Mountain road driving assistance method and device and vehicle

Abstract

The invention provides a mountainous road driving assistance method and device and a vehicle, and relates to the technical field of vehicles. The method is based on a vehicle carrying an unmanned aerial vehicle. The method comprises the following steps: in response to a mountain road navigation instruction, controlling the unmanned aerial vehicle to acquire traffic information about a mountain road where the vehicle is located; constructing a three-dimensional terrain model about the mountain road according to the traffic information; and generating driving path suggestion information and / or early warning information according to vehicle positioning information and the three-dimensional terrain model. On the basis of the vehicle carrying the unmanned aerial vehicle, the traffic information of the mountain road where the vehicle is located is obtained timely and accurately, the three-dimensional terrain model about the mountain road is constructed, and the driving path suggestion information and / or early warning information are / is generated, so that active guidance and risk prompt for the driving behavior of a driver are achieved, and the driving safety of the driver is improved. And the driving safety of the vehicle on the mountainous road and the driving experience of a driver are improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and more specifically, to a driving assistance method and device for mountain roads, and a vehicle. Background Technology

[0002] With the continuous growth of car ownership, traffic safety issues on mountain roads are becoming increasingly prominent. Mountain roads are generally characterized by dense curves, large gradients, significant terrain undulations, and severe obstruction by mountains, which pose challenges to drivers, such as limited visibility, difficulty in prediction, and narrow operating space, making traffic accidents highly likely and seriously threatening driving safety.

[0003] Currently, while some navigation systems have integrated oncoming vehicle warning functions for mountain curves, such as cloud-based warnings, these functions typically rely on oncoming vehicles simultaneously using the same navigation service and require stable communication signals. This limits their practical application and makes it difficult to cover all road users. Furthermore, existing in-vehicle driver assistance systems (such as lane keeping assist, adaptive cruise control, and emergency braking) largely depend on onboard cameras and radar sensors. Their perception range and accuracy are limited by the vehicle's installation location and the surrounding environment, making them particularly ineffective in mountainous areas with severely obstructed visibility. In addition, traditional navigation systems are primarily based on static map data and limited real-time traffic information. Their low update frequency and poor dynamic perception capabilities make them ill-equipped to handle common emergencies on mountain roads, such as landslides, rockfalls, temporary construction, and vehicles stranded ahead. Therefore, drivers often lack sufficient reaction time and decision-making support when facing complex road conditions, resulting in significant safety hazards. Summary of the Invention

[0004] The problem this invention addresses is: how to improve the safety of vehicles driving on mountain roads.

[0005] To address the aforementioned problems, this invention provides a driving assistance method, device, and vehicle for mountain roads.

[0006] In a first aspect, the present invention provides a driving assistance method for mountain roads, based on a vehicle equipped with an unmanned aerial vehicle (UAV); the driving assistance method for mountain roads includes: In response to a navigation command for mountain roads, the drone is controlled to acquire traffic information about the mountain roads where the vehicle is located; Based on the traffic information, construct a three-dimensional terrain model of the mountain roads; Based on the vehicle location information and the three-dimensional terrain model, driving route suggestion information and / or warning information are generated.

[0007] Optionally, the step of controlling the drone to acquire traffic information about the mountain road where the vehicle is located in response to a mountain road navigation command includes: In response to a prompt message about the vehicle entering a mountainous road, or in response to a navigation trigger command input by the driver, the mountainous road navigation command is generated; wherein the prompt message is generated by a navigation system applied to the vehicle; the driver includes at least one of the vehicle's user and the driver assistance control system.

[0008] Optionally, the drone is equipped with an image acquisition mechanism and a spatial position measurement mechanism; controlling the drone to acquire traffic information about the mountain roads where the vehicle is located includes: The drone is controlled to acquire road condition images of the mountain road within a preset distance range in front of the vehicle via the image acquisition mechanism; and spatial feature data of the mountain road within a preset distance range in front of the vehicle is measured via the spatial position measurement mechanism.

[0009] Optionally, the mountain road driving assistance method further includes: The preset distance range is determined based on the first operating parameters of the vehicle and the second operating parameters of the UAV; wherein the preset distance range includes a preset upper limit value and a preset lower limit value; when the first operating parameters meet the risk operation conditions of the vehicle, the preset upper limit value is increased; when the second operating parameters meet the condition of UAV functional decline, the preset lower limit value is decreased.

[0010] Optionally, after constructing a three-dimensional terrain model of the mountain road based on the traffic information, the mountain road driving assistance method further includes: The three-dimensional terrain model is updated based on the latest traffic information and vehicle positioning information acquired by the UAV.

