WO2022100174A1

WO2022100174A1 - Detection method and apparatus for detecting accident of vehicle being stuck in ditch, server, and storage medium

Info

Publication number: WO2022100174A1
Application number: PCT/CN2021/112249
Authority: WO
Inventors: 龙荣深; 何锐邦
Original assignee: 广州小鹏汽车科技有限公司
Priority date: 2020-11-11
Filing date: 2021-08-12
Publication date: 2022-05-19
Also published as: CN112406767A; CN112406767B

Abstract

A detection method and apparatus for detecting an accident of a vehicle being stuck in a ditch, a server, and a storage medium. The detection method comprises: acquiring a vehicle driving signal; determining, according to the vehicle driving signal, a stationary pose of a vehicle so as to determine whether the stationary pose is abnormal; if the stationary pose is abnormal, acquiring a driver behavior signal and a vehicle fault signal; and determining whether an accident of the vehicle being stuck in a ditch has occurred according to the stationary pose, the driver behavior signal and the vehicle fault signal. In the detection method and apparatus for a vehicle, a server, and a storage medium, whether an accident of a vehicle being stuck in a ditch has occurred is determined according to a vehicle driving signal, a driver behavior signal and a vehicle fault signal. Since original equipment of a vehicle is used to perform detection on an accident of the vehicle being stuck in a ditch, data processing load of a vehicle terminal can be reduced. Moreover, when a vehicle is stuck in a ditch, active identification of the accident and corresponding processing can be performed, thereby improving after-sales service of vehicles, and improving user experience.

Description

Monitoring method, monitoring device, server and storage medium for vehicle sinkhole accident

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to Chinese Application No. 202011255021.X, filed on November 11, 2020, which is hereby incorporated by reference in its entirety for all purposes.

technical field

The present application relates to the technical field of vehicles, and in particular, to a monitoring method, monitoring device, server and storage medium for a vehicle sinkhole accident.

Background technique

When driving in bad weather or road conditions, the vehicle may have a sinkhole accident. After the accident, it is often impossible to rely on the driver to solve it. Active detection and timely rescue after the accident can provide great help to the driver. In the related art, additional hardware for collecting data for sinkhole accidents is often used in the vehicle body, and data processing needs to be performed at the vehicle end, which increases the data processing load at the vehicle end.

SUMMARY OF THE INVENTION

In view of this, embodiments of the present application provide a monitoring method, a monitoring device, a server, and a storage medium for a vehicle sinkhole accident.

The application provides a monitoring method for a vehicle sinkhole accident, the monitoring method comprising:

Obtain vehicle driving signals;

Judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal;

In the case that the parking posture is abnormal, obtain the driver behavior signal and the vehicle fault signal;

Whether the sinkhole accident occurs to the vehicle is determined according to the parking posture, the driver behavior signal and the vehicle failure signal.

In some embodiments, the obtaining the vehicle driving signal includes:

The vehicle driving signal of each unit time in the first time period from the first predetermined time before the current time to the current time is acquired.

In some implementations, the judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal includes;

The vehicle driving signal is identified according to the pre-stored parking attitude model to judge the parking attitude of the vehicle to determine whether the parking attitude is abnormal.

In some embodiments, the vehicle driving signal includes a real-time three-axis acceleration signal of the vehicle and a speed signal of each wheel of the vehicle, and the vehicle driving signal is identified according to a pre-stored parking attitude model to Judging the parking posture of the vehicle to determine whether the parking posture is abnormal includes:

Determine whether the vehicle is in a parking state according to the speed signal;

When the vehicle is in the parking state, determining the attitude angle of the vehicle per unit time in the first time period according to the three-axis acceleration signal;

The attitude angle is processed according to the parking attitude model to obtain the parking attitude of the vehicle.

In some embodiments, in the case that the parking posture is abnormal, acquiring the driver behavior signal and the vehicle fault signal includes;

In the case that the parking posture is abnormal, obtain the driver behavior signal of each unit time in the second time period from the second predetermined time before the current time to the third predetermined time after the current time;

Acquire a vehicle fault signal per unit time in the second time period.

