Vehicle control method, vehicle and storage medium

Abstract

The embodiment of the invention provides a vehicle control method, a vehicle and a storage medium. The vehicle control method comprises the steps that unmanned aerial vehicle sensing data and vehicle sensing data are acquired, the unmanned aerial vehicle sensing data are used for representing road environment information outside a target vehicle sight distance sensing range, and the vehicle sensing data are used for representing road environment information within the target vehicle sight distance sensing range; the unmanned aerial vehicle sensing data and the vehicle sensing data are fused, a target fusion result is obtained, and the target fusion result is used for constructing a real-time three-dimensional environment map corresponding to the target vehicle; using the target fusion result to make a driving decision on the target vehicle to obtain a driving decision result, the driving decision result being used for representing a driving strategy to be used by the target vehicle; and controlling the target vehicle based on the driving decision result. The technical problems of insufficient accuracy and low safety of a remote vehicle control strategy caused by a sensing blind area in related technologies are solved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of vehicle control technology, and in particular, to a vehicle control method, a vehicle, and a storage medium. Background Art

[0002] Intelligent connected vehicle technology has flourished in recent years, especially in the field of remote vehicle control. Vehicles rely on their own sensors such as cameras and radars to perceive the environment in order to achieve safe navigation and efficient operation.

[0003] However, in complex or beyond-visual-range scenarios, such as off-road terrain, tunnels, and deep in forests, traditional sensors are limited by their field of view and environmental factors, making it difficult to obtain comprehensive and accurate road condition information. This results in significant blind spots, impacting the accuracy and safety of driving system decisions. Existing vehicle sensors only cover a limited visual range, making it impossible to promptly capture information about obstacles and terrain changes beyond the vehicle's direct perception range. This hinders the accuracy and safety of remote vehicle control strategies. Therefore, improving the accuracy and safety of remote vehicle control strategies remains a key technical challenge in this area.

[0004] There is currently no good solution to the above problems. Summary of the Invention

[0005] The embodiments of the present application provide a vehicle control method, a vehicle, and a storage medium to at least solve the technical problems in the related art of insufficient accuracy and low safety of remote vehicle control strategies due to perception blind spots.

[0006] According to one aspect of an embodiment of the present application, a vehicle control method is provided, including: acquiring drone perception data and vehicle perception data, wherein the drone perception data is used to represent road environment information outside the line of sight perception range of a target vehicle, and the vehicle perception data is used to represent road environment information within the line of sight perception range of the target vehicle; fusing the drone perception data and the vehicle perception data to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle; making driving decisions for the target vehicle using the target fusion result to obtain a driving decision result, wherein the driving decision result is used to represent the driving strategy to be used by the target vehicle; and controlling the target vehicle based on the driving decision result.

[0007] Optionally, obtaining drone perception data includes: obtaining online map data of the Internet of Vehicles, wherein the online map data of the Internet of Vehicles is used to represent the real-time road conditions corresponding to the target vehicle; in response to the online map data of the Internet of Vehicles meeting a preset collection trigger condition, performing planning analysis based on the navigation route information and real-time road condition information of the target vehicle to obtain a planning analysis result, wherein the planning analysis result is used to determine the take-off height, circling radius and scanning path corresponding to the vehicle-mounted drone; obtaining drone scanning data based on the planning analysis result; and performing feature extraction on the drone scanning data to obtain drone perception data.

[0008] Optionally, obtaining drone perception data includes: in response to receiving a data collection instruction from a target user, obtaining drone scanning data according to the data collection instruction, wherein the data collection instruction includes: the take-off height, hovering radius and scanning path corresponding to the vehicle-mounted drone; performing feature extraction on the drone scanning data to obtain drone perception data.

[0009] Optionally, the vehicle control method also includes: obtaining priority information of the drone scanning data, wherein the priority information is used to determine the data urgency corresponding to the drone scanning data; determining a target transmission link based on the priority information; and using the target transmission link to transmit the drone scanning data between the target vehicle and the on-board drone.

[0010] Optionally, determining the target transmission link based on the priority information includes: in response to determining that the data urgency is greater than or equal to a preset threshold based on the priority information, determining the main transmission link as the target transmission link; in response to determining that the data urgency is less than a preset threshold based on the priority information, determining the auxiliary transmission link as the target transmission link, wherein the transmission delay of the auxiliary transmission link is greater than the transmission delay of the main transmission link.

[0011] Optionally, the drone perception data and the vehicle perception data are fused to obtain a target fusion result, including: fusing the drone perception data and the vehicle perception data to obtain an initial fusion result; constructing a real-time three-dimensional environment map based on the initial fusion result; and performing semantic segmentation on the real-time three-dimensional environment map to obtain a target fusion result.

[0012] Optionally, the vehicle control method also includes: uploading the target fusion result to a cloud server, so that the cloud server performs data conversion based on the target fusion result to obtain target perspective map data; receiving the target perspective map data issued by the cloud server; using the target perspective map data to make driving decisions and obtain driving decision results.

[0013] Optionally, the vehicle control method further includes: updating the original navigation path of the target vehicle based on the driving decision result to obtain an updated navigation path; and displaying the updated navigation path on an augmented reality head-up display interface of the target vehicle.

[0014] According to another aspect of an embodiment of the present application, a vehicle control device is also provided, including: an acquisition module for acquiring drone perception data and vehicle perception data, wherein the drone perception data is used to represent road environment information outside the line of sight perception range of the target vehicle, and the vehicle perception data is used to represent road environment information within the line of sight perception range of the target vehicle; a fusion module for fusing the drone perception data and the vehicle perception data to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle; a decision module for making driving decisions for the target vehicle using the target fusion result to obtain a driving decision result, wherein the driving decision result is used to represent the driving strategy to be used by the target vehicle; and a control module for controlling the target vehicle based on the driving decision result.

[0015] According to another aspect of an embodiment of the present application, a vehicle is further provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the vehicle control method in each embodiment of the present application is executed when the program is running.

[0016] According to another aspect of an embodiment of the present application, a computer-readable storage medium is also provided, which includes a stored executable program, wherein when the executable program is running, the device where the readable storage medium is located is controlled to execute the vehicle control method in each embodiment of the present application.

