A reliability evaluation method for a forward collision warning system

Abstract

The application provides a reliability evaluation method of a forward collision warning system, and the reliability evaluation method comprises the following steps: constructing a supply value model and a demand value model of the system respectively through collected vehicle driving data, establishing a reliability evaluation model of the forward collision warning system, calculating a reliability index of the vehicle forward collision warning system and a failure probability function of the forward collision warning system, and judging the reliability of the forward warning system based on the failure probability function. The application can quantitatively evaluate the reliability of the forward collision warning system when the vehicle is driving, provide a theoretical reference for the design optimization of the forward collision warning system, and the reliability evaluation result can also be used for safety risk investigation of the vehicle applying the forward collision warning system before and during actual operation.

Description

Technical Field

[0001] This invention relates to the field of vehicle active safety technology, and in particular to a reliability assessment method for a forward collision warning system. Background Technology

[0002] In recent years, with the rapid development of the economy and society, the number of cars and drivers in China has grown rapidly. As of 2021, the number of cars in China reached 302 million, and the number of drivers exceeded 440 million. This rapid growth in car ownership and the number of drivers has led to increasingly serious road traffic accidents. In terms of accident types, forward collisions account for 90% of all accidents, making them the most common and severe type of accident. Based on this situation, and with the rapid development of intelligent communication technology, forward collision warning (FCW) systems have been widely adopted.

[0003] Currently, numerous FCW (Forward Collision Warning) systems have been developed and widely applied in the automotive market both domestically and internationally. Examples include Mercedes-Benz's Pre-Safe system, Honda's CMBS system, Toyota's PCS system, Mobileye's AWS system, and Tesla's Autopilot system abroad, and SenseDrive system domestically. While most of these systems can perform basic functions and improve driving safety on public roads, some products lack adaptability to complex road environments, exhibiting issues with system reliability and warning accuracy. In some cases, false alarms can even lead to driver misjudgments and traffic accidents. Therefore, conducting comprehensive and effective reliability assessments of FCW systems before and during actual operation is of significant practical importance.

[0004] However, existing evaluation methods for FCW (Forward Collision Warning) systems are relatively simple, relying solely on indicators such as system warning accuracy and false alarm rate, as well as the driver's subjective experience, to assess system reliability. These evaluation results cannot fully reflect the system's performance. In actual road driving, if drivers react based on FCW warning information, unreliability could lead to safety risks, or even system failure causing traffic accidents. Therefore, effectively evaluating the reliability of automotive forward collision warning systems is crucial for ensuring safe driving. Summary of the Invention

[0005] The purpose of this invention is to address the problem that existing evaluation methods are relatively simple and cannot fully and effectively assess the reliability of forward collision warning systems, thus posing certain safety risks. This invention proposes a reliability evaluation method for forward collision warning systems, comprising the following steps:

[0006] S1. Collect driving data of vehicles using the forward collision warning system on the road segment, including warning system operation data, vehicle driving data, and road condition data;

[0007] S2. Based on the operational data of the warning system, the distance between the vehicle and the vehicle in front when the forward collision warning system issues a warning signal, i.e., the warning distance, is used as the supply value for the forward collision warning system. ;

[0008] S3. Based on vehicle driving data, road condition data, driver characteristics, and vehicle braking performance, establish a demand model for a forward collision warning system when vehicles are driving on the road. ;

[0009] S4. Based on the forward collision warning system supply value With demand value model The limit state equation of the forward collision warning system is established, and the function function of the forward collision warning system after linearization is obtained by using the first-order Taylor expansion to process the equation.

[0010] S5. Extract the vehicle random variable state space from the function function of the forward collision warning system, establish a reliability evaluation model for the forward collision system, and calculate the failure probability of the vehicle forward collision warning system.

[0011] S6. Based on the failure probability of the vehicle's forward collision warning system and a preset failure probability threshold, the reliability of the forward collision warning system is obtained.

[0012] Furthermore, in step S3 above, the forward collision warning system demand value model As shown in the following formula:

[0013] ,

[0014] in, The safe warning distance between the vehicle and the vehicle in front when the vehicle is traveling in a straight line; , , These are the speeds of the vehicle itself and the vehicle in front, respectively. To warn of the vehicle's maximum deceleration, The deceleration of the vehicle in front; , , These are the braking system response time and the driver's reaction delay time, respectively. It is the minimum distance between the vehicle and the vehicle in front when they are stationary relative to each other.

[0015] Furthermore, the aforementioned step S4 includes the following sub-steps:

[0016] S401. Establish the limit state equation of the forward collision warning system, as follows:

[0017] ;

[0018] S402. Using the first-order Taylor expansion to process the equations, obtain the linearized function of the forward collision warning system.

