A vehicle active safety warning system threshold adjustment method and system

Abstract

The present invention discloses a threshold adjustment method and system for a vehicle active safety warning system. By collecting real-time vehicle driving data, driving environment data, driver status data, and active safety warning data, a vehicle driving safety warning dataset and a driver abnormal status warning dataset are constructed for warning mode categories coupled with different driving road scenarios and driving weather environments. Warning lag and warning advance threshold evaluation indicators are established for each warning category. The warning threshold is adjusted and judged based on the evaluation indicators, and the warning threshold is adjusted based on the evaluation results. The present invention can continuously quantitatively evaluate and adjust the warning threshold of the active safety system based on the warning data collected in real time during the warning system and vehicle driving, further improving the reliability and robustness of the active safety warning system.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle active safety, and in particular to a method and system for adjusting a threshold value of a vehicle active safety warning system. Background Art

[0002] In recent years, rapid economic and social development has made cars an indispensable part of people's daily lives, leading to a rapid increase in the number of cars and drivers nationwide. According to data from the National Bureau of Statistics, by the end of 2021, China's car ownership had exceeded 300 million, with over 440 million drivers. This has led to severe traffic safety issues, with frequent traffic accidents and particularly severe personal and property damage. Analysis of numerous previous vehicle collision accidents shows that if drivers can foresee dangerous situations in advance, the incidence of accidents can be significantly reduced. For example, if drivers recognize safety risks 0.5 seconds in advance and take appropriate measures, approximately 30% of head-on collisions, 50% of road-related accidents, and 60% of rear-end collisions can be avoided. Based on this fact, with the rapid development of intelligent connected communication technology, active vehicle safety warning systems are being widely used to improve driving safety.

[0003] Currently, numerous active safety warning systems have been developed and widely adopted in the automotive market, including the Mercedes-Benz Pre-Safe, Honda SENSING, Toyota TSS, Mobileye AWS, and Tesla Autopilot systems, as well as the SenseDrive system in China. While most of these systems can achieve basic functions and improve driving safety, the vast majority of these systems employ fixed warning algorithms, i.e., fixed warning thresholds. These algorithms are unable to adapt to complex driving environments, resulting in issues with the reliability and safety of these warnings. These systems may not function properly in certain scenarios, or even interfere with the driver's normal operation, creating new safety hazards. Therefore, real-time analysis of the driving environment and timely adjustment of warning thresholds are crucial for the safe and effective operation of active safety warning systems. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to address the problem that the existing active safety warning system has a fixed built-in warning threshold and thus has certain safety risks.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A method for adjusting a threshold value of a vehicle active safety warning system comprises the following steps:

[0007] Step S1: Based on the warning system data when the warning occurs during vehicle driving and the preset driving data of each type, based on the preset warning modes of each type, obtain the corresponding warning threshold value x of each warning mode. i The early warning data set ξ i ,The data set includes warning system data and corresponding preset driving data of various types, i is the warning mode mark;

[0008] Step S2: Extract the corresponding warning threshold value x of each warning mode i The early warning data set ξ i The evaluation data associated with the early warning mode i is used as the supply data of the early warning mode i, and the supply data set η corresponding to each early warning mode is obtained. i , where the associated evaluation data is the warning system data when warning mode i occurs, and the corresponding indicator data for judging the warning effect in each type of preset driving data;

[0009] Step S3: For each early warning mode, establish the demand function of early warning hysteresis discrimination corresponding to each early warning mode and the demand function for early warning and early identification

[0010] Step S4: Demand function for early warning hysteresis discrimination based on each early warning mode and the demand function for early warning and early identification Combined with the supply data set η corresponding to each early warning mode i , establish the discriminant limit state function of the warning lag corresponding to each warning mode and the limit state function of early warning

[0011] Step S5: Determine the limit state function based on the warning hysteresis corresponding to each warning mode and the limit state function of early warning Establish the function of warning hysteresis discrimination corresponding to each warning mode and early warning function

[0012] Step S6: Based on the function of warning hysteresis discrimination corresponding to each warning mode and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure

[0013] Step S7: Based on the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure The warning threshold χ corresponding to each warning mode i Make assessment adjustments.

