A data processing method and system for intelligent driving, and electronic equipment

Abstract

This invention provides a data processing method, system, and electronic device for intelligent driving. The processing method includes: synchronously acquiring various raw data from the vehicle, including vehicle data, road data, and vehicle perception data; setting a sampling period, including the vehicle's starting time; establishing a driving coordinate system with the starting time as the origin, and forming the vehicle trajectory for the sampling period in the driving coordinate system based on the raw data; acquiring the location information of surrounding traffic participants within the sampling period, and forming the surrounding vehicle trajectories based on the location information; processing the road data to form lanes and vehicle centerlines in the driving coordinate system; acquiring the coordinate point set of lanes, vehicle trajectories, and surrounding vehicle trajectories based on the graph of lanes, vehicle trajectories, and surrounding vehicle trajectories in the driving coordinate system; and setting filtering conditions and selecting training samples from the coordinate point set, and storing the training samples. This invention can improve the control efficiency of intelligent driving.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, specifically to an intelligent driving data processing method and system, and electronic equipment. Background Technology

[0002] Intelligent driving is essentially a cognitive engineering exercise involving attention attraction and attention distraction. The prediction of traffic participant behavior in intelligent driving is achieved based on vehicle trajectory prediction models.

[0003] Currently, data-driven methods combined with deep learning algorithms are commonly used. Taking vehicle trajectory prediction algorithms as an example, a large number of vehicle trajectory samples are needed to train the vehicle trajectory prediction model in order for the model to have good traffic participant prediction performance after training. However, the samples used for training in the vehicle trajectory prediction model are raw data collected from the vehicle. If the raw data is directly used in model training, it will affect the efficiency and accuracy of model training. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the present invention provides a data processing method and system for intelligent driving, as well as an electronic device, to solve the above-mentioned technical problems.

[0005] This invention provides a data processing method for intelligent driving, comprising:

[0006] Simultaneously acquire various raw data from the vehicle, including vehicle data, road data, and vehicle perception data;

[0007] The sampling period is set, which includes the starting time of the vehicle;

[0008] A driving coordinate system is established with the starting time as the origin, and the vehicle trajectory for the sampling period is formed in the driving coordinate system based on the original data.

[0009] The location information of traffic participants around the vehicle during the sampling period is obtained, and the vehicle trajectory around the traffic participants is formed based on the location information.

[0010] Process the road data to form lanes and vehicle centerlines in the driving coordinate system;

[0011] Based on the graphs of the lane, the vehicle trajectory, and the surrounding vehicle trajectory in the driving coordinate system, obtain the coordinate point set of the lane, the vehicle trajectory, and the surrounding vehicle trajectory; and

[0012] Set filtering conditions and filter training samples from the set of coordinate points, and store the training samples.

[0013] In one embodiment of the present invention, the step of synchronously acquiring the raw data includes: synchronizing the sampling frequencies of multiple raw data according to the lowest sampling frequency among the vehicle data, the road data, and the vehicle perception data.

[0014] In one embodiment of the present invention, the step of setting the sampling time period includes:

[0015] Obtain the continuous time period including the stated start time; and

[0016] Based on the synchronized sampling frequency, multiple sampling time steps are set within the continuous time period to form the sampling period.

[0017] In one embodiment of the present invention, the step of obtaining the vehicle trajectory includes:

[0018] Determine whether the vehicle's location information in the vehicle data is reliable;

[0019] If the location information is reliable, then the vehicle trajectory is formed in the driving coordinate system based on the location information; and

[0020] If the positioning information is unreliable, the vehicle's position at the sampling time step is obtained based on the vehicle speed and yaw rate in the vehicle data, thus forming the vehicle trajectory.

[0021] In one embodiment of the present invention, the step of obtaining the circumferential vehicle trajectory includes:

[0022] Obtain images of obstacles around the vehicle at the sampling time step, and determine the obstacle type based on the obstacle images;

[0023] In the driving coordinate system, obtain the obstacle position at each sampling time step within the sampling period; and

[0024] In the driving coordinate system, the circumferential trajectory of the obstacle is formed during the sampling period based on the obstacle type and the obstacle position.

[0025] In one embodiment of the present invention, the step of processing the road data includes:

[0026] Acquire high-precision map information within the sampling period, the high-precision map information including multiple discrete map segments; and

[0027] Based on the sampling time step, the discrete map is stitched together according to the center line and boundary line of the lane to form the lane.

[0028] In one embodiment of the present invention, the step of forming the lane further includes:

[0029] Multiple lanes are divided according to the vehicle's centerline, and different lanes are numbered; and

[0030] The relative positions between different lanes are distinguished based on the lane number.

[0031] In one embodiment of the present invention, the step of obtaining the vehicle body centerline includes:

[0032] Based on the vehicle model, the center of the vehicle body is preset; and

[0033] The vehicle centerline is formed based on the center of the vehicle body and the vehicle trajectory.

[0034] This invention provides a data processing system for intelligent driving, comprising:

[0035] The data collection module is used to acquire raw data of the vehicle, including vehicle data, road data, and vehicle perception data.

[0036] The vehicle trajectory generation module is used to set a sampling period, which includes the starting time of the vehicle. Then, a driving coordinate system is established with the starting time as the origin, and the vehicle trajectory of the sampling period is formed in the driving coordinate system according to the original data.

[0037] The vehicle trajectory generation module is used to obtain the location information of traffic participants around the vehicle during the sampling period, and to form the vehicle trajectory of the surrounding traffic participants based on the location information.

[0038] The road data processing module is used to process the road data and form lanes and vehicle centerlines in the driving coordinate system;

[0039] The coordinate point set acquisition module is used to acquire the coordinate point set of the lane, the vehicle trajectory, and the surrounding vehicle trajectory based on the graphics of the lane, the vehicle trajectory, and the surrounding vehicle trajectory in the driving coordinate system; and

[0040] The sample filtering module is used to set filtering conditions, filter training samples from the set of coordinate points, and store the training samples.

