Auxiliary driving vehicle accident evidence chain tracing method and system based on fault tree

Abstract

The invention discloses an auxiliary driving vehicle accident evidence chain tracing method and system based on a fault tree. The method comprises the following steps: acquiring auxiliary driving function state data, vehicle operation data and time sequence data of a driver state in a certain time period before an accident occurs; analyzing the time series data, and obtaining an abnormal data item list according to a predetermined threshold value; the abnormal data item list is input into a pre-constructed fault tree, and events are connected in series layer by layer through AND gates and OR gates to form a plurality of accident chains; judging an accident chain corresponding to the abnormal data item list according to the logic of the fault tree, finally determining a basic event of the accident chain, and judging an error passing party according to the basic event; and recording a judgment path of the fault tree, and generating a fault analysis report. According to the invention, accurate traceability can be realized from the data level, and the systematicness and accuracy of the accident evidence chain are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle safety assisted driving, and in particular to a method and system for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree. Background Art

[0002] With the support of relevant national policies, the penetration rate of L2-class cars with assisted driving functions reached more than one-third in 2022. With the popularization of assisted driving technologies such as adaptive cruise control, automatic emergency braking, and lane keeping assist, while providing driving convenience, vehicles also increase the complexity of accident responsibility determination. In this case, traditional accident analysis methods, such as relying on eyewitness statements and accident scene investigations, face the problems of insufficient evidence and inaccurate judgments.

[0003] Traditional vehicle accident analysis methods encounter many challenges when facing assisted driving vehicle accidents. These challenges mainly include:

[0004] Relying on subjective evidence and dashcams, these methods rely heavily on eyewitness testimony and driver narratives, as well as dashcam data. Although dashcams can provide visual and audio records of the accident, enhancing the objectivity of accident analysis, these evidences are still affected by personal cognitive biases and memory errors, and dashcams usually only cover limited perspectives, which may lead to incomplete and inaccurate analysis of the cause of the accident.

[0005] The analysis efficiency is low, and a lot of time is required for on-site investigation and evidence collection, which prolongs the cycle of accident handling and responsibility determination.

[0006] The methods lack adaptability. With the continuous development of assisted driving technology, these traditional methods appear to lack adaptability, and it is often difficult to accurately determine the specific role of the driving assistance system in an accident. Summary of the invention

[0007] In view of the problems that existing traffic accident analysis relies on subjective evidence and driving recorders, has low analysis efficiency, and is not applicable to assisted driving vehicles, the present invention proposes a fault tree-based assisted driving vehicle accident evidence chain tracing method and system that can be applied to complex traffic accidents involving assisted driving vehicles and improve the accuracy and efficiency of traffic accident analysis.

[0008] The technical solution adopted by the present invention is:

[0009] A method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree is provided, comprising the following steps:

[0010] When a collision accident occurs, obtain time series data including auxiliary driving function status data, vehicle operation data and driver status for a certain period of time before the accident;

[0011] Analyze time series data and obtain a list of abnormal data items based on pre-determined thresholds;

[0012] Input the abnormal data item list into a pre-built fault tree, which includes a top event, a first-layer intermediate event, a second-layer intermediate event, and a third-layer intermediate event, and each layer of intermediate events corresponds to multiple basic events, and each event is connected layer by layer with AND gates and OR gates to form multiple accident chains; the top event is a collision accident of an assisted driving vehicle, the first-layer intermediate events are multiple events in the impending collision stage, the second-layer intermediate events are events in the emergency avoidance stage, and the third-layer intermediate events are events in the event response stage; according to the logic of the fault tree, the accident chain corresponding to the abnormal data item list is judged, and the basic events of the accident chain are finally determined, and it is judged whether it is the driver's fault or the assisted driving system's fault according to the basic events;

[0013] Record the judgment path of the fault tree and generate a fault analysis report.

[0014] Following the above technical solution, the first-level events include the vehicle motion state being unchanged and the vehicle motion state being insufficiently changed, both of which are judged based on preset thresholds.

[0015] Following the above technical solution, the second-level events include the driver not taking over, the active safety control system not intervening, the driver taking over in time but failing to avoid the risk, the driver taking over too late, and the active safety system intervening but failing to avoid the risk, all of which are judged based on preset thresholds.

