A method for determining a risk condition and an electronic device

Abstract

The application provides a risk working condition determination method and electronic equipment, and relates to the technical field of automobile safety detection. The determination method comprises the following steps: obtaining each shielding working condition corresponding to a to-be-detected vehicle and working condition data corresponding to each shielding working condition; determining working condition parameters corresponding to each shielding working condition based on the working condition data; determining working condition evaluation indexes corresponding to each shielding working condition based on the working condition parameters; sorting the working condition evaluation indexes from large to small; and taking the shielding working conditions corresponding to the working condition evaluation indexes with a preset percentage in the front as the risk working conditions. The application determines the risk working conditions in the shielding working conditions through the working condition evaluation indexes, avoids subjective judgment of the risk working conditions based on experience, and improves the accuracy when the risk working conditions are determined.

Description

Technical Field

[0001] This invention relates to the field of automotive safety testing technology, and more specifically, to a method and electronic device for determining risky operating conditions. Background Technology

[0002] With the development of science and technology, automobiles have gradually become an indispensable part of people's daily lives. However, children are easily left in cars during use. Hot and cold weather can lead to a harsh environment inside the car, which can easily cause safety problems for children left in the car. Therefore, new cars need to meet the China New Car Assessment Program (C-NCAP). For example, vehicles need to meet the testing requirements for the child forgetfulness reminder function in the 2027 version of C-NCAP. When testing the in-vehicle child forgetting reminder function, it is necessary to ensure that all possible occlusion conditions in the vehicle under test are covered, and to identify the high-risk occlusion conditions (i.e. extreme occlusion conditions) that require special attention. However, in the existing technology, the determination of the occlusion conditions of the vehicle under test usually relies on human experience. The selection based on human experience is prone to incomplete acquisition of occlusion conditions, and the accuracy of the high-risk occlusion conditions determined by humans is low. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method and electronic device for determining risky operating conditions, which can solve the technical problems in the prior art that the obstruction conditions of the vehicle under test cannot be fully obtained and the accuracy of the high-risk obstruction conditions is low.

[0004] To achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows: In a first aspect, embodiments of the present invention provide a method for determining risky operating conditions, the method comprising: Acquire the various occlusion conditions corresponding to the vehicle under test and the corresponding condition data for each occlusion condition; wherein, the condition data includes detection radar data, micro-motion feature data, detection confidence data and background clutter data; The target echo energy and background clutter intensity corresponding to each of the aforementioned obstruction conditions are determined based on the detection radar data and background clutter data corresponding to each of the aforementioned obstruction conditions; the micro-motion feature sharpness corresponding to each of the aforementioned obstruction conditions is determined based on the micro-motion feature data corresponding to each of the aforementioned obstruction conditions; and the detection confidence and continuous detection success rate corresponding to each of the aforementioned obstruction conditions are determined based on the detection confidence data corresponding to each of the aforementioned obstruction conditions; wherein, the operating parameters corresponding to the obstruction conditions include the target echo energy, the background clutter intensity, the micro-motion feature sharpness, the detection confidence, and the continuous detection success rate; Based on the operating parameters, determine the operating condition evaluation index corresponding to each of the occlusion conditions. The evaluation indicators for each working condition are sorted from largest to smallest, and the occlusion working condition corresponding to the top-ranked preset percentage of the evaluation indicators is taken as the risk working condition.

[0005] Secondly, embodiments of the present invention also provide an electronic device, comprising: At least one processor, and a memory communicatively connected to at least one of the processors; The memory stores instructions executable by at least one of the processors, which, when executed, enable the at least one processor to perform the aforementioned method for determining risk conditions. The present invention has the following technical effects: The method for determining risk conditions provided by this invention automatically acquires the obstruction conditions and operating condition data of the vehicle under test. Based on the original radar echo data, radar intermediate processing data, micro-motion feature data, detection confidence data, and background clutter data, it obtains operating condition parameters such as target echo energy, background clutter intensity, micro-motion feature clarity, detection confidence (i.e., the fluctuation range of the confidence in the presence of the target), and continuous detection success rate. Based on the values ​​of each operating condition parameter under each obstruction condition, it determines the corresponding operating condition evaluation index for each obstruction condition. This transforms the impact of each obstruction condition of the seat under test on the detection capability of the child detection radar into a quantifiable and comparable objective result. The operating condition evaluation index is sorted, with a larger value indicating a higher risk of the corresponding obstruction condition. Risk conditions (i.e., high-risk obstruction conditions) are screened from the obstruction conditions, avoiding the situation where subjective human experience leads to incomplete acquisition of obstruction conditions. The method improves the accuracy of identifying high-risk obstruction conditions by determining the risk conditions among the obstruction conditions based on the operating condition evaluation index.

[0006] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0007] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0008] Figure 1 A flowchart illustrating a method for determining risky operating conditions provided by an embodiment of the present invention is shown. Figure 2 The diagram illustrates a specific example flowchart of a method for determining risky operating conditions provided by an embodiment of the present invention. Detailed Implementation

[0009] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be described below in conjunction with the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0010] This embodiment provides a method and electronic device for determining risky operating conditions. See [link to documentation]. Figure 1 The diagram shows a flowchart of a method for determining risky operating conditions. This method mainly includes: Step S10: Obtain the occlusion conditions and corresponding condition data for each occlusion condition of the vehicle under test; wherein, the condition data includes detection radar data, micro-motion feature data, detection confidence data and background clutter data. Specifically, a vehicle coordinate system is established based on the vehicle under test. The origin of the vehicle coordinate system is the midpoint of the rear axle floor inside the passenger compartment (i.e., the cabin). The x-axis of the vehicle coordinate system is along the longitudinal direction of the vehicle under test, the y-axis is along the transverse direction of the vehicle under test, and the z-axis is along the vertical direction of the vehicle under test. Based on the vehicle coordinate system, the three-dimensional point cloud data of the passenger compartment of the vehicle under test is obtained. The child test samples are placed on each seat of the vehicle under test, and the placement area of ​​the child test samples on each seat is determined. Based on the seat adjustment parameters and degrees of freedom of each seat, the various occlusion conditions of the vehicle under test are determined. The operating condition data under each occlusion condition is obtained based on the data acquisition device (including pulse per second (PPS) synchronous radar and feature acquisition module).

[0011] Step S20: Determine the target echo energy and background clutter intensity for each occlusion condition based on the detection radar data and background clutter data corresponding to each occlusion condition; determine the micro-motion feature sharpness for each occlusion condition based on the micro-motion feature data corresponding to each occlusion condition; determine the detection confidence and continuous detection success rate for each occlusion condition based on the detection confidence data corresponding to each occlusion condition; wherein, the condition parameters corresponding to the occlusion condition include target echo energy, background clutter intensity, micro-motion feature sharpness, detection confidence, and continuous detection success rate; The system continuously collects operating condition data for a preset number of frames, and determines the operating condition parameters corresponding to each occlusion condition based on the operating condition data for each preset number of frames; among them, each operating condition parameter reflects the core characteristics of each occlusion condition to a certain extent.

