Intelligent all-terrain driving control method and system

Abstract

The invention discloses an intelligent all-terrain driving control method and system, and relates to the technical field of all-terrain driving mode vehicles. Determining that the current road condition belongs to at least one road condition of rain and snow, wading, muddy land, sand land, rock and cross-country; matching a current optimal driving mode according to the determined current road condition, switching to a corresponding optimal driving mode, and sending a control strategy instruction for executing the corresponding optimal driving mode to an execution system; according to the control strategy instruction, a power system, a steering system, a suspension system, a braking system and a cabin system are controlled to execute corresponding actions; and when the current road conditions are two or more, executing the corresponding control strategy according to the set priority of the execution system. Intelligent road condition recognition and self-adaptive driving mode switching are realized, the operation of selecting the driving mode by a user is reduced, and the all-terrain function use experience of the user is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of all-terrain driving mode vehicles, and in particular to an intelligent all-terrain driving control method and system. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Vehicles with all-terrain driving modes typically feature multiple driving modes, including energy-saving, comfort, sport, rain and snow, mud, sand, wading, rock, and off-road. Each driving mode is linked to systems such as powertrain, braking, steering, suspension, and cockpit. When a corresponding driving mode is selected, each system adjusts its matching mode based on the associated strategy. In related technologies, when a vehicle encounters a specific road condition, the user selects the appropriate driving mode based on their needs. The control system receives a signal to switch driving modes, and each system responds accordingly. After navigating the road condition, the user adjusts the driving mode again based on their needs.

[0004] However, manual selection of driving modes is required. Inexperienced drivers often don't know which mode to choose when encountering specific road conditions, resulting in ineffective all-terrain functionality and a poor user experience. Furthermore, drivers can't predict road conditions ahead, often realizing the need to switch to the appropriate driving mode only after they've already entered the desired road. This results in delayed responses from the associated systems and a poor driving experience. For example, if you select mud driving mode after already being stuck in mud, you're likely to be stuck.

[0005] During vehicle driving, accurate perception of road condition information is the core prerequisite for ensuring driving safety and optimizing driving strategies. Existing road condition recognition technology has the following main shortcomings: Limitations of single sensors: Traditional solutions rely on a single camera or radar. Cameras are easily affected by lighting (such as backlight and darkness) and weather (such as rain, snow, and fog), resulting in distortion in road texture recognition. Although radar can obtain three-dimensional information, it lacks semantic features such as color and material, making it difficult to distinguish between similar road conditions such as mud and sand.

[0006] Insufficient trajectory targeting: Existing technologies generally perform generalized recognition of the area in front of the entire vehicle, rather than focusing on the actual wheel trajectory range of the vehicle. This results in a low correlation between the recognition results and driving safety (for example, only recognizing the overall state of the road surface and ignoring potholes in the left and right wheel trajectories).

[0007] Low recognition accuracy for complex road conditions: For dynamic or unstructured road conditions such as rain, snow, wading, and mud, there is a lack of a multi-dimensional feature fusion mechanism. Judgment is based solely on a single indicator (such as reflectivity or color), which can easily lead to misjudgments (such as misjudging the reflection of accumulated water as ice and snow).

[0008] Furthermore, the driving safety and comfort of a vehicle under complex road conditions depends on the precise adaptation of the power, braking, steering and other systems to the road conditions. Existing technologies have the following limitations: Driving mode switching is passive and crude: The driving modes of traditional vehicles (such as "sport mode" and "economy mode") need to be switched manually by the user, and the mode adjustment only involves a simple scaling of the power parameters (such as a fixed torque output ratio), and cannot be dynamically adapted to the specific road conditions identified in real time. Poor system coordination: The powertrain, braking, and steering systems are controlled independently, lacking a unified road response strategy. For example, on rocky roads, the suspension rises without the powertrain simultaneously adjusting to low torque output, potentially causing wheel slip. In rainy or snowy conditions, the braking system increases braking force without reducing steering sensitivity, easily leading to oversteer. Summary of the Invention

[0009] In order to solve the above problems, the present invention proposes an intelligent all-terrain driving control method and system, which can realize intelligent recognition of road conditions and adaptive switching of driving modes, reduce the user's operation of selecting driving modes, and enhance the user's all-terrain function experience.

[0010] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides an intelligent all-terrain driving control method, comprising: Determining, based on the acquired road condition image and radar data corresponding to the location and time, current weather information, environmental information, and location information, whether the current road condition is at least one of rain and snow, wading, mud, sand, rock, and off-road; According to the determined current road conditions, the system matches the current optimal driving mode and switches to the corresponding optimal driving mode, while sending a control strategy instruction for executing the corresponding optimal driving mode to the execution system; According to the control strategy instructions, the power system, steering system, suspension system, braking system and cockpit system are controlled to perform corresponding actions; and when there are two or more current road conditions, the corresponding control strategy is executed according to the set priority of the execution system.