In some embodiments, the determining whether the vehicle has the sinkhole accident according to the parking posture, the driver behavior signal and the vehicle fault signal includes:

The driver's behavior signal is identified according to a pre-stored abnormal driver behavior model to determine the abnormal level of the driver's behavior.

In some embodiments, the driver behavior signal includes a vehicle lock signal, a double flashing light signal, a gear position signal, an accelerator pedal signal, a main driving door signal, and a steering wheel angle signal. The behavior model identifies the driver behavior signal to determine the abnormal level of driver behavior, including:

Determine whether there is a first abnormal behavior of unlocking the vehicle within a predetermined period of time after getting off the vehicle according to the main driving door signal and the vehicle locking signal;

Determine whether there is a second abnormal behavior of turning on the double flashing lights according to the gear signal and the double flashing light signal;

judging whether there is a third abnormal behavior of slamming the accelerator pedal according to the accelerator pedal signal;

Determine whether there is a fourth abnormal behavior of repeatedly operating the steering wheel according to the steering wheel angle signal;

According to the main driving door signal and the steering wheel angle signal, determine whether there is a fifth abnormal behavior of the steering wheel not returning to the right position after getting off the car;

The abnormal level of the driver's behavior is judged according to a preset rating rule, the first abnormal behavior, the second abnormal behavior, the third abnormal behavior, the fourth abnormal behavior and the fifth abnormal behavior .

The parking posture, the abnormal level of the driver's behavior and the vehicle fault signal are identified according to a pre-stored vehicle sinkhole model to determine whether the vehicle has the sinkhole accident.

In some embodiments, the vehicle fault signal includes a slip signal, a chassis fault signal, a tire pressure monitoring system fault signal, and an electrical system fault signal, and the parking attitude, the driving, Identifying the abnormal level of employee behavior and the vehicle fault signal to determine whether the vehicle has the sinkhole accident includes:

According to the pre-stored vehicle sinkhole model, the feature vector formed by the parking posture, the abnormal level of the driver's behavior and the vehicle fault signal is processed to obtain the probability of the vehicle having the sinkhole accident;

It is determined that the sinkhole accident occurs to the vehicle when the probability is greater than a predetermined threshold.

In certain embodiments, the monitoring method further includes:

When it is determined that the vehicle has the sinkhole accident, an alarm signal is sent to a service provider of the vehicle so that the service provider can implement rescue according to the alarm signal.

The application provides a monitoring device for a vehicle, and the monitoring device includes:

an acquisition module, the acquisition module is used to acquire the vehicle driving signal;

a judging module, which is used for judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal;

The obtaining module is further configured to obtain the driver behavior signal and the vehicle fault signal when the parking posture is abnormal;

The judging module is further configured to judge whether the vehicle has the sinkhole accident according to the parking posture, the driver's behavior signal and the vehicle fault signal.

The present application provides a server, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to implement the method for monitoring a vehicle sinkhole accident according to any of the foregoing embodiments when the computer program is executed.

The present application provides one or more non-volatile computer-readable storage media storing a computer program, when the computer program is executed by one or more processors, the monitoring of the vehicle sinkhole accident in any of the above embodiments is implemented method.

In the monitoring method, monitoring device, server and storage medium for a vehicle sinkhole accident according to the embodiments of the present application, it is possible to use the original equipment of the vehicle to determine whether a vehicle sinkhole accident has occurred according to the vehicle driving signal, the driver's behavior signal and the vehicle fault signal. The monitoring of vehicle sinkhole accidents can reduce the data processing load on the vehicle end, and can actively identify the occurrence of accidents after the vehicle sinks in, and carry out corresponding processing, improving vehicle after-sales service and enhancing user experience.

Description of drawings

The above and/or additional aspects and advantages of the present application will become apparent and readily understood from the following description of embodiments taken in conjunction with the accompanying drawings, wherein:

FIG. 1 is a schematic flowchart of a monitoring method for a vehicle sinkhole accident according to some embodiments of the present application.

FIG. 2 is a schematic block diagram of a monitoring device for a vehicle according to some embodiments of the present application.

FIG. 3 is a schematic flowchart of a monitoring device for a vehicle according to some embodiments of the present application.

