Method for evaluating forward collision early warning performance of passenger car in rain and fog environment

Abstract

The invention relates to the technical field of automobile collision early warning performance testing, in particular to a method for evaluating forward collision early warning performance of a passenger vehicle in a rain and fog environment, which comprises the step of analyzing an alarm state frame by frame by synchronously acquiring a vehicle motion state and system alarm audio and video information. In a simulated rain and fog environment, three scenes of a static scene, a constant speed scene and a deceleration scene of a front vehicle are set for multiple tests, collision danger times, alarm times and false alarm times are counted, and the early warning accuracy rate, the missing report rate and the false alarm rate are calculated accordingly. Meanwhile, the alarm time difference is analyzed by combining the vehicle motion information, and the distance collision time is effectively early warned and strengthened. And integrating the multi-dimensional indexes to form a performance index, thereby accurately determining a system safety boundary under different rain and fog conditions, and realizing comprehensive and objective performance evaluation. The method improves the evaluation of the forward collision early warning performance of the passenger car in the rain and fog environment.

Description

Technical Field

[0001] This invention relates to the field of automotive collision warning performance testing technology, and in particular to a method for evaluating the forward collision warning performance of passenger vehicles in rain and fog conditions. Background Technology

[0002] With the widespread application of advanced driver assistance systems (ADAS) in passenger vehicles, forward collision warning systems have become crucial for improving driving safety. Currently, domestic standards such as GB / T 33577-2017 have standardized performance testing for these systems. However, existing evaluation methods primarily target favorable weather conditions such as no rain, no snow, and no fog, which are severely out of touch with the complex real-world driving environment. In common adverse weather conditions such as rain and fog, raindrops and fog significantly interfere with the system's sensor performance: visual sensors suffer from reduced image clarity and contrast, affecting target recognition; and the signals of lidar and millimeter-wave radar are attenuated by scattering and absorption from water droplets, leading to reduced detection accuracy. This makes it difficult for evaluation methods geared towards favorable weather conditions to accurately assess system performance in rain and fog environments. Currently, the industry lacks an effective method specifically designed for rain and fog environments to comprehensively and accurately evaluate the forward collision warning performance of passenger vehicles, representing a pressing technological gap in this field. Summary of the Invention

[0003] This invention provides a method for evaluating the performance of forward collision warning systems for passenger vehicles in rainy and foggy environments, which addresses the problem that existing evaluation methods cannot accurately assess the impact of severe rainy and foggy weather on the performance of forward collision warning systems for passenger vehicles.

[0004] The objective of this invention can be achieved through the following technical solutions: The first aspect of this invention is to provide a method for evaluating the forward collision warning performance of passenger vehicles in rain and fog conditions, comprising: Acquire motion state information of the vehicle under test and the target vehicle ahead; acquire audio and video alarm information issued by the forward collision warning system of the vehicle under test during the test; obtain the alarm light status and whether an alarm signal is issued for each frame of the vehicle under test during the test through the audio and video alarm information. Three test scenarios for vehicles were set up in rain and fog conditions for testing. The test vehicles were tested several times in each rain and fog environment. The number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. Based on the number of collision hazards that occurred, the number of times no collision hazards occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred, the warning accuracy, false alarm rate, and false alarm rate for each test scenario are obtained. Based on the status of the warning lights and whether an alarm signal was issued for each frame of the test vehicle, the alarm time difference is obtained. Based on the motion status information of the test vehicle and the target vehicle ahead, the effective warning distance and enhanced distance collision time are obtained. Based on the warning accuracy, false alarm rate, false alarm rate, alarm time difference, effective warning distance, and enhanced distance collision time for each test scenario, the comprehensive warning performance index of the test vehicle in each scenario is obtained. Based on the comprehensive warning performance index, the safety boundary of the test vehicle in rain and fog environments in each scenario is determined.

[0005] Furthermore, the acquisition of audio and video alarm information emitted by the forward collision warning system of the vehicle under test during the test; and the acquisition of the alarm light status and whether an alarm signal was emitted by the vehicle under test for each frame during the test through the audio and video alarm information, including: Step 1: Using the second pulse output by the Beidou timing board as a reference, information from the microphone and visual sensor is collected using hardware synchronization, and time tags with Beidou time as a reference are added as a synchronization alignment mark; among them, the microphone is used to collect audio alarm information issued by the passenger vehicle forward collision warning system, and the visual sensor is used to collect video alarm information issued by the passenger vehicle forward collision warning system. Step 2: Acquire the image information of the m-th frame through a visual sensor and perform feature extraction to obtain image information features; in parallel, acquire the audio information of the m-th frame through a microphone, perform feature extraction to obtain audio information features; Step 3: Based on the extracted image and audio information features, use a pattern recognition algorithm to detect whether the alarm light is on and whether an alarm signal has been issued; Step 4: When the alarm light is detected to be on, or when the forward collision warning system issues an alarm signal, this is considered a valid alarm. Record the current frame time as the time when the system issued the collision alarm. Step 5: Repeat steps 2 to 4 to process the image and audio information of each subsequent frame, and determine whether the alarm light is on and whether the forward collision warning system issues an alarm signal, until the test ends.