[0011] Optionally, generating driving route suggestion information and / or warning information based on vehicle positioning information and the three-dimensional terrain model includes: Based on the vehicle positioning information and the three-dimensional terrain model, dangerous targets are identified, and the identified dangerous targets are classified into dangerous types. The warning information is generated based on the segmentation results; And / or, the step of generating driving route suggestion information and / or warning information based on vehicle positioning information and the three-dimensional terrain model includes: Based on the vehicle positioning information and the three-dimensional terrain model, an initial path for the vehicle to travel is determined; The initial path is subjected to risk assessment and path optimization to generate the driving path suggestion information.

[0012] Optionally, generating the early warning information based on the division result includes: The warning information corresponding to the division result is generated and output through the vehicle's display mechanism and / or audio interaction mechanism; wherein, the division result includes at least one of oncoming vehicle risk, obstacle risk, road damage risk, and rockfall risk.

[0013] Optionally, the step of performing risk assessment and route optimization on the initial path to generate the driving route suggestion information includes: The initial path is subjected to risk assessment and path optimization to determine the optimal path for the vehicle to travel, and the driving path suggestion information corresponding to the optimal path is generated. The driving path suggestion information is output through the vehicle's display mechanism and / or audio interaction mechanism; wherein, the driving path suggestion information includes navigation information.

[0014] Secondly, the present invention provides a driving assistance device for mountain roads, comprising: The navigation response unit is used to respond to navigation commands on mountain roads and control the drone to acquire traffic information about the mountain roads where the vehicle is located. The model building unit is used to build a three-dimensional terrain model of the mountain roads based on the traffic information. The information generation unit is used to generate driving route suggestion information and / or warning information based on the vehicle positioning information and the three-dimensional terrain model.

[0015] Thirdly, the present invention provides a vehicle including a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the mountain road driving assistance method as described in the first aspect when executing the computer program.

[0016] The beneficial effects of the mountain road driving assistance method, device, and vehicle of the present invention are as follows: Based on a vehicle equipped with a drone, the present invention controls the drone to take off and detect the mountain road where the vehicle is located by responding to a mountain road navigation command. This obtains traffic information about the mountain road, enabling timely and accurate acquisition of traffic information and effectively overcoming safety hazards caused by limited visibility, unpredictable road conditions, and difficulty in warning of sudden risks while driving on mountain roads. By constructing a three-dimensional terrain model of the mountain road based on traffic information, the geometric spatial structure and environmental characteristics of the mountain road are realistically and intuitively reflected to the driver, giving the driver a clear spatial understanding of the undulations and potential risks of the mountain road, thereby enabling early perception and risk prevention in blind spots or complex areas. By generating driving route suggestions and / or warnings based on vehicle positioning information and 3D terrain models, the system aims to proactively guide drivers' behavior and provide risk alerts, thereby improving the scientific rigor, timeliness, and safety of driving decisions, and enhancing vehicle safety and the driver's experience on mountainous roads. Furthermore, by directly providing users with driving route suggestions and / or warnings, the system avoids the cognitive burden and decision-making difficulties that might result from directly providing users with intermediate information (such as unprocessed traffic information). This reduces the cognitive burden and judgment costs for users dealing with complex environmental data during driving, thus enhancing the intuitiveness and practicality of information interaction, helping users quickly understand and adopt relevant suggestions, and further strengthening vehicle intelligence and the driver's experience. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a driving assistance method for mountain roads in an embodiment of the present invention; Figure 2 This is a structural block diagram of the mountain road driving assistance device in an embodiment of the present invention; Figure 3 This is a schematic diagram of the communication connection between the vehicle's memory and processor in an embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in sequences other than those illustrated or described herein.

[0020] Combination Figure 1As shown, this embodiment of the invention provides a driving assistance method for mountain roads, based on a vehicle equipped with an unmanned aerial vehicle (UAV); the driving assistance method for mountain roads includes: Step 100: In response to the mountain road navigation command, control the drone to acquire traffic information about the mountain road where the vehicle is located.

[0021] Specifically, in step 100, when a relevant instruction (referred to as a mountain road navigation instruction) is received to trigger the drone to take off and perform a mountain road navigation (or exploration) task, if the driver (including the vehicle user and the driver assistance control system) actively triggers the instruction or the vehicle automatically generates the instruction when the corresponding conditions are met, the instruction is responded to, the drone is controlled to take off and acquire (or collect) traffic information (such as road condition information, oncoming vehicle information, and surrounding environmental characteristics of the mountain road) of the vehicle, thereby providing corresponding data support for the vehicle's driving decision (or the driver's driving decision) on the mountain road in which it is located.

[0022] Step 200: Based on traffic information, construct a three-dimensional terrain model of mountain roads.