FIG. 4 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

FIG. 5 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

FIG. 6 is a schematic diagram of three-axis coordinates of some embodiments of the present application.

FIG. 7 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

FIG. 8 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

FIG. 9 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

FIG. 10 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

FIG. 11 is a model schematic diagram of a vehicle monitoring method according to some embodiments of the present application.

FIG. 12 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

FIG. 13 is a schematic flowchart of a vehicle monitoring method according to some embodiments of the present application.

Detailed ways

The following describes in detail the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary, and are intended to be used to explain the present application, but should not be construed as a limitation to the present application.

Referring to FIG. 1, the present application provides a method for monitoring a vehicle sinkhole accident, including:

S11: Obtain the vehicle driving signal;

S12: Judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal;

S13: In the case of abnormal parking posture, obtain the driver's behavior signal and the vehicle fault signal;

S14: Determine whether the vehicle has a sinkhole accident according to the parking posture, the driver's behavior signal and the vehicle fault signal.

The embodiment of the present application provides a server 100 . Server 100 includes processor 104 . The processor 104 is used for acquiring the vehicle driving signal, and for judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal, and for acquiring the driver behavior signal and the vehicle fault signal when the parking posture is abnormal, And it is used to judge whether the vehicle has a sinkhole accident according to the parking posture, the driver's behavior signal and the vehicle fault signal. Wherein, the processor 104 may be the processor 104 independently set for implementing the monitoring method for the vehicle sinkhole accident, or may be the processor 104 of the server 100 itself, which is not limited herein.

Referring to FIG. 2 , an embodiment of the present application further provides a monitoring device 110 for a vehicle. The monitoring method for a vehicle sinkhole accident in the embodiment of the present application may be implemented by the monitoring device 110 for a vehicle. The monitoring device 110 of the vehicle includes an acquisition module 112 and a determination module 114 . S11 and S13 may be implemented by the acquisition module 112 , and S12 and S14 may be implemented by the determination module 114 . In other words, the acquisition module 112 is used to acquire the vehicle driving signal, and to acquire the driver behavior signal and the vehicle fault signal when the parking posture is abnormal. The judging module 114 is used for judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal, and for judging whether the vehicle has a sinkhole accident according to the parking posture, the driver behavior signal and the vehicle fault signal.

Specifically, when driving in bad weather, or when the driving road conditions are poor, the vehicle may have a sinkhole accident, and the wheels will sink into the snow, mud pits, ditch or manhole cover, etc., so that the vehicle cannot continue to drive. After a sinkhole accident, it is usually necessary to use the help of foreign objects to get the vehicle out of the predicament, which is often difficult to solve only by the driver.

In the related art, a camera system is used to identify road conditions, a photosensitive system is used to detect road conditions, or a radar system is used for ranging, so as to prevent vehicle sinkhole accidents. The above solutions all need to add additional hardware for data collection for sinkhole accidents, such as camera systems, radar systems, etc., which increase vehicle production costs. In addition, the data collected by the above solution needs to be processed on the vehicle side, which also increases the data processing load on the vehicle side, which will slow down the response speed of the vehicle processor to a certain extent, and the user experience will be poor.

Compared with the solution of using the camera system, the photosensitive system and the radar system to realize the early warning of the vehicle sinkhole, in the monitoring method, the server 100 and the vehicle monitoring device 110 of the vehicle sinkhole accident according to the embodiment of the present application, the vehicle is judged by acquiring the vehicle driving signal. The parking attitude is determined to determine whether the parking attitude of the vehicle is abnormal, and in the case of abnormal parking attitude of the vehicle, the driver behavior signal and the vehicle fault signal are obtained. Then, according to the vehicle driving signal, driver behavior signal and vehicle fault signal, comprehensively determine whether the vehicle has a sinkhole accident. In this way, only the original equipment of the vehicle can be used to monitor the vehicle sinkhole accident. The data size and the amount of data calculation are small, which can reduce the data processing load on the vehicle end, and can actively identify the accident after the vehicle sinks. occurrence, and deal with it accordingly, improve the after-sales service of the vehicle and enhance the user experience.