[0012] Step S30: Determine the operating condition evaluation index corresponding to each shading condition based on the parameters of each operating condition. Based on the parameters of each operating condition, the corresponding operating condition evaluation index is determined for each occlusion condition. The operating condition evaluation index reflects the risk level of the occlusion condition (i.e., the severity of occlusion). The operating condition evaluation index is directly proportional to the risk level of the occlusion condition. The larger the value of the operating condition evaluation index, the more severe the occlusion of the child detection radar by the chair under test under the occlusion condition, and the higher the risk of missed detection under the occlusion condition. The risk level of each occlusion condition is quantified, avoiding the bias of human subjective judgment when judging high-risk occlusion conditions.

[0013] Step S40: Sort the evaluation indicators of each working condition from largest to smallest, and take the occlusion working condition corresponding to the preset percentage of the working condition evaluation indicators at the top of the sort as the risk working condition. The operating condition evaluation indicators corresponding to each occlusion condition are sorted from largest to smallest. Since the larger the operating condition evaluation indicator, the higher the risk level of the corresponding occlusion condition, the occlusion conditions corresponding to the operating condition evaluation indicators at the top of the sorting are selected as risk conditions. In addition, the occlusion conditions corresponding to the operating condition evaluation indicators at the top of the sorting can also be selected as risk conditions. The risk conditions selected in this way take into account both the coverage of the extreme operating conditions and the cost of subsequent verification.

[0014] This invention provides a method for determining risky operating conditions. It automatically acquires the obstruction conditions and operating condition data of the vehicle under test. Based on detection radar data, micro-motion feature data, detection confidence data, and background clutter data, it obtains operating condition parameters such as target echo energy, background clutter intensity, micro-motion feature clarity, detection confidence (i.e., the fluctuation range of target presence confidence), and continuous detection success rate. Based on the values ​​of each operating condition parameter under each obstruction condition, it determines the corresponding operating condition evaluation index. This transforms the impact of each obstruction condition of the seat under test on the detection capability of the child detection radar into quantifiable and comparable objective results. The operating condition evaluation indexes are ranked; a higher index value indicates a higher risk of the corresponding obstruction condition. Risky operating conditions (i.e., high-risk obstruction conditions) are selected from the obstruction conditions, avoiding incomplete acquisition of obstruction conditions due to subjective human experience. Determining risky operating conditions based on operating condition evaluation indicators improves the accuracy of identifying high-risk obstruction conditions.

[0015] In one embodiment, the steps provided in this embodiment for obtaining the occlusion conditions corresponding to the vehicle under test and the corresponding condition data include: Step S101: Determine the seat to be tested in the vehicle based on the radar parameters corresponding to the child detection radar; wherein, the radar parameters include the radar installation position, the direction of the radar main beam, the half-power angle of the main beam, and the effective detection distance; Obtain the three-dimensional radar center coordinates of the child detection radar (i.e., child presence detection radar) in the vehicle coordinate system. The radar installation location is obtained, the direction of the radar main beam is determined based on the radar installation location, and the factory parameters of the child detection radar (including the half-power angle of the radar main beam and the effective detection range of the radar) are obtained. The factory parameters, radar installation location and radar main beam direction are combined to obtain the radar parameters corresponding to the child detection radar, and the test seat in the vehicle is determined based on the radar parameters.

[0016] Step S102: Obtain the various occlusion conditions corresponding to the seat under test based on the seat adjustment parameters and parameter degrees of freedom of the seat under test; wherein, the seat adjustment parameters include the fore-and-aft sliding amount, backrest tilt angle and seat height; Select one or more (usually all) of the fore-and-aft sliding amount, backrest tilt angle, and seat height as seat adjustment parameters. Determine the maximum and minimum values ​​of the degrees of freedom for each parameter, and thus determine the range of values ​​for the degrees of freedom (i.e., physical adjustment range) for the fore-and-aft sliding amount, backrest tilt angle, and seat height. Specifically, the range of values ​​for the degrees of freedom of the fore-and-aft sliding amount is obtained as follows: , Let be the minimum value corresponding to the fore-and-aft sliding amount of the i-th seat to be tested. Let be the maximum value corresponding to the fore-and-aft sliding amount of the i-th seat under test; the range of values ​​for the parametric degrees of freedom of the backrest tilt angle is obtained as follows. , Let be the minimum value corresponding to the backrest tilt angle of the i-th seat to be tested. Let be the maximum value corresponding to the backrest tilt angle of the i-th seat to be tested; the range of values ​​for the seat height parameter is obtained as follows. , Let be the minimum value corresponding to the seat height of the i-th seat to be tested. Let be the maximum value corresponding to the seat height of the i-th seat under test; based on the parameter interval values ​​corresponding to each seat adjustment parameter (i.e., the adjustment step that meets the engineering test accuracy of each seat adjustment parameter), values ​​are taken within the range of parameter degrees of freedom to obtain the parameter degrees of freedom corresponding to each seat adjustment parameter; among them, the parameter interval value corresponding to the fore-and-aft sliding amount can be set to 10mm; the parameter interval value corresponding to the backrest tilt angle can be set to 2°; and the parameter interval value corresponding to the seat height can be set to 5mm. Based on the values ​​of each parameter's degree of freedom, construct the parameter vector corresponding to each test seat: ; in, For the i-th seat to be tested, the i-th... A parameter vector The total number of seats to be tested is I; The total number of parameter vectors corresponding to the i-th test seat is , , Let be the total number of degrees of freedom for the parameter corresponding to the m-th seat adjustment parameter for the i-th seat under test. When the seat adjustment parameters are fore-and-aft sliding, backrest tilt angle, and seat height, M=3, the fore-and-aft sliding is labeled m=1, representing the first seat adjustment parameter; the backrest tilt angle is m=2, representing the second seat adjustment parameter; and the seat height is m=3, representing the third seat adjustment parameter. For the i-th seat to be tested, the i-th... The parametric degrees of freedom (i.e., values) of the forward and backward slip in a parameter vector; For the i-th seat to be tested, the i-th... The parametric degrees of freedom of the backrest tilt angle in the parameter vector; For the i-th seat to be tested, the i-th... The degree of freedom of the seat height parameter in the parameter vector; Each parameter vector corresponding to the aforementioned test seats can be considered as an occlusion condition. Based on all the degrees of freedom of the parameters, the corresponding seat adjustment parameters are assigned values. These assigned seat adjustment parameters are then combined to obtain the parameter vector corresponding to each test seat, thus determining all occlusion conditions corresponding to each test seat. Based on all occlusion conditions, a set of occlusion conditions corresponding to all test seats is constructed. ; in, A set of occlusion conditions is constructed for all occlusion conditions; the occlusion conditions are then optimized without distinguishing between the seats under test, resulting in an optimized set of occlusion conditions: ; in, Let N be the k-th occlusion condition; N is the total number of all occlusion conditions corresponding to all seats under test; the occlusion conditions of each seat adjustment parameter of each seat under test under each parameter degree of freedom are retained in the occlusion condition set to ensure the completeness of the occlusion conditions. Discretizing each seat adjustment parameter can ensure the executability and reproducibility of traversing the occlusion conditions and meet the consistency requirements of the 2027 version of C-NCAP test.

