Collision waveform prediction method, device, medium and equipment based on safety collaboration

Abstract

The collision waveform prediction method, device, medium and equipment based on safety collaboration provided by the present application build an initial collision waveform prediction model based on the test vehicle collision information when the emergency braking system of the test vehicle is involved; and optimize the model parameters of the initial collision waveform prediction model to obtain the final collision waveform prediction model; based on the target vehicle collision information of the target vehicle, obtain the predicted collision waveform of the target vehicle; that is, use the test vehicle collision information of the test vehicle when the emergency braking system is involved to build a collision waveform prediction model, and optimize the model parameters to obtain the final collision waveform prediction model, and use the final collision waveform prediction model to predict the predicted collision waveform of the target vehicle, so as to realize the collision waveform prediction of the vehicle equipped with the emergency braking system, so as to provide data support for the vehicle to optimize the emergency braking system, so as to reduce the probability of collision accidents and reduce the severity of collision accidents.

Description

Technical Field

[0001] The present application relates to the technical field of collision prediction in safety collaboration scenarios, and specifically to a collision waveform prediction method, device, medium and equipment based on safety collaboration. Background Art

[0002] With the continuous increase in the number of cars, the safety performance of cars has always been the focus of attention of all OEMs and consumers. Among them, vehicle active safety has gradually developed and grown, and the development of vehicle automatic emergency braking system (AEB) in my country has been very mature, and most vehicles are equipped with AEB system. At present, the safety test of vehicles mainly tests active safety and passive safety separately, while there are fewer tests for active and passive fusion, which is inconsistent with the technical characteristics of active and passive fusion presented in traffic accidents, resulting in limited accuracy of vehicle safety testing. Therefore, there is an urgent need for a method that can realize active and passive fusion testing and predict the severity of vehicle collisions. Summary of the invention

[0003] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide a collision waveform prediction method, device, medium and equipment based on safety collaboration.

[0004] According to one aspect of the present application, a collision waveform prediction method based on safety collaboration is provided, including: obtaining collision information of a test vehicle when an emergency braking system is engaged; wherein the collision information of the test vehicle includes a first distance between two vehicles of the test vehicle during emergency braking, a first speed of the two vehicles of the test vehicle during emergency braking, a first collision position of the test vehicle, and a first collision angle of the test vehicle; based on the collision information of the test vehicle, building an initial collision waveform prediction model; based on the collision information of the test vehicle and the corresponding real collision waveform, optimizing the model parameters of the initial collision waveform prediction model to obtain a final collision waveform prediction model; inputting the collision information of the target vehicle when the emergency braking system is engaged into the final collision waveform prediction model to obtain a predicted collision waveform of the target vehicle; wherein the collision information of the target vehicle includes a second distance between two vehicles of the target vehicle during emergency braking, a second speed of the two vehicles of the target vehicle during emergency braking, a second collision position of the target vehicle, and a second collision angle of the target vehicle.

[0005] In one embodiment, after obtaining the test vehicle collision information when the emergency braking system of the test vehicle is engaged, the collision waveform prediction method based on safety collaboration also includes: normalizing the test vehicle collision information to obtain normalized collision information; building an initial collision waveform prediction model based on the test vehicle collision information includes: building an initial collision waveform prediction model based on the normalized collision information; optimizing the model parameters of the initial collision waveform prediction model based on the test vehicle collision information and the corresponding real collision waveform to obtain the final collision waveform prediction model includes: optimizing the model parameters of the initial collision waveform prediction model based on the normalized collision information and the corresponding real collision waveform to obtain the final collision waveform prediction model.

[0006] In one embodiment, the normalizing the test vehicle collision information to obtain normalized collision information includes: normalizing the first spacing based on the average value and standard deviation of all first spacings to obtain the normalized first spacing; normalizing the first vehicle speed based on the average value and standard deviation of all first vehicle speeds to obtain the normalized first vehicle speed.

[0007] In one embodiment, the collision information of the test vehicle is normalized to obtain normalized collision information including: the normalized first collision position corresponding to the longitudinal perpendicular bisector of the test vehicle is 0, the normalized first collision position corresponding to the outermost side of the driver's side is 1, the normalized first collision position corresponding to the outermost side of the co-pilot side is -1, and the normalized first collision position between the outermost side of the driver's side and the outermost side of the co-pilot side is calculated by linear interpolation.

[0008] In one embodiment, the collision information of the test vehicle is normalized to obtain normalized collision information including: a normalized first collision angle corresponding to a 0 degree angle between the longitudinal medians of the two vehicles, a normalized first collision angle corresponding to a 90 degree angle between the longitudinal medians of the two vehicles, a normalized first collision angle corresponding to a -90 degree angle between the longitudinal medians of the two vehicles, and a normalized first collision angle corresponding to a -1 angle between the longitudinal medians of the two vehicles between -90 degrees and 90 degrees, and a normalized first collision angle corresponding to a 90 degree angle between the longitudinal medians of the two vehicles obtained by linear interpolation; wherein, the normalized first collision angle corresponding to a first collision position of the test vehicle on the driver's side is positive.