[0017] According to another aspect of the embodiments of the present application, a computer program product is also provided, including a computer program, which implements the vehicle control method in each embodiment of the present application when executed by a processor.

[0018] According to another aspect of an embodiment of the present application, a computer program product is also provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the vehicle control method in each embodiment of the present application is implemented.

[0019] According to another aspect of the embodiments of the present application, a computer program is also provided. When the computer program is executed by a processor, the vehicle control method in each embodiment of the present application is implemented.

[0020] In the embodiment of the present application, firstly, the unmanned aerial vehicle perception data and the vehicle perception data are acquired, wherein the unmanned aerial vehicle perception data is used to represent the road environment information outside the target vehicle visual range perception range, and the vehicle perception data is used to represent the road environment information within the target vehicle visual range perception range; secondly, the unmanned aerial vehicle perception data and the vehicle perception data are fused to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle; further, the target fusion result is used to make a driving decision for the target vehicle to obtain a driving decision result, wherein the driving decision result is used to represent a driving strategy to be used by the target vehicle; finally, the target vehicle is controlled based on the driving decision result. The present application firstly acquires the unmanned aerial vehicle perception data and the vehicle perception data, ensuring that the road environment information within and outside the target vehicle visual range perception range is captured, the unmanned aerial vehicle perception data effectively fills the blind area caused by the limited field of view of the traditional vehicle sensor, and the environmental perception range of the vehicle is expanded; then, the unmanned aerial vehicle perception data and the vehicle perception data are fused to generate a target fusion result, which is used to construct a real-time three-dimensional environment map around the target vehicle, the fusion processing can integrate data from different sources, overcome the limitations of single sensor data, provide more complete and accurate road condition information, improve the accuracy and comprehensiveness of environmental perception, and provide a solid data foundation for subsequent decision-making; further, the target fusion result is used to make a driving decision, and the driving decision result can more accurately predict obstacles, terrain changes, etc., timely adjust the driving route and speed, optimize the driving strategy, and improve the decision-making ability and safety of the vehicle in complex environments; finally, the vehicle is controlled based on the driving decision result, ensuring that the vehicle safely and efficiently travels according to the optimized driving strategy. The above vehicle control process relies on accurate driving decision results, can effectively avoid safety hazards caused by the perception blind area, and improves the reliability of remote vehicle control, thereby improving the driving safety. Therefore, the present application can achieve the technical effect of improving the accuracy and safety of the remote vehicle control strategy, and further solve the technical problems of insufficient accuracy and low safety of the remote vehicle control strategy caused by the perception blind area in the related art. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings, which are included to provide a further understanding of the present application and are incorporated in and constitute a part of this application, illustrate embodiments of the present application and serve to explain the present application, and do not constitute improper limitations on the present application. In the drawings:

[0022] Figure 1 is a flowchart of a vehicle control method according to an embodiment of the present application;

[0023] Figure 2 is a structural block diagram of a vehicle control device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] According to an embodiment of the present application, a method embodiment of a vehicle control method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] In this embodiment, a vehicle control method is provided. Figure 1 is a flow chart of a vehicle control method according to an embodiment of the present application, such as Figure 1 As shown, the process includes the following steps:

[0028] Step S11, acquiring drone perception data and vehicle perception data, wherein the drone perception data is used to represent road environment information outside the line of sight perception range of the target vehicle, and the vehicle perception data is used to represent road environment information within the line of sight perception range of the target vehicle;

[0029] Step S12: fusing the drone perception data and the vehicle perception data to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle;

[0030] Step S13, using the target fusion result to make a driving decision for the target vehicle to obtain a driving decision result, wherein the driving decision result is used to represent the driving strategy to be used by the target vehicle;

[0031] Step S14: Control the target vehicle based on the driving decision result.

[0032] Optionally, the drone perception data obtained above refers to road environment information outside the line of sight perception range of the target vehicle obtained through binocular vision (4K camera), lidar, millimeter wave radar and other sensors carried by the vehicle-mounted drone.

[0033] For example, in an environment where the Global Navigation Satellite System (GNSS) cannot receive GNSS signals (such as tunnels and forests), drones operate in the air and capture road obstacles, traffic conditions, terrain features, etc. through visual-inertial odometry (VIO) and point-line feature fusion algorithms in a GNSS-denied environment, achieving centimeter-level positioning and transmitting terrain elevation and obstacle distribution data in real time.

[0034] Drone perception data includes but is not limited to RGB-D images (a special image format that combines RGB images and depth information), LiDAR point clouds (three-dimensional data sets generated by laser radar sensors), millimeter-wave radar echoes, etc., providing key information for building a detailed map of the vehicle's surrounding environment.

[0035] Vehicle perception data refers to information about the road environment within the vehicle's line of sight, acquired using onboard sensors (such as cameras, ultrasonic radar, and infrared sensors). These sensors monitor obstacles, traffic signs, and road contours ahead in real time, providing a direct basis for vehicle behavior decisions within the vehicle's line of sight.

[0036] In an optional embodiment, the acquired drone perception data is fused with the vehicle perception data via an onboard edge computing unit to produce a target fusion result. The fusion process involves data alignment, feature extraction, and semantic segmentation. Data alignment ensures the spatial consistency of data from different sources, while feature extraction focuses on extracting information critical to describing the environment, such as obstacle shapes, road elevations, and traffic signs. This information is then classified and annotated using a semantic segmentation algorithm to generate map data with semantic labels.

[0037] The goal of the fusion process is to construct a real-time 3D map of the environment corresponding to the target vehicle. This real-time 3D map not only includes road geometry information, but also dynamic obstacles, traffic signs, terrain features, etc., forming a comprehensive, three-dimensional, and dynamic description of the environment.

[0038] Furthermore, based on the target fusion results, a decision-making algorithm is used to optimize the target vehicle's driving strategy, resulting in a driving decision. Optionally, the decision-making algorithm should consider road conditions, traffic conditions, obstacle distribution, weather factors, and other factors to generate a safe and efficient driving strategy, such as preemptive deceleration, lane change, or suspension adjustment.

