Fault detection method, device, electronic device and medium for autonomous driving vehicle

Abstract

The present disclosure provides a fault detection method, device, electronic device and medium for an autonomous driving vehicle, and relates to the field of autonomous driving, and in particular to the field of fault detection in the field of autonomous driving. The specific implementation scheme is: a fault detection method for an autonomous driving vehicle, comprising: in the process of performing a virtual simulation test on an autonomous driving vehicle, when a target abnormality is detected in the autonomous driving vehicle, determining a candidate fault corresponding to the target abnormality in a preset data set, wherein the preset data set includes candidate faults corresponding to at least two preset abnormalities, the target abnormality is any one of the at least two preset abnormalities, and the candidate fault is a fault of a preset functional module in the autonomous driving vehicle; determining a target fault among the candidate faults corresponding to the target abnormality, wherein the target fault is a fault of a preset functional module that occurs during the virtual simulation test. The present disclosure can improve the efficiency of fault detection.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving, and in particular to the field of fault detection in the field of autonomous driving, and specifically to a method, device, electronic device and medium for fault detection of an autonomous driving vehicle. Background Art

[0002] During the virtual simulation test of the autonomous driving vehicle, when the autonomous driving vehicle has an abnormality, it is usually necessary to investigate the cause of the abnormality. In the related art, the method for investigating the cause of the abnormality is usually to investigate the abnormality of each module in the autonomous driving vehicle separately to determine the cause of the abnormality. Summary of the invention

[0003] The present disclosure provides a fault detection method, device, electronic device and medium for an autonomous driving vehicle.

[0004] According to a first aspect of the present disclosure, there is provided a fault detection method for an autonomous driving vehicle, comprising:

[0005] In the process of performing virtual simulation test on the autonomous driving vehicle, when a target abnormality is detected in the autonomous driving vehicle, a candidate fault corresponding to the target abnormality is determined in a preset data set, wherein the preset data set includes candidate faults corresponding to at least two preset abnormalities, the target abnormality is any one of the at least two preset abnormalities, and the candidate faults corresponding to the target abnormality are some candidate faults among the candidate faults included in the preset data set, and the candidate faults are faults of preset functional modules in the autonomous driving vehicle;

[0006] A target fault is determined from candidate faults corresponding to the target abnormality, wherein the target fault is a fault occurring in the preset functional module during the virtual simulation test.

[0007] According to a second aspect of the present disclosure, there is provided a fault detection device for an autonomous driving vehicle, comprising:

[0008] a first determining unit, configured to determine, in a preset data set, a candidate fault corresponding to a target abnormality detected in the autonomous driving vehicle during a virtual simulation test of the autonomous driving vehicle, wherein the preset data set includes candidate faults corresponding to at least two preset abnormalities, the target abnormality is any one of the at least two preset abnormalities, and the candidate faults corresponding to the target abnormality are some candidate faults among the candidate faults included in the preset data set, and the candidate faults are faults of preset functional modules in the autonomous driving vehicle;

[0009] The second determination unit is used to determine a target fault from candidate faults corresponding to the target abnormality, wherein the target fault is a fault occurring in the preset functional module during the virtual simulation test.

[0010] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0011] at least one processor; and

[0012] a memory communicatively connected to the at least one processor; wherein,

[0013] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in the first aspect above.

[0014] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described in the first aspect above.

[0015] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program implements the method described in the first aspect when executed by a processor.

[0016] In the disclosed embodiment, by pre-establishing a correspondence between a preset abnormality and a candidate fault, when a target abnormality is detected in the autonomous driving vehicle, the candidate fault corresponding to the target fault can be determined first, and then the target fault can be determined in the target fault to complete the fault detection process. In the disclosed embodiment, since only the candidate faults corresponding to the target fault need to be checked, and the candidate faults corresponding to the target fault are only some of the candidate faults included in the preset data set, the scope of fault troubleshooting can be effectively narrowed, thereby improving the efficiency of fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0018] Figure 1 is a flow chart of a fault detection method for an autonomous driving vehicle provided by an embodiment of the present disclosure;

[0019] Figure 2 is a structural schematic diagram of a fault detection device for an autonomous driving vehicle provided by an embodiment of the present disclosure;

[0020] Figure 3 is a schematic diagram of the structure of the second determination unit in an embodiment of the present disclosure;

[0021] Figure 4 A block diagram of an electronic device for implementing a fault detection method for an autonomous driving vehicle provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0023] See also Figure 1 , Figure 1 A schematic diagram of a flow chart of a fault detection method for an autonomous driving vehicle provided in an embodiment of the present disclosure, wherein the fault detection method for an autonomous driving vehicle comprises the following steps:

[0024] Step S101: during a virtual simulation test of the autonomous driving vehicle, if a target abnormality is detected in the autonomous driving vehicle, a candidate fault corresponding to the target abnormality is determined in a preset data set, wherein the preset data set includes candidate faults corresponding to at least two preset abnormalities, the target abnormality is any one of the at least two preset abnormalities, and the candidate faults corresponding to the target abnormality are some candidate faults among the candidate faults included in the preset data set, and the candidate faults are faults of preset functional modules in the autonomous driving vehicle;

[0025] Step S102: determining a target fault from candidate faults corresponding to the target abnormality, wherein the target fault is a fault occurring in the preset functional module during the virtual simulation test.

[0026] The above-mentioned virtual simulation test of the autonomous driving vehicle may specifically refer to: performing a virtual simulation test on the autonomous driving vehicle in various test scenarios. Among them, the test scenario may include various driving scenarios that may be encountered during the driving of the autonomous driving vehicle. For example, the test scenario may include driving scenarios such as pedestrians crossing the road, the front vehicle changing lanes, and normal following scenes. By performing simulation tests on the autonomous driving vehicle based on the above-mentioned test scenarios, it is possible to determine whether the corresponding functional modules in the autonomous driving vehicle can accurately identify the above-mentioned test scenarios and make corresponding responses to different test scenarios. For example, when the autonomous driving vehicle is subjected to a virtual simulation test based on the pedestrian crossing the road scenario, it is determined whether the autonomous driving vehicle can slow down or reasonably avoid the pedestrian, so that the subsequent autonomous driving vehicle can make corresponding responses to various driving scenarios.

[0027] Specifically, various traffic elements and traffic parameters in the test scene can be modeled in advance based on simulation software to obtain a virtual simulation world. Among them, modeling can be performed according to actual road traffic scenes, and actual attribute parameters can be assigned to each traffic element and traffic parameter value. In addition, a fictitious road traffic scene can also be created, and actual attribute parameters can be assigned to each traffic element and traffic parameter value.

[0028] It is understandable that after completing the above modeling process, corresponding attribute information can be assigned to each element in the constructed virtual simulation world based on preset simulation parameters, wherein the attribute information includes the position, speed, and direction of each traffic parameter value, and the status information of each traffic sign, etc. Then, virtual simulation is performed.

