Unmanned aerial vehicle assisted driving method and device in non-standard scene and electronic equipment

Abstract

The application discloses a method and device for unmanned aerial vehicle assisted driving in a non-standard scene and electronic equipment. According to the application, the unmanned aerial vehicle is used to break away from the field of view of the vehicle-mounted sensor and perceive the environment of various special non-standard working conditions from the external space, thereby supplementing the information shortage of the vehicle-mounted perception system and outputting a matching driving decision. Therefore, the technical problem that the existing intelligent driving only relies on the vehicle-mounted sensor to identify the environment and cannot fully perceive the non-standard scene such as landslide, water accumulation, narrow road limit and the like, and cannot accurately judge the risk can be solved, and the technical effect that the remote road hazards are identified in advance, the driving control scheme is generated to adapt to the scene, and the driving safety and the reliability of the assisted driving of the vehicle in the non-standard complex road condition are effectively improved can be achieved.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a method, device and electronic equipment for drone-assisted driving in non-standard scenarios. Background Technology

[0002] As a crucial development direction in modern transportation, automotive intelligence is widely applied in scenarios such as autonomous driving on highways and urban road assistance. Among related technologies, the collaborative operation of onboard sensors, high-precision maps, and deep learning algorithms has constructed an intelligent driving system encompassing environmental perception, path planning, and decision-making control.

[0003] Existing intelligent driving methods directly adopt recognition models based on big data, which makes it impossible for vehicles to accurately perceive environmental risks in situations such as road landslides, unknown water depths, and narrow road height restrictions, thus affecting driving safety and user experience. Summary of the Invention

[0004] This disclosure provides a method, device, and electronic device for drone-assisted driving in non-standard scenarios.

[0005] According to a first aspect of this disclosure, a method for drone-assisted driving in non-standard scenarios is provided, comprising: In response to a non-standard scenario trigger signal, the drone is activated and establishes a communication connection with the vehicle. Control the drone to perform environmental perception tasks and acquire environmental data from non-standard scenarios; A driving assistance strategy is generated based on the environmental data; The driver assistance strategy is fed back to the vehicle to control the vehicle to perform corresponding operations.

[0006] Optionally, the non-standard scenario trigger signal includes a mode selection signal actively triggered by the user through the vehicle's interactive interface. Or a trigger signal automatically generated when vehicle sensors detect abnormal operating conditions.

[0007] Optionally, controlling the UAV to perform environmental perception tasks includes at least one of the following methods: The drone is controlled to fly close to the ground and emit light onto the road surface. The smoothness of the road surface is determined by detecting the road surface reflectivity. When the reflectivity exceeds a preset smoothness threshold, a road surface slope model is constructed by measuring the reflectivity at different locations and the distance between the drone and the corresponding reflective location. The drone is controlled to drop a heavy object with a pre-tensioned rope onto the target water surface. A buoyant body is installed on the rope. The water depth is determined by measuring the distance between the buoyant body and the heavy object when it sinks to the bottom. The drone moves and records the data within the water area to generate a water depth distribution map. The drone is controlled to perform a passage space measurement in narrow roads or height-restricted locations, and the measurement is compared with the vehicle's own dimensions to determine whether passage is possible.

[0008] Optionally, controlling the UAV to perform environmental perception tasks includes: The drone is controlled to fly in the area in front of the vehicle at an altitude not lower than a preset safe altitude, to perform obstacle recognition, and to take off and avoid the recognized obstacles. The system also controls the drone to hover above the intersection, perform potential collision object identification in a preset sequence, and after identification, fly to the next target location in advance according to the vehicle's driving direction to repeat the identification.

[0009] Optionally, generating the assisted driving strategy based on the environmental data includes: When the environmental data indicates that the road surface conditions meet the dangerous spraying conditions, a strategy for spraying de-icing agent onto the road surface is generated. The spraying path width is matched with the wheel track of the vehicle. After spraying is completed, the road surface conditions are checked again. When the road surface conditions improve to below the safety threshold, a driving navigation model is established. When the environmental data indicates the presence of water accumulation and a safe passage route exists in the water depth distribution map, a driving navigation model based on the safe passage route is generated; When the environmental data indicates that the passage space is passable, a driving navigation model is generated; When the environmental data indicates that the vehicle is at the edge and needs to reverse, a model of the surrounding environment is generated and the feasibility of reversing and the specific path are evaluated.

[0010] Optionally, feeding back the assisted driving strategy to the vehicle includes: The driving navigation model is transmitted to the vehicle's intelligent driving system, which controls the vehicle to drive along the path indicated by the driving navigation model, and the drone continuously performs environmental monitoring during the driving process.

[0011] Optional, also includes: Before generating the assisted driving strategy, the drone is controlled to hover in the air to detect wind and adjust its own state to maintain stable communication. When communication cannot be maintained stably or when it is not possible to establish connections with both the satellite and the vehicle simultaneously, an information transmission algorithm is executed to encode the information content of the user terminal and the recipient information and transmit it to the UAV. The UAV is then controlled to take off and search for a stable signal source to transmit the encoded information to the external recipient and receive a reply. Subsequently, the UAV flies back and feeds the reply information back to the user.

[0012] According to a second aspect of this disclosure, a drone assisted driving device for non-standard scenarios is provided, comprising: The start-up unit is used to respond to trigger signals in non-standard scenarios, start the drone, and establish a communication connection with the vehicle; The control unit is also used to control the UAV to perform environmental perception tasks and acquire environmental data of non-standard scenarios; The generation unit is also used to generate an assisted driving strategy based on the environmental data; The feedback unit is also used to feed back the assisted driving strategy to the vehicle in order to control the vehicle to perform corresponding operations.

[0013] Optionally, the non-standard scenario trigger signal includes a mode selection signal actively triggered by the user through the vehicle interaction interface, or a trigger signal automatically generated when the vehicle sensors detect abnormal operating conditions.

[0014] Optionally, the control unit is further configured to: The drone is controlled to fly close to the ground and emit light onto the road surface. The smoothness of the road surface is determined by detecting the road surface reflectivity. When the reflectivity exceeds a preset smoothness threshold, a road surface slope model is constructed by measuring the reflectivity at different locations and the distance between the drone and the corresponding reflective location. The drone is controlled to drop a heavy object with a pre-tensioned rope onto the target water surface. A buoyant body is installed on the rope. The water depth is determined by measuring the distance between the buoyant body and the heavy object when it sinks to the bottom. The drone moves and records the data within the water area to generate a water depth distribution map. The drone is controlled to perform a passage space measurement in narrow roads or height-restricted locations, and the measurement is compared with the vehicle's own dimensions to determine whether passage is possible.