[0041] This invention provides an electronic device, the electronic device comprising:

[0042] A memory, wherein computer instructions are stored; and

[0043] A processor that, when executing the computer instructions, implements a data processing method for intelligent driving as described above.

[0044] The beneficial effects of this invention are as follows: The data processing method, system, and electronic device for intelligent driving provided by this invention can simplify the sample acquisition method during intelligent driving, enabling the data perceived by the vehicle to be directly applied to the algorithm model of intelligent driving. The data processing method provided by this invention can also collect and process data within a sampling period to form an effective model reflecting the current state of the vehicle. It can acquire data before the starting time as well as road conditions after the starting time, thus meeting the data volume requirements for machine learning in intelligent driving and facilitating direct computation of the data by the algorithm model. Furthermore, the data processing method provided by this invention can reduce data redundancy during vehicle operation, improve the data processing efficiency of intelligent driving, thereby reducing the iteration cycle of difficult examples, improving the reaction speed of the vehicle, and facilitating the realization of L3 and above intelligent driving assistance.

[0045] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0047] Figure 1 This is an exemplary embodiment of the present application illustrating the interaction of an intelligent driving controller.

[0048] Figure 2 This is a flowchart illustrating a data processing method in an exemplary embodiment of this application.

[0049] Figure 3 This is a schematic diagram of the vehicle's trajectory during a sampling period, as illustrated in an exemplary embodiment of this application.

[0050] Figure 4 yes Figure 2 Step S220 in the illustrated embodiment is a flowchart of an exemplary embodiment.

[0051] Figure 5 This is a coordinate diagram illustrating the construction of a vehicle coordinate system in an exemplary application scenario.

[0052] Figure 6 yes Figure 2 The flowchart of step S230 in the illustrated embodiment is shown in an exemplary embodiment.

[0053] Figure 7 yes Figure 2 Step S240 in the illustrated embodiment is a flowchart of an exemplary embodiment.

[0054] Figure 8 This is a schematic diagram illustrating the stitching of a high-precision map during a sampling period, as shown in an exemplary embodiment of this application.

[0055] Figure 9 This is a block diagram illustrating a data processing system in an exemplary embodiment of this application.

[0056] Figure 10 A schematic diagram of the structure of an electronic device suitable for implementing embodiments of this application is shown. Detailed Implementation

[0057] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0058] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0059] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0060] First, it's important to clarify that intelligent driving and driverless driving are different concepts; intelligent driving is a broader term. Intelligent driving refers to technologies that assist humans in driving, and in special circumstances, completely replace human drivers. Intelligent driving mainly includes three components: network navigation, autonomous driving, and human intervention. Network navigation in intelligent driving addresses questions such as where we are, where we are going, and which lane we should take. Autonomous driving, under the control of an intelligent system, involves lane keeping, overtaking and merging, stopping at red lights and proceeding at green lights, and interacting with traffic signals. Human intervention refers to the driver's appropriate responses to actual road conditions based on a series of prompts from the intelligent system.

[0061] When an intelligent driving vehicle is in autonomous driving mode, it needs to acquire data on dynamic and static targets in the driving environment and predict the behavior of obstacles around the vehicle, i.e., traffic participants, such as the future intentions and trajectories of surrounding vehicles. Traffic participants refer to people in motion or about to enter motion within a vehicle, as well as transporters and traffic managers. Traffic participants include drivers, passengers, travelers, and people directly or indirectly involved in traffic. Taking vehicle trajectory prediction algorithms as an example, a large number of vehicle trajectory samples are needed to train the vehicle trajectory prediction model to achieve good traffic participant prediction performance after training. The foundation of the algorithm model is the training samples; the accuracy of the sample data and the efficiency of the algorithm model directly affect the effectiveness of intelligent driving. Many algorithms are involved in intelligent driving, such as deep neural network (DNN) models; the embodiments of this application do not limit this.

[0062] Figure 1 This is an exemplary embodiment of the present application illustrating the interaction of an intelligent driving controller. Figure 1As shown, the vehicle body 100 refers to the execution entity equipped with intelligent driving functions. The vehicle body includes a control module 101, a perception module 102, and an execution module 103. The control module 101 controls the vehicle's operation and includes drive equipment, thermal management equipment, and power supply equipment, etc. The perception module 102 can acquire information about the driving environment outside the vehicle body 100. The perception module 102 includes various sensors, such as visual cameras and radar ranging sensors. The visual cameras can be monocular or binocular stereo vision, panoramic vision, and infrared cameras, etc. The radar ranging sensors can be lidar, millimeter-wave radar, ultrasonic radar, etc. The perception module 102 is electrically connected to the control module 101, and the control module 101 can receive the sensing information from the perception module 102. The execution module 103 includes braking devices, steering devices, suspension devices, and throttle devices, etc. The execution module 103 is electrically connected to the control module 101 and can execute the operating commands issued by the control module 101.

[0063] like Figure 1 As shown, the vehicle body 100 includes an intelligent driving controller 104 and an in-vehicle network module 105. This invention does not limit the computing power of the intelligent driving controller 104. The intelligent driving controller 104 can be a domain controller with medium to low computing power to achieve, for example, L1 to L2+ level driving assistance functions. The intelligent driving controller 104 can also be a domain controller with high computing power to achieve, for example, L2+ level and above driving assistance functions. The intelligent driving controller 104 is electrically connected to the control module 101 to receive sensing information from the sensing module 102, process the sensing information from the sensing module 102, and output operating commands that control the vehicle body 100 to perform the most appropriate operation at the moment. The in-vehicle network module 105 includes an in-vehicle wireless communication device, enabling the vehicle body 100 to access the internet. The in-vehicle network module 105 is communicatively connected to a cloud server 106 to achieve online big data communication. Map information and navigation information, etc., can be retrieved through the in-vehicle network module 105. The vehicle network module 105 is electrically connected to the intelligent driving controller 104 via the control module 101 to communicate with cloud data and various communication information. The vehicle network module 105 can communicate with the cloud server 106 via wireless networks such as 3G (third-generation mobile information technology), 4G (fourth-generation mobile information technology), and 5G (fifth-generation mobile information technology), but this invention does not limit this.