[0014] As a preferred technical solution of the present invention, in step S3, A minimum index calculation function that satisfies vehicle travel and driving safety requirements when early warning mode i occurs is used;

[0015]

[0016] Where, d ci It indicates the maximum value of the early warning mode i discrimination index that allows early warning.

[0017] As a preferred technical solution of the present invention, in step S4,

[0018]

[0019]

[0020] Where, X i is the function variable space of warning category i, represents the explicit function of the limit state function for the warning lag of warning category i, It represents the explicit function of the limit state function of early warning category i, S i is the supply function of warning category i, which is based on the supply set η i The dependent variable is fitted to obtain a variable function in the demand function. As a preferred technical solution of the present invention, in step S5:

[0021]

[0022]

[0023] Where, X ij Represents the jth variable in the function variable space of early warning mode i, j is the variable identifier, and n is the function variable space X i Total number of variables, For the function variable space in the function A point on the limit state surface is For the function variable space in the function A point on the limit state surface is

[0024] As a preferred technical solution of the present invention, in step S6, for each warning mode, based on the warning hysteresis discrimination function corresponding to each warning mode, and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode through the following steps respectively: and the probability of early warning failure

[0025] Step S6.1: Based on the function variable space X corresponding to the early warning mode, determine whether each variable in X obeys the normal distribution. For variables that obey the normal distribution, calculate the mean value corresponding to the variable. and variance Then execute step S63; for variables that do not obey the normal distribution, execute step S6.2;

[0026] Step S6.2: Substitute the variables X in X that do not follow a normal distribution j Perform equivalent normalization to obtain X' j , calculate X' j Standard deviation and mean As variable X j The mean and standard deviation Execute step S6.3;

[0027]

[0028]

[0029] Where, Represents variable X j The probability density function of Represents variable X j The cumulative distribution function of ; Φ function is the standard normal distribution function; The value of the variable identified by variable j that makes the limit state function equal to 0 for the variable state space X;

[0030] Step S6.3: Based on the mean values ​​of each variable in X, obtain x * , x * =μ X , execute step S6.4;

[0031] Step S6.4: Based on x * , calculate the performance function Z through the following steps L The mean μ Z and standard deviation σ Z , execute step S6.5;

[0032]

[0033] Where, is the variable Xk With X j The correlation coefficient between them, n is the total number of variables in the function variable space X, g(X) is the explicit function for discriminating the limit state function;

[0034] Step S6.5: Based on the performance function Z L The standard deviation σ Z , calculate the variable X through the following steps j Sensitivity coefficient Execute step S6.6;

[0035]

[0036] Step S6.6: Based on the performance function Z L The mean μ Z and standard deviation σ Z ,pass Calculate the system reliability index β and execute step S6.7;

[0037] Step S6.7: Determine |||x * ||-||x^|||<ε, ε is the preset error value, ||x * || is x * If yes, then output the value of β in step S6.6 as the reliability index and execute step S6.8; if no, then update x * , let x * =x^, return to step S6.4;

[0038] Step S6.8: Based on the reliability index β, obtain the failure probability P f For: P f =Φ(-β)=1-Φ(β).

[0039] As a preferred technical solution of the present invention, in step S7, the warning hysteresis failure probability corresponding to each warning mode is calculated. and the probability of early warning failure Through the following steps, the warning threshold χ corresponding to each warning mode is i Make assessment adjustments.

[0040] Step S7.1: Determine the probability of early warning hysteresis failure Is it greater than the preset maximum allowable failure probability P? 0i If it is greater than, then execute step S7.2; if it is not greater than, then execute step S7.3; wherein, the warning threshold χ i Defined as the warning monitoring value when the warning mode i Issue warnings when is the early warning monitoring value;

[0041] Step S7.2: The warning threshold value χ corresponding to the warning category i i Make adjustments to obtain new thresholds for warning category i

[0042] Step S7.3: Determine the probability of early warning failure Is it greater than the preset maximum allowable failure probability P 0i If it is greater than, execute step S7.4; if it is not greater than, keep the warning threshold χ corresponding to the current warning category i i constant;

[0043] Step S7.4: The warning threshold value χ corresponding to the warning category i i Make adjustments to obtain new thresholds for warning category i

[0044] As a preferred technical solution of the present invention, in step S1, the preset various types of driving data include preset various types of vehicle operation data, preset various types of driving environment data, and preset various types of driver status data.