[0029] The preset data set may include the at least two preset anomalies and candidate faults corresponding to the preset anomalies, wherein the preset anomalies and the candidate faults corresponding to the preset anomalies may be stored in association. Thus, when determining that the target anomaly occurs in the autonomous driving vehicle, the candidate fault corresponding to the target anomaly may be determined in the preset data set based on the correspondence between the preset anomaly and the candidate faults.

[0030] The above-mentioned preset functional modules may be various functional modules in an autonomous driving vehicle, for example, may include a perception module, a positioning module, a map module, etc.

[0031] The above-mentioned preset anomalies may be various abnormal driving behaviors during the driving process of the autonomous driving vehicle. For example, the preset anomalies may include: driving in the wrong direction, running a red light, sudden braking, main responsibility collision, vehicle shaking, drawing a dragon, exiting the autonomous driving mode, etc. Correspondingly, the above-mentioned candidate faults may include possible faults of various functional modules in the autonomous driving vehicle. For example, the candidate faults may include: frame loss faults and perception error faults in the process of the perception module detecting the target object; positioning drift faults of the positioning module, etc.

[0032] It can be understood that the candidate faults corresponding to the above-mentioned preset abnormalities may include: all faults that can cause the automatic driving vehicle to have the preset abnormalities. For example, when the preset abnormality is the main responsible collision fault, for the perception module, the candidate faults corresponding to the main responsible collision fault may include: target frame loss, target delayed selection, target position error is too large, target type error and target is a virtual target and other faults. Among them, the target frame loss is a frame loss fault in the process of detecting the target object. The target delayed selection means that the time of locking the target in the process of detecting the target object is too late, the target position error is too large, the target type error is the error in identifying the type of the target object, and the target is a virtual target, that is, a non-existent target is identified as a physical target, for example, the reflection of a tree is identified as a vehicle.

[0033] It is understandable that, based on the detection means in the relevant technology, it is possible to detect whether the above-mentioned preset abnormality occurs in the autonomous driving vehicle during the virtual simulation test, and the preset abnormality that occurs in the autonomous driving vehicle can be determined as the target abnormality. For example, the driving direction of the autonomous driving vehicle can be detected, and the driving direction of the autonomous driving vehicle can be compared with the driving direction indicated by the lane where the autonomous driving vehicle is located. When it is detected that the driving direction of the autonomous driving vehicle is opposite to the driving direction indicated by the lane where the autonomous driving vehicle is located, it is determined that the autonomous driving vehicle has a reverse fault. In addition, since the specific time when the red light of the traffic indicator light at each location in the virtual simulation world turns on can be directly obtained from the simulation information, and the actual time when the autonomous driving vehicle enters and exits a certain traffic intersection can be obtained from the detection data of the autonomous driving vehicle, in this way, when it is detected that the time when the autonomous driving vehicle enters and exits a certain traffic intersection is within the time period when the red light of the traffic intersection turns on, it is determined that the autonomous driving vehicle has a red light running abnormality. When it is detected that the autonomous driving vehicle overlaps with the vehicle in front in the virtual simulation screen, it is determined that a collision abnormality occurs. At the same time, by detecting the position of the autonomous driving vehicle in multiple consecutive frames in the simulation picture, when the change value of the lateral position of the autonomous driving vehicle between adjacent pictures in the multiple consecutive frames exceeds the set threshold, it is determined whether the autonomous driving vehicle has a shaking fault or a dragon drawing fault. When the autonomous driving vehicle exits the autonomous driving mode, a corresponding signal to exit the autonomous driving mode is triggered. When the signal to exit the autonomous driving mode is obtained, it is determined that the autonomous driving vehicle has a fault to exit the autonomous driving mode.

[0034] Accordingly, when determining the candidate fault corresponding to the target abnormality, it is possible to detect whether the candidate fault occurs in the preset functional module in the autonomous driving vehicle during the virtual simulation test based on the detection means in the relevant technology, and the candidate fault output by the autonomous driving vehicle can be determined as the target fault.

[0035] In this implementation, by pre-establishing a correspondence between a preset abnormality and a candidate fault, when a target abnormality is detected in the autonomous driving vehicle, the candidate fault corresponding to the target fault can be determined first, and then the target fault can be determined in the target fault to complete the fault detection process. In the disclosed embodiment, since only the candidate faults corresponding to the target fault need to be checked, and the candidate faults corresponding to the target fault are only some of the candidate faults included in the preset data set, the scope of fault troubleshooting can be effectively narrowed, thereby improving the efficiency of fault detection.

[0036] Optionally, the candidate faults corresponding to the target abnormality include at least two first faults, and determining the target fault from the candidate faults corresponding to the target abnormality includes:

[0037] Acquire a detection parameter set and a simulation parameter set, wherein the detection parameter set includes detection attribute parameters of the target object detected by the autonomous driving vehicle during the virtual simulation test; the simulation parameter set includes true value attribute parameters of the target object during the virtual simulation test; the target object includes at least one of the following: the autonomous driving vehicle and a driving environment object other than the autonomous driving vehicle;

[0038] The target fault is determined among the at least two first faults based on the detection parameter set and the simulation parameter set.

[0039] The first fault may be a fault that may occur in various functional modules of the autonomous driving vehicle, for example, a frame loss fault or a perception error fault in the process of detecting a target object by a perception module, or a positioning drift fault in a positioning module.

[0040] During the virtual simulation test of the autonomous driving vehicle, the autonomous driving vehicle usually perceives the external driving environment information through the perception module, and then outputs the corresponding driving action based on the perceived external driving environment information. However, when the perceived information is wrong, it may cause the above-mentioned preset abnormality of the autonomous driving vehicle. And usually when specific detection data is wrong, it will lead to specific abnormal driving behavior. For example, when the detection of the vehicle in front is wrong, it may cause the autonomous driving vehicle to collide with the vehicle in front.

[0041] Therefore, in the disclosed embodiment, when it is necessary to determine whether a first fault occurs during a virtual simulation test of an autonomous driving vehicle, it is only necessary to determine whether the detection data that can cause the first fault during the virtual simulation test of the autonomous driving vehicle is abnormal, so as to determine whether the autonomous driving vehicle has the first fault during the virtual simulation test. Specifically, when the detection data that can cause the first fault is abnormal, it is determined that the autonomous driving vehicle has the first fault during the virtual simulation test.

[0042] Based on this, during the virtual simulation test of the autonomous driving vehicle, the information sensed by each module in the autonomous driving vehicle can be stored in the detection parameter set, so that when determining whether the autonomous driving vehicle has a first fault, the data corresponding to the first fault in the detection parameter set can be detected to determine whether the first fault has occurred. The information sensed by each module may include: the original information collected by each module, and may also include the information after each module processes the original information, that is, the detection parameter set may include the original information collected by each module, and the information after each module processes the original information.

[0043] Among them, since the autonomous driving vehicle usually needs to perceive its own state and the external driving environment during the virtual simulation test, the detection parameter set includes the detection attribute parameters of the autonomous driving vehicle itself and the detection attribute parameters of the driving environment objects outside the autonomous driving vehicle. The detection attribute parameters are the attribute values ​​detected when detecting the target object. For example, when the target object is the vehicle in front of the autonomous driving vehicle, the detection attribute parameters may include: the position, speed and other attribute information of the vehicle in front.