Referring to Figure 3, in some embodiments, S11 includes:

S111: Acquire a vehicle driving signal of each unit time in a first time period from a first predetermined time before the current time to the current time.

In some embodiments, S111 may be implemented by the obtaining module 112 . In other words, the obtaining module 112 is configured to obtain the vehicle running signal of each unit time in the first time period from the first predetermined time before the current time to the current time.

In some embodiments, the processor 104 is configured to acquire the vehicle running signal of each unit time in the first time period from the first predetermined time before the current time to the current time.

Specifically, during the running of the vehicle, the vehicle running signal of each unit time in the first time period is obtained, and the parking posture of the vehicle is judged according to the vehicle running signal, so as to determine whether the parking posture of the vehicle is abnormal. The first time period may be a time range from a predetermined time t11 before the current time t to the current time t, or a time range from a predetermined time t11 before the current time t to a predetermined time t12 after the current time t.

In some embodiments, the first time period may be a time range from a predetermined time t11 before the current time t to the current time t. During the driving process of the vehicle, obtain the vehicle driving signal of each unit time in the time range of t11-t, analyze and process the vehicle driving signal of each unit time, and judge the parking posture of the vehicle, so as to determine whether the parking posture of the vehicle is not. abnormal.

In this way, the parking posture of the vehicle at the current time and the parking posture of the vehicle before the current time can be compared and analyzed, the parking posture of the vehicle at the current time can be accurately judged, and the accuracy of analyzing the result of the pit accident can be ensured.

In other embodiments, the first time period may be a time range from a predetermined time t11 before the current time t to a predetermined time t12 after the current time t. During the driving process of the vehicle, obtain the vehicle driving signal of each unit time in the time range of t11-t12, analyze and process the vehicle driving signal of each unit time, and judge the parking posture of the vehicle, so as to determine whether the parking posture of the vehicle is not. abnormal.

In this way, the vehicle parking attitude at the current time, the vehicle parking attitude before the current time, and the vehicle parking attitude after the current time can be compared and analyzed, the vehicle parking attitude at the current time can be accurately judged, and the accuracy of the analysis result of the sinkhole accident can be ensured.

It should be noted that the length of the time range of the first time period can be set according to factors such as road conditions, vehicle service life, vehicle maintenance records, vehicle performance, etc., and there is no specific limitation, for example, it can be 3 seconds, 5 seconds, 10 seconds, 13 seconds , 17 seconds, etc. Correspondingly, the predetermined time t11 before the current time t and the predetermined time t12 after the current time t can be set according to factors such as road conditions, vehicle service life, vehicle maintenance records, vehicle performance, etc. 3 seconds, 5 seconds, 8 seconds, 10 seconds, etc., the specific values of t11 and t12 may be equal or unequal.

Referring to FIG. 4, in some embodiments, S12 includes:

S121 : Identify the vehicle driving signal according to the pre-stored parking attitude model to determine the parking attitude of the vehicle to determine whether the parking attitude is abnormal.

In some embodiments, S121 may be implemented by the judgment module 114 . In other words, the judging module 114 is configured to identify the driving signal of the vehicle according to the pre-stored parking posture model to judge the parking posture of the vehicle to determine whether the parking posture is abnormal.

In some embodiments, the processor 104 is configured to identify the vehicle driving signal according to the pre-stored parking posture model to judge the parking posture of the vehicle and determine whether the parking posture is abnormal.

Specifically, the vehicle driving signal can be recognized according to the pre-stored parking attitude model, and the parking attitude of the vehicle can be determined, thereby determining whether the parking attitude is abnormal. The specific processing method of the parking attitude model can be selected according to the data category, data volume and other factors, which is not limited. For example, it can be a gradient boosting tree (Gradient Boosting Decision Tree, GBDT) algorithm or a support vector machine (Support Vector Machine, SVM) algorithm, it can also be a regression algorithm, etc.

The identification process of the vehicle running signal is developed and stored in the form of a model, so that the development efficiency of the monitoring device 110 can be improved. It is also easy to view and modify in case of failure during subsequent use. After the model development is completed, based on the high model reuse rate, the parking attitude model can be applied to a variety of vehicle models and/or systems, reducing development costs.