[0006] Furthermore, the tested vehicle undergoes several test cycles in each rain / fog environment to obtain the number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred, including: When the collision time between the tested vehicle and the target vehicle ahead is equal to If the time between the collision and the target vehicle in front is less than 100 seconds, it is considered a collision hazard situation; while if the time between the collision and the target vehicle in front is less than 100 seconds during a test, it is considered a collision hazard situation. At that time, it is considered a situation where no collision risk occurred; among them, Values ​​set according to Chinese national standards; When a collision hazard occurs and a valid alarm is triggered, it is recorded as the number of times a collision hazard occurs and an alarm is triggered; when a collision hazard occurs but no valid alarm is triggered, it is recorded as the number of times a collision hazard occurs but no alarm is triggered; when a collision hazard does not occur but a valid alarm is triggered, it is recorded as the number of times a collision hazard does not occur but an alarm is triggered.

[0007] Furthermore, the accuracy rate, false alarm rate, and false alarm rate for each test scenario are obtained based on the number of times a collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. This includes: The accuracy rate of the early warning is specifically expressed by the formula:

[0008] In the formula, This indicates the number of times a collision hazard occurred and an alarm was triggered during the test. This indicates the accuracy rate of the warnings in each scenario test. This indicates the number of times a collision hazard occurred during the test; The specific formula for the missed detection rate is as follows:

[0009] In the formula, This indicates the number of times a collision hazard occurred during the test but no alarm was triggered. This represents the false negative rate for each test scenario. The false alarm rate is specifically expressed by the formula:

[0010] In the formula, This indicates the number of times an alarm was triggered during the test when no collision hazard occurred. This represents the false positive rate for each test scenario. This indicates the number of times no collision hazards occurred during the test.

[0011] Furthermore, the alarm time difference is obtained based on the alarm light status and whether an alarm signal is issued in each frame of the test vehicle during the test; the effective warning distance and enhanced distance collision time are obtained based on the motion state information of the test vehicle and the target vehicle ahead, including: The alarm time difference is specifically expressed by the formula:

[0012] In the formula, Indicates the time difference of the alarm. Indicates the time when the system issues a collision alarm. This indicates the time during which a collision hazard occurred during the test. It is the absolute value symbol; The effective warning distance is specifically expressed by the formula:

[0013] In the formula, This represents the horizontal coordinate value of the vehicle being measured. This represents the vertical coordinate value of the vehicle being measured. This represents the horizontal coordinate value of the target vehicle. Represents the vertical coordinate value of the target vehicle. Indicates the effective warning distance; The enhanced distance collision time is specifically expressed by the formula:

[0014] In the formula, Indicates the speed of the vehicle being measured. This indicates the acceleration of the vehicle being tested. Indicates the speed of the target vehicle ahead. This indicates the acceleration of the target vehicle ahead. This indicates the time required to enhance the distance collision.

[0015] Furthermore, based on the warning accuracy, false alarm rate, false alarm rate, alarm time difference, effective warning distance, and enhanced distance collision time in each test scenario, a comprehensive warning performance index for the tested vehicle in each scenario is obtained; based on the comprehensive warning performance index, the safety boundary of the tested vehicle in rain and fog environments in each scenario is determined, including: Based on the warning accuracy, false alarm rate, and false alarm rate, the first warning performance index of the tested vehicle in each scenario is obtained; based on the alarm time difference, effective warning distance, and enhanced distance collision time, the second warning performance index of the tested vehicle in each scenario is obtained. Based on the first and second warning performance indices tested by the vehicle in each scenario, the comprehensive warning performance index of the passenger vehicle in each scenario is obtained. The comprehensive warning performance index is specifically expressed by the formula:

[0016] In the formula, This represents the first warning performance index of the tested vehicle in each scenario. This represents the second warning performance index of the tested vehicle in each scenario. This represents the overall warning performance index of the tested vehicle in each scenario; A three-dimensional early warning performance evaluation space is constructed with horizontal visibility as the x-axis, rainfall intensity as the y-axis, and comprehensive early warning performance index as the z-axis. The comprehensive early warning performance index of the tested vehicle in each scenario is mapped onto the three-dimensional early warning performance evaluation space and then fitted to obtain the early warning performance surface for each scenario. Three thresholds are selected to extract the safety boundary corresponding to the combination of horizontal visibility distance and rainfall intensity. The safety boundary is mapped onto the two-dimensional combination space to obtain the schematic diagram of the system performance safety boundary corresponding to the three thresholds in the three scenarios. The two-dimensional combined space is a space composed of horizontal visibility as the x-axis and rainfall intensity as the y-axis.