[0016] Following the above technical solution, the third-level events include driver failure to intervene, warning system failure to activate, assisted driving function failure to adjust vehicle speed, assisted driving function failure to adjust vehicle speed in time, assisted driving function launched and requested to take over but driver mistakenly took over, and warning system activation too late, all of which are judged based on preset thresholds.

[0017] According to the above technical solution, the basic events include driver fatigue driving, driver distracted driving, driver obstructed vision, sensor failure to detect danger, software system failure, brake system mechanical failure, assisted driving function launched and requested to take over but the driver took over incorrectly, insufficient friction, brake pad wear, driver misjudgment of safety distance, active safety system algorithm is too conservative and active safety system design problems, all of which are judged according to preset thresholds.

[0018] According to the above technical solution, the event response stage is specifically the process from when the front vehicle slows down or an obstacle appears ahead, the assisted driving function has not yet exited and there is time to make adjustments, to when the assisted driving function exits and reminds the driver to take over; the emergency avoidance stage is specifically the transition stage after the assisted driving function exits to the activation of the active safety control system, until the active safety control system has reached the maximum braking or steering force; the impending collision stage is specifically when the active safety control system has reached its maximum effect but the collision cannot be avoided.

[0019] According to the above technical solution, the auxiliary driving function status data mainly includes data from the adaptive cruise control ACC, the anti-lock braking system ABS and the emergency braking system AEB, the vehicle operation data mainly includes the vehicle speed, acceleration and brake pedal data, and the driver status data mainly includes the takeover time and line of sight status data.

[0020] Following the above technical solution, the cause of the impending collision stage when the accident occurs is determined based on the vehicle operating status data. After determining the cause of the impending collision stage, the key data items collected before the accident are compared with the threshold of the predetermined benchmark data to determine whether they exceed the normal range. If they exceed the normal data range, they are considered abnormal, and the abnormal values ​​and corresponding time points are recorded to generate a list of abnormal data items.

[0021] The present invention also provides a fault tree-based assisted driving vehicle accident evidence chain tracing system, comprising:

[0022] The data acquisition module is used to acquire the time series data including the auxiliary driving function status data, vehicle operation data and driver status for a certain period of time before the accident when a collision accident occurs;

[0023] An abnormal data generation module is used to analyze time series data and obtain a list of abnormal data items based on a predetermined threshold;

[0024] The fault tree analysis module inputs the abnormal data item list into a pre-built fault tree, which includes a top event, a first-layer intermediate event, a second-layer intermediate event, and a third-layer intermediate event, and each layer of intermediate events corresponds to multiple basic events, and each event is connected layer by layer with AND gates and OR gates to form multiple accident chains; the top event is a collision accident of an assisted driving vehicle, the first-layer intermediate events are multiple events in the impending collision stage, the second-layer intermediate events are events in the emergency avoidance stage, and the third-layer intermediate events are events in the event response stage; the accident chain corresponding to the abnormal data item list is judged according to the logic of the fault tree, and the basic events of the accident chain are finally determined, and the fault of the driver or the assisted driving system is judged according to the basic events;

[0025] The report generation module is used to record the judgment path of the fault tree and generate a fault analysis report.

[0026] The present invention also provides a computer storage medium, which stores a computer program that can be executed by a processor, and the computer program executes the method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree as described in the above technical solution.

[0027] The beneficial effects of the present invention are as follows: the present invention takes the result of an accident collision as the top event, and the auxiliary driving function status and driver status and other factors in the three stages of the accident process (i.e., the collision stage, the emergency avoidance stage, and the event response stage) as the most direct causes of the vehicle collision accident, deeply analyzes the basic events, and constructs an assisted driving vehicle collision fault tree. The generated abnormal data item list can analyze the corresponding cause events layer by layer through the fault tree to generate an accident chain, so as to finally determine whether it is the driver's fault or the auxiliary driving system's fault. The present invention can systematically reveal the whole process of the accident, thereby reducing the reliance on single evidence and subjective judgment, and improving the accuracy of locating the cause of the accident.

[0028] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 It is a flow chart of a method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to an embodiment of the present invention;

[0031] Figure 2 These are the basic terms and logical symbols of the fault tree in the embodiment of the present invention;

[0032] Figure 3 Schematic diagram of a fault tree of an assisted driving vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0034] It should be noted that the illustrations provided in the embodiments of the present invention are only used to illustrate the basic concept of the present invention in a schematic manner. Therefore, the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout type may also be more complicated.