[0011] As an optional implementation method, the camera is used to collect planar images and stereo images of the road surface within the wheel trajectory range; the lidar is used to collect three-dimensional point cloud data of the road surface and surrounding objects, as well as environmental information such as the distance and speed of surrounding obstacles; the current geographical location information is identified through a map to match the current road section attributes; and the temperature and humidity sensors as well as rain and snow sensors are used to collect weather information such as ambient temperature, precipitation, and humidity.

[0012] As an optional implementation, a pre-trained semantic segmentation model is used to perform pixel-level annotation on the plane image, and output image data including road surface material and state labels; Fuse the planar image and the stereo image to generate a dense point cloud and optimize the point cloud coordinates, outputting point cloud data associated with visual features; Along the center lines of the left and right wheel trajectories, the point cloud elevation values ​​are sampled at intervals to generate two continuous elevation curves.

[0013] As an optional implementation method, a road condition sample library is constructed to store standard image features and point cloud features of different road conditions; the output image data and point cloud data are matched with the road condition sample library to obtain preliminary road surface signals; Leveling quantification based on the left and right wheel track elevation curves, including: If the elevation standard deviation is less than the minimum threshold, the road surface is considered flat. If it is within the range of the minimum and maximum thresholds, the road surface is considered slightly uneven. If it is greater than or equal to the maximum threshold, the road surface is considered severely uneven. If the maximum elevation difference between adjacent sampling points is greater than or equal to the undulation difference threshold, it is marked as avoiding convexity or concavity; The cumulative length of consecutive sampling points with elevation differences greater than the maximum threshold is the continuous abnormal length. If the continuous abnormal length is greater than or equal to the abnormal threshold, a deceleration warning is triggered; Identify current road conditions based on preliminary road surface signals, smoothness quantification results, radar data, current weather information, environmental information, and location information.

[0014] As an optional implementation, the process of controlling the execution of corresponding actions according to the control strategy instruction includes: Control the power system to enter the corresponding torque distribution ratio and torque gradient; Control the steering system to enter the corresponding steering mode; Control the suspension system to adjust to the corresponding suspension damping and suspension height mode; Control the braking system to adjust to the corresponding traction control mode and body stability mode; Control the cockpit system to adjust to the corresponding comfort environment mode.

[0015] As an optional implementation method, when there are two or more current road conditions, the priority order is set as braking system, power system, steering system, suspension system, and cockpit system; and the vehicle status is monitored in real time. If the deviation between the actual status and the expected status exceeds the threshold, the control parameters are dynamically corrected.

[0016] In a second aspect, the present invention provides an intelligent all-terrain driving control system, comprising: a road condition recognition module configured to determine whether a current road condition is at least one of rain and snow, wading, mud, sand, rock, and off-road based on the acquired road condition image and radar data corresponding to the location and time, current weather information, environmental information, and location information; a decision module configured to match the current optimal driving mode according to the determined current road conditions, switch to the corresponding optimal driving mode, and send a control strategy instruction for executing the corresponding optimal driving mode to the execution system; The execution module is configured to control the power system, steering system, suspension system, braking system and cockpit system to perform corresponding actions according to the control strategy instructions; and when the current road conditions are two or more, the corresponding control strategy is executed according to the set priority of the execution system.

[0017] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium for storing computer instructions, wherein when the computer instructions are executed by a processor, the method described in the first aspect is performed.

[0019] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which implements the method described in the first aspect when executed by a processor.

[0020] Compared with the prior art, the present invention has the following beneficial effects: The present invention proposes an intelligent all-terrain driving control method and system that realizes intelligent identification of road conditions and adaptive switching of driving modes. Through the intelligent road condition identification system, the road conditions ahead are perceived in advance, and the driving mode is adaptively switched, reducing the user's selection of driving modes, bringing intelligent perception, and improving the user's all-terrain function experience; at the same time, each system is in the best performance working state in advance, bringing the ultimate performance experience to the user. Compared with traditional all-terrain systems, the method of the present invention realizes intelligent road condition identification and completes closed-loop control of all-terrain road conditions from advance perception, control, and execution, realizing the transition from passive adaptation to active prediction, ensuring that no matter what road conditions the vehicle is in, each system can be in the best working state in advance, ensuring optimal performance, improving the vehicle's passability, safety, and comfort in various terrains, and bringing the user the best driving performance experience in all road conditions.

