Vehicle control method, electronic equipment, vehicle, medium and product

Abstract

The invention discloses a vehicle control method, electronic equipment, a vehicle, a computer readable storage medium and a computer program product. The method comprises the steps of constructing a three-dimensional sand model according to current state information of a sand road surface under the condition that a vehicle runs on the sand road surface; and according to the three-dimensional sand model, controlling the vehicle to run on the sand pavement. Therefore, the three-dimensional sand model can be constructed according to the current state information of the sand road surface when the vehicle runs on the sand road surface, so that the state of the current running environment of the vehicle, such as the positions of quicksand and sand pits, can be determined according to the three-dimensional sand model, and the vehicle can be controlled to run on the sand road surface according to the three-dimensional sand model. Such as quicksand, sandpits and the like are avoided, so that the safe driving of the vehicle in the desert environment is realized.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a vehicle control method, electronic equipment, a vehicle, a computer-readable storage medium, and a computer program product. Background Art

[0002] When vehicles are driving in the desert, they are prone to sand sinking accidents due to loose sand layers or sudden changes in terrain. This not only affects driving safety, but may also cause vehicle damage or even casualties. Therefore, how to ensure the safe driving of vehicles in desert environments has become one of the current urgent problems to be solved. Summary of the Invention

[0003] The present application provides a vehicle control method, an electronic device, a vehicle, a computer-readable storage medium, and a computer program product.

[0004] The present application provides a vehicle control method, which includes:

[0005] When the vehicle is traveling on a sandy road, a three-dimensional sandy model is constructed according to current state information of the sandy road;

[0006] The vehicle is controlled to travel on the sandy road surface according to the three-dimensional sandy model.

[0007] In this way, when a vehicle is driving on a sandy road, a 3D sand model is constructed based on the current state of the sandy road, allowing the vehicle to safely navigate the sandy road based on the 3D sand model. This 3D sand model can be used to determine the location of surrounding dangerous areas, such as quicksand and sand pits. The 3D sand model can then be used to control the vehicle to avoid these dangerous areas, such as planning a route to avoid quicksand and sand pits, thereby ensuring safe driving in desert environments.

[0008] In some embodiments, the current status information is obtained through a first vehicle component, and the first vehicle component includes at least one of a geological radar, a laser radar, and a sounding component.

[0009] In this way, geological radar, lidar, and penetrating components can acquire real-time information about the current state of the sandy road surface. By collecting this information through a variety of sand state acquisition devices, a three-dimensional sand model can be constructed, providing multi-dimensional data for vehicle control and effectively overcoming the detection limitations of a single sensor in complex desert environments.

[0010] In some embodiments, the current state information includes at least one of point cloud image information, sand density, sand flow direction, ground hardness, and ground shear strength.

[0011] The current state information thus includes at least one of point cloud image information, sand density, sand flow direction, ground hardness, and ground shear strength. Based on this current state information, a three-dimensional sand model can be constructed, providing multi-dimensional data for vehicle control.

[0012] In some embodiments, the current state information includes point cloud image information. When the vehicle is traveling on a sandy road, constructing a three-dimensional sand model based on the current state information of the sandy road includes:

[0013] When the vehicle is traveling on a sandy road, pre-processing the point cloud image information to determine processed point cloud image information;

[0014] A three-dimensional sand model is constructed based on the processed point cloud image information.

[0015] In this way, a 3D sand model is constructed based on the pre-processed point cloud image information. In this way, the pre-processing can remove the interference information in the point cloud image, providing a more accurate data basis for constructing the 3D sand model.

[0016] In certain embodiments, the preprocessing includes grayscale processing and / or noise reduction processing.

[0017] By grayscaling and / or performing noise reduction on the point cloud image information, a 3D sand model can be constructed based on the processed point cloud image information. Grayscaling can remove color interference and reduce the amount of data required for computation. Noise reduction can effectively remove noise from the point cloud image while preserving its characteristic information. Grayscaling and noise reduction provide more accurate and computationally efficient data for constructing a 3D sand model.

[0018] In some embodiments, controlling the vehicle to travel on the sandy road surface based on the three-dimensional sandy model includes:

[0019] dividing the three-dimensional sand model to determine a plurality of sand areas;

[0020] The vehicle is controlled to travel on the sandy road surface based on the plurality of sandy areas.

[0021] In this way, the 3D sand model is divided and processed to identify multiple sandy areas; based on these areas, the vehicle is controlled on the sandy road. By dividing the 3D sand model, the global problem can be broken down into multiple local sub-problems, reducing the computational effort and improving computational efficiency. By quantifying the risk of sandy areas, dangerous areas can be identified, providing a basis for vehicle control routes.

[0022] In some embodiments, the dividing and processing of the three-dimensional sand model to determine a plurality of sand areas includes:

[0023] performing edge detection processing on the three-dimensional sand model to determine an edge detection processing result;

[0024] A plurality of sandy areas are determined according to the edge detection processing result.

[0025] In this way, edge detection processing is performed on the 3D sand model to determine the edge detection results. Based on the edge detection results, multiple sand regions are identified. By performing edge detection processing on the 3D sand model, the gradient values of some sudden changes in the 3D sand model can be calculated, thereby locating the region boundaries. Based on the region boundary information, the edge points are connected to form closed edge lines to divide the 3D sand model into multiple sand regions.

[0026] In some embodiments, performing edge detection on the three-dimensional sand model and determining an edge detection result includes:

[0027] Performing target feature detection on the three-dimensional sand model to determine a target feature detection result, wherein the target feature is determined according to a current time and a current position of the vehicle;

[0028] According to the target feature detection result, edge detection processing is performed on the three-dimensional sand model to determine the edge detection processing result.

[0029] In this manner, the 3D sand model is subjected to target feature detection, and a target feature detection result is determined. Based on the target feature detection result, edge detection processing is performed on the 3D sand model, and an edge detection result is determined. This allows for different target feature extraction methods to be used depending on the ambient lighting. By performing edge detection processing on the 3D sand model based on the target features, a complete edge detection result with high detail preservation and strong noise immunity can be obtained.

[0030] In some embodiments, the current state information includes at least one of sand density, sand flow direction, ground hardness, and ground shear strength, and performing edge detection processing on the three-dimensional sand model based on the target feature detection result to determine the edge detection processing result includes:

[0031] According to the target feature detection result and the gradient of the current state information, edge detection processing is performed on the three-dimensional sand model to determine the edge detection processing result.

[0032] In this way, edge detection is performed on the 3D sand model based on the target feature detection results and the gradient of the current state information to determine the edge detection results. This feature extraction method under different lighting modes can retain more edge information in the edge detection results, improving the accuracy of the results and further improving the structure of the edge detection results, providing an accurate data basis for the subsequent division of multiple sand areas.

[0033] In some embodiments, controlling the vehicle to travel on the sandy road surface based on the plurality of sandy areas includes:

[0034] The vehicle is controlled to travel on the sandy road surface according to a target sandy area among the plurality of sandy areas.

[0035] In this way, the vehicle is controlled to drive on the sandy road according to the target sandy area among the multiple sandy areas. In this way, the non-dangerous sandy area obtained by identifying the sandy area can be used as the target sandy area for route planning, providing a range basis for route planning, thereby avoiding dangerous areas and achieving safe driving.

[0036] In certain embodiments, the method further comprises:

[0037] The target sand area among the multiple sand areas is determined according to the sand area and a pre-trained area detection model.

[0038] In this way, the target sand area among multiple sand areas is determined based on the sand area and the pre-trained area detection model. In this way, the pre-trained area detection model can be used to detect the danger level of the sand area and determine the target sand area among multiple sand areas, providing data support for safe driving.

[0039] In some embodiments, the current state information includes at least one of sand density, sand flow direction, ground hardness, and ground shear strength, and the method further includes:

[0040] The target sand area is determined based on the looseness information and roughness information of the sandy ground, as well as the current state information and straight line features corresponding to the sandy area, wherein the looseness information is determined based on the feedback force provided to the vehicle by the road currently located by the vehicle.

[0041] In this way, the target sand area is determined based on the looseness and roughness information of the sand surface, as well as the current state information and straight line characteristics corresponding to the sand area. The looseness information is determined based on the feedback force provided to the vehicle by the road surface it is currently on. By calculating and processing these parameters and characteristics, the sand area can be judged based on multi-dimensional parameters, thus accurately determining the target sand area and providing accurate data for route planning and safe driving.

[0042] In some embodiments, determining the target sand area based on the looseness information and roughness information of the sandy ground, as well as the current state information and straight line features corresponding to the sand area, includes:

[0043] Determining a passability test result according to the looseness information and the roughness information;

[0044] The target sand area is determined according to the passability detection result, the current state information corresponding to the sand area, and the straight line feature.