[0017] Furthermore, based on the warning accuracy rate, false negative rate, and false alarm rate, the first warning performance index of the tested vehicle in each scenario is obtained, specifically expressed by the formula:

[0018] In the formula, This indicates the accuracy rate of the warnings in each scenario test. This represents the false negative rate in each test scenario. This represents the false positive rate for each test scenario. Indicates the first preset weight. This indicates the second preset weight. This indicates the third preset weight. This represents an exponential function with the natural constant as its base. This indicates the first warning performance index of the tested vehicle in each scenario.

[0019] Furthermore, based on the alarm time difference, effective warning distance, and enhanced distance collision time, the second warning performance index of the tested vehicle in each scenario is obtained, specifically expressed by the formula:

[0020] In the formula, This represents the average alarm time difference across several tests in each scenario. This represents the maximum alarm time difference across several tests in each scenario. This represents the average effective warning distance across several tests in each scenario. This represents the maximum effective warning distance across a number of tests in each scenario. This represents the average of the enhanced distance-collision time across several tests in each scenario. This represents the maximum value of the enhanced distance-collision time across a number of tests in each scenario. This indicates the fourth preset weight. This indicates the fifth preset weight. This indicates the sixth preset weight. This represents the second warning performance index of the tested vehicle in each scenario.

[0021] A second aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the aforementioned method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions.

[0022] A third aspect of the present invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions.

[0023] Compared with existing technologies, the beneficial effects of this invention are: by acquiring the status of the warning lights and whether an alarm signal is issued for each frame of the vehicle under test during the test; improving the accuracy of vehicle alarm system monitoring; setting up three vehicle test scenarios in rain and fog environments for testing; the vehicle under test undergoes several test processes in each rain and fog environment, obtaining the number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred; through multi-scenario testing and key event statistics, ensuring the comprehensiveness and statistical significance of the evaluation results; based on the number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. The system obtains the warning accuracy, false alarm rate, and false alarm rate for each test scenario; it obtains the alarm time difference based on the warning light status and whether the alarm signal is issued for each frame of the test vehicle during the test; it obtains the effective warning distance and enhanced distance collision time based on the motion status information of the test vehicle and the target vehicle ahead; it obtains the comprehensive warning performance index of the test vehicle in each scenario based on the warning accuracy, false alarm rate, false alarm rate, alarm time difference, effective warning distance, and enhanced distance collision time for each test scenario, improving the accuracy of the quantitative analysis of warning performance; based on the comprehensive warning performance index, it determines the safety boundary of the test vehicle in each scenario in rain and fog environment, establishes a complete evaluation system from basic indicators to comprehensive performance index, and realizes the scientific quantification of safety boundary; it solves the problem that existing evaluation methods cannot accurately assess the impact of severe weather such as rain and fog on the performance of forward collision warning systems for passenger vehicles. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This invention provides a flowchart illustrating the steps of a method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions. Figure 2 This is a schematic diagram of the system performance safety boundary for a test scenario where the target vehicle ahead is stationary. Figure 3 This is a schematic diagram of the system performance safety boundary for a test scenario where the target vehicle ahead is moving at a constant speed. Figure 4A schematic diagram of the system performance safety boundary corresponding to the test scenario of decelerating the target vehicle ahead. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] To address the problems existing in the background technology, a method for evaluating the forward collision warning performance of passenger vehicles in rain and fog environments has been developed, which has important practical significance.

[0029] like Figure 1 As shown, the first aspect of the present invention is to provide a method for evaluating the forward collision warning performance of passenger vehicles in rain and fog conditions, comprising the following steps: Step S1: Obtain motion state information of the vehicle under test and the target vehicle ahead; obtain audio and video alarm information issued by the forward collision warning system of the vehicle under test during the test; obtain the alarm light status of the vehicle under test and whether an alarm signal is issued in each frame during the test through the audio and video alarm information.

[0030] Specifically, the motion status information of the tested vehicle and the target vehicle ahead is first collected. Advanced sensor technology and data acquisition systems are used to accurately collect this motion status information. This motion status information includes position, speed, and acceleration information. Position information can be precisely measured using a GPS system and onboard positioning devices; speed information is collected using wheel speed sensors to accurately reflect the vehicle's speed; and acceleration information is measured using accelerometers to monitor the vehicle's acceleration or deceleration in real time.