[0009] In one embodiment, the optimizing the model parameters of the initial collision waveform prediction model based on the test vehicle collision information and the corresponding real collision waveform to obtain the final collision waveform prediction model includes: inputting the test vehicle collision information into the initial collision waveform prediction model to obtain an estimated waveform; calculating the difference between the estimated waveform and the real collision waveform; if the difference is greater than a preset difference, optimizing the model parameters of the initial collision waveform prediction model until the difference between the estimated waveform calculated based on the optimized model parameters and the real collision waveform is less than or equal to the preset difference.

[0010] In one embodiment, the optimization of the model parameters of the initial collision waveform prediction model includes: calculating the iteration direction of the model parameters based on the difference between the estimated waveform and the actual collision waveform; dividing the collision process of the test vehicle into multiple time periods based on the acceleration of the test vehicle; setting the iteration step of the model parameters for different time periods; and optimizing the model parameters based on the iteration direction and iteration step of the model parameters.

[0011] According to another aspect of the present application, a collision waveform prediction device based on safety collaboration is provided, including: a test data acquisition module, used to obtain the test vehicle collision information of the test vehicle when the emergency braking system is involved; wherein the test vehicle collision information includes the distance between the test vehicles during emergency braking, the speed of the test vehicle during emergency braking, the collision position of the test vehicle, and the collision angle of the test vehicle; a prediction model building module, used to build an initial collision waveform prediction model based on the test vehicle collision information; a model parameter optimization module, used to optimize the model parameters of the initial collision waveform prediction model based on the test vehicle collision information and the corresponding real collision waveform to obtain a final collision waveform prediction model; a collision waveform prediction module, used to input the target vehicle collision information of the target vehicle when the emergency braking system is involved into the final collision waveform prediction model to obtain the predicted collision waveform of the target vehicle; wherein the target vehicle collision information includes the distance between the target vehicles during emergency braking, the speed of the target vehicle during emergency braking, the collision position of the target vehicle, and the collision angle of the target vehicle.

[0012] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and the computer program is used to execute any of the above methods.

[0013] According to another aspect of the present application, an electronic device is provided, comprising: a processor; a memory for storing instructions executable by the processor; and the processor is used to execute any of the above-described methods.

[0014] The collision waveform prediction method, device, medium and equipment based on safety collaboration provided in the present application obtain the collision information of the test vehicle when the emergency braking system of the test vehicle is involved; wherein the collision information of the test vehicle includes the first distance between the two vehicles during emergency braking, the first speed of the two vehicles during emergency braking, the first collision position of the test vehicle, and the first collision angle of the test vehicle; based on the collision information of the test vehicle, an initial collision waveform prediction model is built; based on the collision information of the test vehicle and the corresponding real collision waveform, the model parameters of the initial collision waveform prediction model are optimized to obtain the final collision waveform prediction model; the collision information of the target vehicle when the emergency braking system of the target vehicle is involved is input into the final collision waveform prediction model , obtaining a predicted collision waveform of the target vehicle; wherein the collision information of the target vehicle includes a second distance between the two vehicles when the target vehicle is in emergency braking, a second speed of the two vehicles when the target vehicle is in emergency braking, a second collision position of the target vehicle, and a second collision angle of the target vehicle; that is, using the collision information of the test vehicle when the emergency braking system of the test vehicle is involved to build a collision waveform prediction model, and optimizing the model parameters to obtain a final collision waveform prediction model, and using the final collision waveform prediction model to predict the predicted collision waveform of the target vehicle, so as to realize the collision waveform prediction of the vehicle equipped with the emergency braking system, so as to provide data support for the vehicle to optimize the emergency braking system, so as to reduce the probability of collision accidents and reduce the severity of collision accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 It is a flowchart of a collision waveform prediction method based on safety collaboration provided by an exemplary embodiment of the present application.

[0017] Figure 2 It is a schematic diagram of the principle structure of the first collision position normalization processing provided by an exemplary embodiment of the present application.

[0018] Figure 3 It is a schematic diagram of the principle structure of the first collision angle normalization processing provided by an exemplary embodiment of the present application.

[0019] Figure 4 It is a schematic diagram of the principle structure of calculating the iteration step length in model parameter optimization provided by an exemplary embodiment of the present application.

[0020] Figure 5 It is a structural diagram of a collision waveform prediction device based on safety collaboration provided by an exemplary embodiment of the present application.