[0021] The method of the present invention integrates vision, radar, map, and environmental data to compensate for the shortcomings of a single sensor and improve the accuracy of identifying complex road conditions. It quantifies the flatness of elevation curves through multiple indicators, and distinguishes between similar road conditions such as mud and sand by combining semantic tags with point cloud reflectivity, with recognition accuracy reaching the centimeter level. Through the coordinated response of multiple systems, the control strategy is precisely adjusted according to the characteristics of road conditions such as rain, snow, and rocks. Without human intervention, it automatically adjusts according to real-time road conditions to adapt to dynamically changing complex scenarios, lowering the driving threshold and improving safety in complex road conditions.

[0022] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0024] Figure 1 This is a flow chart of the intelligent all-terrain driving control method provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0026] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0027] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "comprise" and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0029] Example 1 This embodiment provides an intelligent all-terrain driving control method. Through intelligent recognition of road conditions, it can sense the road conditions ahead in advance and adaptively switch driving modes, reducing the user's need to select driving modes, bringing intelligent perception, and improving the user's all-terrain function experience. At the same time, each system is in the best performance working state in advance, bringing the user an ultimate performance experience.

[0030] like Figure 1 As shown, specifically including: Determining, based on the acquired road condition image and radar data corresponding to the location and time, current weather information, environmental information, and location information, whether the current road condition is at least one of rain and snow, wading, mud, sand, rock, and off-road; According to the determined current road conditions, the system matches the current optimal driving mode and switches to the corresponding optimal driving mode, while sending a control strategy instruction for executing the corresponding optimal driving mode to the execution system; According to the control strategy instructions, the power system, steering system, suspension system, braking system and cockpit system are controlled to perform corresponding actions; and when there are two or more current road conditions, the corresponding control strategy is executed according to the set priority of the execution system.

[0031] In this embodiment, the process of determining the current road condition includes: (1) First, the vehicle collects and judges road condition information through environmental perception modules such as cameras, radars, maps, and weather.

[0032] Specifically: The camera collects the planar image and stereo image of the road surface within the wheel trajectory range (e.g., a 2.5-3m wide area bounded by the left and right wheels of the vehicle); The laser radar collects 3D point cloud data of the road surface and surrounding objects, and the millimeter-wave radar collects the distance and speed information of surrounding obstacles; Identify the current geographical location through the map. The real-time geographic location of the vehicle is obtained through GPS (positioning accuracy ≤ 0.5m) combined with IMU, and the current road section attributes, such as paved / unpaved, speed limit, and historical road condition tags, are obtained by matching with high-precision maps. Weather information such as ambient temperature, precipitation, and humidity is collected through temperature and humidity sensors and rain and snow sensors.

[0033] (2) Then, data preprocessing and spatiotemporal calibration are performed; specifically: Visual data processing: De-noising of planar images, distortion correction based on camera intrinsic parameters, and contrast enhancement (e.g., using the Retinex algorithm); epipolar correction of stereo images to eliminate line deviation between left and right views; Radar data processing: De-noise and downsample the lidar point cloud to retain valid road point clouds; Spatiotemporal calibration: Using GPS timing and a unified timestamp, multi-source data is aligned. The radar point cloud is projected into the camera image coordinate system through external parameter calibration to achieve "visual-3D" spatial alignment.

[0034] (3) Feature extraction and point cloud generation, specifically: Image semantic feature extraction: Use a pre-trained semantic segmentation model to perform pixel-level annotation on planar images and output labels for road surface material (asphalt / cement / soil / rock, etc.) and state (dry / watered / snowy, etc.); Point cloud data generation: Utilize image semantic information combined with a stereo matching algorithm (such as SGBM) to fuse a planar image with a stereo image to generate a dense point cloud. Use a calibration algorithm to optimize the point cloud coordinates and output 3D point cloud data associated with visual features. Elevation curve generation: Along the center lines of the left and right wheel trajectories (determined based on the lane line detection results), the point cloud elevation values ​​are sampled at 0.5m intervals to generate two continuous elevation curves (the horizontal axis is the driving distance, and the vertical axis is the relative elevation).

[0035] (4) Sample matching and flatness judgment, specifically: Sample library matching: Construct a road condition sample library to store standard image features (texture, color, semantic labels) and point cloud features (reflectivity, elevation distribution, material three-dimensional structure) of different road conditions (rain, snow, wading, mud, etc.); perform feature vector matching between the image data and point cloud data output in step (3) and the sample library (e.g., if the cosine similarity is ≥0.85, it is determined to be a successful match), and output preliminary road surface signals.