[0035] In the present invention, it is also necessary to explain that, if the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, if the terms "first" and "second" appear, they are only used for description and distinction purposes, and cannot be understood as indicating or implying relative importance.

[0036] Fault Tree Analysis (FTA) uses a logical deduction method to find various possible causes of a certain fault event through the analysis of logical relationships, and draws a logical relationship diagram to determine the cause of the accident, the combination of causes and the probability of the accident. It is a system safety analysis method with broad application prospects. FTA has been widely used in nuclear power, aerospace, chemical industry, pharmaceuticals and other fields, and has only been introduced to civil engineering, transportation and water conservancy in recent years.

[0037] The present invention divides the accident collision process into three stages: impending collision, emergency avoidance, and incident response, aiming to clarify the responsibility of the assisted driving function and the driver in each stage. Since the definition of the responsibilities of the assisted driving system and the driver is often controversial, the stage division helps to systematically analyze the behavioral impact of each stage and accurately identify the cause of the accident. Through the fault tree analysis model, the abnormal data of each stage can be analyzed layer by layer to determine whether the accident was caused by the driver's fault or the assisted driving system failure.

[0038] Example 1

[0039] like Figure 1 As shown, the method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree in an embodiment of the present invention mainly includes the following steps:

[0040] S1. When a collision accident occurs, obtain time series data including auxiliary driving function status data, vehicle operation data and driver status for a certain period of time before the accident;

[0041] S2, analyzing the time series data and obtaining a list of abnormal data items according to a predetermined threshold;

[0042] S3. Input the abnormal data item list into a pre-built fault tree, which includes a top event, a first-layer intermediate event, a second-layer intermediate event, and a third-layer intermediate event, and each layer of intermediate events corresponds to multiple basic events, and each event is connected layer by layer with an AND gate and an OR gate to form multiple accident chains; wherein the top event is a collision accident of an assisted driving vehicle, the first-layer intermediate events are multiple events in the impending collision stage, the second-layer intermediate events are events in the emergency avoidance stage, and the third-layer intermediate events are events in the event response stage; according to the logic of the fault tree, the accident chain corresponding to the abnormal data item list is judged, and the basic events of the accident chain are finally determined, and it is judged whether it is the fault of the driver or the fault of the assisted driving system according to the basic events;

[0043] S4. Record the judgment path of the fault tree and generate a fault analysis report.

[0044] Specifically, the first-level events include the vehicle motion state being unchanged and the vehicle motion state being insufficiently changed, both of which are judged based on preset thresholds.

[0045] The second level of events includes the driver failing to take over, the active safety control system not intervening, the driver taking over in a timely manner but failing to avoid the risk, the driver taking over too late, and the active safety system intervening but failing to avoid the risk. All are judged based on preset thresholds.

[0046] The third level of events includes driver failure to intervene, warning system failure to activate, assisted driving function failure to adjust vehicle speed, assisted driving function failure to adjust vehicle speed in a timely manner, assisted driving function launched and requested to take over but driver mistakenly took over, and warning system activated too late, all of which are judged based on preset thresholds.

[0047] Among them, basic events include driver fatigue driving, driver distracted driving, obstructed driver vision, sensors failing to detect danger, software system failure, mechanical failure of the braking system, the assisted driving function being launched and requesting to take over but the driver taking over incorrectly, insufficient friction, worn brake pads, driver misjudgment of safety distance, overly conservative active safety system algorithms, and active safety system design problems, all of which are judged based on preset thresholds.

[0048] The event response stage is specifically the process from the time when the vehicle in front slows down or an obstacle appears ahead, the assisted driving function has not yet exited and there is time to make adjustments, to the time when the assisted driving function exits and reminds the driver to take over; the emergency avoidance stage is specifically the transition stage after the assisted driving function exits to the time when the active safety control system is activated, until the active safety control system has reached the maximum braking or steering force; the impending collision stage is specifically the time when the active safety control system has reached its maximum effect but the collision cannot be avoided.

[0049] Optionally, the auxiliary driving function status data mainly includes data from adaptive cruise control ACC, anti-lock braking system ABS and emergency braking system AEB, the vehicle operation data mainly includes vehicle speed, acceleration, and brake pedal data, and the driver status data mainly includes takeover time and line of sight status data.