[0023] Specifically, in step 200, based on the traffic information of the mountain road where the vehicle is located obtained in step 100, a three-dimensional terrain model of the mountain road (partial or complete) is constructed. This model can be displayed intuitively in a three-dimensional view through a display mechanism installed in the vehicle, providing the driver with a real-time and intuitive display of the spatial morphology and potential risk information of the mountain road where the vehicle is located. This provides visual assistance and risk warning support for the driver to ensure the safety of driving on mountain roads. For example, raw data about the mountain road where the vehicle is located can be obtained by drone, such as relevant image information, distance measurement data, and elevation information. This data can be processed accordingly to generate traffic information including road condition information, oncoming vehicle information, and information on the surrounding environment of the mountain road. By employing image recognition and point cloud reconstruction and other related technologies, the acquired traffic information is fused and processed to identify and determine the spatial outline structure of the mountain road where the vehicle is located. Based on this, a three-dimensional terrain model of the mountain road is constructed, thereby presenting the driver with the spatial distribution characteristics, terrain undulations, and potential risk information related to the current driving status of the corresponding mountain road through the three-dimensional terrain model.

[0024] Step 300: Based on the vehicle positioning information and the three-dimensional terrain model, generate driving route suggestion information and / or warning information.

[0025] Specifically, in step 300, based on the three-dimensional terrain model constructed in step 200 and the vehicle positioning information (which can be obtained through an onboard positioning device, such as an onboard GNSS module or inertial navigation system), the spatial characteristics and dynamic risk factors of the mountainous road where the vehicle is located are comprehensively analyzed to generate driving path suggestion information and / or warning information. The driving path suggestion information indicates the recommended driving trajectory, speed, lane change timing, or detour plan for the vehicle in the current or upcoming road segment. The warning information alerts the driver to potential risk factors ahead, such as oncoming vehicles, narrow roads, sharp bends, continuous downhill slopes, rockfalls, landslide areas, and blind spots. The generated driving path suggestion information and warning information can be displayed in a graphical interface through a display mechanism installed in the vehicle, and / or broadcast via voice through an audio interaction mechanism installed in the vehicle, ensuring that the driver can receive key information in a timely and intuitive manner, thereby improving driving decision-making efficiency and driving safety.

[0026] In summary, this embodiment's method, based on a vehicle equipped with a drone, controls the drone to take off and detect the mountain roads where the vehicle is located in response to navigation commands. This allows for the timely and accurate acquisition of traffic information on the mountain roads, effectively overcoming safety hazards caused by limited visibility, unpredictable road conditions, and difficulty in warning of sudden risks while driving on mountain roads. By constructing a three-dimensional terrain model of the mountain roads based on traffic information, the method provides drivers with a realistic and intuitive understanding of the geometric spatial structure and environmental characteristics of the mountain roads. This enables drivers to have a clear spatial awareness of the undulations and potential risks of the mountain roads, thereby achieving early perception and risk prevention in blind spots or complex areas. By generating driving route suggestions and / or warnings based on vehicle positioning information and 3D terrain models, this approach aims to proactively guide drivers' behavior and provide risk alerts, thereby improving the scientific rigor, timeliness, and safety of driving decisions, enhancing vehicle safety on mountainous roads, and improving the driver's experience. Furthermore, by directly providing users with driving route suggestions and / or warnings, this approach avoids the cognitive burden and decision-making difficulties that might result from directly providing users with intermediate information (such as unanalyzed traffic information). It reduces the cognitive burden and judgment costs for users dealing with complex environmental data during driving, thus improving the intuitiveness and practicality of information interaction, helping users quickly understand and adopt corresponding suggestions, and further enhancing the intelligence of vehicles using this method and the driver's experience.

[0027] Furthermore, by equipping vehicles with drones and employing the method described in this embodiment, vehicles can proactively overcome the blind spots caused by the limitations of traditional vehicle-mounted sensors due to vehicle structure and terrain obstruction. This enables beyond-line-of-sight environmental perception even when obstructed by curves, slopes, and mountains, effectively expanding the range of forward road information available to the driver and significantly improving predictive capabilities. Simultaneously, the traffic information acquired by the drones can be transmitted back quickly and used to construct real-time 3D terrain models. Compared to navigation systems that rely on static maps or infrequent updates, this significantly improves the real-time performance and accuracy of mountain road environmental perception, overcoming the problem of information lag. Moreover, by combining traffic information with the 3D terrain model to generate driving path suggestions and warning information, regardless of whether oncoming vehicles are using the same navigation system, potential risks (such as falling rocks, landslides, and blind bends) can be proactively identified, and drivers can be promptly alerted to take appropriate actions, effectively compensating for the shortcomings of related technologies in terms of safety warning coverage and advance warning capabilities. Therefore, the method in this embodiment constructs a closed-loop auxiliary system of dynamic perception, spatial modeling, and intelligent decision-making, which can provide more comprehensive, timely, and accurate driving assistance support in complex mountain road environments, significantly improving driving safety and adaptability, and has good practical application value and promotion prospects.

