A crosswind warning method for vehicle assisted driving

Abstract

This invention relates to the field of vehicle assisted driving technology and provides a crosswind warning method for vehicle assisted driving. The method involves multi-dimensional data acquisition and feature extraction, followed by crosswind interference correlation analysis. Quantitative parameters characterizing the impact of crosswind interference are output and input into a risk analysis model for fusion calculation. The warning level is determined based on the calculation results. By integrating multi-dimensional data such as driver data, real-time wind direction data, driving data, altitude information, and climate information, a dynamic prediction model is constructed to calculate the risk warning coefficient in real time, ensuring prediction accuracy and real-time performance. This proactive prediction method extends the warning response time and effectively improves high-speed driving safety.

Description

Technical Field

[0001] This invention relates to the field of vehicle driver assistance technology, and in particular to a crosswind warning method for vehicle driver assistance. Background Technology

[0002] In scenarios such as highways, sudden strong crosswinds are a significant contributing factor to serious traffic accidents. When vehicles are traveling at high speeds, the aerodynamic interference forces generated by crosswinds can significantly disrupt the vehicle's dynamic balance, causing unexpected lateral displacement, yaw, and even skidding, fishtailing, or rollover, posing a serious threat to the lives of drivers and passengers. Therefore, establishing an effective active safety warning system during vehicle operation to provide drivers with valuable reaction time is a necessary requirement and a key technological path to improving road traffic safety.

[0003] Currently, most mainstream passenger vehicles are equipped with advanced driver assistance systems (ADAS) such as Electronic Stability Program (ESP) and Lane Keeping Assist (LKA). The core intervention logic of these systems relies on onboard inertial sensors and wheel speed sensors to detect vehicle instability by monitoring the vehicle's yaw rate, lateral acceleration, steering wheel angle, and other intrinsic state parameters in real time. When the system detects a deviation between the vehicle's actual trajectory and the driver's expected trajectory (inferred from steering wheel angle, etc.), for example, when crosswinds begin to cause the vehicle to deviate from its lane, ESP will generate corrective torque by braking one or more wheels, while LKA will apply steering assist or pull the vehicle back to the center line of the lane.

[0004] However, since deviation correction is a typical reactive mechanism, intervention is only initiated when crosswind disturbances have already had a substantial impact on the vehicle, and dynamic imbalance has occurred. While it effectively prevents the accident from escalating, it is a passive correction, leaving extremely limited reaction time for the driver. Drivers often only perceive the danger at the moment or after the system intervenes, easily leading to tension and misoperation. Summary of the Invention

[0005] This invention provides a crosswind warning method for vehicle assisted driving, which solves the technical problems of existing high-speed assisted driving systems where crosswind intervention is a passive correction, only intervening and correcting when the crosswind has already caused the vehicle to lose control, resulting in short reaction time, easy operation errors, and low safety performance.

[0006] To address the above technical problems, this invention provides a crosswind warning method for vehicle assisted driving, comprising the following steps: Real-time multidimensional data acquisition is performed, and the multidimensional data includes at least one or more of the following: driver data, real-time wind direction data, driving data, altitude information, and climate information; Interference analysis involves extracting features from the multidimensional data, performing crosswind interference correlation analysis based on the extracted data features, and outputting quantitative parameters characterizing the impact of crosswind interference. The early warning decision-making process involves inputting the quantitative parameters of multidimensional data into a risk analysis model for fusion calculation, determining the early warning level based on the calculation results, and then executing a driving warning according to the early warning level.

[0007] This basic solution involves multi-dimensional data collection and feature extraction, followed by crosswind interference correlation analysis. It outputs quantitative parameters characterizing the impact of crosswind interference and inputs them into a risk analysis model for fusion calculation. Based on the calculation results, the warning level is determined. By integrating multi-dimensional data such as driver data, real-time wind direction data, driving data, altitude information, and climate information, a dynamic prediction model is constructed to calculate the risk warning coefficient in real time, ensuring prediction accuracy and real-time performance. By extending the warning response time through proactive prediction, the safety of high-speed driving is effectively improved.