[0036] Quantification of leveling: Calculate the following indicators based on the left and right wheel track elevation curves: If the elevation standard deviation is less than the minimum threshold (e.g., σ < 2 cm), the road surface is considered flat. If it is within the range of the minimum and maximum thresholds (e.g., 2 cm ≤ σ < 5 cm), the road surface is considered slightly uneven. If it is greater than or equal to the maximum threshold (5 cm ≤ σ), the road surface is considered severely uneven. Maximum elevation difference: This is the maximum elevation difference between adjacent sampling points. If it is greater than or equal to the elevation difference threshold (e.g., 10 cm), it is marked as a convex / concave area that needs to be avoided. Continuous abnormal length: The cumulative length of continuous sampling points with elevation difference greater than the maximum threshold. If it is greater than or equal to the abnormal threshold, a deceleration warning is triggered.

[0037] (5) Comprehensive road condition identification: combining preliminary road surface signals, smoothness indicators and environmental data to identify specific road conditions through multi-dimensional decision logic.

[0038] (5-1) Rainy and snowy road conditions (environmental + visual + radar): When the environmental data detects precipitation > 0.5 mm / h, and there are dynamic rain and snow noise points in the image, and the radar point cloud contains low-reflectivity dynamic points, it is determined to be rainy and snowy road conditions; Specifically: Environmental sensor characteristics: The rain and snow sensor detects precipitation > 0.5 mm / h, the temperature sensor displays > 0°C (rain) or < 0°C (snow); the humidity sensor reading is > 85%, and the wind speed is < level 3 (to prevent rain and snow from being blown away and causing misjudgment).

[0039] Visual image characteristics: There are randomly distributed dynamic noise points in the plane image (the movement of raindrops / snowflakes in front of the camera). Raindrops appear as short vertical lines, and snowflakes appear as irregular dots. Uneven road surface reflectivity: On rainy days, the road surface has enhanced local reflectivity (in areas with accumulated water), while on snowy days, the overall road surface reflectivity increases (>70%) and the texture is blurred (caused by snowflakes).

[0040] Radar characteristics: A large number of low-intensity dynamic points (reflection signals from rain and snow particles) appear in the lidar point cloud, and the proportion of invalid points exceeds 20% as precipitation increases. Millimeter-wave radar echoes contain clutter from non-fixed targets, and the clutter intensity increases with vehicle speed (the movement of rain and snow particles is driven by the vehicle).

[0041] (5-2) Water-crossing conditions (vision + lidar + vehicle status); if the image semantic label contains water, the point cloud reflectivity shows a high and low double peak (water surface reflection and underwater absorption), and the temperature is greater than 0°C, it is determined to be a water-crossing condition, and the water depth is calculated based on the point cloud elevation difference; Specifically: Visual image features: There are continuous specular reflection areas on the road surface. In the HSV color space, H≈120 (blue hue) and S>0.5 (high saturation). As the vehicle approaches, ripple disturbances appear in the reflection area. The boundary between water and non-water areas is clear, and there may be sediment deposition (yellowish color) at the boundary, or there may be water flow direction (such as converging from the roadside to the middle of the road).

[0042] LiDAR characteristics: The reflectivity of the point cloud in the water-crossing area shows a double peak: the reflectivity of some points is extremely high (>90%) due to the mirror reflection of the water surface, while the reflectivity of some points is extremely low (<10%) due to the penetration of the water layer; Point cloud elevation anomaly: The water depth is calculated based on the point cloud elevation difference. When the water depth is greater than 5 cm, the road surface elevation measured by the lidar is lower than the actual road surface (a virtual low elevation caused by the water layer).

[0043] Vehicle status assistance: When the vehicle speed is greater than 30km / h, if the camera captures water splashing from the wheels (white water splash tracks appear at the bottom of the image), it can enhance the water wading judgment.

[0044] (5-3) Muddy road conditions (vision + lidar + driving feedback); If the image color is dark brown / black, the semantic label contains mud / gravel, the point cloud reflectivity is less than 30% (mud) or 30%-50% (sand), and the vehicle wheel speed sensor detects slippage (wheel speed > 10% of vehicle speed), it is judged as mud or sand, respectively; Specifically: Visual image features: The road surface is dark brown / black in color, with a rough and uneven texture. There are continuous ruts (depth > 2cm) formed by wheel rolling, and mud accumulates at the edges of the ruts. When the vehicle is driving, mud may splash onto the lens (irregular dark spots appear in the image), or mud may stick to the bottom of the vehicle body (confirmed by the underbody camera).

