Method and device for determining fault rate of sensor in dangerous accident scene
By introducing weather-influencing factors into hazardous accident scenarios of autonomous driving vehicles and analyzing the sensor failure rate, the problem of the performance of the sensing system in different weather environments in the prior art is solved, and driving safety is improved.
Patent Information
- Application Number
- CN202311113438.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-31
- Publication Date
- 2025-02-28
AI Technical Summary
The existing technology is difficult to truly restore the performance of the autonomous vehicle perception system in different weather environments, and it is impossible to accurately analyze the sensor failure rate, which affects driving safety.
By introducing weather-influencing factors into hazardous accident scenarios, obtain sensor perception data, conduct feature analysis, determine abnormal perception characteristics, and conduct weather impact experiments to analyze sensor failure rates.
It improves the accuracy of sensor failure rate analysis, provides more accurate perception system output data, and provides guarantees for subsequent driving safety.
Smart Images
Figure CN119538488A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of autonomous driving simulation technology, and in particular to a method and device for determining a sensor failure rate in a hazardous accident scenario. Background Art
[0002] At present, simulation tests for autonomous vehicles mainly focus on accident scenarios generated by the test vehicles, mainly to monitor the performance of the perception system in the vehicle under different accident scenarios.
[0003] However, in the existing technology, the autonomous driving data is usually analyzed based on the extreme weather environment given by a given weather model, which has a weak correlation with the perception of autonomous driving. That is, it is impossible to truly restore the performance of the vehicle perception system in different weather environments, and it is impossible to provide the correct perception system output data for the vehicle's subsequent decision-making algorithm in a targeted manner, and it is impossible to ensure subsequent driving safety. Summary of the invention
[0004] In view of this, the purpose of this application is to at least provide a method and device for determining the sensor failure rate in a hazardous accident scenario, by introducing weather factors into the hazardous accident scenario of the test vehicle to analyze the corresponding failure rate of the vehicle sensor, so as to provide guarantee for subsequent driving safety.
[0005] This application mainly includes the following aspects:
[0006] In a first aspect, an embodiment of the present application provides a method for determining a sensor failure rate in a hazardous accident scenario, the method comprising: obtaining a test data set corresponding to a plurality of test vehicles, the test data set comprising actual test data corresponding to each test vehicle; determining at least one target hazardous accident scenario based on the test data set, the target hazardous accident scenario referring to a hazardous accident scenario in which a target sensor mounted on the test vehicle is the cause of the accident; in each target hazardous accident scenario, performing the following processing: obtaining perception data generated by a target sensor on the test vehicle in the target hazardous accident scenario; performing feature analysis on the perception data to determine a plurality of abnormal perception features corresponding to the target sensor; for each abnormal perception feature, performing a weather impact experiment corresponding to the abnormal perception feature in the target hazardous accident scenario according to a preset correspondence table, obtaining test data under the weather impact experiment, the correspondence table indicating a mapping relationship between a sensor category, an abnormal perception feature, and a weather impact factor, and determining, for each weather impact experiment, the device failure rate of the target sensor in the target hazardous accident scenario in which the weather impact experiment is performed according to the test data corresponding to the weather impact experiment.
[0007] In one possible implementation, the actual test data includes autonomous driving data generated by a test vehicle in different test scenarios, the autonomous driving data includes an accident label, and the accident label is used to indicate whether a traffic accident occurs with the test vehicle in the test scenario, wherein at least one target hazardous accident scenario is determined in the following manner: for each test vehicle, based on the accident label of the test vehicle in each test scenario, multiple accident scenarios related to the test vehicle are extracted from multiple test scenarios; accident statistical analysis is performed on multiple accident scenarios corresponding to each test vehicle according to preset analysis dimensions to determine multiple hazardous accident scenarios; and based on the cause of the accident in each hazardous accident scenario, at least one target hazardous accident scenario is screened out from multiple hazardous accident scenarios.
[0008] In a possible implementation, the perception data includes multiple perception parameters, wherein multiple abnormal perception features corresponding to the target sensor in each target hazard accident scenario are determined in the following manner: for each perception parameter, a comparison result between the perception parameter and a parameter standard corresponding to the perception parameter is determined; based on the comparison result corresponding to each perception parameter, multiple abnormal perception features are determined, and the comparison result indicates whether the perception parameter meets the corresponding parameter standard, and the abnormal perception feature is a perception feature corresponding to the perception parameter that does not meet the corresponding parameter standard.
