Vehicle risk prediction method, electronic equipment and vehicle

Abstract

The invention provides a vehicle risk prediction method, electronic equipment and a vehicle, and relates to the technical field of vehicle intelligent driving, and the method comprises the steps: obtaining environment data and vehicle state data in a vehicle driving process, and calculating and determining road surface state parameters through a pre-constructed physical model according to the environment data and the vehicle state data. And based on the environment data and the vehicle state data, correcting the road surface state parameters through a scene deduction model to generate corrected road surface state parameters. Based on the environment data, the vehicle state data, the road surface state parameters and the corrected road surface state parameters, the current target risk type can be accurately determined, and the risk prediction accuracy is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent driving of vehicles, and in particular to a vehicle risk prediction method, an electronic device and a vehicle. BACKGROUND

[0002] With the popularity of off-road activities (such as outdoor exploration, emergency rescue, off-road events, etc.), the demand for risk prediction in off-road scenarios is increasingly urgent. Current risk prediction mainly uses a static strategy or a fixed threshold to provide risk warnings, which has the problem of inaccurate risk prediction. SUMMARY

[0003] Therefore, the present application aims to provide a vehicle risk prediction method, an electronic device and a vehicle to solve the problem of inaccurate risk prediction in off-road scenarios.

[0004] To achieve the above purpose, the first aspect of the present application provides a vehicle risk prediction method, comprising: obtaining environmental data and vehicle state data during vehicle driving; determining road surface state parameters through a pre-constructed physical model according to the environmental data and the vehicle state data; correcting the road surface state parameters through a scenario deduction model based on the environmental data and the vehicle state data to generate corrected road surface state parameters; determining a target risk type based on the environmental data, the vehicle state data, the road surface state parameters and the corrected road surface state parameters.

[0005] Optionally, the determination of the target risk type based on the environmental data, the vehicle state data, the road surface state parameters and the corrected road surface state parameters comprises: determining predicted road surface state parameters through a pre-set prediction algorithm based on the corrected road surface state parameters; calculating risk scores corresponding to each pre-set risk type according to the environmental data, the vehicle state data, the road surface state parameters and / or the predicted road surface state parameters; determining the target risk type according to the risk scores and a pre-set risk threshold.

[0006] Optionally, the calculation of the risk scores corresponding to each pre-set risk type according to the environmental data, the vehicle state data, the road surface state parameters and / or the predicted road surface state parameters comprises: calculating occurrence probabilities corresponding to each pre-set risk type according to the environmental data, the vehicle state data, the road surface state parameters and / or the predicted road surface state parameters; According to the occurrence probability and a preset influence coefficient corresponding to a preset risk type, the risk score is determined.

[0007] Optionally, the method further comprises: obtaining a driving level of a current driver; According to the driving level, the vehicle state data and the target risk type, a corresponding control strategy is determined, and the vehicle is controlled to operate according to the control strategy.

[0008] Optionally, the method further comprises: According to the predicted road surface state parameter, it is determined whether the road in front satisfies a preset road surface wet and slippery condition; In response to determining that the road in front satisfies the preset road surface wet and slippery condition, a driving path of the vehicle is planned according to the environmental data, and the vehicle is controlled to drive according to the planned driving path.

[0009] Optionally, the method further comprises: According to the vehicle state data, the corrected road surface state parameter, the control strategy and the driving level, driving operation prompt information is generated; The driving operation prompt information is displayed.

[0010] Optionally, the method further comprises: According to the driving level, a preset display content to be displayed is determined; According to the vehicle state data, the corrected road surface state parameter, the control strategy and / or the driving level, the driving operation prompt information is generated according to the preset display content.

[0011] Optionally, the method further comprises: Collecting an actual road surface state parameter corresponding to the corrected road surface state parameter; Determining a difference value between the corrected road surface state parameter and the actual road surface state parameter; According to the difference value, the model parameters of the scene deduction model are updated.

[0012] Based on the same inventive concept, the second aspect of the present application also provides a vehicle risk prediction device, comprising: An acquisition module configured to acquire environmental data and vehicle state data during vehicle driving; A calculation module configured to calculate and determine a road surface state parameter through a pre-constructed physical model according to the environmental data and the vehicle state data; The correction module is configured to correct the road surface state parameter based on the environment data and the vehicle state data through a scenario deduction model, to generate a corrected road surface state parameter. The determination module is configured to determine a current target risk type based on the environment data, the vehicle state data, the road surface state parameter and the corrected road surface state parameter.

[0013] Based on the same inventive concept, the third aspect of the present application further provides an electronic device including a memory, a processor and a computer program stored in the memory and executable by the processor, wherein the processor implements the method as described above when executing the computer program.

[0014] Based on the same inventive concept, the fourth aspect of the present application further provides a vehicle including the electronic device as described in the second aspect.

[0015] As can be seen from the above, the vehicle risk prediction method, the electronic device and the vehicle provided by the present application, wherein the method includes: acquiring environment data and vehicle state data in the driving process of the vehicle, and determining a road surface state parameter through a pre-constructed physical model according to the environment data and the vehicle state data. The road surface state parameter calculated through the physical model is a parameter that meets the physical law and can reflect the real-time state of the road surface. However, due to the complexity of the scene, there is often a nonlinear correlation between the environment data, the vehicle state data and the road surface state parameter, and the road surface state parameter calculated through the physical model is inaccurate. Therefore, the present application corrects the road surface state parameter based on the environment data and the vehicle state data through a scenario deduction model, to generate a corrected road surface state parameter. The scenario deduction model can identify the rules that are not accurately described by the physical model, and then can deduce the corrected road surface state parameter that is more consistent with the current terrain characteristics according to the environment data and the vehicle state data, thereby making up for the problem of insufficient scenario deduction in the existing risk prediction process. Then, based on the environment data, the vehicle state data, the road surface state parameter and the corrected road surface state parameter, the current target risk type can be accurately determined, and the accuracy of risk prediction is significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present application or related art, the following will briefly introduce the drawings needed to be used in the embodiments or related art descriptions. Obviously, the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0017] Figure 1A flowchart of a vehicle risk prediction method according to an embodiment of the present application; Figure 2 A structural diagram of a vehicle risk prediction device according to an embodiment of the present application; Figure 3 An electronic device hardware structure diagram according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the present application clearer, the present application will be described in further detail below with reference to specific embodiments and drawings.