[0064] It's important to note that autonomous driving can be categorized into levels L0 to L5. Levels below L2, also known as Advanced Driving Assistance Systems (ADAS), only assist the driver in specific scenarios, with the driver retaining full control over driving decisions. Level L3 autonomous driving can replace human driving in certain situations, and requires driver intervention in case of system failure; the driver still retains full control over driving decisions. Level L4 and above involve vehicle-led driving. For Level L3 and above, the human-vehicle relationship is redefined, with the human transforming from the driver into the regulator of the autonomous vehicle.

[0065] like Figure 1 As shown, in this example, the intelligent driving controller 104 and the cloud server 106 are electrically connected via the vehicle network module 105 to communicate intelligent driving computing data and case models. As intelligent driving technology advances to higher levels, it requires the continuous accumulation of large amounts of perception, decision-making, and execution data, and the continuous iteration of algorithms through simulation models and algorithm training. Cloud services not only provide powerful storage and computing power support, but also connect multiple toolchains, providing a one-stop service capability for algorithm development and iteration. Furthermore, autonomous driving cloud services can selectively collect challenging scenarios from both the vehicle and cloud, forming challenging scenario datasets. Through large-scale concurrent virtual simulations of real-world scenarios in the cloud, autonomous driving algorithms can be effectively tested and verified, achieving efficient algorithm iteration and shortening the iteration cycle from weeks to days. Cloud server 106 can be a server cluster or distributed system composed of multiple physical servers. It can provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. There are no restrictions on this.

[0066] like Figure 1As shown, it should be noted that for higher levels of autonomous driving, the accuracy requirements of the algorithm models are higher, the computing power requirements of the autonomous driving controller 104 are also higher, and the amount of data involved is too large, resulting in a longer model iteration cycle and a higher difficulty in machine learning. Cloud services can help solve difficult model problems, but to achieve high-precision autonomous driving, the problem of excessively large amounts of computational data still exists. Furthermore, most of the sensor information acquired by the perception module 102 in the vehicle body 100 cannot be directly used as samples in the algorithm model. Therefore, the data utilization efficiency is low, the algorithm redundancy is high, which can easily affect the efficiency and accuracy of the algorithm model, the iteration cycle of the algorithm model, the machine learning time, and also the vehicle's reaction speed to current events during high-level autonomous driving.

[0067] The problems mentioned above are universally applicable to common travel scenarios. It can be seen that in intelligent driving, improper sample selection and processing can lead to various problems, especially in high-level intelligent driving environments and when difficult example models are required. To address these problems, embodiments of this application propose an intelligent driving data processing method and an intelligent driving data processing system, which will be described in detail below.

[0068] Please see Figure 2 , Figure 2 This is a flowchart illustrating a data processing method in an exemplary embodiment of this application. This method can be applied to... Figure 1 The implementation environment shown is specifically executed by the intelligent driving controller 104 within that implementation environment. It should be understood that this method can also be applied to other exemplary implementation environments and executed by devices in other implementation environments; this embodiment does not limit the implementation environment to which the method is applicable.

[0069] Figure 2 This is a flowchart illustrating a data processing method in an exemplary embodiment of this application. Figure 2 As shown, in an exemplary embodiment, the data processing method includes at least steps S210 to S250, which are described in detail below:

[0070] Step S210: Obtain the vehicle's raw data, which includes vehicle data, road data, and vehicle perception data.

[0071] Please see Figure 1 and Figure 2As shown, it's important to first clarify that the raw data refers to data that can be directly acquired by modules such as the vehicle's perception module 102, control module 101, execution module 103, and in-vehicle network module 105 during the intelligent driving process. Furthermore, the data acquired during the raw data acquisition process can be of various types, such as sound data, image data, light signal data, and numerical data. In this example, the acquired raw data can be accurately identified by the processing system for collection. For instance, traffic light signal information can be captured by a camera, and then image recognition technology can be used to convert the traffic light image into signal light data that can be directly recognized by the processing system. Similarly, the flashing light information of a vehicle ahead can be converted into flashing light data using signal conversion. Both the traffic light data and the flashing light data can then be collected by the perception module 102 as raw data.

[0072] Please see Figure 1 and Figure 2 As shown, the vehicle data includes the vehicle's speed, acceleration, and location information. Road data can specifically be high-precision map information. Vehicle perception data includes information about surrounding vehicles received by the perception module 102, such as the distance between the vehicle and surrounding vehicles, the relative position of the vehicle and surrounding vehicles, and the relative speed of the vehicle and surrounding vehicles. The relative position of the vehicle and surrounding vehicles includes the lateral and longitudinal positions of the surrounding vehicles relative to the vehicle.

[0073] Please see Figure 1 and Figure 2 As shown, in step S210, the receiving frequencies of the vehicle data, road data, and vehicle perception data are adjusted to be consistent. In the application environment of this example, execution information such as vehicle speed and acceleration in the vehicle data can be obtained by the execution module 103, and positioning information can be obtained through the vehicle's GPS positioning system. The positioning information of the vehicle body 100 can include the latitude and longitude of the vehicle body 100 to accurately locate its position. The vehicle data is updated in real time based on road conditions and travel objectives. As the vehicle data changes and the status of surrounding vehicles changes, the vehicle perception data is also updated in real time. The sampling frequency of road data is affected by factors such as road shape and length, resulting in a relatively long update frequency. Therefore, the transmission frequencies of the vehicle data, road data, and vehicle perception data are inconsistent. In the application environment of this example, the frequencies at which the intelligent driving controller 104 receives the vehicle data, road data, and vehicle perception data are adjusted to be consistent to improve the data processing efficiency of the intelligent driving controller 104 and reduce redundant data.