[0050] The specific generation process of the abnormal data item list mainly includes the following steps: judging the cause of the impending collision stage when the accident occurs based on the vehicle operation status data; after determining the cause of the impending collision stage, comparing the key data items collected before the accident with the threshold of the predetermined benchmark data to determine whether they exceed the normal range. If they exceed the normal data range, they are considered abnormal, and the abnormal values ​​and corresponding time points are recorded to generate a list of abnormal data items.

[0051] Then, the direct cause of the vehicle's status before the accident is confirmed based on the list of abnormal data items. For the lower-level node causes under the direct cause, find out whether there are matching abnormal items in the abnormal data list, trace down layer by layer until the bottom-level abnormal cause data item is found, and record the path completely.

[0052] The present invention takes the result of an accident collision as the top event, and the auxiliary driving function status and driver status and other factors in the three stages of the accident process (i.e., the collision stage, the emergency avoidance stage and the event response stage) as the most direct causes of the vehicle collision accident. The basic events are obtained through in-depth analysis, and an assisted driving vehicle collision fault tree is constructed. The generated abnormal data item list can be used to analyze the corresponding cause events layer by layer through the fault tree to generate an accident chain, so as to finally determine whether it is the driver's fault or the auxiliary driving system's fault. The present invention can systematically reveal the whole process of the accident, more accurately locate the cause of the accident from the data level, and reduce the reliance on single evidence and subjective judgment.

[0053] Example 2

[0054] This embodiment is based on Embodiment 1, and the main difference is that a specific fault tree determination process is provided.

[0055] The construction in the case of a fault in this embodiment mainly includes the following steps:

[0056] Step 1: For assisted driving vehicle collision accidents, in order to understand its collision mechanism and avoid or mitigate collision events at various stages in the collision development process, the assisted driving vehicle collision process is divided into three stages, and a fault tree model of assisted driving vehicle collision is constructed;

[0057] Step 2: Perform a qualitative evaluation of the fault tree. Qualitative evaluation can determine the minimum cut set of the top event. The minimum cut set is the minimum combination of basic events that lead to the top event. It helps to gain a deeper understanding of the causal relationship between events and the causes of the top event. Each minimum cut set represents a possible path for basic events to lead to the top event;

[0058] Step 3: Perform a quantitative assessment of the fault tree to determine the importance of the basic events in contributing to the top event. The results of the quantitative assessment help prioritize actions and resources based on the importance of the basic events that led to the top event.

[0059] Furthermore, the three stages of the assisted driving car collision process in step 1 are: 1) event response stage (such as 66.6 to 97.2 seconds before the collision), that is, the front vehicle slows down or an obstacle appears in front, the assisted driving function has not yet exited and has time to make adjustments, to the process of the assisted driving function exiting and reminding the driver to take over; 2) emergency avoidance stage (such as 20 to 66.6 seconds before the collision), that is, the transition stage after the assisted driving function exits to the activation of the active safety control system, until the active safety control system has reached the maximum braking or steering force; 3) the collision stage (such as 0 to 20 seconds before the collision), that is, the active safety control system has reached the maximum effect but still cannot avoid the collision. Taking the result of the accident collision as the top event, the assisted driving function status and driver status in the three stages of the accident process as the most direct cause of the vehicle collision accident, the basic events are obtained through in-depth analysis, and the assisted driving vehicle collision fault tree is constructed.

[0060] The collision of the assisted driving vehicle is taken as the top event of the fault tree for analysis and a fault tree model is constructed. The information of each event is shown in Table 1. The fault tree model includes all the contents before the collision of the assisted driving vehicle. Each event is connected layer by layer with AND gates and OR gates to form an accident chain, as shown in the attached figure. Figure 3 shown.

[0061] Table 1: Event names and meanings

[0062]

[0063]

[0064] Furthermore, the Boolean algebra simplification method is used to perform qualitative evaluation of the fault tree in step 2. Figure 2 , 3As shown, this method first lists the Boolean expression of the fault tree, starting from the first layer of input events in the fault tree, the input events of the "OR gate" are represented by logical addition, and the input events of the "AND gate" are represented by logical product; then the second layer of input events replaces the first layer, and the third layer of input events replaces the second layer, until all events in the fault tree are replaced. In the replacement process, the conditions and events are always represented by logical products. After the Boolean expression is sorted, the logical sum of several logical products is obtained. Each logical product is a cut set. Then, using the relevant operation laws of Boolean algebra to simplify, the minimum cut set can be found. Specifically, Figure 3 As shown, the fault tree is qualitatively evaluated based on the Boolean algebra simplification method, and the minimum cut set of the assisted driving vehicle collision fault tree is calculated. The Boolean expression is T=G1+G2, and finally 12 minimum cut sets are simplified.