[0019] It should be noted that, unless otherwise defined, technical terms or scientific terms used in the embodiments of the present application should be understood as their common meanings to those of ordinary skill in the art to which the present application belongs. The terms "first", "second", and similar terms used in the embodiments of the present application do not represent any order, number, or importance, but are only used to distinguish different components. The terms "include" or "contain" and similar terms mean that the elements or objects before the terms encompass the elements or objects listed after the terms and their equivalents, without excluding other elements or objects. The terms "connect" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right", and the like only represent relative positional relationships, which can change when the absolute positions of the described objects change.

[0020] As described in the background, with the wide popularity of off-road exploration, outdoor exploration, emergency rescue, and various off-road events, the complex and unpredictable outdoor environment poses a serious challenge to the safety of personnel and equipment. In this context, accurate and forward-looking risk prediction for off-road scenes has become an increasingly urgent core demand.

[0021] Traditional off-road risk prediction systems lack the ability to forwardly deduce the dynamic evolution of the scene. Specifically, the system's judgment of risk always stays on a static snapshot of "now" or "past", and cannot simulate and predict the changes caused by dynamic factors such as rainfall and strong wind in the next few minutes or hours. This lack of prediction of dynamic changes in the scene significantly reduces the accuracy of the system's risk prediction, and cannot guarantee the safety of driving in off-road scenes.

[0022] Therefore, the application provides a vehicle risk prediction method, which first calculates a road surface state parameter through a pre-constructed physical model, and then corrects the road surface state parameter through a scene deduction model to generate a corrected road surface state parameter. The scene deduction model can identify rules that are not accurately described by the physical model, and then deduce a corrected road surface state parameter that is more consistent with the current terrain characteristics according to environmental data and vehicle state data, thereby making up for the problem of insufficient scene deduction in the existing risk prediction process. Based on the environmental data, the vehicle state data, the road surface state parameter and the corrected road surface state parameter, the current target risk type can be accurately determined, and the accuracy of risk prediction is significantly improved.

[0023] The embodiments of the application will be described in detail below with reference to the accompanying drawings.

[0024] The embodiments of the application provide a vehicle risk prediction method, which refers to Figure 1 , comprising the following steps: Step 102, acquiring environmental data and vehicle state data in a vehicle driving process.

[0025] Specifically, the environmental data includes three-dimensional scene data and meteorological data. The three-dimensional scene data can be 3DGS (3D Gaussian Splatting) scene data, which is a real-time scene reconstruction and rendering technology based on explicit three-dimensional Gaussian point cloud representation. The scene is modeled by a 3D Gaussian function, and each Gaussian point contains position, covariance matrix, color, opacity and other attributes. The 3DGS scene data can include terrain parameters and obstacle parameters, such as terrain type, slope, terrain roughness in the terrain parameters, and obstacle type, obstacle coordinates and obstacle volume in the obstacle parameters. The meteorological data is real-time data obtained through communication connection between a vehicle-mounted sensor and a weather station, including rainfall, wind speed and temperature.

[0026] The vehicle state data is real-time data obtained through a vehicle CAN (Controller Area Network) bus, including vehicle weight, tire type, current power output and the like.

[0027] Step 104, calculating and determining a road surface state parameter through a pre-constructed physical model according to the environmental data and the vehicle state data.

[0028] Specifically, the physical model is a model constructed in advance according to physical laws, and the road surface state parameters can be calculated through the physical model to provide an interpretable physical quantitative benchmark. The calculated road surface state parameters are parameters that meet objective physical laws and provide a data basis for subsequent correction processes. The road surface state parameters can include soil bearing coefficient, road friction coefficient, and temporary sand barrier formation probability, etc.

[0029] For example, the calculation method of the soil bearing coefficient K is as follows: K = K0× (1 – a1×Q) Wherein, K0 represents the initial soil bearing coefficient, and K0 is determined according to the terrain type in the 3DGS scene data. When the terrain type is dry sand, K0=0.9, and when the terrain type is wet mud, K0=0.5. Q represents rainfall, with a unit of mm / h. a1 is a parameter, and the value can be adjusted according to actual needs, for example, a1=0.05. The lower the K value, the softer the ground, and the greater the possibility of vehicle sinking.

[0030] The calculation method of the road friction coefficient μ is as follows: μ= μ0× (1 - a2×Q) × f(T) Wherein, μ0 represents the friction coefficient under dry road surface, and μ0=0.8 when the tire type is off-road tire. When the tire type is ordinary tire, μ0=0.6. f(T) represents the temperature correction coefficient, and if the current environment temperature <0℃, f(T)=0.7, if 0℃≤current environment temperature <25℃, f(T)=1, and if the current environment temperature ≥25℃, f(T)=1.1. a2 is a parameter, and the value can be adjusted according to actual needs, for example, a2=0.1. The μ value directly determines the vehicle's grip, and is the core data for predicting skidding, sliding, and braking distance.

[0031] The calculation method of the temporary sand barrier formation probability S is as follows: S = a3×V× (1 - R) Wherein, V represents the wind speed, with a unit of m / s, and R represents the terrain roughness, R∈[0,1]. The higher the terrain roughness of the ground, the larger the value of R, and the smaller the S. a3 is a scaling coefficient, and the value can be adjusted according to actual needs, for example, a3=0.1, to ensure that the value of S is within a reasonable range. The higher the S value, the greater the possibility that the path ahead is blocked by new sand dunes.

[0032] The visibility Visi can also be calculated through the physical model, and the calculation method is as follows: Visi=1000 / (1+ a4×Q+ a5×v) Wherein, the unit of Visi is meter, the smaller the value of Visi is, the worse the line of sight is, and the more dangerous the driving environment is. v represents the vehicle speed, and a4 and a5 are both attenuation coefficients, the values of which can be adjusted according to actual needs, for example, a4=0.1 and a5=0.05.

[0033] Step 106, based on the environment data and the vehicle state data, correcting the road surface state parameter through a scene deduction model to generate a corrected road surface state parameter.