[0074] Please see Figure 1 and Figure 2As shown, in step S210, time synchronization is performed based on the absolute time within the vehicle body 100, and the high sampling frequency of the intelligent driving controller 104 is reduced to match the low frequency. Here, processing state information can use time-proximity values, and processing numerical information can use linear interpolation methods. For example, the sampling frequency of the vehicle's positioning information is, for example, 10ms / time to 20ms / time, specifically, 20ms / time. The sampling frequency of vehicle perception data is, for example, 50ms / time to 60ms / time, specifically, 50ms / time. The sampling frequency of road data is, for example, 90ms / time to 100ms / time, specifically, 100ms / time. Therefore, in step S220, the data sampling frequency of the intelligent driving controller 104 can be unified to, for example, 100ms / time or 200ms / time, etc.

[0075] Step S220: Set the sampling period according to the starting time of the vehicle, establish a driving coordinate system with the starting time as the origin, and establish the vehicle trajectory corresponding to the sampling period in the driving coordinate system.

[0076] Please see Figure 1 and Figure 2 As shown, in the application environment of this example, the intelligent driving controller 104 acquires raw data and then processes it to obtain the vehicle's driving trajectory. The collection period for the raw data can be T1, and the processing period for the raw data by the intelligent driving controller 104 can be T2. T1 ≤ T2, ensuring that the intelligent driving processor 104 can form an effective sampling period when processing the raw data. When the intelligent driving controller 104 analyzes the data, the starting time is the current time of the vehicle. The sampling period includes the starting time and takes continuous time periods before and after the starting time. For example, if the vehicle's current time corresponds to 0 seconds, the sampling period is set to -10s to 0s. In step S220, the set sampling period can be T0, where T1 ≤ T2 ≤ T0.

[0077] Figure 3 This is a schematic diagram illustrating the trajectory of a vehicle during a sampling period, as shown in an exemplary embodiment of this application. Figure 3 As shown, T0 is the sampling period, t1 is the first moment of the sampling period, t2 is the second moment of the sampling period, and 0 is the starting moment, which corresponds to the current position of the vehicle. Where T0 = t2 - t1. The moments t2 and 0 can coincide. In this embodiment, t2 is, for example, moment 0, T0 is, for example, 10 seconds, and t1 is, for example, -10 seconds. In step S220, a driving coordinate system is established, and the vehicle's trajectory is formed in the driving coordinate system based on the original data. In step S220, the established driving coordinate system can refer to the vehicle's coordinate system.

[0078] It is important to note that in this embodiment, the vehicle is used as the origin of the coordinate system, which could be the vehicle's center or center of gravity, etc. The vehicle's direction of travel is taken as the positive direction of the driving coordinate system, specifically the direction of travel at the starting moment as the x-axis. The xy-plane can be a plane parallel to the vehicle's chassis. In this embodiment, the vehicle's route and its relative position to obstacles can be analyzed using a two-dimensional xy-plane coordinate system. In other embodiments of the invention, a three-dimensional coordinate system can be formed using a direction perpendicular to the xy-plane as the z-axis. This allows for the analysis of road gradient and vehicle speed changes, as well as timely analysis of potential collisions between the vehicle and other traffic participants. The driving coordinate system is always matched to the vehicle. By considering the sampling period and the sampling time steps of multiple synchronous sampling data, the accuracy of the sampling data is improved, and the redundancy of the sampling data is reduced. During the driving journey, relevant data on traffic participants can be continuously acquired and analyzed, which is beneficial for selecting samples that are needed by the algorithm model and can be directly used, meeting various computing power requirements in the intelligent driving process, and reducing the algorithm's iteration cycle.

[0079] like Figure 3 As shown, in one embodiment of the present invention, the sampling period is T0 and the sampling period is T1. Then, within the sampling period, for example, (T0 / T1+1) sampling time steps are set. In this embodiment, T1 is, for example, 100ms, and T0 is, for example, 10s. Therefore, in this embodiment, there are, for example, 101 sampling time steps. Figure 4 This is a flowchart for generating the vehicle's driving trajectory. For example... Figure 4 As shown, the steps for obtaining the trajectory of the vehicle body 100 include steps S410 to S430.

[0080] Step S410: Determine whether the location information in the vehicle data is reliable.

[0081] like Figures 1 to 4As shown, in one embodiment of the present invention, the vehicle data includes the positioning information of the vehicle body 100. In the application environment of this example, the positioning information can be obtained through the Global Positioning System (GPS). Therefore, in step S410, the reliability of the positioning information is determined based on the value of the Position Dilution of Precision (PDOP). The smaller the PDOP, the higher the confidence level of the positioning information. Therefore, when the PDOP is less than the confidence threshold, the positioning information of the vehicle is considered reliable. When the PDOP is greater than the confidence threshold, the positioning information of the vehicle is considered reliable. The confidence threshold is determined based on the PDOP data. In this embodiment, the confidence threshold can be, for example, 3 to 5, specifically, 3. That is, when the value of the PDOP is less than, for example, 3, step S420 is executed. When the value of the PDOP is greater than or equal to, for example, 3, step S430 is executed.

[0082] Step S420: Based on the vehicle's positioning information, form the vehicle's trajectory in the driving coordinate system.