[0045] In this way, the traversability test results are determined based on the looseness and roughness information. The target sand area is then determined based on the traversability test results, along with the current state information and line characteristics corresponding to the sand area. In this way, by combining various sand area information such as looseness, roughness, current state information, and line characteristics, the danger level of the current sand area can be accurately determined through calculation, thereby identifying non-dangerous areas as target sand areas.

[0046] In some embodiments, controlling the vehicle to travel on the sandy road surface according to a target sandy area among the plurality of sandy areas includes:

[0047] Planning a driving route according to the target sandy area;

[0048] The vehicle is controlled to travel along the travel route on the sandy road surface.

[0049] In this way, a driving route is planned based on the target sandy area, and the vehicle is controlled to drive along the driving route on the sandy road surface. In this way, route planning within the target sandy area can effectively improve the off-road vehicle's maneuverability in complex desert environments, avoiding movement obstructions caused by sandy terrain, and thus achieving safe driving when the vehicle follows the planned route.

[0050] In some embodiments, planning a driving route according to the target sandy area includes:

[0051] The vehicle driving parameters are determined according to the target sand area to plan the driving route, wherein the vehicle driving parameters include route curvature parameters and / or vehicle steering angle parameters.

[0052] In this way, the vehicle driving parameters are determined based on the target sand area to plan the driving route, wherein the vehicle driving parameters include the route curvature parameter and / or the vehicle steering angle parameter. In this way, the optimal driving route can be obtained by solving the route planning formula using the route curvature parameter and the vehicle steering angle parameter.

[0053] In certain embodiments, the method further comprises:

[0054] According to other areas except the target sand area among the multiple sand areas, first preset prompt information is fed back to the target object.

[0055] In this way, the first preset prompt information is fed back to the target object based on the other areas in the multiple sand areas other than the target sand area. In this way, based on the detected dangerous areas, the current route status can be fed back in real time, reminding the user to pay attention to driving safety, and route planning can be carried out based on the detection results to achieve safe driving.

[0056] In some embodiments, the current state information is determined based on a sampling result of a first vehicle component on the vehicle, and the method further includes:

[0057] When the vehicle is traveling on a sandy road and the current speed of the vehicle is less than or equal to a preset threshold, controlling the first vehicle component to perform the sampling at a first time interval; and / or,

[0058] When a vehicle is traveling on a sandy road and the current speed of the vehicle is greater than the preset threshold, the first vehicle component is controlled to perform the sampling at a second time interval, where the second time interval is greater than the first time interval, and the current state information is determined by interpolating the sampling results.

[0059] In this way, when the vehicle is traveling on a sandy road and its current speed is less than or equal to a preset threshold, the first vehicle component is controlled to perform sampling at a first time interval. When the vehicle is traveling on a sandy road and its current speed is greater than the preset threshold, the first vehicle component is controlled to perform sampling at a second time interval, which is greater than the first time interval. The current state information is determined by interpolating the sampling results. In this way, by updating the sampling frequency in real time, more accurate sand state data can be collected, providing an accurate data basis for subsequent 3D sand modeling.

[0060] In certain embodiments, the method further comprises:

[0061] When the vehicle is in a sand-trapped condition, the operating state of the second vehicle component is adjusted to exit the sand-trapped condition.

[0062] In this way, when the vehicle is stuck in sand, the operating state of the second vehicle component is adjusted to escape the sand condition. In this way, in the event that the vehicle is inevitably stuck in sand, the vehicle components can be adjusted to escape the sand, thereby assisting the user in escaping the predicament and ensuring safety in sandy driving.

[0063] In certain embodiments, the second vehicle component includes at least one of a tire, an engine, and a suspension.

[0064] In this case, the second vehicle component includes at least one of a tire, an engine, and a suspension. By adjusting multiple components, such as reducing tire pressure, reducing the engine's maximum torque speed, and adjusting the suspension height, the interaction force, dynamics, and center of gravity between the vehicle and the sand can be adjusted, intelligently freeing the vehicle from the sand.

[0065] In some embodiments, the current state information includes multiple sub-information. When the vehicle is traveling on a sandy road, constructing a three-dimensional sand model based on the current state information of the sandy road includes:

[0066] When the vehicle is traveling on the sandy road, acquiring the sub-information;

[0067] When the acquisition time of each sub-information satisfies a preset condition, the three-dimensional sand model is constructed according to the multiple sub-information.

[0068] In this way, when a vehicle is driving on a sandy road, sub-information is acquired. If the acquisition time of each sub-information satisfies a preset condition, a 3D sand model is constructed based on the multiple sub-information. By performing consistency checks on the sub-information timestamps, the temporal consistency and availability of the sub-information can be ensured, thereby ensuring the construction of an accurate 3D sand model.

[0069] In certain embodiments, the method further comprises:

[0070] If the acquisition time of each sub-information does not meet the preset condition, performing a status detection on the vehicle and determining a status detection result;

[0071] The vehicle is controlled according to the state detection result.

[0072] In this way, if the acquisition time of each sub-information item does not meet the preset conditions, the vehicle status is checked and the status detection results are determined. The vehicle is then controlled based on the status detection results. This can also detect vehicle faults by determining the acquisition time of the sub-information, thereby performing vehicle status detection. The status detection results can be fed back to the user to prevent safety accidents.

[0073] In some embodiments, controlling the vehicle according to the state detection result includes:

[0074] When the status detection result is the first result, obtaining new sub-information; and / or,

[0075] When the state detection result is the first result, second preset prompt information is fed back to the target object.

[0076] In this way, if the status detection result is the first result, new sub-information is obtained; and / or if the status detection result is the second result, a second preset prompt information is fed back to the target object. In this way, by judging the status detection result, that is, judging the vehicle's health data, it is possible to determine whether the vehicle is in a normal state. If the vehicle is in a normal state, sand information collection begins. If it is in an abnormal state, a prompt is issued to the user to inform the user that the vehicle's sand driving function has malfunctioned and needs to be repaired.

[0077] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0078] An embodiment of the present application provides a vehicle, including the above-mentioned electronic device, to implement the steps of the above-mentioned method.

[0079] An embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by one or more processors, the steps of the above method are implemented.

[0080] An embodiment of the present application provides a computer program product, including a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0081] The electronic device, vehicle, computer-readable storage medium, and computer program product provided in the embodiments of the present application construct a three-dimensional sand model based on the current state information of the sandy road surface when the vehicle is traveling on a sandy road surface; and control the vehicle's travel on the sandy road surface based on the three-dimensional sand model. Thus, when the vehicle is traveling on a sandy road surface, a three-dimensional sand model can be constructed based on the current state information of the sandy road surface. The three-dimensional sand model can then be used to determine the state of the vehicle's current driving environment, such as the location of quicksand or sand pits. The three-dimensional sand model can then be used to control the vehicle's travel on the sandy road surface, such as avoiding quicksand or sand pits, thereby ensuring safe driving in desert environments.

[0082] Additional aspects and advantages of the embodiments of the present application will be given in part in the description below, and in part will become obvious from the description below, or will be learned through practice of the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0084] Figure 1 This is one of the flowcharts of the vehicle control method according to certain embodiments of the present application;

[0085] Figure 2 This is a second flow chart of a vehicle control method according to certain embodiments of the present application;

[0086] Figure 3 is a data processing diagram of a vehicle control method according to certain embodiments of the present application;

[0087] Figure 4 This is a third flow chart of a vehicle control method according to certain embodiments of the present application;

[0088] Figure 5 This is a fourth flow chart of a vehicle control method according to certain embodiments of the present application;

[0089] Figure 6 This is a fifth flow chart of a vehicle control method according to certain embodiments of the present application;

[0090] Figure 7 This is the sixth flow chart of the vehicle control method according to certain embodiments of the present application;

[0091] Figure 8 This is the seventh flow chart of the vehicle control method according to certain embodiments of the present application;

[0092] Figure 9 This is the eighth flow chart of the vehicle control method according to certain embodiments of the present application;

[0093] Figure 10 This is a ninth flowchart of a vehicle control method according to certain embodiments of the present application;

[0094] Figure 11 This is a tenth flowchart of a vehicle control method according to certain embodiments of the present application;

[0095] Figure 12 This is one of the schematic diagrams of part of the principle of the vehicle control method in certain embodiments of the present application;

[0096] Figure 13 This is a flowchart of a vehicle control method according to certain embodiments of the present application.