[0034] Specifically, the 3DGS scene data in the environment data can provide scene context for the scene deduction model. The rainfall, temperature and wind speed in the environment data input into the scene deduction model are all data in the past period of time, forming an environment time sequence and representing the environmental changes in the past period of time. The vehicle state data such as vehicle speed and power output input into the scene deduction model are all data in the past period of time, forming a state time sequence and representing the vehicle state changes in the past period of time.

[0035] The structure of the scene deduction model of the present application includes a Transformer module and an LSTM module. The Transformer module can process scene context data, and the LSTM module can process each time sequence data. The scene deduction model uses an internal attention mechanism to weigh the importance of different information. Based on the environment data and the vehicle state data, the road surface state parameter is corrected to obtain a corrected road surface state parameter.

[0036] For example, if the road surface friction coefficient μ in the road surface state parameter is 0.6, the scene deduction model, according to the historical rainfall (outputting a large amount of rainfall in the past 10 minutes through the LSTM module), the current terrain (outputting that the vehicle is in backlight and the low-lying land with poor drainage through the Transformer module) and the temperature (the current temperature is low, and the water evaporation is slow), etc., through complex nonlinear calculation, forms an inference and judgment that the road surface friction coefficient calculated by the physical model is too high, and the influence of rainfall, terrain and temperature will further reduce the road surface friction coefficient, and thus outputs the corrected road surface friction coefficient μ1=0.48.

[0037] In addition to correcting the road surface state parameter through the scene deduction model, the slope and the terrain roughness can also be corrected through the scene deduction model to output the corrected slope and the corrected terrain roughness.

[0038] The physical algorithm in the physical model is universal, but due to the particularity of the local environment, the scene deduction model can compensate for the errors of the physical model in a specific local scene. Learning cannot describe complex relationships that cannot be simply described by the physical model, such as the drainage rate of a specific soil type, etc. The changes in off-road scenes are the result of the nonlinear coupling of dozens of factors. Scene deduction models are good at handling such high-dimensional, nonlinear problems, which are problems that physical models cannot solve.

[0039] Among them, the scene deduction model is trained by a large amount of off-road data, and the scene deduction model learns the changes of the road surface state under various environments, various terrains and climate conditions through training. Therefore, the road surface state parameters can be accurately corrected according to the characteristics of the local environment to provide accurate data basis for subsequent prediction.

[0040] It should be noted that the output result of the scene deduction model will be directly fed back to the 3DGS scene model to dynamically update the details of the terrain, obstacles, road surface state, etc. in the 3DGS scene data, so as to further improve the accuracy of subsequent risk prediction.

[0041] Step 108, determining a current target risk type based on the environment data, the vehicle state data, the road surface state parameter and the corrected road surface state parameter.

[0042] Specifically, the target risk type is the highest risk type in the current scene, which is the risk type that needs to take corresponding control strategy immediately. Determining the target risk type in combination with the corrected road surface state parameter can significantly improve the accuracy of determining the target risk type, and further improve the accuracy of risk prediction.

[0043] Based on the steps 102 to 108, the vehicle risk prediction method provided by the embodiment comprises: acquiring environment data and vehicle state data in the vehicle driving process, and determining a road surface state parameter through a pre-constructed physical model according to the environment data and the vehicle state data. The road surface state parameter calculated through the physical model is a parameter that meets the physical law and can reflect the real-time state of the road surface. However, due to the complexity of the scene, there is often a nonlinear correlation between the environment data, the vehicle state data and the road surface state parameter, and the road surface state parameter calculated through the physical model is inaccurate. Therefore, based on the environment data and the vehicle state data, the scene deduction model is used to correct the road surface state parameter to generate a corrected road surface state parameter. The scene deduction model can identify the law that is not accurately described by the physical model, and then can deduce the corrected road surface state parameter that is more consistent with the current terrain characteristics according to the environment data and the vehicle state data, thereby making up for the problem of insufficient scene deduction in the existing risk prediction process. Then, based on the environment data, the vehicle state data, the road surface state parameter and the corrected road surface state parameter, the current target risk type can be accurately determined, and the accuracy of risk prediction is significantly improved.

[0044] In some embodiments, determining the current target risk type based on the environment data, the vehicle state data, the road surface state parameter and the corrected road surface state parameter comprises: Step 1081, determining a predicted road surface state parameter through a preset prediction algorithm based on the corrected road surface state parameter.

[0045] Specifically, in order to predict the road surface state parameter in the future period, the predicted road surface state parameter is calculated through the prediction algorithm based on the corrected road surface state parameter in this step.

[0046] The predicted road surface friction coefficient μ(t) after t minutes can be calculated by the following formula: μ(t) = μ1× (1 – b1×t) Wherein, μ1 represents the corrected road surface friction coefficient, t represents minutes, t ∈ [1, 5], b1 represents the attenuation coefficient, and the numerical value can be adjusted according to requirements, for example, b1 = 0.02. The calculation method of the predicted road surface friction coefficient μ(t) can simulate the process of continuous decline of the road surface friction coefficient, which may be caused by continuous rainfall or repeated rolling of vehicles. The predicted road surface friction coefficient μ(t) provides a kind of aftercare prediction strategy.

[0047] The calculation method of the soil bearing coefficient K(t) after t minutes is as follows: K(t) = K1× (1 - b3× Q × t) wherein, K1 represents the corrected soil bearing coefficient, Q represents the rainfall, t represents the minute, t ∈ [1, 5], b3 represents the attenuation coefficient, the value of which can be adjusted according to the requirement, for example, b3 = 0.008. The calculation of the soil bearing coefficient K(t) can simulate the influence of rainfall on the soil bearing coefficient over time.

[0048] In addition, the slope can also be predicted by the prediction algorithm, for example, the predicted slope a(t) after t minutes can be calculated by the following formula: a(t) = a1 x (1 + b2 x Q x t) wherein, a1 represents the corrected slope, Q represents the rainfall, t represents the minute, t ∈ [1, 5], b2 represents the coupling coefficient, the value of which can be adjusted according to the requirement, for example, b2 = 0.01. The calculation method of the predicted slope a(t) can simulate the hydraulic erosion effect and further cause a certain influence on the slope.