[0021] Figure 6 It is a structural diagram of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] Figure 1 FIG. 1 is a flow chart of a collision waveform prediction method based on safety collaboration provided by an exemplary embodiment of the present application. Figure 1 As shown, the collision waveform prediction method based on safety collaboration includes the following steps:

[0024] Step 110: Obtain the test vehicle collision information when the emergency braking system of the test vehicle is engaged.

[0025] The test vehicle collision information includes the first distance between the two vehicles during emergency braking, the first speed of the two vehicles during emergency braking, the first collision position of the test vehicle, and the first collision angle of the test vehicle. The present application can obtain relevant information about the collision of the test vehicle under the action of the emergency braking system through simulation or direct real vehicle collision experiments to obtain the collision situation of the test vehicle under the action of the emergency braking system.

[0026] Step 120: Building an initial collision waveform prediction model based on the test vehicle collision information.

[0027] After obtaining the test vehicle collision information, the present application removes the damaged data and obviously unreasonable data, and classifies the data according to small cars, SUVs, MPVs and other models. Based on each type of data, an initial collision waveform prediction model suitable for a specific model is built, wherein the model parameters of the initial collision waveform prediction model can be default values ​​or empirical values ​​determined based on experience. Specifically, the initial collision waveform prediction model can be:

[0028] ;

[0029] in, is the output predicted collision waveform, which is a value that changes over time. is the input column vector, , is the first distance, To test the vehicle's emergency braking speed, is the speed of the hit vehicle, is the first collision position, is the first collision angle, is the vector of corresponding coefficients, , arrive For the corresponding input arrive The coefficient of is a constant term, arrive are all time-related column vectors, whose elements correspond to the coefficient values ​​of each time point. Specifically, The expression of can also be written as a matrix: ,in, n Equal to the number of collision waveform sampling points, when the sampling time and sampling frequency are determined, n The value of can be determined.

[0030] Step 130: Based on the test vehicle collision information and the corresponding real collision waveform, the model parameters of the initial collision waveform prediction model are optimized to obtain the final collision waveform prediction model.

[0031] This application uses the test vehicle collision information and the corresponding real collision waveform as training samples, optimizes and learns the initial collision waveform prediction model to optimize its model parameters, and thus obtains the final collision waveform prediction model.

[0032] Step 140: Inputting the target vehicle collision information when the emergency braking system of the target vehicle is engaged into the final collision waveform prediction model to obtain the predicted collision waveform of the target vehicle.

[0033] The target vehicle collision information includes the second distance between the target vehicle and the target vehicle during emergency braking, the second speed of the target vehicle and the target vehicle during emergency braking, the second collision position of the target vehicle, and the second collision angle of the target vehicle. After optimizing the final collision waveform prediction model, the present application collects the target vehicle collision information under the action of the emergency braking system and inputs the target vehicle collision information into the final collision waveform prediction model to generate a predicted collision waveform of the target vehicle, so that the function of the emergency braking system can be adjusted according to the predicted collision waveform to avoid or reduce the collision as much as possible.

[0034] The collision waveform prediction method based on safety collaboration provided by the present application obtains the collision information of the test vehicle when the emergency braking system of the test vehicle is involved; wherein the collision information of the test vehicle includes the first distance between the two vehicles during emergency braking, the first speed of the two vehicles during emergency braking, the first collision position of the test vehicle, and the first collision angle of the test vehicle; based on the collision information of the test vehicle, an initial collision waveform prediction model is constructed; based on the collision information of the test vehicle and the corresponding real collision waveform, the model parameters of the initial collision waveform prediction model are optimized to obtain a final collision waveform prediction model; the collision information of the target vehicle when the emergency braking system of the target vehicle is involved is input into the final collision waveform prediction model to obtain the target vehicle. The predicted collision waveform of the vehicle; wherein the target vehicle collision information includes the second distance between the two vehicles of the target vehicle during emergency braking, the second vehicle speed of the two vehicles of the target vehicle during emergency braking, the second collision position of the target vehicle, and the second collision angle of the target vehicle; that is, a collision waveform prediction model is built using the collision information of the test vehicle when the emergency braking system of the test vehicle is involved, and the model parameters are optimized to obtain a final collision waveform prediction model, and the predicted collision waveform of the target vehicle is predicted using the final collision waveform prediction model to achieve collision waveform prediction of vehicles equipped with emergency braking systems, thereby providing data support for optimizing the emergency braking system of the vehicle to reduce the probability of collision accidents and reduce the severity of collision accidents.

[0035] In one embodiment, after step 110, the above-mentioned collision waveform prediction method based on safety collaboration may further include: normalizing the collision information of the test vehicle to obtain normalized collision information; correspondingly, the specific implementation method of the above-mentioned step 120 may be: building an initial collision waveform prediction model based on the normalized collision information; correspondingly, the specific implementation method of the above-mentioned step 130 may be: optimizing the model parameters of the initial collision waveform prediction model based on the normalized collision information and the corresponding real collision waveform to obtain the final collision waveform prediction model.