[0065] Furthermore, in step 3, the fault tree is quantitatively evaluated based on probability importance and critical importance. Probability importance analysis is a quantitative analysis based on the probability importance coefficient of each basic event, that is, the degree to which the change in the probability of the top event caused by the change in the probability of the occurrence of the i-th basic event causes the change in the probability of the occurrence of the top event. i By taking a partial derivative, we can get the probability importance coefficient of the basic event. The probability importance coefficient of the basic event is:

[0066]

[0067] In the formula, P(T) represents the probability of the top event occurring, q i represents the probability of occurrence of the i-th basic event;

[0068] Critical importance analysis indicates the rate of change of the probability of the top event caused by the change in the probability of the occurrence of the ith basic event. The critical importance coefficient of the basic event is:

[0069]

[0070] In the formula, I g (i) represents the probability importance coefficient of the i-th basic event, represents the critical importance coefficient of the ith basic event, P(T) represents the probability of the top event, and q i represents the probability of occurrence of the i-th basic event;

[0071] The probability of basic events can be calculated from historical assisted driving vehicle accident samples.

[0072] Therefore, the first-stage cause of the accident, that is, whether the vehicle's motion state has changed, can be determined based on the vehicle's operating status data. After determining the cause of the impending collision, the key data items collected before the accident are compared with the threshold of the benchmark data to determine whether they are beyond the normal range. If a data item is beyond the normal range, the abnormal value and corresponding time point of the data are recorded, and an abnormal data list is generated. The example is as follows:

[0073] Abnormal data item 1: The driver took over at T1 (the threshold is T1'), and the time point was t1 seconds before the accident.

[0074] Abnormal data item 2: The driver’s pedal braking force is less than the threshold value P1, and the time point is t2 seconds before the accident.

[0075] Abnormal data item 3: The driver's line of sight is not focused on the front, and the duration is T2 (the threshold is T2 ′ ), the time point is t3 seconds before the accident.

[0076] Abnormal data item 4: The driver misjudged the distance between vehicles. The abnormal value is d1 and the time point is t4 seconds before the accident.

[0077] Abnormal data item 5: Vehicle speed exceeds threshold V1 ′ , the time point is t5 seconds before the accident.

[0078] Abnormal data item 6: The vehicle's sideslip angle exceeds the threshold θ1 ′ , the time point is t6 seconds before the accident.

[0079] Abnormal data item 7: The sensor output data is always empty or a fixed value (such as zero) and there is no other alarm signal. The abnormal time point is t7 seconds before the accident.

[0080] Abnormal data item 8: Active safety system intervention time exceeds threshold T3 ′ , risk control failed to be completed in time, and the abnormal time point was t8 seconds before the accident.

[0081] Abnormal data item 9: The system detects software fault code E1 ′ , resulting in partial functional failure, the time point is t before the accident g Second.

[0082] Abnormal data item 10: The exit time of the assisted driving function is T4, and the start time of the driver taking over is T5. The time difference exceeds the threshold value T4. ′ , the time point is t before the accident 10 Second.

[0083] Determine the second-level cause, that is, the abnormal situation in the emergency avoidance stage. If T1>T1', it means that the driver's takeover time has exceeded, and the "driver takes over too late" node is confirmed to be abnormal.

[0084] Further check its subordinate nodes such as "driver fatigue status data" and "driver sight deviation data". If there is abnormal data, mark the subordinate node as abnormal. For example, if the driver's sight is deviated for a long time, mark the "driver distraction" node as abnormal. If there is no abnormal data, further trace the third-level cause, that is, the abnormal situation in the event response stage, determine the third-stage abnormal point based on the abnormal data of this stage, trace it down layer by layer until the lowest level abnormal cause data item is found, and record the path completely. If multiple paths are found to have abnormal nodes, integrate multiple fault path chains, analyze whether there is a correlation, and generate a fault analysis report.