[0045] LiDAR characteristics: The reflectivity of the point cloud is generally low (<30%) because the loose soil surface absorbs light; The elevation curve shows low-frequency fluctuations: the mud ground is uneven as a whole, with randomly distributed small bumps (mud blocks) and depressions (ruts), and the elevation standard deviation σ>5cm.

[0046] Vehicle feedback characteristics: The wheel speed sensor detects an abnormal wheel speed difference (unilateral wheel slip, wheel speed > 10% of vehicle speed); the vehicle posture sensor (IMU) measures increased vertical vibration (acceleration > 0.5g) and increased driving resistance (vehicle speed decreases under the same throttle opening).

[0047] (5-4) Sandy road conditions (vision + lidar + material characteristics); Visual image characteristics: The road surface is light yellow / white in color, has a strong grainy feel, has no obvious rutting (or the rutting is shallow and easily restored), and may show bright spots of sand reflective under sunlight; Blurred edges: The boundary between sandy land and non-sandy land (such as hard shoulders) transitions smoothly, without sharp edges, and may have sand dunes / sand ridges formed by wind.

[0048] LiDAR characteristics: The point cloud has a medium reflectivity (30%-50%), and the echo intensity is dispersed due to sand scattering; Irregular small undulations in the elevation curve: tiny bumps (1-3 cm in height) formed by sand accumulation, and the point cloud density is lower than that of hard road surface (gaps between sand grains cause partial laser penetration).

[0049] (5-5) Rocky road surface (vision + lidar + vibration detection); if the image contains stone texture with sharp edges, the point cloud elevation standard deviation σ ≥ 10cm, and the reflectivity is greater than 60%, it is judged as a rocky road condition; Specifically: Visual image characteristics: The road surface is composed of stones of different sizes with sharp edges and various colors (gray, black, red), and obvious joints (width > 1 cm); the stone surface texture is hard (such as the granularity of granite and the layered structure of limestone), without soil coverage, and may have moss (shady and humid areas).

[0050] LiDAR features: drastic fluctuations in point cloud elevation: σ>10 cm, with isolated high protrusions (single rock height 5-30 cm) and crack depressions (depth 3-10 cm); High and uneven reflectivity: The hard rock surface results in a reflectivity greater than 60%, but the reflectivity in the cracks is less than 10% (no reflection).

[0051] (5-6) Off-road conditions: When the conditions of unpaved road (map attribute) + a combination of two or more complex road conditions (such as rock + mud) + an elevation standard deviation σ ≥ 15 cm are met, the road is considered off-road.

[0052] Specifically: Off-road conditions are not a single condition, but a combination of multiple and complex conditions. At least three of the following conditions must be met: Poor road surface flatness: elevation standard deviation σ>15cm, with continuous undulations (slope>10°); Contains two or more special road conditions (such as rock + mud, sand + wading); There are no regular lane lines, and the map location shows unpaved roads (such as wilderness and mining areas); While the vehicle is driving, the IMU detects continuous severe vibration (vertical acceleration > 0.8g) and frequent changes in the steering wheel angle (the driver frequently corrects the direction); The LiDAR point cloud shows that the road surface boundaries are blurred (no clear shoulder, and it blends with the surrounding terrain).

[0053] (6) Finally, the output is a road condition signal containing the following information: basic road condition type (rain and snow / wading / mud / sand / rock / off-road), flatness index, etc.

[0054] In this embodiment, the output image data and point cloud data are compared with the image data and point cloud data of road sample targets collected and learned previously. The road surface flatness is determined using elevation curve data from the left and right wheel trajectories, thereby determining road surface information and outputting relevant road surface signals. Finally, the system integrates the collected road condition information to complete road surface recognition and determine whether the current road conditions are rain, snow, wading, mud, sand, rocky, or off-road. For example, if the vehicle is currently traveling on a snowy road, when the camera captures an image of the snowy road, the processed image information is compared with the snow surface targets in the road sample, and if the current weather information is below 0°C, the vehicle is determined to be traveling on a snowy road, and rain and snow road information is output.

[0055] It is understandable that the above-mentioned cameras, radars, weather, maps, etc. are only examples of the perception systems recommended in this embodiment, and actual applications are not limited to this solution. Road perception can also be achieved through other methods such as wheel speed and vehicle posture.

[0056] In this embodiment, a domain controller is used to make judgments based on the identified current road condition information and the preset characteristic logic of each road condition, complete the judgment and decision of the current optimal driving mode, adaptively switch to the corresponding driving mode, and send control strategy instructions for executing the corresponding driving mode to each subsystem of the execution system.