[0017] Step S103: Adjust the seat under test to each occlusion condition and collect the condition data under each occlusion condition; wherein, the condition data includes detection radar data, micro-motion feature data, detection confidence data and background clutter data; Based on the degrees of freedom of the seat adjustment parameters under various obstruction conditions, target adjustment parameters are determined. These parameters are then transmitted to the electric actuator of the seat under test. The electric actuator (i.e., the seat controller) adjusts the fore-and-aft sliding amount, backrest tilt angle, and seat height of the seat under test according to the target adjustment parameters. After adjustment, the difference between the actual values ​​of the fore-and-aft sliding amount, backrest tilt angle, and seat height and the target adjustment parameters is used to determine whether the seat under test is properly adjusted. Specifically, the absolute value of the difference between the actual value of the fore-and-aft sliding amount and the target adjustment parameter is less than or equal to the sliding amount difference (which can be set to 5mm); the absolute value of the difference between the actual value of the backrest tilt angle and the target adjustment parameter is less than or equal to the tilt angle difference (which can be set to 0.5°); and the absolute value of the difference between the actual value of the seat height and the target adjustment parameter is less than or equal to the height difference (which can be set to 2mm). After the test seats are adjusted to their proper positions under various occlusion conditions, the child test samples are placed on the test seats according to the placement rules. The data acquisition device then collects the operating condition data of each test seat under each occlusion condition. This operating condition data includes detection radar data, micro-motion characteristic data, detection confidence data, and background clutter data. The detection radar data includes raw radar echo data (including raw radar echo...). The data includes radar intermediate processing data (including range-dimensional Fast Fourier Transform (FFT) results, moving target detection output, target range gate, and target detection confidence, etc.); among which, the original radar echo data refers to the complex baseband sampling sequence output by the child detection radar receiving channel without range-dimensional Fast Fourier Transform, filtering, target detection, and feature extraction processing; the sampling rate of the original radar echo is consistent with the native sampling rate of the child detection radar; the child test samples can be selected from the four groups of child test samples of 0 years old, 1 year old, 3 years old, and 6 years old as specified in the 2027 version of C-NCAP, and the size parameters of the standard child dummy corresponding to the child test samples in the same group are the same to avoid the size difference of the child test samples from interfering with the subsequent evaluation of the occlusion condition; The micro-motion feature data includes the time-domain amplitude sequence and frequency-domain feature spectrum of breathing and micro-body movements extracted from the slow-time echo of the child test sample at the target distance gate, with the target being a child; the detection confidence data includes the continuous output status of the child detection radar, outputting continuous target detection confidence (usually output once every 1 second), continuously collected for 30 seconds under various obstruction conditions; the background clutter data includes clutter echo signals. When acquiring background clutter data, the child test sample on the seat to be tested needs to be removed first, keeping the other parameters inside the vehicle unchanged (including the position of each seat, the status of the doors and windows, the operating parameters of the child detection radar, and the environmental conditions inside the vehicle unchanged); among them, the target detection confidence refers to the confidence (i.e., probability) of the existence of the child test sample output by the child detection radar when detecting the child test sample. Specifically, during the data collection process, all other non-test variables are kept constant. The interior temperature of the vehicle under test is controlled at 23℃±2℃, the doors and windows are kept completely closed, there are no other moving objects inside the vehicle, and the power supply voltage of the child detection radar is kept stable at 13.5V±0.1V. Based on the child detection radar (i.e., second-pulse synchronous radar), the original radar echo data, radar intermediate processing data, and background clutter data are collected. Micro-motion feature data and detection confidence data are obtained based on the feature acquisition module. When collecting data, each data acquisition device and the electric actuator of the seat are uniformly transmitted through the second pulse, and the correlation between data is completed according to the timestamp (alignment error is usually ≤10ms) to ensure the correspondence between different data.

[0018] In one embodiment, the steps of adjusting the seat under test to various occlusion conditions and collecting operating data under each occlusion condition provided in this embodiment include: Step S1031: Adjust the degree of freedom of the seat adjustment parameters of the seat under test according to each occlusion condition, and collect the working condition data under each occlusion condition. Step S1032: Determine the standard deviation of clutter energy corresponding to each occlusion condition based on the background clutter data. If the standard deviation of clutter energy is greater than the preset clutter energy threshold, re-acquire the condition data; if the standard deviation of clutter energy is less than or equal to the preset clutter energy threshold, retain the condition data. The degrees of freedom of the seat adjustment parameters of the test seat were adjusted according to the target adjustment parameters for each occlusion condition. Operating data under each occlusion condition was collected. The energy consistency of the background clutter was verified based on the clutter echo signal in the background clutter data, and the standard deviation of the clutter energy was calculated. ; in, Let $\frac{k}{k}$ be the standard deviation of clutter energy for the $k$-th operating condition. This refers to the clutter echo signal obtained at sampling point t under the k-th operating condition; This represents the clutter echo power obtained at sampling time t under the k-th operating condition; This represents the average clutter echo power at all sampling times under the k-th operating condition; This represents the total number of sampling points of the single-frame clutter echo signal in the fast time dimension under the k-th operating condition. The decibels corresponding to the standard deviation of clutter energy are compared with the preset clutter energy threshold (which can be set to 0.5dB). If the decibels corresponding to the standard deviation of clutter energy are greater than 0.5dB, the operating condition data is re-acquired; if the decibels corresponding to the standard deviation of clutter energy are less than or equal to 0.5dB, the operating condition data is retained. Among them, the operating condition data collected under the same obstruction operating condition are associated with the same operating condition number and the same timestamp.