[0015] Optionally, the control unit is further configured to: The drone is controlled to fly in the area in front of the vehicle at an altitude not lower than a preset safe altitude, to perform obstacle recognition, and to take off and avoid the recognized obstacles. The system also controls the drone to hover above the intersection, perform potential collision object identification in a preset sequence, and after identification, fly to the next target location in advance according to the vehicle's driving direction to repeat the identification.

[0016] Optionally, the generation unit is further configured to: When the environmental data indicates that the road surface conditions meet the dangerous spraying conditions, a strategy for spraying de-icing agent onto the road surface is generated. The spraying path width is matched with the wheel track of the vehicle. After spraying is completed, the road surface conditions are checked again. When the road surface conditions improve to below the safety threshold, a driving navigation model is established. When the environmental data indicates the presence of water accumulation and a safe passage route exists in the water depth distribution map, a driving navigation model based on the safe passage route is generated; When the environmental data indicates that the passage space is passable, a driving navigation model is generated; When the environmental data indicates that the vehicle is at the edge and needs to reverse, a model of the surrounding environment is generated and the feasibility of reversing and the specific path are evaluated.

[0017] Optionally, the feedback unit is further configured to: The driving navigation model is transmitted to the vehicle's intelligent driving system, which controls the vehicle to drive along the path indicated by the driving navigation model, and the drone continuously performs environmental monitoring during the driving process.

[0018] Optional, also includes: The detection unit is used to control the UAV to hover in the air to detect wind and adjust its own state before generating the assisted driving strategy, and to maintain stable communication. The transmission unit is also used to execute an information transmission algorithm when communication cannot be maintained stably or when it is not possible to establish a connection with both the satellite and the vehicle at the same time. The algorithm encodes the information content of the user terminal and the information of the recipient and transmits it to the UAV. The UAV is then controlled to take off and search for a stable signal source to transmit the encoded information to the external recipient and receive the reply information. The UAV then flies back and feeds the reply information back to the user.

[0019] According to a third aspect of this disclosure, an electronic device is provided, comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect above.

[0020] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are configured to cause the computer to perform the method described in the first aspect above.

[0021] According to a fifth aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0022] The unmanned aerial vehicle (UAV) assisted driving method, device, and electronic equipment disclosed herein for non-standard scenarios, by utilizing the UAV to break free from the field of vision limitations of onboard sensors, can achieve full-domain environmental perception of various special non-standard working conditions from external space, supplementing the information shortcomings of the onboard perception system and outputting corresponding driving decisions. Therefore, it can solve the technical problems of existing intelligent driving relying solely on onboard sensors to identify the environment, resulting in incomplete perception and inaccurate risk assessment in non-standard scenarios such as landslides, water accumulation, and narrow road height restrictions. It can achieve early identification of distant road hazards and generate driving control schemes adapted to the scenarios, effectively improving the technical effect of vehicle driving safety and assisted driving reliability under non-standard and complex road conditions.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0024] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein: Figure 1 This is a flowchart illustrating a method for assisted driving of a drone in a non-standard scenario, as provided in an embodiment of this disclosure. Figure 2 This is a schematic diagram of the structure of a drone assisted driving device in a non-standard scenario provided by an embodiment of this disclosure; Figure 3 This is a schematic diagram of the structure of an unmanned aerial vehicle (UAV) assisted driving device in another non-standard scenario provided by an embodiment of this disclosure; Figure 4 A schematic block diagram of an example electronic device provided for embodiments of this disclosure. Detailed Implementation

[0025] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0026] The following description, with reference to the accompanying drawings, outlines a method, apparatus, and electronic device for assisted driving of unmanned aerial vehicles (UAVs) in non-standard scenarios according to embodiments of this disclosure.

[0027] Figure 1 This is a flowchart illustrating a non-standard drone assisted driving method provided in an embodiment of this disclosure.

[0028] like Figure 1As shown, the method includes the following steps: Step 101: In response to the non-standard scenario trigger signal, start the drone and establish a communication connection with the vehicle; The source of non-standard scenario trigger signals can be either active operation commands issued by the driver through the in-vehicle interactive interface, or trigger signals automatically generated by the vehicle's intelligent driving system based on environmental perception results. During operation, the vehicle's intelligent driving system continuously collects and analyzes surrounding environmental data. When the system determines that the characteristics of the environment exceed the coverage of its preset database and cannot reliably model and make decisions based on existing data models, it autonomously generates this trigger signal. For example, when a vehicle enters an unmaintained mountainous road or a sparsely populated natural environment, the system enters a state of lack of auxiliary decision-making due to a lack of matching standardized scenario data. In this case, it automatically issues a trigger signal to introduce external perception resources for intervention and judgment.

[0029] In response to the trigger signal, the system sends a start command to the drone, which then completes the startup process, including power-on, flight control system self-check, and mission parameter initialization. During startup, the system loads a flight mode and detection mission configuration suitable for the current non-standard scenario's environmental attributes, enabling the drone to have preliminary operational capabilities for the current scenario before takeoff. After startup, the drone establishes a two-way communication connection with the vehicle's intelligent driving system. This communication connection constitutes the physical channel for all information exchange between the drone and the vehicle, responsible for transmitting environmental perception data collected by the drone back to the intelligent driving system in real time, while simultaneously transmitting control commands and status information from the vehicle to the drone. The specific method of establishing the communication link can be selected according to the actual application conditions, such as using a point-to-point direct connection between the vehicle's wireless communication module and the drone's onboard communication module, or using the vehicle's existing network communication architecture for bridging and relaying. The configuration of the communication protocol and operating frequency band depends on the transmission distance, data bandwidth requirements, and the actual electromagnetic environment.

[0030] Step 102: Control the UAV to perform environmental perception tasks and acquire environmental data of non-standard scenarios; The sensor system onboard the drone carries the core data acquisition function for the entire environmental perception mission. The specific configuration of this sensor system can be flexibly configured according to the perception requirements of the target scene, and its data acquisition covers multi-dimensional environmental elements directly related to vehicle traffic safety. When performing perception tasks, the drone plans its flight path based on the spatial scale and structural characteristics of the target scene, and systematically scans the area along the predetermined path and the target region to acquire continuous and complete environmental data.

