Method and device for testing and evaluating performance of automobile blind area monitoring and early warning system

Abstract

The invention relates to an automobile blind area monitoring and early warning system test and performance evaluation method and device, and the method comprises the steps: carrying out the scene modeling of a preset test road through an evaluation system, and adding a to-be-tested vehicle and a detected target to a test road scene; visible areas and various blind areas relative to the evaluation system are identified by using the geometrical relationship between the to-be-detected vehicle and the target and the semantics of the target; obtaining a to-be-detected vehicle attitude, and obtaining current vehicle design blind area distribution according to the to-be-detected vehicle attitude; in the standard measurement range of the to-be-measured vehicle blind area monitoring and early warning system, counting the overlapping condition of the current vehicle design blind area distribution and each measured blind area and the alarm condition of the to-be-measured vehicle blind area monitoring and early warning system; and calculating missing report evaluation, false report evaluation, matching evaluation between the alarm degree and the risk value and matching evaluation between the alarm time and the overlapping time of the to-be-detected vehicle blind area monitoring and early warning system, and configuring weights according to evaluation requirements to carry out weighted integration on the evaluations to obtain overall evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of blind spot monitoring performance evaluation, and in particular to a method and device for testing and performance evaluation of a vehicle blind spot monitoring and early warning system. Background Art

[0002] Due to the large size of heavy-duty vehicles and the high position of the cab, there are inevitable blind spots in the design of the vehicle itself, and it is often difficult for the driver to directly observe all the conditions around the vehicle. If pedestrians, non-motor vehicles or small vehicles in these blind spots are not discovered in time, it is very easy to cause traffic accidents. Therefore, blind spot monitoring and early warning systems have emerged, aiming to assist drivers in perceiving targets in blind spots through technical means, giving prompts and avoiding dangers. At present, there are many heavy-duty vehicle blind spot monitoring and early warning systems on the market. Most of these systems use sensors such as radars and cameras for blind spot monitoring, and process sensor data through recognition algorithms to identify targets in blind spots. Corresponding evaluation systems for evaluating the performance of blind spot monitoring and early warning systems have emerged. In terms of testing and performance evaluation, the existing technical solutions still have the following deficiencies: many systems only evaluate the performance of automobile blind spot monitoring and early warning systems through simple scene simulations or road tests to statistically calculate the missed alarm rate and false alarm rate, and cannot evaluate the potential risks brought by missed alarms, nor can they evaluate whether the risk and alarm intensity match during the alarm process. Summary of the invention

[0003] In order to solve the above technical problems or at least partially solve the above technical problems, the present invention provides a method and device for testing and performance evaluation of a vehicle blind spot monitoring and warning system.

[0004] In a first aspect, the present invention provides a method for testing and evaluating the performance of a vehicle blind spot monitoring and warning system, comprising: pre-establishing a connection between a vehicle posture and a vehicle design blind spot distribution;

[0005] An evaluation system is set up on the vehicle to be tested. The evaluation system includes: evaluation cameras covering the front and both sides of the vehicle, evaluation radars, and a host computer for analysis. The evaluation system performs scene modeling on the test road and adds the vehicle to be tested and the detected targets to the test road scene.

[0006] The evaluation system uses the geometric relationship between the vehicle to be tested and the target and the semantics of the target to identify the visible area and various blind areas relative to the evaluation system, and tracks the dynamic changes of the visible area and various blind areas relative to the evaluation system as the vehicle and the target move;

[0007] The evaluation system obtains the posture of the vehicle to be tested, and obtains the distribution of the blind area of ​​the current vehicle design according to the posture of the vehicle to be tested;

[0008] Within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, the evaluation system counts the overlap between the current vehicle design blind spot distribution and each blind spot measured by the evaluation system, as well as the alarm status of the blind spot monitoring and warning system of the vehicle to be tested;

[0009] Based on the statistics of the current vehicle design blind spot distribution, the overlap of each blind spot measured by the evaluation system, and the alarm status of the blind spot monitoring and early warning system of the vehicle to be tested, the underreporting evaluation, false alarm evaluation, matching evaluation between the alarm degree and the risk value, and matching evaluation between the alarm time and the overlapping time of the blind spot monitoring and early warning system of the vehicle to be tested are determined, and weights are configured according to the evaluation requirements to weight and integrate each evaluation to obtain an overall evaluation.