[0044] In addition, since the detection parameter set includes detection attribute parameters of the target object, in order to determine whether the data in the detection parameter set is abnormal, it is necessary to obtain the real attribute parameters of the target object. In this way, the parameters in the detection parameter set can be compared with the corresponding real attribute parameters in the simulation parameter set to determine whether the detection parameters are correct.

[0045] Specifically, since the real properties of each target object in the virtual world at the beginning of the simulation are usually manually input by relevant personnel during the virtual simulation test, the real properties of the target object can be directly obtained from the simulation. In this way, the simulation parameter set can be obtained from the real properties of the target object given in the simulation software.

[0046] In this implementation, when it is necessary to determine whether a first abnormality occurs in the autonomous driving vehicle during the virtual simulation test, the detection attribute parameters corresponding to the first abnormality can be compared with the corresponding real attribute parameters to determine whether the first abnormality occurs, thereby realizing the target abnormality determination process.

[0047] Optionally, the determining a target fault from the at least two first faults based on the detection parameter set and the simulation parameter set includes:

[0048] In the case where the first attribute parameter does not match the second attribute parameter, determining the first candidate fault as the target fault;

[0049] Among them, the first candidate fault is a fault among the at least two first faults, and the first candidate fault is used to characterize the abnormality of the target attribute of the preset functional module; the first attribute parameter is an attribute parameter in the detection parameter set, which is used to characterize the target attribute of the preset functional module; the second attribute parameter is an attribute parameter in the simulation parameter set, which is used to characterize the target attribute of the preset functional module.

[0050] The first candidate faults mentioned above may be various types of faults that may occur in various functional modules in the autonomous driving vehicle. For example, the first candidate fault may be any fault in the candidate faults corresponding to the perception module, that is, the first candidate fault may be any fault in "target frame loss, target delayed selection, target position error is too large, target type error and target is a virtual target".

[0051] The mismatch between the first attribute parameter and the second attribute parameter may mean that the first attribute parameter and the second attribute parameter are not equal, or that the difference between the first attribute parameter and the second attribute parameter exceeds a preset threshold.

[0052] In one embodiment of the present disclosure, the test scenario may be a scenario where the front vehicle changes lanes to the lane of the main vehicle, wherein the main vehicle is the above-mentioned autonomous driving vehicle. The first candidate fault may be a delayed target selection, the preset function module is a perception module, and the target attribute is the time when the perception module selects the target object. Since the front vehicle changes lanes and the main vehicle selects the front vehicle too late, it may cause the main vehicle to collide with the front vehicle, or the main vehicle may have abnormalities such as sudden braking, so the target abnormality may be a main vehicle collision or sudden braking.

[0053] Specifically, when the front vehicle starts to change lanes, the main vehicle will select the front vehicle and identify it. Therefore, in order to determine whether the main vehicle has a delayed selection fault, the first attribute parameter can be the time point when the main vehicle's perception module starts to select the front vehicle, and the first attribute parameter can be directly obtained from the detection attribute parameter. The second attribute parameter can be the time point when the front vehicle starts to change lanes, and the time point can be directly obtained from the simulation parameter set. In this way, a judgment threshold can be set in advance. When the time difference between the first attribute parameter and the second attribute parameter is greater than the judgment threshold, it is determined that the main vehicle has a delayed selection fault, and the delayed selection fault is determined as a target fault and output.

[0054] In this implementation, when determining whether a certain abnormality occurs in a preset functional module of the autonomous driving vehicle, since the abnormality specifically indicates that an abnormality occurs in a target attribute of a specific functional module in the autonomous driving vehicle, a first attribute parameter corresponding to the target attribute can be obtained from a detection parameter set, and a second attribute parameter corresponding to the target attribute can be obtained from a simulation parameter set. Then, it is determined whether the first attribute parameter matches the second attribute parameter to determine whether the autonomous driving vehicle has the abnormality during the virtual simulation test.

[0055] Optionally, the preset function module includes a perception module and a positioning module, and before determining the first candidate fault as the target fault, the method further includes:

[0056] In the case where the first candidate fault is a fault of the perception module, determining a matching state between the first attribute parameter and the second attribute parameter based on a first matching condition; in the case where the first candidate fault is a fault of the positioning module, determining a matching state between the first attribute parameter and the second attribute parameter based on a second matching condition;

[0057] The first matching condition is different from the second matching condition.

[0058] The first matching condition may include a judgment condition for determining whether each attribute parameter in the perception module is abnormal, and correspondingly, the second matching condition may include a judgment condition for determining whether each attribute parameter in the positioning module is abnormal. The first matching condition and the second matching condition may be preset.

[0059] It is understandable that the above target abnormality may be an abnormality caused by a fault in the perception module, and at the same time, it may also be an abnormality caused by a fault in the positioning module. In addition, the fault of the perception module and the fault of the positioning module may correspond to the same preset abnormality. Therefore, the at least two first faults may include a fault of the perception module and / or a fault of the positioning module.

[0060] In this embodiment, by presetting the first matching condition and the second matching condition, when it is necessary to determine whether a corresponding first fault occurs in the perception module and the positioning module during the virtual simulation test of the autonomous driving vehicle, a judgment can be made based on the first matching condition and the second matching condition to realize the target fault determination process.

[0061] Optionally, the first matching condition includes at least one of the following:

[0062] When the first candidate fault is a detection frame loss fault and the number of detection frames of the target object does not match the number of shooting frames of the target object, determining that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the number of detection frames, and the second attribute parameter includes the number of shooting frames;

[0063] In a case where the first candidate fault is a delayed selection fault and the difference between a first time point and a second time point is greater than a first threshold, determining that the first attribute parameter does not match the second attribute parameter, the first time point is a time point when the autonomous driving vehicle actually selects the target object, the second time point is a time point when the target object enters a detection range of the autonomous driving vehicle, the first attribute parameter includes the first time point, and the second attribute parameter includes the second time point;

[0064] In a case where the first candidate fault is a detection fault of an attribute of a target obstacle, and a first attribute detection parameter of the target obstacle does not match a first attribute true parameter of the target obstacle, determining that the first attribute parameter does not match the second attribute parameter, the target object includes the target obstacle, the first attribute parameter includes the first attribute detection parameter, the second attribute parameter includes the first attribute true parameter, and the attribute of the target obstacle includes at least one of the following: a position attribute, a speed attribute, and a type attribute;

[0065] When the first candidate fault is a detection fault of the attribute of the lane line, and the second attribute detection parameter of the lane line does not match the second attribute true parameter of the lane line, it is determined that the first attribute parameter does not match the second attribute parameter, the target object includes the lane line, the first attribute parameter includes the second attribute detection parameter, the second attribute parameter includes the second attribute true parameter, and the attribute of the lane line includes at least one of the following: the morphological attribute of the lane line and the position attribute of the lane line.

[0066] The target object may be a pre-set object. For example, in a lane change test scenario, the target object may be the vehicle in front. The method provided by the embodiment of the present disclosure is further explained below by taking the target object as the vehicle in front in the lane change test scenario as an example.