[0045] A system based on the threshold adjustment method of the vehicle active safety warning system includes a data acquisition module, a warning mode supply module, a demand function module, a limit state judgment function module, and a warning threshold evaluation model.

[0046] The data acquisition module is used to obtain the warning system data when the warning occurs during vehicle driving, as well as the preset various types of driving data, based on the preset various types of warning modes, and obtain the corresponding warning threshold value χ of each warning mode. i The data set ξ i ;

[0047] The early warning mode supply module is based on the corresponding early warning threshold χ i The data set ξ i , the early warning system data when the early warning occurs is used as the supply data of the early warning mode, and the supply data set S corresponding to each early warning mode is obtained i ;

[0048] The demand function module is used to establish the demand function of warning hysteresis discrimination corresponding to each warning mode. and the demand function for early warning and early identification

[0049] Step S4: Demand function for early warning hysteresis discrimination based on each early warning mode and the demand function for early warning and early identification Combined with the supply data set S corresponding to each early warning mode i , establish the discriminant limit state function of the warning lag corresponding to each warning mode and the limit state function of early warning

[0050] The limit state function module is used to determine the limit state function based on the warning lag corresponding to each warning mode. and the limit state function of early warning Establish the function of warning hysteresis discrimination corresponding to each warning mode and early warning function

[0051] The warning threshold evaluation model is used to determine the function of warning hysteresis based on each warning mode. and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure Then the warning threshold χ corresponding to each warning mode is i Make assessment adjustments.

[0052] A vehicle active safety warning system threshold adjustment terminal includes a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to execute a vehicle active safety warning system threshold adjustment method.

[0053] The beneficial effects of the present invention are as follows: (1) The active safety system threshold adjustment method of the present invention adopts a system reliability theory method to establish a system risk assessment model based on the failure mode of the active safety warning system when the vehicle is driving, quantitatively assesses the reliability of the active safety warning system when the vehicle is driving, and provides a theoretical reference for the dynamic adjustment of the warning threshold of the active safety system;

[0054] (2) The design method of the present invention realizes real-time collection and analysis of warning data and continuous adjustment of the warning threshold of the active safety system. The warning threshold meets the safety requirements under different road environments, thereby improving the safety and reliability of vehicles using the active safety warning system in actual operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A flow chart of a threshold adjustment method for an active safety warning system designed for the present invention. DETAILED DESCRIPTION

[0056] The present invention will be further described below in conjunction with the accompanying drawings. The following examples can enable those skilled in the art to more fully understand the present invention, but are not intended to limit the present invention in any way.

[0057] A method for adjusting a threshold value of a vehicle active safety warning system comprises the following steps:

[0058] Step S1: Based on the warning system data when the warning occurs during vehicle driving and the preset driving data of each type, based on the preset warning modes of each type, obtain the corresponding warning threshold value x of each warning mode. i The early warning data set ξ i ,The data set includes warning system data and corresponding preset driving data of various types, and i is the warning mode mark.

[0059] In this embodiment, the early warning system data based on the occurrence of early warnings during vehicle driving, and the preset various types of driving data are data when early warnings occur multiple times within a preset historical time period. Various types of driving data are temporally and spatially matched with the early warning system data during early warnings. In step S1, the preset various types of driving data include preset various types of vehicle operation data, preset various types of driving environment data, and preset various types of driver status data. The preset various types of vehicle operation data include speed, acceleration, and GPS. The preset various types of driving environment data include road geometry and weather conditions. The preset various types of driver status data include eye movement data, heart rate data, steering wheel control data, and pedal control data. The preset various types of early warning modes are divided into early warning modes based on road geometry and weather conditions, including forward collision (sunny day, straight ahead) warning mode, driving fatigue warning (sunny day, straight ahead) warning mode, based on Figure 1 The warning categories shown in the figure are divided into warning modes based on road geometry and weather conditions. The warning categories include forward collision warning category, side collision warning category, fatigue driving warning category, distracted driving warning category, and smoking driving warning category.