[0009] In one possible implementation, in each target hazard accident scenario, the test data generated by the weather impact experiment corresponding to each abnormal perception feature is determined in the following manner: traversing a preset correspondence table to determine at least one target weather impact factor corresponding to the abnormal perception feature and the sensor category corresponding to the target sensor; determining a target perception parameter related to the abnormal perception feature; for each target weather impact factor corresponding to the abnormal perception feature, performing the following processing: simulating the target hazard accident scenario, and adding a preset degree of interference clutter corresponding to the target weather impact factor to the target perception parameter in the formed simulation environment; and determining, in the simulation environment corresponding to the target hazard accident scenario, the test data generated by the target sensor under the preset degree of disturbance clutter corresponding to the target weather impact factor.
[0010] In one possible implementation, the test data includes a sensor power attenuation value and a test time; wherein, in each target hazard accident scenario, the equipment failure rate of the target sensor under each weather impact experiment is determined in the following manner: based on the sensor power attenuation value and the test time under the weather impact experiment, the failure rate of the target sensor's equipment to be processed under the weather impact experiment is determined; based on the failure rate of the equipment to be processed under the weather impact experiment, the equipment failure rate of the target sensor under the weather impact experiment is determined.
[0011] In one possible implementation, the failure rate of the target sensor under each weather impact experiment is determined in the following manner: during the test time, the sensor power attenuation value is monitored and recorded in real time; the duration of the sensor power attenuation value being greater than a preset threshold is counted; and the ratio of the duration to the test time is determined as the failure rate of the target sensor under the weather impact experiment.
[0012] In a possible implementation, based on the failure rate of the equipment to be processed under the weather impact experiment, the step of determining the equipment failure rate of the target sensor under the weather impact experiment includes: if the failure rate of the equipment to be processed is greater than the preset equipment failure rate lower limit corresponding to the target sensor and is less than the preset equipment failure rate upper limit, then the failure rate of the equipment to be processed is determined as the equipment failure rate corresponding to the target sensor; if the failure rate of the equipment to be processed is greater than the preset equipment failure rate upper limit and less than the preset equipment failure rate limit, then the equipment failure rate corresponding to the target sensor is determined by the preset equipment failure rate lower limit, the failure rate of the equipment to be processed and the preset equipment failure rate limit; if the failure rate of the equipment to be processed is greater than the preset equipment failure rate limit, then the equipment failure rate corresponding to the target sensor is determined to be 1.
[0013] In a possible implementation, if the failure rate of the device to be processed is greater than the preset device failure rate lower limit and less than the preset device failure rate limit, the device failure rate corresponding to the target sensor is determined by the following formula:
[0014] t=L+(1-L) / L max -L lim max ×(LL lim max )
[0015] In this formula, t represents the device failure rate corresponding to the target sensor, L represents the failure rate of the device to be processed, and L max Indicates the preset device failure rate limit corresponding to the target sensor, L lim max Indicates the preset upper limit of the device failure rate corresponding to the target sensor.
[0016] In a second aspect, an embodiment of the present application further provides a device for determining a sensor failure rate in a hazardous accident scenario, the device comprising: a first acquisition module for acquiring a test data set corresponding to a plurality of test vehicles, the test data set comprising actual test data corresponding to each test vehicle; a determination module for determining at least one target hazardous accident scenario based on the test data set, the target hazardous accident scenario being a hazardous accident scenario in which a target sensor mounted on the test vehicle is the cause of the accident; a second acquisition module for acquiring, in each target hazardous accident scenario, the perception data generated by the target sensor on the test vehicle in the target hazardous accident scenario; a feature analysis module for analyzing the perception data in each target hazardous accident scenario. Perform feature analysis to determine multiple abnormal perception features corresponding to the target sensor; a weather impact test module is used to, in each target hazard accident scenario, for each abnormal perception feature, according to a preset correspondence table, conduct a weather impact experiment corresponding to the abnormal perception feature in the target hazard accident scenario to obtain test data under the weather impact experiment, and the correspondence table indicates the mapping relationship between the sensor category, the abnormal perception feature and the weather influence factor; a failure rate determination module is used to, in each target hazard accident scenario, for each weather impact experiment, determine the equipment failure rate of the target sensor in the target hazard accident scenario where the weather impact experiment is conducted according to the test data corresponding to the weather impact experiment.
[0017] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the sensor failure determination method in a hazardous accident scenario in the above-mentioned first aspect or any possible implementation scheme of the first aspect.