[0008] In a further implementation, feature extraction is performed on the multidimensional data, and crosswind interference correlation analysis is conducted based on the extracted data features to output quantitative parameters characterizing the impact of crosswind interference, including: Based on the driver data in the multidimensional data, the driver status is analyzed, and then the crosswind interference is assessed based on the driver status to obtain the first quantitative parameter. Based on the real-time wind direction data and driving data in the multidimensional data, a crosswind sensitivity analysis is performed to assist in the assessment of crosswind interference and obtain the second quantitative parameter. Based on the real-time wind direction data and driving data in the multidimensional data, a path crosswind force analysis is performed to assist in crosswind interference assessment and obtain a third quantitative parameter. Based on the driving data and altitude information in the multidimensional data, altitude wind pressure adaptation analysis is performed to assist in crosswind interference assessment and obtain the fourth quantitative parameter. Based on the climate information in the multidimensional data, regional wind field characteristics are analyzed to assist in the assessment of crosswind interference and obtain the fifth quantitative parameter.

[0009] This solution breaks through the limitations of traditional methods that rely solely on vehicle sensors. It uses driver data, real-time wind direction data, driving data, altitude information, and climate information as basic data. Based on the correlation between crosswind interference, it conducts multi-dimensional cross-analysis (multi-dimensional analysis of user status, vehicle driving status, and regional altitude / climate achieves scenario fusion) to construct a panoramic risk assessment model. This model is highly adaptable to various scenarios and greatly improves the system's early warning capabilities and coverage.

[0010] In a further implementation scheme, the driver's state is analyzed based on the driver data in the multidimensional data, and then a first quantitative parameter is obtained by assessing crosswind interference based on the driver's state, including: Driver data is acquired for feature extraction, including eye movement frequency, duration of eye closure, and head posture data. Based on the eye movement frequency, eye closure duration, and head posture data, a multimodal feature fusion analysis of the driver's fatigue state is performed to obtain a fatigue weighting factor, which serves as the first quantitative parameter characterizing the impact of crosswind interference.

[0011] This solution collects drivers' physiological characteristics and assesses their fatigue state using real-time data on eye movement frequency, eye closure duration, and head posture. This data serves as the primary quantitative parameter characterizing the impact of crosswind interference, enabling personalized risk assessment and providing earlier and stronger warning signals, thus significantly improving the accuracy and safety of the warning system.

[0012] In a further implementation scheme, based on the real-time wind direction data and driving data in the multidimensional data, a path crosswind sensitivity analysis is performed to assist in the assessment of crosswind interference and obtain a second quantitative parameter, including: Navigation routes are segmented based on driving data; Obtain the path direction, real-time wind direction, and real-time wind speed of the navigation route on each road segment; Calculate the real-time crosswind angle based on the path direction and real-time wind direction; Based on the real-time crosswind angle and the real-time wind speed, determine whether the corresponding navigation path is a crosswind sensitive section; The total length of the navigation path segments belonging to the crosswind sensitive section is statistically analyzed and normalized to obtain the crosswind interference factor, which serves as the second quantitative parameter characterizing the impact of crosswind interference.

[0013] When performing crosswind-sensitive interference analysis, this solution divides the road into segments and performs granular analysis of real-time crosswind angles and real-time wind speeds. This allows for the accurate identification of crosswind-sensitive road segments (such as bridges and wind gaps) in the navigation path. Consequently, the safety assessment of the navigation path is used as a second quantitative parameter to characterize the impact of crosswind interference, effectively enhancing driving safety.

[0014] In a further implementation scheme, based on the real-time wind direction data and driving data in the multidimensional data, a path crosswind force analysis is performed to assist in crosswind interference assessment and obtain a third quantitative parameter, including: Obtain the real-time crosswind angle of the navigation path on each road segment; The effective intensity ratio of the crosswind is determined by force analysis based on the real-time crosswind angle, and the average value of the effective intensity ratio of the crosswind within a preset distance is calculated to obtain the effectiveness coefficient. The wind disturbance factor is calculated based on the effectiveness coefficient and real-time wind speed, serving as the third quantitative parameter characterizing the impact of crosswind interference.