[0060] Step S2: Extract the corresponding warning threshold value x of each warning mode i The early warning data set ξ i The evaluation data associated with the early warning mode i is used as the supply data of the early warning mode i, and the supply data set η corresponding to each early warning mode is obtained. i , where the associated evaluation data is the warning system data when warning mode i occurs, and the corresponding indicator data for distinguishing the warning effect in each type of preset driving data, that is, the various types of driving data for distinguishing the warning effect; the indicator data for distinguishing the warning effect is the indicator data for distinguishing the warning effect specified in the industry for different warning categories. For example, when a forward collision (sunny day, straight driving) warning occurs, η iis the data set of the relative distance between the two vehicles in the vehicle driving data when the warning occurs; when the driving fatigue warning (rainy day, turning) occurs, η i This is a dataset of driver blink counts and cardiovascular and cerebrovascular-related fatigue indicators in driver status data.

[0061] Step S3: For each early warning mode, establish the demand function of early warning hysteresis discrimination corresponding to each early warning mode and the demand function for early warning and early identification

[0062] In the step S3, The minimum indicator calculation function that meets the vehicle's driving and driving safety requirements when warning mode i occurs is used; for example, when i is a forward collision warning, This model is used to predict the distance between vehicles and the road, ensuring they brake to the minimum safe distance after a warning warning when a straight-ahead collision is imminent. Reference can be made to industry standards for safety calculation functions for various types of travel and driving behaviors.

[0063]

[0064] Where, d ci It indicates the maximum value of the early warning allowed by the early warning mode i discrimination index. That is, the maximum value of the early warning allowed by the driver when the warning occurs.

[0065] Step S4: Demand function for early warning hysteresis discrimination based on each early warning mode and the demand function for early warning and early identification Combined with the supply data set η corresponding to each early warning mode i , establish the discriminant limit state function of the warning lag corresponding to each warning mode and the limit state function of early warning

[0066] In the step S4,

[0067]

[0068]

[0069] Where, X i is the function variable space of warning category i, represents the explicit function of the limit state function for the warning lag of warning category i, It represents the explicit function of the limit state function of early warning category i, S i is the supply function of warning category i, which is based on the supply set η iFitting the dependent variable to obtain the variable function in the demand function, that is, the supply function is based on the supply set η i and early warning data collectionξ i The variable function in the demand function is obtained by fitting. Step S5: The limit state function of the warning hysteresis corresponding to each warning mode is determined and the limit state function of early warning Establish the function of warning hysteresis discrimination corresponding to each warning mode and early warning function

[0070] In the step S5:

[0071]

[0072]

[0073] Where, X ij Represents the jth variable in the function variable space of early warning mode i, j is the variable identifier, and n is the function variable space X i Total number of variables, For the function variable space in the function A point on the limit state surface is For the function variable space in the function A point on the limit state surface is for The jth variable in for The jth variable in .

[0074] Step S6: Based on the function of warning hysteresis discrimination corresponding to each warning mode and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure

[0075] In the step S6, for each warning mode, the function function of the warning hysteresis judgment corresponding to each warning mode is respectively determined. and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode through the following steps respectively: and the probability of early warning failure

[0076] Functional function based on S5 early warning hysteresis discrimination and early warning function Execute step S6 respectively to obtain the probability of delayed failure of warning for warning category i and the probability of early warning failure

[0077] Probability of warning hysteresis failure for warning category i Obtained by the following steps:

[0078] Step S6.1: Based on the function variable space X corresponding to the early warning mode, determine whether each variable in X obeys the normal distribution. For variables that obey the normal distribution, calculate the mean value corresponding to the variable. and variance Then execute step S63; for variables that do not obey the normal distribution, execute step S6.2;

[0079] Step S6.2: Substitute the variables X in X that do not follow a normal distribution j Perform equivalent normalization to obtain X' j , calculate X' j Standard deviation and mean As variable X j The mean and standard deviation Execute step S6.3;