Referring to FIG. 5 , in some embodiments, the vehicle driving signal includes the real-time three-axis acceleration signal of the vehicle and the speed signal of each wheel of the vehicle, and S121 includes:

S1211: Determine whether the vehicle is in a parked state according to the speed signal;

S1212: When the vehicle is in a parked state, determine the attitude angle of the vehicle per unit time in the first time period according to the three-axis acceleration signal;

S1213: Process the attitude angle according to the parking attitude model to obtain the parking attitude of the vehicle.

In some embodiments, S1211-S1213 may be implemented by the determination module 114 . In other words, the determination module 114 is configured to determine whether the vehicle is in a parked state according to the speed signal, and to determine the attitude angle of the vehicle per unit time in the first time period according to the three-axis acceleration signal when the vehicle is in a parked state, and It is used to process the attitude angle according to the parking attitude model to obtain the parking attitude of the vehicle.

In some embodiments, the processor 104 is configured to determine whether the vehicle is in a parked state according to the speed signal, and to determine the speed of the vehicle per unit time in the first time period according to the three-axis acceleration signal when the vehicle is in a parked state. attitude angle, and is used to process the attitude angle according to the parking attitude model to obtain the parking attitude of the vehicle.

Specifically, whether the vehicle is in a parked state can be determined according to the speed signals of each wheel of the vehicle. Among them, the speed signal can be the speed signal collected from the driving wheel, or the speed signal collected from all normal tires.

In some embodiments, if the vehicle is a front-wheel drive vehicle, the speed signals of the left front wheel and the right front wheel may be collected, and the speed signals of all the wheels in normal use may also be collected. In some embodiments, if the vehicle is a rear-wheel-drive vehicle, the speed signals of the left rear wheel and the right front wheel may be collected, and the speed signals of all the wheels in normal use may also be collected. In other embodiments, if the vehicle is a four-wheel drive vehicle, the speed signals of all the wheels in normal use, including the driven wheels and the driving wheels, can be collected.

In this way, by identifying the speed signals of each vehicle in the vehicle, it can more accurately reflect whether the vehicle is in a parked state, thereby ensuring the accuracy of the signals.

In addition, it can also be judged whether the vehicle is in a parked state according to the Global Positioning System (GPS). When the GPS shows that the vehicle position moves, it can be considered that the vehicle is not in a parked state. When the GPS shows that the position of the vehicle remains unchanged, it can be considered that the vehicle is in a parked state.

When it is judged that the vehicle is in a parked state, the three-axis acceleration signal of the vehicle per unit time in the first period of time is obtained, the attitude angle of the vehicle can be determined according to the three-axis acceleration signal, and the attitude angle is processed by using the parking attitude model to obtain The parking status of the vehicle. Referring to FIG. 6 , specifically, the three-axis acceleration signal includes a pitch angle signal of the x-axis, a yaw angle signal of the y-axis, and a roll angle signal of the z-axis. Among them, the pitch angle pitch can represent the angle between the x-axis of the vehicle coordinate system and the horizontal plane, the yaw angle yaw can represent the angle between the y-axis of the vehicle coordinate system and the horizontal plane, and the roll angle roll can represent the vehicle coordinate system. The angle between the z-axis and the horizontal plane .

In some embodiments, the yaw angle yaw and the roll angle roll can be measured using the LIS3DH three-axis accelerometer. According to the calculation formula of pitch angle pitch=arctan(-y,z)*180/3.14159 and the calculation formula of roll angle roll=arctan(x,z)*180/3.14159, each unit of the vehicle in the first time period can be obtained Time yaw angle yaw and roll angle roll.

When judging that the vehicle is in a parked state, the attitude angle of the vehicle per unit time in the first time period is obtained, and each obtained attitude angle is input into the parking attitude model, and the attitude angle is processed by the parking attitude model to output whether the vehicle has occurred. There is no abnormal probability of leaning forward, leaning backward, leaning left, leaning right, or the vehicle is parked.

For example, after the parking attitude model processes the input pitch angle, the parking state of the vehicle is obtained: the probability of the vehicle tilting forward is 0.2, the probability of the vehicle tilting backward is 0.9, and the probability of no abnormality is 0.1. Then, according to the maximum value of the above results, it can be determined that the vehicle is tilted backwards at the current time.