[0019] In one embodiment, the step of determining the seat to be tested in the vehicle based on the radar parameters corresponding to the child detection radar provided in this embodiment includes: Step S1011: Determine the effective detection range and propagation path length of the child detection radar based on radar parameters; Step S1012: Determine the seat in the vehicle to be tested that is located within the effective detection range as the seat to be tested; The effective detection range and propagation path length of the child detection radar are determined based on radar parameters, including the radar's installation location, main beam direction, and main beam half-power angle. The effective detection range (i.e., effective coverage space) and propagation path length of the child detection radar are determined based on the effective detection range and effective detection distance. The occlusion conditions corresponding to each seat are determined based on the seat adjustment parameters and parameter degrees of freedom of each seat. The child test samples are placed under the corresponding occlusion conditions of each seat according to the placement rules. The sample placement area of ​​the child test samples under each occlusion condition of each seat is determined. It is determined whether there is an intersection between each sample placement area and the effective detection range (i.e., there is an intersecting area). If there is no sample placement area that intersects with the effective detection range in each sample placement area corresponding to the seat, the seat is discarded and will not be used as a test seat in the future. If at least one sample placement area in each sample placement area corresponding to the seat intersects with the effective detection interval, then the seat is taken as the test seat, and all degrees of freedom of the seat adjustment parameters of the test seat are retained; based on the sample placement area of ​​the test seat under each occlusion condition and the effective detection interval, it is determined whether there are invalid conditions in each occlusion condition. Specifically, based on the geometric pre-judgment marking of the sample placement area under each occlusion condition, an occlusion judgment function is established: ; in, This is the occlusion detection function; For the i-th seat to be tested, the i-th... The three-dimensional coordinates of the center point of the sample placement area in the vehicle coordinate system under the occlusion condition; The propagation path segment between the child detection radar and the sample placement area; For the i-th seat to be tested, the i-th... The sample three-dimensional envelope surface of the sample placement area under the occlusion condition is generated from the outer contour point set of the corresponding seat in the three-dimensional point cloud data of the vehicle under test, and determined by coordinate transformation according to the values ​​of the parameter degrees of freedom of the seat under test, such as the front and rear sliding amount, backrest tilt angle, and seat height. This is a function to determine the intersection of the propagation path segment and the three-dimensional envelope of the sample. Represents the empty set; If the result of the intersection judgment function is Then it represents the propagation path segment corresponding to the i-th test seat. Under occlusion conditions, the three-dimensional envelope surfaces of the samples do not intersect, and the i-th test seat corresponds to the i-th... The three-dimensional envelope of the sample under the occlusion condition did not obstruct the propagation path of the child detection radar. If the value is 0, the occlusion condition is only retained in the test record and does not participate in the selection of subsequent risk conditions (i.e., the occlusion condition is marked as an occlusion condition outside the effective detection range of the child detection radar), distinguishing between insufficient coverage of the child detection radar and the degradation of the detection performance of the child detection radar caused by occlusion conditions; if the result of the intersection judgment function is not... Then it represents the propagation path segment corresponding to the i-th test seat. There is an intersection between the three-dimensional envelope surfaces of the samples under the occlusion condition. The i-th test seat corresponds to the i-th... The three-dimensional envelope of the sample under occlusion conditions obstructs the propagation path of the child detection radar. The value of is 1, which retains the current occlusion condition (which can be marked as the radar main path occlusion condition). Adjacent or consecutive occlusion conditions are merged into the radar main path occlusion interval. Further filtering of occlusion conditions based on the occlusion judgment function helps to improve the filtering efficiency when filtering risk conditions in the future and avoids missing potential occlusion conditions.

[0020] In one embodiment, the steps provided in this embodiment for obtaining the various occlusion conditions corresponding to the seat under test based on the seat adjustment parameters and parameter degrees of freedom of the seat under test include: Step S1021: Determine the degree of freedom weights corresponding to each seat adjustment parameter based on the proportion of the test seat to the propagation path length under each occlusion condition. Step S1022: Determine the high-weight parameters and low-weight parameters in each seat adjustment parameter corresponding to the seat under test based on the weights of each degree of freedom. Step S1023: Assign values ​​to each seat adjustment parameter and combine them to obtain each occlusion condition; wherein, if the seat adjustment parameter is a high-weight parameter, the assigned value range of the seat adjustment parameter includes each value in the parameter degree of freedom; if the seat adjustment parameter is a low-weight parameter, the assigned value range of the seat adjustment parameter includes the maximum and minimum values ​​in the parameter degree of freedom. An operating condition database is established based on the occlusion judgment function-filtered operating conditions, and the total number of occlusion operating conditions in the database is determined. ,like If the fully automatic traversal mode is selected, the electric actuator of the test seat will be automatically controlled according to the sequence of occlusion conditions, so that the test seat will move to the position of each occlusion condition in the condition library in sequence, and the test seat will automatically complete the verification of the adjustment of the test seat, the collection of condition data and the recording of condition under each occlusion condition. The whole process does not require manual selection or switching of occlusion conditions. like If the semi-automatic scanning mode is selected, the influence weight of each seat adjustment parameter on the propagation path of the child detection radar is divided into high-weight parameters and low-weight parameters. The high-weight parameters are automatically controlled to complete the full-degree-of-freedom scanning combination to generate occlusion conditions. The low-weight parameters are only scanned at the extreme value level of the degree of freedom (i.e.,) to generate occlusion conditions. At the same time, the tester needs to complete the occlusion condition confirmation, the test seat status verification and the handling of abnormal occlusion conditions according to the prompts. Specifically, the scanning priority (i.e., occlusion condition priority) is determined according to the weight of the influence of the seat adjustment parameters of each test seat on the degree of obstruction of the propagation path of the child detection radar. For each test seat, the scanning priority is determined based on the weight of the influence of the seat adjustment parameters on the degree of obstruction of the propagation path of the child detection radar. Given a seat adjustment parameter, and assuming the degrees of freedom of the remaining seat adjustment parameters are taken as the midpoint of their physical adjustment ranges, the first seat adjustment parameter... For each seat adjustment parameter, the degrees of freedom are sequentially taken. The proportion of the intersection length between the sample's three-dimensional envelope (i.e., the seat's three-dimensional envelope) and the propagation path under each degree of freedom is calculated to the total propagation path length. This yields the occlusion ratio of the m-th seat adjustment parameter of the i-th seat under test when the v-th degree of freedom is taken. v= ; Calculate the first degree of occlusion of each test seat based on the average change in the occlusion ratio of adjacent parameter degrees of freedom. Weights of the degrees of freedom for each seat adjustment parameter: ; ; in, The weights of the degrees of freedom corresponding to the m-th seat adjustment parameter of the i-th seat under test; The average degree of freedom weights for the m-th seat adjustment parameter corresponding to the seat under test; Let m be the occlusion ratio of the m-th seat adjustment parameter of the i-th seat under test when taking the (v+1)-th parameter degree of freedom; determine the weight ratio corresponding to each seat adjustment parameter based on the average degree of freedom weight of each seat adjustment parameter: ; in, This represents the weighting ratio corresponding to the m-th seat adjustment parameter; if If the value is ≥0.8, then the degree of freedom weight of the m-th seat adjustment parameter is ranked first and is designated as a high-weight parameter, retaining all values ​​of the degree of freedom corresponding to the high-weight parameter; the other parameters in the seat adjustment parameters besides the high-weight parameter are designated as low-weight parameters, retaining the maximum and minimum values ​​of the degree of freedom corresponding to the low-weight parameter; the degree of freedom retained by the high-weight parameter and the degree of freedom retained by the low-weight parameter are combined to update the occlusion case set (i.e., the semi-automatic scanning case set): ; in, The total number of occlusion conditions after being filtered by weight ratio; priority is given to collecting condition data from the radar main path occlusion interval marked by geometric pre-judgment; the degree of freedom of seat adjustment parameters is filtered based on weight ratio, reducing the number of occlusion conditions to 30%~50% of the number of occlusion conditions corresponding to full parameter degrees of freedom.