[0080]

[0081]

[0082] Where, Represents variable X j The probability density function of Represents variable X j The cumulative distribution function of ; Φ function is the standard normal distribution function; The value of the variable identified as j that makes the discriminant limit state function of the warning hysteresis equal to 0 for the variable state space X;

[0083] Step S6.3: Based on the mean values ​​of each variable in X, obtain x * , x * =μ X , execute step S6.4;

[0084] Step S6.4: Based on x * , calculate the warning hysteresis discrimination function Z through the following steps L The mean μ Z and standard deviation σ Z , execute step S6.5;

[0085]

[0086]

[0087] Where, is the variable X k With X j The correlation coefficient n between them is the total number of variables in the function variable space X, and g(X) is the explicit function of the discriminant limit state function of the warning lag;

[0088] Step S6.5: Function Z based on early warning hysteresis discrimination L The standard deviation σ Z , calculate the variable X through the following steps j Sensitivity coefficient Execute step S6.6;

[0089]

[0090] Step S6.6: Function Z based on early warning hysteresis discrimination L The mean μ Z and standard deviation σ Z ,pass Calculate the system reliability index β and execute step S6.7;

[0091] Step S6.7: Determine |||x * ||-||x^|||<ε, ε is the preset error value, is the jth variable in x^, ||x * || is x * If yes, then output the value of β in step S6.6 as the reliability index and execute step S6.8; if no, then update x * , let x * =x^, return to step S6.4;

[0092] Step S6.8: Based on the reliability index β, obtain the failure probability P f For: P f =Φ(-β)=1-Φ(β). That is, the probability of warning hysteresis failure of warning category i is obtained

[0093] Probability of early failure of warning for warning category i Obtained by the following steps:

[0094] Step S6.1: Based on the function variable space X corresponding to the early warning mode, determine whether each variable in X obeys the normal distribution. For variables that obey the normal distribution, calculate the mean value corresponding to the variable. and variance Then execute step S63; for variables that do not obey the normal distribution, execute step S6.2;

[0095] Step S6.2: Substitute the variables X in X that do not follow a normal distribution j Perform equivalent normalization to obtain X' j , calculate X' j Standard deviation and mean As variable X j The mean and standard deviation Execute step S6.3;

[0096]

[0097]

[0098] Where, Represents variable X j The probability density function of Represents variable X j The cumulative distribution function of ; Φ function is the standard normal distribution function; The value of the variable identified as j that makes the limit state function of early warning equal to 0 for the variable state space X;

[0099] Step S6.3: Based on the mean values ​​of each variable in X, obtain x * , x * =μ X , execute step S6.4;

[0100] Step S6.4: Based on x * , calculate the performance function Z through the following steps L The mean μ Z and standard deviation σ Z , execute step S6.5;

[0101]

[0102] Where, is the variable X k With X j The correlation coefficient between them, n is the total number of X variables in the function variable space, g(x * ) The early warning function is the explicit function of the limit state function;

[0103] Step S6.5: Based on the early warning function Z L The standard deviation σ Z , calculate the variable X through the following steps j Sensitivity coefficient Execute step S6.6;

[0104]

[0105] Step S6.6: Based on the early warning function Z L The mean μ Z and standard deviation σ Z ,pass Calculate the system reliability index β and execute step S6.7;

[0106] Step S6.7: Determine ||x * ||-||x^|||<ε, ε is the preset error value, ||x * || is x * If yes, then output the value of β in step S6.6 as the reliability index and execute step S6.8; if no, then update x * , let x * =x^, return to step S6.4;

[0107] Step S6.8: Based on the reliability index β, obtain the failure probability P f For: P f =Φ(-β)=1-Φ(β). That is, the probability of early failure of warning for warning category i is obtained

[0108] Step S7: Based on the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure The warning threshold χ corresponding to each warning mode i Make assessment adjustments.

[0109] In step S7, the warning hysteresis failure probability corresponding to each warning mode is and the probability of early warning failure Through the following steps, the warning threshold χ corresponding to each warning mode is i Make assessment adjustments.