[0010] Furthermore, the test road is pre-constructed, and the test road covers a straight line, a left-turn line, and a right-turn line for cars. Static obstacles are set on the test road, and dynamically moving vehicles and pedestrians are arranged on the test road. The dynamically moving vehicles and pedestrians form dynamic obstacles, forming a unified test environment.

[0011] Furthermore, the evaluation system detects targets in front of and on both sides of the vehicle through an evaluation camera and an evaluation radar, and the evaluation system recognizes the semantics of the targets detected by the evaluation system through an image-based semantic recognition algorithm, and uses the radar data and image data of the target to obtain the motion state and appearance of the target; the evaluation system obtains the state vector of the vehicle under test during operation, and the state vector includes: the position coordinates of the vehicle under test on the test road, the driving direction θ, the driving speed v, the length, width, and wheelbase of the vehicle under test;

[0012] The evaluation system places the vehicle under test and the detected target into a test road scene according to the position of the vehicle under test and the relative position between the vehicle under test and the target.

[0013] Furthermore, the evaluation system uses the geometric relationship between the vehicle under test and the objects on the road and the semantics of the objects to identify the visible area and various blind areas relative to the evaluation system, including:

[0014] In the test road scenario, first determine the local test road area without obstacles based on the effective detection range of the evaluation system;

[0015] Take the targets in the local test road area in the test road scene, obtain all the corner point coordinates of each target, traverse all its corner point pairs for each target, calculate the angle between the rays from the position of the vehicle to be tested to the corner point, find the two corner points with the largest angle, and use them as the two projection points of the target. The rays downstream of each projection point form the blind spot generated by the current target.

[0016] The change of variable blind spots is related to the dynamic and static states of the targets serving as the blind spot sources. The blind spots are divided into static source blind spots and dynamic source blind spots according to the dynamic and static states of the blind spot sources. The maximum distance of the blind spots in the direction perpendicular to the trajectory of the vehicle to be tested is determined, and the ability of the blind spots to accommodate various types of targets is determined according to the maximum distance to classify the blind spots.

[0017] Furthermore, within the standard measurement range of the blind spot monitoring and early warning system of the vehicle to be tested, the evaluation system counts the start time and end time of the overlapping process between each sub-self-designed blind spot in the current vehicle design blind spot distribution and each target; counts the position and speed of the target blind spot relative to the vehicle to be tested during the overlapping process; counts the target semantics of the target blind spot; counts the type evolution of the target blind spot from the start to the end of the overlap, and the type evolution of the main target blind spot caused by the change in size; and counts the alarm start time, alarm end time, and alarm degree of the blind spot monitoring and early warning system of the vehicle to be tested for each sub-self-designed blind spot.

[0018] Furthermore, the underreporting evaluation is based on the underreporting. The underreporting situation is as follows: within the standard measurement range of the blind spot monitoring and early warning system of the vehicle to be tested, if the evaluation system detects that any target blind spot overlaps with the current vehicle design blind spot distribution, and the blind spot monitoring and early warning system of the vehicle to be tested does not alarm; in the event of a underreporting, the collision risk is evaluated based on the motion state of the underreported target and the motion state of the vehicle to be tested, and the collision risk of possible unobserved targets in the target blind spot and the vehicle to be tested is evaluated based on the state change of the target blind spot of the underreported target and the motion state of the vehicle to be tested. The two collision risks are multiplied by the underreporting collision risk coefficient respectively, and the sum of the underreporting rate and the underreporting coefficient is multiplied to obtain the underreporting evaluation.

[0019] Furthermore, the false alarm evaluation is based on the generation of false alarms. The false alarm situation is: within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, if the evaluation system does not detect that any target blind spot overlaps with the current vehicle design blind spot distribution, and the blind spot monitoring and warning system of the vehicle to be tested alarms, in the case of a false alarm, the false alarm rate is multiplied by the false alarm coefficient to obtain the false alarm evaluation.