[0049] In order to avoid the scenario deduction model generating deduction results that violate the physical laws in extreme cases, after obtaining the corrected road surface state parameters, the predicted road surface state parameters are calculated according to the prediction algorithm that conforms to the physical change law, so as to ensure that the predicted road surface state parameters have physical rationality. At the same time, without integrating the prediction process in the scenario deduction model, the efficiency of the scenario deduction model training can be improved and the difficulty of the scenario deduction model training can be reduced. By predicting the road surface state parameters through the prediction algorithm, the prediction efficiency can be improved. Further, the response rate of the subsequent vehicle is improved, and the driving safety in the off-road scene is further ensured.

[0050] Step 1082, according to the environment data, the vehicle state data, the road surface state parameter and / or the predicted road surface state parameter, the risk score corresponding to each preset risk type is calculated.

[0051] Specifically, in this embodiment, different risk types are pre-configured according to various driving risks in off-road scenes, such as skidding risk, vehicle sinking risk, tire damage risk, obstacle collision risk, and tire hardening risk, etc. The skidding risk is the risk of insufficient vehicle grip due to the decrease of road friction coefficient. The vehicle sinking risk is the risk of sinking of tires due to insufficient soil bearing capacity of the area (such as muddy land, soft sandy land) where the vehicle travels, which often occurs in low bearing coefficient road sections after rain. The tire damage risk is the risk of tire tread damage or tire burst when the vehicle runs over sharp obstacles (such as rock edges, broken wood, metal fragments) or in high-terrain roughness terrain (such as gravel land). The obstacle collision risk is the risk of collision of the vehicle body due to the failure to avoid static obstacles (such as dry trees, boulders) or dynamic obstacles (such as wild animals, rolling stones) in front of the vehicle during driving, which often occurs in scenes with low visibility or missed route planning. The tire hardening risk is the risk of further reduction of grip (friction coefficient) due to the hardening of tire rubber and the decrease of elasticity when the ambient temperature is lower than 0℃, which indirectly causes skidding and understeering, and often occurs in high-altitude off-road or winter low-temperature scenes.

[0052] According to the environmental data, vehicle state data, road surface state parameters and / or predicted road surface state parameters, the risk scores corresponding to various pre-configured risk types can be calculated. The risk score can represent the risk degree of the pre-configured risk type in the current scene. The higher the risk score, the higher the risk degree of the corresponding pre-configured risk type in the current scene, and certain control strategies need to be taken to deal with the high-risk type to improve driving safety.

[0053] Step 1083, determining the target risk type according to the risk score and the pre-configured risk threshold.

[0054] Specifically, after the risk score is determined, if the risk score exceeds the pre-configured risk threshold, the risk type is determined as the target risk type, and if the risk score does not exceed the pre-configured risk threshold, the risk type is determined as not the target risk type. Exemplarily, the pre-configured risk threshold can be 0.6. When the risk score is greater than or equal to 0.6, the risk type is a high-risk type, and a control strategy needs to be generated immediately to ensure the driving safety of the vehicle. If 0.3≤risk score<0.6, the risk type belongs to a medium-risk type, and needs to be continuously monitored. If the risk score is less than 0.3, the risk type belongs to a low-risk type, and no intervention is needed.

[0055] This embodiment calculates predicted road surface state parameters for a future time period, enabling advance prediction of potential road risks and allowing drivers sufficient time to react, thus preventing problems arising after control measures are implemented. Risk scoring quantifies the risk level of different preset risk types, accurately identifying the target risk type with the highest risk at present, improving the accuracy of risk prediction and enhancing driving safety. In some embodiments, calculating the risk score corresponding to each preset risk type based on the environmental data, the vehicle state data, the road surface state parameters, and / or the predicted road surface state parameters includes: Based on the environmental data, vehicle status data, road surface status parameters, and / or predicted road surface status parameters, calculate the probability of occurrence for each preset risk type; determine the risk score based on the probability of occurrence and the preset influence coefficient corresponding to the preset risk type.

[0056] Specifically, the risk score is determined by the probability of occurrence (P) corresponding to the risk type and the preset influence coefficient (I). The probability of occurrence is calculated based on environmental data, vehicle condition data, road surface condition parameters, and / or predicted road surface condition parameters. The preset coefficient is predetermined based on the degree of impact of the risk type. Risk score = P × I. The probability of occurrence (P) and the preset influence coefficient (I) for different risk types are explained below.

[0057] Slippage risk: probability of occurrence =1 - μ(t), where μ(t) represents the predicted road surface friction coefficient after t minutes. The smaller μ(t) is, the higher the probability of skidding. The higher the preset influence coefficient. =0.9.

[0058] Risk of getting stuck: probability of occurrence = 1 - (K × O) / (M × c1), where K represents the soil bearing capacity coefficient calculated by the physical model, K∈[0,1]. O represents the tire contact area, determined by the tire type; for off-road tires, O=0.2m. 2 For a standard tire, O = 0.15m 2 M represents the vehicle weight in kg. c1 represents the scaling factor, which can be used to scale the vehicle weight. Scale to a reasonable range to avoid If the value of c1 exceeds a reasonable range, it can be 0.001. (Preset influence coefficient) =0.8. For example, if K=0.3, M=2000kg, S=0.2 m 2 ,but = 1-(0.3*0.2) / (2000*0.001)=1-0.06 / 2=0.97, which belongs to high probability of sinking.

[0059] Tire damage risk: occurrence probability = RxCxW, R represents terrain roughness, which is obtained by 3DGS scene data, R ∈ [0, 1]. For example, R = 0.8 in the gravel terrain and R = 0.2 in the flat grassland terrain. C represents the sharpness of the obstacle, which is determined by the terrain type obtained by 3DGS scene data, such as C = 0.9 in the rock corner terrain, C = 0.6 in the broken wood terrain, and C = 0.3 in the ordinary gravel terrain. W represents the tire wear degree, which is collected by the vehicle sensor, W ∈ [0, 1], such as W = 0.2 for a brand new tire. For example, if R = 0.8, C = 0.9, and W = 0.8, = 0.8*0.9*0.8=0.576, which belongs to medium-high damage probability. Preset influence coefficient = 0.7.