[0021] In one embodiment, the operating parameters provided in this embodiment include target echo energy, micro-motion feature clarity, detection confidence, background clutter intensity, and continuous detection success rate. In one embodiment, the steps provided in this embodiment for determining the target echo energy and background clutter intensity corresponding to each obstruction condition based on the detection radar data and background clutter data corresponding to each obstruction condition include: Step S201: Determine the target distance gates and non-target distance gates corresponding to each occlusion condition; Step S202: Determine the target echo power corresponding to each target range gate based on the detection radar data corresponding to each occlusion condition; determine the clutter echo power corresponding to each non-target range gate based on the background clutter data corresponding to each occlusion condition. Step S203: Calculate the sum of the target echo power for each blocking condition to determine the target echo energy corresponding to each blocking condition; calculate the average value of the clutter echo power to determine the background clutter intensity corresponding to each blocking condition. A range-dimensional Fast Fourier Transform (FFT) is performed on each frame of the raw radar echo data to obtain the raw radar echo (i.e., complex echo amplitude) for each range gate. Based on the squared modulus of each raw radar echo, the target echo power (i.e., raw radar echo power) corresponding to each range gate is obtained. The maximum range gate corresponding to the maximum target echo power is determined within a preset target range search range based on the actual location of the child test sample. A preset number of range gates are then extended before and after this maximum range gate. Based on the maximum range gate and the extended range gates, the target range gates are obtained, forming the target range gate intervals. The target echo power of each target range gate within the target range gate interval is summed to obtain the target echo energy (i.e., the target echo energy characteristic).

[0022] ; in, This represents the target echo energy corresponding to the kth occlusion condition in the set of occlusion conditions updated with weighted ratios. For the k-th occlusion condition, the first... The first frame of raw radar echo data The target echo power corresponding to each target range gate =1,2,……, , Preset frame count (can be set to 100); =1,2,……, ; For the target distance gate interval Total number of doors for internal target distance; = , For the k-th occlusion condition, the first... The first frame of raw radar echo data The original radar echo corresponding to each target range gate; Excluding the target range threshold from each frame of raw radar echo data In addition to the distance gates located within the effective detection range of the child detection radar, the pre-calibrated strong reflection distance gates corresponding to the seats, doors, and other fixed structures of the vehicle under test are removed, resulting in the non-target distance gates and the non-target distance gate range (i.e., the non-living object area). Based on background clutter data, the clutter echo power corresponding to each non-target range gate in each frame is determined, and the background clutter intensity (i.e., background clutter intensity characteristics) is determined based on each clutter echo power. : ; in, The background clutter intensity corresponding to the kth occlusion condition in the set of occlusion conditions updated with weighted ratios; For the k-th occlusion condition, the first... The first frame of raw radar echo data The clutter echo power corresponding to each non-target range gate =1,2,……, ; For the target distance gate interval The total number of non-target distance gates; = , For the k-th occlusion condition, the first... The first frame of raw radar echo data The clutter echo signal corresponding to the non-target range gate.

[0023] In one embodiment, the step of determining the sharpness of micro-motion features corresponding to each occlusion condition based on the micro-motion feature data corresponding to each occlusion condition provided in this embodiment includes: Step S204: Determine the peak value of the breathing feature and the noise floor corresponding to each target distance gate based on the micro-motion feature data corresponding to each occlusion condition; Step S205: Calculate the ratio of the peak value of the breathing feature to the noise floor for each occlusion condition to obtain the micro-motion feature clarity corresponding to each occlusion condition. Slow-time series extraction was performed on the raw radar echo data (i.e., complex echo data) within each target range gate under various obstruction conditions. The target micro-motion signal corresponding to the child test sample was obtained through static clutter suppression and phase expansion processing. The target micro-motion signal was subjected to spectral analysis to extract the respiratory feature peak value in the 0.1Hz to 0.8Hz frequency band. The average spectral amplitude of the frequency band outside this band or the adjacent target-free frequency band was used as the noise floor. The ratio of the respiratory feature peak value to the noise floor was used as the micro-motion feature clarity (i.e., the clarity of the micro-motion feature of living organisms) under each obstruction condition. The clarity of the micro-motion feature of living organisms cannot be directly obtained from the detection confidence.

[0024] In one embodiment, the steps provided in this embodiment for determining the detection confidence level and continuous detection success rate for each occlusion condition based on the detection confidence level data corresponding to each occlusion condition include: Step S206: Obtain the target presence confidence of a consecutive preset number of frames in the detection confidence data corresponding to each occlusion condition, calculate the standard deviation of the target presence confidence of each occlusion condition, and obtain the detection confidence corresponding to each occlusion condition. Step S207: Determine the number of targets with a confidence level greater than a preset confidence level in each occlusion condition, obtain the first number, calculate the ratio of each first number to the preset number of frames, and obtain the continuous detection success rate corresponding to each occlusion condition. The target presence confidence (i.e., the presence confidence of the child test samples) is read from the radar intermediate processing data (or radar detection output data) of each occlusion condition for a consecutive preset number of frames (i.e., 100 consecutive frames) to obtain the detection confidence data. The standard deviation of the target presence confidence of 100 consecutive frames under each occlusion condition is calculated to obtain the detection confidence (i.e., the fluctuation range of the target presence confidence). The number of frames under each occlusion condition where the target presence confidence is higher than the preset confidence is divided by 100 to obtain the continuous detection success rate corresponding to each occlusion condition. It should be noted that the detection confidence and continuous detection success rate cannot be directly calculated by fast Fourier transform.

[0025] In one embodiment, the steps for determining the operating condition evaluation index corresponding to each occlusion condition based on each operating condition parameter provided in this embodiment include: Step S301: Normalize each working condition parameter to obtain each normalized parameter; Step S302: Determine the working condition evaluation index corresponding to each occlusion working condition based on the preset weights and normalized parameters corresponding to each working condition parameter. For the total quantity For all occlusion conditions, the condition parameters are divided into positive and negative feature parameters. Each condition parameter is normalized within the [0,1] interval to obtain the normalized parameters. Among the condition parameters, background clutter intensity and detection confidence are positive feature parameters (higher values ​​indicate more severe occlusion). Target echo energy, micro-motion feature clarity, and continuous detection success rate are negative feature parameters (lower values ​​indicate more severe occlusion). Specifically, the normalization formula for the positive feature parameters is: ; The normalization formula for negative feature parameters is: ; in, Let d be the value of the normalized parameter corresponding to the d-th condition parameter under the k-th occlusion condition, where d = 1, 2, ..., 5 (there are five condition parameters in total). This represents the value of the parameter corresponding to the d-th condition under the k-th occlusion condition; This represents the maximum value of the parameter corresponding to the d-th condition under the k-th occlusion condition; This represents the minimum value of the parameter corresponding to the d-th condition under the k-th occlusion condition.