[0015] In a further embodiment, the formula for calculating the wind disturbance factor is as follows:

[0016] In the formula, Indicates the wind disturbance factor. Indicates real-time wind speed. Indicates the effectiveness coefficient. This indicates the maximum wind speed.

[0017] This solution performs force analysis based on the real-time crosswind angle of the navigation path. Through physical modeling, the geometric influence of wind direction is transformed into dynamic parameters, quantifying the interference of crosswinds on vehicles. This significantly improves the scientific rigor and accuracy of crosswind interference assessment, providing a more reliable quantitative basis for the early warning system and thus enhancing the accuracy of crosswind warnings.

[0018] In a further implementation scheme, based on the driving data and altitude information in the multidimensional data, an altitude-wind-pressure adaptation analysis is performed to assist in the assessment of crosswind interference and obtain a fourth quantitative parameter, including: Navigation routes are segmented based on driving data; Obtain the altitude of the navigation path on each segment and generate an altitude sequence; Calculate the average altitude of the navigation path within a preset distance based on the altitude sequence; The altitude disturbance factor is determined based on the average altitude and normalized to serve as the fourth quantitative parameter characterizing the impact of crosswind interference.

[0019] This scheme improves the stability of the algorithm by calculating the average altitude, and determines the altitude disturbance factor based on the average altitude as the fourth quantitative parameter to characterize the impact of crosswind interference. This enables the quantitative characterization of regional terrain features, provides a key environmental correction benchmark for crosswind risk assessment, and achieves scene adaptation in different altitude areas. This significantly improves the accuracy and reliability of the crosswind early warning system in different geographical environments.

[0020] In a further implementation scheme, based on the climate information in the multidimensional data, regional wind field characteristic analysis is performed to assist in crosswind interference assessment and obtain a fifth quantitative parameter, including: Climate data from different regions are acquired in advance, and big data analysis is performed based on the climate data to obtain the climate interference coefficient of different climate zones on crosswinds. Different weather types are obtained in advance, and big data analysis is performed based on the weather types to obtain the weather risk coefficient of different weather environments on the impact of crosswinds; Based on the real-time collected climate information, the climate interference coefficient and weather risk coefficient under the current driving environment are determined. The meteorological disturbance value is calculated based on the climate interference coefficient and weather risk coefficient, and then normalized as the fifth quantitative parameter characterizing the impact of crosswind interference.

[0021] In a further implementation plan, the meteorological disturbance value is calculated based on the climate disturbance coefficient and the weather risk coefficient, specifically as follows: Based on the aforementioned climate disturbance coefficient, the weather risk coefficient, which directly affects driving safety, is introduced as the disturbance growth amplitude. The meteorological disturbance value is calculated using the following formula:

[0022] In the formula, Indicates the value of meteorological disturbance. Indicates the climate disturbance coefficient. This indicates the weather risk factor.

[0023] This scheme uses the climate disturbance coefficient as a basis and introduces the weather risk coefficient as the disturbance growth amplitude to calculate the meteorological disturbance value. It can significantly improve the environmental adaptability of crosswind warnings. By quantifying the synergistic amplification effect of severe weather such as rainstorms and snow on crosswind effects, it breaks through the limitation of traditional models that only consider a single wind condition, realizes a comprehensive assessment of the risk of multiple meteorological elements coupled together, and thus improves driving safety under different weather conditions.

[0024] In a further implementation scheme, the calculation formula for the risk analysis model is as follows:

[0025] In the formula, , , , , , represent the first quantization parameter to the fifth quantization parameter, respectively. , , , , These represent the weight coefficients of the first to fifth quantization parameters, respectively.

[0026] This solution comprehensively considers five major factors: driver status, real-time wind conditions, route characteristics, altitude, and climate, to achieve a panoramic risk assessment and significantly improve the coverage and reliability of early warnings. By allocating different weights, it highlights the dominant risk factors in different scenarios, making the assessment results more consistent with the actual risk level and reducing false alarms and missed alarms. Attached Figure Description

[0027] Figure 1This is a flowchart of a crosswind warning method for vehicle assisted driving provided by an embodiment of the present invention.