[0020] Furthermore, the matching evaluation method between the alarm degree and the risk value is as follows: the collision risk is evaluated according to the motion state of the prompted target and the motion state of the vehicle to be tested, and the collision risk between the possible unobserved target in the target blind spot and the vehicle to be tested is evaluated according to the state change of the target blind spot of the prompted target and the motion state of the vehicle to be tested. The two collision risks are multiplied by the collision risk coefficient to obtain a risk score, and the cross entropy between the risk score and the alarm degree is used as the matching evaluation between the alarm degree and the risk value.

[0021] Furthermore, the matching between the alarm time and the overlapping time is evaluated as the cross entropy of the alarm time and the overlapping time.

[0022] In a second aspect, the present invention provides a vehicle blind spot monitoring and warning system testing and performance evaluation device, comprising: at least one processing unit, the processing unit is connected to a storage unit and a collection unit via a bus unit, the storage unit stores a computer program, and when the computer program is executed by the processing unit, the vehicle blind spot monitoring and warning system testing and performance evaluation method is implemented.

[0023] In a third aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the vehicle blind spot monitoring and warning system testing and performance evaluation method is implemented.

[0024] The above technical solution provided by the embodiment of the present invention has the following advantages compared with the prior art:

[0025] The present invention pre-establishes a connection between a vehicle posture and a vehicle design blind spot distribution; an evaluation system is arranged on the vehicle to be tested, the evaluation system performs scene modeling on a test road, and adds the vehicle to be tested and a detected target to the test road scene; the evaluation system uses the geometric relationship between the vehicle to be tested and the target and the semantics of the target to identify the visible area and various blind spots relative to the evaluation system, and tracks the dynamic changes of the visible area and various blind spots relative to the evaluation system as the vehicle and the target move; the evaluation system obtains the posture of the vehicle to be tested, and obtains the distribution of the current vehicle design blind spots according to the posture of the vehicle to be tested; within the standard measurement range of the blind spot monitoring and early warning system of the vehicle to be tested, the evaluation system counts the overlapping situation of the current vehicle design blind spot distribution and each blind spot measured by the evaluation system and the alarm situation of the blind spot monitoring and early warning system of the vehicle to be tested; based on the statistical overlapping situation of the current vehicle design blind spot distribution and each blind spot measured by the evaluation system and the alarm situation of the blind spot monitoring and early warning system of the vehicle to be tested, the underreporting evaluation, the false alarm evaluation, the matching evaluation between the alarm degree and the risk value, and the matching evaluation between the alarm time and the overlapping time of the blind spot monitoring and early warning system of the vehicle to be tested are determined, and the weights are configured according to the evaluation requirements to weight and integrate each evaluation to obtain an overall evaluation. Among them, when there is a missed report, the collision risk is evaluated according to the motion state of the missed target and the motion state of the vehicle to be tested. According to the state change of the target blind spot of the missed target and the motion state of the vehicle to be tested, the collision risk of the possible unobserved target in the target blind spot and the vehicle to be tested is evaluated. The two collision risks are multiplied by the missed collision risk coefficient, and the missed report rate is multiplied by the sum of the missed report coefficients to obtain the missed report evaluation. The collision risk is evaluated according to the motion state of the prompted target and the motion state of the vehicle to be tested. According to the state change of the target blind spot of the prompted target and the motion state of the vehicle to be tested, the collision risk of the possible unobserved target in the target blind spot and the vehicle to be tested is evaluated. The two collision risks are multiplied by the collision risk coefficient to obtain the risk score, and the cross entropy between the risk score and the alarm level is used as the matching evaluation between the alarm level and the risk value. When evaluating, this application considers both the false alarm rate and the impact of the risk caused by false alarm on the shaping performance; when evaluating, this application determines the time sensitivity of the blind spot alarm system based on the matching between the alarm time and the actual overlap; when evaluating, this application evaluates the correct alarm part based on the matching between the alarm intensity and the direct and indirect risks brought by the correct alarm target. This allows for a more comprehensive evaluation of the automobile blind spot monitoring and warning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0028] Figure 1 A flowchart of a method for testing and evaluating the performance of a vehicle blind spot monitoring and warning system provided by an embodiment of the present invention;