[0060] Obstacle collision risk: occurrence probability = (1-Visi / 1000)*D*(v / 100), Visi represents visibility, which is calculated by a physical model. D represents the size of the obstacle, which is obtained according to 3DGS scene data. When the maximum cross-sectional diameter of the obstacle is ≥1 m, D = 1.0. When 0.3 m ≤ the maximum cross-sectional diameter of the obstacle <1 m, D = 0.6. When the maximum cross-sectional diameter of the obstacle <0.3 m, D = 0.2. (1-Visi / 1000) represents the visibility attenuation coefficient, v represents the vehicle speed, and v / 100 represents the vehicle speed coefficient. For example, if Visi = 200 m, D = 1.0, and v = 60 km / h, = (1-200 / 1000)*1.0*(60 / 100)=0.8*0.6=0.48, which belongs to medium collision probability. When the maximum cross-sectional diameter of the obstacle is ≥1 m, the preset influence coefficient = 0.9. When the maximum cross-sectional diameter of the obstacle <1 m, the preset influence coefficient = 0.5.

[0061] Tire hardening risk: occurrence probability = max(0, (0 - T) / 20) x Ttype, wherein T represents an ambient temperature, Ttype represents a tire anti-low temperature coefficient, when the tire type is a cross-country snow tire, Ttype = 0.3, when the tire type is a general cross-country tire, Ttype = 0.6, and when the tire type is a general tire, Ttype = 1.0. max(0, (0 - T) / 20) as a whole represents a temperature attenuation coefficient, the lower the temperature T, the harder the tire rubber, and the greater the temperature attenuation coefficient. For example, if T = -10℃ and Ttype = 0.6, then = (0 - (-10)) / 20 x 0.6 = 0.5 x 0.6 = 0.3, which belongs to a medium-low hardening probability. The preset influence coefficient = 0.6.

[0062] Through the method of the embodiment, a calculation method of the risk score is given, and a basis for risk quantification and unified comparison is formed. The risk score is determined in combination with the occurrence probability and the preset influence coefficient, which can filter out risks with high occurrence probability but small harm, or risks with great harm but extremely low occurrence probability, so that the vehicle system can focus on the risk type that really exists. According to the accurate evaluation of risks of different preset risk types, the target risk type can be accurately determined subsequently.

[0063] After the target risk type is determined, a control strategy corresponding to the target risk type can be taken to control the vehicle to stably cope with the risk and improve driving safety. The following specific embodiments are described.

[0064] In some embodiments, the method further comprises: obtaining a driving level of a current driver; determining a corresponding control strategy according to the driving level, the vehicle state data and the target risk type, and controlling the vehicle to operate according to the control strategy.

[0065] Specifically, for drivers with different off-road experience, the degree of active intervention of the vehicle is different. For drivers with rich off-road experience, the vehicle can reduce the degree of active intervention, and for drivers with poor off-road experience, the vehicle can increase the degree of active intervention. Therefore, before determining the control strategy, the driving level of the driver also needs to be obtained. The higher the driving level, the more off-road experience, and the lower the driving level, the less off-road experience. For example, the driving level is divided into L1 level, L2 level and L3 level.

[0066] Afterwards, the corresponding control strategy is determined according to the driving level, vehicle state data and target risk type. In specific implementation, different control strategies are adopted for different risk types. When the target risk type is the slipping risk, the corresponding control strategy is to reduce the power output of the vehicle. The reduced power output of the vehicle is = P x (0.8 - 0.1 x (1 - μ(t))), P represents the current power output of the vehicle. When μ(t) = 1, the predicted road surface friction coefficient is the maximum, the road surface condition is excellent, = 0.8P, at this time, the system also reduces the power output to 80% of P preventively, which is equivalent to providing a basic safety margin to prevent the driver from suddenly stepping on the accelerator to damage the balance of the vehicle. When μ(t) = 0, the road surface is extremely wet and slippery, = 0.7P, the system reduces the power output to 70% of P. The reduced power output varies in the range of 0.7P to 0.8P, achieving stepless adjustment of the power output.

[0067] The control strategy can also include adjusting the four-wheel drive torque distribution ratio and increasing the front wheel torque ratio. For example, the front wheel torque ratio is increased from 50% to 60%, more torque is distributed to the front wheel, which is equivalent to applying a slight “pulling” force to the vehicle instead of a “pushing” force. This can effectively suppress the restlessness of the tail and increase the driving stability of the vehicle, especially when accelerating, which can better prevent the rear wheel (usually the main driving force) from losing grip due to excessive torque.

[0068] It should be noted that for the slipping risk, the control strategy corresponding to the driver of different driving levels is the same.

[0069] When the target risk type is the vehicle sinking risk, the control strategy intervenes from three aspects of power control, transmission control and suspension control. For the driver of L1 level, the power output of the vehicle is limited to be less than or equal to 50%, so as to ensure that the driver of L1 level will not cause serious risk even if he mistakenly steps on the accelerator. The driving mode of the vehicle is switched to 4L (Four-Wheel Drive Low, low-speed four-wheel drive) mode, which is beneficial to the vehicle to get out of trouble. At the same time, the suspension is lifted by 20 mm to provide a large safety margin for the vehicle. For the driver of L2 level, the power output of the vehicle is limited to be less than or equal to 60%, which is appropriately relaxed compared with the driver of L1 level. The driving mode of the vehicle is switched to 4L mode, and the suspension is lifted by 15 mm. For the driver of L3 level, only risk information (such as the soil bearing coefficient K(t), the target risk type, etc.) is provided to the driver, and no active intervention is made, so that the driver can make decisions on the control operation of the vehicle by himself.

[0070] When the target risk type is tire damage risk, the control strategy intervenes from three aspects of power control, tire control and ESP enhancement. For L1 level drivers, limit the increase rate of throttle pedal opening ≤ 3%, at this time the throttle response is relatively dull, and the sudden acceleration caused by misoperation is eliminated. Reduce the tire pressure by 0.3 bar, increase the tire cushion and ground contact area. At the same time, the ESP (Electronic Stability Program) enhancement function is turned on. ESP can sense and correct the out-of-control trend of the vehicle, such as understeering or oversteering. After the ESP enhancement function is activated, it can correct the vehicle to the maximum extent to avoid vehicle out-of-control, spin or roll. For L2 level drivers, limit the increase rate of throttle pedal opening ≤ 5%, reduce the tire pressure by 0.2 bar, and turn on the ESP enhancement function. For L3 level drivers, only provide risk information (such as road surface terrain in front, terrain roughness tire pressure suggestion, target risk type, etc.) to the driver, and do not actively intervene, so that the driver can make his own decision on the control operation of the vehicle.