[0031] In this embodiment, the vehicle under test refers to a passenger car for which forward collision warning performance is evaluated, and the target vehicle ahead refers to the target closest to the vehicle under test on the driving trajectory line in front of the vehicle under test. It is the target of the forward collision warning system when it is working.

[0032] Secondly, the audio and video alarm information emitted by the forward collision warning system of the vehicle under test is collected, and then it is checked whether a valid warning is issued and the corresponding time is recorded; this includes the following 5 sub-steps: Step 1: Using the second pulse output by the BeiDou timing board as a reference, acquire information from the microphone and visual sensor through hardware synchronization, and affix a time tag with BeiDou time as a synchronization alignment identifier. The microphone is used to acquire audio alarm information from the forward collision warning system of the vehicle under test, and the visual sensor is used to acquire video alarm information from the forward collision warning system (audio alarm information is the warning or voice prompt issued by the forward collision warning system, and video alarm information is the visual warning displayed on the screen by the forward collision warning system, such as flashing icons, red collision symbols, etc.). Step 2: Acquire the image information of the m-th frame through a visual sensor and perform feature extraction to obtain image information features; in parallel, acquire the audio information of the m-th frame through a microphone, perform feature extraction to obtain audio information features; Step 3: Based on the extracted image and audio information features, use a pattern recognition algorithm to detect whether the alarm light is on and whether an alarm signal has been issued; Step 4: When the alarm light is detected to be on, or when the forward collision warning system issues an alarm signal, this is considered a valid alarm. Record the current frame time as the time when the system issued the collision alarm. Step 5: Repeat steps 2 to 4 to process the image and audio information of each subsequent frame, and determine whether the alarm light is on and whether the forward collision warning system issues an alarm signal, until the test ends.

[0033] At this point, the status of the warning lights and whether an alarm signal was issued for each frame of the test vehicle were determined.

[0034] Step S2: Set up three test scenarios for vehicles in rain and fog environments to conduct the test; the test vehicle performs several test processes in each rain and fog environment, and obtains the number of times a collision hazard occurs, the number of times no collision hazard occurs, the number of times a collision hazard occurs and an alarm is triggered, the number of times a collision hazard occurs but no alarm is triggered, and the number of times an alarm is triggered even though no collision hazard occurs.

[0035] Specifically, (1) Test scenario where the target vehicle in front is stationary.

[0036] The tested vehicle and the target vehicle ahead were located in a closed test area with artificially simulated rain and fog capabilities, with a horizontal visibility distance of [missing information]. Rainfall intensity is The tested vehicle and the target vehicle ahead are in the same lane, and the target vehicle is on the trajectory line in front of the tested vehicle. The tested vehicle is driven at the test speed. The vehicle moves forward and gradually approaches a stationary target vehicle ahead. During the test, the collision time between the tested vehicle and the target vehicle is equal to... During the test, record the number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. Also record the time when a collision hazard occurred and the time when the system issued a collision alarm.

[0037] (2) Test scenario of the target vehicle in front traveling at a constant speed.

[0038] The tested vehicle and the target vehicle ahead were located in a closed test area with artificially simulated rain and fog capabilities, with a horizontal visibility distance of [missing information]. Rainfall intensity is The tested vehicle and the target vehicle ahead are traveling in the same direction along the center line of their respective lanes. The tested vehicle is traveling at the test speed. Drive and gradually approach at 20 A target vehicle moving at a constant speed ahead. During the test, when the collision time between the tested vehicle and the target vehicle is equal to... During the test, record the number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. Also record the time when a collision hazard occurred and the time when the system issued a collision alarm.

[0039] (3) Test scenario of the target vehicle slowing down.

[0040] The tested vehicle and the target vehicle ahead were located in a closed test area with artificially simulated rain and fog capabilities, with a horizontal visibility distance of [missing information]. Rainfall intensity is The tested vehicle and the target vehicle ahead are traveling in the same direction along the center line of their respective lanes. Both the tested vehicle and the target vehicle are traveling at the test speed. Driving, the target vehicle ahead is Braking is achieved by decelerating the vehicle. The collision time between the tested vehicle and the target vehicle ahead is equal to... When the test begins, record the number of times a collision hazard occurs, the number of times no collision hazard occurs, the number of times a collision hazard occurs and an alarm is triggered, the number of times a collision hazard occurs but no alarm is triggered, and the number of times an alarm is triggered even though no collision hazard occurs. Also record the time when a collision hazard occurs and the time when the system issues a collision alarm.