[0036] In order to facilitate subsequent calculations and unify the magnitude, after obtaining the test vehicle collision information, the present application normalizes the test vehicle collision information to obtain normalized collision information, and builds an initial collision waveform prediction model based on the normalized collision information. At the same time, based on the normalized collision information and the corresponding real collision waveform, the model parameters of the initial collision waveform prediction model are optimized to obtain the final collision waveform prediction model.

[0037] In one embodiment, the specific implementation method of normalizing the test vehicle collision information may be: based on the average value and standard deviation of all first intervals, the first intervals are normalized to obtain the normalized first intervals; based on the average value and standard deviation of all first vehicle speeds, the first vehicle speed is normalized to obtain the normalized first vehicle speed.

[0038] The first distance between the two vehicles and the first speed of the two vehicles during emergency braking are normalized using the following formulas:

[0039] The first spacing after normalization = (the first spacing mean of all first intervals) / standard deviation of all first intervals,

[0040] Normalized first vehicle speed = (first vehicle speed The mean of all first vehicle speeds) / the standard deviation of all first vehicle speeds;

[0041] The ranges of the normalized first spacing and the normalized first vehicle speed are both between [-1,1].

[0042] In one embodiment, a specific implementation method for normalizing the test vehicle collision information may be: the normalized first collision position corresponding to the longitudinal mid-perpendicular line of the test vehicle is 0, the normalized first collision position corresponding to the outermost side of the driver's side is 1, the normalized first collision position corresponding to the outermost side of the co-pilot side is -1, and the normalized first collision position between the outermost side of the driver's side and the outermost side of the co-pilot side is calculated by linear interpolation.

[0043] For the normalization of the first collision position, Figure 2 As shown, in the present application, the normalized first collision position corresponding to the longitudinal median perpendicular line of the test vehicle is 0, the normalized first collision position corresponding to the outermost side of the driver's side is 1, and the normalized first collision position corresponding to the outermost side of the co-pilot side is -1. The normalized first collision position between the outermost side of the driver's side and the outermost side of the co-pilot side is calculated by linear interpolation according to the ratio of the collision distance to the vehicle width, thereby normalizing the first collision position to a value between [-1,1].

[0044] In one embodiment, a specific implementation method for normalizing the test vehicle collision information may be: when the angle between the longitudinal medians of the two vehicles is 0 degrees, the corresponding normalized first collision angle is 0; when the angle between the longitudinal medians of the two vehicles is 90 degrees, the corresponding normalized first collision angle is 1; when the angle between the longitudinal medians of the two vehicles is -90 degrees, the corresponding normalized first collision angle is -1; when the angle between the longitudinal medians of the two vehicles is between -90 degrees and 90 degrees, the corresponding normalized first collision angle is calculated by linear interpolation; wherein, when the first collision position of the test vehicle is located on the driver's side, the corresponding normalized first collision angle is positive.

[0045] For the normalization of the first collision angle, Figure 3 As shown, the present application takes the normalized first collision angle corresponding to when the angle between the longitudinal medians of the two vehicles is 0 degree as 0, the normalized first collision angle corresponding to when the angle between the longitudinal medians of the two vehicles is 90 degrees as 1, and the normalized first collision angle corresponding to when the angle between the longitudinal medians of the two vehicles is -90 degrees as -1. The normalized first collision angle corresponding to when the angle between the longitudinal medians of the two vehicles is between -90 degrees and 90 degrees is calculated by linear interpolation, and the first collision angle close to the driver's direction is defined as positive, and the first collision angle close to the co-driver's direction is defined as negative, thereby normalizing the first collision angle to a value between [-1,1].

[0046] In one embodiment, the specific implementation method of the above step 130 can be: inputting the test vehicle collision information into the initial collision waveform prediction model to obtain an estimated waveform; calculating the difference between the estimated waveform and the actual collision waveform; if the difference is greater than a preset difference, optimizing the model parameters of the initial collision waveform prediction model until the difference between the estimated waveform calculated based on the optimized model parameters and the actual collision waveform is less than or equal to the preset difference.

[0047] In order to evaluate whether the initial collision waveform prediction model is accurate, it is necessary to evaluate its accuracy and evaluate the difference between it and the true value. This application evaluates the fitting degree of the initial collision waveform prediction model by building a fitting function. Specifically, the fitting function is:

[0048] ;

[0049] in, m is the number of data in the training set, i For its corner mark, n is the number of sampling points for each data in the training set, j For its corner mark, X i For the training set iThe input value of the data, For this input, the predicted value is y ij For the training set i The data j The true value of the sampling points, Characterizes the difference between the predicted value and the true value. The smaller its value is, the higher the fitting degree of the initial collision waveform prediction model is. In order to make the initial collision waveform prediction model more accurate, just let The minimum is sufficient, then we only need to The smallest value value.