[0057] For example, if the road surface information sent by the camera is snow surface information and the temperature signal is also below 0℃, the decision-making system will determine that the current optimal driving mode is rain and snow mode based on internal logical operations, and will automatically switch the driving mode to rain and snow mode, and at the same time send the corresponding working strategy signal under rain and snow mode to the relevant execution system.

[0058] In this embodiment, the execution system includes a power system, a steering system, a suspension system, a braking system and a cockpit system. After receiving the control strategy instructions, the power, braking, steering, suspension, cockpit and other systems execute specific actions according to the corresponding driving mode strategy.

[0059] Specifically: (1) Power system: Adjust the power system to the corresponding torque distribution ratio and torque gradient. For example, in rainy and snowy conditions, a low torque gradient and equal distribution of front and rear axle torque are used to achieve low traction starting.

[0060] Specifically: (1-1) Rainy and snowy road conditions: The torque gradient is reduced to 60% of the normal mode (to avoid slipping during sudden acceleration), the front and rear axle torque is evenly distributed at 50:50 (to enhance stability), the maximum speed is limited to 80 km / h, and the throttle response delay is increased by 200ms (to prevent misoperation). (1-2) Water-crossing conditions: If the water depth is less than 10 cm, the torque output is limited to 70% of the rated value, and a constant speed is maintained (to avoid splashing water into the air intake); if the depth is ≥ 10 cm, the vehicle automatically switches to low-speed four-wheel drive mode, with the torque distributed 30% to the front and 70% to the rear (the rear wheels take the lead in propulsion to reduce the paddling resistance of the front wheels). (1-3) Mud: The torque gradient is reduced to 50%, and the torque pulsation mode is activated (torque fluctuation ±10% every 0.5 seconds), using vibration to reduce the risk of tire sinking. (1-4) Sand: Torque output is increased by 10% (to overcome the rolling resistance of sand), and the front and rear axle torque is distributed at a ratio of 40:60 to prevent excessive sand-digging by the front wheels. (1-5) Rocky / off-road conditions: Switch to low-speed, high-torque mode with a torque amplification factor of ≥2.5 times. Use single-wheel brake differential lock logic (when a wheel slips, brake that wheel and distribute torque to other wheels). Maximum speed is limited to 30 km / h. (2) The steering system enters the corresponding steering mode.

[0061] Specifically: (2-1) Rainy, snowy, or wading conditions: Steering assist is reduced by 20% (to increase road feedback), and the steering ratio is adjusted to 15:1 (normally 12:1, meaning a 15° steering wheel turn corresponds to a 1° wheel turn) to reduce the risk of oversteer. The return torque is increased by 30% to ensure that the wheels quickly return to center after steering. (2-2) Mud / Sandy Road Conditions: Steering sensitivity is reduced (steering wheel rotation must be increased by 20% to reach the normal steering angle) to avoid lateral wheel slip caused by frequent steering; steering assist is automatically increased at low speeds (<10 km / h) to facilitate large-angle steering when getting out of trouble. (2-3) Rocky / off-road conditions: Activate off-road steering mode, and the rear wheel steering angle increases to 5° (normally 2°), reducing the turning radius; when it detects that the wheel is about to contact the rock, the steering motor is temporarily locked (0.3 seconds) to prevent the steering mechanism from overloading. (3) The suspension system is adjusted to the corresponding suspension damping and suspension height modes; for example, when encountering rocky or off-road conditions, if the vehicle cannot pass through at its current ground clearance and there is a risk of collision, the suspension will adjust the suspension height mode, such as using a high suspension mode that allows the vehicle to pass, so that the suspension height is in a high position to pass through.

[0062] Specifically: (3-1) Rainy, snowy / water-crossing road conditions: The suspension damping is adjusted to medium hardness (damping coefficient 3000N•s / m) to reduce body roll; when wading in water, if the water depth is greater than 20cm, the suspension will automatically raise 5cm (increase ground clearance) to prevent water splashing into the engine compartment. (3-2) Muddy / sandy road conditions: The suspension damping is reduced to soft mode (damping coefficient 1500N•s / m), allowing the vehicle body to vibrate up and down more significantly (±5cm). The elastic buffer reduces the hard impact between the tires and the ground to prevent the vehicle from getting stuck. (3-3) Rocky / off-road conditions: Based on the road elevation predicted by the LiDAR point cloud, the suspension height is adjusted 500ms in advance: when a bump height greater than 15cm is detected, the suspension is raised to the highest position (ground clearance ≥ 25cm); when continuous potholes (depth greater than 10cm) are detected, the suspension damping switches to adaptive mode (damping is adjusted in real time according to vehicle acceleration, ranging from 1000-5000N•s / m). (4) Braking system: Adjust to the corresponding traction control mode and body stability mode, enter the limited slip state in advance, and improve the vehicle's passability.