[0067] The above-mentioned detection frame loss fault may be a fault of missed detection during the identification process of the target object. Specifically, after the main vehicle selects the target object, the target object will continue to be photographed, and the photographed image will be transmitted to the perception module, and the perception module will identify the target object in the photographed image to determine the real-time position of the target object. However, in the process of the perception module identifying the target object in the photographed image, there may be a problem of missed detection, that is, a certain frame of the photographed image includes the image content of the target object, but the perception module does not output the position of the target object for the frame of the photographed image, which may lead to wrong driving decisions of the autonomous driving vehicle. Among them, the detection frame number of the above-mentioned target object can be the number of photographed images containing the target object detected by the perception module during the virtual simulation test, and this number can be directly obtained from the detection parameter set. The shooting frame number of the target object is the number of times the target object is photographed after the main vehicle selects the target object, and this number can be directly obtained from the simulation parameter set. In this way, by determining whether the detection frame number of the target object is equal to the shooting frame number of the target object, it can be determined whether the detection frame loss fault occurs.

[0068] The above-mentioned delayed selection fault is a fault that the target object is selected too late. The above-mentioned detection range may refer to the lane where the host vehicle is located, that is, the second time point is the actual time point when the target object starts to change lanes. In this way, by comparing the first time point when the host vehicle selects the target object with the time point when the target object actually starts to change lanes, the delay in the host vehicle selecting the target object can be determined. When the delay is greater than the first threshold, it is determined that the host vehicle has a delayed selection fault.

[0069] The above-mentioned detection failure of the attribute of the target obstacle may specifically include: position detection failure, speed detection failure and type detection failure. The above-mentioned position detection failure may refer to that the error between the detected position of the target obstacle and the actual position of the target obstacle exceeds the preset position error. The above-mentioned speed detection failure may refer to that the error between the detected speed of the target obstacle and the actual speed of the target obstacle exceeds the preset speed error. The above-mentioned type detection failure is the detection error of the category of the target obstacle, for example, pedestrians are mistakenly detected as vehicles. The first attribute detection parameters may include position detection parameters, speed detection parameters and type detection parameters for the target obstacle. The first attribute real parameters may include real position parameters, real speed parameters and real type parameters of the target obstacle.

[0070] Accordingly, when the first candidate fault is a position detection fault for a target obstacle, and the position detection parameter of the target obstacle does not match the real position parameter of the target obstacle, it is determined that the first attribute parameter does not match the second attribute parameter. Wherein, the mismatch between the position detection parameter and the real position parameter may mean that the distance between the position indicated by the position detection parameter and the position indicated by the real position parameter exceeds a preset distance threshold.

[0071] When the first candidate fault is a speed detection fault for the target obstacle, and the speed detection parameter of the target obstacle does not match the real speed parameter of the target obstacle, it is determined that the first attribute parameter does not match the second attribute parameter. The speed detection parameter does not match the real speed parameter, which may mean that the distance between the speed indicated by the speed detection parameter and the speed indicated by the real speed parameter exceeds a preset speed threshold.

[0072] When the first candidate fault is a type detection fault for the target obstacle, and the type detection parameter is different from the true type parameter, it is determined that the first attribute parameter does not match the second attribute parameter.

[0073] The above-mentioned lane line morphological attributes may include the length of the lane line and whether the lane line is missing. The lane line position attributes may include: lane line change state attribute, relative position attribute of two lane lines in the same lane, and lane line position error attribute. Accordingly, the lane line attribute detection failure may include: lane line length detection failure, lane line jitter failure, lane line inside and outside eight failure, lane line single and double side missing failure, and lane line position error failure.

[0074] Accordingly, the second attribute detection parameter may include: the detected length of the lane line, the detected lane line position change state, the detected lane line form, the detected lane line shape, and the detected lane line position. The second attribute real parameter may include: the real length of the lane line, the real lane line position, the real lane line form, and the real lane line shape.

[0075] When the first candidate fault is a lane line length detection fault, and the difference between the detected length of the lane line and the actual length of the lane line is less than a preset length threshold, it is determined that the first attribute parameter does not match the second attribute parameter.

[0076] When the first candidate fault is a lane line jitter fault, and the detected lane line position change status indicates that the standard deviation of the position error change between the lane line positions obtained by continuous multi-frame shooting exceeds a preset threshold, it is determined that the first attribute parameter does not match the second attribute parameter.

[0077] When the first candidate fault is a lane line inner-outer eight fault, and the detected lane line shape indicates that the slope of the lane line in multiple consecutive frames of images exceeds a preset slope threshold, it is determined that the first attribute parameter does not match the second attribute parameter.

[0078] When the first candidate fault is a lane line missing fault on one or both sides, and within the simulation time, the actual lane line exists for several consecutive frames but the sensed lane line is not detected, it is determined that the first attribute parameter does not match the second attribute parameter.

[0079] When the first candidate fault is a lane line position error fault, and the difference between the detected lane line position and the actual lane line position is greater than a set position error threshold, it is determined that the first attribute parameter does not match the second attribute parameter.

[0080] Please refer to Table 1 below for a definition of a fault present in a sensing module in an embodiment of the present disclosure. Please refer to Table 2 below for a corresponding relationship between the detection strategy corresponding to each fault type in the first matching condition:

[0081] Table 1

[0082]

[0083]

[0084] Table 2

[0085]

[0086] Please refer to Table 3, which shows the candidate faults of the perception module corresponding to the above-mentioned preset anomalies.

[0087] Table 3

[0088]

[0089]

[0090] In Table 3, "O" indicates that there is a corresponding relationship, and "X" indicates that there is no corresponding relationship. For example, the candidate faults in the perception module corresponding to the preset anomaly "main responsible collision" include: target frame loss, target delayed selection, target speed error is too large, target position error is too large, and target type error.

[0091] In this implementation, the possible fault types of the perception module are determined in advance, and the detection strategies corresponding to each fault are pre-set. In this way, when it is necessary to determine whether the perception module has a corresponding fault during the virtual simulation test, the pre-set detection strategy can be directly used for detection to realize the fault detection process of the perception module.

[0092] Optionally, the second matching condition includes at least one of the following:

[0093] When the first candidate fault is a positioning drift fault of the autonomous driving vehicle, and the detected positioning parameters of the autonomous driving vehicle do not match the real positioning parameters of the autonomous driving vehicle, it is determined that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the detected positioning parameter, the second attribute parameter includes the real positioning parameter, and the positioning drift fault includes a position drift fault and a heading angle drift fault;

[0094] When the first candidate fault is a positioning jump fault of the autonomous driving vehicle, and the first position change parameter does not match the second position change parameter, it is determined that the first attribute parameter does not match the second attribute parameter, the first position change parameter is the position change parameter detected by the autonomous driving vehicle, the second position change parameter is the actual position change parameter of the autonomous driving vehicle, the first attribute parameter includes the first position change parameter, the second attribute parameter includes the actual positioning parameter, and the positioning jump fault includes a positioning jump fault and a heading angle jump fault.