[0041] In this embodiment, the horizontal visibility distance... The value ranges from [500, 1000], and the unit is m (meters). Rainfall intensity The value ranges from [2, 20], and the unit is mm / h (millimeters per hour). This represents the test speed. The value ranges from [20, 90], and the unit is km / h.

[0042] In this embodiment, the time when the forward collision warning system of the tested vehicle issues a collision alarm is obtained through step S001. The time when a collision hazard occurs during the test is calculated as the ratio between the relative distance and the relative speed between the tested vehicle and the target vehicle ahead (wherein, the relative speed is the speed of the tested vehicle minus the speed of the target vehicle ahead).

[0043] When the collision time between the tested vehicle and the target vehicle ahead is equal to If the time between the collision and the target vehicle in front is less than 100 seconds, it is considered a collision hazard situation; while if the time between the collision and the target vehicle in front is less than 100 seconds during a test, it is considered a collision hazard situation. If a collision does not occur, it is considered a situation where no danger of collision has been detected. This is directly quoted from the Chinese national standard GB / T 33577-2017. This standard provides a unified testing benchmark, based on engineering calculations of driver reaction time and vehicle braking distance, and specifies that... =4.4 seconds is used as the uniform starting time for triggering the forward collision warning test to ensure the consistency and comparability of the evaluation.

[0044] When a collision hazard occurs and a valid alarm is triggered, it is recorded as the number of times a collision hazard occurs and an alarm is triggered; when a collision hazard occurs but no valid alarm is triggered, it is recorded as the number of times a collision hazard occurs but no alarm is triggered; when a collision hazard does not occur but a valid alarm is triggered, it is recorded as the number of times a collision hazard does not occur but an alarm is triggered.

[0045] Thus, we obtain the time when a collision alarm is issued, the collision time when a collision alarm is issued, the number of collision warnings issued, and the number of times no warning is issued in the performance test scenario.

[0046] Step S3: Based on the number of collision hazards that occurred, the number of times no collision hazards occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred, obtain the warning accuracy, false alarm rate, and false alarm rate for each test scenario; based on the status of the warning lights and whether an alarm signal was issued by the vehicle under test in each frame during the test, obtain the alarm time difference; based on the motion state information of the vehicle under test and the target vehicle ahead, obtain the effective warning distance and enhanced distance collision time; based on the warning accuracy, false alarm rate, false alarm rate, alarm time difference, effective warning distance, and enhanced distance collision time for each test scenario, obtain the comprehensive warning performance index of the vehicle under test in each scenario; based on the comprehensive warning performance index, determine the safety boundary of the vehicle under test in rain and fog environments in each scenario.

[0047] Specifically, in order to comprehensively and scientifically evaluate the forward collision warning function and performance of the tested vehicles in rain and fog environments, multi-dimensional evaluation indicators are proposed, including: (1) Early warning accuracy;

[0048] In the formula, This indicates the number of times a collision hazard occurred and an alarm was triggered during the test. This indicates the accuracy rate of the warnings in each scenario test. This indicates the number of times a collision hazard occurred during the test.

[0049] (2) Underreporting rate;

[0050] In the formula, This indicates the number of times a collision hazard occurred during the test but no alarm was triggered. This indicates the number of times a collision hazard occurred during the test. This represents the false negative rate for each test scenario.

[0051] (3) False alarm rate;

[0052] In the formula, This indicates the number of times an alarm was triggered during the test when no collision hazard occurred. This represents the false positive rate for each test scenario. This indicates the number of times no collision hazards occurred during the test.

[0053] (4) Alarm time difference;

[0054] In the formula, This indicates the alarm time difference (i.e., the time difference between the time when a collision hazard occurs and the time when the system issues a collision alarm). Indicates the time when the system issues a collision alarm. This indicates the time during which a collision hazard occurred during the test. It is the absolute value symbol.

[0055] It should be noted that a collision hazard occurs during the test when the time between the tested vehicle and the target vehicle ahead is equal to... When this occurs, it constitutes a collision hazard.

[0056] (5) Effective warning distance;

[0057] In the formula, This represents the horizontal coordinate value of the vehicle being measured. This represents the vertical coordinate value of the vehicle being measured. This represents the horizontal coordinate value of the target vehicle. Represents the vertical coordinate value of the target vehicle. Indicates the effective warning distance (the straight-line distance between the tested vehicle and the target vehicle ahead at the instant the forward collision warning system issues a collision warning).

[0058] (6) Enhance distance collision time;

[0059] In the formula, Indicates the speed of the vehicle being measured. This indicates the acceleration of the vehicle being tested. Indicates the speed of the target vehicle ahead. This indicates the acceleration of the target vehicle ahead. The enhanced distance collision time (refers to the time elapsed from the moment the forward collision warning system issues a warning to the theoretical critical point of collision between the two vehicles, taking into account the relative speed and relative acceleration of the two vehicles).