[0085] In summary, the method of tracing the evidence chain of assisted driving vehicle accidents based on the fault tree of the present invention can trace the cause of the accident more systematically and accurately, and enhance the objectivity and reliability of accident analysis. By constructing an accident fault tree model, the accident process of the assisted driving vehicle is subdivided into three stages: event response, emergency avoidance, and impending collision, and the key factors and causal chains of each stage are analyzed layer by layer. In qualitative analysis, the minimum cut set that causes the accident is identified by Boolean algebra simplification, and the basic event combination and its path to cause the accident are clarified, thereby forming a clear chain of evidence. Quantitative analysis further calculates the degree of influence of each basic event on the top event, providing a scientific basis for the determination of accident responsibility. At the same time, combined with the on-board electronic data acquisition equipment, the assisted driving function and driver status data are recorded in real time, and the fault tree model is used to determine the source of the abnormal data and its correlation with the accident, so as to achieve accurate tracing from the data level. In general, the present invention greatly improves the systematicness and accuracy of the accident evidence chain, and is suitable for efficient tracing and responsibility analysis of assisted driving vehicle accidents.

[0086] Example 3

[0087] This embodiment is to implement the above method embodiment. The assisted driving vehicle accident evidence chain tracing system based on the fault tree mainly includes:

[0088] The data acquisition module is used to acquire the time series data including the auxiliary driving function status data, vehicle operation data and driver status for a certain period of time before the accident when a collision accident occurs;

[0089] An abnormal data generation module is used to analyze time series data and obtain a list of abnormal data items based on a predetermined threshold;

[0090] The fault tree analysis module inputs the abnormal data item list into a pre-built fault tree, which includes a top event, a first-layer intermediate event, a second-layer intermediate event, and a third-layer intermediate event, and each layer of intermediate events corresponds to multiple basic events, and each event is connected layer by layer with AND gates and OR gates to form multiple accident chains; the top event is a collision accident of an assisted driving vehicle, the first-layer intermediate events are multiple events in the impending collision stage, the second-layer intermediate events are events in the emergency avoidance stage, and the third-layer intermediate events are events in the event response stage; the accident chain corresponding to the abnormal data item list is judged according to the logic of the fault tree, and the basic events of the accident chain are finally determined, and the fault of the driver or the assisted driving system is judged according to the basic events;

[0091] The report generation module is used to record the judgment path of the fault tree and generate a fault analysis report.

[0092] The data acquisition module can be based on the electronic data reading equipment on the assisted driving car. The data reading equipment mainly includes the customized car driving recorder (built-in GPS), DMS and ADAS equipment. The acquired data includes the vehicle assisted driving function activation status, vehicle operation data, driver status data, etc. The data of the assisted driving function status mainly includes data such as adaptive cruise control (ACC), anti-lock braking system (ABS) and emergency braking system (AEB). The vehicle operation data mainly includes data such as vehicle speed, acceleration, brake pedal, etc. The driver status data mainly includes data such as takeover time and line of sight status. After acquiring the data, the integrity of the data can be verified first to ensure that each data has an accurate timestamp and complete record.

[0093] Each module is mainly used to implement each step of the above method embodiment, which will not be described in detail here.

[0094] Example 4

[0095] The present application also provides a computer-readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a disk, an optical disk, a server, an App application store, etc., on which a computer program is stored, and the program is executed by a processor to implement the corresponding function. The computer-readable storage medium of this embodiment is executed by a processor to implement the method embodiment of the assisted driving vehicle accident evidence chain tracing method based on a fault tree.

[0096] It should be pointed out that, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, and two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0097] The order of execution of each step in the above embodiment does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0098] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree, characterized in that: The following steps are involved: When a collision accident occurs, obtain time series data including auxiliary driving function status data, vehicle operation data and driver status for a certain period of time before the accident; Analyze time series data and obtain a list of abnormal data items based on pre-determined thresholds; Input the list of abnormal data items into a pre-built fault tree, which includes a top event, a first-level intermediate event, a second-level intermediate event, and a third-level intermediate event, and each intermediate event corresponds to multiple basic events, and each event is connected layer by layer with an AND gate and an OR gate to form multiple accident chains; the top event is a collision accident of an assisted driving vehicle, the first-level intermediate event is multiple events in the impending collision stage, the second-level intermediate event is an event in the emergency avoidance stage, and the third-level intermediate event is an event in the event response stage; According to the logic of the fault tree, the accident chain corresponding to the abnormal data item list is judged, and the basic events of the accident chain are finally determined. According to the basic events, it is judged whether it is the fault of the driver or the auxiliary driving system; Record the judgment path of the fault tree and generate a fault analysis report.