[0039] Optionally, the driving decision result is specifically expressed as a series of control instructions for instructing the vehicle how to adjust its driving strategy according to environmental changes.

[0040] Finally, based on the driving decision results, the vehicle's driving strategy is adjusted in real time, and the vehicle is controlled through actuators (such as the steering, brakes, and suspension system). For example, if a landslide or falling rocks are detected ahead, the Automatic Emergency Braking (AEB) system is immediately triggered, and a detour is planned to ensure vehicle safety.

[0041] Based on the above steps S11 to S14, first, the drone perception data and the vehicle perception data are obtained, wherein the drone perception data is used to represent the road environment information outside the line of sight perception range of the target vehicle, and the vehicle perception data is used to represent the road environment information within the line of sight perception range of the target vehicle; secondly, the drone perception data and the vehicle perception data are fused to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle; further, the target fusion result is used to make driving decisions for the target vehicle to obtain a driving decision result, wherein the driving decision result is used to represent the driving strategy to be used by the target vehicle; finally, the target vehicle is controlled based on the driving decision result. This application first obtains drone perception data and vehicle perception data, ensuring that road environment information within and outside the target vehicle's visual range is captured. The drone perception data effectively fills the perception blind spots caused by the limited field of view of traditional vehicle sensors and expands the vehicle's environmental perception range. Then, the drone perception data and vehicle perception data are fused to generate a target fusion result. This result is used to construct a real-time three-dimensional environmental map around the target vehicle. The fusion process can integrate data from different sources, overcome the limitations of single sensor data, provide more complete and accurate road condition information, improve the accuracy and comprehensiveness of environmental perception, and provide a solid data foundation for subsequent decision-making. Further, the target fusion result is used to make driving decisions. The generated driving decision results can more accurately predict obstacles, terrain changes, etc., adjust the driving route and speed in time, optimize the driving strategy, and improve the vehicle's decision-making ability and safety in complex environments. Finally, the vehicle is controlled based on the driving decision results to ensure that the vehicle drives safely and efficiently according to the optimized driving strategy. The above vehicle control process relies on accurate driving decision results, which can effectively avoid safety hazards caused by perception blind spots, improve the reliability of remote vehicle control, and thus improve driving safety. Therefore, the present application can achieve the technical effect of improving the accuracy and safety of remote vehicle control strategies, thereby solving the technical problems in related technologies of insufficient accuracy and low safety of remote vehicle control strategies due to perception blind spots.

[0042] The vehicle control method in the embodiment of the present application is further introduced below.

[0043] In an optional embodiment, in step S11, obtaining drone perception data includes the following steps:

[0044] Step S111, obtaining online map data of the Internet of Vehicles, wherein the online map data of the Internet of Vehicles is used to represent the real-time road conditions corresponding to the target vehicle;

[0045] Step S112: In response to the IoV online map data satisfying a preset collection trigger condition, a planning analysis is performed based on the navigation route information and real-time road condition information of the target vehicle to obtain a planning analysis result, wherein the planning analysis result is used to determine the corresponding take-off height, hovering radius, and scanning path of the vehicle-mounted UAV;

[0046] Step S113, obtaining drone scanning data based on the planning analysis results;

[0047] Step S114: extract features from the drone scanning data to obtain drone perception data.

[0048] The above-mentioned Internet of Vehicles online map data refers to the online map information downloaded in real time after the vehicle is connected to the Internet. The online map information includes but is not limited to road network structure, real-time traffic conditions, weather conditions, special event reports (such as construction, accidents), etc.

[0049] In an optional embodiment, before the vehicle enters driving mode, the onboard system will automatically connect to the cloud server to request the latest online map data for the current route. The requested online map data includes static road information as well as real-time dynamic information such as traffic flow speed, congestion length, and accident location.

[0050] The above-mentioned preset collection trigger conditions define the specific circumstances under which the vehicle-mounted drone needs to be activated to collect additional perception data. For example, if there is a long distance congestion ahead, the vehicle-mounted drone needs to be activated to collect additional perception data.

[0051] Optionally, when the analysis of the online map data of the Internet of Vehicles meets the preset collection trigger conditions, the drone perception mission planning is initiated and the planning analysis results are obtained. Based on the current position of the target vehicle, navigation route information, and real-time road conditions, the corresponding takeoff height of the vehicle-mounted drone (e.g., 30-50 meters) is calculated to cover a certain range of vision above the target vehicle. The corresponding hovering radius (e.g., 50 meters) and scanning path (e.g., vertical scanning along the front of the navigation route) are also calculated to ensure that the drone can fully scan the road conditions ahead, including lane congestion, road obstacles, etc.

[0052] For example, when the target vehicle's Internet of Vehicles online map displays that the road ahead is congested, the on-board artificial intelligence automatically wakes up the on-board drone and plans the drone's take-off height, hovering radius and scanning path (such as a hovering radius of 50 meters and a take-off height of 30 meters) based on the navigation route information and real-time traffic information.

[0053] For example, the acquired online map data of the Internet of Vehicles is used to analyze the terrain characteristics of the route to be traveled. For example, for open areas, a higher take-off altitude (such as 50 meters) is selected to expand the field of view; while in forests or mountainous areas, the flight altitude should be appropriately lowered (such as 30 meters) to avoid being blocked by trees and rocks. Alternatively, the type and location of obstacles that may be encountered ahead are estimated through real-time road condition information. If it is predicted that there is an overpass, tunnel or high-rise building ahead, a lower flight altitude is selected to avoid collision. On open roads, the flight altitude can be increased to increase the scanning range.

[0054] Optionally, in an urban environment, a smaller circling radius may be selected to reduce the impact on other vehicles; whereas in rural or uninhabited areas, a larger circling radius helps cover a wider area.

[0055] Optionally, a scanning path is determined using a path planning algorithm based on the acquired online map data of the Internet of Vehicles. The determined scanning path is a flight path that not only meets the data collection requirements but also avoids potential obstacles and saves energy.

[0056] The above-mentioned drone scanning data refers to the high-definition images and three-dimensional point cloud data collected by the camera or sensors carried by the drone, which is used to build a real-time three-dimensional environmental map around the vehicle.