[0018] The embodiment of the present application provides a method and device for determining the sensor failure rate in a hazardous accident scenario, including: in each target hazardous accident scenario: obtaining the perception data generated by the target sensor on the test vehicle in the target hazardous accident scenario; performing feature analysis on the perception data to determine multiple abnormal perception features; according to a preset correspondence table, respectively performing weather impact experiments corresponding to multiple abnormal perception features in the target hazardous accident scenario to obtain test data for each abnormal perception feature under the corresponding weather impact experiment; for each weather impact experiment, determining the equipment failure rate of the target sensor in the target hazardous accident scenario where the weather impact experiment is performed based on the test data corresponding to the weather impact experiment. The present application is implemented by introducing weather influences in the hazardous accident scenario of the test vehicle, comprehensively analyzing the equipment failure rate of the vehicle sensor in this situation, and improving the accuracy of the equipment failure rate analysis.
[0019] The benefits of this application are:
[0020] In order to avoid the occurrence of driving accidents to the greatest extent, this application obtains the perception data generated by the target sensor on the test vehicle in the target hazard accident scenario, introduces the corresponding weather environment factors into the subsequent test through table lookup, and obtains the failure rate of the target sensor in the target hazard accident scenario through experimental analysis. When the failure rate meets certain conditions, an alarm is issued in time to ensure driving safety.
[0021] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0023] Figure 1 A flow chart of a method for determining a sensor failure rate in a hazardous accident scenario provided by an embodiment of the present application is shown;
[0024] Figure 2 A schematic diagram of the structure of a device for determining a sensor failure rate in a hazardous accident scenario provided by an embodiment of the present application is shown;
[0025] Figure 3 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0026] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.
[0027] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.
[0028] In the prior art, when a vehicle is operated in a special hazardous scenario, such as operating under strong sunlight conditions, the braking system is not activated in time, causing the vehicle to collide with the body of the vehicle in front, which may cause the driver to be killed.
[0029] From the previous case, it is not sufficient to only analyze extreme weather. In fact, it was the strong light that encountered the white target vehicle in front. The visual sensor of the autonomous driving system failed to recognize the white vehicle in front and failed to activate the brake system in time, causing the vehicle to collide with the body of the vehicle in front. In other words, the visual perception sensor failure was also required to cause an accident. Therefore, the subsequent analysis of driving safety based solely on the weather environment will be inaccurate.
[0030] Based on this, the embodiment of the present application provides a method and device for determining the sensor failure rate in a hazardous accident scenario, which introduces weather factors into the hazardous accident scenario of the test vehicle to analyze the corresponding failure rate of the vehicle sensor, thereby providing a guarantee for subsequent driving safety, as follows:
[0031] See also Figure 1 , Figure 1 FIG. 1 is a flow chart showing a method for determining a sensor failure rate in a hazardous accident scenario provided by an embodiment of the present application. Figure 1 As shown, the method for determining the sensor failure rate in a hazardous accident scenario provided by an embodiment of the present application includes the following steps:
[0032] S100: Obtain test data sets corresponding to multiple test vehicles.
[0033] The test dataset includes the actual test data corresponding to each test vehicle.
[0034] S200: Determine at least one target hazard accident scenario based on a test data set.
[0035] A target hazardous accident scenario refers to a hazardous accident scenario in which the target sensor mounted on the test vehicle is the cause of the accident.
[0036] S300: Under each target hazard accident scenario, obtain perception data generated by a target sensor on the test vehicle under the target hazard accident scenario.
[0037] S400: In each target hazard accident scenario, feature analysis is performed on the perception data to determine multiple abnormal perception features corresponding to the target sensor.
[0038] The perception data includes multiple perception features generated by the target sensor.
[0039] S500. In each target hazard accident scenario, according to a preset correspondence table, weather impact experiments corresponding to multiple abnormal perception features are respectively performed in the target hazard accident scenario to obtain test data of each abnormal perception feature under the corresponding weather impact experiment.
[0040] The correspondence table indicates the mapping relationship between sensor categories, abnormal perception features, and weather influencing factors.
[0041] S600. In each target hazard accident scenario, for each weather impact experiment, according to the test data corresponding to the weather impact experiment, determine the equipment failure rate of the target sensor in the target hazard accident scenario in which the weather impact experiment is conducted.
[0042] In the above steps S100 to S600, the present application obtains actual test data corresponding to each test vehicle from the vehicle actual operation database carried by the intelligent connected vehicle test monitoring and operation platform, and then screens out at least one target hazard accident scenario corresponding to each test vehicle from the actual test data corresponding to each test vehicle, and then conducts a corresponding weather impact experiment in the target hazard accident scenario through the perception data generated by the target sensor in the target hazard accident scenario and the preset correspondence table, that is, the impact of weather factors on the target sensor is added to the target hazard accident scenario, so that the equipment failure rate of the target sensor in the target hazard accident scenario with the weather factors added can be determined, so that in the same target hazard accident scenario, the failure rate output of the target sensor under the influence of different weather factors can be quickly determined, the granularity of the sensor's equipment failure rate affected by the weather can be accurately determined, the accuracy of the equipment failure rate analysis can be improved, and the calculation accuracy of subsequent related links that rely on the sensor output data can be improved.