[0028] Figure 2 This is a weather type table provided in an embodiment of the present invention; Figure 3 This is a climate type table provided in an embodiment of the present invention. Detailed Implementation

[0029] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are given for illustrative purposes only and should not be construed as limiting the present invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0030] This invention provides a crosswind warning method for vehicle assisted driving, such as... Figure 1 As shown, in this embodiment, steps S1 to S3 are included: S1. Real-time multi-dimensional data acquisition, wherein the multi-dimensional data includes at least one or more of the following: driver data, real-time wind direction data, driving data, altitude information, and climate information; Driver data is collected through the CMS (Driver Monitoring System); Real-time wind direction data includes real-time wind direction and real-time wind speed. Real-time wind direction is measured by a wind vane, whose blades rotate with the airflow direction and are converted into an electrical signal output by a sensor. Real-time wind speed is obtained by a three-cup anemometer or an ultrasonic anemometer.

[0031] Driving data is acquired through the navigation system, including but not limited to navigation route and vehicle attitude, navigation positioning; Altitude information is obtained based on navigation positioning, for example, based on GPS positioning and a DEM (Digital Elevation Model) elevation database; Climate information includes climate data and weather types. Climate data is obtained through navigation positioning, and weather types are collected and identified through vehicle-mounted cameras.

[0032] S2. Interference analysis: Feature extraction is performed on the multidimensional data, and crosswind interference correlation analysis is conducted based on the extracted data features. Quantitative parameters characterizing the impact of crosswind interference are output, including: A. Analyze the driver's state based on the driver data in the multidimensional data, and then perform crosswind interference assessment based on the driver's state to obtain a first quantitative parameter, including: Driver data is acquired for feature extraction, including eye movement frequency, duration of eye closure, and head posture data. Based on the eye movement frequency, eye closure duration, and head posture data, a multimodal feature fusion analysis of the driver's fatigue state is performed to obtain a fatigue weighting factor, which serves as the first quantitative parameter characterizing the impact of crosswind interference.

[0033] The formula for calculating the fatigue weighting factor is as follows:

[0034] in, Indicates eye movement frequency (unit: times / minute); Indicates the duration of eye closure (unit: seconds); The head offset angle represents the head pose data; Indicates the speed of neck movement; These are the eye movement frequency attenuation coefficient, the eye closure time enhancement coefficient, and the head / neck offset linkage coefficient, respectively. This represents the normal reference eye movement frequency (18 times / minute). This indicates the normal reference time for closing your eyes (0.3 seconds). This indicates the maximum reference value for head offset (45°). This indicates the reference value for the neck angular velocity (90° / second).

[0035] This fatigue weighting factor is used as the "first quantitative parameter" (fatigue factor) in the subsequent risk level as a behavioral sensitivity adjustment item. The weighting of the impact on the overall risk level R determines the intensity of the risk warning.

[0036] For example, eye movement frequency =12 times / minute, duration of eye closure =0.4 seconds, head offset angle =8 degrees, neck movement speed =25 degrees / second, then:

[0037] This embodiment collects the driver's physiological characteristics and assesses the driver's fatigue state through real-time eye movement frequency, eye closure duration, and head posture data. This data serves as the first quantitative parameter characterizing the impact of crosswind interference, enabling personalized risk assessment and providing earlier and stronger warning signals, significantly improving the accuracy and safety of the warning system.

[0038] B. Based on the real-time wind direction data and driving data in the multidimensional data, perform path crosswind sensitivity analysis to assist in crosswind interference assessment and obtain second quantitative parameters, including: Navigation routes are segmented based on driving data; Obtain the path direction, real-time wind direction, and real-time wind speed of the navigation route on each road segment; Calculate the real-time crosswind angle based on the path direction and real-time wind direction; Based on the real-time crosswind angle and the real-time wind speed, determine whether the corresponding navigation path is a crosswind sensitive section; The total length of the navigation path segments belonging to the crosswind sensitive section is statistically analyzed and normalized to obtain the crosswind interference factor, which serves as the second quantitative parameter characterizing the impact of crosswind interference.