[0028] Optionally, vehicle location information can be used to present information such as the vehicle's real-time location and direction of travel.

[0029] Optionally, step 100 includes: In response to a prompt message about the vehicle entering a mountainous road, or in response to a navigation trigger command input by the driver, a mountainous road navigation command is generated; wherein the prompt message is generated by a navigation system applied to the vehicle; the driver includes at least one of the vehicle's user and the driver assistance control system.

[0030] Specifically, before the drone mounted on the vehicle can take off and perform navigation tasks on mountain roads, it needs to receive a corresponding navigation command for mountain roads. For example, this could be triggered when the vehicle's navigation system (such as an in-vehicle navigation system or an external navigation system connected to the vehicle) detects that the vehicle has entered a mountainous area, automatically generating a notification message and thus triggering the navigation command. For instance, the navigation system could determine whether the vehicle is currently in a mountainous area based on its real-time location and road type; if so, it would output the corresponding notification message. Alternatively, the command could be triggered when the driver (such as the vehicle's user or at least one of the driver assistance system's inputs the navigation trigger command. This could be achieved by the user performing input through the vehicle's human-machine interface (such as a voice interface, touchscreen interface, or physical buttons), or by the driver assistance system automatically generating the command based on its operating status and environmental perception during driving. This would allow the drone to provide navigation services on mountain roads.

[0031] Thus, by introducing a process for generating mountain road navigation instructions before controlling the drone to navigate mountain roads, and supporting triggering methods such as automatic triggering based on the vehicle navigation system and active input by the driver, the method in this embodiment has higher scene adaptability and triggering flexibility.

[0032] Optionally, the drone is equipped with an image acquisition mechanism and a spatial positioning measurement mechanism; controlling the drone to acquire traffic information about the mountain roads where the vehicle is located includes: The drone is controlled to acquire road condition images of the mountainous road within a preset distance range in front of the vehicle via an image acquisition mechanism; and to measure spatial feature data of the mountainous road within a preset distance range in front of the vehicle via a spatial position measurement mechanism.

[0033] Specifically, the UAV is equipped with an image acquisition mechanism and a spatial position measurement mechanism to achieve multi-source information perception of mountainous road areas (or road sections) within a preset distance range in front of the vehicle, thereby enhancing the ability to identify and model complex road conditions. The image acquisition mechanism includes a high-resolution camera, infrared thermal imaging equipment, and a multispectral imaging module, used to acquire road condition images of mountainous roads, such as lane layout, traffic markings, vehicles, temporary obstacles, and landslide areas. The spatial position measurement mechanism includes a LiDAR, infrared ranging sensor, structured light sensor, ToF depth camera, and binocular stereo vision system, used to acquire three-dimensional spatial structure data of the mountainous road surface and its surrounding environment, such as terrain undulations, road curvature, slope height, obstacle size and relative position, and other spatial feature data.

[0034] The drone detection area (i.e., the area where the drone acquires traffic information) is a preset distance range in front of the vehicle on the mountainous road where the vehicle is located. On the one hand, this ensures that the drone detection area is highly correlated with the actual travel path of the vehicle, avoiding ineffective detection of irrelevant road sections, improving resource utilization efficiency and drone operation efficiency, and reducing the complexity and energy consumption of drone flight missions. On the other hand, by continuously and segmentedly exploring the detection area (which changes dynamically as the vehicle moves) by the drone, replacing the one-time large-scale detection mode of the drone, unnecessary long-term flight and data collection redundancy can be reduced, communication latency and communication burden between the vehicle and the drone can be reduced, and the real-time and targeted nature of the detection data can be enhanced. In some embodiments, the road where the vehicle is located and the drone detection area can be updated in conjunction with the real-time location and navigation path updates of the vehicle, so that the drone detection area always focuses on the high-priority road section that the vehicle is about to enter, realizing fine-grained scheduling of sensing resources and closed-loop execution of detection tasks, thereby improving the decision support capability and energy efficiency management capability of the whole vehicle (and the vehicle carrying the drone) in complex traffic scenarios.

[0035] Thus, based on the aforementioned multimodal perception, it is convenient to further integrate road condition images and spatial feature data to construct a high-precision traffic condition perception model (i.e., a three-dimensional terrain model) for mountainous road scenarios. This will improve the detection accuracy and road risk assessment capabilities under typical mountainous terrain conditions such as narrow roads, limited visibility, and complex slopes, providing more reliable data support and judgment basis for subsequent decision-making processes such as UAV-assisted navigation, lane selection, and path adjustment.