[0043] In step S100, the actual test data includes the autonomous driving data generated by the test vehicle in different test scenarios. The autonomous driving data includes but is not limited to at least one of the following items: an accident label, perception data corresponding to different sensors, vehicle speed, weather environment data, and obstacle information around the test vehicle. The accident label is used to indicate whether a traffic accident occurs to the test vehicle in the test scenario.
[0044] In a preferred embodiment, in step S200, at least one target hazardous accident scenario is determined by:
[0045] For each test vehicle, based on the accident label of the test vehicle in each test scenario, multiple accident scenarios related to the test vehicle are extracted from multiple test scenarios, and accident statistical analysis is performed on the multiple accident scenarios corresponding to each test vehicle according to preset analysis dimensions to determine multiple hazardous accident scenarios. Based on the cause of the accident in each hazardous accident scenario, at least one target hazardous accident scenario is screened out from the multiple hazardous accident scenarios.
[0046] Specifically, multiple test scenarios corresponding to each test vehicle are screened layer by layer. First, test scenarios in which no traffic accidents occurred are eliminated from the multiple test scenarios through accident labels, and the remaining test scenarios are determined as accident scenarios. Accident statistical analysis is performed on the multiple accident scenarios corresponding to each test vehicle according to preset analysis dimensions. The preset analysis dimensions here include the test vehicle before entering the intersection, in the intersection, leaving the intersection, in the emergency driving area, and in the unresponsive area.
[0047] For example, a statistical analysis of multiple accident scenarios was conducted with intersections and travel directions (left turns, right turns, and straight ahead) as preset analysis dimensions. The result showed that among the multiple accident scenarios, intersection accident scenarios accounted for 3,261 accidents, non-intersection accident scenarios accounted for 2,980 accidents, and intersection accident scenarios accounted for 52.25%. Furthermore, among the intersection accident scenarios, right turn accident scenarios accounted for 22%, straight ahead accident scenarios accounted for 38%, and left turn accident scenarios accounted for 40%. Therefore, it was concluded that the left turn accident scenarios accounted for a higher proportion. Therefore, the left turn accident scenario was determined to be a hazardous accident scenario obtained with intersections and travel directions (left turns, right turns, and straight ahead) as preset analysis dimensions.
[0048] Then, from the multiple hazardous accident scenarios determined, the cause of the accident in each hazardous accident scenario is analyzed (for example, a problem with a sensor causes a vehicle collision accident), and the target hazardous accident scenarios with the target sensor loaded on the test vehicle as the cause of the accident are screened out, thereby further narrowing the scope of accident scenarios for analysis of the target sensor. For the present application, each target hazardous accident scenario corresponds to a test vehicle.
[0049] In step S300, the perception data includes multiple perception parameters, which include the sensor's own attribute parameters and detection parameters, wherein the own attribute parameters include but are not limited to at least one of the following items: sampling frequency, resolution, number of wiring harnesses and FOV, and the detection parameters include but are not limited to at least one of the following items: target object point cloud, distance to the target object, target object motion state and current driving road state.
[0050] In a preferred embodiment, in step S400, a plurality of abnormal perception features corresponding to the target sensor in each target hazard accident scenario are determined in the following manner:
[0051] For each perception parameter, determine the comparison result between the perception parameter and the parameter standard corresponding to the perception parameter. According to the comparison result corresponding to each perception parameter, determine multiple abnormal perception features. The comparison result indicates whether the perception parameter meets the corresponding parameter standard. The abnormal perception feature is the perception feature corresponding to the perception parameter that does not meet the corresponding parameter standard.
[0052] In a specific implementation, if the target sensor is a lidar, taking the perception parameter corresponding to the lidar - the number of feature points in the point cloud as an example, if the number of feature points in the point cloud does not reach the standard point cloud number known by the parameter standard, the corresponding abnormal perception feature is determined to be missing point cloud.