[0060] It should be noted that, in order to analyze the warning performance of the tested vehicle under different rain and fog conditions, the warning performance under different combinations of horizontal visibility distance and rainfall intensity can be compared, thereby determining the safety boundary of the tested vehicle under different combinations of horizontal visibility distance and rainfall intensity.

[0061] The following analysis examines rain and fog conditions corresponding to a combination of horizontal visibility distance and rainfall intensity: Based on the warning accuracy, false alarm rate, and false alarm rate, the first warning performance index of the tested vehicle in each scenario is obtained. The first warning performance index of the tested vehicle in each scenario is specifically expressed by the following formula:

[0062] In the formula, This indicates the accuracy rate of the warnings in each scenario test. This represents the false negative rate in each test scenario. This represents the false positive rate for each test scenario. Indicates the first preset weight. This indicates the second preset weight. This indicates the third preset weight. This represents an exponential function with the natural constant as its base. This indicates the first warning performance index of the tested vehicle in each scenario.

[0063] In this embodiment , , In this embodiment, for , and No specific restrictions are imposed; implementers can decide based on the specific circumstances.

[0064] Based on the alarm time difference, effective warning distance, and enhanced distance collision time, the second warning performance index of the tested vehicle in each scenario is obtained. The second warning performance index of the tested vehicle in each scenario is specifically expressed by the following formula:

[0065] In the formula, This represents the average alarm time difference across several tests in each scenario. This represents the maximum alarm time difference across several tests in each scenario. This represents the average effective warning distance across several tests in each scenario. This represents the maximum effective warning distance across a number of tests in each scenario. This represents the average of the enhanced distance-collision time across several tests in each scenario. This represents the maximum value of the enhanced distance-collision time across a number of tests in each scenario. This indicates the fourth preset weight. This indicates the fifth preset weight. This indicates the sixth preset weight. This represents the second warning performance index of the tested vehicle in each scenario.

[0066] In this embodiment , , In this embodiment, for , and No specific restrictions are imposed; implementers can decide based on the specific circumstances.

[0067] It should be noted that the smaller the alarm time difference, the faster the system reacts, and the better the warning performance; the greater the effective warning distance, the better the warning performance, and the longer the available distance for the driver; the greater the enhanced distance collision time, the longer the available time for the driver, and the better (greater) the warning performance.

[0068] Based on the first and second warning performance indices tested by the vehicle in each scenario, the comprehensive warning performance index of the vehicle in each scenario is obtained; the comprehensive warning performance index of the vehicle in each scenario is specifically expressed by the formula:

[0069] In the formula, This represents the first warning performance index of the tested vehicle in each scenario. This represents the second warning performance index of the tested vehicle in each scenario. This represents the comprehensive warning performance index of the tested vehicle in each scenario.

[0070] Thus, a comprehensive warning performance index was obtained for the tested vehicle in each scenario under a combination of horizontal visibility distance and rainfall intensity.

[0071] A three-dimensional early warning performance evaluation space is constructed with horizontal visibility as the x-axis, rainfall intensity as the y-axis, and the comprehensive early warning performance index as the z-axis. The comprehensive early warning performance index of the tested vehicle in each scenario is mapped onto the three-dimensional early warning performance evaluation space and then fitted to obtain the early warning performance surface for each scenario. Three thresholds are selected to extract the safety boundaries corresponding to the combinations of horizontal visibility distance and rainfall intensity (one threshold corresponds to one contour line). These safety boundaries are mapped onto a two-dimensional combination space (composed of horizontal visibility as the x-axis and rainfall intensity as the y-axis), yielding schematic diagrams of the system performance safety boundaries corresponding to the three thresholds in the three scenarios, as shown in the diagram. Figure 2 , Figure 3 as well as Figure 4 As shown. The three thresholds are: the first security threshold... Second safety threshold The third safety threshold Among them, the first safety threshold Second safety threshold and the third security threshold All are preset values. In this embodiment, the three thresholds are not specifically limited, and the implementer can determine them according to the specific situation.

[0072] in, Figure 2 This is a schematic diagram of the system performance safety boundary for a test scenario where the target vehicle ahead is stationary. Figure 3 This is a schematic diagram of the system performance safety boundary for a test scenario where the target vehicle ahead is moving at a constant speed. Figure 4 This is a schematic diagram of the system performance safety boundaries for a test scenario involving decelerating a target vehicle ahead. Figure 2 , Figure 3 as well as Figure 4 The three lines from top to bottom in the middle correspond to the first security threshold. Two safety thresholds and the third safety threshold .