[0050] In one embodiment, the specific implementation method of the above step 130 can be: based on the difference between the estimated waveform and the actual collision waveform, calculating the iteration direction of the model parameters; dividing the collision process of the test vehicle into multiple time periods based on the acceleration of the test vehicle; setting the iteration step of the model parameters for different time periods; optimizing the model parameters based on the iteration direction and iteration step of the model parameters.

[0051] Fitting function If it is a convex function in form, it has a minimum value, and when iterating downward along its tangent direction, it is The value decreases in the direction of The tangent line of is its derivative. Therefore, this application constructs an iterative equation for iterative training, so that along Iterate downward in the tangent direction, where The iteration equation is:

[0052] ;

[0053] in, p =0,1,2,3,4,5, correspond ~ , For the training set i In the data , r =1,2,3,4,5, correspond ~ , is the iteration step length.

[0054] The constructed iterative equation is Iterate downward in the tangent direction. The initial value of is 0. Not too big nor too small. If it is too large, it may cause the iteration to be too fast and miss the The minimum value of If the value is too small, it may lead to slow iteration speed and long iteration time. It is also a time-related vector. For collision waveforms, the sampling frequency is high, resulting in many data points in the entire collision cycle. If the iteration function uses a single iteration step, when the iteration step is large, some features with high importance may be missed. When the iteration step is small, information with low importance may be repeatedly iterated, increasing the computational cost and storage cost. At the same time, collision waveforms generally have the characteristics of small initial values, large intermediate values, and then returning to small values ​​(such as Figure 4 As shown in the figure, some smaller values ​​at the beginning and later stages have a weaker effect on vehicle performance analysis than the middle part. If a single iteration step is used, the features of the important middle part may be ignored, while the values ​​of the weaker parts at both ends are iterated multiple times. Therefore, this application adopts a segmented iteration method for setting. Specifically, Figure 4 As shown, for acceleration (absolute value) less than phase (i.e. from the beginning to and from The iteration step size of the data points within the time period from 1 to the end is set to ; For acceleration (absolute value) at arrive The stage between arrive as well as arrive The iteration step size of the data points within the time period is set to ; And so on, multiple stages can be set. Among them, is the maximum acceleration (absolute value), > , For the maximum value and The acceleration value is set to a longer iteration step for the smaller value and a smaller iteration step for the larger value. and Can be set according to waveform characteristics, special, when =100, it is a single step iteration.

[0055] The data points at different stages are iterated according to the corresponding iteration step, and the error of each iteration result is evaluated. This application can use the root mean square error to evaluate the iteration result. The specific formula is as follows:

[0056] ;

[0057] in, yj For the j The true value of the sampling points, RMSE is the root mean square error. By setting the acceptable error limit, if this iteration The value of RMSE If the value of meets the error limit requirement, the iteration ends; if it does not meet the error limit requirement, the iteration calculation continues until The value of RMSE Satisfy the requirements, the iteration ends, and the The value of is the value of the final collision waveform prediction model.

[0058] Figure 5 FIG. 1 is a schematic diagram of a collision waveform prediction device based on safety collaboration provided by an exemplary embodiment of the present application. Figure 5 As shown, the collision waveform prediction device 50 based on safety collaboration includes: a test data acquisition module 51, which is used to obtain the test vehicle collision information when the test vehicle is involved in the emergency braking system; wherein the test vehicle collision information includes the distance between the test vehicles during emergency braking, the speed of the test vehicle during emergency braking, the collision position of the test vehicle, and the collision angle of the test vehicle; a prediction model building module 52, which is used to build an initial collision waveform prediction model based on the test vehicle collision information; a model parameter optimization module 53, which is used to optimize the model parameters of the initial collision waveform prediction model based on the test vehicle collision information and the corresponding real collision waveform to obtain a final collision waveform prediction model; a collision waveform prediction module 54, which is used to input the target vehicle collision information when the emergency braking system is involved into the final collision waveform prediction model to obtain the predicted collision waveform of the target vehicle; wherein the target vehicle collision information includes the distance between the target vehicles during emergency braking, the speed of the target vehicle during emergency braking, the collision position of the target vehicle, and the collision angle of the target vehicle.