[0053] In a preferred embodiment, in step S500, in each target hazard accident scenario, the test data generated by the weather impact experiment corresponding to each abnormal perception feature is determined in the following manner:
[0054] Traverse the preset correspondence table, determine at least one target weather influencing factor corresponding to the abnormal perception feature and the sensor category corresponding to the target sensor, determine the target perception parameter related to the abnormal perception feature, and for each target weather influencing factor corresponding to the abnormal perception feature, perform the following processing: simulate the target hazard accident scenario, and add a preset degree of interference clutter corresponding to the target weather influencing factor to the target perception parameter in the formed simulation environment; in the simulation environment corresponding to the target hazard accident scenario, determine the test data generated by the target sensor under the preset degree of disturbance clutter corresponding to the target weather influencing factor.
[0055] In an example, as shown in Table 1:
[0056]
[0057]
[0058] Table 1 is a correspondence table provided in an embodiment of the present application. If the target sensor is a lidar and the abnormal perception feature is large point cloud noise, then by looking up Table 1, at least one corresponding target weather influencing factor can be obtained, which is precipitation and air particles.
[0059] When the abnormal perception feature is large point cloud noise, the target perception parameter related to the large point cloud noise is the distance between the test vehicle and the target object detected by the lidar. For the abnormal perception feature, its corresponding target perception parameter is a perception parameter that has a computational relationship with the acquisition of the abnormal perception feature.
[0060] The distance between the test vehicle and the target object detected by the lidar is large, the transmission power attenuation is large, and the attenuation is sensitive to the distance and the influence of different weather conditions.
[0061] After determining at least one target weather influencing factor, a weather impact test is performed for each target weather influencing factor. Specifically, it is necessary to simulate the target hazard accident scenario first, and then perform corresponding weather impact tests in the simulated environment. The so-called weather impact test is to add a preset degree of interference clutter corresponding to the weather influencing factor under the target perception parameters in the simulated target hazard accident scenario. The input degree of the interference clutter can be pre-entered according to the observation task to be performed, and then the simulation environment will feed back the test data generated by the target sensor under the preset degree of disturbance clutter corresponding to the target weather influencing factor.
[0062] The test data includes the sensor power attenuation value and the test time.
[0063] In step S600, in each target hazard accident scenario, the device failure rate of the target sensor in each weather impact experiment is determined in the following manner:
[0064] According to the sensor power attenuation value and test time under the weather impact experiment, the failure rate of the target sensor's equipment to be processed under the weather impact experiment is determined. According to the failure rate of the equipment to be processed under the weather impact experiment, the equipment failure rate of the target sensor under the weather impact experiment is determined.
[0065] In another preferred embodiment, the failure rate of the target sensor under each weather impact experiment is determined by the following method:
[0066] During the test time, the sensor power attenuation value is monitored and recorded in real time, and the duration of the sensor power attenuation value greater than the preset threshold is counted. The ratio of the duration to the test time is determined as the failure rate of the target sensor under the weather influence experiment.
[0067] In one example, according to the failure rate of the device to be processed under the weather impact experiment, the step of determining the device failure rate of the target sensor under the weather impact experiment includes:
[0068] If the failure rate of the device to be processed is greater than the preset device failure rate lower limit corresponding to the target sensor and is less than the preset device failure rate upper limit, the failure rate of the device to be processed is determined as the device failure rate corresponding to the target sensor; if the failure rate of the device to be processed is greater than the preset device failure rate upper limit and less than the preset device failure rate limit, the device failure rate corresponding to the target sensor is determined by the preset device failure rate lower limit, the failure rate of the device to be processed and the preset device failure rate limit; if the failure rate of the device to be processed is greater than the preset device failure rate limit, the device failure rate corresponding to the target sensor is determined to be 1.
[0069] Specifically, the device failure rate of the target sensor under a weather impact experiment can be determined by the following function:
[0070]
[0071] In this function, t represents the device failure rate corresponding to the target sensor, L represents the failure rate of the device to be processed, and L max Indicates the preset device failure rate limit corresponding to the target sensor, L lim max Indicates the preset upper limit of the device failure rate corresponding to the target sensor, L lim min Indicates the preset lower limit of the device failure rate corresponding to the target sensor.
[0072] In this application, when the determined equipment failure rate is 1, an alarm needs to be issued and the automatic driving needs to be exited, and the test vehicle needs to stop running.
[0073] In order to avoid the occurrence of driving accidents to the greatest extent, the present application obtains the perception data generated by the target sensor on the test vehicle in the target hazard accident scenario, such as the sensor generating lens glare, and reads the image contrast at this time; by looking up the table, it is seen that the glare part of the abnormal perception feature of the visual perception sensor corresponds to the lighting condition, and then in subsequent test experiments, the effect of adding a strong light source is simulated, such as circular ghosting, and the introduction of different colors of mist is used to adjust the range of tonal values in the image, lower the image contrast, and form a lens glare phenomenon. Through analysis, the failure rate of the target sensor, i.e., the camera, in the target hazard accident scenario is obtained. When the failure rate meets certain conditions, an alarm is given in time to ensure driving safety.