[0063] Specifically: (4-1) Rainy and snowy road conditions: Activate the ABS anti-lock braking system in advance (normal mode requires a wheel speed difference > 20km / h to trigger, but this is reduced to 10km / h), increase the brake pedal travel gain by 30% (a slight step can generate a greater braking force), and shorten the braking distance by 15%-20%. (4-2) When driving on flooded roads: the brake disc will automatically vent air (lightly apply the brake every 5 seconds) to expel the water film between the brake disc and the brake pad after driving on flooded roads; the braking force output will be adjusted gradually (the initial braking force is 70% of the normal force, which increases linearly with the increase of pedal travel) to avoid brake failure caused by water film. (4-3) Mud / Sand Road Conditions: The brake intervention threshold is increased (allowing wheel speed differences of up to 1.5 times that of normal mode) to prevent tire locking and vehicle bogging caused by frequent braking; 5% of the torque output (drag torque) is retained after braking to prevent the tires from sinking into mud and sand after they have completely stopped. (4-4) Rocky / off-road conditions: Hill Descent Control (HDC) is activated to automatically maintain the vehicle speed at ≤5 km / h, and the braking frequency is synchronized with the road surface bumps (the position of the bumps is predicted 0.5 seconds in advance through the lidar, and the braking is enhanced in a targeted manner). (5) The cockpit system is adjusted to the corresponding comfort environment mode, such as different air-conditioning modes and different seat angles, heights and postures.

[0064] Specifically: (5-1) Rain, snow, and off-road conditions: The seats automatically switch to sports support mode (the side wings are inflated and the lumbar support strength is increased by 50%) to prevent the occupants from sliding due to vehicle bumps; the air conditioner switches to internal circulation + defogger mode to keep the windows clear. (5-2) Water / mud road conditions: The display screen switches to a road condition visualization interface, showing the wheel track and obstacle distance in real time. (6) System collaborative execution and dynamic adjustment.

[0065] Real-time priority determination: When multiple road conditions coexist (such as rain, snow, and mud), the strategy is executed according to safety priority (brakes > power > steering > suspension > cabin). Feedback closed-loop control: The vehicle status (such as slip rate and body roll angle) is monitored in real time through wheel speed sensors and acceleration sensors. If the actual status deviates from the expected status by more than 10% (such as continued slip under the expected torque distribution), the control parameters are dynamically corrected (such as further reducing the torque gradient). Smooth mode switching: When switching from one road condition to another (such as driving from mud to a rocky area), each system parameter completes a smooth transition within 1.5 seconds (such as the suspension height linearly increasing from medium to maximum) to avoid impact.

[0066] It can be understood that some of the data given in the above scheme are only example data given in this embodiment, and the actual application is not limited to this scheme. Those skilled in the art can also make adaptive adjustments according to actual conditions.

[0067] Example 2 This embodiment provides an intelligent all-terrain driving control system, including: a road condition recognition module configured to determine whether a current road condition is at least one of rain and snow, wading, mud, sand, rock, and off-road based on the acquired road condition image and radar data corresponding to the location and time, current weather information, environmental information, and location information; a decision module configured to match the current optimal driving mode according to the determined current road conditions, switch to the corresponding optimal driving mode, and send a control strategy instruction for executing the corresponding optimal driving mode to the execution system; The execution module is configured to control the power system, steering system, suspension system, braking system and cockpit system to perform corresponding actions according to the control strategy instructions; and when the current road conditions are two or more, the corresponding control strategy is executed according to the set priority of the execution system.

[0068] It should be noted that the above modules correspond to the steps described in Example 1, and the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above Example 1. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0069] In further embodiments, there is also provided: An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed by the processor, wherein when the computer instructions are executed by the processor, the method described in Example 1 is performed. For the sake of brevity, no further details are given here.

[0070] It should be understood that in this embodiment, the processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0071] The memory may include a read-only memory and a random access memory, and provides instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0072] A computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the method described in Example 1 is performed.

[0073] The method in Example 1 can be directly implemented as a hardware processor, or can be implemented using a combination of hardware and software modules in the processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, it will not be described in detail here.

[0074] A computer program product includes a computer program, which implements the method described in embodiment 1 when executed by a processor.

[0075] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions contained in program modules, which are executed in a device on a real or virtual processor of a target to perform the process / method described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided between program modules as needed. The machine-executable instructions for the program modules can be executed in local or distributed devices. In distributed devices, program modules can be located in local and remote storage media.