For another example, after the parking attitude model processes the input roll angle, the parking state of the vehicle is obtained: the probability of the vehicle leaning to the left is 0.7, the probability of the vehicle leaning right is 0.4, and the probability of no abnormality is 0.1. Then, according to the maximum value of the above results, it can be determined that the vehicle leans to the left at the current time.

In this way, the attitude angle of the vehicle per unit time in the first time period can be compared and analyzed, and the parking attitude of the vehicle at the current time can be accurately determined.

Referring to Figure 7, in some embodiments, S13 includes:

S131: In the case of abnormal parking posture, obtain the driver behavior signal of each unit time in the second time period from the second predetermined time before the current time to the third predetermined time after the current time;

S132: Acquire a vehicle fault signal per unit time in the second time period.

In some embodiments, S131 and S132 may be implemented by the obtaining module 112 . In other words, the obtaining module 112 is configured to obtain the driver behavior signal per unit time in the second time period from the second predetermined time before the current time to the third predetermined time after the current time in the case of abnormal parking posture, and use in acquiring the vehicle fault signal per unit time in the second time period.

In some embodiments, the processor 104 is configured to obtain the driver behavior per unit time in the second time period from the second predetermined time before the current time to the third predetermined time after the current time when the parking posture is abnormal signal, and a vehicle fault signal for each unit time in the second time period.

Specifically, in the case of judging that the parking posture is abnormal, the driver behavior signal and the vehicle fault signal of each unit time in the second time period are obtained, and according to the parking posture, the driver behavior signal and the vehicle fault signal, it is judged whether the vehicle has occurred. sinkhole accident. The second time period may be a time range from a predetermined time t21 before the current time t to the current time t, or a time range from a second predetermined time t21 before the current time t to a predetermined time t22 after the current time t.

In some embodiments, the second time period may be a time range from a predetermined time t21 before the current time t to the current time t. In the case of judging that the parking posture is abnormal, obtain the driver behavior signal and vehicle fault signal per unit time within the time range of t21-t, and analyze and process the driver behavior signal and vehicle fault signal per unit time , so as to determine whether the vehicle has a sinkhole accident.

In this way, it is possible to compare and analyze the driver's behavior at the current time and the driver's behavior before the current time, accurately determine whether the driver's behavior at the current time is abnormal, and compare the vehicle fault signal at the current time with the vehicle fault signal before the current time. Analysis, accurately determine whether the vehicle is faulty at the current time, and ensure the accuracy of the results of the analysis of the sinkhole accident.

In other embodiments, the second time period may be a time range from a second predetermined time t21 before the current time t to a predetermined time t22 after the current time t. In the case of judging that the parking posture is abnormal, obtain the driver behavior signal and vehicle fault signal per unit time within the time range of t21-t22, and analyze and process the driver behavior signal and vehicle fault signal per unit time , so as to determine whether the vehicle has a sinkhole accident.

In this way, it is possible to compare and analyze the driver's behavior before the current time, the current time and after the current time, accurately determine whether the driver's behavior at the current time is abnormal, and compare the vehicle fault signals before the current time, the current time and after the current time. Analysis, accurately determine whether the vehicle is faulty at the current time, and ensure the accuracy of the results of the analysis of the sinkhole accident.

It should be noted that the length of the time range of the second time period can be set according to factors such as road conditions, vehicle service life, vehicle maintenance records, vehicle performance, etc., and is not specifically limited, for example, it can be 10 seconds, 30 seconds, 50 seconds, 60 seconds , 100 seconds, etc. The second predetermined time t21 before the current time t and the third predetermined time t22 after the current time t can be set according to factors such as road conditions, vehicle age, vehicle maintenance records, vehicle performance, etc. Seconds, 30 seconds, 50 seconds, 80 seconds, etc., the specific values of t21 and t22 may be equal or unequal.

Referring to FIG. 8, in some embodiments, S14 includes:

S141: Identify the driver's behavior signal according to the pre-stored abnormal driver behavior model to determine the abnormal level of the driver's behavior.