[0036] Alternatively, driving assistance methods for mountain roads may also include: Based on the vehicle's first operating parameters and the drone's second operating parameters, a preset distance range is determined; wherein, the preset distance range includes a preset distance upper limit value and a preset distance lower limit value; when the first operating parameters meet the vehicle's risky operating conditions, the preset distance upper limit value is increased; when the second operating parameters meet the drone's functional descent conditions, the preset distance lower limit value is decreased.

[0037] Specifically, the preset distance range depends on the minimum detection distance (i.e., the lower limit of the preset distance range, denoted as the lower limit of the preset distance, which is also the closest distance between the drone's detection area and the vehicle) and the maximum detection distance (i.e., the upper limit of the preset distance range, denoted as the upper limit of the preset distance, which is also the farthest distance between the drone's detection area and the vehicle). The setting of the preset distance range needs to consider the following key factors: First, the lower limit of the preset distance affects the reaction time of the vehicle and driver in response to corresponding situations (such as sudden dangerous situations). Therefore, the setting of the lower limit of the preset distance should ensure that it provides sufficient safety buffer for the vehicle and driver, thereby providing sufficient reaction time. Based on this, the size of the lower limit of the preset distance can be dynamically adjusted in conjunction with the vehicle's operating parameters (denoted as the first operating parameter, such as vehicle speed, braking / braking distance, etc.) to ensure sufficient reaction time. Secondly, considering that the detection capability of UAVs is affected by factors such as hardware performance and communication environment, the setting of the preset distance limit needs to take into account the operating parameters of the UAV (referred to as the second operating parameter, such as the effective detection range of the UAV, remaining battery power, communication distance with vehicles, communication delay, etc.) to ensure that the UAV can work normally, continuously collect traffic information and communicate normally with vehicles; in some embodiments, it is also necessary to take into account the energy efficiency optimization of the UAV, and reduce the energy consumption of the UAV while ensuring the detection effect.

[0038] Furthermore, considering that the vehicle's first operating parameters and the drone's second operating parameters may change dynamically, the upper and lower limits of the preset distance range can be dynamically adjusted according to changes in the first and second operating parameters to obtain the best detection effect. For example, under conditions where vehicle operation poses a significant risk, such as high-speed driving or low-adhesion road surfaces (referred to as vehicle risk operating conditions), the vehicle may require a longer detection distance to ensure earlier risk identification and warning; therefore, the upper limit of the preset distance can be increased accordingly. Conversely, under conditions where the drone's performance declines, such as weak drone communication signals or low drone battery power (referred to as drone performance decline conditions), the lower limit of the preset distance can be appropriately reduced to ensure normal operation and communication stability of the drone, thereby avoiding data loss or excessive battery consumption and reducing the burden on the drone. In some embodiments, for terrain features unique to mountainous areas, such as continuous curves and steep slopes, the detection area can be specifically adjusted using high-precision map data to ensure that the drone always focuses on the most critical potential risk areas, thus achieving efficient utilization of drone resources while ensuring driving safety.

[0039] Optionally, after step 200, the mountain road driving assistance method further includes: The 3D terrain model is updated based on the latest traffic and vehicle location information acquired by the drone.

[0040] Specifically, the drone acquires traffic information on mountainous roads within a preset distance in front of the vehicle. As the vehicle continues to move, the drone's detection area also moves forward accordingly. In other words, the drone can maintain a certain distance in front of the vehicle for guided flight detection through appropriate dynamic path control logic. During this process, the drone can dynamically adjust its flight trajectory and altitude in real time based on the vehicle's current position, direction of travel, speed, and road topology information to ensure that the detection area always covers the target road segment the vehicle is about to travel on, thereby achieving continuous perception and early warning of the traffic conditions ahead.

[0041] As the vehicle's real-time location (determined based on vehicle positioning information) changes and traffic information acquired by the drone is updated, the constructed 3D terrain model is simultaneously updated. This includes expanding terrain data for newly added road areas ahead, optimizing the caching of terrain data for previously traversed areas, and updating the temporal marking and changes of dynamic traffic elements (such as vehicle distribution, obstacles, construction areas, etc.). This results in a traffic scene model with spatiotemporal continuity and dynamically evolving accuracy. This ensures that subsequently constructed new 3D terrain models (or newly added parts of the 3D terrain model) can reflect the traffic conditions of the road segment currently ahead of the vehicle in real time. It enhances the ability to identify risk areas in advance in changing environments, strengthens the foresight and accuracy of path planning, lane selection, and early warning, and thus provides more precise and effective driving assistance support in complex mountainous road environments.