[0097] Figure 14 This is one of the effect schematic diagrams of the vehicle control method of certain embodiments of the present application;

[0098] Figure 15 This is a twelfth flowchart of a vehicle control method according to certain embodiments of the present application;

[0099] Figure 16 This is the thirteenth flow chart of the vehicle control method according to certain embodiments of the present application;

[0100] Figure 17 This is the fourteenth flow chart of the vehicle control method according to certain embodiments of the present application;

[0101] Figure 18 FIG15 is a flowchart of a vehicle control method according to certain embodiments of the present application;

[0102] Figure 19 This is the second schematic diagram of the effect of the vehicle control method of certain embodiments of the present application;

[0103] Figure 20 This is a sixteenth flow chart of a vehicle control method according to certain embodiments of the present application;

[0104] Figure 21 FIG17 is a flow chart of a vehicle control method according to certain embodiments of the present application;

[0105] Figure 22 This is the second schematic diagram of the partial principle of the vehicle control method in certain embodiments of the present application;

[0106] Figure 23 FIG18 is a flow chart of a vehicle control method according to certain embodiments of the present application;

[0107] Figure 24 This is the nineteenth flow chart of the vehicle control method of certain embodiments of the present application. DETAILED DESCRIPTION

[0108] The embodiments of the present application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of the present application, and should not be understood as limiting the embodiments of the present application.

[0109] Due to the loose geological structure of desert areas, vehicles driving through them often face the problem of getting stuck in the sand. The adhesion between sand grains in the desert is extremely weak, making it easy for vehicle wheels to sink into the sand. This increases the contact area between the tires and the sand surface, preventing the vehicle from moving forward properly. Furthermore, sand is fluid, so driving a vehicle stuck in a sand pit will cause the sand around the wheels to rapidly migrate, forming a sand pit and further exacerbating the vehicle's sinking. For example, when a vehicle attempts to accelerate to escape, the rapidly spinning tires will throw sand backwards, making the sand layer beneath the wheels even looser, creating a vicious cycle.

[0110] Furthermore, desert terrain is highly undulating, with widespread dunes, ridges, and other features with significant differences in height and slope. When driving on sandy terrain, sudden and frequent uphill and downhill climbs or crossing dune ridges can lead to dangerous situations such as chassis scrapes, vehicle tilt, and even rollovers. Furthermore, hidden sand valleys or windswept depressions can be covered by surface shifting sand, making them difficult for users to identify and easily lead to dangerous areas. This can lead to vehicles being stuck in the desert sand, not just a superficial obstacle to driving but also a series of safety issues. For example, a vehicle stuck in the sand can lose power. If the vehicle is in a blind spot, such as the leeward side of a dune, it is vulnerable to collisions with following vehicles. Reckless attempts to free the vehicle, such as aggressively pressing the accelerator, can further stall the vehicle, leading to mechanical failures such as driveshaft breakage and transmission overheating. Furthermore, prolonged periods of time stranded in the desert can put occupants at risk of heatstroke, dehydration, and other life-threatening conditions. Using inadequate towing equipment or the wrong towing angle during a rescue can cause structural deformation, damage to chassis components, and even secondary accidents.

[0111] Based on the above questions, please refer to Figure 1 , an embodiment of the present application provides a vehicle control method, the method comprising:

[0112] 01: When a vehicle is driving on a sandy road, a 3D sand model is constructed based on the current state information of the sandy road.

[0113] 02: Control the vehicle to drive on the sandy road based on the 3D sand model.

[0114] Embodiments of the present application provide a vehicle control device. The vehicle control method of the embodiments of the present application can be implemented by the vehicle control device of the embodiments of the present application. Specifically, the vehicle control device includes a three-dimensional sand model construction module and a vehicle control module. The three-dimensional sand model construction module is configured to construct a three-dimensional sand model based on current state information of the sandy road surface when the vehicle is traveling on the sandy road surface. The vehicle control module is configured to control the vehicle's travel on the sandy road surface based on the three-dimensional sand model.

[0115] The present application also provides an electronic device comprising a memory and a processor. The vehicle control method of the present application can be implemented by the electronic device of the present application. Specifically, the memory stores a computer program, and the processor is configured to construct a three-dimensional sand model based on current state information of the sandy road surface when the vehicle is traveling on the sandy road surface. The processor is further configured to control the vehicle's travel on the sandy road surface based on the three-dimensional sand model.

[0116] Specifically, current state information refers to sand feature data collected by various onboard sand state acquisition devices, such as radars. When a vehicle travels on a sandy road, these devices collect sand features and obtain current sand state information, such as point clouds, sand hardness, and sand shear strength.

[0117] A 3D sand model is a data model composed of a large number of discrete point clouds, constructed from current state information. This model can be used to identify dangerous areas in the sand, such as quicksand and sand pits, and control vehicle movement to avoid these dangerous areas, preventing accidents involving vehicles becoming trapped in the sand.

[0118] When a vehicle is driving on a sandy road, the current state information is calculated and processed to construct a three-dimensional sand model. Compared to the limitations of traditional single-point perception of the sandy environment, the three-dimensional sand model of the present embodiment can combine multi-dimensional sand feature data to comprehensively perceive the sand layer environment and identify dangerous areas in the sand, thereby providing strong data support for intelligent driving, path planning, and terrain adaptation.

[0119] In summary, in the embodiments of the present application, when a vehicle is traveling on a sandy road, a three-dimensional sand model can be constructed based on the current state information of the sandy road surface, thereby controlling the vehicle's safe driving on the sandy road surface based on the three-dimensional sand model. Thus, a three-dimensional sand model can be constructed based on the current state information of the sandy road surface. Based on the three-dimensional sand model, the location of surrounding dangerous areas, such as quicksand and sand pits, can be determined. Furthermore, based on the three-dimensional sand model, the vehicle can be controlled to avoid dangerous areas, such as planning a driving route to avoid quicksand and sand pits, thereby ensuring safe driving in desert environments.

[0120] In some embodiments, the current status information is obtained through a first vehicle component, and the first vehicle component includes at least one of a geological radar, a lidar, and a sounding component.

[0121] Specifically, the first vehicle component refers to a multi-type sand status collection device on the vehicle, which is used to collect the current status information of the sand in real time, including at least one of a geological radar, a lidar, and a probing component such as a mechanical probing.

[0122] Geological radar can measure and collect data on the sandy terrain where a vehicle is traveling, providing information on the sand's density. Using contact probes, the hardness and shear strength of the sand can be determined. LiDAR can scan the sandy terrain where a vehicle is traveling, generating point cloud information. Multiple types of state acquisition devices can provide three-dimensional perception of the area in front of the vehicle, providing multi-dimensional data for vehicle control.

[0123] In this way, geological radar, lidar, and penetrating components can acquire real-time information about the current state of the sandy road surface. By collecting this information through a variety of sand state acquisition devices, a three-dimensional sand model can be constructed, providing multi-dimensional data for vehicle control and effectively overcoming the detection limitations of a single sensor in complex desert environments.

[0124] In some embodiments, the current state information includes at least one of point cloud image information, sand density, sand flow direction, ground hardness, and ground shear strength.

[0125] Specifically, point cloud image information is a discrete point data set in three-dimensional space. Each discrete point contains three-dimensional spatial coordinates and characteristic attributes such as color, reflection intensity, texture information, etc., which are used to describe the spatial structure of the sand.

[0126] Sand density is the mass of sand per unit volume and is used to assess the bearing capacity of sand, such as the risk of sinking when a vehicle passes through. The higher the sand density, the greater the bearing capacity. Sand flow direction is the direction and trend of movement of sand particles on the surface of the sand, which is mainly affected by wind direction and is used to specify headwind / tailwind driving strategies. For example, when driving against the wind, sand particles in front of the vehicle are blown towards the tires, which may increase resistance and require changing direction. Ground hardness is the ability of sand to resist external pressure and is used to assess the suitability of vehicle driving. Surfaces with high hardness are suitable for vehicle driving, while areas with low hardness require low speed driving or detours. Ground shear strength is the ultimate ability of sand to resist shear failure and is used for driving force design. Sand with low shear strength, that is, fine sand, requires greater torque to drive the tires.

[0127] Current state information, including point cloud image information, sand density, sand flow direction, ground hardness, and ground shear strength, is acquired by the first vehicle component. Based on this current state information, a three-dimensional sand model can be constructed, providing multi-dimensional data for vehicle control.

[0128] The current state information thus includes at least one of point cloud image information, sand density, sand flow direction, ground hardness, and ground shear strength. Based on this current state information, a three-dimensional sand model can be constructed, providing multi-dimensional data for vehicle control.

[0129] See also Figure 2 In some embodiments, step 01 (constructing a three-dimensional sand model based on current state information of the sand road when the vehicle is traveling on the sand road) includes:

[0130] 011: When the vehicle is driving on a sandy road, pre-process the point cloud image information and determine the processed point cloud image information;

[0131] 012: Construct a 3D sand model based on the processed point cloud image information.

[0132] In certain embodiments, the 3D sand model construction module is further configured to pre-process the point cloud image information and determine processed point cloud image information when the vehicle is traveling on a sandy road. The 3D sand model construction module is further configured to construct a 3D sand model based on the processed point cloud image information.