[0095] Please refer to Table 4 below for a definition of a fault existing in a positioning module in an embodiment of the present disclosure. Please refer to Table 5 below for a corresponding relationship between the detection strategy corresponding to each fault type in the second matching condition:

[0096] Table 4

[0097]

[0098] Table 5

[0099]

[0100] The position error described in Table 5 refers to the coordinates of the center point of the rear axle of the vehicle output by the positioning module in a Cartesian coordinate system with the center of the rear axle of the autonomous driving vehicle in the simulation as the origin, the front of the vehicle as the x-axis, the side as the y-axis, and the vertical as the z-axis. The same applies to the heading angle error.

[0101] In addition, according to the situation of road test problems, positioning drift may cause abnormal driving behaviors such as vehicle shaking / drawing, collision risk or hitting the shoulder of the road. Positioning jump may cause the vehicle to exit abnormal autonomous driving behavior. Therefore, similar to the analysis of perception problems, the connection between the abnormal behavior of the main vehicle and the positioning problem can be established, which is convenient for tracing the source of the abnormal behavior of the main vehicle.

[0102] In this implementation, the possible fault types of the positioning module are determined in advance, and the detection strategies corresponding to each fault are pre-set. In this way, when it is necessary to determine whether the positioning module has a corresponding fault during the virtual simulation test, the pre-set detection strategy can be directly used for detection to realize the fault detection process of the positioning module.

[0103] Optionally, the preset function module further includes a map module, the candidate faults corresponding to the target abnormality further include at least two second faults, the second faults are faults of the map module, and determining the target fault from the candidate faults corresponding to the target abnormality further includes:

[0104] When, based on the first matching condition and the second matching condition, it is determined that the first candidate fault is a fault of a module other than the perception module and the positioning module, and no fault exists in other modules other than the map module, the target fault is determined among the at least two second faults.

[0105] Since the map module usually provides instructions for other modules, it does not need to collect data or process data. Therefore, it is impossible to perform fault detection in a similar way to the above-mentioned positioning module and perception module, that is, it is impossible to detect whether there is a fault by comparing the detection value with the true value. In the related art, the main thing is to provide maps, and its failure rate is relatively low, while the failure rate of the perception module and the positioning module is relatively high. Therefore, when an autonomous driving vehicle has a target abnormality, it can be determined that the cause of the target abnormality is the map module when it is determined that there is no fault in the perception module and the positioning module. And perform fault detection in the candidate faults corresponding to the map module.

[0106] Based on the first matching condition and the second matching condition, it is determined that the first candidate fault is a fault of a module other than the perception module and the positioning module, that is, based on the first matching condition and the second matching condition, it is determined that during the virtual simulation test, the perception module and the positioning module have not failed.

[0107] It is understandable that the absence of faults in other modules other than the map module means that: during the virtual simulation test, there are no faults in other modules other than the map module, the perception module and the positioning module. Specifically, to ensure that there are no faults in other modules other than the map module, the perception module and the positioning module during the virtual simulation test, the map module, the perception module and the positioning module can be set to an ideal state in the simulation software. In this way, during the virtual simulation test, when the above-mentioned target abnormality occurs, it is only possible that an abnormality occurs in a module among the map module, the perception module and the positioning module. In the case where it is determined that there are no faults in the perception module and the positioning module based on the above-mentioned first matching condition and the second matching condition, it can be considered that the above-mentioned target abnormality is an abnormality caused by a fault in the map module. Therefore, the target fault can be determined among the at least two second faults.

[0108] In this implementation, by determining, based on the first matching condition and the second matching condition, that the first candidate fault is a fault of a module other than the perception module and the positioning module, and that no fault exists in modules other than the map module, the target fault is determined among the at least two second faults to ensure that fault detection can be performed quickly even when a fault occurs in the map module.

[0109] Optionally, the at least two second faults include a second candidate fault, the autonomous driving vehicle performs the virtual simulation test based on map information in the map module, and determining the target fault from the at least two second faults includes:

[0110] When the second candidate fault is a target map element labeling fault and it is detected that the target map element in the map information has a labeling fault, determining that the second candidate fault is the target fault, and the target map element includes at least one of the following: a speed limit sign, a lane line, a traffic light, and a lane;

[0111] When the second candidate fault is a binding relationship fault between map elements, and the binding relationship fault is detected between the map elements in the map information, the second candidate fault is determined to be the target fault, and the binding relationship includes at least one of the following: a binding relationship between different lanes, and a binding relationship between a traffic light and a lane.

[0112] Please refer to Table 6 below. Possible faults of the map module may include: element missing labeling fault, element wrong labeling fault, element binding relationship missing fault, element binding relationship error fault and element labeling error too large fault.

[0113] Table 6

[0114]

[0115] The missing speed limit in the map means that the speed limit sign of the lane where the main vehicle is located is missing, which may cause the main vehicle to exceed the speed limit. The missing speed limit in the lane of the main vehicle means that the road in front of or behind the lane where the main vehicle is located is missing in the map.

[0116] Specifically, a correspondence between candidate faults and preset abnormalities in the map module can be established in advance. In this way, when it is determined that the target abnormality occurs in the main vehicle during the virtual simulation test, and it is determined that the abnormality is not caused by the perception module and the positioning module, the at least two second faults can be determined from the possible faults of the above-mentioned map module based on the abnormality type according to the target abnormality and using the correspondence. And the means in the relevant technology are used to detect whether the map module has the second fault to determine the target fault.

[0117] Among them, in the virtual simulation test, when the main vehicle stalls / stutters, it may be that the map lane is not smooth enough, or the lane inheritance relationship is bound incorrectly; when the main vehicle exits the automatic driving mode, it may be that the lane inheritance relationship is missing; when the vehicle reverses, it may be because the lane direction is marked incorrectly; when the vehicle runs a red light, it may be that the traffic light corresponding to the main vehicle lane is not bound or is bound incorrectly.

[0118] In this implementation, by pre-determining the possible fault types of the map module and pre-setting the preset abnormalities corresponding to each fault type, when a target abnormality occurs in the main vehicle, at least two second faults can be screened out from the possible faults of the map module, so as to improve the efficiency of fault detection.

[0119] Optionally, the preset abnormality includes at least one of the following: collision, braking deceleration exceeding a preset deceleration, vehicle shaking amplitude exceeding a preset amplitude, exiting the automatic driving mode, driving in the wrong direction, and running a red light.

[0120] Among them, the collision may refer to a collision between the main vehicle and the main vehicle, the braking deceleration exceeds the preset deceleration, that is, emergency braking occurs, and the vehicle shaking amplitude exceeds the preset amplitude, that is, vehicle shaking.

[0121] It is understandable that means in related technologies can be used to detect whether the above-mentioned preset abnormalities occur in the autonomous driving vehicle during the virtual simulation test. When the above-mentioned abnormalities occur, abnormal information can be output to facilitate subsequent relevant personnel to determine the cause of the above-mentioned preset abnormalities based on the fault detection method provided by the present disclosure.

[0122] In this embodiment, by pre-setting the exception types of various preset exceptions and setting the candidate faults corresponding to each preset exception, when the above preset exceptions occur during the virtual simulation test of the autonomous driving vehicle, the scope of troubleshooting can be narrowed, thereby improving the efficiency of fault detection.