[0029] Figure 2 A schematic diagram of the test road, target, design blind spot and target blind spot distribution provided by an embodiment of the present invention;

[0030] Figure 3 A flow chart of adding a vehicle to be tested and a detected target to a test road scene provided by an embodiment of the present invention;

[0031] Figure 4 A flow chart of an evaluation system provided by an embodiment of the present invention using the geometric relationship between the vehicle to be tested and the target on the road and the target semantics to identify the visible area and various blind spots relative to the evaluation system;

[0032] Figure 5 A schematic diagram of a vehicle blind spot monitoring and warning system testing and performance evaluation device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0035] Example 1

[0036] like Figure 1As shown, the technology of the present invention implements a method for testing and evaluating the performance of a vehicle blind spot monitoring and warning system, including:

[0037] S100, pre-establishing a connection between the vehicle posture and the distribution of the vehicle's designed blind area. Changes in the vehicle posture will affect the vehicle's designed blind area. For example, when the vehicle turns, one side of the vehicle body will bulge outward, changing the distribution of the vehicle's designed blind area.

[0038] S200, such as Figure 2 As shown, a test road is pre-constructed, and the test road covers a straight line, a left turn line, and a right turn line for cars. Static obstacles are set on the test road, and dynamically moving vehicles and pedestrians are arranged on the test road. The dynamically moving vehicles and pedestrians form dynamic obstacles, forming a unified test environment. Figure 2 The red area in the middle is the A-pillar blind spot, and the rectangular dotted line around the vehicle is the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested. In actual applications, the standard measurement range can be the area formed by other polygons. Figure 2 The middle cross-shaped area is a local test road area in the test road area determined by the effective detection range of the evaluation system. The black solid line connecting the local test road area and the target projection point cooperates with the local test road area to form a blind spot.

[0039] S300, setting up an evaluation system on the vehicle to be tested, the evaluation system including: evaluation cameras covering the front and both sides of the vehicle, evaluation radars and a host computer for analysis, the evaluation system performs scene modeling on the test road, and adds the vehicle to be tested and the detected targets to the test road scene.

[0040] Specifically, Figure 3 As shown, step S300 includes:

[0041] The evaluation system detects objects in front of and on both sides of the vehicle using an evaluation camera and an evaluation radar.

[0042] The evaluation system recognizes the semantics of the target detected by the evaluation system through an image-based semantic recognition algorithm, and uses the radar data and image data of the target to obtain the motion state and shape of the target.

[0043] The evaluation system obtains the state vector of the vehicle under test during operation, and the state vector includes: the position coordinates of the vehicle under test on the test road, the driving direction θ, the driving speed v, the length, width and wheelbase of the vehicle under test.

[0044] The evaluation system places the vehicle under test and the detected target into a test road scene according to the position of the vehicle under test and the relative position between the vehicle under test and the target.

[0045] S400, the evaluation system uses the geometric relationship between the vehicle to be tested and the target and the semantics of the target to identify the visible area and various blind spots relative to the evaluation system, and tracks the dynamic changes of the visible area and various blind spots relative to the evaluation system as the vehicle and the target move.

[0046] Among them, Figure 4 As shown, the evaluation system uses the geometric relationship between the vehicle under test and the target on the road and the target semantics to identify the visible area and various blind areas relative to the evaluation system, including:

[0047] In the test road scenario, the local test road area is first determined in an obstacle-free condition based on the effective detection range of the evaluation system. In this application, the area of ​​interest is the area within the test road range, so both the visible area and the blind spot belong to the test road area. The local test road area includes the standard measurement range of the blind spot monitoring and early warning system of the vehicle to be tested, and can obtain the situation of the target and the target blind spot within a larger range.

[0048] Take the targets in the local test road area in the test road scene and obtain the coordinates of all corner points of each target; for each target, traverse all its corner point pairs and calculate the angle between the rays from the position of the vehicle to be tested to the corner points; find the two corner points with the largest angle as the two projection points of the target, that is, observe the edge points of the target from the position of the vehicle to be tested, and the rays downstream of each projection point form the blind spot generated by the current target.