[0133] In certain embodiments, the processor is further configured to pre-process the point cloud image information when the vehicle is traveling on a sandy road surface to determine processed point cloud image information. The processor is further configured to construct a three-dimensional sand model based on the processed point cloud image information.

[0134] Specifically, preprocessing point cloud image information is a key prerequisite for building a 3D point cloud model. The point cloud image information acquired by LiDAR may contain various interference factors. Direct use can lead to insufficient model accuracy and low computational efficiency. Preprocessing the point cloud image information, such as grayscale conversion and noise reduction, is necessary to remove interference from the point cloud image and obtain accurate and easily computable processed point cloud image information, thereby improving the accuracy and computational efficiency of the 3D sand model.

[0135] The following Figure 3 Take the following as an example to explain the construction of a three-dimensional sand model:

[0136] Taking a single cycle as an example, a LiDAR scan acquires raw ADC data, or raw point cloud image information. This raw point cloud information undergoes pipeline processing, such as range- and Doppler-dimensional FFT processing, followed by layered density mapping. The final output is a density point cloud, or accurate and easily calculated processed point cloud image information. This processed point cloud information can then be used to construct a 3D sand model.

[0137] In this way, a 3D sand model is constructed based on the pre-processed point cloud image information. In this way, the pre-processing can remove the interference information in the point cloud image, providing a more accurate data basis for constructing the 3D sand model.

[0138] In some embodiments, pre-processing includes grayscale processing and / or noise reduction processing.

[0139] Specifically, grayscale processing is used to reduce data processing workload and highlight the structural features of point cloud images. Grayscale processing extracts the brightness data from the point cloud image information and converts the color point cloud image into a grayscale point cloud image. This allows subsequent processing to only focus on the brightness data of the point cloud image, reducing data processing workload and improving the system's computational efficiency. Grayscale processing can also remove color noise, such as color casts caused by light reflections, and highlight structural features such as shapes and contours within the point cloud image.

[0140] Denoising can remove noise from an image while preserving its key features. In one example, denoising can smooth a grayscale image using a Gaussian filter, correcting outliers in the point cloud image to reasonable values consistent with domain characteristics. By removing interference obscured by the point cloud image and preserving its features, feature extraction efficiency can be improved.

[0141] In one example, when the point cloud image information is affected by both color and noise, grayscale processing and noise reduction processing can be performed sequentially to obtain accurate and easily calculated processed point cloud image information. If the point cloud image information is only affected by color, grayscale processing alone can be used to obtain processed point cloud shadow information; if the point cloud image information is only affected by noise, noise reduction processing alone can be used to obtain processed point cloud shadow information.

[0142] By grayscaling and / or performing noise reduction on the point cloud image information, a 3D sand model can be constructed based on the processed point cloud image information. Grayscaling can remove color interference and reduce the amount of data required for computation. Noise reduction can effectively remove noise from the point cloud image while preserving its characteristic information. Grayscaling and noise reduction provide more accurate and computationally efficient data for constructing a 3D sand model.

[0143] See also Figure 4 In some embodiments, step 02 (controlling the vehicle to travel on the sandy road according to the three-dimensional sandy model) includes:

[0144] 021: Divide and process the three-dimensional sand model to determine multiple sand areas;

[0145] 022: Control the vehicle to drive on the sandy road based on multiple sandy areas.

[0146] In some embodiments, the vehicle control module is further configured to divide and process the three-dimensional sand model to determine a plurality of sand areas; the vehicle control module is further configured to control the vehicle to travel on the sandy road surface based on the plurality of sand areas.

[0147] In certain embodiments, the processor is further configured to divide and process the three-dimensional sand model to determine a plurality of sand areas; and the processor is further configured to control the vehicle to travel on the sandy road surface based on the plurality of sand areas.

[0148] Specifically, sandy areas refer to subregions within a 3D sand model, obtained by segmenting the 3D sand model. 3D sand models typically contain millions of point cloud data points. For example, a LiDAR generates approximately 100,000 points per second. Directly processing this global data would result in an explosion in computational complexity. Therefore, adaptive segmentation of the 3D sand model based on terrain features is required. Edge detection is performed on the 3D sand model to divide the model into multiple sandy areas, reducing computational complexity.

[0149] By extracting and statistically analyzing the characteristic information of each sandy area, the area can be quantified, and it can be determined whether it is a dangerous area, providing a route basis for subsequent vehicle control.

[0150] In this way, the 3D sand model is divided and processed to identify multiple sandy areas; based on these areas, the vehicle is controlled on the sandy road. By dividing the 3D sand model, the global problem can be broken down into multiple local sub-problems, reducing the computational effort and improving computational efficiency. By quantifying the risk of sandy areas, dangerous areas can be identified, providing a basis for vehicle control routes.

[0151] See also Figure 5 In some embodiments, step 021 (dividing the three-dimensional sand model to determine multiple sand areas) includes:

[0152] 0211: Perform edge detection processing on the three-dimensional sand model and determine the edge detection processing result;

[0153] 0212: Based on the edge detection processing results, multiple sand areas are determined.

[0154] In certain embodiments, the vehicle control module is further configured to perform edge detection on the three-dimensional sand model and determine edge detection results. The vehicle control module is further configured to determine multiple sand areas based on the edge detection results.

[0155] In some embodiments, the processor is further configured to perform edge detection processing on the three-dimensional sand model and determine edge detection processing results. The processor is further configured to determine multiple sand areas based on the edge detection processing results.

[0156] Specifically, edge detection processing is used to detect the regional boundaries of the 3D sand model, allowing it to be divided into different regions based on these boundaries. The edge detection processing results represent regional boundary information. Edge detection processing can calculate the gradient values of target features of the model, such as height, slope, and curvature, capturing sudden changes in target feature values, thereby accurately locating the boundaries of different regions. In this way, the 3D sand model can be divided into multiple sand regions based on this regional boundary information.

[0157] In one example, for sand elevation grid data, the gradient values of slope and curvature are calculated for each point. Points with slopes greater than a set threshold constitute preliminary edge points. These points typically correspond to the ridges of sand dunes and the steep edges of sand pits, thereby identifying ridges and steep slope edges. Combined with curvature analysis, areas with negative curvature (depressions) and large slopes can be identified as the boundaries of sand pits, thereby identifying the edges of depressions. Using edge points as boundaries, the sand model can be divided into multiple closed areas.

[0158] In this way, edge detection processing is performed on the 3D sand model to determine the edge detection results. Based on the edge detection results, multiple sand regions are identified. By performing edge detection processing on the 3D sand model, the gradient values of some sudden changes in the 3D sand model can be calculated, thereby locating the region boundaries. Based on the region boundary information, the edge points are connected to form closed edge lines to divide the 3D sand model into multiple sand regions.

[0159] See also Figure 6 In some embodiments, step 0211 (performing edge detection processing on the three-dimensional sand model and determining edge detection processing results) includes:

[0160] 02111: Detecting target features on the three-dimensional sand model and determining target feature detection results, wherein the target features are determined based on the current time and the current position of the vehicle;

[0161] 02112: Based on the target feature detection results, perform edge detection processing on the three-dimensional sand model to determine the edge detection processing results.

[0162] In certain embodiments, the vehicle control module is further configured to detect target features on the three-dimensional sand model and determine a target feature detection result, wherein the target features are determined based on the current time and the current position of the vehicle. The vehicle control module is further configured to perform edge detection processing on the three-dimensional sand model based on the target feature detection result and determine an edge detection processing result.

[0163] In certain embodiments, the processor is further configured to detect target features on the three-dimensional sand model and determine a target feature detection result, wherein the target features are determined based on the current time and the current location of the vehicle. The processor is further configured to perform edge detection processing on the three-dimensional sand model based on the target feature detection result and determine an edge detection processing result.

[0164] Specifically, target features refer to different feature extraction methods for the 3D sand model, which can be determined by the current time and vehicle position. The current time and position refer to the real-time time and absolute position of the vehicle traveling on the sand. Based on the time and position, it is possible to determine whether the current sand scene is daytime or nighttime, that is, the light intensity. Based on the light intensity, different feature extraction methods are used to determine the target features.