[0026] Based on the preset weight corresponding to the dth working condition parameter (satisfy Each preset weight is set according to the contribution of each working condition parameter in the 2027 C-NCAP test requirements to the risk of missed detection under occlusion conditions; among them, the preset weight of micro-motion feature clarity accounts for the highest proportion, which is used to ensure a strong correlation between the working condition evaluation index and the actual detection capability of the child detection radar. The normalized parameters are weighted and fused to obtain the working condition evaluation index (i.e., the worst occlusion evaluation index): ; in, For the kth occlusion condition, the condition evaluation index is used. The value range is [0,1]. The higher the value of the working condition evaluation index, the stronger the suppression effect of the occlusion of the seat under the test on the detection capability of the child detection radar under the occlusion working condition, and the higher the risk of missing the child test sample. It directly realizes the quantifiable horizontal comparison of the occlusion effect of the occlusion working condition, and avoids the deviation caused by the subjective judgment of risk working conditions by human experience. Record all the occlusion conditions corresponding to each seat under test as follows: , The performance evaluation index (i.e., the worst-case shading evaluation index) for each shading condition is: The higher the value of the performance evaluation index, the greater the degree of occlusion of the child test sample under that occlusion condition, the worse the detectability of living beings, and the higher the risk of missed detection. The performance evaluation indices are sorted from largest to smallest (i.e., descending order) to obtain the descending order sequence of each occlusion condition corresponding to each performance evaluation index: ; in, The descending sorted sequence corresponding to each occlusion condition; For sorting functions, A descending order rule is specified, where the higher the occlusion condition appears in the descending sort sequence (i.e., the smaller its rank), the higher the risk of missed detection for the child test samples under that occlusion condition. A predetermined percentage of occlusion conditions at the top of the descending sort sequence are designated as risk conditions. This method filters out risk conditions from the occlusion conditions, ensuring coverage of high-risk occlusion conditions and reducing subsequent verification costs. A preliminary set of risk conditions (i.e., a candidate set of conditions) is established. Under each risk condition in the preliminary risk condition set, a set of condition data (i.e., multiple condition data) is collected. Based on the condition evaluation indicators corresponding to each condition data, the risk conditions in the preliminary risk condition set (i.e., ...) are calculated. The volatility coefficient of the indicator: ; in, The index volatility coefficient corresponding to the kth risk condition in the risk condition set; For the k-th risk condition in the set of risk conditions The operating condition evaluation index obtained from each data collection; the index fluctuation coefficient, used to reflect the relative dispersion of the operating condition evaluation index obtained from each data collection under this risky operating condition. This represents the standard deviation of the evaluation index for each working condition under the k-th risk condition. This represents the average value of the evaluation indicators for each working condition under the k-th risk condition; based on Determine whether the performance indicators corresponding to this risky operating condition are stable. The larger the value, the more unstable the performance index (i.e., the evaluation result) of the risk condition. Greater than the preset reasonable fluctuation threshold When the value is 0.15, it indicates that the dispersion of the operating condition evaluation index under this risky operating condition exceeds the acceptable range. Random errors during data collection may cause distortion of the operating condition evaluation index, triggering secondary data collection and verification. A new set of operating condition data under this risky operating condition is collected. When there are at least three sets of operating condition data under the risky operating condition with index fluctuation coefficients ≤ the preset reasonable fluctuation threshold, the average value of the operating condition evaluation index corresponding to each risky operating condition in each set of valid operating condition data is used. : ; Each risk condition is sorted in descending order according to the average value of its corresponding condition evaluation index, resulting in a list of risk conditions with clearly defined priorities. The degree of high risk (i.e., priority) of each risk condition is correlated with its corresponding risk level. The values ​​are positively correlated, that is The smaller the value, the higher the risk level of the working condition. Finally, the list of the working conditions is output to complete the screening of the occlusion working conditions. Based on the index fluctuation coefficient, the misjudgment of the working condition caused by occasional noise or collection deviation can be effectively avoided, and the reliability of the working condition judgment result can be improved.

[0027] The method for determining risk conditions provided in this embodiment of the invention further includes the step of: using the worst-case occlusion condition as the core test condition for the child forgetfulness reminder function of the target vehicle model; firstly, using a sample of children of a specified age group to complete the functional performance verification under extreme occlusion conditions; then, directly reusing this fixed condition for testing samples of children of other age groups, eliminating the need to repeat the worst-case screening process, thereby reducing the overall testing cycle and improving the reliability of the test results while meeting regulatory testing requirements. The specific implementation method of this step is as follows: Combine the seat adjustment parameters corresponding to each risk condition in the risk condition list into a seat adjustment parameter group: ; in, For seat adjustment parameter groups; This refers to the fore-aft sliding amount in the seat adjustment parameter group; This refers to the backrest tilt angle in the seat adjustment parameter group; For the seat height parameter group, first select children's test samples from the specified age group (usually 1 year old) as specified in the 2027 version of C-NCAP, and fix the children's test samples in place according to the placement rules. Continuous data collection under various risk conditions The detection results of the child detection radar (usually Rt=300, corresponding to a test duration of 60 seconds) are used to calculate functional verification indicators: ; in, For functional verification metrics; For the alarm status of frame rt, the value is 1 when the in-vehicle child detection system triggers a valid alarm, and 0 when no alarm is triggered. The requirement is... Only after passing the basic performance verification under risky operating conditions can the functional verification indicators be considered to have passed; the functional verification indicators are directly aligned with the alarm reliability thresholds specified in the 2027 version of C-NCAP, avoiding verification bias caused by subjective human judgment and improving the consistency of test results; after passing the basic verification, the basic performance verification will be directly implemented. The corresponding risk conditions were reused for testing samples from children in other age groups (0, 1, 3, and 6 years old), and samples were collected from each age group in the same manner. Frame data calculation correspondence That's it; there's no need to iterate through the occlusion conditions again to filter for risky conditions. The method for determining risky operating conditions provided in this embodiment of the invention differs from the traditional age-group screening testing model, which has a total testing time of [missing information]. Where Ma is the total number of age groups covered by the test. The average time for a single round of risk condition screening; the total test time of the method provided in this embodiment of the invention is ,in, The testing duration for risky working conditions for a single age group is typically... Only This method reduces the overall testing cycle time by 40% or more, by 1 / 10 of the time required for occlusion testing. Based on the engineering prior that the attenuation effect of occlusion on the propagation path of the child detection radar is not strongly correlated with the age of the target child, this method significantly reduces redundant operations of repeated traversal without reducing the rigor of the test. At the same time, it ensures that all age group tests are carried out under the most stringent risk conditions, and the reliability of the test results is not affected by the omission of the screening process.