[0031] The type of environmental data varies depending on the hazard category of the non-standard scenario involved. For unknown environments such as road surface conditions, drones can use active optical detection methods to acquire information about the physical properties of the road surface. For example, a drone carrying a light source illuminates the road surface and receives reflected signals. By analyzing optical parameters such as the intensity and angle of reflected light, key geometric features such as the material state, smoothness, and slope distribution of the road surface can be inferred. For road areas with the risk of water cover or terrain collapse, drones can use physical contact detection methods to acquire depth and structural information. By deploying detection components with pre-defined connections to the water surface or suspicious areas, the depth distribution or load-bearing status of the target area can be calculated based on the physical quantities returned. When the scenario involves spatial passage conditions around vehicles, drones can measure the lateral width, longitudinal clearance height, and structural stability of the road edges one by one to form a geometric basis for judging whether vehicles can pass safely.

[0032] In scenarios where visibility is obstructed or collision risks exist, the focus of the drone's perception tasks shifts to obstacle recognition and motion monitoring. The drone flies in a pre-defined airspace in front of or behind the vehicle, identifying and classifying various objects that may affect vehicle passage. This includes the position and size of stationary obstacles, the speed and trajectory of moving objects, and the presence of potential threats behind obstructions. The drone encodes these identification results, along with its own spatial positioning information, into a structured environmental data packet, which is transmitted in real-time to the vehicle's intelligent driving system via an established communication connection. This provides information support regarding the external environment for subsequent path planning and safety decisions.

[0033] Step 103: Generate an assisted driving strategy based on the environmental data; After receiving environmental data from the drone, the vehicle's intelligent driving system analyzes and comprehensively evaluates the data, transforming the environmental characteristics into assisted driving strategies that can be directly invoked by the vehicle's execution system. The environmental data analysis process revolves around the vehicle's feasibility and safety in the current non-standard scenario. Based on the type and content of the received data, the system identifies, classifies, and analyzes the hazardous elements in the scenario, extracting constraints and feasible path information directly related to the vehicle's driving decisions.

[0034] In terms of road surface condition assessment, the system uses road surface optical feature parameters or depth structure data collected by UAVs to determine the load-bearing capacity and friction characteristics of the area the vehicle intends to pass through. When the analysis results indicate that the road surface conditions pose a risk exceeding the vehicle's safe passage threshold, the assisted driving strategy includes restrictive constraints or path correction instructions on the passage path. For example, when the combined relationship between road surface reflectivity and slope exceeds the safe range, the strategy may include a timing constraint requiring the vehicle to wait for road surface treatment before proceeding; when water depth distribution analysis indicates that the depth of a local area exceeds the vehicle's wading capacity, the strategy may generate path restriction instructions for detour or prohibition; when width or clearance measurements show insufficient space, the strategy may output control suggestions for deceleration or prohibition. The constraints are prioritized according to their impact on vehicle safety, with higher-priority constraints covering lower-priority constraints.

[0035] In the obstacle and motion risk assessment dimension, the system performs spatiotemporal correlation analysis on the positions of stationary obstacles identified by the UAV, the trajectories of moving objects, and the potential threats in obstructed areas, constructing a dynamic risk situation map of the vehicle's surroundings in the current scenario. Based on this, the assisted driving strategy generates corresponding motion control suggestions, including parameters such as speed adjustment, directional deflection, and braking intervention timing, ensuring that the vehicle maintains a preset safe distance from the identified risk elements during passage. As the UAV continuously updates environmental data during reconnaissance, the assisted driving strategy also dynamically adjusts accordingly, forming a closed-loop decision-making process that remains synchronized with the real-time state of the scene.

[0036] Once generated, the assisted driving strategy is output to the vehicle's execution control layer in a standardized data format. Its content includes not only the real-time control parameters at the current moment, but also pre-planning suggestions for the vehicle's driving status in the future journey, thereby providing phased decision support for the vehicle's continuous passage in non-standard scenarios.

[0037] Step 104: Feedback the assisted driving strategy to the vehicle to control the vehicle to perform corresponding operations.

[0038] The intelligent driving system transforms the generated driver assistance strategy into control signals that the vehicle can directly execute, transmitting them to various actuators via the vehicle's internal control bus. This feedback process involves protocol adaptation between the strategy data format and the vehicle's execution system control interface. After completing data encoding and instruction encapsulation, the intelligent driving system distributes the strategy content item by item in a manner conforming to the vehicle control bus communication specification. The type of control signal depends on the specific constraints and action commands contained in the driver assistance strategy, covering the coordinated control of various functional subsystems of the vehicle, such as the drive system, braking system, steering system, and warning devices.

[0039] During the execution of feedback control, the vehicle sequentially invokes the corresponding actuators according to the priority order of the various instructions in the driver assistance strategy. Core constraints related to driving safety have the highest execution priority; for example, braking deceleration or emergency avoidance instructions are responded to immediately by the actuators after being issued, while routine control instructions such as path correction or speed fine-tuning are executed sequentially according to the preset timing of the strategy. When the driver assistance strategy contains multiple control instructions simultaneously, the intelligent driving system arbitrates and coordinates the logical relationships between the instructions to ensure that the actions of each actuator are compatible in time and space, avoiding conflicts between control instructions or disorder in the execution timing.

[0040] During the feedback control execution, the drone continuously perceives and updates environmental data, which the intelligent driving system uses to periodically iterate and correct its assisted driving strategy. The corrected strategy is fed back to the vehicle's execution system along the same path, ensuring that the vehicle's actual driving state always matches the latest environmental situation. When the scene's environmental characteristics return to the range that the vehicle's intelligent driving system can make autonomous decisions, or when the driver actively deactivates the non-standard scene assistance mode, the system terminates the output of the assisted driving strategy and simultaneously issues a mission termination and recovery command to the drone. The drone automatically returns to the onboard take-off and landing platform and completes its storage according to the command.

[0041] In some embodiments, the non-standard scenario trigger signal includes a mode selection signal actively triggered by the user through the vehicle's interactive interface. Or a trigger signal automatically generated when vehicle sensors detect abnormal operating conditions.

