understood as a kind of ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), or the like. In addition, the computing platform 120 can also include a memory for storing instructions, and some or all of the processors 121 through 12n can call the instructions in the memory to implement corresponding functions.

[0065] The display device 130 in the cabin is mainly divided into two categories, the first category is a vehicle display screen, and the second category is a projection display screen, such as a head up display (HUD). The vehicle display screen is a physical display screen and is an important component of the in-vehicle infotainment system. Multiple display screens can be provided in the cabin, such as a digital instrument display screen, a center control screen, a display screen in front of a passenger (also referred to as a front passenger) at a co-driver position, a display screen in front of a left rear passenger, and a display screen in front of a right rear passenger, or even a vehicle window can be used as a display screen for display. The head up display, also known as a head-up display system, is mainly used for displaying driving information such as speed, navigation, etc. on a display device (such as a windshield) in front of the driver. This reduces the time for the driver to change his line of sight and avoids changes in the pupil caused by the driver changing his line of sight, thereby improving driving safety and comfort. The HUD includes, for example, a combiner-HUD (C-HUD) system, a windshield-HUD (W-HUD) system, and an augmented reality HUD (AR-HUD). It should be understood that other types of systems can also appear as the technology evolves, and the present application does not limit this.

[0066] The display device 130 described above is illustrated by taking the vehicle display screen and the projection display screen as examples, and embodiments of the present application are not limited thereto. For example, the display device 130 can also be a light display screen or a projection screen.

[0067] Optionally, the structure of the vehicle 100 described above is only schematic, and in actual applications, various components in the vehicle 100 described above can be added or deleted according to actual needs.

[0068] The vehicle 100 can include an intelligent driving system, which can include an advanced driving assistant system (ADAS) and an autonomous driving system (ADS). The intelligent driving system uses various sensors (including but not limited to laser radar, millimeter wave radar, camera, ultrasonic sensor, global positioning system, and inertial measurement unit) on the vehicle to obtain information from the surroundings of the vehicle, and analyzes and processes the obtained information to realize functions such as obstacle perception, target recognition, vehicle positioning, path planning, driver monitoring / reminding, etc., thereby improving the safety, automation level, and comfort of vehicle driving.

[0069] For example, FIG. 2 shows a schematic block diagram of an intelligent driving system according to an embodiment of the present application. The intelligent driving system can include three functional modules: a perception module 210, a planning module 220, and a control module 230. The perception module 210 perceives the environment around the vehicle body through sensors and outputs corresponding perception data to the planning module 220. The planning module 220 obtains information of road elements based on the information obtained by the perception module 210. The planning module 220 can determine the physical connectivity of the vehicle from the current position to a sampling point based on the current position of the vehicle and the information of the road elements, and plan a driving trajectory of the vehicle from the current position to the sampling point when the vehicle is physically connected from the current position to the sampling point. The planning module 220 can determine a strategy space of the vehicle according to the driving trajectory. The planning module 220 can send the strategy space to the control module 230. The control module 230 can evaluate in the Euclidean space based on the strategy space, thereby making a behavior decision or an interaction decision of the vehicle.

[0070] The perception module 210 described above can be the perception system 110 described above, and the planning module 220 and the control module 230 can be located in the computing platform 120 described above.

[0071] The degree to which a vehicle's driving automation system is capable of performing dynamic driving tasks is divided into levels 0 to 5 (or L0-L5) according to the role allocation in performing dynamic driving tasks and the presence or absence of design operational range (ODD) restrictions, such as external conditions suitable for the functional operation of the driving automation system as determined when the driving automation system is designed, such as roads, traffic, weather, lighting, etc. Among the six levels of driving automation, levels 0-2 are driving assistance, and the system assists humans in performing dynamic driving tasks, and the driving subject is still the driver. Levels 3-5 are autonomous driving, and the system replaces humans to perform dynamic driving tasks under the design operating conditions, and when the function is activated, the driving subject is the system. The names and definitions of each level are as follows:

[0072] A level 0 driving automation (may also be referred to as emergency assistance) system is not capable of sustained driving task execution of vehicle lateral or longitudinal motion control, but has the capability for partial goal and event detection and response in dynamic driving tasks. A level 1 driving automation (may also be referred to as partial driver assistance) system is capable of sustained driving task execution of vehicle lateral or longitudinal motion control, and has partial goal and event detection and response capability commensurate with the vehicle lateral or longitudinal motion control performed, under the conditions for which it is designed to operate. A level 2 driving automation (may also be referred to as combined driver assistance) system is capable of sustained driving task execution of vehicle lateral and longitudinal motion control, and has partial goal and event detection and response capability commensurate with the vehicle lateral and longitudinal motion control performed, under the conditions for which it is designed to operate. A level 3 driving automation (may also be referred to as conditionally automated driving) system is capable of sustained driving task execution of all dynamic driving tasks under the conditions for which it is designed to operate. A level 4 driving automation (may also be referred to as highly automated driving) system is capable of sustained driving task execution of all dynamic driving tasks and the automated execution of a minimal risk strategy under the conditions for which it is designed to operate. A level 5 driving automation (may also be referred to as fully automated driving) system is capable of sustained driving task execution of all dynamic driving tasks and the automated execution of a minimal risk strategy under any and all conditions for which it is designed to operate. Generally, intelligent driving systems are generally L2-L5, such as ADAS is L2, and ADS is L3-L5.

[0073] As described above, with the development and wide application of intelligent driving technology, the demand for data of intelligent driving perception technology is increasingly diversified. The current intelligent driving simulation data generation method is mainly generated by simulation software. The data generated by the simulation software has poor authenticity, and the method of adding foreground to the real background is a way of pasting pictures, which requires more human intervention. This will result in low efficiency and high cost of simulation data generation.

[0074] Embodiments of the present application provide an intelligent driving data generation method, device and vehicle, which does not require human intervention in the process of intelligent driving simulation data generation, helps to improve the efficiency of simulation data generation, and also helps to reduce the cost of simulation data generation.

[0075] FIG. 3 shows a schematic flowchart of an intelligent driving data generation method 300 provided by embodiments of the present application. The method can be executed by a data processing device (for example, a server or a computer). The method 300 comprises:

[0076] S301, acquire data collected by sensors of the vehicle.

[0077] Exemplarily, the sensors include one or more of a camera, a laser radar or a millimeter wave radar.

[0078] Exemplarily, the method 300 is executed by a server. The server can acquire data collected by sensors of the vehicle, and generate intelligent driving data based on the following steps.

[0079] S302, determine a drivable area of the vehicle according to the data.

[0080] In one embodiment, taking the sensor as a camera as an example, determining the drivable area of the vehicle according to the data includes: inputting the image data into a prediction model to obtain a semantic map, the semantic map including lane lines, road edges, drivable areas and the like.

[0081] Exemplarily, the semantic map can be a three-dimensional semantic map.

[0082] Exemplarily, the prediction model can be trained by training data. For example, the training data set includes image data 1 and information of lane lines, road edges and a drivable area 1 corresponding to the image data 1, image data 2 and information of lane lines, road edges and a drivable area 2 corresponding to the image data 2, and the like. The prediction model can be trained based on the training data set.

[0083] Exemplarily, FIG. 4 shows a schematic diagram of an intelligent driving scene provided by an embodiment of the present application.

[0084] As shown in (a) of FIG. 4, the vehicle 100 around includes a vehicle 200 in a driving state, lane lines, guardrails and the like.

[0085] As shown in (b) of FIG. 4, according to image data collected by a camera of the vehicle 100, a drivable area of the vehicle 100 at this time can be determined.

[0086] S303, acquire a driving trajectory of the vehicle and a moving trajectory of a dynamic target around the vehicle.

[0087] Exemplarily, taking the scene shown in (a) of FIG. 4 as an example, the vehicle 100 can acquire a driving trajectory of the vehicle 100 in a future period of time and a driving trajectory of the vehicle 200 in the future period of time.