[0071] When the target risk type is obstacle collision risk, the control strategy intervenes from three aspects of power control, brake control and steering control. For L1 level drivers, stop sending oil to the transmitter when the vehicle is judged to collide with the obstacle ≤ 3s, cut off the power source, and prevent the driver from stepping on the accelerator due to nervousness and aggravating the accident. A hierarchical emergency braking action is adopted, that is, different braking forces are used to balance the relationship between "avoiding collision" and "maintaining stability". At the same time, steering correction is performed, with a correction number of 5°, that is, the steering wheel is controlled to turn 5° to the left or right to bypass the obstacle and reduce the risk of head-on collision. For L2 level drivers, stop sending oil to the transmitter when the vehicle is judged to collide with the obstacle ≤ 2s, cut off the power source, and intervene actively slightly later than L1 level drivers. A reduced braking action is adopted, such as the vehicle end system may first perform emergency braking, but if it is monitored that the driver has made the correct braking operation, the system will reduce the braking force of active intervention and cooperate with the driver to complete the braking, which avoids the discomfort of competing for control with the driver. At the same time, steering correction is performed, with a correction number of 3°, that is, the steering wheel is controlled to turn 3° to the left or right. For L3 level drivers, stop sending oil to the transmitter when the vehicle is judged to collide with the obstacle ≤ 1s, cut off the power source. A slow braking action is adopted to leave room for the driver to operate himself. At the same time, steering correction is performed, but the steering is not intervened.

[0072] When the target risk type is tire hardening risk, the control strategy intervenes from three aspects of power control, tire control and brake control. For L1 level driver, the vehicle speed is controlled to be less than or equal to 40 km / h to reduce the kinetic energy of the vehicle. The tire is forced to heat to actively increase the temperature of the tire tread, so as to directly improve the road friction coefficient by restoring the elasticity and viscosity of the frozen hardened rubber. At the same time, the expected braking distance is increased by 30% to prompt the driver to take braking operation in advance. For L2 level driver, the vehicle speed is controlled to be less than or equal to 50 km / h to reduce the kinetic energy of the vehicle. The driver is suggested to heat the tire to improve the road friction coefficient. At the same time, the expected braking distance is increased by 30% to prompt the driver to take braking operation in advance. For L3 level driver, only risk information (such as vehicle speed, heating suggestion, target risk type, etc.) is provided to the driver, and the expected braking distance is increased by 30% to prompt the driver to take braking operation in advance.

[0073] By the method of the embodiment, different control strategies can be formulated for different risk types, and personalized control strategies can be provided for drivers of different driving levels, thereby improving the satisfaction of the driver on the premise of ensuring the driving safety of the vehicle.

[0074] In some embodiments, the method further comprises: determining whether the front road surface meets a preset road wet condition according to the predicted road surface state parameter; in response to determining that the front road surface meets the preset road wet condition, planning a driving path of the vehicle according to the environmental data, and controlling the vehicle to drive according to the planned driving path.

[0075] Specifically, whether the front road surface meets the preset road wet condition can be determined according to the predicted soil bearing coefficient K (t) in the predicted state parameter. For example, the preset road wet condition includes that the predicted soil bearing coefficient K (t) is less than or equal to 0.3. When the preset road wet condition is met, it indicates that the front road surface is wet and slippery, which is not conducive to vehicle driving. At this time, the driving path of the vehicle can be re-planned to enable the vehicle to avoid the front wet and slippery road section, thereby avoiding the risk from the root and reducing the probability of risk occurrence. In specific implementation, the path is re-planned based on the 3DGS scene data in combination with a path planning algorithm. The 3DGS scene data can provide passable area identification, terrain details and obstacle accurate coordinates for the path planning process to assist in accurate path planning. The path planning algorithm can be algorithm, Dijkstra's algorithm is an efficient algorithm for finding the shortest path in a graph, which guides the search direction through an evaluation function. After obtaining the re-planned driving path, the vehicle is controlled to drive according to the re-planned driving path, which significantly reduces the risk occurrence rate of the vehicle and improves the driving safety of the vehicle.

[0076] In some embodiments, the method further comprises: generating driving operation prompt information according to the vehicle state data, the modified road surface state parameter, the control strategy and the driving level; displaying the driving operation prompt information.

[0077] Specifically, in addition to actively intervening according to the driving level, the vehicle system can also provide the driver with operation prompt information containing rich information. The driver can control the vehicle according to the operation prompt information, improve the accuracy of the control action, avoid misoperation due to insufficient off-road experience, and effectively improve the driving safety. The driving operation prompt information can include the current risk type, vehicle operation suggestion, environmental temperature, wind speed, rainfall, road surface friction coefficient, road roughness, etc. The driving operation prompt information can be displayed through a head-up display, or can be displayed through a car screen. Alternatively, it can be conveyed to the driver through a voice prompt method to reduce the probability of the driver's driving distraction.

[0078] Further, the generating driving operation prompt information according to the vehicle state data, the modified road surface state parameter, the control strategy and the driving level comprises: determining preset display content to be displayed according to the driving level; generating the driving operation prompt information according to the vehicle state data, the modified road surface state parameter, the control strategy and / or the driving level according to the preset display content.

[0079] Specifically, different driving operation prompt information can be provided for different driving levels of drivers due to different off-road experiences. For drivers with low driving levels, as much detailed driving operation prompt information as possible can be provided to make up for the risks caused by insufficient off-road experience. For drivers with high driving levels, concise driving operation prompt information can be provided to reduce the interference with the driver's driving operation.

[0080] In this embodiment, the preset display content to be displayed is determined according to the driving level. Different driving levels of drivers correspond to different preset display content under different risk types. For example, the preset display content corresponding to low driving levels includes detailed operation steps and operation parameters, and the preset display content corresponding to high driving levels includes risk parameters but does not include operation steps and operation parameters. Different risk types are illustrated below.