[0042] In this way, by dynamically linking and coordinating paths between vehicles and drones, a traffic perception model of forward exploration and backward control is constructed. This not only improves the timeliness and completeness of traffic information acquisition in mountainous road environments, but also provides forward-looking and high-precision data support for subsequent risk identification, lane decision-making and path adjustment.

[0043] Optionally, a navigation path is planned for the drone based on the vehicle's navigation information; subsequently, based on the vehicle's navigation information, the drone's corresponding flight path is updated in real time when the vehicle changes its route, and the drone continues to guide the vehicle, so as to ensure that the drone's detection area continuously covers the target road segment that the vehicle is about to travel, thereby improving the effectiveness of traffic information collection.

[0044] Optionally, step 300 includes: Based on vehicle location information and 3D terrain model, dangerous targets are identified, and the identified dangerous targets are classified into dangerous types. Early warning information is generated based on the classification results.

[0045] Specifically, based on the 3D terrain model constructed in step 200 and the vehicle positioning information, precise driving decision support is achieved through multimodal data fusion and intelligent analysis. For example, the raw data collected by the UAV (such as image information, ranging data, elevation information, etc.) is preprocessed (e.g., denoising, distortion correction, illumination equalization, point cloud filtering, etc.), and deep learning feature extraction based on convolutional neural networks is performed (e.g., using a ResNet50 backbone network to extract multi-scale features). The extracted features of different types (e.g., visual semantic features corresponding to image information, structural features corresponding to ranging data, and terrain geometric features corresponding to elevation data, etc.) are cross-modal aligned and fused for use in the construction of the 3D terrain model. This allows the 3D terrain model to intuitively and realistically reflect the spatial structure and potential risk distribution of the mountain road detection area, improving the convenience of identifying dangerous targets in traffic information and classifying the identified dangerous targets by hazard type. Based on the constructed 3D terrain model, and by linking the spatial coordinate information within the model with vehicle positioning information, the system can accurately determine the current vehicle's location and potential risk targets along its forward path. This enables the identification of hazardous targets and the classification of hazard types, allowing for the generation of targeted warning information based on the classification results. For example, based on the spatial distribution and behavioral characteristics of objects existing on the vehicle's predicted path in the 3D terrain model, it can be determined whether they constitute hazardous targets. Then, combined with road terrain information (such as slope undulations, sharp turns, and obstructed areas), the hazard types of hazardous targets can be classified, such as oncoming vehicle risk, obstacle risk, road damage risk, and rockfall risk. Based on the hazard type classification results, corresponding warning information can be generated, such as an oncoming vehicle warning if there is an oncoming vehicle, an obstacle collision warning if there is an obstacle ahead, a road damage warning if there is a landslide ahead, and a rockfall area warning if there are falling rocks ahead, and so on.

[0046] Optionally, step 300 includes: Based on vehicle positioning information and a 3D terrain model, an initial path for vehicle travel is determined. The initial route is assessed for risk and optimized to generate suggested driving routes.

[0047] Specifically, for the generation of driving path suggestion information, based on the three-dimensional terrain model constructed in step 200 and the vehicle positioning information, the mountain roads in the UAV exploration area that can be used for vehicle driving (denoted as the initial path) are searched and determined; then, risk assessment and path optimization are performed on all determined initial paths to determine the optimal path for vehicle driving and generate the corresponding driving path suggestion information. For example, by combining elevation data, slope information, and road boundary constraints of the three-dimensional terrain, an improved algorithm is used to quickly generate multiple initial paths in three-dimensional space that meet the vehicle passability requirements. Secondly, a multi-dimensional risk assessment model is established. For each initial path, a quantitative assessment is performed from multiple dimensions, including static risks (such as road collapse areas and slope stability), dynamic risks (such as predicted trajectories of oncoming vehicles and rockfall probability), and environmental risks (such as low-visibility road sections and slippery coefficient), to calculate the comprehensive risk value of the path. Then, based on the risk-benefit trade-off, the initial paths are smoothed and locally optimized using the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm to generate a Pareto optimal path set. Finally, based on vehicle type (such as trucks, which need to consider load and turning radius) and driving mode (autonomous driving / manual driving), the best recommended path is selected from the optimal set, and the path details (such as length, estimated time, and risk level) and alternative solutions are visualized through the vehicle's infotainment system. In addition, the system continuously monitors changes in the route environment. When new risks are detected (such as sudden rockfalls), the system will trigger real-time route replanning to ensure driving safety.

[0048] Optionally, generating early warning information based on the classification results includes: The system generates warning information corresponding to the classification results and outputs the warning information through the vehicle's display mechanism and / or audio interaction mechanism; wherein the classification results include at least one of the following: oncoming vehicle risk, obstacle risk, road damage risk, and rockfall risk.