[0057] The vehicle-mounted drone automatically takes off to perform its perception mission according to the takeoff altitude, hovering radius, and scanning path specified in the planned analysis results. For example, while the vehicle-mounted drone is performing its perception mission, its onboard RGB-D camera and LiDAR sensor begin operating, continuously capturing images and depth information of the road ahead. The images collected by the RGB-D camera contain rich color information, while the LiDAR generates point cloud data through laser ranging. The combination of the two can construct an accurate three-dimensional map of the environment. The raw data collected by the vehicle-mounted drone is transmitted to the vehicle-mounted system via a wireless communication link, resulting in the drone's scanning data.

[0058] Furthermore, feature extraction is performed on the received drone scanning data to obtain drone perception data. Feature extraction includes image edge detection, texture recognition, obstacle classification, and 3D spatial feature extraction of point cloud data, such as obstacle location, size, and motion trend.

[0059] In an optional embodiment, feature extraction is performed using computer vision and machine learning algorithms. For example, obstacles in RGB-D images are identified and classified using depth information and color contrast, while LiDAR point cloud data is used to distinguish different types of objects, such as pedestrians, vehicles, trees, and road boundaries, through cluster analysis and feature point detection. The extracted feature information is integrated to generate drone perception data.

[0060] The drone perception data acquisition process disclosed in this application initiates drone perception mission planning when the analysis of online IoV map data meets preset collection trigger conditions. Using navigation route information and real-time road conditions, the drone's takeoff altitude, hovering radius, and scanning path are planned to ensure effective data capture for specific scenarios. Subsequently, the drone executes the mission according to the planned results, obtaining drone scanning data. Feature extraction is performed on this data, converting the drone scanning data into structured drone perception data, providing critical environmental awareness and decision-making basis for the vehicle, thereby achieving efficient and accurate environmental perception by vehicle-mounted drones.

[0061] In an optional embodiment, in step S11, obtaining drone perception data includes the following steps:

[0062] Step S115, in response to receiving the data collection instruction from the target user, obtaining the drone scanning data according to the data collection instruction, wherein the data collection instruction includes: the take-off height, hovering radius and scanning path corresponding to the vehicle-mounted drone;

[0063] Step S116: extract features from the drone scanning data to obtain drone perception data.

[0064] The above data collection instructions include parameters such as the vehicle-mounted drone's take-off height, hovering radius, and scanning path, which are used to guide the drone to perform customized environmental perception tasks.

[0065] Optionally, the data collection instruction is triggered based on the immediate needs of the target user. For example, the target user wants to know the road conditions within 2 kilometers ahead and therefore requires the drone to perform a detailed survey.

[0066] For example, the user manually launches the drone to take off and scan the length of the real-time congestion ahead and the congestion status of each lane according to immediate needs.

[0067] After receiving the data collection instruction from the target user, the flight parameters included in the data collection instruction, such as take-off altitude, hovering radius, and scanning path, are parsed. The vehicle-mounted drone performs the perception task according to the take-off altitude, hovering radius, and scanning path specified in the data collection instruction, thereby obtaining drone scanning data. The process of the vehicle-mounted drone performing the perception task is set by the target user according to current needs and environmental conditions, which improves the flexibility and practicality of the vehicle-mounted drone application. For example, in an emergency, the target user can request the vehicle-mounted drone to increase the flight altitude to cover a wider area faster and obtain a panoramic image of the accident scene. When detailed road condition analysis is required, a smaller hovering radius can be set to ensure in-depth detection of a specific area.

[0068] Through computer vision and machine learning algorithms, feature extraction is performed on drone scanning data, and useful information such as road features, obstacle locations, pedestrian activities, etc. is identified and extracted from drone scanning data, and the extracted useful information is converted into structured drone perception data.

[0069] Optionally, the use of vehicle-mounted drones increases the target vehicle perception radius from the traditional 200 meters to 500 meters, reducing the coverage blind spot rate by 80%.

[0070] The drone perception data acquisition process disclosed in this application enables users to flexibly adjust the flight mode of a vehicle-mounted drone based on real-time needs, acquiring detailed environmental information about a specific area. The feature extraction process ensures the structured and understandable nature of drone perception data, providing critical data support for applications such as intelligent driving and remote control.

[0071] In an optional embodiment, the vehicle control method further includes the following steps:

[0072] Step S117, obtaining priority information of the drone scanning data, wherein the priority information is used to determine the urgency of the data corresponding to the drone scanning data;

[0073] Step S118, determining a target transmission link based on the priority information;

[0074] Step S119: The drone scanning data is transmitted between the target vehicle and the vehicle-mounted drone using the target transmission link.

[0075] Priority information refers to the classification results of drone scanning data according to its importance or urgency. Data in emergency situations (such as the sudden appearance of an obstacle ahead) is generally given the highest priority, while routine environmental monitoring data is given a lower priority.

[0076] For example, when a vehicle-mounted drone performs a perception task, each frame of data or point cloud collected by its onboard sensors is assigned a priority value. For example, safety-related data such as obstacle detection and pedestrian intrusion warnings are marked as high priority, while general traffic flow monitoring and environmental monitoring data are marked as low priority.

[0077] According to the priority information of the drone scanning data, the appropriate data transmission path, namely the target transmission link, is selected to ensure that the drone scanning data can be transmitted in the shortest time and with the highest quality.

[0078] Among multiple available links, a low-latency, high-bandwidth link is selected for transmission of high-priority data.

[0079] Exemplarily, available links include the fifth generation mobile communication network (5th Generation Mobile Network, 5G), wireless local area network (Wireless Fidelity, Wi-Fi), etc.

[0080] In an optional embodiment, the target transmission link is intelligently selected based on the priority information of the drone scanning data. For example, for high-priority emergency data (such as emergency obstacle avoidance alerts), low-latency links (such as the main link of the 5G network) are preferentially used for data transmission to ensure that the emergency data can be received and processed in less than 50 milliseconds. For low-priority general data (such as ambient temperature and humidity information), it can be transmitted through auxiliary links with lower bandwidth but lower power consumption (such as Wi-Fi) to save energy.