[0042] One way non-standard scenario trigger signals are generated is through active commands issued by the driver via the vehicle's interface. The vehicle's human-machine interface system integrates a scenario mode selection module, presented as a graphical interface on the in-vehicle display terminal. When the driver observes non-standard characteristics in the current driving environment, such as missing road markings, abnormal road conditions, or a geographical environment significantly deviating from typical urban road conditions, they can select the corresponding assisted driving mode option on the display terminal. The driver's confirmation of this option constitutes a mode selection signal, which is transmitted via the vehicle's internal control bus to the decision-making layer of the intelligent driving system, thereby triggering the initiation of the drone-assisted driving process. As another generation method, the vehicle's intelligent driving system can automatically generate trigger signals based on real-time monitoring data from the in-vehicle sensor network. During vehicle operation, the vehicle continuously collects environmental data from visual sensors, radar sensors, and the positioning system; the intelligent driving system performs online analysis and pattern matching on this data.

[0043] When the system compares real-time environmental features with its internally stored standard scene database, if the matching degree remains below a preset threshold, or if abnormal conditions such as abnormal road structure, sudden drop in visibility, or obstacle distribution characteristics exceeding the known model range are detected, the system determines that the vehicle has entered a non-standard scenario. At this point, the intelligent driving system's decision module autonomously generates a trigger signal and activates the drone-assisted program. This automatic generation mechanism allows the vehicle to proactively introduce external sensing resources for collaborative detection at the initial stage of increased environmental uncertainty, without waiting for the driver's subjective judgment, thus improving the system's timeliness in responding to non-standard environments. The two triggering methods are logically independent. The driver's active selection is suitable for situations where the complexity of the environment is already known, while the system's automatic generation provides the ability to automatically intervene in sudden or concealed non-standard conditions. Together, they constitute a complete trigger signal generation system.

[0044] In some embodiments, controlling the drone to perform environmental perception tasks includes at least one of the following methods: The drone is controlled to fly close to the ground and emit light onto the road surface. The smoothness of the road surface is determined by detecting the road surface reflectivity. When the reflectivity exceeds a preset smoothness threshold, a road surface slope model is constructed by measuring the reflectivity at different locations and the distance between the drone and the corresponding reflective location. The drone is controlled to drop a heavy object with a pre-tensioned rope onto the target water surface. A buoyant body is installed on the rope. The water depth is determined by measuring the distance between the buoyant body and the heavy object when it sinks to the bottom. The drone moves and records the data within the water area to generate a water depth distribution map. The drone is controlled to perform a passage space measurement in narrow roads or height-restricted locations, and the measurement is compared with the vehicle's own dimensions to determine whether passage is possible.

[0045] The detection of road surface smoothness and slope characteristics relies on the active optical sensing unit (ASI) carried by the UAV. The UAV travels at a preset flight altitude close to the road surface along the target road segment at low speed. The ASI projects a specific wavelength of illumination beam onto the road surface, and the reflected echo signal from the road surface is captured by the photoelectric receiver within the same unit. The system calculates the reflectivity of the road surface at the current illumination point based on the ratio of emitted light intensity to reflected light intensity. When the detected reflectivity value exceeds a preset smoothness threshold, it indicates that the road surface is covered with ice or snow, or has a large area of ​​water accumulation, resulting in a significant reduction in the friction coefficient. The system then initiates the slope measurement process. During the slope measurement, the UAV selects multiple discrete detection points along the longitudinal and transverse directions of the road surface. Reflectivity data is collected at each detection point, and the UAV's own spatial coordinates output by the flight control system are recorded simultaneously. The differences in reflectivity at each detection point reflect the non-uniformity of the road surface material or ice thickness, while the spatial distance relationship between the UAV and each reflection point provides geometric constraints on the three-dimensional shape of the road surface. The system performs spatial fitting on the above multi-point data to establish a mathematical model of the slope of the target road segment. This model characterizes the inclination of the road surface in the longitudinal and transverse directions. When the tilt angle in any direction of the slope model and the reflectivity of the corresponding area meet specific combination conditions, such as a slope exceeding 15 degrees and a reflectivity exceeding 50%, or a slope exceeding 20 degrees and a reflectivity exceeding 40%, the system determines that there is a risk of vehicle loss of control on the road segment, and the assisted driving strategy will generate a speed limit or prohibition instruction for the corresponding road segment.

[0046] The depth of the accumulated water is measured using a physical contact detection component carried by a drone. This component includes a pre-tensioned rope connected to the drone's body at one end, a weight fixed to the other end of the rope, and buoyancy bodies spaced along the length of the rope. The density and mass of the weight are selected to allow it to sink to the bottom after entering the water, overcoming the combined effects of buoyancy and the rope's pre-tension. The volume and buoyancy of the buoyancy bodies are designed to maintain tension on the section of rope and keep it vertical while floating on the surface. When the drone flies over the target area of ​​accumulated water, it releases the weight vertically below the water surface. The rope, driven by the pre-tension, descends synchronously with the weight. Once the weight touches the bottom, it stops sinking. At this point, the rope forms a vertical tension section between the weight and the buoyancy bodies. The buoyancy bodies float on the surface while the weight sinks to the bottom. The straight-line distance between them along the rope's direction is the depth of the accumulated water at that measurement point.

[0047] The drone, carrying the detection component, moves sequentially along a pre-defined grid pattern within the flooded area, repeatedly deploying and retrieving the device at each waypoint, recording the coordinates and depth of each measurement point. After summarizing all the data, spatial interpolation is performed to generate a continuous water depth distribution map covering the entire flooded area. The system then performs path analysis on the water depth distribution map to determine if a continuous channel traverses the flooded area, ensuring that the water depth at each point along this channel does not exceed the vehicle's maximum wading depth threshold. If such a channel exists, the assisted driving strategy generates navigation path data along this channel and feeds it back to the vehicle.

[0048] Measuring the passage space is used to determine whether a vehicle can safely pass through a narrow section of road or a location with height restrictions. After the drone arrives at the target location, it uses an onboard distance sensor or visual ranging module to measure the lateral distance between the two boundaries of the narrow road. The start and end points of the ranging are aligned with the narrowest constraint surfaces on both sides of the narrow road. The obtained width data is compared with the vehicle's pre-calibrated and stored outer width. The difference between the two is the margin available on one side when the vehicle passes through the narrow road. When the sum of the margins on both sides is positive and meets the minimum safe distance requirement, the system determines that the vehicle can pass safely. The assisted driving strategy will generate navigation guidance data for driving along the centerline of the narrow road. The drone will hover and monitor above the narrow road, monitoring the changes in the distance between the vehicle body and the two side boundaries in real time during the vehicle's passage. Once the distance approaches the warning threshold, it will issue a deceleration or stopping command to the vehicle.