[0019] ,

[0020] in, Let the state space of the random variable be the state space of the random variable. There are a total of 7 basic variables. This represents the number of basic variables in the random variable space. For the random variable in the state space, the first... One variable, Let i be the specific value of the i-th variable on the limit state surface. A point in the state space that is on the limit state surface. .

[0021] Furthermore, the aforementioned step S5 includes the following sub-steps:

[0022] S501, Judgment If each random variable follows a normal distribution, then calculate the mean of each variable. and variance Then proceed to step S503; otherwise, proceed to step S502.

[0023] S502, will Nonnormal random variables in Perform equivalent normalization to Calculate the standard deviation of the equivalent normal distribution using the following formula. and mean ,use and As a variable Standard deviation and mean :

[0024] ,

[0025] ,

[0026] in and Random variables The probability density function and cumulative distribution function, The function is the standard normal distribution function;

[0027] S503, Set initial value , ,in for The mean;

[0028] S504. Calculate the function according to the following formula. mean and standard deviation :

[0029] ,

[0030] ,

[0031] in, For variables and The correlation coefficient between them;

[0032] S505, Calculating Variables Sensitivity coefficient As shown in the following formula:

[0033] ;

[0034] S506 calculates the reliability index of a vehicle's forward collision warning system. As shown in the following formula:

[0035] ,

[0036] in, and and are the function functions respectively. The mean and standard deviation;

[0037] S507 judgment Is the value less than ,in , for If the modulo function is true, then the output in step S506 is... If the value is used as a reliability indicator, then proceed to step S509; otherwise, proceed to step S508.

[0038] S508, Update ,make and return to step S504;

[0039] S509. Calculate the failure probability of the forward collision warning system. As shown in the following formula:

[0040] .

[0041] Furthermore, step S6 described above specifically involves: comparing the obtained failure probability with a preset system failure probability threshold. If a comparison is made, If so, the forward collision warning system is unreliable; If the forward collision warning system operates reliably, then the forward collision warning system will function reliably. The forward collision warning system is then in a critical state of failure.

[0042] Furthermore, in the aforementioned step S1, the warning system operation data includes the system warning time, the actual distance between the vehicle and the vehicle in front at the time of the warning, the vehicle driving data includes the vehicle's driving speed and maximum deceleration, the braking system delay time of different vehicles and the reaction time of different drivers, and the road condition data includes the speed data of the vehicle in front.

[0043] Furthermore, the aforementioned step S2 specifically involves: based on the early warning system operation data collected in step S1, the first... The actual distance between the two vehicles when the second warning signal was issued Construct a dataset of actual vehicle distances for early warning purposes. Among them, if a collision occurs between the vehicle and the vehicle in front but the forward collision warning system does not issue a warning signal, it is recorded as follows: Vehicle distance data at the time of the warning is used as a continuous variable in the probability distribution. The sample values ​​are used to infer the actual distribution of the variable using nonparametric tests, and this variable is then used as the supply value for the forward collision warning system. .

[0044] Furthermore, the aforementioned system failure probability threshold =85%.

[0045] Compared with the prior art, the technical solution of the present invention has the following beneficial technical effects:

[0046] (1) The reliability assessment method of the present invention adopts the system reliability theory method to establish a system risk assessment model based on the failure mode of the forward collision warning system when the vehicle is in motion, and quantitatively assesses the reliability of the forward collision warning system when the vehicle is in motion, providing a theoretical reference for the design optimization of the forward collision warning system.

[0047] (2) The failure probability of the forward collision warning system when the vehicle is in motion can be obtained by the design method of the present invention. This can be used to conduct safety risk screening for vehicles using the forward collision warning system before actual operation. At the same time, the performance of the forward collision warning system in operation can be tested and updated based on historical data. Attached Figure Description

[0048] Figure 1This is a flowchart illustrating a reliability assessment method for a forward collision warning system designed according to the present invention. Detailed Implementation

[0049] To better understand the technical content of the present invention, specific embodiments are described below in conjunction with the accompanying drawings.

[0050] In this invention, various aspects of the invention are described with reference to the accompanying drawings, in which numerous illustrative embodiments are shown. Embodiments of the invention are not limited to those depicted in the drawings. It should be understood that the invention is implemented through any of the various concepts and embodiments described above, as well as the concepts and embodiments described in detail below, because the concepts and embodiments disclosed herein are not limited to any particular implementation. Furthermore, some aspects of the invention disclosed may be used alone or in any suitable combination with other aspects of the invention disclosed.

[0051] like Figure 1 As shown in the overall flowchart of the present invention, a reliability assessment method for a forward collision warning system includes...