[0049] The change of the variable blind spot is related to the static or dynamic state of the target as the blind spot source. According to the static or dynamic state of the blind spot source, the blind spot is divided into a static source blind spot and a dynamic source blind spot. If the target is stationary relative to the ground, the blind spot generated by the target is a static source blind spot, and the state change of the static source blind spot is only related to the motion state of the vehicle to be tested. If the target moves relative to the ground, the blind spot generated by the target is a dynamic source blind spot, and the change of the dynamic source blind spot is related to the motion state of the vehicle to be tested and the target point.

[0050] In addition to the changes in blind spots, this application also focuses on the size structure of blind spots. The larger the blind spot, the more targets it can accommodate, the greater the uncertainty and risk it contains, and the risk mostly comes from the direction perpendicular to the trajectory of the vehicle to be tested. When the blind spot alarm system of the vehicle to be tested itself performs a blind spot alarm, the size structure of the blind spot of the missed target is different, so the risk assessment will also be different during the evaluation. Therefore, the maximum distance of the blind spot in the direction perpendicular to the trajectory of the vehicle to be tested is determined, and the ability of the blind spot to accommodate various types of targets is determined based on the maximum distance to classify the blind spots.

[0051] S500, the evaluation system obtains the posture of the vehicle to be tested, and obtains the distribution of the blind area of ​​the current vehicle design according to the posture of the vehicle to be tested.

[0052] S600: within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, the evaluation system counts the overlap between the current vehicle design blind spot distribution and each blind spot measured by the evaluation system and the alarm status of the blind spot monitoring and warning system of the vehicle to be tested.

[0053] In the specific implementation process, the blind spot warning system often alarms when the target is within a certain distance from the vehicle and is in the vehicle's designed blind spot to provide a reminder. Within the standard measurement range, there is a certain risk of collision between the vehicle and the target.

[0054] Within the standard measurement range of the blind spot monitoring and early warning system of the vehicle to be tested, the evaluation system counts the start time and end time of the overlapping process between each sub-self-designed blind spot in the current vehicle design blind spot distribution and each target; counts the position and speed of the target blind spot relative to the vehicle to be tested during the overlapping process; counts the target semantics of the target blind spot; counts the type evolution of the target blind spot from the start to the end of the overlap, and the type evolution caused by the change of the size of the main target blind spot. Count the alarm start time, alarm end time, and alarm degree of each sub-self-designed blind spot of the blind spot monitoring and early warning system of the vehicle to be tested.

[0055] S700, based on the statistical distribution of the current vehicle design blind spots, the overlap of each blind spot measured by the evaluation system, and the alarm status of the blind spot monitoring and early warning system of the vehicle to be tested, determine the underreporting evaluation, false alarm evaluation, matching evaluation between the alarm level and the risk value, and matching evaluation between the alarm time and the overlapping time of the blind spot monitoring and early warning system of the vehicle to be tested, and configure weights according to the evaluation requirements to weight and integrate each evaluation to obtain an overall evaluation.

[0056] Among them, the missed alarm evaluation is based on the missed alarm. The missed alarm situation is: within the standard measurement range of the blind spot monitoring and early warning system of the vehicle to be tested, if the evaluation system detects that any target blind spot overlaps with the current vehicle design blind spot distribution, and the blind spot monitoring and early warning system of the vehicle to be tested does not alarm. In the case of missed alarm, the collision risk is evaluated according to the motion state of the missed target and the motion state of the vehicle to be tested. According to the state change of the target blind spot of the missed target and the motion state of the vehicle to be tested, the collision risk between the possible unobserved target in the target blind spot and the vehicle to be tested is evaluated. The collision risk of the two is multiplied by the missed collision risk coefficient, and the missed alarm rate is multiplied by the sum of the missed alarm coefficients to obtain the missed alarm evaluation. The possible unobserved targets are selected according to the capacity of the target blind spot. For example, the targets that may be included in the target blind spot with large capacity are vehicles, electric donkeys, motorcycles, pedestrians, etc., the targets that may be included in the target blind spot with medium capacity are electric bicycles, motorcycles, pedestrians, etc., and the targets that may be included in the target blind spot with small capacity are pedestrians, etc. Within the possible speed range of the unobserved target, the collision risk is calculated according to the maximum collision risk path.