[0123] It should be noted that, in the specific examples of the embodiments of the present disclosure, only the perception module, the map module and the positioning module are described. However, the method of the embodiments of the present disclosure can also be applied to the fault detection process of other modules of the autonomous driving vehicle.

[0124] In addition, in another embodiment of the present disclosure, during the virtual simulation test, the above fault detection method can be used to test each module in the autonomous driving vehicle one by one. For example, when testing the map module, other modules other than the map module are set to an ideal state, and then the map module is tested based on various test scenarios. Then, in the event of a target fault, based on the correspondence between the above preset abnormality and the candidate fault, the candidate fault in the map module corresponding to the target abnormality is determined, and fault detection is performed from the determined candidate faults.

[0125] The fault detection method provided by the disclosed embodiment also has the following beneficial effects: First, since fault detection is performed only when an abnormality occurs in the autonomous driving vehicle, relevant personnel can first focus on the abnormality of the autonomous driving vehicle itself, improve the efficiency of problem location, and ignore those situations where the indicators exceed the limit but have no impact on the operation of the autonomous driving vehicle. Second, since each module is only detected when abnormal driving behavior occurs, this can improve the efficiency of simulation operation and save computing resources.

[0126] See also Figure 2 , Figure 2 The present invention provides a schematic diagram of a fault detection device 200 for an autonomous driving vehicle according to an embodiment of the present invention. The fault detection device 200 for an autonomous driving vehicle includes:

[0127] A first determining unit 201 is configured to determine, in a preset data set, a candidate fault corresponding to a target abnormality detected in the autonomous driving vehicle during a virtual simulation test of the autonomous driving vehicle, wherein the preset data set includes candidate faults corresponding to at least two preset abnormalities, the target abnormality is any one of the at least two preset abnormalities, and the candidate faults corresponding to the target abnormality are some of the candidate faults included in the preset data set, and the candidate faults are faults of preset functional modules in the autonomous driving vehicle;

[0128] The second determining unit 202 is used to determine a target fault from the candidate faults corresponding to the target abnormality, wherein the target fault is a fault occurring in the preset functional module during the virtual simulation test.

[0129] Optionally, see Figure 3 , the candidate faults corresponding to the target anomaly include at least two first faults, and the second determining unit 202 includes:

[0130] The acquisition subunit 2021 is used to acquire a detection parameter set and a simulation parameter set, wherein the detection parameter set includes detection attribute parameters of the target object detected by the autonomous driving vehicle during the virtual simulation test; the simulation parameter set includes true value attribute parameters of the target object during the virtual simulation test; the target object includes at least one of the following: the autonomous driving vehicle and a driving environment object other than the autonomous driving vehicle;

[0131] The determination subunit 2022 is configured to determine the target fault among the at least two first faults based on the detection parameter set and the simulation parameter set.

[0132] Optionally, the determining subunit 2022 is specifically configured to determine the first candidate fault as the target fault when the first attribute parameter does not match the second attribute parameter;

[0133] Among them, the first candidate fault is a fault among the at least two first faults, and the first candidate fault is used to characterize the abnormality of the target attribute of the preset functional module; the first attribute parameter is an attribute parameter in the detection parameter set, which is used to characterize the target attribute of the preset functional module; the second attribute parameter is an attribute parameter in the simulation parameter set, which is used to characterize the target attribute of the preset functional module.

[0134] Optionally, the preset function module includes a perception module and a positioning module, and the determination subunit 2022 is further configured to determine, when the first candidate fault is a fault of the perception module, a matching state between the first attribute parameter and the second attribute parameter based on a first matching condition; and, when the first candidate fault is a fault of the positioning module, determine, when the first candidate fault is a fault of the positioning module, a matching state between the first attribute parameter and the second attribute parameter based on a second matching condition;

[0135] The first matching condition is different from the second matching condition.

[0136] Optionally, the first matching condition includes at least one of the following:

[0137] When the first candidate fault is a detection frame loss fault and the number of detection frames of the target object does not match the number of shooting frames of the target object, determining that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the number of detection frames, and the second attribute parameter includes the number of shooting frames;

[0138] In a case where the first candidate fault is a delayed selection fault and the difference between a first time point and a second time point is greater than a first threshold, determining that the first attribute parameter does not match the second attribute parameter, the first time point is a time point when the autonomous driving vehicle actually selects the target object, the second time point is a time point when the target object enters a detection range of the autonomous driving vehicle, the first attribute parameter includes the first time point, and the second attribute parameter includes the second time point;

[0139] In a case where the first candidate fault is a detection fault of an attribute of a target obstacle, and a first attribute detection parameter of the target obstacle does not match a first attribute true parameter of the target obstacle, determining that the first attribute parameter does not match the second attribute parameter, the target object includes the target obstacle, the first attribute parameter includes the first attribute detection parameter, the second attribute parameter includes the first attribute true parameter, and the attribute of the target obstacle includes at least one of the following: a position attribute, a speed attribute, and a type attribute;

[0140] When the first candidate fault is a detection fault of the attribute of the lane line, and the second attribute detection parameter of the lane line does not match the second attribute true parameter of the lane line, it is determined that the first attribute parameter does not match the second attribute parameter, the target object includes the lane line, the first attribute parameter includes the second attribute detection parameter, the second attribute parameter includes the second attribute true parameter, and the attribute of the lane line includes at least one of the following: the morphological attribute of the lane line and the position attribute of the lane line.

[0141] Optionally, the second matching condition includes at least one of the following:

[0142] When the first candidate fault is a positioning drift fault of the autonomous driving vehicle, and the detected positioning parameters of the autonomous driving vehicle do not match the real positioning parameters of the autonomous driving vehicle, it is determined that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the detected positioning parameter, the second attribute parameter includes the real positioning parameter, and the positioning drift fault includes a position drift fault and a heading angle drift fault;

[0143] When the first candidate fault is a positioning jump fault of the autonomous driving vehicle, and the first position change parameter does not match the second position change parameter, it is determined that the first attribute parameter does not match the second attribute parameter, the first position change parameter is the position change parameter detected by the autonomous driving vehicle, the second position change parameter is the actual position change parameter of the autonomous driving vehicle, the first attribute parameter includes the first position change parameter, the second attribute parameter includes the actual positioning parameter, and the positioning jump fault includes a positioning jump fault and a heading angle jump fault.

[0144] Optionally, the preset function module also includes a map module, and the candidate faults corresponding to the target abnormality also include at least two second faults, and the second fault is a fault of the map module. The determination subunit 2022 is specifically used to determine the target fault among the at least two second faults when, based on the first matching condition and the second matching condition, the first candidate fault is determined to be a fault of modules other than the perception module and the positioning module, and there is no fault in other modules other than the map module.