[0052] S1: Collects driving data of vehicles using the forward collision warning system on the road segment, including warning system operation data, vehicle driving data, and road condition data. Among them, warning system operation data includes system warning time and distance to the vehicle in front at the time of warning; vehicle driving data includes the vehicle's speed and maximum deceleration, braking system delay time of different vehicles, and reaction time of different drivers; road condition data includes the speed data of the vehicle in front.

[0053] S2. Based on the operational data of the warning system, the distance between the vehicle and the vehicle in front when the forward collision warning system issues a warning signal, i.e., the warning distance, is used as the supply value for the forward collision warning system. According to the early warning system operation data collected in step S1, the first... The actual distance between the two vehicles when the second warning signal was issued Construct a dataset of actual vehicle distances for early warning purposes. Among them, if a collision occurs between the vehicle and the vehicle in front but the forward collision warning system does not issue a warning signal, it is recorded as follows: Vehicle distance data at the time of the warning is used as a continuous variable in the probability distribution. The sample values ​​are used to infer the actual distribution of the variable using nonparametric tests, and this variable is then used as the supply value for the forward collision warning system. .

[0054] S3. Based on vehicle driving data, road condition data, driver characteristics, and vehicle braking performance, establish a demand model for a forward collision warning system when vehicles are driving on the road. : ,

[0055] in, The safe warning distance between the vehicle and the vehicle in front when the vehicle is traveling in a straight line; , , These are the speeds of the vehicle itself and the vehicle in front, respectively. To warn of the vehicle's maximum deceleration, The deceleration of the vehicle in front; , , These are the braking system response time and the driver's reaction delay time, respectively. This is the minimum distance between the vehicle and the vehicle in front when they are stationary relative to each other; a value of 3-5m is recommended.

[0056] S4. Based on the forward collision warning system supply value With demand value model The limit state equations of the forward collision warning system are established, and the linearized function of the forward collision warning system is obtained by using the first-order Taylor expansion to process the equations.

[0057] Based on the supply and demand model of the forward collision warning system, and using reliability theory, the limit state equation of the forward collision warning system is established as follows:

[0058] ;

[0059] Then, using a first-order Taylor expansion to process the equations, the linearized function of the forward collision warning system is obtained:

[0060] ,

[0061] in, Let the state space of the random variable be the state space of the random variable. There are a total of 7 basic variables. This represents the number of basic variables in the random variable space. For the random variable in the state space, the first... One variable, Let i be the specific value of the i-th variable on the limit state surface. A point in the state space that is on the limit state surface. .

[0062] S5. Extract the vehicle random variable state space from the function of the forward collision warning system, establish a reliability assessment model for the forward collision warning system, and calculate the failure probability of the vehicle's forward collision warning system. The specific steps are as follows:

[0063] S501, Judgment If each random variable follows a normal distribution, then calculate the mean of each variable. and variance Then proceed to step S503; otherwise, proceed to step S502.

[0064] S502, will Nonnormal random variables in Perform equivalent normalization to Assuming that the normalization process does not significantly alter the correlation of variables, the standard deviation of the equivalent normal distribution is calculated using the following formula. and mean ,use and As a variable Standard deviation and mean :

[0065] ,

[0066] ,

[0067] in, and Random variables The probability density function and cumulative distribution function, The function is the standard normal distribution function;

[0068] S503, Set initial value , ,in for The mean;

[0069] S504. Calculate the function according to the following formula. mean and standard deviation :

[0070] ,

[0071] ,

[0072] in, For variables and The correlation coefficient between them;

[0073] S505, Calculating Variables Sensitivity coefficient As shown in the following formula:

[0074] ;

[0075] S506 calculates the reliability index of a vehicle's forward collision warning system. As shown in the following formula:

[0076] ,

[0077] in, and and are the function functions respectively. The mean and standard deviation;

[0078] S507 judgment Is the value less than , Recommended value ,in , for If the modulo function is true, then the output in step S506 is... If the value is used as a reliability indicator, then proceed to step S509; otherwise, proceed to step S508.

[0079] S508, Update ,make and return to step S504;

[0080] S509. Calculate the failure probability of the forward collision warning system. As shown in the following formula:

[0081] .

[0082] S6. Based on the failure probability of the vehicle's forward collision warning system and a preset failure probability threshold, obtain the reliability of the forward collision warning system. Compare the obtained failure probability with the preset system failure probability threshold. Comparison, The recommended value is 85%. If so, the forward collision warning system is unreliable; If the forward collision warning system operates reliably, then the forward collision warning system will function reliably. The forward collision warning system is then in a critical state of failure.