[0083] like Figures 1 to 4 As shown, in one embodiment of the present invention, the positioning information includes the position of the vehicle body 100 during the sampling period. In this embodiment, the position of the vehicle body 100 at each sampling time step is obtained to form the vehicle trajectory of the vehicle body 100. The position of the vehicle body 100 corresponding to each sampling time step can be represented by the latitude and longitude of the vehicle body 100, or by the lateral distance x and longitudinal distance y between the vehicle body 100 at the corresponding sampling time step and the vehicle body 100 at time 0. Figure 5 This is a coordinate diagram illustrating the construction of a vehicle coordinate system in an exemplary application scenario. For example... Figure 5 As shown, in the driving coordinate system, the driving coordinate system and the actual scale can be modeled one-to-one. Taking time t1 as an example, the lateral distance x1 between the vehicle at time t1 and the vehicle at time 0, and the longitudinal distance y1 between the vehicle at time t1 and the vehicle at time 0 are obtained based on the positioning information, thus obtaining the vehicle position at time t1. This process is repeated to obtain the vehicle position at each sampling time step, forming the vehicle trajectory in the driving coordinate system.

[0084] Step S430: Based on the vehicle speed and yaw rate, obtain the position of the vehicle during the sampling period and form the vehicle trajectory.

[0085] like Figures 1 to 5 As shown, in one embodiment of the present invention, when the confidence level of the positioning information is not high, the vehicle body 100 is calculated at each moment of the sampling period based on the vehicle speed and yaw rate of the vehicle data in each sampling moment. Figure 5 As shown, t11 is the next time after t1, and the time interval between t11 and t1 is one sampling period T1. Taking t1 as an example, given the current position of the vehicle body 100, based on the vehicle speed and yaw rate at t1 and t11, the relative positions of the vehicle body 100 at t11 ​​and t1, i.e., the lateral and longitudinal distances, can be deduced. Similarly, given the actual position of the vehicle body 100 at time 0, the trajectory of the vehicle body 100 can be obtained in the driving coordinate system.

[0086] Step S230: Obtain the location information of traffic participants around the vehicle during the sampling period, and form the vehicle trajectory of the surrounding traffic participants based on the location information.

[0087] like Figure 1 and Figure 2 As shown, in one embodiment of the present invention, the sensing module 102 can obtain the distance between the vehicle and surrounding obstacles, as well as the type of obstacles. Specifically, the sensing module 102 obtains images of the vehicle's surroundings at each sampling time step within the sampling period using a camera. Specifically, step S230 includes steps S610 to S630.

[0088] Step S610: Obtain images of obstacles around the vehicle at each sampling time step during the sampling period, and analyze the obstacle images to obtain the obstacle type and number of obstacles.

[0089] like Figure 1 , Figure 5 and Figure 6 As shown, in one embodiment of the present invention, the perception module 102 can acquire environmental images around the vehicle body 100 through a camera component. Specifically, during the sampling period, each sampling time step corresponds to the intelligent driving controller 104 receiving raw data once, and the raw data includes vehicle perception data. In step S610, the vehicle perception data includes environmental images of the vehicle body 100. By analyzing the surrounding vehicle environment images, the number and type of obstacles around the vehicle body 100 can be obtained. For example, obstacles can be motor vehicles, pedestrians, bicycles, and inanimate objects, etc. Among them, obstacles are traffic participants. In step S610, the vehicle perception data also includes the distance between the vehicle body 100 and the obstacles, as well as the angle between the obstacles and the vehicle body 100. Among them, the angle between the obstacles and the vehicle body 100 can be the angle between the line connecting the center of the obstacle and the center of the vehicle body 100 in the horizontal plane and the central axis of the vehicle body 100. To identify different obstacles, the obstacles can also be numbered.

[0090] Step S620: In the driving coordinate system, obtain the obstacle position at each sampling time step within the sampling period.

[0091] like Figure 1 , Figure 5 and Figure 6 As shown, in one embodiment of the present invention, in step S620, at one sampling time step, the relative position of the vehicle body 10 and the obstacle can be represented by the lateral and longitudinal distances between the obstacle and the vehicle body 100. Based on the vehicle perception data, after obtaining the distance between the vehicle body 100 and the obstacle, as well as the angle between the vehicle body 100 and the obstacle, the lateral and longitudinal distances between the vehicle body 100 and the obstacle are calculated. This forms an image of the obstacle in the driving coordinate system.

[0092] Step S630: In the driving coordinate system, obtain the circumferential trajectory of the obstacle during the sampling period.

[0093] like Figure 1 , Figure 5 and Figure 6 As shown, in one embodiment of the present invention, by combining the position of the obstacle in the driving coordinate system at each sampling time step, the circumferential trajectory of the vehicle body 100 at the starting time during the sampling period can be obtained. It should be noted that the sensing range of the vehicle body 100 is limited. Therefore, in this embodiment, the type of obstacle at each sampling time step can be different, the number of obstacles at each sampling time step can be different, and the same obstacle can appear in different sampling time steps in a skipping manner. During the sampling period, the circumferential trajectory of the same obstacle may only appear in a portion of the sampling time steps. For example, the trajectory of vehicle body number one surrounding vehicle body 100 during a sampling period of, for example, -10s to 0, may be in the interval of, for example, -5s to -2s, or in the intervals of, for example, -10s to -8s and, for example, -5s to -2s, or even within the sampling period of, for example, -10s to 0. The present invention does not limit this.

[0094] Step S240: Based on the road data, form the lane and vehicle centerline in the driving coordinate system.

[0095] like Figure 1 , Figure 2 and Figure 5As shown, in one embodiment of the present invention, the original data includes road data. For L3 and above autonomous driving levels, GPS positioning information is insufficient to meet the accuracy requirements of autonomous driving. Therefore, in this embodiment, the road data includes high-precision map information. A high-precision map (High Definition Map, HD Map, or Highly Automated Driving Map, HAD Map) refers to a high-precision, high-freshness, and high-rich electronic map with both absolute and relative accuracy within 1 meter. In this embodiment, the high-precision map information includes road information such as road type, road curvature, and lane line positions, as well as environmental object information such as roadside infrastructure, obstacles, and traffic signs, and real-time dynamic information such as traffic flow and traffic light status. In this embodiment, the high-precision map information is combined with the driving coordinate system. Specifically, step S240 includes steps S710 to S740.