[0049] For measuring height restrictions, the drone adjusts the distance sensor's ranging direction to vertical and measures the clearance height from the ground to the lower edge of the height restriction structure. This height value is compared with the vehicle's pre-stored outer profile height, and the feasibility of passage is judged using the same logic as for measuring narrow road width.

[0050] In some embodiments, controlling the drone to perform an environmental perception task includes: The drone is controlled to fly in the area in front of the vehicle at an altitude not lower than a preset safe altitude, to perform obstacle recognition, and to take off and avoid the recognized obstacles. The system also controls the drone to hover above the intersection, perform potential collision object identification in a preset sequence, and after identification, fly to the next target location in advance according to the vehicle's driving direction to repeat the identification.

[0051] When the drone performs obstacle recognition tasks in the airspace in front of the vehicle, its flight altitude is controlled within a preset safe range to ensure that the drone has sufficient longitudinal obstacle avoidance margin and avoids collisions with overhead objects during low-altitude flight. The drone cruises along the vehicle's predetermined direction of travel in the area ahead, and the onboard perception module detects and classifies various obstacles appearing in the flight path in real time. The obstacle recognition range covers objects in the space above the road that may affect the passage of the vehicle or drone, such as overhead cables, fixed supports, and hanging decorations. When the drone identifies an obstacle in its flight path, the flight control system autonomously plans vertical ascent and descent maneuvers to avoid it, adjusts its flight altitude to bypass the obstacle area, and then returns to the original altitude to continue its forward reconnaissance mission. The entire avoidance process requires no manual intervention.

[0052] After the drone arrives at the intersection area, it enters intersection recognition mode, stops its horizontal movement, and hovers above the intersection. The chosen hovering position ensures that the drone's field of view completely covers all traffic directions at the intersection, providing ample spatial observation conditions for subsequent all-around impact object recognition. Impact object recognition is performed step-by-step according to a preset hierarchical sequence. First, moving objects are identified. The drone uses its onboard visual sensors to detect moving objects within the intersection area, determining whether their trajectories might intersect with the paths of vehicles about to enter the intersection. After moving object recognition, the drone extends its perception range to the ground area of ​​the intersection, scanning for ground integrity, obstacle distribution, and traffic conditions. Next, it enters the occlusion recognition stage, detecting structures or vegetation within the intersection area that might obstruct the driver's view, analyzing whether there are any undetected moving objects or fixed obstacles hidden behind the obstructed areas. If the occlusion recognition result indicates the existence of an occluded area, but the previous moving object recognition failed to detect any moving targets, the drone will lower its hovering altitude and activate its onboard acoustic warning device to emit a sound, deterring and warning pedestrians or animals that may be behind the occluded area. It will then perform another round of moving object recognition at the lowered altitude to compensate for the initial blind spot caused by the occlusion. After all recognition results are summarized, the drone transmits the comprehensive situational information of the intersection to the vehicle's intelligent driving system via a communication link.

[0053] As a vehicle approaches an intersection, the system needs to determine its upcoming direction of travel so that the drone can fly in advance to the next intersection to perform a new round of identification tasks. If the vehicle's navigation system is operational and a planned path has been established, the drone extracts the vehicle's turning intention at the current intersection from the navigation path data. If the navigation system is not activated, the drone determines the vehicle's upcoming direction by sensing the vehicle's turn signal. After the vehicle passes through the intersection, the drone, based on the determined direction of travel, flies in advance to the next intersection or the next curve, repeating the entire process from hovering at the intersection to collision object identification. While the drone is flying in advance to the next target location, the vehicle's own sensors simultaneously monitor the current path, comparing its actual trajectory with the drone's previously predicted direction of movement. If a deviation is found between the drone's predicted direction of movement and the vehicle's actual path, the system will correct the drone's flight path, realigning it with the correct forward position, thus ensuring that the drone's forward reconnaissance remains consistent with the vehicle's actual direction of travel throughout the entire process.

[0054] In some embodiments, generating the assisted driving strategy based on the environmental data includes: When the environmental data indicates that the road surface conditions meet the dangerous spraying conditions, a strategy for spraying de-icing agent onto the road surface is generated. The spraying path width is matched with the wheel track of the vehicle. After spraying is completed, the road surface conditions are checked again. When the road surface conditions improve to below the safety threshold, a driving navigation model is established. When the environmental data indicates the presence of water accumulation and a safe passage route exists in the water depth distribution map, a driving navigation model based on the safe passage route is generated; When the environmental data indicates that the passage space is passable, a driving navigation model is generated; When the environmental data indicates that the vehicle is at the edge and needs to reverse, a model of the surrounding environment is generated and the feasibility of reversing and the specific path are evaluated.

[0055] Road surface reflectivity data and slope models together form the basis for determining hazardous spraying conditions. When the road surface reflectivity test result exceeds 35%, the system initiates the slope measurement process, fitting the multi-point reflectivity and spatial distance data collected by the UAV to establish a mathematical model of the slope of the target road section. The system jointly determines the slope angle in the slope model with the reflectivity value of the corresponding area, and identifies the road surface as being under hazardous spraying conditions if any of the following combinations of conditions are met: slope exceeding 15 degrees and reflectivity exceeding 50%; slope exceeding 20 degrees and reflectivity exceeding 40%; slope exceeding 25 degrees and reflectivity exceeding 35%; or reflectivity alone exceeding 70%. After the determination is completed, the assisted driving strategy generates a de-icing agent spraying command, and the de-icing agent storage and release device carried by the UAV is activated.

[0056] The spraying operation proceeds longitudinally along the road surface, with a spray width set at 25 centimeters. The drone performs one spray pass each on the left and right sides of the road surface, ensuring the center distance between the two spray paths is equal to the wheelbase of the target vehicle. This allows the de-icing agent to precisely cover the area where the vehicle's wheels will soon be driven over. After spraying on both sides is completed, the drone performs a second reflectivity test on the treated road surface, comparing the new reflectivity data with the baseline data before spraying. When the reflectivity drops below a preset safety threshold, it indicates that the road surface friction conditions have improved to a level where vehicles can drive safely. Based on this, the system establishes a driving navigation model for that section of road and feeds back the coordinates of the safe driving trajectory confirmed by the drone to the vehicle's navigation system. The vehicle then drives along this trajectory through the treated road surface area.