[0081] For the risk of skidding, if the driving level is L1 level or L2 level, the driving operation prompt information includes brake pedal opening and steering angle of the steering wheel, and can also include synchronous voice prompts such as “slowly step on the brake and keep the steering wheel stable”. If the driving level is L3 level, the driving operation prompt information includes the risk type.

[0082] For the risk of getting stuck, if the driving level is L1 level or L2 level, the driving operation prompt information includes throttle pedal opening, vehicle driving direction, and recommended detailed driving operation steps. If the driving level is L3 level, the driving operation prompt information includes risk parameters related to the risk of getting stuck, such as risk parameters and power system parameters identified through 3DGS scene data.

[0083] For the risk of tire damage, if the driving level is L1 level or L2 level, the driving operation prompt information includes a safe route and a target tire pressure, forces the vehicle to travel along the safe route, and locks the tire pressure to the target tire pressure. If the driving level is L3 level, the driving operation prompt information includes an obstacle heat map. The driver can choose the driving route by himself.

[0084] For the risk of obstacle collision, if the driving level is L1 level or L2 level, the driving operation prompt information includes auxiliary braking operation type and steering wheel steering angle. If the driving level is L3 level, the driving operation prompt information includes the relative speed between the vehicle and the obstacle, and the driver judges the avoidance mode by himself.

[0085] For the risk of tire hardening, if the driving level is L1 level or L2 level, the driving operation prompt information includes vehicle speed, expected braking distance, and tire heating identifier. If the driving level is L3 level, the driving operation prompt information includes the predicted road friction coefficient μ(t).

[0086] The method of the embodiment can provide personalized driving operation prompt information for drivers of different driving levels, meet the off-road driving needs of different drivers, and improve the driving experience of the driver.

[0087] In some embodiments, the method further comprises: collecting actual road surface state parameters corresponding to the corrected road surface state parameters; determining a difference value between the corrected road surface state parameters and the actual road surface state parameters; updating the model parameters of the scene deduction model according to the difference value.

[0088] Specifically, during vehicle operation, the scenario simulation model continuously corrects the road surface condition parameters while simultaneously collecting actual road surface condition parameters. By comparing the corrected road surface condition parameters with the actual road surface condition parameters, the difference between the two can be determined. Based on this difference, the model parameters of both the physical model and the scenario simulation model are updated to continuously optimize their accuracy.

[0089] For example, based on the actual road surface friction coefficient and the corrected road surface friction coefficient Calculate the difference value Δμ= According to the actual slope and the corrected slope Calculate the difference value Δ = . Δμ and Δ The input is fed into the optimization module of the scene deduction model, where the weight parameters of the scene deduction model are adjusted using the gradient descent algorithm.

[0090] It should be noted that the differences between the actual road surface condition parameters and the road surface condition parameters calculated by the physical model can also be compared. Based on these differences, adjustments can be made to the parameters or coefficients in the physical model to better adapt it to the current climate and terrain. Through the method of updating model parameters in this embodiment, the physical model and the scenario simulation model can be optimized in real time, making them more adaptable to actual road conditions and improving the accuracy of model calculations and simulations.

[0091] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0092] It should be noted that some embodiments of this application have been described above. In some cases, the actions or steps described in the above embodiments can be performed in a different order than that shown in the above embodiments and the desired result can still be achieved. In addition, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] Based on the same inventive concept, corresponding to any of the above embodiments, this application also provides a vehicle risk prediction device.

[0094] refer toFigure 2 The vehicle risk prediction apparatus comprises: An acquisition module 202 configured to acquire environment data and vehicle state data during vehicle driving; A calculation module 204 configured to calculate a road surface state parameter according to the environment data and the vehicle state data through a pre-constructed physical model; A correction module 206 configured to correct the road surface state parameter through a scenario deduction model based on the environment data and the vehicle state data, to generate a corrected road surface state parameter; A determination module 208 configured to determine a current target risk type based on the environment data, the vehicle state data, the road surface state parameter, and the corrected road surface state parameter.

[0095] In some embodiments, the determination module 208 is configured to calculate a predicted road surface state parameter through a preset prediction algorithm based on the corrected road surface state parameter. According to the environment data, the vehicle state data, the road surface state parameter, and / or the predicted road surface state parameter, a risk score corresponding to each preset risk type is calculated. The target risk type is determined according to the risk score and a preset risk threshold.

[0096] In some embodiments, the determination module 208 is configured to calculate an occurrence probability corresponding to each preset risk type according to the environment data, the vehicle state data, the road surface state parameter, and / or the predicted road surface state parameter. The risk score is determined according to the occurrence probability and a preset influence coefficient corresponding to the preset risk type.

[0097] In some embodiments, a generation module is further included, configured to acquire a driving level of a current driver. According to the driving level, the vehicle state data, and the target risk type, a corresponding control strategy is determined, and the vehicle is controlled to operate according to the control strategy.

[0098] In some embodiments, a planning module is further included, configured to determine whether a front road surface meets a preset road surface wet and slippery condition according to the predicted road surface state parameter. In response to determining that the front road surface meets the preset road surface wet and slippery condition, a vehicle driving path is planned according to the environment data, and the vehicle is controlled to drive according to the planned driving path.

[0099] In some embodiments, the device further comprises a display module configured to generate driving operation prompt information according to the vehicle state data, the modified road surface state parameter, the control strategy and the driving level; and display the driving operation prompt information.

[0100] In some embodiments, the display module is configured to determine preset display content to be displayed according to the driving level. According to the preset display content, the driving operation prompt information is generated according to the vehicle state data, the modified road surface state parameter, the control strategy and / or the driving level.

[0101] In some embodiments, the device further comprises an updating module configured to collect an actual road surface state parameter corresponding to the modified road surface state parameter. determine a difference value between the modified road surface state parameter and the actual road surface state parameter; update a model parameter of the scenario deduction model according to the difference value.

[0102] For the convenience of description, the above device is described as various modules in function. Of course, in the implementation of the present application, the functions of each module can be implemented in one or more software and / or hardware.

[0103] The device of the above embodiments is used to implement the corresponding vehicle risk prediction method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be described here.