[0028] This invention also provides an electronic device, comprising: at least one processor, and a memory communicatively connected to at least one processor; wherein the memory stores instructions executable by at least one processor, the instructions being executed by at least one processor to enable at least one processor to perform the above-described method for determining risk conditions.

[0029] The method for determining risk conditions provided in this invention combines the vehicle interior layout of the vehicle under test, the installation location of the child detection radar, and the degree of freedom of the parameters of each seat adjustment to construct an occlusion condition library. By pre-marking the occlusion interval of the radar main path, the test boundary of risk conditions is effectively narrowed, and the screening efficiency of occlusion conditions is improved without reducing coverage. Based on raw radar echo data, intermediate radar data, micro-motion feature data, detection confidence data, and background clutter data, operating parameters such as target echo energy, background clutter intensity, micro-motion feature clarity, detection confidence (i.e., the fluctuation range of target presence confidence) and continuous detection success rate are obtained. Based on the values ​​of each operating parameter under various obstruction conditions, operating condition evaluation indicators are determined for each obstruction condition. This transforms the impact of various obstruction conditions of the tested seat on the detection capability of the child detection radar into quantifiable and comparable objective results, avoiding subjective human judgment bias. It accurately captures various hidden risk conditions, meets the stringent testing requirements of the 2027 C-NCAP regulations, and realistically reproduces the actual operating performance of the in-vehicle child detection system under risk conditions, improving the reliability and professionalism of testing in-vehicle child detection systems. This method can identify risky operating conditions (i.e., concealed extreme occlusion scenarios) in multi-dimensional coupling occlusion scenarios, improving the accuracy (accuracy can be improved by 60%) and reproducibility of risky operating condition judgment. It also reduces the risk of misjudgment caused by occasional noise and acquisition errors by using a candidate operating condition secondary verification mechanism. The worst occlusion operating condition among the selected risky operating conditions can be directly reused for testing of children in other age groups without repeating full-condition scanning, significantly shortening the overall testing cycle and reducing testing costs. It can provide a unified, standardized, and reproducible testing basis for different vehicle models and different radar installation schemes, and has good engineering application value and regulatory compliance value.

[0030] Based on the foregoing embodiments, this embodiment provides an example of using the aforementioned method for determining risky operating conditions to identify risky operating conditions within occlusion conditions. For details, please refer to... Figure 2 The flowchart shown illustrates a specific example of a method for determining risky operating conditions. This method includes: Step S501: Establish a vehicle coordinate system based on the vehicle under test and determine the radar parameters of the child detection radar; A vehicle coordinate system is established based on the vehicle under test. The three-dimensional point cloud data of the vehicle's cabin is obtained. The coordinates of the three-dimensional radar center of the child detection radar in the vehicle coordinate system are determined, and the radar position of the child detection radar is determined. The direction of the radar's main beam is determined based on the radar's installation location. The factory parameters of the child detection radar (including the half-power angle of the radar's main beam and the radar's effective detection range) are obtained. The radar parameters are then constructed based on the radar's installation location, the direction of the radar's main beam, the half-power angle, and the effective detection range. Step S502: Determine the seat to be tested in the vehicle under test based on the radar parameters of the child detection radar, and obtain the seat adjustment parameters and parameter degrees of freedom corresponding to the seat to be tested. The effective detection range of the child detection radar is determined based on the radar parameters of the child detection radar. Based on the effective detection range and the sample placement area of ​​the child test sample under various occlusion conditions of each seat, the seat to be tested is determined, and the degrees of freedom of each parameter corresponding to each seat adjustment parameter (including fore-and-aft sliding amount, backrest tilt angle and seat height) of the seat to be tested is obtained. Step S503: Based on the occlusion judgment function, filter the occlusion conditions, establish a condition database, and determine the radar main path occlusion interval based on the occlusion conditions in the condition database. The sample three-dimensional envelope surface corresponding to the sample placement area of ​​the child test sample under each occlusion condition of each test seat is determined. Based on the occlusion judgment function, the intersection of the three-dimensional envelope surface of each sample and the propagation path line segment of the child detection radar is determined. If the intersection is an empty set, the occlusion condition is regarded as an invalid condition; if the intersection is not an empty set, the occlusion condition is retained and a condition library is established. Adjacent or consecutive blocking conditions in the operating condition database are grouped into the main radar path blocking interval. Step S504: Determine the acquisition mode based on the total number of occluded working conditions in the working condition database; If the total number of occlusion conditions in the condition database If the total number of occlusion conditions in the condition database is..., then select the fully automatic traversal mode; If so, select the semi-automatic scanning mode; Step S505: Adjust each seat to be tested to the corresponding occlusion conditions, and collect the working condition data under each occlusion condition based on the acquisition mode. The adjustment parameters of each seat under test are adjusted according to each occlusion condition, and the working condition data (including detection radar data, micro-motion feature data, detection confidence data and background clutter data) under each occlusion condition are obtained based on the acquisition mode. Step S506: Filter the clutter energy standard deviation corresponding to each operating condition data. Calculate the standard deviation of clutter energy corresponding to each operating condition data. If the decibel level corresponding to the standard deviation of clutter energy is greater than the preset clutter energy threshold (i.e., 0.5dB), then re-acquire the operating condition data; if the decibel level corresponding to the standard deviation of clutter energy is less than or equal to the preset clutter energy threshold (i.e., 0.5dB), then retain the operating condition data. Step S507: Determine the operating parameters based on the operating data of each occlusion condition; The data from each working condition determine the parameter values ​​of the working condition parameters (including target echo energy, background clutter intensity, micro-motion feature clarity, detection confidence, and continuous detection success rate) corresponding to each occlusion working condition; Step S508: Normalize the parameters of each working condition to obtain the normalized parameters corresponding to each occlusion working condition, and determine the evaluation index of each working condition based on the normalized parameters. The normalization formula based on positive feature parameters normalizes the background clutter intensity and detection confidence, while the normalization formula based on negative feature parameters normalizes the target echo energy, micro-motion feature sharpness, and continuous detection success rate; thus obtaining the normalized parameters. The weighted summation of each normalized parameter is used to obtain the working condition evaluation index corresponding to each occlusion condition. Step S509: Sort the various shading conditions according to the condition evaluation indicators to determine the risk conditions. All occlusion conditions are sorted from largest to smallest according to the condition evaluation index, and the occlusion conditions with the highest preset percentage after sorting are selected as risk conditions.