[0049] Specifically, based on the hazard type classification results, such as at least one of the following: oncoming vehicle risk, obstacle risk, road damage risk, and rockfall risk, corresponding warning information is generated and output through the vehicle's display mechanism and / or audio interaction mechanism to ensure timely and effective reminders to users. For example, the warning information is displayed to the user through the vehicle's display mechanism and audio interaction mechanism. For instance, the oncoming vehicle warning information is displayed as a flashing red box on the display mechanism and the audio interaction mechanism announces "Oncoming vehicle 500 meters ahead"; the obstacle collision warning information is displayed as a yellow triangle on the display mechanism and the audio interaction mechanism announces "Obstacle ahead on the right, avoid immediately"; the road damage warning information is displayed as a dark red warning symbol on the display mechanism and the audio interaction mechanism announces "Road collapse ahead, detour recommended"; and the road damage warning information is displayed as a dynamic warning area on the display mechanism and the audio interaction mechanism announces "Beware of falling rocks, slow down."

[0050] Optionally, a risk assessment and route optimization are performed on the initial route to generate suggested driving route information, including: The system performs risk assessment and path optimization on the initial path to determine the optimal path for vehicle travel and generates driving path suggestion information corresponding to the optimal path. The driving path suggestion information is output through the vehicle's display mechanism and / or audio interaction mechanism; wherein, the driving path suggestion information includes navigation information.

[0051] Specifically, for the generation of driving path suggestion information, based on the constructed 3D terrain model and vehicle positioning information, the mountain roads in the UAV exploration area that can be used for vehicle driving (denoted as the initial path) are searched and determined. Subsequently, risk assessment and path optimization are performed on all determined initial paths to achieve multi-level path optimization with driving safety as the premise, thereby determining the optimal path for vehicle driving, and generating corresponding driving path suggestion information based on the optimal path. The driving path suggestion information is output through the vehicle's display mechanism and / or audio interaction mechanism to ensure that the optimization can obtain the driving path suggestion information in a timely and effective manner. For example, risk assessment includes identifying hazardous targets and terrain risk factors (such as oncoming vehicles, obstacles, road damage, falling rocks, etc.) in the path, and scoring the path safety based on vehicle operating parameters, terrain structure, and traffic environment information. On this basis, a path optimization algorithm avoids high-risk road sections and selects safer alternative routes to generate the optimal path. Subsequently, driving path suggestion information including navigation information (including navigation route, driving suggestions, speed prompts, etc.) can be generated and output through the vehicle's display mechanism and / or audio interaction mechanism. For example, the vehicle's display mechanism indicates the path direction with a graphical interface, or the audio interaction mechanism broadcasts prompts such as "It is recommended to turn right to detour around the construction area 800 meters ahead" or "Please keep driving in the current lane" in a voice manner, thereby improving the driver's path decision-making efficiency and driving safety.

[0052] Optionally, the vehicle's display system includes an augmented reality (AR) display system, which can be used to provide the driver with driving route suggestions and / or warning information in a more intuitive way. Specifically, the AR display system may include a head-up display (HUD) or AR glasses integrated into the vehicle's windshield area. Through image overlay and environmental scene fusion technology, it overlays virtual content such as navigation information and risk warning signs onto the actual road conditions in the driver's field of vision in real time, thereby achieving a combined virtual and real-world assisted driving prompt. For example, when there is a risk of falling rocks ahead, the AR display system can mark the rockfall area with a warning icon on the windshield in real time; when route adjustments are needed, the AR display system can intuitively indicate the direction of change of the suggested route through virtual arrows or route guidance lines. This helps improve the perception efficiency and intuitiveness of information prompts, enhancing driving safety and user experience in complex mountainous road scenarios.

[0053] Combination Figure 2 As shown, another embodiment of the present invention provides a driving assistance device for mountain roads, comprising: The navigation response unit is used to respond to navigation commands on mountain roads and control the drone to acquire traffic information about the mountain roads where the vehicle is located. The model building unit is used to construct a three-dimensional terrain model of mountain roads based on traffic information. The information generation unit is used to generate driving route suggestion information and / or warning information based on vehicle positioning information and three-dimensional terrain model.

[0054] The mountain road driving assistance device of this embodiment is used to implement the above-mentioned mountain road driving assistance method. Its advantages over the prior art are the same as the advantages of the above-mentioned mountain road driving assistance method over the prior art, and will not be repeated here.

[0055] Combination Figure 3 As shown, another embodiment of the present invention provides a vehicle, including a memory 301 and a processor 302; Memory 301 is used to store computer programs; Processor 302 is used to implement the above-mentioned driving assistance method for mountain roads when executing a computer program.