[0145] Optionally, the at least two second faults include a second candidate fault, the autonomous driving vehicle performs the virtual simulation test based on the map information in the map module, and the determining subunit 2022 is further configured to determine that the second candidate fault is the target fault when the second candidate fault is a target map element labeling fault and it is detected that the target map element in the map information has a labeling fault, and the target map element includes at least one of the following: a speed limit sign, a lane line, a traffic light, and a lane;

[0146] The determination subunit 2022 is further specifically used to determine that the second candidate fault is the target fault when the second candidate fault is a binding relationship fault between map elements and the binding relationship fault is detected between the map elements in the map information, and the binding relationship includes at least one of the following: a binding relationship between different lanes, and a binding relationship between a traffic light and a lane.

[0147] Optionally, the preset abnormality includes at least one of the following: collision, braking deceleration exceeding a preset deceleration, vehicle shaking amplitude exceeding a preset amplitude, exiting the automatic driving mode, driving in the wrong direction, and running a red light.

[0148] It should be noted that the fault detection device 200 for the autonomous driving vehicle provided in this embodiment can implement all the technical solutions of the above-mentioned fault detection method embodiment for the autonomous driving vehicle, and therefore can at least achieve all the above-mentioned technical effects, which will not be repeated here.

[0149] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0150] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0151] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0152] like Figure 4 As shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0153] Multiple components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the device 400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0154] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the fault detection method of the autonomous driving vehicle. For example, in some embodiments, the fault detection method of the autonomous driving vehicle may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the fault detection method of the autonomous driving vehicle described above are performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the fault detection method of the autonomous driving vehicle in any other appropriate manner (e.g., by means of firmware).

[0155] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0156] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0157] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0158] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0159] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0160] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0161] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0162] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A fault detection method for an autonomous driving vehicle, comprising: In the process of performing virtual simulation test on the autonomous driving vehicle, when a target abnormality is detected in the autonomous driving vehicle, a candidate fault corresponding to the target abnormality is determined in a preset data set, wherein the preset data set includes candidate faults corresponding to at least two preset abnormalities, the target abnormality is any one of the at least two preset abnormalities, and the candidate faults corresponding to the target abnormality are some candidate faults among the candidate faults included in the preset data set, and the candidate faults are faults of preset functional modules in the autonomous driving vehicle; Determining a target fault from candidate faults corresponding to the target abnormality, wherein the target fault is a fault occurring in the preset functional module during the virtual simulation test; The candidate faults corresponding to the target abnormality include at least two first faults, and determining the target fault from the candidate faults corresponding to the target abnormality includes: Acquire a detection parameter set and a simulation parameter set, wherein the detection parameter set includes detection attribute parameters of the target object detected by the autonomous driving vehicle during the virtual simulation test; the simulation parameter set includes true value attribute parameters of the target object during the virtual simulation test; the target object includes at least one of the following: the autonomous driving vehicle and a driving environment object other than the autonomous driving vehicle; The target fault is determined among the at least two first faults based on the detection parameter set and the simulation parameter set.

2. The method according to claim 1, wherein: The determining a target fault among the at least two first faults based on the detection parameter set and the simulation parameter set comprises: In the case where the first attribute parameter does not match the second attribute parameter, determining the first candidate fault as the target fault; Among them, the first candidate fault is a fault among the at least two first faults, and the first candidate fault is used to characterize the abnormality of the target attribute of the preset functional module; the first attribute parameter is an attribute parameter in the detection parameter set, which is used to characterize the target attribute of the preset functional module; the second attribute parameter is an attribute parameter in the simulation parameter set, which is used to characterize the target attribute of the preset functional module.

3. The method according to claim 2, wherein: The preset function module includes a perception module and a positioning module. Before determining the first candidate fault as the target fault, the method further includes: In the case where the first candidate fault is a fault of the perception module, determining a matching state between the first attribute parameter and the second attribute parameter based on a first matching condition; in the case where the first candidate fault is a fault of the positioning module, determining a matching state between the first attribute parameter and the second attribute parameter based on a second matching condition; The first matching condition is different from the second matching condition.

4. The method according to claim 3, wherein: The first matching condition includes at least one of the following: When the first candidate fault is a detection frame loss fault and the number of detection frames of the target object does not match the number of shooting frames of the target object, determining that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the number of detection frames, and the second attribute parameter includes the number of shooting frames; In a case where the first candidate fault is a delayed selection fault and the difference between a first time point and a second time point is greater than a first threshold, determining that the first attribute parameter does not match the second attribute parameter, the first time point is a time point when the autonomous driving vehicle actually selects the target object, the second time point is a time point when the target object enters a detection range of the autonomous driving vehicle, the first attribute parameter includes the first time point, and the second attribute parameter includes the second time point; In a case where the first candidate fault is a detection fault of an attribute of a target obstacle, and a first attribute detection parameter of the target obstacle does not match a first attribute true parameter of the target obstacle, determining that the first attribute parameter does not match the second attribute parameter, the target object includes the target obstacle, the first attribute parameter includes the first attribute detection parameter, the second attribute parameter includes the first attribute true parameter, and the attribute of the target obstacle includes at least one of the following: a position attribute, a speed attribute, and a type attribute; When the first candidate fault is a detection fault of the attribute of the lane line, and the second attribute detection parameter of the lane line does not match the second attribute true parameter of the lane line, it is determined that the first attribute parameter does not match the second attribute parameter, the target object includes the lane line, the first attribute parameter includes the second attribute detection parameter, the second attribute parameter includes the second attribute true parameter, and the attribute of the lane line includes at least one of the following: the morphological attribute of the lane line and the position attribute of the lane line.

5. The method according to claim 3, wherein: The second matching condition includes at least one of the following: When the first candidate fault is a positioning drift fault of the autonomous driving vehicle, and the detected positioning parameters of the autonomous driving vehicle do not match the real positioning parameters of the autonomous driving vehicle, it is determined that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the detected positioning parameter, the second attribute parameter includes the real positioning parameter, and the positioning drift fault includes a position drift fault and a heading angle drift fault; When the first candidate fault is a positioning jump fault of the autonomous driving vehicle, and the first position change parameter does not match the second position change parameter, it is determined that the first attribute parameter does not match the second attribute parameter, the first position change parameter is the position change parameter detected by the autonomous driving vehicle, the second position change parameter is the actual position change parameter of the autonomous driving vehicle, the first attribute parameter includes the first position change parameter, the second attribute parameter includes the actual positioning parameter, and the positioning jump fault includes a positioning jump fault and a heading angle jump fault.

6. The method according to claim 3, wherein: The preset function module further includes a map module, the candidate faults corresponding to the target abnormality further include at least two second faults, the second faults are faults of the map module, and the target fault is determined from the candidate faults corresponding to the target abnormality, further including: When, based on the first matching condition and the second matching condition, it is determined that the first candidate fault is a fault of a module other than the perception module and the positioning module, and no fault exists in other modules other than the map module, the target fault is determined among the at least two second faults.