[0110] Step S7.1: Determine the probability of early warning hysteresis failure Is it greater than the preset maximum allowable failure probability P? 0i In this embodiment, the maximum allowable failure probability P 0i If it is greater than 85%, it means that the system has a built-in preset threshold standard χ i If the system threshold is too large, the forward collision warning is delayed, then step S7.2 is executed; if it is not greater than, then step S7.3 is executed; wherein the warning threshold χ i Defined as the warning monitoring value when the warning mode i Issue warnings when is the early warning monitoring value; that is, the early warning threshold value in the present invention is changed, and the judgment standard is unified as Issue early warning.

[0111] Step S7.2: The warning threshold value χ corresponding to the warning category i i Make adjustments to obtain new thresholds for warning category i

[0112] Step S7.3: Determine the probability of early warning failure Is it greater than the preset maximum allowable failure probability P 0i If it is greater than, it means that the system has a built-in preset threshold standard χ i If the threshold value is too small, the system warning is too sensitive, and the system threshold value is set too small, resulting in an early forward collision warning, then execute step S7.4; if it is not greater than, maintain the warning threshold value χ corresponding to the current warning category i i constant;

[0113] Step S7.4: The warning threshold value χ corresponding to the warning category i i Make adjustments to obtain new thresholds for warning category i

[0114] In addition, the establishment of the data set and in step S1 of the present invention includes obtaining data records η under different warning categories z zp , including vehicle driving data, warning data, and driving environment data, where p is the data under warning category z and is identified as the pth record. Based on machine learning or deep learning algorithms, data records η under different warning types z are respectively zp , based on the recognition and classification of driving environment data, different driving scene warnings are obtained, that is, each warning mode is preset, and then the data set of different warning modes i is obtained. i The driving environment data includes road geometry and weather conditions. The road geometry includes straight driving, turning, uphill and downhill straight driving, and uphill and downhill turning. The weather conditions include rainy days, sunny days, thundering days, and snowy days. Each road geometry condition is combined with each weather condition one by one and combined with the warning category to form a warning mode.

[0115] The present invention also designs a system based on the threshold adjustment method of the vehicle active safety warning system, including a data acquisition module, a warning mode supply module, a demand function module, a limit state judgment function module, and a warning threshold evaluation model.

[0116] The data acquisition module is used to obtain the warning system data when the warning occurs during vehicle driving, as well as the preset various types of driving data, based on the preset various types of warning modes, and obtain the corresponding warning threshold value χ of each warning mode. i The data set ξ i ;

[0117] The early warning mode supply module is based on the corresponding early warning threshold χ i The data set ξ i , the early warning system data when the early warning occurs is used as the supply data of the early warning mode, and the supply data set S corresponding to each early warning mode is obtained i ;

[0118] The demand function module is used to establish the demand function D corresponding to each warning mode for each warning mode. i * and the demand function for early warning and early judgment

[0119] Step S4: Demand function for early warning hysteresis discrimination based on each early warning mode and the demand function for early warning and early identification Combined with the supply data set S corresponding to each early warning mode i , establish the discriminant limit state function of the warning lag corresponding to each warning mode and the limit state function of early warning

[0120] The limit state function module is used to determine the limit state function based on the warning lag corresponding to each warning mode. and the limit state function of early warning Establish the function of warning hysteresis discrimination corresponding to each warning mode and early warning function

[0121] The warning threshold evaluation model is used to determine the function of warning hysteresis based on each warning mode. and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure Then the warning threshold χ corresponding to each warning mode is i Make assessment adjustments.

[0122] A vehicle active safety warning system threshold adjustment terminal includes a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to execute a vehicle active safety warning system threshold adjustment method.

[0123] The present invention designs a method and system for adjusting the threshold of a vehicle active safety warning system. The active safety system threshold adjustment method of the present invention adopts a system reliability theory method to establish a system risk assessment model based on the failure mode of the active safety warning system when the vehicle is driving, quantitatively evaluates the reliability of the active safety warning system when the vehicle is driving, and provides a theoretical reference for the dynamic adjustment of the warning threshold of the active safety system; through the design method of the present invention, real-time collection and analysis of warning data and continuous adjustment of the warning threshold of the active safety system are achieved, and the warning threshold meets the safety requirements in different road environments, thereby improving the safety and reliability of vehicles using the active safety warning system in actual operation.