[0056] Alternatively, a vehicle includes a memory 301 and a processor 302 coupled to the memory 301; the memory 301 is configured to store a computer program; the processor 302 is configured to perform the following operations when the computer program is executed: In response to navigation commands for mountain roads, the drone is controlled to acquire traffic information about the mountain roads where the vehicle is located; Based on traffic information, construct a three-dimensional terrain model of mountain roads; Based on vehicle location information and 3D terrain model, generate driving route suggestion information and / or warning information.

[0057] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A driving assistance method for mountain roads, characterized in that, Based on vehicles equipped with drones; The driving assistance methods for mountain roads include: In response to a navigation command for mountain roads, the drone is controlled to acquire traffic information about the mountain roads where the vehicle is located; Based on the traffic information, construct a three-dimensional terrain model of the mountain roads; Based on the vehicle location information and the three-dimensional terrain model, driving route suggestion information and / or warning information are generated.

2. The driving assistance method for mountain roads as described in claim 1, characterized in that, The step of responding to a mountain road navigation command and controlling the drone to acquire traffic information about the mountain road where the vehicle is located includes: In response to a prompt message about the vehicle entering a mountainous road, or in response to a navigation trigger command input by the driver, the mountainous road navigation command is generated; wherein the prompt message is generated by a navigation system applied to the vehicle; the driver includes at least one of the vehicle's user and the driver assistance control system.

3. The driving assistance method for mountain roads as described in claim 1 or 2, characterized in that, The drone is equipped with an image acquisition mechanism and a spatial position measurement mechanism; controlling the drone to acquire traffic information about the mountain roads where the vehicle is located includes: The drone is controlled to acquire road condition images of the mountain road within a preset distance range in front of the vehicle via the image acquisition mechanism; and spatial feature data of the mountain road within a preset distance range in front of the vehicle is measured via the spatial position measurement mechanism.

4. The driving assistance method for mountain roads as described in claim 3, characterized in that, Also includes: The preset distance range is determined based on the first operating parameters of the vehicle and the second operating parameters of the drone; wherein the preset distance range includes a preset upper limit value and a preset lower limit value; When the first operating parameter meets the vehicle risk operation conditions, the preset distance upper limit value is increased; when the second operating parameter meets the drone's descent conditions, the preset distance lower limit value is decreased.

5. The driving assistance method for mountain roads as described in claim 1 or 2, characterized in that, After constructing a three-dimensional terrain model of the mountain road based on the traffic information, the mountain road driving assistance method further includes: The three-dimensional terrain model is updated based on the latest traffic information and vehicle positioning information acquired by the UAV.

6. The driving assistance method for mountain roads as described in claim 1 or 2, characterized in that, The step of generating driving route suggestion information and / or warning information based on vehicle positioning information and the three-dimensional terrain model includes: Based on the vehicle positioning information and the three-dimensional terrain model, dangerous targets are identified, and the identified dangerous targets are classified into dangerous types. The warning information is generated based on the segmentation results; And / or, the step of generating driving route suggestion information and / or warning information based on vehicle positioning information and the three-dimensional terrain model includes: Based on the vehicle positioning information and the three-dimensional terrain model, an initial path for the vehicle to travel is determined; The initial path is subjected to risk assessment and path optimization to generate the driving path suggestion information.

7. The driving assistance method for mountain roads as described in claim 6, characterized in that, The step of generating the early warning information based on the division results includes: The warning information corresponding to the division result is generated and output through the vehicle's display mechanism and / or audio interaction mechanism; wherein, the division result includes at least one of oncoming vehicle risk, obstacle risk, road damage risk, and rockfall risk.

8. The driving assistance method for mountain roads as described in claim 6, characterized in that, The step of performing risk assessment and route optimization on the initial route to generate the driving route suggestion information includes: The initial path is subjected to risk assessment and path optimization to determine the optimal path for the vehicle to travel, and the driving path suggestion information corresponding to the optimal path is generated. The driving path suggestion information is output through the vehicle's display mechanism and / or audio interaction mechanism; wherein, the driving path suggestion information includes navigation information.

9. A driving assistance device for mountain roads, characterized in that, include: The navigation response unit is used to respond to navigation commands on mountain roads and control the drone to acquire traffic information about the mountain roads where the vehicle is located. The model building unit is used to build a three-dimensional terrain model of the mountain roads based on the traffic information. The information generation unit is used to generate driving route suggestion information and / or warning information based on the vehicle positioning information and the three-dimensional terrain model.

10. A vehicle, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the mountain road driving assistance method as described in any one of claims 1-8 when executing the computer program.

Images