[0039] Specifically: 1) Divide the navigation path within a preset distance L (5km) into a sequence of path segments. ; 2) Obtain the path direction (i.e., direction angle) of navigation path Pi on each road segment. (0°~360°), Real-time wind direction of the area corresponding to navigation path Pi on each road segment. (0°~360°), wind speed in the area corresponding to navigation path Pi on each road segment. (0~20m / s); 3) Calculate the real-time crosswind angle between the path direction of navigation path Pi and its corresponding real-time wind direction on each road segment. :

[0040] 4) If the following conditions are met, the navigation path segment is determined to be a crosswind sensitive segment:

[0041] An angle greater than 45° indicates that the wind direction has a strong "lateral component" (an angle exceeding 45°) relative to the path. The wind speed has reached a level that would interfere with vehicle stability. Meeting both of these conditions, this section of the road is therefore classified as a "crosswind sensitive section".

[0042] 5) The total length of road sections sensitive to crosswinds is accumulated.

[0043] The total length of multiple road segments is calculated as Ls. The larger the total length of Ls, the longer the driver will be in the crosswind environment, and the higher the driving risk.

[0044] 6) Normalize the total length Ls of the crosswind-sensitive section to obtain the path-level crosswind interference intensity index. The normalization formula is as follows:

[0045] Second quantization parameter It is used to determine whether there are sections of road ahead with strong crosswind interference, and provides path length data for the overall risk level.

[0046] In this embodiment, when performing crosswind-sensitive interference analysis, road segments are divided, and real-time crosswind angle and real-time wind speed are analyzed in a granular manner. This can accurately identify crosswind-sensitive road segments (such as bridges and wind gaps) in the navigation path, and then use the safety assessment of the navigation path as a second quantitative parameter to characterize the impact of crosswind interference, effectively enhancing driving safety.

[0047] C. Based on the real-time wind direction data and driving data in the multidimensional data, perform path crosswind force analysis to assist in crosswind interference assessment and obtain third quantitative parameters, including: Obtain the real-time crosswind angle of the navigation path on each road segment; The effective intensity ratio of the crosswind is determined by force analysis based on the real-time crosswind angle, and the average value of the effective intensity ratio of the crosswind within a preset distance is calculated to obtain the effectiveness coefficient. The formula for calculating the effective intensity ratio of crosswinds is as follows:

[0048] The formula for calculating the effectiveness coefficient is as follows:

[0049] In the formula, The crosswind effectiveness coefficient is represented by the coefficient of performance. Indicates the real-time crosswind angle. This represents the effectiveness coefficient regarding crosswinds. This indicates the maximum wind speed (usually set to 20 m / s; exceeding this speed will cause the wind to become saturated and interfere with vehicles).

[0050] The wind disturbance factor is calculated based on the effectiveness coefficient and real-time wind speed, serving as the third quantitative parameter characterizing the impact of crosswind interference.

[0051] In this embodiment, the formula for calculating the wind disturbance factor is as follows:

[0052] In the formula, Indicates the wind disturbance factor. Indicates real-time wind speed. Indicates the effectiveness coefficient. This indicates the maximum wind speed.

[0053] This embodiment performs force analysis based on the real-time crosswind angle of the navigation path. Through physical modeling, the geometric influence of wind direction is transformed into dynamic parameters, quantifying the interference of crosswind on vehicles. This significantly improves the scientificity and accuracy of crosswind interference assessment, provides a more reliable quantitative basis for the early warning system, and thus improves the accuracy of crosswind warning.

[0054] D. Based on the driving data and altitude information in the multidimensional data, perform altitude-wind-pressure adaptation analysis to assist in crosswind interference assessment and obtain a fourth quantitative parameter, including: Navigation routes are segmented based on driving data; Obtain the altitude of the navigation path on each segment and generate an altitude sequence; Calculate the average altitude of the navigation path within a preset distance based on the altitude sequence; The altitude disturbance factor is determined based on the average altitude and normalized to serve as the fourth quantitative parameter characterizing the impact of crosswind interference.