[0081] Once the target transmission link is determined, the vehicle-mounted drone transmits the drone-scanned data to the target vehicle via the target transmission link. During the transmission process, data encoding and compression are performed based on the characteristics of the target transmission link and the nature of the data, adapting to the transmission rate and bandwidth limitations of each link. For example, high-priority data undergoes only light compression before transmission to preserve key features, while low-priority data undergoes deep compression to reduce transmission overhead.

[0082] In an optional embodiment, the quality of the target transmission link, such as signal strength and packet loss rate, is monitored in real time to ensure the continuity and stability of data transmission. When the quality of the target transmission link degrades, the link can be dynamically switched to maintain efficient data transmission.

[0083] Through the above steps disclosed in this application, the priority division and link selection of vehicle-mounted drone scanning data are realized, ensuring that high-priority data can be processed in the first time, providing timely information support for key decisions such as emergency avoidance and safe driving. At the same time, the rational use of different transmission links effectively balances the relationship between data transmission speed, quality and energy consumption, and improves the energy utilization efficiency and overall performance of the entire system. The above method greatly improves the response speed and reliability of intelligent connected vehicles in real-time perception and remote control, and is of great significance to promoting the development of autonomous driving technology and improving road traffic safety.

[0084] In an optional embodiment, in step S118, determining the target transmission link based on the priority information includes the following steps:

[0085] Step S1181, in response to determining that the urgency of the data is greater than or equal to a preset threshold based on the priority information, determining the primary transmission link as a target transmission link;

[0086] Step S1182: In response to determining that the data urgency is less than a preset threshold based on the priority information, the auxiliary transmission link is determined as a target transmission link, wherein the transmission delay of the auxiliary transmission link is greater than the transmission delay of the main transmission link.

[0087] The above-mentioned data urgency refers to the degree of data urgency assessed based on the content of drone scanning data and its impact on driving safety or emergency decision-making.

[0088] The above-mentioned preset threshold is a pre-set urgency reference value, which is used to distinguish high-priority data from low-priority data.

[0089] The above-mentioned main transmission link refers to a transmission link with high bandwidth and low latency characteristics, which is used to transmit high-priority data. It is usually a direct link or Uu interface link of a 5G network. The Uu interface link refers to the wireless interface between the user terminal (UserEquipment, UE) and the evolved Node B (Evolved Node B, eNodeB). In a mobile communication system, the Uu interface is responsible for carrying all wireless communications between the user equipment and the base station, including data transmission on the control plane and the user plane. Specifically for 5G networks, the Uu interface link is a wireless channel connecting 5G user equipment (such as smartphones, communication modules of smart cars) and 5G base stations. It supports high-speed data transmission, low-latency communication and large-scale device connections.

[0090] Optionally, the urgency of the drone scan data is first assessed based on its priority information. For example, when a vehicle-mounted drone detects an accident on the road ahead, the relevant data is marked as high priority. If the assessment indicates that the urgency has reached a preset threshold, the primary transmission link is prioritized for data transmission. The 5G network's primary link, with its high speed and low latency, is well-suited for transmitting urgent drone scan data, ensuring the immediacy and reliability of the information.

[0091] For example, when a vehicle-mounted drone detects a landslide or falling rocks in front of a target vehicle, it immediately triggers the vehicle's Automatic Emergency Braking (AEB) system through a low-latency link (<50ms).

[0092] Compared with the main transmission link, the auxiliary transmission link has lower bandwidth and higher latency, but lower power consumption and is suitable for transmitting low-priority data.

[0093] Optionally, for drone scan data whose data urgency is below a preset threshold, an auxiliary transmission link is selected as the target transmission link. Low-urgency drone scan data includes ambient temperature and humidity monitoring information, or traffic flow data in non-emergency situations. While the auxiliary transmission link has higher transmission latency, it consumes less power, making it suitable for long-term, continuous data transmission. This transmission method, through the auxiliary transmission link, enables continuous environmental monitoring to be maintained at a lower cost while also reducing the occupancy of the primary transmission link, ensuring that the primary transmission link is always ready to handle sudden, high-priority data transmission needs.

[0094] Exemplarily, the main transmission link transmits high-priority control instructions (such as emergency obstacle avoidance), and the auxiliary transmission link transmits vehicle-to-everything (V2X) traffic guidance information.

[0095] By intelligently evaluating the urgency of drone scanning data based on priority information and dynamically selecting the main transmission link or auxiliary transmission link accordingly, the method disclosed in this application effectively balances the efficiency and resource consumption of information transmission. While ensuring that high-priority data can arrive quickly, it also fully utilizes the low-cost advantage of the auxiliary transmission link to transmit daily monitoring data, thereby achieving the optimal state of collaborative work between the target vehicle and the on-board drone. The above method disclosed in this application not only improves the response speed of intelligent connected vehicles when facing emergencies and reduces the risks caused by information delays, but also improves the overall operating efficiency and economic benefits of the system through refined data transmission management.

[0096] In an optional embodiment, in step S12, fusing the drone perception data and the vehicle perception data to obtain a target fusion result includes the following steps:

[0097] Step S121, performing fusion processing based on the UAV perception data and the vehicle perception data to obtain an initial fusion result;

[0098] Step S122, constructing a real-time three-dimensional environment map based on the initial fusion result;

[0099] Step S123: Perform semantic segmentation processing on the real-time three-dimensional environment map to obtain a target fusion result.

[0100] The above-mentioned fusion processing refers to combining data from different sensors through a fusion algorithm to form unified environmental perception information.

[0101] In an optional embodiment, the drone perception data and the vehicle perception data are first transmitted to the vehicle-mounted edge computing unit or cloud server. Through time synchronization and spatial calibration, the drone perception data and the vehicle perception data are aligned on the timeline and matched in the coordinate system. Subsequently, data fusion processing is performed to obtain the initial fusion result. The data fusion processing process includes data preprocessing (such as noise filtering and outlier removal), feature matching (such as data correspondence through feature points, edges, etc.), and the application of data fusion algorithms (such as Kalman filtering, particle filtering, etc.).