[0165] In one example, if the vehicle is currently in a daytime sandy scene with sufficient light, Haar (Oriented FAST and Rotated BRIEF) and ORB (Oriented FAST and Rotated BRIEF) feature extraction technologies can be used as target features to perform feature extraction processing on the 3D sand model to obtain target feature detection results. If the vehicle is currently in a nighttime sandy scene with insufficient light, the 3D point cloud model can use Forward-Looking Infrared (FLIR) fusion technology to obtain target features. The key parameters of the Haar and ORB feature extraction technologies and the FLIR infrared fusion technology are shown in Table 1:

[0166] Table 1

[0167] model Resolution Frame rate Processing algorithm Daytime Mode 1920*1080 30fps Haar and ORB feature extraction Night Mode 1280*720 15fps FLIR Infrared Fusion

[0168] Haar features, by calculating pixel grayscale differences, can quickly locate local features, reduce false edges, and focus on high-contrast structural areas for edge detection. ORB features can enhance the description of edge feature points, improving edge connectivity and feature differentiation for subsequent edge detection. FLIR infrared fusion captures ambient infrared radiation to create a visual thermal distribution image, providing nighttime feature information for subsequent edge detection.

[0169] In day mode, edge detection is performed on 3D sand models based on strong Haar response areas and ORB keypoints, resulting in complete, highly detail-retaining, and noise-resistant edge detection results. In night mode, edge detection is performed even in dark environments based on FLIR thermal imagery, achieving accurate results.

[0170] In this manner, the 3D sand model is subjected to target feature detection, and a target feature detection result is determined. Based on the target feature detection result, edge detection processing is performed on the 3D sand model, and an edge detection result is determined. This allows for different target feature extraction methods to be used depending on the ambient lighting. By performing edge detection processing on the 3D sand model based on the target features, a complete edge detection result with high detail preservation and strong noise immunity can be obtained.

[0171] See also Figure 7 In some embodiments, step 02112 (performing edge detection processing on the three-dimensional sand model based on the target feature detection result to determine the edge detection processing result) includes:

[0172] 021121: Perform edge detection on the three-dimensional sand model based on the target feature detection result and the gradient of the current state information to determine the edge detection result.

[0173] In certain embodiments, the vehicle control module is further configured to perform edge detection processing on the three-dimensional sand model based on the target feature detection result and the gradient of the current state information to determine the edge detection processing result.

[0174] In certain embodiments, the vehicle control module is further configured to perform edge detection processing on the three-dimensional sand model based on the target feature detection result and the gradient of the current state information to determine the edge detection processing result.

[0175] Specifically, the gradient refers to the rate of change of pixels in an image. The gradient of the current state information includes horizontal and vertical gradients. Based on the target feature detection results, the edge detection results can be determined by performing gradient calculation on the 3D sand model.

[0176] In daytime mode, by assigning weights to strong Haar response areas and ORB key points, edges can be refined along the gradient direction to obtain accurate and continuous edge detection processing results.

[0177] In night mode, based on the thermal distribution map, by calculating the temperature gradient mutation, the boundary formed by the thermal radiation difference can be detected to obtain the edge detection processing result.

[0178] Compared with traditional edge detection processing that only performs gradient calculation, the implementation scheme of this application performs gradient calculation based on target feature detection results such as strong Haar response areas, ORB key points and thermal distribution images. It can retain more edge information for the edge detection processing results, improve the accuracy of the results, and make the structure of the edge detection results more complete, thereby providing accurate data basis for the subsequent division of multiple sand areas.

[0179] In this way, edge detection is performed on the 3D sand model based on the target feature detection results and the gradient of the current state information to determine the edge detection results. This feature extraction method under different lighting modes can retain more edge information in the edge detection results, improving the accuracy of the results and further improving the structure of the edge detection results, providing an accurate data basis for the subsequent division of multiple sand areas.

[0180] See also Figure 8 In some embodiments, step 022 (controlling the vehicle to travel on a sandy road surface according to multiple sandy areas) includes:

[0181] 0221: Control the vehicle to travel on a sandy road surface according to a target sandy area among multiple sandy areas.

[0182] In some embodiments, the vehicle control module is further configured to control the vehicle to travel on a sandy road surface according to a target sandy area among the plurality of sandy areas.

[0183] In some embodiments, the processor is further configured to control the vehicle to travel on a sandy road surface according to a target sandy area among the plurality of sandy areas.

[0184] Specifically, the target sand area refers to a non-dangerous sand area, which is obtained by identifying the sand area. By analyzing the sand status information in the sand area, it can be determined whether the sand area is dangerous. If the area is not dangerous, it can be marked as the target sand area. After all surrounding sand areas have been identified, route planning can be performed within the scope of the target sand area. It can be assumed that the vehicle is safe to drive in the target sand area, thus allowing the vehicle to avoid dangerous sand areas and achieve safe driving.

[0185] In this way, the vehicle is controlled to drive on the sandy road according to the target sandy area among the multiple sandy areas. In this way, the non-dangerous sandy area obtained by identifying the sandy area can be used as the target sandy area for route planning, providing a range basis for route planning, thereby avoiding dangerous areas and achieving safe driving.

[0186] See also Figure 9 , in some embodiments, further comprising:

[0187] 0222: Determine the target sand area among multiple sand areas based on the sand area and the pre-trained area detection model.

[0188] In certain embodiments, the vehicle control module is further configured to determine a target sand area among the plurality of sand areas based on the sand area and a pre-trained area detection model.

[0189] In certain embodiments, the processor is further configured to determine a target sand area among the plurality of sand areas based on the sand area and a pre-trained area detection model.

[0190] Specifically, the area detection model is a pre-trained machine learning model that is used to detect and process sandy areas and identify whether the sandy areas belong to non-dangerous sandy areas or dangerous areas.

[0191] By inputting the sandy area into the pre-trained area detection model, the area detection model can detect the parameters of each sandy area, identify the danger level of the area, and ultimately determine the target sandy area.

[0192] In this way, the target sand area among multiple sand areas is determined based on the sand area and the pre-trained area detection model. In this way, the pre-trained area detection model can be used to detect the danger level of the sand area and determine the target sand area among multiple sand areas, providing data support for safe driving.

[0193] See also Figure 10 , in some embodiments, further comprising:

[0194] 0223: Determine the target sand area based on the looseness information and roughness information of the sandy ground, as well as the current state information and straight line features corresponding to the sandy area, wherein the looseness information is determined based on the feedback force provided to the vehicle by the road currently on which the vehicle is located.

[0195] In certain embodiments, the vehicle control module is further configured to determine the target sand area based on looseness information and roughness information of the sandy ground, as well as current state information and straight line features corresponding to the sandy area.

[0196] In some embodiments, the processor is further configured to determine the target sand area based on looseness information and roughness information of the sandy ground, as well as current state information and straight line features corresponding to the sandy area.

[0197] Specifically, looseness refers to the compactness of sand grains. Higher looseness indicates larger gaps between sand grains, weaker bonding, and a greater tendency for the sand to flow or deform. Roughness refers to the microscopic undulations and irregularities of the sandy surface. Looseness and roughness can be used to determine whether an area is passable. If looseness and roughness do not meet passability requirements, the area is considered dangerous and not the target sand area, meaning it is a non-hazardous area.

[0198] Linear features refer to linear geometric features that characterize sandy areas. These features can be extracted and processed using line detection algorithms, such as the Hough algorithm. These linear features can be used as a reference for determining whether a sandy area is dangerous. Examples include the boundary between soft sand and solid ground, the ridgeline of a sand dune, and the edge of a dry riverbed. These naturally formed linear boundaries appear as straight or nearly straight lines in an image. By observing the density and height of sand ripples, it can be determined that sand with low, dense ripples is more compact, while sand with sparse, tall ripples is more loose.

[0199] After determining that vehicles can pass through the area based on the looseness information and roughness, the state information and straight line characteristics of the sand area are analyzed and processed based on the above-mentioned regional monitoring model to predict whether the sand area is a dangerous sand area, thereby providing a basis for path planning.

[0200] It should be noted that, in the embodiment of the present application, the straight line feature is one of the features for determining the target sand area, and other features that can achieve regional division can also be used.

[0201] In this way, the target sand area is determined based on the looseness and roughness information of the sand surface, as well as the current state information and straight line characteristics corresponding to the sand area. The looseness information is determined based on the feedback force provided to the vehicle by the road surface it is currently on. By calculating and processing these parameters and characteristics, the sand area can be judged based on multi-dimensional parameters, thus accurately determining the target sand area and providing accurate data for route planning and safe driving.

[0202] See also Figure 11 In some embodiments, step 0223 (determining a target sandy area based on the looseness information and roughness information of the sandy ground, as well as the current state information and straight line features corresponding to the sandy area) includes:

[0203] 02231: Determine the passability test result based on the looseness information and roughness information;

[0204] 02232: Determine the target sand area based on the passability detection results, as well as the current status information and straight line features corresponding to the sand area.

[0205] In some embodiments, the vehicle control module is further configured to determine a passability detection result based on the looseness information and the roughness information. The vehicle control module is further configured to determine a target sand area based on the passability detection result, current state information corresponding to the sand area, and straight line features.

[0206] In some embodiments, the processor is further configured to determine a passability detection result based on the looseness information and the roughness information. The processor is further configured to determine a target sand area based on the passability detection result, current state information corresponding to the sand area, and straight line features.