[0088] In one embodiment, the moving track of the dynamic target around the vehicle is obtained by predicting the moving track of the dynamic target in a future period of time according to a historical moving track of the dynamic target and a current moving parameter (e.g., position, speed, acceleration, heading angle, etc.) of the dynamic target.

[0089] In S304, the insertable region is determined according to the drivable region, the driving track of the vehicle, and the moving track of the dynamic target.

[0090] The insertable region can be understood as a region in which the foreground target can be inserted. For example, the foreground target includes, but is not limited to, one or more of a boulder, a water barrier, a road sign, a manhole cover, a construction sign, a road arrow, and a road stop line.

[0091] The driving track of the vehicle and the moving track of the dynamic target can also be referred to as traffic flow information.

[0092] In one embodiment, the insertable region is determined according to the drivable region, the driving track of the vehicle, and the moving track of the dynamic target, including: determining a moving region of the vehicle and the dynamic target according to the driving track of the vehicle and the moving track of the dynamic target; and determining the insertable region according to the drivable region and the moving region, the insertable region including other regions in the drivable region except the moving region.

[0093] For example, FIG. 5 shows a schematic diagram of an intelligent driving scene provided by an embodiment of the present application.

[0094] As shown in FIG. 5, the moving region 1 and the moving region 2 can be determined according to the driving track of the vehicle 100 in a future period of time and the driving track of the vehicle 200 in a future period of time, respectively. The insertable region can be obtained by removing the moving region 1 and the moving region 2 from the drivable region.

[0095] In one embodiment, the insertable region is determined according to the drivable region, the driving track of the vehicle, and the moving track of the dynamic target, including: determining the insertable region according to the drivable region, the driving track of the vehicle, the moving track of the dynamic target, and the sizes of the vehicle and the dynamic target.

[0096] For example, if the dynamic target is a construction vehicle or a passenger vehicle, the 3D box of the dynamic target occupies a larger space, and the corresponding moving region is also larger. For example, if the dynamic target is a motorcycle, the 3D box of the dynamic target occupies a smaller space, and the corresponding moving region is also smaller.

[0097] Exemplarily, the size of the vehicle and the dynamic target can be represented by a polygonal contour of the vehicle and the dynamic target.

[0098] S305, pre-sampling N sampling points in the insertable region, N being a positive integer greater than 1.

[0099] In an embodiment, pre-sampling N sampling points in the insertable region includes: determining a first region according to the driving track of the vehicle; and pre-sampling N sampling points in a region where the first region and the insertable region overlap.

[0100] Exemplarily, FIG. 6 shows a schematic diagram of an intelligent driving scene provided by an embodiment of the present application.

[0101] Exemplarily, as shown in (a) of FIG. 6, the first region can be a rectangular region 1 of a middle and rear section of the driving track of the vehicle 100 and one-lane-width apart left and right.

[0102] Exemplarily, by taking the intersection between the rectangular region 1 shown in (a) of FIG. 6 and the insertable region shown in (b) of FIG. 6, the overlapping region shown in (c) of FIG. 6 can be obtained. In this way, 10 sampling points can be pre-sampled in the overlapping region.

[0103] S306, screening the N sampling points according to the occlusion relationship between the foreground target inserted at each of the N sampling points and the background, to obtain M sampling points, M being a positive integer less than or equal to N.

[0104] In an embodiment, the N sampling points include a sampling point 1, and screening the N sampling points according to the occlusion relationship between the foreground target inserted at each of the N sampling points and the background includes: determining whether there is occlusion between the foreground target inserted at the sampling point 1 and the background according to whether a line connecting the current position of the vehicle and the sampling point 1 intersects with the contour of the drivable region.

[0105] Exemplarily, if the line connecting the current position of the vehicle and the sampling point 1 intersects with the contour of the drivable region, it can be determined that there is occlusion between the foreground target inserted at the sampling point 1 and the background, and the sampling point 1 can be removed at this time.

[0106] Exemplarily, FIG. 7 shows a schematic diagram of an intelligent driving scene provided by an embodiment of the present application.