[0096] Step S710: Obtain a high-precision map of the road data, and stitch the vehicle centerline and lane boundary lines according to the vehicle's driving direction to form a stitched lane.

[0097] Figure 7 yes Figure 2 The flowchart of step S240 in the illustrated embodiment is shown in an exemplary embodiment. Figure 8 This is a schematic diagram illustrating the stitching of a high-precision map during a sampling period, as shown in an exemplary embodiment of this application. Figure 1 , Figure 7 and Figure 8 As shown, in one embodiment of the present invention, the high-precision map includes lane data, wherein the lane data includes lane position, lane shape, lane curvature, etc. Figure 8 The lane shape shown is a straight road, but this invention is not limited to straight roads; it can also be a lane with curvature. In this embodiment, during the sampling period, the high-precision maps collected at each sampling time step are stitched together according to the time sequence of the sampling time steps. Specifically, the lane shape is stitched together according to the vehicle centerline and lane boundary lines based on the sampling time steps.

[0098] like Figure 1 , Figure 7 and Figure 8As shown, in one embodiment of the present invention, in step S710, when stitching lanes, the segment where the vehicle is located at the starting time is taken as the starting segment. The segment where the vehicle is located at the starting time is considered positive along the driving direction and negative against the driving direction. Specifically, in the driving coordinate system, roads can be stitched together based on road data, such as segmented discrete map information in high-precision map information and the road connection relationships in each segment of map information. Specifically, the endpoints of the vehicle centerlines of two road segments in adjacent sampling time steps are aligned, and the endpoints of the lane boundary lines are aligned. Specifically, the segment where the vehicle is located at the starting time can be obtained first, then the high-precision map information of the most recent time period can be obtained and stitched together to form partial road information. This process can be repeated by advancing along the timeline and sequentially obtaining high-precision map information, and then stitching together the road information for the sampling time period.

[0099] Step S720: Based on the road data, establish the vehicle centerline and vehicle reference line in the driving coordinate system.

[0100] like Figure 1 , Figure 7 and Figure 8 As shown, in one embodiment of the present invention, the vehicle body centerline can be a line connecting the centers of the vehicle body 100 at each sampling time step. The center of the vehicle body 100 can be preset according to the vehicle model, for example, set as the chassis center of gravity, chassis center, geometric center, center of gravity, and the position of key components of the vehicle body 100, etc., and the present invention does not limit this. The vehicle body centerline can be obtained in a three-dimensional coordinate system using the vehicle's positioning information. Due to differences in sampling data, the interval of the vehicle reference lines is variable. In this embodiment, the point set spacing of the vehicle reference lines can be corrected to less than or equal to, for example, 0.2m using interpolation encryption.

[0101] Step S730: Number the different lanes and obtain the relative positional relationship of the different lanes.

[0102] like Figure 1 , Figure 7 and Figure 8 As shown, in one embodiment of the present invention, the lane where the vehicle body 100 is located at the starting time is used as the reference lane. In the driving coordinate system, one side of the reference lane can be designated as a first number, and the other side of the reference lane can be designated as a second number. For example, the reference lane can be designated as number 0. Along the driving direction of the vehicle body 100, the left side of the driving direction is designated as the left lane, and the right side of the driving direction is designated as the right lane. For example, the first left lane is designated as 1, the first right lane is designated as -1, the second left lane on the left is designated as, for example, 2, and the second right lane on the right is designated as, for example, -2.

[0103] like Figure 1 , Figure 7 and Figure 8 As shown, in one embodiment of the present invention, during the sampling period, in the driving coordinate system, based on the established road information, a first reserved distance, for example, 50m to 100m, can be obtained at the first moment in the opposite direction of travel. In this embodiment, during the sampling period from -10s to 0s, road data for, for example, 100m before -10s can be obtained, and the road data can be converted into a point set to form a lane in the driving coordinate system. At the second moment, a second reserved distance, for example, 250m to 300m, is obtained along the direction of travel. In this embodiment, during the sampling period from -10s to 0s, road data for, for example, 300m after 0s can be obtained, and the road data can be formed in the driving coordinate system.

[0104] Step S740: Obtain the point set data of lane boundary lines and vehicle center lines.

[0105] like Figure 1 , Figure 7 and Figure 8 As shown, in one embodiment of the present invention, a lane is formed based on the relative positional relationship with the starting segment at the starting time, and the precision of the vehicle centerline and lane boundary lines is improved by numerical encryption, thus enriching the point set data of the vehicle centerline and lane boundary lines. In this embodiment, for example, a point set A for the vehicle centerline and a point set B for the lane boundary lines are formed, and the coordinate data corresponding to point set A and point set B are collected for use in the calculation and retrieval of the algorithm model during intelligent driving.

[0106] Step S250: Based on the graphics of the lane, the vehicle trajectory, and the surrounding vehicle trajectory in the driving coordinate system, obtain the coordinate point set of the lane, the vehicle trajectory, and the surrounding vehicle trajectory.

[0107] like Figure 1 and Figure 2 As shown, in one embodiment of the present invention, the lane, vehicle trajectory, and surrounding vehicle trajectory formed in steps S210 to S240 can be lines connecting multiple coordinate points. The multiple coordinate points corresponding to the vehicle trajectory can be the location of a point on the vehicle, such as the spatial geometric center of the vehicle body 100, the center of the chassis of the vehicle body 100, and the center of gravity of the vehicle body 100, etc. After determining the coordinate points of the vehicle trajectory, the set of coordinate points corresponding to the outer shell of the vehicle body 100 can be obtained in the driving coordinate system based on the vehicle model data of the vehicle body 100. The coordinate points of the surrounding vehicle trajectory can be the point closest to the obstacle and the vehicle body 100.