[0207] Specifically, the feasibility test result refers to the rule for data fusion of looseness and roughness information, which can be expressed by the following formula:

[0208] P pass (x,y)=0.75S radar +0.3V visionr

[0209] Among them, P pass (x,y) is the result of the feasibility test, S radarLooseness information, the value range is 0-1, 0 is completely loose, V visionr is the surface roughness coefficient, and its value range is 0.85-1.15.

[0210] The passability test result verifies whether the current road surface can support vehicle movement, providing data for subsequent determination of target sand areas. When the passability test result is greater than or equal to a preset value, the current road surface is considered to be supportive of vehicle movement; when the passability test result is less than a preset threshold, the current road surface is considered unsupportive of vehicle movement. The preset value can be set based on actual conditions.

[0211] Based on the passability detection results, that is, when the current road surface can support vehicle movement, the current state information and straight line features corresponding to the sand area are calculated to determine whether the area is the target sand area.

[0212] In some embodiments, the passability detection result, as well as the current state information and straight line features corresponding to the sand area can be input into a pre-trained danger level detection model for judgment.

[0213] In response to the passability detection results, the model can perform directional statistics on the detected straight line features and construct a directional histogram. By analyzing the distribution characteristics of the histogram and the corresponding current state information, it can analyze the optimal travel direction in the sand area, thereby determining that the area is passable and determining that the area is the target sand area, providing the optimal travel direction basis for path planning, and avoiding dangerous areas in the desert, such as soft sand, sand dunes and other obstacles.

[0214] The following Figure 12 Take the following as an example to explain how to determine the optimal direction of travel:

[0215] First, the original image, or point cloud image information, is grayscaled to remove color interference for subsequent processing. Then, Gaussian filtering is performed to remove noise interference to obtain an accurate three-dimensional sand model. The three-dimensional sand model is then gradient calculated and divided into multiple sand areas, mapping the three-dimensional space to a two-dimensional plane area. Hough line detection is then performed on the sand area to analyze the sand ripple characteristics. Finally, histogram statistics of the sand ripple characteristics can be used to determine the optimal travel direction, or the main direction.

[0216] In this way, the traversability test results are determined based on the looseness and roughness information. The target sand area is then determined based on the traversability test results, along with the current state information and line characteristics corresponding to the sand area. In this way, by combining various sand area information such as looseness, roughness, current state information, and line characteristics, the optimal direction of travel for the current sand area can be accurately determined through calculation, thereby identifying the area as the target sand area.

[0217] See also Figure 13 In some embodiments, step 0221 (controlling the vehicle to travel on a sandy road surface according to a target sandy area among a plurality of sandy areas) includes:

[0218] 02211: Plan the driving route based on the target sand area;

[0219] 02212: Control the vehicle to drive along the driving route on the sandy road.

[0220] In some embodiments, the vehicle control module is further configured to plan a driving route based on the target sandy area. The vehicle control module is further configured to control the vehicle to travel along the driving route on the sandy road surface.

[0221] In some embodiments, the processor is further configured to plan a driving route based on the target sandy area. The processor is further configured to control the vehicle to travel along the driving route on the sandy road surface.

[0222] Specifically, a driving route refers to the specific path a vehicle takes, including driving parameters such as direction and distance. Within the defined target sand area, a driving route can be planned. Controlling vehicle movement according to the driving route allows for safe driving, avoiding hazardous areas.

[0223] The following Figure 14 Taking the example of FIG. 1 as an example to explain the effect of route planning, the safe path of the vehicle route planning is in the target sand area, avoiding the dangerous area shown in the figure and driving safely. In this way, the vehicle can follow the safe path during driving to achieve safe driving.

[0224] In this way, a driving route is planned based on the target sandy area, and the vehicle is controlled to drive along the driving route on the sandy road surface. In this way, route planning within the target sandy area can effectively improve the off-road vehicle's maneuverability in complex desert environments, avoiding movement obstructions caused by sandy terrain, and thus achieving safe driving when the vehicle follows the planned route.

[0225] See also Figure 15 In some embodiments, step 02211 (planning a driving route based on the target sand area) includes:

[0226] 022111: Determine vehicle driving parameters according to the target sand area to plan a driving route, wherein the vehicle driving parameters include route curvature parameters and / or vehicle steering angle parameters.

[0227] In certain embodiments, the vehicle control module is further configured to determine vehicle driving parameters according to the target sandy area to plan a driving route.

[0228] In certain embodiments, the processor is further configured to determine vehicle driving parameters according to the target sand area to plan a driving route.

[0229] Vehicle driving parameters include path curvature parameters and vehicle steering angle parameters. Path planning can be expressed by the following formula:

[0230]

[0231] where φ i is the rate of change of path curvature, δ steer is the steering angle change, ω1ω2ω3 is the weight. The weight matrix setting is expressed by the following formula:

[0232] Q=diag(0.1,0.1,0.05,0.3,0.3,0.2,0.1)R=diag(0.5,0.5,0.3)

[0233] ω1ω2ω3 are obtained from the weight matrix Q.

[0234] According to the non-dangerous sand area, the driving route can be solved by determining the path curvature parameters and vehicle steering angle parameters.

[0235] In this way, the vehicle driving parameters are determined based on the target sand area to plan the driving route, wherein the vehicle driving parameters include the route curvature parameter and / or the vehicle steering angle parameter. In this way, the optimal driving route can be obtained by solving the route planning formula using the route curvature parameter and the vehicle steering angle parameter.

[0236] See also Figure 16 , in some embodiments, further comprising:

[0237] 021123: Feedback first preset prompt information to the target object based on other areas other than the target sand area among the multiple sand areas.

[0238] In some embodiments, the vehicle control module is further configured to feed back first preset prompt information to the target object based on other areas among the multiple sand areas except the target sand area.

[0239] In some embodiments, the processor is further configured to feed back first preset prompt information to the target object based on other areas other than the target sand area among the multiple sand areas.

[0240] Specifically, "other areas" refer to dangerous areas. The first preset information refers to user feedback. When a dangerous sandy area is identified, the system can provide key information to the user, such as "There is a high risk of vehicles getting stuck ahead. Intelligent avoidance is in progress. Please pay attention to safety." The system also displays a recommended speed range and safe distance based on real-time assessment results, providing comprehensive driving advice to the driver.

[0241] In this way, the first preset prompt information is fed back to the target object based on the other areas in the multiple sand areas other than the target sand area. In this way, based on the detected dangerous areas, the current route status can be fed back in real time, reminding the user to pay attention to driving safety, and route planning can be carried out based on the detection results to achieve safe driving.

[0242] See also Figure 17 , in some embodiments, further comprising:

[0243] 03: When the vehicle is traveling on a sandy road and the current speed of the vehicle is less than or equal to a preset threshold, controlling the first vehicle component to perform sampling at a first time interval; and / or,

[0244] 04: When the vehicle is traveling on a sandy road and the current speed of the vehicle is greater than a preset threshold, the first vehicle component is controlled to be sampled at a second time interval, where the second time interval is greater than the first time interval, and the current state information is determined by interpolating the sampling results.

[0245] In certain embodiments, the vehicle control module is further configured to, when the vehicle is traveling on a sandy road surface and the current vehicle speed is less than or equal to a preset threshold, control the first vehicle component to perform sampling at a first time interval. The vehicle control module is further configured to, when the vehicle is traveling on a sandy road surface and the current vehicle speed is greater than the preset threshold, control the first vehicle component to perform sampling at a second time interval, the second time interval being greater than the first time interval, and the current state information is determined by interpolating the sampling results.

[0246] In certain embodiments, the processor is further configured to, when the vehicle is traveling on a sandy road surface and the current vehicle speed is less than or equal to a preset threshold, control the first vehicle component to perform sampling at a first time interval. The processor is further configured to, when the vehicle is traveling on a sandy road surface and the current vehicle speed is greater than the preset threshold, control the first vehicle component to perform sampling at a second time interval, the second time interval being greater than the first time interval, and the current state information is determined by interpolating the sampling results.

[0247] Specifically, the preset threshold refers to a preset vehicle speed. When a user exceeds the preset speed, the frequency of sand status information collection may be affected by the vehicle speed. The first time interval refers to the sampling interval for the first component when the vehicle speed does not affect the collected status information. The second time interval refers to a sampling interval that is longer than the first time interval.

[0248] In one example, when a vehicle is traveling on a sandy road and the current vehicle speed is less than or equal to 20 km / h, the sandy state collecting device collects current state information at a normal collection frequency.

[0249] When the current vehicle speed is greater than 20 km / h, the sand state acquisition device will reduce the acquisition frequency and process the sampled data based on the interpolation algorithm as the current state information. This ensures that the sand information obtained can be processed accordingly at every moment. The pre-interpolation method is used to increase the refresh rate based on the following formula to implement the model:

[0250]

[0251] Where f() is the 14-DOF vehicle dynamics model and h is the nonlinear observation equation.