[0107] As shown in FIG. 7, taking the sampling point 1 located at a gate (or, a ramp entrance) as an example, the line connecting the current position 1 of the vehicle 100 and the sampling point 1 intersects with the contour of the drivable region, and it can be determined that there is occlusion between the foreground target inserted at the sampling point 1 and the background. At this time, the sampling point 1 can be removed.

[0108] Taking sample point 2 as an example, the line connecting the current position 1 of the vehicle 100 and the sample point 2 does not intersect with the contour of the drivable area, and it can be determined that there is no occlusion between the foreground object inserted at the sample point 2 and the background. At this time, the sample point 2 can be retained.

[0109] The above is an example of removing the sample point when there is an occlusion relationship between the foreground object inserted at the sample point and the background in S306. The embodiments of the present application are not limited thereto. For example, when there is an occlusion relationship between the foreground object inserted at the sample point and the background and the data processing device can segment the foreground object and the background, the sample point can also be retained.

[0110] For example, FIG. 8 shows a schematic diagram of the separation of the inserted foreground object and the background in the embodiments of the present application.

[0111] If a boulder is inserted at the sample point 1, since there is a background (for example, a guardrail) at the contour of the drivable area, the background will cause occlusion to the boulder. At this time, only part of the boulder can be seen from the cabin of the vehicle 100. As shown in FIG. 8, if the data processing device can separate the foreground object and the background through depth estimation, the information of the partial area of the boulder at the current position of the vehicle can also be obtained. In this way, the data processing device can also choose not to remove the sample point.

[0112] S307, clustering the M sample points to obtain L sample points, L is less than M and L is a positive integer.

[0113] For example, the M sample points can be clustered by a K-Nearest Neighbor (KNN) classification algorithm or a Kmeans algorithm to obtain the L sample points.

[0114] Through the above screening and clustering analysis, the insertable area and the position of the insertable point can be automatically generated, the intelligent driving data can be accurately generated, the availability and quality of the intelligent driving data can be improved, meanwhile, the cost of manual verification can be reduced and the data production efficiency can be improved. Meanwhile, the clustering algorithm can be used to screen the sample points far apart, which helps to improve the diversity of the intelligent driving data.

[0115] S308, outputting the coordinates of the L sample points.

[0116] In one embodiment, the semantic map described above can also include a lane line (boundary) layer. By fitting the road equation around the sample points, the pose of the inserted foreground object can be obtained, and thus the coordinates of the L sample points and the pose of the foreground object at each sample point can be output.

[0117] Exemplarily, FIG. 9 shows a schematic diagram of a smart driving scene provided by an embodiment of the present application.

[0118] As shown in (a) of FIG. 9, the sampling point 3 is a sampling point obtained through the above steps.

[0119] As shown in (b) of FIG. 9, by fitting the road equation around the sampling point 3, the pose of the inserted road speed limit sign (for example, speed limit 40 km / h) can be obtained, so that the side with pictures and words in the road speed limit sign faces the vehicle.

[0120] S309, rendering the foreground target and fusing the foreground target and the background to obtain smart driving data.

[0121] Exemplarily, according to the coordinates of the sampling points output in S308, the foreground target can be inserted at each sampling point. After inserting the foreground target, the foreground target can be rendered. After rendering the foreground target, the inserted foreground target and the background can be fused to obtain smart driving data.

[0122] The above smart driving data can also be referred to as smart driving simulation data.

[0123] The foreground target in the above smart driving data can include an inserted foreground target (which can be understood as a virtual foreground target) and a foreground target determined by sensor collected data (for example, a traffic participant such as a vehicle or a pedestrian). The foreground target determined by the sensor collected data can also be understood as a foreground target existing in the real physical world. The background can refer to a target in the smart driving data other than the foreground target.

[0124] FIG. 10 shows a schematic flowchart of a smart driving data generation method 1000 provided by an embodiment of the present application. The method 1000 can be executed by a data processing device (for example, a server or a computer). The method 1000 includes:

[0125] S1010, obtaining a drivable area of a vehicle.

[0126] Optionally, obtaining the drivable area of the vehicle includes: obtaining data collected by one or more cameras outside the vehicle cabin; inputting the data into a prediction model to obtain the drivable area.