[0055] Among them, the fourth quantization parameter It is used to supplement the uncertainty of wind disturbance caused by altitude factors in the path, and works with the second quantitative parameter to improve the accuracy of "environmental wind disturbance potential".

[0056] Specifically: 1) Obtain the elevation sequence corresponding to the path segment sequence: ; 2) Calculate the average altitude:

[0057] 3) Determine the altitude disturbance factor based on the average altitude. :

[0058] in, The altitude disturbance amplification factor is 0.1 to 0.3, with 0.2 recommended to better reflect the real feeling of the enhancement of wind disturbance by altitude and ensure moderate sensitivity of the alert. 4) Altitude disturbance factor Normalization is performed using the following formula:

[0059] In the formula, This represents the maximum altitude disturbance value.

[0060] In this embodiment, The altitude is set to 4000m. This corresponds to high altitudes (e.g., 4000m), where winds are stronger and the risks are greater.

[0061] This embodiment improves the stability of the algorithm by calculating the average altitude, and determines the altitude disturbance factor as the fourth quantitative parameter to characterize the impact of crosswind interference based on the average altitude. This enables the quantitative characterization of regional terrain features, provides a key environmental correction benchmark for crosswind risk assessment, and achieves scene adaptation in different altitude areas. This significantly improves the accuracy and reliability of the crosswind early warning system in different geographical environments.

[0062] E. Based on the climate information in the multidimensional data, perform regional wind field characteristic analysis to assist in crosswind interference assessment and obtain the fifth quantitative parameter, including: Climate data from different regions are acquired in advance, and big data analysis is performed based on the climate data to obtain the climate interference coefficient of different climate zones on crosswinds. Different weather types are obtained in advance, and big data analysis is performed based on the weather types to obtain the weather risk coefficient of different weather environments on the impact of crosswinds; Based on the real-time collected climate information, the climate interference coefficient and weather risk coefficient under the current driving environment are determined. The meteorological disturbance value is calculated based on the climate interference coefficient and weather risk coefficient, and then normalized as the fifth quantitative parameter characterizing the impact of crosswind interference.

[0063] Specifically: 1) Based on the climate zone of the navigation route, refer to a table (e.g., Figure 2 Determine the climate disturbance coefficient .

[0064] For example, if the navigation path is located in a temperate continental climate zone, then It is 1.0.

[0065] 2) Based on the current weather type, use the table lookup method (e.g.) Figure 3 Determine the corresponding risk coefficient Risk(T).

[0066] For example, if the current weather is foggy, then Risk(T) is 0.8.

[0067] 3) Calculate the meteorological disturbance value based on the aforementioned climate disturbance coefficient and weather risk coefficient, specifically as follows: Based on the aforementioned climate disturbance coefficient, the weather risk coefficient, which directly affects driving safety, is introduced as the disturbance growth amplitude. The meteorological disturbance value is calculated using the following formula:

[0068] In the formula, Indicates the value of meteorological disturbance. Indicates the climate disturbance coefficient. This indicates the weather risk factor.

[0069] 4) Meteorological disturbance values Normalization yields the meteorological disturbance factor. The normalization formula is as follows:

[0070] In the formula, This represents the maximum combined meteorological disturbance value, which can be determined based on the weather type table and the climate type table under the most extreme climate zone and the most extreme weather combination. Substitute values ​​such as 1.4 (polar ice cap climate) and 0.9 (hail) into the equation. Formula acquisition.

[0071] This embodiment uses the climate disturbance coefficient as a basis and introduces the weather risk coefficient as the disturbance growth amplitude to calculate the meteorological disturbance value, which can significantly improve the environmental adaptability of crosswind warning. By quantifying the synergistic amplification effect of severe weather such as rainstorms and snow on crosswind effects, it breaks through the limitation of traditional models that only consider a single wind condition, realizes the comprehensive assessment of the risk of multiple meteorological elements coupled together, and thus improves driving safety under different weather conditions.

[0072] The steps A through E above are not in any particular order and can be executed synchronously or asynchronously depending on actual needs.