[0076] The computer program code for implementing the method of the present invention can be written in one or more programming languages. These computer program codes can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the computer or other programmable data processing device, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on a computer, partially on a computer, as an independent software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0077] In the context of the present invention, computer program code or related data can be carried by any appropriate carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals include electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals, and the like.

[0078] Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0079] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. An intelligent all-terrain driving control method, characterized in that: include: Determining, based on the acquired road condition image and radar data corresponding to the location and time, current weather information, environmental information, and location information, whether the current road condition is at least one of rain and snow, wading, mud, sand, rock, and off-road; According to the determined current road conditions, the system matches the current optimal driving mode and switches to the corresponding optimal driving mode, while sending a control strategy instruction for executing the corresponding optimal driving mode to the execution system; According to the control strategy instructions, the power system, steering system, suspension system, braking system and cockpit system are controlled to perform corresponding actions; and when there are two or more current road conditions, the corresponding control strategy is executed according to the set priority of the execution system.

2. The intelligent all-terrain driving control method according to claim 1, characterized in that: The camera collects planar and stereo images of the road surface within the wheel trajectory range; the lidar collects three-dimensional point cloud data of the road surface and surrounding objects, as well as environmental information such as the distance and speed of surrounding obstacles; the map identifies the current geographical location and matches the current road section attributes; and the temperature and humidity sensors as well as rain and snow sensors collect weather information such as ambient temperature, precipitation, and humidity.

3. The intelligent all-terrain driving control method according to claim 2, characterized in that: Use a pre-trained semantic segmentation model to perform pixel-level annotation on plane images and output image data including road surface material and status labels; Fuse the planar image and the stereo image to generate a dense point cloud and optimize the point cloud coordinates, outputting point cloud data associated with visual features; Along the center lines of the left and right wheel trajectories, the point cloud elevation values ​​are sampled at intervals to generate two continuous elevation curves.

4. The intelligent all-terrain driving control method according to claim 3, characterized in that: Build a road condition sample library to store standard image features and point cloud features of different road conditions; Match the output image data and point cloud data with the road condition sample library to obtain preliminary road surface signals; Leveling quantification based on the left and right wheel track elevation curves, including: If the elevation standard deviation is less than the minimum threshold, the road surface is considered flat. If it is within the range of the minimum and maximum thresholds, the road surface is considered slightly uneven. If it is greater than or equal to the maximum threshold, the road surface is considered severely uneven. If the maximum elevation difference between adjacent sampling points is greater than or equal to the undulation difference threshold, it is marked as avoiding convexity or concavity; The cumulative length of consecutive sampling points with elevation differences greater than the maximum threshold is the continuous abnormal length. If the continuous abnormal length is greater than or equal to the abnormal threshold, a deceleration warning is triggered; Identify current road conditions based on preliminary road surface signals, smoothness quantification results, radar data, current weather information, environmental information, and location information.

5. The intelligent all-terrain driving control method according to claim 1, characterized in that: According to the control strategy instructions, the process of controlling the execution of corresponding actions includes: Control the power system to enter the corresponding torque distribution ratio and torque gradient; Control the steering system to enter the corresponding steering mode; Control the suspension system to adjust to the corresponding suspension damping and suspension height mode; Control the braking system to adjust to the corresponding traction control mode and body stability mode; Control the cockpit system to adjust to the corresponding comfort environment mode.

6. The intelligent all-terrain driving control method according to claim 1, characterized in that: When there are two or more road conditions, the priority order is set as braking system, power system, steering system, suspension system, and cockpit system; and the vehicle status is monitored in real time. If the deviation between the actual status and the expected status exceeds the threshold, the control parameters are dynamically corrected.

7. An intelligent all-terrain driving control system, characterized in that: include: a road condition recognition module configured to determine whether a current road condition is at least one of rain and snow, wading, mud, sand, rock, and off-road based on the acquired road condition image and radar data corresponding to the location and time, current weather information, environmental information, and location information; a decision module configured to match the current optimal driving mode according to the determined current road conditions, switch to the corresponding optimal driving mode, and send a control strategy instruction for executing the corresponding optimal driving mode to the execution system; The execution module is configured to control the power system, steering system, suspension system, braking system and cockpit system to perform corresponding actions according to the control strategy instructions; and when the current road conditions are two or more, the corresponding control strategy is executed according to the set priority of the execution system.

8. An electronic device, characterized in that: The method comprises a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method according to any one of claims 1 to 6 is completed.

9. A computer-readable storage medium, characterized in that Used to store computer instructions, which, when executed by a processor, complete the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program, which is used to implement the method according to any one of claims 1 to 6 when the computer program is executed by a processor.

Images