[0127] Exemplarily, after inputting the data collected by the camera into the prediction model, a semantic map around the vehicle can be obtained, and the drivable area can be obtained by hierarchical clustering of the semantic map.

[0128] Exemplarily, by road extraction of the semantic map, a road equation can be obtained.

[0129] S1020, determining an insertable region according to the drivable region, the driving trajectory of the vehicle, and the moving trajectory of the dynamic target around the vehicle.

[0130] Optionally, the determining the insertable region according to the drivable region, the driving trajectory of the vehicle, and the moving trajectory of the dynamic target around the vehicle comprises: determining the insertable region according to the drivable region, the driving trajectory of the vehicle, the moving trajectory of the dynamic target around the vehicle, the size of the vehicle, and the size of the dynamic target.

[0131] For example, the size of the vehicle can be represented by a three-dimensional rectangular region (3D box) of the vehicle, or a polygonal contour of the vehicle.

[0132] S1030, inserting a foreground target in the insertable region.

[0133] For example, the foreground target comprises one or more of a bollard, a water horse, a road surface mark, a manhole cover, a construction sign, a road surface arrow, a lane line, a traffic indicator, and a road surface stop line.

[0134] Optionally, the inserting the foreground target in the insertable region comprises: obtaining a first sample point set comprising a plurality of sample points; screening the first sample point set according to the occlusion relationship between the foreground target and the background at the plurality of sample points to obtain a second sample point set; and inserting the foreground target at each sample point in the second sample point set.

[0135] Optionally, the line between the position of the vehicle and each sample point in the second sample point set does not intersect with the contour of the drivable region.

[0136] For example, the screening the first sample point set according to the occlusion relationship between the foreground target and the background at the plurality of sample points to obtain a second sample point set comprises: when the line between the position of the vehicle and a certain sample point in the first sample point set intersects with the polygonal contour of the drivable region, the sample point is removed.

[0137] For example, as shown in FIG. 7, sample point 1 is located at a gate (or, a ramp entrance), and the line between the current position 1 of the vehicle 100 and the sample point 1 intersects with the contour of the drivable region, which can determine that there is occlusion between the foreground target inserted at the sample point 1 and the background. At this time, the sample point 1 can be removed. For another example, sample point 2 is located at a non-gate (or, on a road), and the line between the current position 1 of the vehicle 100 and the sample point 2 does not intersect with the contour of the drivable region, which can determine that there is no occlusion between the foreground target inserted at the sample point 2 and the background. At this time, the sample point 2 can be retained.

[0138] Optionally, the first set of sampling points is filtered according to the occlusion relationship between the plurality of sampling points and the background to obtain a second set of sampling points, including: the first set of sampling points is filtered according to the occlusion relationship between the plurality of sampling points and the background to obtain a third set of sampling points; and the sampling points in the third set of sampling points are clustered to obtain the second set of sampling points.

[0139] Optionally, the third set of sampling points can include the M sampling points. The M sampling points are clustered by a KNN classification algorithm or a Kmeans algorithm to obtain the L sampling points.

[0140] In the embodiments of the present application, the distance between the randomly sampled points can be relatively close, which can affect the diversity of the generated intelligent driving data. By clustering the sampling points, sampling points with relatively far distances can be output, which helps to improve the diversity of the intelligent driving data.

[0141] Optionally, the inserting the foreground object in the insertable region includes: obtaining a first sampling point; when there is occlusion between the foreground object and the background at the first sampling point and the foreground object and the background at the first sampling point can be segmented, inserting the foreground object at the first sampling point.

[0142] Optionally, the inserting the foreground object in the insertable region includes: obtaining a first sampling point; when there is occlusion between the foreground object and the background at the first sampling point and the foreground object and the background at the first sampling point can be segmented, inserting the foreground object at the first sampling point.

[0143] Optionally, the inserting the plurality of sampling points in the insertable region includes: determining a first region according to the driving trajectory of the vehicle; and inserting the plurality of sampling points in a region where the first region and the insertable region overlap.