[0059] The collision waveform prediction device based on safety collaboration provided by the present application obtains the collision information of the test vehicle when the emergency braking system of the test vehicle is involved through the test data acquisition module 51; wherein the collision information of the test vehicle includes the first distance between the two vehicles of the test vehicle during emergency braking, the first speed of the two vehicles of the test vehicle during emergency braking, the first collision position of the test vehicle, and the first collision angle of the test vehicle; the prediction model building module 52 builds an initial collision waveform prediction model based on the collision information of the test vehicle; the model parameter optimization module 53 optimizes the model parameters of the initial collision waveform prediction model based on the collision information of the test vehicle and the corresponding real collision waveform to obtain the final collision waveform prediction model; the collision waveform prediction module 54 calculates the target vehicle collision waveform when the emergency braking system is involved. The collision information is input into the final collision waveform prediction model to obtain the predicted collision waveform of the target vehicle; wherein the collision information of the target vehicle includes the second distance between the two vehicles when the target vehicle is in emergency braking, the second speed of the two vehicles when the target vehicle is in emergency braking, the second collision position of the target vehicle, and the second collision angle of the target vehicle; that is, the collision waveform prediction model is built using the collision information of the test vehicle when the emergency braking system of the test vehicle is involved, and the model parameters are optimized to obtain the final collision waveform prediction model, and the predicted collision waveform of the target vehicle is predicted using the final collision waveform prediction model to achieve collision waveform prediction of vehicles equipped with emergency braking systems, thereby providing data support for optimizing the emergency braking system of the vehicle to reduce the probability of collision accidents and reduce the severity of collision accidents.

[0060] In one embodiment, the above-mentioned collision waveform prediction device 50 based on safety collaboration can be further configured as: normalizing the test vehicle collision information to obtain normalized collision information; correspondingly, the above-mentioned prediction model building module 52 can be further configured as: building an initial collision waveform prediction model based on the normalized collision information; correspondingly, the above-mentioned model parameter optimization module 53 can be further configured as: optimizing the model parameters of the initial collision waveform prediction model based on the normalized collision information and the corresponding real collision waveform to obtain the final collision waveform prediction model.

[0061] In one embodiment, the above-mentioned collision waveform prediction device 50 based on safety collaboration can be further configured as: based on the average value and standard deviation of all first intervals, the first intervals are normalized to obtain the normalized first intervals; based on the average value and standard deviation of all first vehicle speeds, the first vehicle speed is normalized to obtain the normalized first vehicle speed.

[0062] In one embodiment, the above-mentioned collision waveform prediction device 50 based on safety collaboration can be further configured as: the normalized first collision position corresponding to the longitudinal mid-perpendicular line of the test vehicle is 0, the normalized first collision position corresponding to the outermost side of the driver's side is 1, and the normalized first collision position corresponding to the outermost side of the co-pilot side is -1, and the normalized first collision position between the outermost side of the driver's side and the outermost side of the co-pilot side is calculated by linear interpolation.

[0063] In one embodiment, the above-mentioned collision waveform prediction device 50 based on safety collaboration can be further configured as follows: when the angle between the longitudinal medians of the two vehicles is 0 degrees, the corresponding normalized first collision angle is 0, when the angle between the longitudinal medians of the two vehicles is 90 degrees, the corresponding normalized first collision angle is 1, when the angle between the longitudinal medians of the two vehicles is -90 degrees, the corresponding normalized first collision angle is -1, and when the angle between the longitudinal medians of the two vehicles is between -90 degrees and 90 degrees, the corresponding normalized first collision angle is calculated by linear interpolation; wherein, when the first collision position of the test vehicle is located on the driver's side, the corresponding normalized first collision angle is positive.

[0064] In one embodiment, the above-mentioned model parameter optimization module 53 can be further configured as follows: inputting the test vehicle collision information into the initial collision waveform prediction model to obtain an estimated waveform; calculating the difference between the estimated waveform and the actual collision waveform; if the difference is greater than a preset difference, optimizing the model parameters of the initial collision waveform prediction model until the difference between the estimated waveform calculated based on the optimized model parameters and the actual collision waveform is less than or equal to the preset difference.

[0065] In one embodiment, the above-mentioned model parameter optimization module 53 can be further configured as: calculating the iteration direction of the model parameters based on the difference between the estimated waveform and the actual collision waveform; dividing the collision process of the test vehicle into multiple time periods based on the acceleration of the test vehicle; setting the iteration step of the model parameters for different time periods; optimizing the model parameters based on the iteration direction and iteration step of the model parameters.

[0066] Below, reference Figure 6 The electronic device according to the embodiment of the present application is described. The electronic device may be any one or both of the first device and the second device, or a stand-alone device independent of them, and the stand-alone device may communicate with the first device and the second device to receive the collected input signal from them.

[0067] Figure 6 A block diagram of an electronic device according to an embodiment of the present application is illustrated.

[0068] like Figure 6As shown, the electronic device 10 includes one or more processors 11 and a memory 12 .