[0057] In flooded areas, the drone has obtained a complete water depth distribution map of the flooded area through physical contact detection. Each grid node in the map is marked with a measured water depth value. The assisted driving strategy generation module performs path search analysis on the water depth distribution map. Using the vehicle's maximum wading depth as a constraint, it searches the distribution map for a continuous channel from the near end to the far end of the flooded area. The channel must have a water depth at all nodes that does not exceed the vehicle's wading capacity limit, and the total length of the channel should be as short as possible with minimal turning. If the search algorithm finds a feasible channel in the water depth distribution map that satisfies all the above constraints, the assisted driving strategy generates a driving navigation model based on the centerline coordinate sequence of that channel. The precise direction of the channel, the water depth margin at each key node, and the recommended crossing speed are encoded as navigation data and transmitted back to the vehicle, which then uses this data to safely cross the flooded area. If the search fails to find a channel that meets the constraints, the strategy will generate a no-passage or detour suggestion.

[0058] After measuring the passage space, the system subtracts the measured width of the narrow road or the height restriction clearance from the vehicle's pre-stored dimensions. When the margin in both the narrow road and height restriction directions is not less than the corresponding minimum safe clearance threshold, the assisted driving strategy determines that the location is passable and generates a driving navigation model based on the centerline of the narrow road or height restriction location. This navigation model includes lateral position control parameters during vehicle passage, ensuring the vehicle remains in the central area of ​​the passage space with equal safety margins on both sides. The drone continuously monitors from above during vehicle passage; if it detects that the dynamic change in the distance between the vehicle body and the constraint surface approaches a warning value, the navigation model immediately issues lateral correction or deceleration commands to the vehicle.

[0059] When environmental data indicates that the vehicle is near an edge such as a cliff or ditch and needs to reverse, the drone first performs a comprehensive spatial scan of a large area around the vehicle, collecting data on terrain undulations, road width, and the distribution of surrounding obstacles to construct a surrounding environment model covering the vehicle's reversing path. Based on this environmental model, the system plans multiple candidate paths in the reversing direction. For each path, it evaluates whether its turning radius meets the vehicle's maneuverability, whether there are obstacles interfering on either side of the path, and whether there is safe parking space at the end of the path. The path with the highest overall score is then selected as the recommended reversing plan. During the reversing process, the drone hovers above the edge, continuously detecting the stability of the road edge structure behind the vehicle. The drone uses onboard sensors to monitor surface displacement or vibration in the edge area. When it detects signs of localized collapse, loose gravel, or other structural instability at the road edge, it immediately transmits the abnormal status to the vehicle. The assisted driving strategy then interrupts the current reversing plan and reassesses the surrounding environment model, planning a new reversing position and path for the vehicle.

[0060] In some embodiments, feeding back the driver assistance strategy to the vehicle includes: The driving navigation model is transmitted to the vehicle's intelligent driving system, which controls the vehicle to drive along the path indicated by the driving navigation model, and the drone continuously performs environmental monitoring during the driving process.

[0061] The navigation model is encoded in a standardized data format, containing the geographical coordinate sequence of the path, recommended driving speeds at each waypoint, safety distance constraints on both sides of the path, and risk labeling information at key locations. The encoded navigation data packet is transmitted to the navigation data receiving port of the vehicle's intelligent driving system via the established communication link between the UAV and the vehicle. Upon receiving the navigation data packet, the intelligent driving system parses the coordinate sequence and constraint parameters, converting the path coordinates into a reference trajectory in the vehicle's local coordinate system. Based on the constraints of each path segment, it generates a sequence of underlying execution commands to coordinate the vehicle's steering mechanism, drive system, and braking system, ensuring that the lateral deviation between the vehicle's actual driving trajectory and the reference trajectory indicated by the navigation model remains within acceptable limits. While the vehicle travels along the navigation path, the UAV does not move synchronously with the vehicle but remains in flight or hovering in a pre-defined airspace in front of or above the vehicle, continuously monitoring the area the vehicle is about to enter using its onboard perception module. The monitoring range covers changes in road surface conditions ahead of the navigation path, newly appearing obstacles, the stability of curb structures, and the movement trends of surrounding dynamic objects.

[0062] The drone transmits incremental environmental data acquired during surveillance back to the intelligent driving system in real time via streaming. The system dynamically compares this data with the path constraints in the current navigation model. If the incremental data shows a new risk on the path ahead that was not present when the navigation model was established, such as a new local road collapse or a change in depth due to rising water levels, the intelligent driving system corrects the navigation model online based on the latest surveillance data. This adjusts the direction or speed parameters of the subsequent path, and the corrected instructions are simultaneously sent to the actuators to ensure that the vehicle's driving status remains adapted to the real-time environment. This cycle of monitoring and correction continues throughout the entire journey along the navigation path until the vehicle safely passes the target area.

[0063] In some embodiments, it also includes: Before generating the assisted driving strategy, the drone is controlled to hover in the air to detect wind and adjust its own state to maintain stable communication. When communication cannot be maintained stably or when it is not possible to establish connections with both the satellite and the vehicle simultaneously, an information transmission algorithm is executed to encode the information content of the user terminal and the recipient information and transmit it to the UAV. The UAV is then controlled to take off and search for a stable signal source to transmit the encoded information to the external recipient and receive a reply. Subsequently, the UAV flies back and feeds the reply information back to the user.

[0064] While hovering, the UAV activates its onboard wind detection module to measure the wind speed, direction, and wind force trends in the current airspace in real time. The data collected by the wind detection module is input into the attitude control loop of the flight control system. The flight control system calculates the attitude compensation required for hovering based on the real-time wind parameters and drives each rotor motor to independently adjust its speed, enabling the UAV to maintain a stable hovering position and fuselage attitude under wind disturbances. The criterion for attitude stability is that the UAV's positional deviation in all three axes does not exceed a preset allowable range, and this condition is continuously met within a continuous sampling period. When the wind detection results indicate that the wind speed or wind force fluctuation in the current airspace exceeds the compensation capability of the UAV's attitude control system, causing the UAV to be unable to maintain hovering accuracy within the preset allowable range, the UAV autonomously performs horizontal displacement maneuvers to search for a location with more stable wind conditions in the surrounding airspace. The UAV sequentially samples wind force and tests attitude stability at candidate hovering points until a hovering position that meets communication stability requirements is found. Then, it resumes hovering and reconfirms the communication link quality with the vehicle's intelligent driving system and the user terminal.