[0074] Based on the same application concept, the embodiments of the present application also provide a sensor fault determination device in a hazardous accident scenario corresponding to the sensor fault determination method in a hazardous accident scenario provided in the above embodiments. Since the principle of solving the problem by the device in the embodiments of the present application is similar to the sensor fault determination method in a hazardous accident scenario in the above embodiments of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0075] See also Figure 2 , Figure 2 FIG. 1 is a schematic diagram showing a structure of a device for determining a sensor failure rate in a hazardous accident scenario provided by an embodiment of the present application. Figure 2 As shown, the device comprises:
[0076] The first acquisition module 700 is used to acquire a test data set corresponding to a plurality of test vehicles, where the test data set includes actual test data corresponding to each test vehicle.
[0077] The determination module 710 is used to determine at least one target hazardous accident scenario based on the test data set, where the target hazardous accident scenario refers to a hazardous accident scenario in which the target sensor mounted on the test vehicle is the cause of the accident.
[0078] The second acquisition module 720 is used to acquire, in each target hazard accident scenario, the perception data generated by the target sensor on the test vehicle in the target hazard accident scenario.
[0079] The feature analysis module 730 is used to perform feature analysis on the perception data in each target hazard accident scenario to determine multiple abnormal perception features corresponding to the target sensor.
[0080] The weather impact test module 740 is used to conduct a weather impact experiment corresponding to each abnormal perception feature in each target hazard accident scenario according to a preset correspondence table, and obtain test data under the weather impact experiment. The correspondence table indicates the mapping relationship between the sensor category, the abnormal perception feature and the weather influencing factors.
[0081] The failure rate determination module 750 is used to determine the equipment failure rate of the target sensor in the target hazard accident scenario where the weather impact experiment is being conducted, for each weather impact experiment, based on the test data corresponding to the weather impact experiment.
[0082] Preferably, the actual test data includes autonomous driving data generated by the test vehicle in different test scenarios, the autonomous driving data includes accident labels, and the accident labels are used to indicate whether a traffic accident occurs in the test vehicle in the test scenario, wherein the determination module 710 is also used to determine at least one target hazardous accident scenario in the following manner: for each test vehicle, based on the accident label of the test vehicle in each test scenario, extracting multiple accident scenarios related to the test vehicle from multiple test scenarios; performing accident statistical analysis on the multiple accident scenarios corresponding to each test vehicle according to preset analysis dimensions to determine multiple hazardous accident scenarios; and screening out at least one target hazardous accident scenario from multiple hazardous accident scenarios based on the cause of the accident in each hazardous accident scenario.
[0083] Preferably, the perception data includes multiple perception parameters, wherein the feature analysis module 730 is also used to determine multiple abnormal perception features corresponding to the target sensor in each target hazard accident scenario in the following manner: for each perception parameter, determine a comparison result between the perception parameter and a parameter standard corresponding to the perception parameter; based on the comparison result corresponding to each perception parameter, determine multiple abnormal perception features, the comparison result indicates whether the perception parameter meets the corresponding parameter standard, and the abnormal perception feature is a perception feature corresponding to the perception parameter that does not meet the corresponding parameter standard.
[0084] Preferably, in each target hazard accident scenario, the weather impact test module 740 is also used to determine the test data generated by the weather impact experiment corresponding to each abnormal perception feature in the following manner: traverse the preset correspondence table to determine at least one target weather impact factor corresponding to the abnormal perception feature and the sensor category corresponding to the target sensor; determine the target perception parameter related to the abnormal perception feature; for each target weather impact factor corresponding to the abnormal perception feature, perform the following processing: simulate the target hazard accident scenario, and add a preset degree of interference clutter corresponding to the target weather impact factor to the target perception parameter in the formed simulation environment; determine the test data generated by the target sensor under the preset degree of disturbance clutter corresponding to the target weather impact factor in the simulation environment corresponding to the target hazard accident scenario.
[0085] Preferably, the test data includes the sensor power attenuation value and the test time; wherein, in each target hazard accident scenario, the failure rate determination module 750 is also used to determine the equipment failure rate of the target sensor under each weather impact experiment in the following manner: according to the sensor power attenuation value and the test time under the weather impact experiment, determine the failure rate of the equipment to be processed of the target sensor under the weather impact experiment; according to the failure rate of the equipment to be processed under the weather impact experiment, determine the equipment failure rate of the target sensor under the weather impact experiment.