[0057] The false alarm evaluation is based on the generation of false alarms. The false alarm situation is: within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, if the evaluation system does not detect any target blind spot overlapping with the current vehicle design blind spot distribution, and the blind spot monitoring and warning system of the vehicle to be tested alarms. In the case of a false alarm, the false alarm rate is multiplied by the false alarm coefficient to obtain the false alarm evaluation.

[0058] The matching evaluation method between the alarm level and the risk value is as follows: the collision risk is evaluated according to the motion state of the prompted target and the motion state of the vehicle to be tested; the collision risk between the possible unobserved target in the target blind spot and the vehicle to be tested is evaluated according to the state change of the target blind spot of the prompted target and the motion state of the vehicle to be tested; the risk of the two collision risks are multiplied by the collision risk coefficient to obtain the risk score; the cross entropy between the risk score and the alarm level is used as the matching evaluation between the alarm level and the risk value.

[0059] The matching between the alarm time and the overlapping time is evaluated as the cross entropy of the alarm time and the overlapping time.

[0060] In the embodiment, the collision risk is calculated by using a risk field based on trajectory prediction in the prior art.

[0061] Example 2

[0062] See also Figure 5 As shown, an embodiment of the present invention provides a vehicle blind spot monitoring and warning system test and performance evaluation device, including: at least one processing unit, the processing unit is connected to a storage unit through a bus unit, the storage unit is a computer-readable storage medium, and can be used to store software programs, computer executable programs and modules, such as the software programs, computer executable programs and modules corresponding to a vehicle blind spot monitoring and warning system test and performance evaluation method in an embodiment of the present invention. The processing unit implements the above-mentioned vehicle blind spot monitoring and warning system test and performance evaluation method by running the software programs, computer executable programs and modules stored in the storage unit, including:

[0063] Establish the connection between vehicle posture and vehicle design blind spot distribution in advance;

[0064] An evaluation system is set up on the vehicle to be tested. The evaluation system includes: evaluation cameras covering the front and both sides of the vehicle, evaluation radars, and a host computer for analysis. The evaluation system performs scene modeling on the test road and adds the vehicle to be tested and the detected targets to the test road scene.

[0065] The evaluation system uses the geometric relationship between the vehicle to be tested and the target and the semantics of the target to identify the visible area and various blind areas relative to the evaluation system, and tracks the dynamic changes of the visible area and various blind areas relative to the evaluation system as the vehicle and the target move;

[0066] The evaluation system obtains the posture of the vehicle to be tested, and obtains the distribution of the blind area of ​​the current vehicle design according to the posture of the vehicle to be tested;

[0067] Within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, the evaluation system counts the overlap between the current vehicle design blind spot distribution and each blind spot measured by the evaluation system, as well as the alarm status of the blind spot monitoring and warning system of the vehicle to be tested;

[0068] Based on the statistics of the current vehicle design blind spot distribution, the overlap of each blind spot measured by the evaluation system, and the alarm status of the blind spot monitoring and early warning system of the vehicle to be tested, the underreporting evaluation, false alarm evaluation, matching evaluation between the alarm degree and the risk value, and matching evaluation between the alarm time and the overlapping time of the blind spot monitoring and early warning system of the vehicle to be tested are determined, and weights are configured according to the evaluation requirements to weight and integrate each evaluation to obtain an overall evaluation.

[0069] Of course, the computer program stored in the storage unit of the vehicle blind spot monitoring and warning system testing and performance evaluation device provided in an embodiment of the present invention is not limited to the method operations described above, and can also execute related operations in the vehicle blind spot monitoring and warning system testing and performance evaluation method provided in any embodiment of the present invention.

[0070] Example 3

[0071] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program. When the computer program is executed, the vehicle blind spot monitoring and warning system testing and performance evaluation method is implemented, including:

[0072] Establish the connection between vehicle posture and vehicle design blind spot distribution in advance;

[0073] An evaluation system is set up on the vehicle to be tested. The evaluation system includes: evaluation cameras covering the front and both sides of the vehicle, evaluation radars, and a host computer for analysis. The evaluation system performs scene modeling on the test road and adds the vehicle to be tested and the detected targets to the test road scene.