[0031] Step S510: Optimize the risk conditions based on the index fluctuation coefficients corresponding to each risk condition; Calculate the fluctuation coefficient of the indicators corresponding to each risk condition. If the fluctuation coefficient of the indicator is greater than the preset reasonable fluctuation threshold, the operating data of the corresponding risk condition will be collected again. If the fluctuation coefficient of each indicator is less than or equal to the preset reasonable fluctuation threshold, the list of risk conditions will be output.

[0032] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0033] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for determining risky operating conditions, characterized in that, The method includes: Acquire the various occlusion conditions corresponding to the vehicle under test and the corresponding condition data for each occlusion condition; wherein, the condition data includes detection radar data, micro-motion feature data, detection confidence data and background clutter data; The target echo energy and background clutter intensity corresponding to each of the aforementioned obstruction conditions are determined based on the detection radar data and background clutter data corresponding to each of the aforementioned obstruction conditions; the micro-motion feature sharpness corresponding to each of the aforementioned obstruction conditions is determined based on the micro-motion feature data corresponding to each of the aforementioned obstruction conditions; and the detection confidence and continuous detection success rate corresponding to each of the aforementioned obstruction conditions are determined based on the detection confidence data corresponding to each of the aforementioned obstruction conditions; wherein, the operating parameters corresponding to the obstruction conditions include the target echo energy, the background clutter intensity, the micro-motion feature sharpness, the detection confidence, and the continuous detection success rate; Based on the operating parameters, determine the operating condition evaluation index corresponding to each of the occlusion conditions. The evaluation indicators for each working condition are sorted from largest to smallest, and the occlusion working condition corresponding to the top-ranked preset percentage of the evaluation indicators is taken as the risk working condition.

2. The method for determining risky operating conditions according to claim 1, characterized in that, The steps of obtaining the various occlusion conditions corresponding to the vehicle under test and the corresponding condition data for each occlusion condition include: The test seat in the vehicle is determined based on the radar parameters corresponding to the child detection radar; wherein, the radar parameters include the radar installation position, radar main beam direction, main beam half power angle, and effective detection range; Based on the seat adjustment parameters and degrees of freedom of the seat under test, the various occlusion conditions corresponding to the seat under test are obtained; wherein, the seat adjustment parameters include the fore-and-aft sliding amount, the backrest tilt angle, and the seat height; The test seat is adjusted to each of the aforementioned occlusion conditions, and the operating data under each of the aforementioned occlusion conditions is collected.

3. The method for determining risky operating conditions according to claim 2, characterized in that, The step of adjusting the seat under test to each of the occlusion conditions and collecting operating data under each of the occlusion conditions includes: The degree of freedom of the seat adjustment parameters of the seat under test is adjusted according to each of the occlusion conditions, and the working condition data under each of the occlusion conditions is collected. Based on the background clutter data, the standard deviation of clutter energy corresponding to each of the occlusion conditions is determined. If the standard deviation of clutter energy is greater than a preset clutter energy threshold, the condition data is re-acquired; if the standard deviation of clutter energy is less than or equal to the preset clutter energy threshold, the condition data is retained.

4. The method for determining risky operating conditions according to claim 2, characterized in that, The step of determining the seat to be tested in the vehicle based on the radar parameters corresponding to the child detection radar includes: The effective detection range and propagation path length of the child detection radar are determined based on the radar parameters. The seat in the vehicle under test that is located within the effective detection range is identified as the seat under test.

5. The method for determining risky operating conditions according to claim 4, characterized in that, The step of obtaining the various occlusion conditions corresponding to the seat under test based on the seat adjustment parameters and parameter degrees of freedom of the seat under test includes: The degree of freedom weights corresponding to each seat adjustment parameter are determined based on the proportion of the test seat to the length of the propagation path under each of the occlusion conditions. Based on the weights of each degree of freedom, determine the high-weight parameters and low-weight parameters among the seat adjustment parameters corresponding to the seat under test; The seat adjustment parameters are assigned values ​​and combined to obtain the various occlusion conditions; wherein, if the seat adjustment parameter is a high-weight parameter, the assigned value range of the seat adjustment parameter includes each value in the parameter degree of freedom; if the seat adjustment parameter is a low-weight parameter, the assigned value range of the seat adjustment parameter includes the maximum and minimum values ​​in the parameter degree of freedom.

6. The method for determining risky operating conditions according to claim 1, characterized in that, The step of determining the target echo energy and background clutter intensity corresponding to each of the aforementioned obstruction conditions based on the detection radar data and background clutter data includes: Determine the target distance gate and non-target distance gate corresponding to each of the aforementioned occlusion conditions; The target echo power corresponding to each target range gate is determined based on the detection radar data corresponding to each of the aforementioned occlusion conditions; the clutter echo power corresponding to each non-target range gate is determined based on the background clutter data corresponding to each of the aforementioned occlusion conditions. Calculate the sum of the target echo power for each of the aforementioned blocking conditions to determine the target echo energy corresponding to each of the aforementioned blocking conditions; calculate the average value of the clutter echo power to determine the background clutter intensity corresponding to each of the aforementioned blocking conditions.

7. The method for determining risky operating conditions according to claim 6, characterized in that, The step of determining the sharpness of the micro-motion features corresponding to each of the occlusion conditions based on the micro-motion feature data corresponding to each of the occlusion conditions includes: Based on the micro-motion feature data corresponding to each of the aforementioned occlusion conditions, determine the respiratory feature peak and noise floor corresponding to each of the aforementioned target distance gates; Calculate the ratio of the peak value of the breathing feature to the noise floor for each of the occlusion conditions to obtain the micro-motion feature clarity corresponding to each of the occlusion conditions.

8. The method for determining risky operating conditions according to claim 1, characterized in that, The step of determining the detection confidence level and continuous detection success rate corresponding to each of the occlusion conditions based on the detection confidence level data corresponding to each of the occlusion conditions includes: Obtain the target presence confidence of a consecutive preset number of frames in the detection confidence data corresponding to each of the occlusion conditions, calculate the standard deviation of the target presence confidence of each of the occlusion conditions, and obtain the detection confidence corresponding to each of the occlusion conditions. The number of targets with a confidence level greater than a preset confidence level in each of the occlusion conditions is determined to obtain a first number. The ratio of each first number to the preset number of frames is calculated to obtain the continuous detection success rate corresponding to each of the occlusion conditions.

9. The method for determining risky operating conditions according to claim 1, characterized in that, The step of determining the operating condition evaluation index corresponding to each of the occlusion operating conditions based on each of the operating condition parameters includes: Each of the aforementioned operating condition parameters is normalized to obtain the normalized parameters. The evaluation index for each occlusion condition is determined based on the preset weights corresponding to each operating condition parameter and the normalized parameters.

10. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to at least one of the processors; The memory stores instructions executable by at least one of the processors, which are executed to enable the at least one processor to perform the method of any one of claims 1-9.

Images