[0086] Preferably, the failure rate determination module 750 is also used to determine the failure rate of the target sensor in the equipment to be processed under each weather influence experiment in the following manner: during the test time, real-time monitoring and recording of the sensor power attenuation value; counting the duration of the sensor power attenuation value being greater than a preset threshold; and determining the ratio between the duration and the test time as the failure rate of the target sensor in the equipment to be processed under the weather influence experiment.
[0087] Preferably, the failure rate determination module 750 is also used for: if the failure rate of the device to be processed is greater than the preset device failure rate lower limit corresponding to the target sensor and is less than the preset device failure rate upper limit, then the failure rate of the device to be processed is determined as the device failure rate corresponding to the target sensor; if the failure rate of the device to be processed is greater than the preset device failure rate upper limit and less than the preset device failure rate limit, then the device failure rate corresponding to the target sensor is determined by the preset device failure rate lower limit, the failure rate of the device to be processed and the preset device failure rate limit; if the failure rate of the device to be processed is greater than the preset device failure rate limit, then the device failure rate corresponding to the target sensor is determined to be 1.
[0088] Preferably, if the failure rate of the device to be processed is greater than the preset device failure rate lower limit and less than the preset device failure rate limit, the failure rate determination module 750 is further used to determine the device failure rate corresponding to the target sensor by the following formula:
[0089] t=L+(1-L) / L max -L lim max ×(LL lim max
[0090] In this formula, t represents the device failure rate corresponding to the target sensor, L represents the failure rate of the device to be processed, and L max Indicates the preset device failure rate limit corresponding to the target sensor, L lim max Indicates the preset upper limit of the device failure rate corresponding to the target sensor.
[0091] Based on the same application idea, please refer to Figure 3 , Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown, wherein the electronic device 800 comprises: a processor 810, a memory 820 and a bus 830, wherein the memory 820 stores machine-readable instructions executable by the processor 810, and when the electronic device 800 is running, the processor 810 communicates with the memory 820 via the bus 830, and the machine-readable instructions are executed by the processor 810 when running, such as the steps of the method for determining the sensor failure rate in a hazardous accident scenario provided in any of the above embodiments.
[0092] Based on the same application concept, an embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for determining the sensor failure rate in a hazardous accident scenario provided in the above embodiment are executed.
[0093] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, 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 through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0094] 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.
[0095] In addition, each functional unit in each embodiment of the present application 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.
[0096] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.
[0097] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for determining the failure rate of a sensor in a hazardous accident scenario, characterized in that: The method comprises: Acquire a test data set corresponding to a plurality of test vehicles, wherein the test data set includes actual test data corresponding to each test vehicle; Determine at least one target hazardous accident scenario based on the test data set, where the target hazardous accident scenario refers to a hazardous accident scenario in which a target sensor mounted on the test vehicle is the cause of the accident; In each target hazard incident scenario, the following processing is performed: Acquire perception data generated by a target sensor on a test vehicle in a target hazard accident scenario; Performing feature analysis on the perception data to determine a plurality of abnormal perception features corresponding to the target sensor; For each abnormal perception feature, according to a preset correspondence table, a weather impact experiment corresponding to the abnormal perception feature is conducted under the target hazard accident scenario to obtain test data under the weather impact experiment, wherein the correspondence table indicates the mapping relationship between the sensor category, the abnormal perception feature, and the weather influencing factor. For each weather impact experiment, the equipment failure rate of the target sensor in the target hazard accident scenario of the weather impact experiment is determined according to the test data corresponding to the weather impact experiment.
2. The method according to claim 1, characterized in that The actual test data includes autonomous driving data generated by the test vehicle in different test scenarios, and the autonomous driving data includes an accident label, and the accident label is used to indicate whether a traffic accident occurs to the test vehicle in the test scenario. Wherein, at least one target hazard accident scenario is determined by: For each test vehicle, extracting multiple accident scenarios related to the test vehicle from multiple test scenarios according to the accident label of the test vehicle in each test scenario; Perform accident statistics analysis on multiple accident scenarios corresponding to each test vehicle according to preset analysis dimensions to determine multiple hazardous accident scenarios; At least one target hazardous accident scene is selected from the plurality of hazardous accident scenes according to the cause of the accident in each hazardous accident scene.