In some embodiments, S141 may be implemented by the judgment module 114 . In other words, the judging module 114 is configured to identify the driver's behavior signal according to the pre-stored abnormal driver behavior model to judge the abnormal level of the driver's behavior.

In some embodiments, the processor 104 is configured to identify the driver's behavior signal according to the pre-stored abnormal driver behavior model to determine the abnormal level of the driver's behavior.

Specifically, the driver's behavior signal can be identified according to the pre-stored abnormal driver behavior model, to determine whether the driver's behavior is abnormal, and to evaluate the abnormal level of the driver's behavior, so as to determine whether the vehicle has a sinkhole accident. The specific processing method of the driver's abnormal behavior model can be selected according to the data type, data volume and other factors, and there is no specific limitation. For example, it can be a counting model, a GBDT algorithm, an SVM algorithm, or a regression algorithm.

The identification process of the driver's behavior signal is developed and stored in the form of a model, so that the development efficiency of the monitoring device 110 can be improved. It is also easy to view and modify in case of failure during subsequent use. After the model development is completed, based on the characteristics of high model reuse rate, the abnormal driver behavior model can be applied to a variety of vehicle models and/or systems, reducing development costs.

Referring to FIG. 9 , in some embodiments, the driver behavior signal includes a vehicle lock signal, a double flashing light signal, a gear position signal, an accelerator pedal signal, a main driving door signal and a steering wheel angle signal, and S141 includes:

S1411: Determine whether there is a first abnormal behavior of unlocking the car within a predetermined period of time after getting off the car according to the main driving door signal and the vehicle locking signal;

S1412: Determine whether there is a second abnormal behavior of turning on the double flashing light according to the gear signal and the double flashing light signal;

S1413: Determine whether there is a third abnormal behavior of slamming the accelerator pedal according to the accelerator pedal signal;

S1414: Determine whether there is a fourth abnormal behavior of repeatedly operating the steering wheel according to the steering wheel angle signal;

S1415: According to the main driving door signal and the steering wheel angle signal, determine whether there is a fifth abnormal behavior of the steering wheel not returning to the right position after getting off the car;

S1416: Determine the abnormal level of the driver's behavior according to the preset rating rule, the first abnormal behavior, the second abnormal behavior, the third abnormal behavior, the fourth abnormal behavior, and the fifth abnormal behavior.

In some embodiments, S1411-S1416 may be implemented by the determination module 114 . In other words, the judging module 114 is used for judging whether there is a first abnormal behavior of not locking the car within a predetermined period of time after getting off the car according to the main driving door signal and the vehicle locking signal, and for judging whether it is not according to the gear signal and the double flashing light signal. There is a second abnormal behavior of turning on the double flashing lights, and a third abnormal behavior for judging whether there is a slam on the accelerator pedal according to the accelerator pedal signal, and a fourth abnormal behavior for judging whether there is a repeated steering wheel operation according to the steering wheel angle signal, And it is used to judge whether there is a fifth abnormal behavior in which the steering wheel is not returned after getting off the car according to the main driving door signal and the steering wheel angle signal, and is used to judge whether there is a fifth abnormal behavior according to the preset rating rules, the first abnormal behavior, the second abnormal behavior, the first abnormal behavior, the first abnormal behavior The third abnormal behavior, the fourth abnormal behavior and the fifth abnormal behavior determine the abnormal level of the driver's behavior.

In some embodiments, the processor 104 is configured to determine whether there is a first abnormal behavior in which the vehicle is not locked within a predetermined period of time after getting off the vehicle according to the main driving door signal and the vehicle locking signal, and is configured to determine whether there is a first abnormal behavior in which the vehicle is not locked within a predetermined period of time after getting off the vehicle, and is configured to determine whether there is a first abnormal behavior in which the vehicle is not locked within a predetermined period of time after getting off the vehicle, and is configured to determine whether there is a first abnormal behavior in which the vehicle is not locked within a predetermined period of time after getting off the vehicle, and is configured to use the gear signal and the double flashing The light signal judges whether there is a second abnormal behavior of turning on the double flashing lights, and the third abnormal behavior for judging whether there is a hard pressing of the accelerator pedal according to the accelerator pedal signal, and the third abnormal behavior for judging whether the steering wheel is repeatedly operated according to the steering wheel angle signal. Four abnormal behaviors, and a fifth abnormal behavior for judging whether there is a steering wheel not returning after getting off the car according to the main driving door signal and the steering wheel angle signal, and a fifth abnormal behavior for judging whether the steering wheel does not return after getting off the car, and for according to the preset rating rules, the first abnormal behavior, the second abnormal behavior The abnormal behavior, the third abnormal behavior, the fourth abnormal behavior, and the fifth abnormal behavior determine the abnormal level of the driver's behavior.