[0083] While the present invention has been described above with reference to preferred embodiments, it is not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A reliability assessment method for a forward collision warning system, characterized in that, Includes the following steps: S1. Collect driving data of vehicles using the forward collision warning system on the road segment, including warning system operation data, vehicle driving data, and road condition data; S2. Based on the operational data of the warning system, the distance between the vehicle and the vehicle in front when the forward collision warning system issues a warning signal, i.e., the warning distance, is used as the supply value for the forward collision warning system. ; S3. Based on vehicle driving data, road condition data, driver characteristics, and vehicle braking performance, establish a demand model for a forward collision warning system when vehicles are driving on the road. ; S4. Based on the forward collision warning system supply value With demand value model The limit state equation of the forward collision warning system is established, and the function function of the forward collision warning system after linearization is obtained by using the first-order Taylor expansion to process the equation. S5. Extract the vehicle random variable state space from the function function of the forward collision warning system, establish a reliability evaluation model for the forward collision system, and calculate the failure probability of the vehicle forward collision warning system. S6. Based on the failure probability of the vehicle's forward collision warning system and a preset failure probability threshold, the reliability of the forward collision warning system is obtained.

2. The reliability assessment method for a forward collision warning system according to claim 1, characterized in that, In step S3, the forward collision warning system demand value model As shown in the following formula: ， in, The safe warning distance between the vehicle and the vehicle in front when the vehicle is traveling in a straight line; , , These are the speeds of the vehicle itself and the vehicle in front, respectively. To warn of the vehicle's maximum deceleration, The deceleration of the vehicle in front; , , These are the braking system response time and the driver's reaction delay time, respectively. It is the minimum distance between the vehicle and the vehicle in front when they are stationary relative to each other.

3. The reliability assessment method for a forward collision warning system according to claim 2, characterized in that, Step S4 includes the following sub-steps: S401. Establish the limit state equation of the forward collision warning system, as follows: ； S402. Using the first-order Taylor expansion to process the equations, obtain the linearized function of the forward collision warning system. , in, Let the state space of the random variable be the state space of the random variable. There are a total of 7 basic variables. This represents the number of basic variables in the random variable space. For the random variable in the state space, the first... One variable, Let i be the specific value of the i-th variable on the limit state surface. A point in the state space that is on the limit state surface. .

4. The reliability assessment method for a forward collision warning system according to claim 3, characterized in that, Step S5 includes the following sub-steps: S501, Judgment If each random variable follows a normal distribution, then calculate the mean of each variable. and variance Then proceed to step S503; otherwise, proceed to step S502. S502, will Nonnormal random variables in Perform equivalent normalization to Calculate the standard deviation of the equivalent normal distribution using the following formula. and mean ,use and As a variable Standard deviation and mean : ， ， in and Random variables The probability density function and cumulative distribution function, The function is the standard normal distribution function; S503, Set initial value , ,in for The mean; S504. Calculate the function according to the following formula. mean and standard deviation : ， ， in, For variables and The correlation coefficient between them; S505, Calculating Variables Sensitivity coefficient As shown in the following formula: ； S506 calculates the reliability index of a vehicle's forward collision warning system. As shown in the following formula: ， in, and and are the function functions respectively. The mean and standard deviation; S507 judgment Is the value less than ,in , for If the modulo function is true, then the output in step S506 is... If the value is used as a reliability indicator, then proceed to step S509; otherwise, proceed to step S508. S508, Update ,make and return to step S504; S509. Calculate the failure probability of the forward collision warning system. As shown in the following formula: 。 5. The reliability evaluation method for a forward collision warning system according to claim 4, characterized in that, Step S6 specifically involves: comparing the obtained failure probability with a preset system failure probability threshold. If a comparison is made, If so, the forward collision warning system is unreliable; If the forward collision warning system operates reliably, then the forward collision warning system will function reliably. The forward collision warning system is then in a critical state of failure.

6. The reliability assessment method for a forward collision warning system according to claim 1, characterized in that, In step S1, the warning system operation data includes the system warning time, the actual distance between the vehicle and the vehicle in front at the time of the warning, the vehicle driving data includes the vehicle's driving speed and maximum deceleration, the braking system delay time of different vehicles and the reaction time of different drivers, and the road condition data includes the speed data of the vehicle in front.

7. The reliability assessment method for a forward collision warning system according to claim 1, characterized in that, Step S2 specifically involves: based on the early warning system operation data collected in step S1, the first... The actual distance between the two vehicles when the second warning signal was issued Construct a dataset of actual vehicle distances for early warning purposes. Among them, if a collision occurs between the vehicle and the vehicle in front but the forward collision warning system does not issue a warning signal, it is recorded as follows: Vehicle distance data at the time of the warning is used as a continuous variable in the probability distribution. The sample values ​​are used to infer the actual distribution of the variable using nonparametric tests, and this variable is then used as the supply value for the forward collision warning system. .

8. The reliability evaluation method for a forward collision warning system according to claim 5, characterized in that, The system failure probability threshold =85%.

Images