[0065] When a drone, after repeated adjustments during hovering, fails to maintain a stable communication link quality above the operating threshold, or when the drone's location prevents simultaneous bidirectional connections with both the satellite communication system and the vehicle's intelligent driving system, the system automatically switches to a message relay mode. In this mode, the user edits the message content to be sent and specifies the recipient's identity via a terminal device. The terminal device then submits the message content and recipient's identity to the communication protocol layer for encoding. The encoding process converts the raw text information into structured data codes, embedding information type markers to distinguish between emergency assistance messages and routine communication messages. Emergency assistance codes include fields for the number of personnel on site, each person's identity code, current geographic coordinates and altitude, terrain type, required supplies, type of hazard encountered, and the personnel's travel status. Routine communication codes include fields for the number of personnel, recipient's identity, geographic coordinates, terrain information, and direction of travel. Both code types maintain a unified frame format in their encoding structure, differing only in the identifier value of the information type marker field to facilitate automatic classification by the receiving end.

[0066] After encoding, the terminal device transmits the code data to the drone via the vehicle's intelligent driving system. Upon receiving the code, the drone releases its current hovering state and begins a wide-range high-altitude flight, continuously scanning the surrounding airspace for satellite signal strength. When the drone reaches a location where the satellite signal strength meets the data transmission requirements, it initiates transmission of the stored code data to a remote receiver via a satellite communication link. The code data arrives at the receiver in digital signal form after being relayed via the satellite link. The receiver decodes the code, restoring the structured data to readable text information, and classifies and forwards it according to the information type markers and recipient identifiers in the code. For emergency assistance codes, the decoded text information is automatically pushed to the nearest receiving terminal of the public security or fire department; for routine communication codes, the decoded information is forwarded to the private recipient specified in the code. The receiver has uninterrupted operation capabilities, maintaining a continuous monitoring state during brief disconnections due to signal instability. Once the drone re-establishes its satellite connection, data reception can be resumed immediately, without requiring the drone to remain airborne for extended periods.

[0067] After forwarding the emergency distress message, the receiver continuously receives response codes from the rescue system. These codes contain the estimated arrival time of the rescue team, their mode of transport, and a list of supplies they will be carrying. The receiver continuously transmits this response code to the drone via satellite link, 24 hours a day, to ensure the drone can capture the response data on its next signal search. After completing the satellite transmission of the coded information, the drone returns to the airspace near the vehicle and takes off again when the next communication opportunity arises to search for satellite signals and receive the response code forwarded by the receiver. The drone then transmits the received response code back to the vehicle's intelligent driving system, which presents the decoded response to the driver via the in-vehicle display terminal or audio channel. When the response includes the estimated arrival time of the rescue team, the drone automatically records the time period parameters and sets a timed task. After driver confirmation, it autonomously takes off within the designated time period, releasing an emergency beacon signal operating in the UHF band to provide precise location guidance for the rescue team.

[0068] Corresponding to the aforementioned drone-assisted driving method for non-standard scenarios, this invention also proposes a drone-assisted driving device for non-standard scenarios. Since the device embodiments of this invention correspond to the aforementioned method embodiments, details not disclosed in the device embodiments can be referred to in the aforementioned method embodiments, and will not be repeated here.

[0069] Figure 2 This is a schematic diagram of the structure of a drone assisted driving device in a non-standard scenario provided by an embodiment of this disclosure, such as... Figure 2 As shown, it includes: The start-up unit 21 is used to start the drone and establish a communication connection with the vehicle in response to a non-standard scenario trigger signal; Control unit 22 is also used to control the UAV to perform environmental perception tasks and acquire environmental data of non-standard scenarios; The generation unit 23 is also used to generate an assisted driving strategy based on the environmental data; Feedback unit 24 is also used to feed back the assisted driving strategy to the vehicle in order to control the vehicle to perform corresponding operations.

[0070] Furthermore, in one possible implementation of this disclosure embodiment, the non-standard scenario trigger signal includes a mode selection signal actively triggered by the user through the vehicle interaction interface, or a trigger signal automatically generated when the vehicle sensors detect abnormal operating conditions.

[0071] Furthermore, in one possible implementation of this disclosure, the control unit 22 is further configured to: The drone is controlled to fly close to the ground and emit light onto the road surface. The smoothness of the road surface is determined by detecting the road surface reflectivity. When the reflectivity exceeds a preset smoothness threshold, a road surface slope model is constructed by measuring the reflectivity at different locations and the distance between the drone and the corresponding reflective location. The drone is controlled to drop a heavy object with a pre-tensioned rope onto the target water surface. A buoyant body is installed on the rope. The water depth is determined by measuring the distance between the buoyant body and the heavy object when it sinks to the bottom. The drone moves and records the data within the water area to generate a water depth distribution map. The drone is controlled to perform a passage space measurement in narrow roads or height-restricted locations, and the measurement is compared with the vehicle's own dimensions to determine whether passage is possible.

[0072] Furthermore, in one possible implementation of this disclosure, the control unit 22 is further configured to: The drone is controlled to fly in the area in front of the vehicle at an altitude not lower than a preset safe altitude, to perform obstacle recognition, and to take off and avoid the recognized obstacles. The system also controls the drone to hover above the intersection, perform potential collision object identification in a preset sequence, and after identification, fly to the next target location in advance according to the vehicle's driving direction to repeat the identification.

[0073] Furthermore, in one possible implementation of this disclosure embodiment, the generation unit 23 is further configured to: When the environmental data indicates that the road surface conditions meet the dangerous spraying conditions, a strategy for spraying de-icing agent onto the road surface is generated. The spraying path width is matched with the wheel track of the vehicle. After spraying is completed, the road surface conditions are checked again. When the road surface conditions improve to below the safety threshold, a driving navigation model is established. When the environmental data indicates the presence of water accumulation and a safe passage route exists in the water depth distribution map, a driving navigation model based on the safe passage route is generated; When the environmental data indicates that the passage space is passable, a driving navigation model is generated; When the environmental data indicates that the vehicle is at the edge and needs to reverse, a model of the surrounding environment is generated and the feasibility of reversing and the specific path are evaluated.

[0074] Furthermore, in one possible implementation of this disclosure, the feedback unit 24 is further configured to: The driving navigation model is transmitted to the vehicle's intelligent driving system, which controls the vehicle to drive along the path indicated by the driving navigation model, and the drone continuously performs environmental monitoring during the driving process.

[0075] Furthermore, in one possible implementation of the embodiments of this disclosure, such as Figure 3As shown, it also includes: The detection unit 25 is used to control the UAV to hover in the air to detect wind and adjust its own state before generating the assisted driving strategy, and to maintain stable communication. The transmission unit 26 is also used to execute an information transmission algorithm when communication cannot be maintained stably or when it is not possible to establish a connection with the satellite and the vehicle at the same time. The algorithm encodes the information content of the user terminal and the information of the receiver and transmits it to the UAV. The UAV is then controlled to take off and search for a stable signal source to transmit the encoded information to the external receiver and receive the reply information. The UAV then flies back and feeds the reply information back to the user.