[0252] It should be noted that the preset threshold of 20 km / h in this application is for illustrative purposes only and should not be understood as limiting the speed of a vehicle traveling on sand. In other examples, the vehicle speed value may also be 15 km / h, 25 km / h, etc., which is not limited here and is set based on actual conditions.

[0253] In this way, when the vehicle is traveling on a sandy road and its current speed is less than or equal to a preset threshold, the first vehicle component is controlled to perform sampling at a first time interval. When the vehicle is traveling on a sandy road and its current speed is greater than the preset threshold, the first vehicle component is controlled to perform sampling at a second time interval, which is greater than the first time interval. The current state information is determined by interpolating the sampling results. In this way, by updating the sampling frequency in real time, more accurate sand state data can be collected, providing an accurate data basis for subsequent 3D sand modeling.

[0254] See also Figure 18 , in some embodiments, further comprising:

[0255] 05: When the vehicle is in a sand-trapped condition, adjust the operating state of the second vehicle component to escape from the sand-trapped condition.

[0256] In some embodiments, the vehicle control module is further configured to adjust an operating state of a second vehicle component to exit the sand-trapped condition when the vehicle is in the sand-trapped condition.

[0257] In certain embodiments, the processor is further configured to, when the vehicle is in a sand-trapped condition, adjust an operating state of a second vehicle component to exit the sand-trapped condition.

[0258] Specifically, the "sand-stuck" condition refers to a situation where the vehicle is stuck in sand and cannot travel normally. When the vehicle inevitably gets stuck in sand, the system can automatically detect abnormalities such as persistent wheel slippage, a tendency for the vehicle to sink, or significant loss of driving force, and thus automatically identify it as a "sand-stuck" condition. Secondary vehicle components refer to those that can adjust the vehicle's state, such as tires, the engine, and the suspension.

[0259] In response to the sand-trapped condition, the second vehicle component can intelligently adjust its operating state without user operation to allow the vehicle to escape from the sand-trapped condition.

[0260] In this way, when the vehicle is stuck in sand, the operating state of the second vehicle component is adjusted to escape the sand condition. In this way, in the event that the vehicle is inevitably stuck in sand, the vehicle components can be adjusted to escape the sand, thereby assisting the user in escaping the predicament and ensuring safety in sandy driving.

[0261] In certain embodiments, the second vehicle component includes at least one of a tire, an engine, and a suspension.

[0262] Specifically, Figure 19 Taking the example of the second vehicle, the operating status of each component is explained. By reducing tire pressure, the tire deformation increases, thereby increasing the contact area between the tire and the sand. The interlocking effect between sand particles improves adhesion, preventing slipping and idling, and providing a basis for escape. The engine can adjust power output by amplifying low-speed torque. For example, reducing the maximum torque output speed to 1500-2000rpm makes power output more gradual, avoiding sudden impacts that exceed the adhesion limit of the sand surface, while ensuring that the vehicle still has sufficient driving force under low-adhesion conditions. The suspension can optimize vehicle stability by adjusting its height and prevent further loss of control due to center of gravity shift. For example, by raising the vehicle body (e.g., by 2-3cm), ground clearance is increased, reducing the resistance of the chassis scraping against the sand surface, and at the same time changing the center of gravity to avoid "diving" or "lifting" due to the front or rear of the vehicle sinking.

[0263] In this case, the second vehicle component includes at least one of a tire, an engine, and a suspension. By adjusting multiple components, such as reducing tire pressure, reducing the engine's maximum torque speed, and adjusting the suspension height, the interaction force, dynamics, and center of gravity between the vehicle and the sand can be adjusted, intelligently freeing the vehicle from the sand.

[0264] See also Figure 20In some embodiments, the current state information includes multiple sub-information. Step 01 (constructing a three-dimensional sand model based on the current state information of the sandy road surface when the vehicle is traveling on the sandy road surface) includes:

[0265] 013: When the vehicle is driving on a sandy road, obtain sub-information;

[0266] 014: Construct a three-dimensional sand model based on multiple sub-information when the acquisition time of each sub-information meets the preset conditions.

[0267] In certain embodiments, the 3D sand model construction module is further configured to acquire sub-information when the vehicle is traveling on a sandy road surface. The 3D sand model construction module is further configured to construct a 3D sand model based on the plurality of sub-information when the acquisition time of each sub-information satisfies a preset condition.

[0268] In some embodiments, the processor is further configured to obtain sub-information when the vehicle is traveling on a sandy road surface. The processor is further configured to construct a three-dimensional sand model based on the plurality of sub-information when the acquisition time of each sub-information satisfies a preset condition.

[0269] Specifically, sub-information refers to sand data collected by the sand condition acquisition device, including point cloud image information, sand density, sand flow direction, ground hardness, and ground shear strength. The acquisition time of the sub-information refers to the timestamp of the sub-information. Factor information is generated by different sand condition acquisition devices, and their timestamps may vary, resulting in some impact on vehicle control after fusion.

[0270] The precondition is that the difference between the timestamps of any two sub-information items is within a preset tolerance. Temporal alignment maps the spatiotemporal consistency of the sub-information items to the digital world, providing reliable data for the subsequent construction of a 3D sand model. The precondition can be expressed as the following formula:

[0271] |t radar -t vision |<=3ms

[0272] Only when the difference between the timestamps of any two sub-information is within 3ms can the signal be considered to meet the fusion conditions, thus allowing subsequent data fusion processing to construct a three-dimensional sand model.

[0273] It should be noted that the difference between the timestamps of any two sub-messages in this application is for illustrative purposes only and should not be understood as a limitation on the corresponding quantity. In other examples, the difference between the timestamps of any two sub-messages can also be 0.5ms, 2.5ms, 5ms, etc., which is not limited here and is set according to actual circumstances.

[0274] In this way, when a vehicle is driving on a sandy road, sub-information is acquired. If the acquisition time of each sub-information satisfies a preset condition, a 3D sand model is constructed based on the multiple sub-information. By performing consistency checks on the sub-information timestamps, the temporal consistency and availability of the sub-information can be ensured, thereby ensuring the construction of an accurate 3D sand model.

[0275] See also Figure 21 , in some embodiments, further comprising:

[0276] 015: If the acquisition time of each sub-information does not meet the preset conditions, perform a status detection on the vehicle and determine the status detection result;

[0277] 016: Control the vehicle based on the status detection results.

[0278] In some embodiments, the vehicle control module is further configured to perform a status detection on the vehicle and determine a status detection result when the acquisition time of each sub-information does not meet a preset condition. The vehicle control module is further configured to control the vehicle according to the status detection result.

[0279] In some embodiments, the processor is further configured to perform a status detection on the vehicle and determine a status detection result when the acquisition time of each sub-information does not meet a preset condition. The processor is further configured to control the vehicle based on the status detection result.

[0280] Specifically, the status detection result refers to the vehicle's health data, obtained by performing a status check on the vehicle. If the acquisition time for each sub-information item does not meet the preset conditions, it indicates that a device on the vehicle is malfunctioning. For example, the sand condition acquisition device may be experiencing distorted data due to bumpy driving. Therefore, a status check is required to obtain overall vehicle health data.

[0281] By analyzing and processing the health data, it is possible to determine the specific device in the vehicle that has malfunctioned, and thus control the vehicle, such as providing reminders to the user.

[0282] The following Figure 22 Take the following as an example to explain how to control the vehicle based on the status detection results:

[0283] In one example, the main control unit will pre-process the signals of the sub-information data obtained by the vehicle. If the acquisition time of each sub-information does not meet the preset conditions, the vehicle's model computing chip will perform status detection on the vehicle to obtain the vehicle's health data and transmit it back to the main control unit. The main control unit will analyze and process the health data to determine the specific device in the vehicle that has malfunctioned, and thus control the vehicle, such as reminding the user.

[0284] In this way, if the acquisition time of each sub-information item does not meet the preset conditions, the vehicle status is checked and the status detection results are determined. The vehicle is then controlled based on the status detection results. This can also detect vehicle faults by determining the acquisition time of the sub-information, thereby performing vehicle status detection. The status detection results can be fed back to the user to prevent safety accidents.

[0285] See also Figure 23 In some embodiments, step 06 (controlling the vehicle according to the state detection result) includes:

[0286] 061: When the status detection result is the first result, obtain new sub-information; and / or,

[0287] 062: When the status detection result is the second result, second preset prompt information is fed back to the target object.

[0288] In some embodiments, the vehicle control module is further configured to obtain new sub-information when the state detection result is a first result, and to feed back second preset prompt information to the target object when the state detection result is a second result.