[0144] Optionally, the first region is a rectangular region that is a middle and rear section of the driving trajectory of the vehicle and extends outward by a preset distance along the vertical direction of the driving trajectory. For example, the preset distance is 3m.

[0145] Optionally, the first region is a rectangular region 1 as shown in (a) of FIG. 6. The overlapping region can be an overlapping region as shown in (c) of FIG. 6.

[0146] Optionally, the inserting the foreground object in the insertable region includes: obtaining information of a road where a second sampling point is located in the insertable region; determining a pose of the foreground object at the second sampling point according to the information of the road; and inserting the foreground object at the second sampling point according to the pose.

[0147] Optionally, the information of the road can be the road equation.

[0148] Optionally, the method 1000 further includes fusing the foreground object and the background to obtain the intelligent driving data.

[0149] FIG. 11 shows a schematic block diagram of an intelligent driving data generation apparatus 1100 provided by an embodiment of the present application. The apparatus 1100 includes an acquisition unit 1110 configured to acquire a drivable area of a vehicle; a determination unit 1120 configured to determine an insertable area according to the drivable area, a driving trajectory of the vehicle, and a moving trajectory of a dynamic object around the vehicle; and a foreground object insertion unit 1130 configured to insert a foreground object in the insertable area.

[0150] Optionally, the acquisition unit 1110 is further configured to acquire a first sample point set including a plurality of sample points; a screening unit is configured to screen the first sample point set according to an occlusion relationship between the foreground object and the background at the plurality of sample points to obtain a second sample point set; and the foreground object insertion unit 1130 is specifically configured to insert a foreground object at each sample point in the second sample point set.

[0151] Optionally, a line between the position of the vehicle and each sample point in the second sample point set does not intersect with an outline of the drivable area.

[0152] Optionally, the screening unit is specifically configured to screen the first sample point set according to an occlusion relationship between the plurality of sample points and the background to obtain a third sample point set; and to cluster the sample points in the third sample point set to obtain the second sample point set.

[0153] Optionally, the acquisition unit 1110 is further configured to acquire a first sample point; and the foreground object insertion unit is specifically configured to insert a foreground object at the first sample point when there is an occlusion between the foreground object and the background at the first sample point and the foreground object and the background at the first sample point can be segmented.

[0154] Optionally, the determination unit 1120 is further configured to determine a first area according to the driving trajectory of the vehicle; and the foreground object insertion unit is specifically configured to insert the plurality of sample points in an area where the first area and the insertable area overlap.

[0155] Optionally, the acquisition unit 1110 is further configured to acquire information of a road where a second sample point is located in the insertable area; the determination unit is further configured to determine a pose of the foreground object at the second sample point according to the information of the road; and the foreground object insertion unit is specifically configured to insert a foreground object at the second sample point according to the pose.

[0156] Optionally, the foreground object comprises one or more of a boulder, a water barrier, a road surface mark, a manhole cover, a construction sign, a road surface arrow, a lane line, a traffic light, and a road surface stop line.

[0157] Optionally, the acquisition unit 1110 is specifically configured to: acquire data collected by one or more camera devices outside the vehicle cabin; and input the data into a prediction model to obtain the drivable area.

[0158] Optionally, the device 1100 further comprises a fusion unit configured to fuse the foreground object and the background to obtain the intelligent driving data.

[0159] It should be understood that the division of each unit in the above device is only a logical functional division, and all or part of the units can be integrated into one physical entity, or can be physically separated. In addition, the units in the device can be implemented in the form of processor calling software; for example, the device comprises a processor, the processor is connected with a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to implement any one of the above methods or to realize the functions of each unit of the device, wherein the processor is, for example, a general processor such as CPU or microprocessor, and the memory is an internal memory of the device or an external memory of the device. Alternatively, the units in the device can be implemented in the form of hardware circuit, and the functions of part or all of the units can be realized by the design of the hardware circuit, which can be understood as one or more processors; for example, in one implementation, the hardware circuit is ASIC, and the functions of part or all of the units are realized by the design of the logical relationship of elements in the circuit; for example, in another implementation, the hardware circuit is PLD, and taking FPGA as an example, it can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to realize the functions of part or all of the units. All units of the above device can be implemented in the form of processor calling software, or all units can be implemented in the form of hardware circuit, or part of the units can be implemented in the form of processor calling software, and the remaining part can be implemented in the form of hardware circuit.