2. The method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to claim 1 is characterized in that: The first level of events includes the vehicle motion state remaining unchanged and the vehicle motion state not changing enough, both of which are judged based on preset thresholds.

3. The method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to claim 1 is characterized in that: The second level of events includes the driver not taking over, the active safety control system not intervening, the driver taking over in time but failing to avoid the risk, the driver taking over too late, and the active safety system intervening but failing to avoid the risk. All are judged based on preset thresholds.

4. The method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to claim 1 is characterized in that: The third level of events includes driver failure to intervene, warning system failure to activate, assisted driving function failure to adjust vehicle speed, assisted driving function failure to adjust vehicle speed in a timely manner, assisted driving function launched and requested to take over but driver mistakenly took over, and warning system activated too late, all of which are judged based on preset thresholds.

5. The method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to claim 1 is characterized in that: Basic events include driver fatigue driving, driver distraction, obstructed driver vision, sensors failing to detect danger, software system failure, mechanical failure of the braking system, the assisted driving function being launched and requesting to take over but the driver taking over incorrectly, insufficient friction, worn brake pads, driver misjudgment of safe distance, overly conservative active safety system algorithms and active safety system design problems, all of which are judged based on preset thresholds.

6. The method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to claim 1 is characterized in that: The event response stage is specifically the process from when the vehicle ahead slows down or an obstacle appears ahead, the assisted driving function has not yet exited and has time to make adjustments, to when the assisted driving function exits and reminds the driver to take over; The emergency avoidance stage is specifically the transition stage from the exit of the assisted driving function to the activation of the active safety control system, until the active safety control system has reached the maximum braking or steering force; the impending collision stage is specifically when the active safety control system has reached its maximum effect but still cannot avoid a collision.

7. The method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to any one of claims 1 to 6, characterized in that: The assisted driving function status data mainly includes data from the adaptive cruise control ACC, anti-lock braking system ABS and emergency braking system AEB. The vehicle operation data mainly includes vehicle speed, acceleration and brake pedal data. The driver status data mainly includes takeover time and vision status data.

8. The method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree according to any one of claims 1 to 6, characterized in that: Specifically, the cause of the impending collision stage when the accident occurs is judged based on the vehicle operating status data. After the cause of the impending collision stage is determined, the key data items collected before the accident are compared with the threshold of the predetermined benchmark data to determine whether they exceed the normal range. If they exceed the normal data range, they are considered abnormal, and the abnormal values ​​and corresponding time points are recorded to generate a list of abnormal data items.

9. A fault tree-based assisted driving vehicle accident evidence chain tracing system, characterized in that: include: The data acquisition module is used to acquire the time series data including the auxiliary driving function status data, vehicle operation data and driver status for a certain period of time before the accident when a collision accident occurs; An abnormal data generation module is used to analyze time series data and obtain a list of abnormal data items based on a predetermined threshold; The fault tree analysis module inputs the abnormal data item list into a pre-built fault tree, which includes a top event, a first-level intermediate event, a second-level intermediate event, and a third-level intermediate event, and each intermediate event corresponds to multiple basic events, and each event is connected layer by layer with an AND gate and an OR gate to form multiple accident chains; the top event is a collision accident of an assisted driving vehicle, the first-level intermediate event is multiple events in the impending collision stage, the second-level intermediate event is an event in the emergency avoidance stage, and the third-level intermediate event is an event in the event response stage; According to the logic of the fault tree, the accident chain corresponding to the abnormal data item list is judged, and the basic events of the accident chain are finally determined. According to the basic events, it is judged whether it is the fault of the driver or the auxiliary driving system; The report generation module is used to record the judgment path of the fault tree and generate a fault analysis report.

10. A computer storage medium, characterized in that: A computer program executable by a processor is stored therein, and the computer program executes the method for tracing the evidence chain of an assisted driving vehicle accident based on a fault tree as described in any one of claims 1 to 8.

Images