[0289] In some embodiments, the processor is further configured to obtain new sub-information when the state detection result is the first result. The processor is further configured to feed back second preset prompt information to the target object when the state detection result is the second result.

[0290] Specifically, the first result refers to the vehicle being in a normal state. The second result refers to the vehicle being in an abnormal state. The second preset prompt information refers to a fault prompt.

[0291] In one example, if the self-test result is normal, the sub-information is retrieved again. If the self-test result is abnormal, a prompt is issued to the user to inform the user that the vehicle's sand driving function has failed and needs to be repaired.

[0292] In this way, if the status detection result is the first result, new sub-information is obtained; and / or if the status detection result is the second result, a second preset prompt information is fed back to the target object. In this way, by judging the status detection result, that is, judging the vehicle's health data, it is possible to determine whether the vehicle is in a normal state. If the vehicle is in a normal state, sand information collection begins. If it is in an abnormal state, a prompt is issued to the user to inform the user that the vehicle's sand driving function has malfunctioned and needs to be repaired.

[0293] Please see below Figure 24 , a complete example is used to explain the vehicle control method:

[0294] The vehicle control method is mainly implemented by the sensor layer, processing layer, decision layer and execution layer.

[0295] The sensor layer uses multiple sensors, including lidar and millimeter-wave radar, to provide three-dimensional perception of the area ahead. Geological radar, lidar, and other vehicle components can collect information such as point cloud images of the sand, sand density, sand flow direction, ground hardness, and ground shear strength.

[0296] The processing layer processes the information acquired by the sensor layer. This includes pre-processing the point cloud image information to construct a 3D sand model and model sand density. It also includes identifying the 3D sand model based on surface texture and dividing the sand into areas. Furthermore, it includes calculating sand information and identifying non-hazardous sand areas, among other data processing operations to estimate vehicle dynamics.

[0297] The decision-making layer is mainly based on the data processed by the processing layer. It performs route planning and vehicle control through vehicle passability prediction, path planning, and human-machine collaborative strategies to enable users to drive safely in the sand.

[0298] In response to the sand-trapping accident, the executive layer adjusts components such as the tires, engine, and suspension, actively raising the suspension, adjusting tire pressure, and distributing power, thereby freeing the user.

[0299] The present application also provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the vehicle control method described above when the computer program is executed by a processor.

[0300] It is understood that a computer program includes computer program code. The computer program code may be in source code form, object code form, executable file, or some intermediate form. Computer-readable storage media may include any entity or device capable of carrying computer program code, recording media, USB flash drives, removable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media.

[0301] In the description of this specification, the descriptions with reference to the terms "particularly", "further", "particularly", "understandably", etc. are intended to mean that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms are not intended to refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0302] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code that includes one or more executable requests for implementing a specific logical function or step of a process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed in a different order than shown or discussed, including performing functions in a substantially simultaneous manner or in a reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0303] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A vehicle control method, characterized in that: include: When the vehicle is traveling on a sandy road, a three-dimensional sandy model is constructed according to current state information of the sandy road; The vehicle is controlled to travel on the sandy road surface according to the three-dimensional sandy model.

2. The method according to claim 1, characterized in that The current status information is obtained through a first vehicle component, and the first vehicle component includes at least one of a geological radar, a laser radar, and a probe component.

3. The method according to claim 1, characterized in that The current state information includes at least one of point cloud image information, sand density, sand flow direction, ground hardness, and ground shear strength.

4. The method according to claim 1, wherein The current state information includes point cloud image information. When the vehicle is traveling on a sandy road, constructing a three-dimensional sand model based on the current state information of the sandy road includes: When the vehicle is traveling on a sandy road, pre-processing the point cloud image information to determine processed point cloud image information; A three-dimensional sand model is constructed based on the processed point cloud image information.

5. The method according to claim 4, characterized in that The pre-processing includes grayscale processing and / or noise reduction processing.

6. The method according to claim 1, characterized in that The step of controlling the vehicle to travel on the sandy road surface according to the three-dimensional sandy model includes: dividing the three-dimensional sand model to determine a plurality of sand areas; The vehicle is controlled to travel on the sandy road surface based on the plurality of sandy areas.

7. The method according to claim 6, characterized in that The dividing process of the three-dimensional sand model to determine a plurality of sand areas includes: performing edge detection processing on the three-dimensional sand model to determine an edge detection processing result; A plurality of sandy areas are determined according to the edge detection processing result.

8. The method according to claim 7, characterized in that The performing edge detection processing on the three-dimensional sand model and determining the edge detection processing result includes: Performing target feature detection on the three-dimensional sand model to determine a target feature detection result, wherein the target feature is determined according to a current time and a current position of the vehicle; According to the target feature detection result, edge detection processing is performed on the three-dimensional sand model to determine the edge detection processing result.

9. The method according to claim 8, characterized in that The current state information includes at least one of sand density, sand flow direction, ground hardness, and ground shear strength. The edge detection processing is performed on the three-dimensional sand model based on the target feature detection result to determine the edge detection processing result, including: According to the target feature detection result and the gradient of the current state information, edge detection processing is performed on the three-dimensional sand model to determine the edge detection processing result.

10. The method according to claim 6, characterized in that The step of controlling the vehicle to travel on the sandy road surface according to the plurality of sandy areas includes: The vehicle is controlled to travel on the sandy road surface according to a target sandy area among the plurality of sandy areas.

11. The method according to claim 10, characterized in that The method further comprises: The target sand area among the multiple sand areas is determined according to the sand area and a pre-trained area detection model.

12. The method according to claim 10, characterized in that The current state information includes at least one of sand density, sand flow direction, ground hardness, and ground shear strength. The method further includes: The target sand area is determined based on the looseness information and roughness information of the sandy ground, as well as the current state information and straight line features corresponding to the sandy area, wherein the looseness information is determined based on the feedback force provided to the vehicle by the road currently located by the vehicle.

13. The method according to claim 12, characterized in that The step of determining the target sandy area according to the looseness information and roughness information of the sandy ground, and the current state information and straight line features corresponding to the sandy area includes: Determining a passability test result according to the looseness information and the roughness information; The target sand area is determined according to the passability detection result, the current state information corresponding to the sand area, and the straight line feature.

14. The method according to claim 10, characterized in that The controlling the vehicle to travel on the sandy road surface according to the target sandy area among the plurality of sandy areas comprises: Planning a driving route according to the target sandy area; The vehicle is controlled to travel along the travel route on the sandy road surface.

15. The method according to claim 14, characterized in that The step of planning a driving route according to the target sandy area includes: The vehicle driving parameters are determined according to the target sand area to plan the driving route, wherein the vehicle driving parameters include route curvature parameters and / or vehicle steering angle parameters.

16. The method according to claim 9, characterized in that The method further comprises: According to other areas except the target sand area among the multiple sand areas, first preset prompt information is fed back to the target object.

17. The method according to claim 1, wherein The current state information is determined based on a sampling result of a first vehicle component on the vehicle, and the method further includes: When the vehicle is traveling on a sandy road and the current speed of the vehicle is less than or equal to a preset threshold, controlling the first vehicle component to perform the sampling at a first time interval; and / or, When a vehicle is traveling on a sandy road and the current speed of the vehicle is greater than the preset threshold, the first vehicle component is controlled to perform the sampling at a second time interval, where the second time interval is greater than the first time interval, and the current state information is determined by interpolating the sampling results.

18. The method according to claim 1, wherein The method further comprises: When the vehicle is in a sand-trapped condition, the operating state of the second vehicle component is adjusted to exit the sand-trapped condition.

19. The method according to claim 18, characterized in that The second vehicle component includes at least one of a tire, an engine, and a suspension.

20. The method according to claim 1, wherein The current state information includes multiple sub-information. When the vehicle is traveling on a sandy road, constructing a three-dimensional sand model according to the current state information of the sandy road includes: When the vehicle is traveling on the sandy road, acquiring the sub-information; When the acquisition time of each sub-information satisfies a preset condition, the three-dimensional sand model is constructed according to the multiple sub-information.

21. The method according to claim 20, characterized in that The method further comprises: If the acquisition time of each sub-information does not meet the preset condition, performing a status detection on the vehicle and determining a status detection result; The vehicle is controlled according to the state detection result.

22. The method according to claim 21, characterized in that The controlling of the vehicle according to the state detection result includes: When the status detection result is the first result, obtaining new sub-information; and / or, When the state detection result is the first result, second preset prompt information is fed back to the target object.

23. An electronic device, characterized in that: The method comprises a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method according to any one of claims 1 to 22 is implemented.

24. A vehicle, characterized in that: The vehicle includes the electronic device of claim 23.

25. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by one or more processors, the method according to any one of claims 1 to 22 is implemented.

26. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 22 is implemented.

Images