[0074] The evaluation system uses the geometric relationship between the vehicle to be tested and the target and the semantics of the target to identify the visible area and various blind areas relative to the evaluation system, and tracks the dynamic changes of the visible area and various blind areas relative to the evaluation system as the vehicle and the target move;

[0075] The evaluation system obtains the posture of the vehicle to be tested, and obtains the distribution of the blind area of ​​the current vehicle design according to the posture of the vehicle to be tested;

[0076] Within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, the evaluation system counts the overlap between the current vehicle design blind spot distribution and each blind spot measured by the evaluation system, as well as the alarm status of the blind spot monitoring and warning system of the vehicle to be tested;

[0077] Based on the statistics of the current vehicle design blind spot distribution, the overlap of each blind spot measured by the evaluation system, and the alarm status of the blind spot monitoring and early warning system of the vehicle to be tested, the underreporting evaluation, false alarm evaluation, matching evaluation between the alarm degree and the risk value, and matching evaluation between the alarm time and the overlapping time of the blind spot monitoring and early warning system of the vehicle to be tested are determined, and weights are configured according to the evaluation requirements to weight and integrate each evaluation to obtain an overall evaluation.

[0078] A computer-readable storage medium provided in an embodiment of the present invention stores a computer program which is not limited to the method operations described above, but can also execute related operations in a vehicle blind spot monitoring and warning system testing and performance evaluation method provided in any embodiment of the present invention.

[0079] In the embodiments provided by the present invention, it should be understood that the disclosed structures and methods can be implemented in other ways. For example, the structural embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, structures or units, which can be electrical, mechanical or other forms.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0082] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for testing and evaluating the performance of a vehicle blind spot monitoring and warning system, characterized in that: include: Establish the connection between vehicle posture and vehicle design blind spot distribution in advance; An evaluation system is set up on the vehicle to be tested. The evaluation system includes: evaluation cameras covering the front and both sides of the vehicle, evaluation radars, and a host computer for analysis. The evaluation system performs scene modeling on the test road and adds the vehicle to be tested and the detected targets to the test road scene. The evaluation system uses the geometric relationship between the vehicle to be tested and the target and the semantics of the target to identify the visible area and various blind areas relative to the evaluation system, and tracks the dynamic changes of the visible area and various blind areas relative to the evaluation system as the vehicle and the target move; The evaluation system obtains the posture of the vehicle to be tested, and obtains the distribution of the blind area of ​​the current vehicle design according to the posture of the vehicle to be tested; Within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, the evaluation system counts the overlap between the current vehicle design blind spot distribution and each blind spot measured by the evaluation system, as well as the alarm status of the blind spot monitoring and warning system of the vehicle to be tested; Based on the statistics of the current vehicle design blind spot distribution, the overlap of each blind spot measured by the evaluation system, and the alarm status of the blind spot monitoring and early warning system of the vehicle to be tested, the underreporting evaluation, false alarm evaluation, matching evaluation between the alarm degree and the risk value, and matching evaluation between the alarm time and the overlapping time of the blind spot monitoring and early warning system of the vehicle to be tested are determined, and weights are configured according to the evaluation requirements to weight and integrate each evaluation to obtain an overall evaluation.

2. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 1 is characterized in that: The test road is pre-constructed, and the test road covers a straight line, a left-turn line, and a right-turn line for cars. Static obstacles are set on the test road, and dynamically moving vehicles and pedestrians are arranged on the test road. The dynamically moving vehicles and pedestrians form dynamic obstacles, forming a unified test environment.

3. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 1 is characterized in that: The evaluation system detects the targets in front and on both sides of the vehicle through the evaluation camera and the evaluation radar. The evaluation system recognizes the semantics of the targets detected by the evaluation system through the image-based semantic recognition algorithm, and uses the radar data and image data of the targets to obtain the motion state and appearance of the targets; the evaluation system obtains the state vector of the vehicle under test during operation, and the state vector includes: the position coordinates of the vehicle under test on the test road, the driving direction θ, the driving speed v, the length, width and wheelbase of the vehicle under test; The evaluation system places the vehicle under test and the detected target into a test road scene according to the position of the vehicle under test and the relative position between the vehicle under test and the target.

4. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 1 is characterized in that: The evaluation system uses the geometric relationship between the vehicle under test and the target on the road and the target semantics to identify the visible area and various blind areas relative to the evaluation system, including: In the test road scenario, first determine the local test road area without obstacles based on the effective detection range of the evaluation system; Take the targets in the local test road area in the test road scene, obtain all the corner point coordinates of each target, traverse all its corner point pairs for each target, calculate the angle between the rays from the position of the vehicle to be tested to the corner point, find the two corner points with the largest angle, and use them as the two projection points of the target. The rays downstream of each projection point form the blind spot generated by the current target. The change of variable blind spots is related to the dynamic and static states of the targets serving as the blind spot sources. The blind spots are divided into static source blind spots and dynamic source blind spots according to the dynamic and static states of the blind spot sources. The maximum distance of the blind spots in the direction perpendicular to the trajectory of the vehicle to be tested is determined, and the ability of the blind spots to accommodate various types of targets is determined according to the maximum distance to classify the blind spots.

5. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 1 is characterized in that: Within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, the evaluation system counts the start time and end time of the overlapping process of the blind spots formed by each sub-self-designed blind spot in the current vehicle design blind spot distribution and each target; During the overlapping process, the position and speed of the target blind spot relative to the vehicle to be tested are counted; the target semantics that form the target blind spot are counted; the type evolution of the target blind spot from the beginning to the end of the overlapping is counted, and the type evolution caused by the change of the size of the main target blind spot; Statistics are collected on the alarm start time, alarm end time and alarm degree of each blind spot designed by the blind spot monitoring and early warning system of the vehicle under test.

6. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 5 is characterized in that: The underreporting evaluation is based on underreporting. The underreporting situation is as follows: within the standard measurement range of the blind spot monitoring and early warning system of the vehicle to be tested, if the evaluation system detects that any target blind spot overlaps with the current vehicle design blind spot distribution, and the blind spot monitoring and early warning system of the vehicle to be tested does not alarm; in the event of a underreporting, the collision risk is evaluated based on the motion state of the underreported target and the motion state of the vehicle to be tested, and the collision risk between the possible unobserved target in the target blind spot and the vehicle to be tested is evaluated based on the state change of the target blind spot of the underreported target and the motion state of the vehicle to be tested. The two collision risks are multiplied by the underreporting collision risk coefficient respectively, and the sum of the underreporting rate and the underreporting coefficient is multiplied to obtain the underreporting evaluation.

7. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 5, characterized in that: The false alarm evaluation is based on the generation of false alarms. The false alarm situation is: within the standard measurement range of the blind spot monitoring and warning system of the vehicle to be tested, if the evaluation system does not detect that any target blind spot overlaps with the current vehicle design blind spot distribution, and the blind spot monitoring and warning system of the vehicle to be tested alarms, in the case of a false alarm, the false alarm rate is multiplied by the false alarm coefficient to obtain the false alarm evaluation.

8. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 1 is characterized in that: The matching evaluation method between the alarm level and the risk value is as follows: the collision risk is evaluated according to the motion state of the prompted target and the motion state of the vehicle to be tested; the collision risk between the possible unobserved target in the target blind spot and the vehicle to be tested is evaluated according to the state change of the target blind spot of the prompted target and the motion state of the vehicle to be tested; the risk of the two collision risks are multiplied by the collision risk coefficient to obtain the risk score; the cross entropy between the risk score and the alarm level is used as the matching evaluation between the alarm level and the risk value.

9. The vehicle blind spot monitoring and warning system testing and performance evaluation method according to claim 1, characterized in that: The matching between the alarm time and the overlapping time is evaluated as the cross entropy of the alarm time and the overlapping time.

10. A vehicle blind spot monitoring and warning system testing and performance evaluation device, characterized in that: include: At least one processing unit, the processing unit is connected to the storage unit and the acquisition unit through a bus unit, the storage unit stores a computer program, and when the computer program is executed by the processing unit, the automobile blind spot monitoring and warning system testing and performance evaluation method as described in any of claims 1-9 is implemented.

Images