[0073] This embodiment breaks through the limitations of traditional methods that rely solely on vehicle-mounted sensors. It uses driver data, real-time wind direction data, driving data, altitude information, and climate information as basic data. Based on the correlation between crosswind interference, it performs multi-dimensional cross-analysis (multi-dimensional analysis of user status, vehicle driving status, and regional altitude / climate achieves scenario fusion) to construct a panoramic risk assessment model. This model has strong scenario adaptability and greatly improves the system's early warning capability and coverage.

[0074] S3. Early warning decision: Input the quantitative parameters of multi-dimensional data into the risk analysis model for fusion calculation, determine the early warning level based on the calculation results, and execute the driving warning according to the early warning level.

[0075] In this embodiment, the calculation formula of the risk analysis model is as follows:

[0076] In the formula, , , , , , represent the first quantization parameter to the fifth quantization parameter, respectively. , , , , These represent the weight coefficients of the first to fifth quantization parameters, respectively.

[0077] This embodiment comprehensively considers five major factors: driver status, real-time wind conditions, route characteristics, altitude, and climate, to achieve a panoramic risk assessment, significantly improving the coverage and reliability of early warnings. By differentiating weight allocation, it highlights the dominant risk factors in different scenarios, making the assessment results more consistent with the actual risk level and reducing false alarms and missed alarms.

[0078] This invention performs multi-dimensional data acquisition and feature extraction, then conducts crosswind interference correlation analysis, outputs quantitative parameters characterizing the impact of crosswind interference, and inputs them into a risk analysis model for fusion calculation. Based on the calculation results, the warning level is determined. By integrating multi-dimensional data such as driver data, real-time wind direction data, driving data, altitude information, and climate information, a dynamic prediction model is constructed to calculate the risk warning coefficient in real time, ensuring prediction accuracy and real-time performance. By extending the warning response time through advance prediction, the safety of high-speed driving is effectively improved.

[0079] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A crosswind warning method for vehicle assisted driving, characterized in that, Including the following steps: Real-time multidimensional data acquisition is performed, and the multidimensional data includes at least one or more of the following: driver data, real-time wind direction data, driving data, altitude information, and climate information; Interference analysis involves extracting features from the multidimensional data, performing crosswind interference correlation analysis based on the extracted data features, and outputting quantitative parameters characterizing the impact of crosswind interference. The early warning decision-making process involves inputting the quantitative parameters of multidimensional data into a risk analysis model for fusion calculation, determining the early warning level based on the calculation results, and then executing a driving warning according to the early warning level. Feature extraction is performed on the multidimensional data, and crosswind interference correlation analysis is conducted based on the extracted data features to output quantitative parameters characterizing the impact of crosswind interference, including: Based on the driver data in the multidimensional data, the driver status is analyzed, and then the crosswind interference is assessed based on the driver status to obtain the first quantitative parameter. Based on the real-time wind direction data and driving data in the multidimensional data, a crosswind sensitivity analysis is performed to assist in the assessment of crosswind interference and obtain the second quantitative parameter. Based on the real-time wind direction data and driving data in the multidimensional data, a path crosswind force analysis is performed to assist in crosswind interference assessment and obtain a third quantitative parameter. Based on the driving data and altitude information in the multidimensional data, altitude wind pressure adaptation analysis is performed to assist in crosswind interference assessment and obtain the fourth quantitative parameter. Based on the climate information in the multidimensional data, regional wind field characteristics are analyzed to assist in the assessment of crosswind interference and obtain the fifth quantitative parameter.

2. The crosswind warning method for vehicle assisted driving as described in claim 1, characterized in that, Based on the driver data analysis in the multidimensional data, the driver's state is analyzed, and then crosswind interference is assessed based on the driver's state to obtain the first quantitative parameter, including: Driver data is acquired for feature extraction, including eye movement frequency, duration of eye closure, and head posture data. Based on the eye movement frequency, eye closure duration, and head posture data, a multimodal feature fusion analysis of the driver's fatigue state is performed to obtain a fatigue weighting factor, which serves as the first quantitative parameter characterizing the impact of crosswind interference.