[0160] In this application embodiment, a processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction reading and execution capabilities, such as a CPU, microprocessor, GPU, or DSP. In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor may be a hardware circuit implemented as an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the processor loading instructions to implement the functions of some or all of the above units. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, or DPU.

[0161] As can be seen, each unit in the above device can be one or more processors (or processing circuits) configured to implement the above methods, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0162] Furthermore, the units in the above devices can be integrated in whole or in part, or they can be implemented independently. In one implementation, these units are integrated together as a System-on-a-Chip (SoC). The SoC may include at least one processor for implementing any of the above methods or implementing the functions of the units in the device. The at least one processor may be of different types, such as CPU and FPGA, CPU and AI processor, CPU and GPU, etc.

[0163] This application also provides an intelligent driving data generation device, which includes a processing unit and a storage unit. The storage unit is used to store instructions, and the processing unit executes the instructions stored in the storage unit to enable the device to perform the methods or steps described in the above embodiments.

[0164] Optionally, if the intelligent driving device is located in a data processing device, the aforementioned processing unit may be one or more of the processors 121-12n shown in FIG1.

[0165] This application also provides a data processing device, which may include the above-described apparatus 1000.

[0166] This application also provides a computer program product, which includes computer program code that, when run on a computer, causes the computer to perform the methods described in the above embodiments.

[0167] This application also provides a computer-readable medium storing program code that, when run on a computer, causes the computer to perform the methods described in the above embodiments.

[0168] This application also provides a chip, which includes a circuit for performing the methods described in the above embodiments.

[0169] In implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software. The method disclosed in the embodiments of this application can be directly implemented by a hardware processor, or by a combination of hardware and software modules within the processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, power-on erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, detailed descriptions are omitted here.

[0170] It should be understood that in the embodiments of this application, the memory may include read-only memory and random access memory, and provides instructions and data to the processor.

[0171] It should also be understood that, in the various embodiments of this application, the order of the above-mentioned processes does not imply 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 this application.

[0172] Those skilled in the art will recognize that the units and algorithm steps of the various examples 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 implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0173] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0174] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0175] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0176] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0177] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0178] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be covered. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for generating intelligent driving data, characterized in that, The method comprises: acquiring a drivable area of a vehicle; determining an insertable area according to the drivable area, a driving track of the vehicle, and a moving track of a dynamic target around the vehicle; inserting a foreground target in the insertable area.

2. The method of claim 1, wherein, The inserting the foreground target in the insertable area comprises: acquiring a first sample point set comprising a plurality of sample points; screening the first sample point set according to an occlusion relationship between the foreground target and a background at the plurality of sample points to obtain a second sample point set; inserting the foreground target at each sample point in the second sample point set.

3. The method of claim 2, wherein, A line between a position where the vehicle is located and each sample point in the second sample point set does not intersect with an outline of the drivable area.

4. The method according to claim 2 or 3, characterized in that, The screening the first sample point set according to the occlusion relationship between the plurality of sample points and the background to obtain the second sample point set comprises: screening the first sample point set according to the occlusion relationship between the plurality of sample points and the background to obtain a third sample point set; clustering sample points in the third sample point set to obtain the second sample point set.

5. The method of claim 1, wherein, The inserting the foreground target in the insertable area comprises: acquiring a first sample point; when there is occlusion between the foreground target and the background at the first sample point and the foreground target and the background at the first sample point can be segmented, inserting the foreground target at the first sample point.

6. The method according to any one of claims 2 to 5, characterized in that, The inserting the plurality of sample points in the insertable area comprises: determining a first area according to the driving track of the vehicle; inserting the plurality of sample points in an area where the first area and the insertable area overlap.

7. The method according to any one of claims 1 to 6, characterized in that, The inserting the foreground target in the insertable area comprises: acquiring information of a road where a second sample point is located in the insertable area; determining a pose of the foreground target at the second sample point according to the information of the road; inserting the foreground target at the second sample point according to the pose.