7. The method according to claim 6, wherein: The at least two second faults include a second candidate fault, the autonomous driving vehicle performs the virtual simulation test based on the map information in the map module, and determining the target fault from the at least two second faults includes: When the second candidate fault is a target map element labeling fault and it is detected that the target map element in the map information has a labeling fault, determining that the second candidate fault is the target fault, and the target map element includes at least one of the following: a speed limit sign, a lane line, a traffic light, and a lane; When the second candidate fault is a binding relationship fault between map elements, and the binding relationship fault is detected between the map elements in the map information, the second candidate fault is determined to be the target fault, and the binding relationship includes at least one of the following: a binding relationship between different lanes, and a binding relationship between a traffic light and a lane.

8. The method according to any one of claims 1 to 7, wherein: The preset abnormalities include at least one of the following: collision, braking deceleration exceeding a preset deceleration, vehicle shaking amplitude exceeding a preset amplitude, exiting the automatic driving mode, driving in the wrong direction, and running a red light.

9. A fault detection device for an autonomous driving vehicle, comprising: a first determining unit, configured to determine, in a preset data set, a candidate fault corresponding to a target abnormality detected in the autonomous driving vehicle during a virtual simulation test of the autonomous driving vehicle, wherein the preset data set includes candidate faults corresponding to at least two preset abnormalities, the target abnormality is any one of the at least two preset abnormalities, and the candidate faults corresponding to the target abnormality are some candidate faults among the candidate faults included in the preset data set, and the candidate faults are faults of preset functional modules in the autonomous driving vehicle; A second determining unit is used to determine a target fault from the candidate faults corresponding to the target abnormality, wherein the target fault is a fault occurring in the preset functional module during the virtual simulation test; The candidate faults corresponding to the target abnormality include at least two first faults, and the second determination unit includes: an acquisition subunit, configured to acquire a detection parameter set and a simulation parameter set, wherein the detection parameter set includes detection attribute parameters of the target object detected by the autonomous driving vehicle during the virtual simulation test; the simulation parameter set includes true value attribute parameters of the target object during the virtual simulation test; the target object includes at least one of the following: the autonomous driving vehicle and a driving environment object other than the autonomous driving vehicle; A determination subunit is configured to determine the target fault among the at least two first faults based on the detection parameter set and the simulation parameter set.

10. The device according to claim 9, wherein: The determination subunit is specifically configured to determine, when the first attribute parameter does not match the second attribute parameter, that the first candidate fault is the target fault; Among them, the first candidate fault is a fault among the at least two first faults, and the first candidate fault is used to characterize the abnormality of the target attribute of the preset functional module; the first attribute parameter is an attribute parameter in the detection parameter set, which is used to characterize the target attribute of the preset functional module; the second attribute parameter is an attribute parameter in the simulation parameter set, which is used to characterize the target attribute of the preset functional module.

11. The device according to claim 10, wherein: The preset function module includes a perception module and a positioning module, and the determination subunit is specifically used to determine the matching state of the first attribute parameter and the second attribute parameter based on a first matching condition when the first candidate fault is a fault of the perception module; In a case where the first candidate fault is a fault of the positioning module, determining a matching state between the first attribute parameter and the second attribute parameter based on a second matching condition; The first matching condition is different from the second matching condition.

12. The device according to claim 11, wherein The first matching condition includes at least one of the following: When the first candidate fault is a detection frame loss fault and the number of detection frames of the target object does not match the number of shooting frames of the target object, determining that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the number of detection frames, and the second attribute parameter includes the number of shooting frames; In a case where the first candidate fault is a delayed selection fault and the difference between a first time point and a second time point is greater than a first threshold, determining that the first attribute parameter does not match the second attribute parameter, the first time point is a time point when the autonomous driving vehicle actually selects the target object, the second time point is a time point when the target object enters a detection range of the autonomous driving vehicle, the first attribute parameter includes the first time point, and the second attribute parameter includes the second time point; In a case where the first candidate fault is a detection fault of an attribute of a target obstacle, and a first attribute detection parameter of the target obstacle does not match a first attribute true parameter of the target obstacle, determining that the first attribute parameter does not match the second attribute parameter, the target object includes the target obstacle, the first attribute parameter includes the first attribute detection parameter, the second attribute parameter includes the first attribute true parameter, and the attribute of the target obstacle includes at least one of the following: a position attribute, a speed attribute, and a type attribute; When the first candidate fault is a detection fault of the attribute of the lane line, and the second attribute detection parameter of the lane line does not match the second attribute true parameter of the lane line, it is determined that the first attribute parameter does not match the second attribute parameter, the target object includes the lane line, the first attribute parameter includes the second attribute detection parameter, the second attribute parameter includes the second attribute true parameter, and the attribute of the lane line includes at least one of the following: the morphological attribute of the lane line and the position attribute of the lane line.

13. The device according to claim 11, wherein: The second matching condition includes at least one of the following: When the first candidate fault is a positioning drift fault of the autonomous driving vehicle, and the detected positioning parameters of the autonomous driving vehicle do not match the real positioning parameters of the autonomous driving vehicle, it is determined that the first attribute parameter does not match the second attribute parameter, the first attribute parameter includes the detected positioning parameter, the second attribute parameter includes the real positioning parameter, and the positioning drift fault includes a position drift fault and a heading angle drift fault; When the first candidate fault is a positioning jump fault of the autonomous driving vehicle, and the first position change parameter does not match the second position change parameter, it is determined that the first attribute parameter does not match the second attribute parameter, the first position change parameter is the position change parameter detected by the autonomous driving vehicle, the second position change parameter is the actual position change parameter of the autonomous driving vehicle, the first attribute parameter includes the first position change parameter, the second attribute parameter includes the actual positioning parameter, and the positioning jump fault includes a positioning jump fault and a heading angle jump fault.

14. The device according to claim 11, wherein: The preset function module also includes a map module, and the candidate faults corresponding to the target abnormality also include at least two second faults, and the second fault is a fault of the map module. The determination subunit is specifically used to determine the target fault among the at least two second faults when, based on the first matching condition and the second matching condition, it is determined that the first candidate fault is a fault of modules other than the perception module and the positioning module, and there is no fault in other modules other than the map module.

15. The device according to claim 14, wherein: The at least two second faults include a second candidate fault, the autonomous driving vehicle performs the virtual simulation test based on the map information in the map module, and the determination subunit is further configured to determine that the second candidate fault is the target fault when the second candidate fault is a target map element labeling fault and the target map element in the map information has a labeling fault, and the target map element includes at least one of the following: a speed limit sign, a lane line, a traffic light, and a lane; The determination subunit is further specifically used to determine that the second candidate fault is the target fault when the second candidate fault is a binding relationship fault between map elements and the binding relationship fault is detected between the map elements in the map information, and the binding relationship includes at least one of the following: a binding relationship between different lanes, and a binding relationship between a traffic light and a lane.

16. The device according to any one of claims 9 to 15, wherein: The preset abnormalities include at least one of the following: collision, braking deceleration exceeding a preset deceleration, vehicle shaking amplitude exceeding a preset amplitude, exiting the automatic driving mode, driving in the wrong direction, and running a red light.

17. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the fault detection method for the autonomous driving vehicle described in any one of claims 1-8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the steps of the fault detection method for an autonomous driving vehicle according to any one of claims 1-8.

19. A computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the fault detection method for an autonomous driving vehicle according to any one of claims 1 to 8.

Images