[0102] Sensor data is often affected by electromagnetic interference, physical jitter, or inherent sensor uncertainty, introducing noise. To reduce the impact of noise on data fusion, various filtering techniques, such as mean filtering, median filtering, and Gaussian filtering, can be used to smooth the raw data.

[0103] Optionally, data rationality boundaries can be set to identify and exclude data values ​​outside the normal range. For example, for RGB-D images, a reasonable range of pixel values ​​can be set, and data points outside this range need to be removed. For LiDAR point clouds, the average distance and standard deviation between point clouds can be calculated to identify and remove points whose distances deviate significantly from the average. These points may be caused by false detections or abnormal reflections.

[0104] Optionally, stable feature points such as corners, edges, texture patterns, etc. are found in the image and point cloud data. Through feature point matching algorithms such as scale-invariant feature transformation, a one-to-one or many-to-one correspondence can be established between the drone perception data and the vehicle perception data to ensure the spatial consistency of the data.

[0105] For image data, edge detection algorithms (such as Canny edge detection and the Sobel operator) are used to identify object boundaries or road edges. In point cloud data, similar edge information can also be identified by searching for points with sudden changes in density. Edge matching helps find common environmental structures, such as road layouts and building outlines, across different data sources.

[0106] Alternatively, the Kalman filter is a recursive prediction-correction algorithm widely used in multi-sensor data fusion. It combines sensor measurement errors with the system dynamics model to optimally estimate the state vector. When fusing drone and vehicle perception data, the Kalman filter can predict the vehicle's motion state and correct for measurement errors caused by sensor noise, resulting in a more accurate estimate of the vehicle and environment states.

[0107] Alternatively, particle filtering, a nonlinear state estimation technique based on the Monte Carlo method, is suitable for handling non-Gaussian noise and nonlinear system models. During the fusion process, particle filtering can simulate the dynamic environmental changes of drone and vehicle perception data. By updating particle weights and resampling, it continuously adjusts the system state to achieve real-time tracking and estimation of the changing environment.

[0108] The purpose of fusion processing is to resolve data redundancy, eliminate information conflicts, and combine scattered sensor information into a complete set of environmental perception information. Through fusion processing, the complementary advantages of vehicle-mounted drone and vehicle sensors can be utilized to obtain more comprehensive environmental perception data.

[0109] The above-mentioned real-time three-dimensional environment map is a three-dimensional model built based on the initial fusion results that can reflect the vehicle's surrounding environment in real time, including elements such as roads, obstacles, pedestrians, and traffic signs.

[0110] Optionally, the initial fusion result contains rich environmental feature information. This initial fusion result, combined with the vehicle's real-time position and posture data, is used to construct a real-time 3D environmental map. This process includes dynamically updating map data and identifying and tracking obstacles in real time.

[0111] In an optional embodiment, the algorithm for constructing a real-time three-dimensional environment map can be point cloud-based three-dimensional modeling, image-based three-dimensional reconstruction, or an environment understanding algorithm combined with deep learning.

[0112] The semantic segmentation process described above is used to classify each pixel or point in an image or point cloud data into a specific category, such as road, vehicle, pedestrian, tree, etc.

[0113] Optionally, semantic segmentation processing is performed on the real-time 3D environment map to obtain a target fusion result. By performing semantic segmentation processing on the real-time 3D environment map, information can be further refined, and various environmental elements can be identified and classified. Semantic segmentation processing usually relies on deep learning models, such as convolutional neural networks. The deep learning model that implements semantic segmentation processing has been trained with a large amount of data and can accurately identify and distinguish different objects and environmental features. Through semantic segmentation, the real-time 3D environment map is converted into a set of structured environmental element labels, clarifying the location and type of roads, obstacles, pedestrians, etc.

[0114] This application discloses a process for fusing drone and vehicle perception data to generate a target fusion result. Through precise data fusion and environmental modeling, it enables intelligent connected vehicles to achieve real-time, comprehensive perception of complex environments. Semantic segmentation processing can intelligently distinguish and identify different environmental elements, providing intuitive and effective decision support for drivers.

[0115] In an optional embodiment, the vehicle control method further comprises the following steps:

[0116] Step S131, uploading the target fusion result to a cloud server to enable the cloud server to perform data conversion based on the target fusion result to obtain target perspective map data;

[0117] Step S132, receiving the target perspective map data issued by the cloud server;

[0118] Step S133, using the target perspective map data to make driving decisions to obtain driving decision results.

[0119] The cloud server mentioned above refers to a remote server that provides computing resources and storage capabilities, used to handle complex computing tasks such as data conversion and advanced environment modeling.

[0120] In an optional embodiment, after obtaining the target fusion result, the target vehicle will upload the target fusion result to the cloud server through a high-speed, low-latency wireless communication link (such as a 5G network).

[0121] The process of uploading the target fusion result to the cloud server needs to ensure the secure transmission of data, and usually adopts data encryption technology such as Advanced Encryption Standard (AES) encryption to protect sensitive driving environment information. The cloud server has strong computing power and can handle a large amount of data and perform complex computing tasks such as three-dimensional environment map conversion and semantic segmentation.

[0122] In an optional embodiment, after receiving the target fusion result, the cloud server will first preprocess and integrity check the target fusion result to ensure data quality and availability. Preprocessing includes data format conversion, data deduplication, and data integrity verification to ensure smooth subsequent processing. Then the three-dimensional environment perception data corresponding to the target fusion result is converted into a two-dimensional image or map from the top (i.e. bird's eye view image), so that objects, road layout, etc. in the environment can be more intuitively displayed, facilitating understanding and analysis. Further, the cloud server uses a pre-trained deep learning model to perform semantic segmentation on the converted bird's eye view image, and further obtains target perspective map data.

[0123] Optionally, to ensure the efficiency and real-time performance of data transmission, the cloud server needs to optimize and compress the target perspective map data, retaining key information while reducing data volume.

[0124] The cloud server securely transmits the target viewpoint map data through encryption, ensuring data security and privacy during transmission. It then sends it to the vehicle via a low-latency communication link for driving decision-making. The target vehicle receives the target viewpoint map data from the cloud server, decodes it, and verifies it to ensure data integrity and accuracy.