8. The method according to any one of claims 1 to 7, characterized in that, The foreground target comprises one or more of a stone pier, a water horse, a road surface mark, a manhole cover, a construction mark, a road surface arrow, a lane line, a traffic indicator, and a road surface stop line.

9. The method according to any one of claims 1 to 8, characterized in that, The acquiring the drivable area of the vehicle comprises: acquiring data collected by one or more cameras outside a cabin of the vehicle; inputting the data into a prediction model to obtain the drivable area.

10. The method according to any one of claims 1 to 9, characterized in that, The method further comprises: fusing the foreground target and the background to obtain intelligent driving data. 11.An intelligent driving data generation apparatus, characterized by comprising: The apparatus comprises: an acquiring unit configured to acquire a drivable area of a vehicle; a determining unit configured to determine an insertable area according to the drivable area, a driving track of the vehicle, and a moving track of a dynamic target around the vehicle; a foreground target inserting unit configured to insert a foreground target in the insertable area.

12. The apparatus according to claim 11, wherein the acquiring unit is further configured to acquire a first sample point set comprising a plurality of sample points; The screening unit is configured to screen the first set of sampling points according to an occlusion relationship between foreground objects and a background at the plurality of sampling points, to obtain a second set of sampling points. The foreground object insertion unit is specifically configured to insert a foreground object at each sampling point in the second set of sampling points.

13. The apparatus of claim 12, wherein, A line between a position of the vehicle and each sampling point in the second set of sampling points does not intersect with an outline of the drivable area.

14. The apparatus of claim 12 or 13, wherein, The screening unit is specifically configured to: screen the first set of sampling points according to an occlusion relationship between the plurality of sampling points and a background, to obtain a third set of sampling points; and cluster the sampling points in the third set of sampling points, to obtain the second set of sampling points.

15. The apparatus of claim 11, wherein: The acquisition unit is further configured to acquire a first sampling point. The foreground object insertion unit is specifically configured to insert a foreground object at the first sampling point when an occlusion exists between a foreground object and a background at the first sampling point and the foreground object and the background at the first sampling point can be segmented.

16. The apparatus of any one of claims 12 to 15, wherein: The determination unit is further configured to determine a first region according to a driving trajectory of the vehicle. The foreground object insertion unit is specifically configured to insert the plurality of sampling points in a region in which the first region and the insertable region overlap.

17. The apparatus of any one of claims 11 to 16, wherein: The acquisition unit is further configured to acquire, in the insertable region, information of a road at a second sampling point. The determination unit is further configured to determine a pose of a foreground object at the second sampling point according to the information of the road. The foreground object insertion unit is specifically configured to insert a foreground object at the second sampling point according to the pose.

18. The apparatus of any one of claims 11-17, wherein, The foreground object includes one or more of a boulder, a water barrier, a road surface mark, a manhole cover, a construction mark, a road surface arrow, a lane line, a traffic indicator light, and a road stop line.

19. The apparatus of any one of claims 11-18, wherein, The acquisition unit is specifically configured to: acquire data collected by one or more cameras outside a cabin of the vehicle; and input the data into a prediction model, to obtain the drivable area.

20. The apparatus of any one of claims 11-19, wherein, The apparatus further includes: a fusion unit configured to fuse the foreground object and the background, to obtain the intelligent driving data.

21. An intelligent driving data generation apparatus, characterized by comprising: including: a processor configured to execute a computer program stored in a memory, to cause the apparatus to perform the method of any one of claims 1 to 10.

22. The apparatus of claim 21, wherein, The apparatus further includes the memory.

23. A computer-readable storage medium, characterized in that, instructions stored thereon, which, when executed by a processor, cause the processor to implement the method of any one of claims 1 to 10.

24. A computer program product, characterised in that, The computer program product includes computer program code that, when executed on a computer, causes the computer to implement the method of any one of claims 1 to 10.

25. A chip, characterized by The chip includes a circuit configured to perform the method of any one of claims 1 to 10.

Images