[0073] This concludes the embodiment.

[0074] A second aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions.

[0075] A third aspect of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions.

[0076] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0077] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0078] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0079] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for evaluating the forward collision warning performance of passenger vehicles in rain and fog environments, characterized in that, include: Acquire motion status information of the tested vehicle and the target vehicle ahead; Acquire audio and video alarm information issued by the forward collision warning system of the vehicle under test during the test; By using audio and video alarm information, the status of the alarm lights and whether an alarm signal was issued for each frame of the vehicle under test during the test can be obtained. Three test scenarios for vehicles were set up in rain and fog conditions for testing. The test vehicles were tested several times in each rain and fog environment. The number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. Based on the number of collision hazards that occurred, the number of times no collision hazards occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred, the warning accuracy, false alarm rate, and false alarm rate for each test scenario are obtained. Based on the status of the warning lights and whether an alarm signal was issued for each frame of the test vehicle, the alarm time difference is obtained. Based on the motion status information of the test vehicle and the target vehicle ahead, the effective warning distance and enhanced distance collision time are obtained. Based on the warning accuracy, false alarm rate, false alarm rate, alarm time difference, effective warning distance, and enhanced distance collision time for each test scenario, the comprehensive warning performance index of the test vehicle in each scenario is obtained. Based on the comprehensive warning performance index, the safety boundary of the test vehicle in rain and fog environments in each scenario is determined.

2. The method for evaluating the forward collision warning performance of passenger vehicles in rain and fog environments according to claim 1, characterized in that, The audio and video alarm information issued by the forward collision warning system of the vehicle under test during the test is acquired. By using audio and video alarm information, the status of the alarm lights and whether an alarm signal was issued for each frame of the tested vehicle during the test are obtained, including: Step 1: Using the second pulse output by the Beidou timing board as a reference, information from the microphone and visual sensor is collected using hardware synchronization, and time tags with Beidou time as a reference are added as a synchronization alignment mark; among them, the microphone is used to collect audio alarm information issued by the passenger vehicle forward collision warning system, and the visual sensor is used to collect video alarm information issued by the passenger vehicle forward collision warning system. Step 2: Acquire the image information of the m-th frame through a visual sensor and perform feature extraction to obtain image information features; in parallel, acquire the audio information of the m-th frame through a microphone, perform feature extraction to obtain audio information features; Step 3: Based on the extracted image and audio information features, use a pattern recognition algorithm to detect whether the alarm light is on and whether an alarm signal has been issued; Step 4: When the alarm light is detected to be on, or when the forward collision warning system issues an alarm signal, this is considered a valid alarm. Record the current frame time as the time when the system issued the collision alarm. Step 5: Repeat steps 2 to 4 to process the image and audio information of each subsequent frame, and determine whether the alarm light is on and whether the forward collision warning system issues an alarm signal, until the test ends.

3. The method for evaluating the forward collision warning performance of passenger vehicles in rain and fog environments according to claim 2, characterized in that, The tested vehicle underwent several test cycles in each rain and fog environment to obtain the number of times a collision hazard occurred, the number of times no collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. This includes: When the collision time between the tested vehicle and the target vehicle ahead is equal to If the time between the collision and the target vehicle in front is less than 100 seconds, it is considered a collision hazard situation; while if the time between the collision and the target vehicle in front is less than 100 seconds during a test, it is considered a collision hazard situation. At that time, it is considered a situation where no collision risk occurred; among them, Values ​​set according to Chinese national standards; When a collision hazard occurs and a valid alarm is triggered, it is recorded as the number of times a collision hazard occurs and an alarm is triggered; when a collision hazard occurs but no valid alarm is triggered, it is recorded as the number of times a collision hazard occurs but no alarm is triggered; when a collision hazard does not occur but a valid alarm is triggered, it is recorded as the number of times a collision hazard does not occur but an alarm is triggered.

4. The method for evaluating the forward collision warning performance of passenger vehicles in rain and fog environments according to claim 1, characterized in that, The accuracy rate, false alarm rate, and false alarm rate for each test scenario are obtained based on the number of times a collision hazard occurred, the number of times a collision hazard occurred and an alarm was triggered, the number of times a collision hazard occurred but no alarm was triggered, and the number of times an alarm was triggered even though no collision hazard occurred. The accuracy rate of the early warning is specifically expressed by the formula: In the formula, This indicates the number of times a collision hazard occurred and an alarm was triggered during the test. This indicates the accuracy rate of the warnings in each scenario test. This indicates the number of times a collision hazard occurred during the test; The specific formula for the missed detection rate is as follows: In the formula, This indicates the number of times a collision hazard occurred during the test but no alarm was triggered. This represents the false negative rate for each test scenario. The false alarm rate is specifically expressed by the formula: In the formula, This indicates the number of times an alarm was triggered during the test when no collision hazard occurred. This represents the false positive rate for each test scenario. This indicates the number of times no collision hazards occurred during the test.