[0125] After receiving the target-view map data, the target vehicle performs real-time path planning and decision-making based on information contained in the target-view map, such as obstacle locations, road conditions, and traffic participant behavior, resulting in a driving decision. For example, on a winding mountain road, the curvature and slope of the next curve can be predicted in advance, optimizing the vehicle's suspension settings and improving vehicle stability and ride comfort.

[0126] By uploading the target fusion results to the cloud and utilizing its powerful computing resources for data conversion, we generate more easily understandable target-view map data. Cloud-based processing avoids excessive load on the onboard computing unit, ensuring stable operation of the target vehicle's control system. Furthermore, the target-view map data provides a global perspective of the environment, enabling the target vehicle to pre-plan its route, avoid potential risks, and adapt to complex and changing driving conditions.

[0127] In an optional embodiment, the vehicle control method further includes the following steps:

[0128] Step S134, updating the original navigation path of the target vehicle based on the driving decision result to obtain an updated navigation path;

[0129] Step S135 : Displaying the updated navigation route on the augmented reality head-up display interface of the target vehicle.

[0130] Optionally, the original navigation path refers to a driving route preset by the target vehicle before departure, which is usually based on an in-vehicle navigation system or an online map.

[0131] The updated navigation path mentioned above refers to the driving route adjusted according to the driving decision results, aiming to avoid obstacles, reduce driving time or improve driving safety.

[0132] In an optional embodiment, upon receiving the driving decision result, the target vehicle's navigation system immediately updates the original navigation path to obtain an updated navigation path. Optionally, the navigation path is dynamically adjusted using a path planning algorithm based on obstacle information, road conditions, and recommended actions in the driving decision result. For example, if the driving decision result indicates an impassable obstacle ahead, the navigation system will calculate the shortest detour route. If the driving decision result recommends accelerating on a specific road section to avoid congestion, the navigation system will adjust the acceleration timing and path based on the current vehicle speed and road conditions ahead.

[0133] Augmented reality head-up display is a transparent display technology that projects navigation information, vehicle status and environmental information directly in front of the driver, allowing the driver to intuitively see driving assistance information integrated with the actual road environment.

[0134] Optionally, the updated navigation path will be projected onto the target vehicle's augmented reality head-up display (AR-HUD) interface. The AR-HUD system uses the front camera and the target vehicle's positioning system to determine the target vehicle's position relative to the updated navigation path, and aligns the updated navigation path information with the actual road scene in the driver's line of sight. Key points on the path (such as turns and obstacle avoidance points) will be highlighted, prompting the driver in the form of color, icons or animation to ensure that the driver can quickly notice changes in the path. In addition, AR-HUD can also display real-time traffic conditions, obstacle types, distances and other information to provide all-round driving assistance.

[0135] By updating the navigation path and visually displaying it using AR-HUD, the real-time and effectiveness of driving decisions are effectively improved. First, the feedback on driving decision results enables the vehicle to instantly adjust its driving strategy to avoid potential dangers or delays, improving driving safety and efficiency. Second, the introduction of AR-HUD not only presents complex information to the driver in an intuitive manner, reducing driver distraction, but also enhances the driver's awareness of the surrounding environment. This is especially evident at night, in bad weather, or in complex road conditions, where the driving assistance effect is more pronounced.

[0136] In an optional embodiment, the vehicle control method includes: first, before the intelligent connected vehicle and the drone system work together, the system needs to be initialized and configured. The initialization and configuration process includes checking the status of on-board sensors (such as cameras, radars, terrain perspective systems, etc.) and binocular vision, lidar, millimeter-wave radar and other equipment carried by the drone to ensure their normal operation. At the same time, it is necessary to configure a vehicle-grade 5G+V2X communication module to ensure unimpeded PC5 direct communication and Uu dual-channel communication. In addition, it is necessary to set the drone's take-off conditions, cruising altitude, scanning path and other parameters to enable quick response when needed.

[0137] When the connected car's online car map indicates traffic congestion or other complex terrain ahead, the onboard AI system automatically activates the drone. Based on navigation information and road condition data, the onboard AI plans the drone's hovering altitude and scanning path. For example, on congested roads, the drone is programmed to scan at an altitude of 30 meters and a hovering radius of 50 meters to obtain the real-time length of the congestion ahead and the congestion status of each lane. Users can also manually launch the drone based on actual conditions for more flexible sensing tasks. Users can control the drone's takeoff, altitude adjustment, and scanning direction through the onboard display or voice commands to adapt to different road conditions and driving needs.

[0138] During flight, drones collect RGB-D images, LiDAR point clouds, and other data in real time. This data is compressed and feature extracted by the onboard edge computing unit to reduce data transmission and improve processing efficiency. The computing power of the onboard edge computing unit must reach or exceed 10TOPS to ensure real-time processing capabilities. After compression and feature extraction, the drone's perception data is fused with data from the onboard sensors. This fusion process constructs a 360° real-time 3D map of the vehicle's surroundings and identifies terrain features such as mud and gravel. Alternatively, a pure vision + LiDAR fusion solution can reduce costs by 60% compared to a single high-precision radar solution. Semantic segmentation algorithms can further extract key information such as roads, obstacles, and traffic signs, providing a basis for subsequent decision-making.

[0139] Based on the global environmental data collected by drones, intelligent connected vehicles will adjust their driving strategies. For example, on a winding mountain road, the vehicle can predict the curvature of a curve and use the adaptive suspension system to optimize the vehicle's posture, improving driving stability and safety.

[0140] To provide a more comprehensive basis for decision-making, some data will be synchronized to the cloud for further processing. In the cloud, the data will be converted into a bird's-eye view, allowing the remote control center to have a more intuitive understanding of the vehicle's surrounding environment.

[0141] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0142] According to another aspect of the embodiment of the present application, a vehicle control device is further provided. Figure 2 is a structural block diagram of a vehicle control device 200 according to an embodiment of the present application, such as Figure 2 As shown, the device includes: an acquisition module 201, which is used to acquire drone perception data and vehicle perception data, wherein the drone perception data is used to represent road environment information outside the visual range of the target vehicle, and the vehicle perception data is used to represent road environment information within the visual range of the target vehicle; a fusion module 202, which is used to fuse the drone perception data and the vehicle perception data to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle; a decision module 203, which is used to make a driving decision for the target vehicle using the target fusion result to obtain a driving decision result, wherein the driving decision result is used to represent the driving strategy to be used by the target vehicle; and a control module 204, which is used to control the target vehicle based on the driving decision result.