[0124] The above are only preferred embodiments of the present invention, but do not limit the scope of the patent of the present invention. Although the present invention has been described in detail with reference to the above embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the above embodiments or to replace some of the technical features therein with equivalents. Any equivalent structure made by using the contents of the present invention specification and drawings, directly or indirectly applied to other related technical fields, is also within the scope of protection of the patent of the present invention.

Claims

1. A method for adjusting a threshold value of a vehicle active safety warning system, characterized by: The following steps are involved: Step S1: Based on the warning system data when the warning occurs during vehicle driving and the preset driving data of each type, based on the preset warning modes of each type, obtain the corresponding warning threshold value x of each warning mode. i The early warning data set ξ i ,The data set includes warning system data and corresponding preset driving data of various types, i is the warning mode mark; Step S2: Extract the corresponding warning threshold value x of each warning mode i The early warning data set ξ i The evaluation data associated with the early warning mode i is used as the supply data of the early warning mode i, and the supply data set η corresponding to each early warning mode is obtained. i , where the associated evaluation data is the warning system data when warning mode i occurs, and the corresponding indicator data for judging the warning effect in each type of preset driving data; Step S3: For each early warning mode, establish the demand function of early warning hysteresis discrimination corresponding to each early warning mode and the demand function for early warning and early identification Step S4: Demand function for early warning hysteresis discrimination based on each early warning mode and the demand function for early warning and early identification Combined with the supply data set η corresponding to each early warning mode i , establish the discriminant limit state function of the warning lag corresponding to each warning mode and the limit state function of early warning Step S5: Determine the limit state function based on the warning hysteresis corresponding to each warning mode and the limit state function of early warning Establish the function of warning hysteresis discrimination corresponding to each warning mode and early warning function Step S6: Based on the function of warning hysteresis discrimination corresponding to each warning mode and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure Step S7: Based on the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure The warning threshold χ corresponding to each warning mode i Make assessment adjustments.

2. The method for adjusting the threshold value of a vehicle active safety warning system according to claim 1, characterized in that: In the step S3, A minimum index calculation function that satisfies vehicle travel and driving safety requirements when early warning mode i occurs is used; Where, d ci It indicates the maximum value of the early warning mode i discrimination index that allows early warning.

3. The method for adjusting the threshold value of a vehicle active safety warning system according to claim 1, characterized in that: In the step S4, Where, X i is the function variable space of warning category i, represents the explicit function of the limit state function for the warning lag of warning category i, It represents the explicit function of the limit state function of early warning category i, S i is the supply function of warning category i, which is based on the supply set η i Fit the dependent variable to obtain a function of the variables in the demand function.

4. The method for adjusting the threshold value of a vehicle active safety warning system according to claim 1, characterized in that: In the step S5: Where, X ij Represents the jth variable in the function variable space of early warning mode i, j is the variable identifier, and n is the function variable space X i Total number of variables, For the function variable space in the function A point on the limit state surface is For the function variable space in the function A point on the limit state surface is 5. The method for adjusting the threshold value of a vehicle active safety warning system according to claim 1, characterized in that: In the step S6, for each warning mode, the function function of the warning hysteresis judgment corresponding to each warning mode is respectively determined. and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode through the following steps respectively: and the probability of early warning failure Step S6.1: Based on the function variable space X corresponding to the early warning mode, determine whether each variable in X obeys the normal distribution. For variables that obey the normal distribution, calculate the mean value corresponding to the variable. and variance Then execute step S63; for variables that do not obey the normal distribution, execute step S6.2; Step S6.2: Substitute the variables X in X that do not follow a normal distribution j Perform equivalent normalization to obtain X j ', calculate X j Standard deviation of ' and mean As variable X j The mean and standard deviation Execute step S6.3; Where, Represents variable X j The probability density function of Represents variable X j The cumulative distribution function of ; Φ function is the standard normal distribution function; The value of the variable identified by variable j that makes the limit state function equal to 0 for the variable state space X; Step S6.3: Based on the mean values ​​of each variable in X, obtain x * , x * =μ X , execute step S6.4; Step S6.4: Based on x * , calculate the performance function Z through the following steps L The mean μ Z and standard deviation σ Z , execute step S6.5; Where, is the variable X k With X j The correlation coefficient between them, n is the total number of variables in the function variable space X, g(X) is the explicit function for discriminating the limit state function; Step S6.5: Based on the performance function Z L The standard deviation σ Z , calculate the variable X through the following steps j Sensitivity coefficient Execute step S6.6; Step S6.6: Based on the performance function Z L The mean μ Z and standard deviation σ Z ,pass Calculate the system reliability index β and execute step S6.7; Step S6.7: Determine |||x * ||-||x^|||<ε, ε is the preset error value, ||x * || is x * If yes, then output the value of β in step S6.6 as the reliability index and execute step S6.8; if no, then update x * , let x * =x^, return to step S6.4; Step S6.8: Based on the reliability index β, obtain the failure probability P f For: P f =Φ(-β)=1-Φ(β).