3. The crosswind warning method for vehicle assisted driving as described in claim 2, characterized in that, Based on the real-time wind direction data and driving data in the multidimensional data, a crosswind sensitivity analysis is performed to assist in the assessment of crosswind interference and obtain a second quantitative parameter, including: Navigation routes are segmented based on driving data; Obtain the path direction, real-time wind direction, and real-time wind speed of the navigation route on each road segment; Calculate the real-time crosswind angle based on the path direction and real-time wind direction; Based on the real-time crosswind angle and the real-time wind speed, determine whether the corresponding navigation path is a crosswind sensitive section; The total length of the navigation path segments belonging to the crosswind sensitive section is statistically analyzed and normalized to obtain the crosswind interference factor, which serves as the second quantitative parameter characterizing the impact of crosswind interference.

4. A crosswind warning method for vehicle assisted driving as described in claim 3, characterized in that, Based on the real-time wind direction data and driving data in the multidimensional data, a path crosswind force analysis is performed to assist in the assessment of crosswind interference and obtain a third quantitative parameter, including: Obtain the real-time crosswind angle of the navigation path on each road segment; The effective intensity ratio of the crosswind is determined by force analysis based on the real-time crosswind angle, and the average value of the effective intensity ratio of the crosswind within a preset distance is calculated to obtain the effectiveness coefficient. The wind disturbance factor is calculated based on the effectiveness coefficient and real-time wind speed, serving as the third quantitative parameter characterizing the impact of crosswind interference.

5. A crosswind warning method for vehicle assisted driving as described in claim 4, characterized in that, The formula for calculating the wind disturbance factor is as follows: In the formula, Indicates the wind disturbance factor. Indicates real-time wind speed. Indicates the effectiveness coefficient. This indicates the maximum wind speed.

6. A crosswind warning method for vehicle assisted driving as described in claim 5, characterized in that, Based on the driving data and altitude information in the multidimensional data, an altitude-wind-pressure adaptation analysis is performed to assist in the assessment of crosswind interference and obtain a fourth quantitative parameter, including: Navigation routes are segmented based on driving data; Obtain the altitude of the navigation path on each segment and generate an altitude sequence; Calculate the average altitude of the navigation path within a preset distance based on the altitude sequence; The altitude disturbance factor is determined based on the average altitude and normalized to serve as the fourth quantitative parameter characterizing the impact of crosswind interference.

7. A crosswind warning method for vehicle assisted driving as described in claim 6, characterized in that, Based on the climate information in the multidimensional data, regional wind field characteristics are analyzed to assist in crosswind interference assessment and obtain the fifth quantitative parameter, including: Climate data from different regions are acquired in advance, and big data analysis is performed based on the climate data to obtain the climate interference coefficient of different climate zones on crosswinds. Different weather types are obtained in advance, and big data analysis is performed based on the weather types to obtain the weather risk coefficient of different weather environments on the impact of crosswinds; Based on the real-time collected climate information, the climate interference coefficient and weather risk coefficient under the current driving environment are determined. The meteorological disturbance value is calculated based on the climate interference coefficient and weather risk coefficient, and then normalized as the fifth quantitative parameter characterizing the impact of crosswind interference.

8. A crosswind warning method for vehicle assisted driving as described in claim 7, characterized in that, The meteorological disturbance value is calculated based on the aforementioned climate disturbance coefficient and weather risk coefficient, specifically as follows: Based on the aforementioned climate disturbance coefficient, the weather risk coefficient, which directly affects driving safety, is introduced as the disturbance growth amplitude. The meteorological disturbance value is calculated using the following formula: In the formula, Indicates the value of meteorological disturbance. Indicates the climate disturbance coefficient. This indicates the weather risk factor.

9. A crosswind warning method for vehicle assisted driving as described in claim 1, characterized in that, The calculation formula for the risk analysis model is as follows: In the formula, , , , , , represent the first quantization parameter to the fifth quantization parameter, respectively. , , , , These represent the weight coefficients of the first to fifth quantization parameters, respectively.

Images