[0076] It should be noted that the foregoing explanation of the method embodiments also applies to the apparatus of the embodiments of this disclosure, and the principle is the same. Therefore, the embodiments of this disclosure are not limited thereto.

[0077] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0078] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0079] like Figure 4 As shown, device 400 includes a computing unit 401, which can perform various appropriate actions and processes based on a computer program stored in ROM (Read-Only Memory) 402 or a computer program loaded from storage unit 408 into RAM (Random Access Memory) 403. RAM 403 may also store various programs and data required for the operation of device 400. The computing unit 401, ROM 402, and RAM 403 are interconnected via bus 404. I / O (Input / Output) interface 405 is also connected to bus 404.

[0080] Multiple components in device 400 are connected to I / O interface 405, including: input unit 406, such as keyboard, mouse, etc.; output unit 407, such as various types of monitors, speakers, etc.; storage unit 408, such as disk, optical disk, etc.; and communication unit 409, such as network card, modem, wireless transceiver, etc. Communication unit 409 allows device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0081] The computing unit 401 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, CPUs (Central Processing Units), GPUs (Graphics Processing Units), various special-purpose AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, DSPs (Digital Signal Processors), and any suitable processor, controller, microcontroller, etc. The computing unit 401 performs the various methods and processes described above, such as drone-assisted driving methods in non-standard scenarios. For example, in some embodiments, drone-assisted driving methods in non-standard scenarios can be implemented as computer software programs tangibly contained in a machine-readable medium, such as storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by the computing unit 401, one or more steps of the methods described above can be performed. Alternatively, in other embodiments, computing unit 401 may be configured to perform the aforementioned unmanned aerial vehicle assisted driving method in non-standard scenarios by any other suitable means (e.g., by means of firmware).

[0082] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application-Specific Standard Products), SOCs (System-on-Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0083] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0084] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, EPROM (Electrically Programmable Read-Only Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0085] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0086] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include LANs (Local Area Networks), WANs (Wide Area Networks), the Internet, and blockchain networks.

[0087] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0088] It's important to note that artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily include computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graph technologies.

[0089] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0090] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for assisted driving of unmanned aerial vehicles in non-standard scenarios, characterized in that, include: In response to a non-standard scenario trigger signal, the drone is activated and establishes a communication connection with the vehicle. Control the drone to perform environmental perception tasks and acquire environmental data from non-standard scenarios; A driving assistance strategy is generated based on the environmental data; The driver assistance strategy is fed back to the vehicle to control the vehicle to perform corresponding operations.

2. The method according to claim 1, characterized in that, The non-standard scenario trigger signals include mode selection signals actively triggered by the user through the vehicle's interactive interface. Or a trigger signal automatically generated when vehicle sensors detect abnormal operating conditions.

3. The method according to claim 1, characterized in that, Controlling the UAV to perform environmental perception tasks includes at least one of the following methods: The drone is controlled to fly close to the ground and emit light onto the road surface. The smoothness of the road surface is determined by detecting the road surface reflectivity. When the reflectivity exceeds a preset smoothness threshold, a road surface slope model is constructed by measuring the reflectivity at different locations and the distance between the drone and the corresponding reflective location. The drone is controlled to drop a heavy object with a pre-tensioned rope onto the target water surface. A buoyant body is installed on the rope. The water depth is determined by measuring the distance between the buoyant body and the heavy object when it sinks to the bottom. The drone moves and records the data within the water area to generate a water depth distribution map. The drone is controlled to perform a passage space measurement in narrow roads or height-restricted locations, and the measurement is compared with the vehicle's own dimensions to determine whether passage is possible.

4. The method according to claim 1, characterized in that, The control of the UAV to perform environmental perception tasks includes: The drone is controlled to fly in the area in front of the vehicle at an altitude not lower than a preset safe altitude, to perform obstacle recognition, and to take off and avoid the recognized obstacles. The system also controls the drone to hover above the intersection, perform potential collision object identification in a preset sequence, and after identification, fly to the next target location in advance according to the vehicle's driving direction to repeat the identification.

5. The method according to claim 1, characterized in that, The step of generating the assisted driving strategy based on the environmental data includes: When the environmental data indicates that the road surface conditions meet the dangerous spraying conditions, a strategy for spraying de-icing agent onto the road surface is generated. The spraying path width is matched with the wheel track of the vehicle. After spraying is completed, the road surface conditions are checked again. When the road surface conditions improve to below the safety threshold, a driving navigation model is established. When the environmental data indicates the presence of water accumulation and a safe passage route exists in the water depth distribution map, a driving navigation model based on the safe passage route is generated; When the environmental data indicates that the passage space is passable, a driving navigation model is generated; When the environmental data indicates that the vehicle is at the edge and needs to reverse, a model of the surrounding environment is generated and the feasibility of reversing and the specific path are evaluated.

6. The method according to claim 1, characterized in that, The step of feeding back the driver assistance strategy to the vehicle includes: The driving navigation model is transmitted to the vehicle's intelligent driving system, which controls the vehicle to drive along the path indicated by the driving navigation model, and the drone continuously performs environmental monitoring during the driving process.

7. The method according to claim 1, characterized in that, Also includes: Before generating the assisted driving strategy, the drone is controlled to hover in the air to detect wind and adjust its own state to maintain stable communication. When communication cannot be maintained stably or when it is not possible to establish connections with both the satellite and the vehicle simultaneously, an information transmission algorithm is executed to encode the information content of the user terminal and the recipient information and transmit it to the UAV. The UAV is then controlled to take off and search for a stable signal source to transmit the encoded information to the external recipient and receive a reply. Subsequently, the UAV flies back and feeds the reply information back to the user.

8. A drone-assisted piloting device for non-standard scenarios, characterized in that, include: The start-up unit is used to respond to trigger signals in non-standard scenarios, start the drone, and establish a communication connection with the vehicle; The control unit is also used to control the UAV to perform environmental perception tasks and acquire environmental data of non-standard scenarios; The generation unit is also used to generate an assisted driving strategy based on the environmental data; The feedback unit is also used to feed back the assisted driving strategy to the vehicle in order to control the vehicle to perform corresponding operations.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.

Images