6. The method for adjusting the threshold value of a vehicle active safety warning system according to claim 1, characterized in that: In step S7, the warning hysteresis failure probability corresponding to each warning mode is and the probability of early warning failure Through the following steps, the warning threshold χ corresponding to each warning mode is i Make assessment adjustments; Step S7.1: Determine the probability of early warning hysteresis failure Is it greater than the preset maximum allowable failure probability P? 0i , if it is greater, proceed to step S7.2; If not, execute step S7.3; wherein, the warning threshold χ i Defined as the warning monitoring value when the warning mode i Issue warnings when is the early warning monitoring value; Step S7.2: The warning threshold value χ corresponding to the warning category i i Make adjustments to obtain new thresholds for warning category i Step S7.3: Determine the probability of early warning failure Is it greater than the preset maximum allowable failure probability P 0i , if it is greater, execute step S7.4; If it is not greater than, keep the warning threshold χ corresponding to the current warning category i i constant; Step S7.4: Press The warning threshold χ corresponding to warning category i i Make adjustments to obtain new thresholds for warning category i 7. The method for adjusting the threshold value of a vehicle active safety warning system according to claim 1, characterized in that: In step S1 , the preset various types of driving data include preset various types of vehicle operation data, preset various types of driving environment data, and preset various types of driver status data.

8. A system based on the vehicle active safety warning system threshold adjustment method according to any one of claims 1 to 7, characterized in that: Including data acquisition module, early warning mode supply module, demand function module, limit state judgment function module, early warning threshold evaluation model, The data acquisition module is used to obtain the warning system data when the warning occurs during vehicle driving, as well as the preset various types of driving data, based on the preset various types of warning modes, and obtain the corresponding warning threshold value χ of each warning mode. i The data set ξ i ; The early warning mode supply module is based on the corresponding early warning threshold χ of each early warning mode. i The data set ξ i , the early warning system data when the early warning occurs is used as the supply data of the early warning mode, and the supply data set S corresponding to each early warning mode is obtained i ; The demand function module is used to establish the demand function of warning hysteresis discrimination corresponding to each warning mode. and the demand function for early warning and early identification Step S4: Demand function for early warning hysteresis discrimination based on each early warning mode and the demand function for early warning and early identification Combined with the supply data set S corresponding to each early warning mode i , establish the discriminant limit state function of the warning lag corresponding to each warning mode and the limit state function of early warning The limit state function module is used to determine the limit state function based on the warning lag corresponding to each warning mode. and the limit state function of early warning Establish the function of warning hysteresis discrimination corresponding to each warning mode and early warning function The warning threshold evaluation model is used to determine the function of warning hysteresis based on each warning mode. and early warning function Obtain the warning hysteresis failure probability corresponding to each warning mode and the probability of early warning failure Then the warning threshold χ corresponding to each warning mode is i Make assessment adjustments.

9. A threshold adjustment terminal for a vehicle active safety warning system, characterized in that: It includes a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the vehicle active safety warning system threshold adjustment method described in any one of claims 1 to 7 by executing the computer instructions.

Images