3. The method according to claim 1, characterized in that The perception data includes a plurality of perception parameters. Among them, multiple abnormal perception features corresponding to the target sensor in each target hazard accident scenario are determined in the following way: For each perception parameter, determining a comparison result between the perception parameter and a parameter standard corresponding to the perception parameter; According to the comparison result corresponding to each perception parameter, a plurality of abnormal perception features are determined, wherein the comparison result indicates whether the perception parameter meets the corresponding parameter standard, and the abnormal perception feature is a perception feature corresponding to the perception parameter that does not meet the corresponding parameter standard.
4. The method according to claim 1, characterized in that: Under each target hazard accident scenario, the test data generated by the weather impact experiment corresponding to each abnormal perception feature is determined in the following way: Traversing a preset correspondence table, determining at least one target weather influencing factor corresponding to the abnormal perception feature and the sensor category corresponding to the target sensor; determining a target perception parameter associated with the abnormal perception feature; For each target weather influencing factor corresponding to the abnormal perception feature, the following processing is performed: Simulating the target hazard accident scenario, and adding a preset degree of interference clutter corresponding to the target weather influencing factor to the target perception parameters in the formed simulation environment; In a simulation environment corresponding to the target hazardous accident scenario, test data generated by the target sensor under a preset degree of disturbance clutter corresponding to the target weather influencing factor is determined.
5. The method according to claim 4, characterized in that The test data includes sensor power attenuation value and test time; Wherein, in each target hazard accident scenario, the equipment failure rate of the target sensor in each weather impact experiment is determined by the following method: Determine the failure rate of the target sensor under the weather impact experiment according to the sensor power attenuation value and the test time; According to the failure rate of the equipment to be processed under the weather influence experiment, the equipment failure rate of the target sensor under the weather influence experiment is determined.
6. The method according to claim 5, characterized in that The failure rate of the target sensor under each weather impact experiment is determined by the following method: During the test time, real-time monitoring and recording of the sensor power attenuation value; Counting the duration that the power attenuation value of the sensor is greater than a preset threshold; The ratio between the duration and the test time is determined as the failure rate of the target sensor under the weather impact experiment.
7. The method according to claim 5, characterized in that The step of determining the device failure rate of the target sensor under the weather impact experiment according to the failure rate of the device to be processed under the weather impact experiment comprises: If the failure rate of the device to be processed is greater than the preset lower limit of the device failure rate corresponding to the target sensor and less than the preset upper limit of the device failure rate, the failure rate of the device to be processed is determined as the device failure rate corresponding to the target sensor; If the failure rate of the device to be processed is greater than the preset device failure rate upper limit and less than the preset device failure rate limit, the device failure rate corresponding to the target sensor is determined by the preset device failure rate lower limit, the failure rate of the device to be processed and the preset device failure rate limit; If the failure rate of the device to be processed is greater than the preset device failure rate limit, the device failure rate corresponding to the target sensor is determined to be 1.
8. The method according to claim 7, characterized in that If the failure rate of the device to be processed is greater than the preset device failure rate lower limit and less than the preset device failure rate limit, the device failure rate corresponding to the target sensor is determined by the following formula: t=L+(1-L) / L max -L limmax ×(LL limmax ) In this formula, t represents the device failure rate corresponding to the target sensor, L represents the failure rate of the device to be processed, and L max represents the preset device failure rate limit corresponding to the target sensor, L lim max Indicates the preset upper limit of the device failure rate corresponding to the target sensor.
9. A device for determining sensor failure rate in a hazardous accident scenario, characterized in that: The device comprises: A first acquisition module, used to acquire a test data set corresponding to a plurality of test vehicles, wherein the test data set includes actual test data corresponding to each test vehicle; A determination module, configured to determine at least one target hazardous accident scenario based on the test data set, wherein the target hazardous accident scenario refers to a hazardous accident scenario in which a target sensor mounted on the test vehicle is the cause of the accident; A second acquisition module is used to acquire, in each target hazard accident scenario, the perception data generated by the target sensor on the test vehicle in the target hazard accident scenario; A feature analysis module, used to perform feature analysis on the perception data in each target hazard accident scenario to determine a plurality of abnormal perception features corresponding to the target sensor; A weather impact test module is used to perform a weather impact experiment corresponding to each abnormal perception feature in each target hazard accident scenario according to a preset correspondence table, and obtain test data under the weather impact experiment, wherein the correspondence table indicates a mapping relationship between sensor categories, abnormal perception features, and weather influencing factors; The failure rate determination module is used to determine the equipment failure rate of the target sensor in the target hazard accident scenario where the weather impact experiment is being conducted, for each weather impact experiment, according to the test data corresponding to the weather impact experiment in each target hazard accident scenario.
10. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to perform the steps of the method for determining the sensor failure rate in a hazardous accident scenario as described in any one of claims 1 to 8.