[0108] like Figure 1 and Figure 2As shown, in one embodiment of the present invention, in step S250, during the sampling period, the lane, vehicle trajectory, and surrounding vehicle trajectory for each sampling time step are acquired, as well as the coordinate point set of the vehicle body 100. The coordinate point set includes the x-coordinate x of the traffic participant at time t. t y coordinates t Longitudinal velocity Lateral velocity Trajectory motion state information X t In this embodiment, trajectory motion state information can be represented by an array, specifically...

[0109] like Figure 1 and Figure 2 As shown, in one embodiment of the present invention, in step S250, the coordinate point set includes information on the i-th vehicle centerline, the i-th road left boundary line, and the i-th road right boundary line within the sampling time period. During a single trip, roads can be numbered according to the order in which the vehicle 100 enters, to distinguish the coordinate data corresponding to each road. Here, i is a natural number, and the present invention does not limit the specific value of i. In this embodiment, the information on the i-th vehicle centerline, the left boundary line, and the right boundary line are all stored in the form of a point set. Specifically, the information on the i-th vehicle centerline includes, for example, coordinate point data of the i-th vehicle centerline in the driving coordinate system, stitched together based on high-precision map information within a period from -10s to 0s, and is denoted by L. i The information of the left boundary line of the i-th road includes, for example, information from -10s to 0s, stitched together based on high-precision map information, and stores the set of coordinate points of the left lane boundary line in the driving coordinate system as a point set. The information of the left boundary line of the i-th road can be represented by Ll. i The information of the right boundary line of the i-th road includes, for example, information from -10s to 0s, stitched together based on high-precision map information, and stores the set of coordinate points of the right lane boundary line in the driving coordinate system as a point set. The information of the right boundary line of the i-th road can be represented by Lr. i In this embodiment, the set of coordinate points can be Z, and Z = (H, L) i ,Ll i Lr i ), where H represents the trajectory motion state information during the sampling period.

[0110] Step S260: Set filtering conditions, filter out training samples from the set of coordinate points, and store the training samples.

[0111] like Figure 1 and Figure 2As shown, in one embodiment of the present invention, the coordinate point set obtained during the sampling period can lead to wasted computing power and unbalanced distribution of computing power during the algorithm model's calculation process. This can cause difficult and urgently needed cases to be delayed by unnecessary data, affecting the efficiency of the algorithm model. Specifically, the coordinate point set obtained in steps S210 to S250 contains redundant data. If there are no roads ahead on the current map, or if the roads are invalid, much of the data in the coordinate point set, although collected, may not be used in the algorithm model. Therefore, the filtering conditions can be set according to the needs of the algorithm model.

[0112] like Figure 1 and Figure 2 As shown, in one embodiment of the present invention, the filtering condition can be set to whether there is map information within a first reserved distance behind the vehicle body 100 at the first moment and a second reserved distance in front of the vehicle body 100 at the second moment during the sampling period. For example, whether there is map information within a range of 100m behind the vehicle body 100 at -10s, and whether there is map information within a range of 300m in front of the vehicle body 100 at 0s. If there is no matching map information, the acquired data is invalid, and this part of the data is removed when filtering training samples. If there is map information within the corresponding map range, the coordinate point set corresponding to the sampling period can be saved as training samples. In another embodiment of the present invention, the filtering condition can also be set to determine whether the lane curvature radius is greater than a preset threshold. The preset threshold is, for example, 30-50m, specifically 50m. When the lane curvature radius is greater than, for example, 50m, the corresponding coordinate point set is retained as training samples. In another embodiment of the present invention, the filtering condition can also be set to determine whether the distance to the travel target is within an effective range. The travel target can be a preset target before the vehicle travels, or a target modified during the vehicle's journey. In this embodiment, the effective range can be a preset judgment range, such as 500m to 700m. When the distance to the travel target is, for example, within 500m, the corresponding set of coordinate points is retained as training samples. The filtering process targets lane information.

[0113] like Figure 1 and Figure 2As shown, in one embodiment of the present invention, after the training samples are screened, the samples can be stored in the intelligent driving controller 104 and directly retrieved for calculation when needed by the algorithm model. The present invention does not limit the algorithm model; it can be a positioning model, trajectory model, etc. When storing samples, the sampling period includes multiple sampling time steps, each containing trajectory motion state information. For example, within -10s to 0s, the trajectory information interval is, for example, 100ms, resulting in, for example, 101 sampling time steps. For example, after screening, 10 lane information entries are stored. Then, in step S260, the stored training samples can be Z = (H0, H1, ..., H...). 100 L i1 ,Ll i1 Lr i1 ···L i10 ,Ll i10 Lr i10 ).

[0114] Figure 9 This is a block diagram illustrating a data processing system as shown in an exemplary embodiment of this application. The present invention also provides a data processing system for intelligent driving. Figure 9 As shown, the data processing system 900 includes a data collection module 901, a vehicle trajectory generation module 902, a surrounding vehicle trajectory generation module 903, a road data processing module 904, a coordinate point set acquisition module 905, and a sample filtering module 906. The data collection module 901 acquires the vehicle's raw data and synchronizes the sampling frequencies of various raw data sources, including vehicle data, road data, and vehicle perception data. The vehicle trajectory generation module 902 sets the sampling period based on the vehicle's starting time, establishes a driving coordinate system with the starting time as the origin, and creates the vehicle trajectory corresponding to the sampling period within the driving coordinate system. The surrounding vehicle trajectory generation module 903 acquires the location information of surrounding traffic participants within the sampling period and, based on this location information, generates the surrounding vehicle trajectories. The road data processing module 904 processes the road data, forming lanes and vehicle centerlines in the driving coordinate system. The coordinate point set acquisition module 905 is used to acquire the coordinate point sets of the lane, the vehicle trajectory, and the surrounding vehicle trajectory based on the graphics of the lane, the vehicle trajectory, and the surrounding vehicle trajectory in the driving coordinate system. The sample filtering module 906 is used to set filtering conditions, filter training samples from the coordinate point set, and store the training samples.