[0069] The processor 11 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0070] The memory 12 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may run the program instructions to implement the methods of the various embodiments of the present application described above and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0071] In one example, the electronic device 10 may further include: an input device 13 and an output device 14 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0072] When the electronic device is a stand-alone device, the input device 13 may be a communication network connector, which is used to receive the collected input signals from the first device and the second device.

[0073] In addition, the input device 13 may also include, for example, a keyboard, a mouse, and the like.

[0074] The output device 14 can output various information to the outside, including the determined distance information, direction information, etc. The output device 14 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0075] Of course, to simplify, Figure 6 Only some of the components related to the present application in the electronic device 10 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 10 may also include any other appropriate components.

[0076] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method according to various embodiments of the present application described in the above-mentioned "Exemplary Method" section of this specification.

[0077] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0078] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the method according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0079] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0080] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0081] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0082] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0083] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0084] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A collision waveform prediction method based on safety collaboration, characterized in that: include: Acquire the test vehicle collision information when the test vehicle's emergency braking system is engaged; wherein the test vehicle collision information includes a first distance between the two vehicles of the test vehicle when the emergency braking is performed, a first vehicle speed of the two vehicles of the test vehicle when the emergency braking is performed, a first collision position of the test vehicle, and a first collision angle of the test vehicle; Based on the test vehicle collision information, an initial collision waveform prediction model is constructed; wherein the initial collision waveform prediction model is: ； in, is the output predicted collision waveform, which is a value that changes over time. is the input column vector, , is the first distance, is the speed of the test vehicle during emergency braking, is the speed of the hit vehicle, is the first collision position, is the first collision angle, is a vector of corresponding coefficients, , arrive For the corresponding input arrive The coefficient of is a constant term, arrive are all time-related column vectors, whose elements correspond to the coefficient values ​​at each time point. The expression is: ,in, n Equal to the number of collision waveform sampling points; Based on the test vehicle collision information and the corresponding real collision waveform, optimizing the model parameters of the initial collision waveform prediction model to obtain a final collision waveform prediction model; Inputting the target vehicle collision information when the emergency braking system of the target vehicle is engaged into the final collision waveform prediction model to obtain the predicted collision waveform of the target vehicle; wherein the target vehicle collision information includes the second distance between the two vehicles of the target vehicle when the emergency braking is performed, the second vehicle speed of the two vehicles of the target vehicle when the emergency braking is performed, the second collision position of the target vehicle, and the second collision angle of the target vehicle; After obtaining the test vehicle collision information when the emergency braking system of the test vehicle is engaged, the collision waveform prediction method based on safety collaboration further includes: Normalizing the collision information of the test vehicle to obtain normalized collision information; The constructing of an initial collision waveform prediction model based on the test vehicle collision information includes: Based on the normalized collision information, an initial collision waveform prediction model is constructed; The step of optimizing the model parameters of the initial collision waveform prediction model based on the test vehicle collision information and the corresponding real collision waveform to obtain the final collision waveform prediction model comprises: Based on the normalized collision information and the corresponding real collision waveform, optimizing the model parameters of the initial collision waveform prediction model to obtain a final collision waveform prediction model; The normalizing the collision information of the test vehicle to obtain the normalized collision information includes: The normalized first collision position corresponding to the longitudinal mid-perpendicular line of the test vehicle is 0, the normalized first collision position corresponding to the outermost side of the driver's side is 1, and the normalized first collision position corresponding to the outermost side of the co-pilot side is -1. The normalized first collision position between the outermost side of the driver's side and the outermost side of the co-pilot side is calculated by linear interpolation; The normalizing the collision information of the test vehicle to obtain the normalized collision information includes: The normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is 0 degrees is 0, the normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is 90 degrees is 1, the normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is -90 degrees is -1, and the normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is between -90 degrees and 90 degrees is calculated by linear interpolation; wherein, the normalized first collision angle corresponding to the first collision position of the test vehicle located on the driver's side is positive; The step of optimizing the model parameters of the initial collision waveform prediction model based on the test vehicle collision information and the corresponding real collision waveform to obtain the final collision waveform prediction model comprises: Inputting the test vehicle collision information into the initial collision waveform prediction model to obtain an estimated waveform; calculating a difference between the estimated waveform and the actual collision waveform; If the difference is greater than a preset difference, optimizing the model parameters of the initial collision waveform prediction model until the difference between the estimated waveform calculated based on the optimized model parameters and the actual collision waveform is less than or equal to the preset difference; The model parameters of the optimization of the initial collision waveform prediction model include: Calculating the iteration direction of the model parameters based on the difference between the estimated waveform and the actual collision waveform; dividing the collision process of the test vehicle into a plurality of time periods based on the acceleration of the test vehicle; The iteration step length of the model parameters is set for different time periods; wherein, for the absolute value of the acceleration less than The iteration step size of the data points in the stage is set to , for the absolute value of acceleration arrive The iteration step size of the data points within the phase between is set to ,in, is the absolute value of the maximum acceleration, , , The maximum value of acceleration and The acceleration value of The model parameters are optimized based on the iteration direction and iteration step size of the model parameters.