[0104] Based on the same inventive concept, corresponding to any of the above method embodiments, the present application further provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle risk prediction method of any one of the above embodiments.

[0105] Figure 3 A more specific hardware structure of an electronic device is shown in the embodiment, which can include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040 and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030 and the communication interface 1040 are connected to each other through the bus 1050 for communication within the device.

[0106] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, etc., for executing relevant programs to implement the technical solutions provided by the embodiments of the present specification.

[0107] The memory 1020 can be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 1020 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the relevant program codes are saved in the memory 1020 and called and executed by the processor 1010.

[0108] The input / output interface 1030 is configured to connect input / output modules to implement information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. The input devices can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices can include a display, a speaker, a vibrator, an indicator light, etc.

[0109] The communication interface 1040 is configured to connect a communication module (not shown in the figure) to implement the communication interaction between the device and other devices. The communication module can realize communication through a wired manner (such as USB, network cable, etc.) or through a wireless manner (such as mobile network, WIFI, Bluetooth, etc.).

[0110] The bus 1050 includes a channel for transmitting information between various components (such as the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040) of the device.

[0111] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in the specific implementation process, the device can also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device can also only include the components necessary to implement the solutions of the embodiments of the present specification, and does not have to include all the components shown in the figure.

[0112] The electronic device of the above embodiments is used to implement the corresponding vehicle risk prediction method in any of the preceding embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0113] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the vehicle risk prediction method of any of the above embodiments.

[0114] The computer-readable medium of the present embodiment includes permanent and non-permanent, removable and non-removable media, which can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0115] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to perform the vehicle risk prediction method of any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0116] Based on the same concept, corresponding to the method of any of the above embodiments, the present application also provides a computer program product comprising computer program instructions which, when executed on a computer, cause the computer to perform the method of any of the above embodiments, have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0117] Those of ordinary skill in the art will understand that the above discussion of any of the embodiments is only exemplary and is not intended to imply that the scope of the present application is limited to these examples; the above embodiments or technical features between different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes to the aspects of the embodiments of the present application as described above. In order to be brief, they are not provided in detail.

[0118] Additionally, to simplify the description and discussion, and so as not to obscure the embodiments of the application being presented, the well-known functions or constructions of integrated circuit (IC) chips and other components can or can not be shown in the figures and will be omitted as not to unnecessarily obscure the embodiments of the application being presented. Moreover, the devices can be shown in block diagram form in order to avoid obscuring the embodiments of the application, and this also acknowledges the fact that the details in regards to the implementation of such block devices are highly dependent on the platform upon which the embodiments of the application are being implemented (i.e., these details should be well within the purview of one of ordinary skill in the art). Where specific details are set forth in order to describe an illustrative embodiment of the application, it will be apparent to one of ordinary skill in the art that the embodiments of the application can be practiced without, or with variation of, these specific details. Thus, the description is to be considered as illustrative and not restrictive, and the scope of the application should be determined not with reference to the above description, but should be given to the appended claims.

[0119] While the application has been described in connection with specific embodiments thereof, many alternatives, modifications and variations will be apparent to those skilled in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) can use the embodiments discussed.

[0120] Embodiments of the application are intended to cover all such alternatives, modifications and variations as falling within the scope of the application. Accordingly, any omission, modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the application should be included in the scope of protection of the application.

Claims

1. A vehicle risk prediction method, characterized in that, include: Acquire environmental and vehicle status data during vehicle operation; Based on the environmental data and the vehicle status data, road surface status parameters are calculated and determined using a pre-built physical model; Based on the environmental data and the vehicle status data, the road surface status parameters are corrected using a scenario inference model to generate corrected road surface status parameters. Based on the environmental data, the vehicle status data, the road surface status parameters, and the corrected road surface status parameters, the current target risk type is determined.

2. The method according to claim 1, characterized in that, The determination of the current target risk type based on the environmental data, vehicle status data, road surface status parameters, and the corrected road surface status parameters includes: Based on the corrected road surface condition parameters, the predicted road surface condition parameters are calculated and determined using a preset prediction algorithm. Based on the environmental data, the vehicle status data, the road surface status parameters, and / or the predicted road surface status parameters, calculate the risk score corresponding to each preset risk type; The target risk type is determined based on the risk score and the preset risk threshold.

3. The method according to claim 2, characterized in that, The step of calculating the risk score corresponding to each preset risk type based on the environmental data, the vehicle status data, the road surface condition parameters, and / or the predicted road surface condition parameters includes: Based on the environmental data, the vehicle status data, the road surface status parameters, and / or the predicted road surface status parameters, calculate the probability of occurrence for each preset risk type. The risk score is determined based on the probability of occurrence and the preset impact coefficient corresponding to the preset risk type.

4. The method according to claim 1, characterized in that, The method further includes: Get the current driver's driving level; Based on the driving level, the vehicle status data, and the target risk type, a corresponding control strategy is determined, and the vehicle operation is controlled according to the control strategy.

5. The method according to claim 2, characterized in that, The method further includes: Determine whether the road surface ahead meets the preset slippery road conditions based on the predicted road surface condition parameters; In response to determining that the road surface ahead meets the preset slippery road conditions, the vehicle travel path is planned based on the environmental data, and the vehicle is controlled to travel along the planned path.

6. The method according to claim 4, characterized in that, The method further includes: Based on the vehicle status data, the corrected road surface status parameters, the control strategy, and the driving level, driving operation prompts are generated. The driving operation prompts are displayed.

7. The method according to claim 6, characterized in that, The step of generating driving operation prompts based on the vehicle status data, the corrected road surface parameters, the control strategy, and the driving level includes: The preset display content to be displayed is determined based on the driving level; According to the preset display content, the driving operation prompt information is generated based on the vehicle status data, the corrected road surface status parameters, the control strategy, and / or the driving level.

8. The method according to claim 1, characterized in that, The method further includes: Collect the actual road surface state parameters corresponding to the corrected road surface state parameters; Determine the difference between the corrected road surface condition parameters and the actual road surface condition parameters; The model parameters of the scenario inference model are updated based on the difference value.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 8.

10. A vehicle, characterized in that, The vehicle includes the electronic equipment as described in claim 9.

Images