[0115] It should be noted that the data processing system and the data processing method provided in the above embodiments belong to the same concept. The specific ways in which each module and unit performs operations have been described in detail in the method embodiments and will not be repeated here. In practical applications, the traffic condition refresh device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0116] Embodiments of this application also provide an electronic device, including a memory 101 and one or more processors 103. The memory stores one or more computer instructions 102, which, when executed by the one or more processors 103, cause the electronic device to implement the data processing methods provided in the various embodiments described above. The computer instructions may be stored in a computer-readable storage medium.

[0117] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. A computer program contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0118] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0119] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A data processing method for intelligent driving, characterized in that, include: Simultaneously acquire various raw data from the vehicle, including vehicle data, road data, and vehicle perception data; The sampling period is set, which includes the starting time of the vehicle and takes continuous time periods before and after the starting time. A driving coordinate system is established with the starting time as the origin, and the vehicle trajectory for the sampling period is formed in the driving coordinate system based on the original data. The location information of traffic participants around the vehicle during the sampling period is obtained, and the vehicle trajectory around the traffic participants is formed based on the location information. Process the road data to form lanes and vehicle centerlines in the driving coordinate system; Based on the graphs of the lane, the vehicle trajectory, and the surrounding vehicle trajectory in the driving coordinate system, obtain the coordinate point set of the lane, the vehicle trajectory, and the surrounding vehicle trajectory; and Set filtering conditions and filter training samples from the set of coordinate points, and store the training samples; The filtering criteria include at least one of the following: If map information exists within the first reserved distance behind the vehicle body at the first moment and within the second reserved distance in front of the vehicle body at the second moment during the sampling period, then the set of coordinate points corresponding to the sampling period is saved as training samples. If the radius of curvature of the lane is greater than a preset threshold, the corresponding set of coordinate points will be retained as training samples. If the distance of the vehicle's travel target is within the valid range, then the corresponding set of coordinate points is retained as training samples.

2. The data processing method for intelligent driving according to claim 1, characterized in that, The step of synchronously acquiring the raw data includes: synchronizing the sampling frequencies of multiple raw data based on the lowest sampling frequency among the vehicle data, the road data, and the vehicle perception data.

3. The data processing method for intelligent driving according to claim 2, characterized in that, The steps for setting the sampling period include: Obtain the continuous time period including the stated start time; and Based on the synchronized sampling frequency, multiple sampling time steps are set within the continuous time period to form the sampling period.

4. The data processing method for intelligent driving according to claim 3, characterized in that, The steps for obtaining the vehicle trajectory include: Determine whether the vehicle's location information in the vehicle data is reliable; If the location information is reliable, then the vehicle trajectory is formed in the driving coordinate system based on the location information; and If the positioning information is unreliable, the vehicle's position at the sampling time step is obtained based on the vehicle speed and yaw rate in the vehicle data, thus forming the vehicle trajectory.

5. The data processing method for intelligent driving according to claim 3, characterized in that, The steps for obtaining the circumferential vehicle trajectory include: Obtain images of obstacles around the vehicle at the sampling time step, and determine the obstacle type based on the obstacle images; In the driving coordinate system, obtain the obstacle position at each sampling time step within the sampling period; and In the driving coordinate system, the circumferential trajectory of the obstacle is formed during the sampling period based on the obstacle type and the obstacle position.

6. The data processing method for intelligent driving according to claim 3, characterized in that, The steps for processing the road data include: Acquire high-precision map information within the sampling period, the high-precision map information including multiple discrete map segments; and Based on the sampling time step, the discrete map is stitched together according to the center line and boundary line of the lane to form the lane.

7. The data processing method for intelligent driving according to claim 1, characterized in that, The steps of forming the lane also include: Multiple lanes are divided according to the vehicle's centerline, and different lanes are numbered; and The relative positions between different lanes are distinguished based on the lane number.

8. The data processing method for intelligent driving according to claim 1, characterized in that, The steps for obtaining the vehicle body centerline include: Based on the vehicle model, the center of the vehicle body is preset; and The vehicle centerline is formed based on the center of the vehicle body and the vehicle trajectory.

9. A data processing system for intelligent driving, characterized in that, include: The data collection module is used to acquire raw data of the vehicle, including vehicle data, road data, and vehicle perception data. The vehicle trajectory generation module is used to set a sampling period, which includes the starting time of the vehicle and takes a continuous time period before and after the starting time. Then, a driving coordinate system is established with the starting time as the origin, and the vehicle trajectory of the sampling period is formed in the driving coordinate system according to the original data. The vehicle trajectory generation module is used to obtain the location information of traffic participants around the vehicle during the sampling period, and to form the vehicle trajectory of the surrounding traffic participants based on the location information. The road data processing module is used to process the road data and form lanes and vehicle centerlines in the driving coordinate system; The coordinate point set acquisition module is used to acquire the coordinate point set of the lane, the vehicle trajectory, and the surrounding vehicle trajectory based on the graphics of the lane, the vehicle trajectory, and the surrounding vehicle trajectory in the driving coordinate system; as well as A sample filtering module is used to set filtering conditions, filter training samples from the coordinate point set, and store the training samples; wherein, the filtering conditions include at least one of the following: if map information exists within a first reserved distance behind the vehicle body at the first moment and within a second reserved distance in front of the vehicle body at the second moment during the sampling period, then the coordinate point set corresponding to the sampling period is saved as a training sample; if the lane curvature radius is greater than a preset threshold, then the corresponding coordinate point set is retained as a training sample. If the distance of the vehicle's travel target is within the valid range, then the corresponding set of coordinate points is retained as training samples.

10. An electronic device, characterized in that, The electronic device includes: A memory, wherein computer instructions are stored; and A processor that, when executing the computer instructions, implements a data processing method for intelligent driving as described in any one of claims 1 to 8.

Images