2. The collision waveform prediction method based on safety collaboration according to claim 1 is characterized in that: The normalizing the collision information of the test vehicle to obtain the normalized collision information includes: Based on the average value and standard deviation of all first intervals, the first intervals are normalized to obtain normalized first intervals; Based on an average value and a standard deviation of all first vehicle speeds, the first vehicle speeds are normalized to obtain a normalized first vehicle speed.

3. A collision waveform prediction device based on safety collaboration, characterized in that: include: A test data acquisition module, used to acquire the test vehicle collision information when the test vehicle emergency braking system is engaged; wherein the test vehicle collision information includes the distance between the test vehicles during emergency braking, the speed of the test vehicle during emergency braking, the collision position of the test vehicle, and the collision angle of the test vehicle; The prediction model building module is used to build an initial collision waveform prediction model based on the test vehicle collision information; wherein the initial collision waveform prediction model is: ； in, is the output predicted collision waveform, which is a value that changes over time. is the input column vector, , is the first distance, is the speed of the test vehicle during emergency braking, is the speed of the hit vehicle, is the first collision position, is the first collision angle, is a vector of corresponding coefficients, , arrive For the corresponding input arrive The coefficient of is a constant term, arrive are all time-related column vectors, whose elements correspond to the coefficient values ​​at each time point. The expression is: ,in, n Equal to the number of collision waveform sampling points; A model parameter optimization module, used to optimize the model parameters of the initial collision waveform prediction model based on the test vehicle collision information and the corresponding real collision waveform to obtain a final collision waveform prediction model; A collision waveform prediction module, used for inputting the target vehicle collision information when the emergency braking system of the target vehicle is engaged into the final collision waveform prediction model to obtain the predicted collision waveform of the target vehicle; wherein the target vehicle collision information includes the distance between the target vehicles during emergency braking, the speed of the target vehicle during emergency braking, the collision position of the target vehicle, and the collision angle of the target vehicle; The collision waveform prediction device based on safety collaboration is further configured as follows: Normalizing the collision information of the test vehicle to obtain normalized collision information; The collision waveform prediction device based on safety collaboration is further configured as follows: Based on the normalized collision information, an initial collision waveform prediction model is constructed; The model parameter optimization module is further configured as follows: Based on the normalized collision information and the corresponding real collision waveform, optimizing the model parameters of the initial collision waveform prediction model to obtain a final collision waveform prediction model; The collision waveform prediction device based on safety collaboration is further configured as follows: The normalized first collision position corresponding to the longitudinal mid-perpendicular line of the test vehicle is 0, the normalized first collision position corresponding to the outermost side of the driver's side is 1, and the normalized first collision position corresponding to the outermost side of the co-pilot side is -1. The normalized first collision position between the outermost side of the driver's side and the outermost side of the co-pilot side is calculated by linear interpolation; The collision waveform prediction device based on safety collaboration is further configured as follows: The normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is 0 degrees is 0, the normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is 90 degrees is 1, the normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is -90 degrees is -1, and the normalized first collision angle corresponding to the angle between the longitudinal medians of the two vehicles is between -90 degrees and 90 degrees is calculated by linear interpolation; wherein, the normalized first collision angle corresponding to the first collision position of the test vehicle located on the driver's side is positive; The model parameter optimization module is further configured as follows: Inputting the test vehicle collision information into the initial collision waveform prediction model to obtain an estimated waveform; calculating a difference between the estimated waveform and the actual collision waveform; If the difference is greater than a preset difference, optimizing the model parameters of the initial collision waveform prediction model until the difference between the estimated waveform calculated based on the optimized model parameters and the actual collision waveform is less than or equal to the preset difference; The model parameter optimization module is further configured as follows: Calculating the iteration direction of the model parameters based on the difference between the estimated waveform and the actual collision waveform; dividing the collision process of the test vehicle into a plurality of time periods based on the acceleration of the test vehicle; The iteration step length of the model parameters is set for different time periods; wherein, for the absolute value of the acceleration less than The iteration step size of the data points in the stage is set to , for the absolute value of acceleration arrive The iteration step size of the data points within the phase between is set to ,in, is the absolute value of the maximum acceleration, , , The maximum value of acceleration and The acceleration value of The model parameters are optimized based on the iteration direction and iteration step size of the model parameters.

4. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the method according to any one of claims 1 to 2.

5. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is used to execute the method described in any one of claims 1-2.

Images