5. The method for evaluating the forward collision warning performance of passenger vehicles in rain and fog environments according to claim 3, characterized in that, The alarm time difference is obtained based on the alarm light status and whether an alarm signal is issued in each frame of the test vehicle during the test process; Based on the motion status information of the tested vehicle and the target vehicle ahead, the effective warning distance and enhanced collision time are obtained, including: The alarm time difference is specifically expressed by the formula: In the formula, Indicates the time difference of the alarm. Indicates the time when the system issues a collision alarm. This indicates the time during which a collision hazard occurred during the test. It is the absolute value symbol; The effective warning distance is specifically expressed by the formula: In the formula, This represents the horizontal coordinate value of the vehicle being measured. This represents the vertical coordinate value of the vehicle being measured. This represents the horizontal coordinate value of the target vehicle. Represents the vertical coordinate value of the target vehicle. Indicates the effective warning distance; The enhanced distance collision time is specifically expressed by the formula: In the formula, Indicates the speed of the vehicle being measured. This indicates the acceleration of the vehicle being tested. Indicates the speed of the target vehicle ahead. This indicates the acceleration of the target vehicle ahead. This indicates the time required to enhance the distance collision.

6. The method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions according to claim 1, characterized in that, The comprehensive warning performance index of the tested vehicle in each scenario is obtained based on the warning accuracy, false alarm rate, alarm time difference, effective warning distance, and enhanced distance collision time in each test scenario. Based on the comprehensive early warning performance index, the safety boundaries of the tested vehicle in rain and fog environments in each scenario are determined, including: Based on the warning accuracy, false alarm rate, and false alarm rate, the first warning performance index of the tested vehicle in each scenario is obtained; based on the alarm time difference, effective warning distance, and enhanced distance collision time, the second warning performance index of the tested vehicle in each scenario is obtained. Based on the first and second warning performance indices tested by the vehicle in each scenario, the comprehensive warning performance index of the passenger vehicle in each scenario is obtained. The comprehensive warning performance index is specifically expressed by the formula: In the formula, This represents the first warning performance index of the tested vehicle in each scenario. This represents the second warning performance index of the tested vehicle in each scenario. This represents the overall warning performance index of the tested vehicle in each scenario; A three-dimensional early warning performance evaluation space is constructed with horizontal visibility as the x-axis, rainfall intensity as the y-axis, and comprehensive early warning performance index as the z-axis. The comprehensive early warning performance index of the tested vehicle in each scenario is mapped onto the three-dimensional early warning performance evaluation space and then fitted to obtain the early warning performance surface for each scenario. Three thresholds are selected to extract the safety boundary corresponding to the combination of horizontal visibility distance and rainfall intensity. The safety boundary is mapped onto the two-dimensional combination space to obtain the schematic diagram of the system performance safety boundary corresponding to the three thresholds in the three scenarios. The two-dimensional combined space is a space composed of horizontal visibility as the x-axis and rainfall intensity as the y-axis.

7. The method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions according to claim 6, characterized in that, The first warning performance index of the tested vehicle in each scenario is obtained based on the warning accuracy rate, false negative rate, and false alarm rate, and is specifically expressed by the formula: In the formula, This indicates the accuracy rate of the warnings in each scenario test. This represents the false negative rate in each test scenario. This represents the false positive rate for each test scenario. Indicates the first preset weight. This indicates the second preset weight. This indicates the third preset weight. This represents an exponential function with the natural constant as its base. This indicates the first warning performance index of the tested vehicle in each scenario.

8. The method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions according to claim 6, characterized in that, The second warning performance index of the tested vehicle in each scenario is obtained based on the alarm time difference, effective warning distance, and enhanced distance collision time, and is specifically expressed by the formula: In the formula, This represents the average alarm time difference across several tests in each scenario. This represents the maximum alarm time difference across several tests in each scenario. This represents the average effective warning distance across several tests in each scenario. This represents the maximum effective warning distance across a number of tests in each scenario. This represents the average of the enhanced distance-collision time across several tests in each scenario. This represents the maximum value of the enhanced distance-collision time across a number of tests in each scenario. This indicates the fourth preset weight. This indicates the fifth preset weight. This indicates the sixth preset weight. This represents the second warning performance index of the tested vehicle in each scenario.

9. An electronic device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for evaluating the forward collision warning performance of a passenger vehicle in rain and fog conditions as described in any one of claims 1-8.

Images