[0143] According to another aspect of an embodiment of the present application, a vehicle is further provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the vehicle control method in each embodiment of the present application is executed when the program is running.

[0144] Optionally, in this embodiment, the processor in the vehicle may be configured to run an executable program to perform the following steps:

[0145] Step S11, acquiring drone perception data and vehicle perception data, wherein the drone perception data is used to represent road environment information outside the line of sight perception range of the target vehicle, and the vehicle perception data is used to represent road environment information within the line of sight perception range of the target vehicle;

[0146] Step S12: fusing the drone perception data and the vehicle perception data to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle;

[0147] Step S13, using the target fusion result to make a driving decision for the target vehicle to obtain a driving decision result, wherein the driving decision result is used to represent the driving strategy to be used by the target vehicle;

[0148] Step S14: Control the target vehicle based on the driving decision result.

[0149] According to another aspect of an embodiment of the present application, a computer-readable storage medium is also provided, which includes a stored executable program, wherein when the executable program is running, the device where the readable storage medium is located is controlled to execute the vehicle control method in each embodiment of the present application.

[0150] Optionally, in this embodiment, the executable program may be configured to store an executable program for performing the following steps:

[0151] Step S11, acquiring drone perception data and vehicle perception data, wherein the drone perception data is used to represent road environment information outside the line of sight perception range of the target vehicle, and the vehicle perception data is used to represent road environment information within the line of sight perception range of the target vehicle;

[0152] Step S12: fusing the drone perception data and the vehicle perception data to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle;

[0153] Step S13, using the target fusion result to make a driving decision for the target vehicle to obtain a driving decision result, wherein the driving decision result is used to represent the driving strategy to be used by the target vehicle;

[0154] Step S14: Control the target vehicle based on the driving decision result.

[0155] An embodiment of the present application further provides a computer program product, including a computer program, which implements the vehicle control method in each embodiment of the present application when executed by a processor.

[0156] An embodiment of the present application also provides a computer program product, including a non-volatile computer-readable storage medium, which is used to store a computer program. When the computer program is executed by a processor, it implements the vehicle control method in each embodiment of the present application.

[0157] An embodiment of the present application further provides a computer program, which, when executed by a processor, implements the vehicle control method in each of the above-mentioned embodiments of the present application.

[0158] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0159] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0160] The units described as separate components may or may not be physically separate, and 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 units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0161] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0162] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0163] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A vehicle control method, characterized in that: include: Acquiring drone perception data and vehicle perception data, wherein the drone perception data is used to represent road environment information outside the line of sight perception range of the target vehicle, and the vehicle perception data is used to represent road environment information within the line of sight perception range of the target vehicle; Fusing the drone perception data and the vehicle perception data to obtain a target fusion result, wherein the target fusion result is used to construct a real-time three-dimensional environment map corresponding to the target vehicle; Making a driving decision for the target vehicle using the target fusion result to obtain a driving decision result, wherein the driving decision result is used to represent a driving strategy to be used by the target vehicle; The target vehicle is controlled based on the driving decision result.

2. The method according to claim 1, characterized in that Acquiring the drone perception data includes: Obtaining online map data of the Internet of Vehicles, wherein the online map data of the Internet of Vehicles is used to represent the real-time road conditions corresponding to the target vehicle; In response to the IoV online map data satisfying a preset collection trigger condition, performing planning analysis based on the navigation route information and real-time road condition information of the target vehicle to obtain a planning analysis result, wherein the planning analysis result is used to determine a corresponding take-off height, hovering radius, and scanning path of the vehicle-mounted drone; Acquiring drone scanning data based on the planning analysis results; Feature extraction is performed on the drone scanning data to obtain the drone perception data.

3. The method according to claim 1, characterized in that Acquiring the drone perception data includes: In response to receiving a data collection instruction from a target user, acquiring drone scanning data according to the data collection instruction, wherein the data collection instruction includes: a take-off height, a hovering radius, and a scanning path corresponding to the vehicle-mounted drone; Feature extraction is performed on the drone scanning data to obtain the drone perception data.

4. The method according to claim 2 or 3, characterized in that The method further comprises: Obtaining priority information of the drone scanning data, wherein the priority information is used to determine the urgency of data corresponding to the drone scanning data; determining a target transmission link based on the priority information; The target transmission link is used to transmit drone scanning data between the target vehicle and the vehicle-mounted drone.

5. The method according to claim 4, characterized in that Determining the target transmission link based on the priority information includes: In response to determining, based on the priority information, that the urgency of the data is greater than or equal to a preset threshold, determining the target transmission link as the primary transmission link; In response to determining that the data urgency is less than the preset threshold based on the priority information, an auxiliary transmission link is determined as the target transmission link, wherein a transmission delay of the auxiliary transmission link is greater than a transmission delay of the main transmission link.

6. The method according to claim 1, characterized in that The UAV perception data and the vehicle perception data are fused to obtain the target fusion result, including: Performing fusion processing based on the UAV perception data and the vehicle perception data to obtain an initial fusion result; Constructing the real-time three-dimensional environment map according to the initial fusion result; Perform semantic segmentation processing on the real-time three-dimensional environment map to obtain the target fusion result.

7. The method according to claim 1, characterized in that The method further comprises: Uploading the target fusion result to a cloud server, so that the cloud server performs data conversion based on the target fusion result to obtain target perspective map data; Receiving the target perspective map data sent by the cloud server; The target viewpoint map data is used to make a driving decision to obtain the driving decision result.

8. The method according to claim 1, characterized in that The method further comprises: updating the original navigation path of the target vehicle based on the driving decision result to obtain an updated navigation path; The updated navigation route is displayed on an augmented reality head-up display interface of the target vehicle.

9. A vehicle, characterized in that: include: a memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 8 when running.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 8.

Images