Specifically, according to the main driving door signal and the vehicle locking signal, it is determined whether the driver has a first abnormal behavior X1 in which the vehicle is not locked within a predetermined period of time after getting off the vehicle within the second time period. Wherein, within a predetermined time after the main driving door is opened or closed, the vehicle lock signal is 0, that is, the vehicle is not locked. At this time, it can be determined that the driver has the first abnormal behavior X1 in the second time period.

When the first abnormal behavior X1 exists, the abnormal level of driver behavior is set to 1, and when the first abnormal behavior X1 does not exist, the abnormal level of driver behavior is set to 0.

In the second time period, when the driver has the first abnormal behavior X1, according to the gear signal and the double flashing light signal, it is judged whether the driver has the second abnormal behavior X2 of turning on the double flashing light in the second time period . Wherein, the current gear of the vehicle is the parking gear, and the double flashing light signal is 1, that is, the double flashing light is in a light-on state. At this time, it can be determined that the driver has a second abnormal behavior X2 in the second time period.

When the first abnormal behavior X1 and the second abnormal behavior X2 exist, the abnormal level of the driver's behavior is 2. In the case where the first abnormal behavior X1 exists but the second abnormal behavior X2 does not exist, the abnormal level of the driver's behavior is still 1.

In the second time period, when it is determined that the driver has the first abnormal behavior X1 in the second time period, according to the accelerator pedal signal, the main driving door signal and the steering wheel angle signal, it is determined that the driver is in the second time period. Whether there is a third abnormal behavior X3 of pressing the accelerator pedal hard, a fourth abnormal behavior X4 of repeatedly operating the steering wheel, or a fifth abnormal behavior X5 of the steering wheel not returning to the right position after getting off the car. Among them, the number of times a that the depth of the accelerator pedal is stepped over the predetermined depth threshold in the second time period is calculated, and when the value of a is greater than or equal to the first predetermined number of times threshold, it is determined that the driver has a third abnormal behavior X3. Calculate the number of times b that the difference between the steering wheel angle at the current time and the steering wheel angle at a predetermined time before the current time is greater than the predetermined difference threshold. When the value of b is greater than or equal to the second predetermined number of times threshold, it is determined that the driver has a fourth abnormal behavior X4. In the second time period after the main driving door is opened, if the degree of the steering wheel angle exceeds the predetermined angle threshold, it is determined that the driver has a fifth abnormal behavior X5.

In the presence of the first abnormal behavior X1, when the driver has two or more abnormal behaviors among the third abnormal behavior X3, the fourth abnormal behavior X4 and/or the fifth abnormal behavior X5, the abnormal behavior level of the driver is 2. In the presence of the first abnormal behavior X1, but not the third abnormal behavior X3, the fourth abnormal behavior X4 and the fifth abnormal behavior X5, or the existence of the third abnormal behavior X3, the fourth abnormal behavior X4 or the fifth abnormal behavior X5 In the case of less than two types, the abnormal driver behavior level is still 1.

Further, in the second time period, when the driver has the first abnormal behavior X1 and the second abnormal behavior X2, there is a third abnormal behavior X3, the fourth abnormal behavior X4 and/or the fifth abnormal behavior X5. Two or more abnormal behaviors, the driver behavior abnormal level is 3.

For example, in the second time period, if the driver has abnormal behaviors of X3, X4 and X5, the abnormal level of the driver's behavior is 0. If X1 exists, the abnormal level of driver behavior is 1. If X1 and X2 exist, the abnormal level of driver behavior is 2. If X1, X4 and X5 exist, the abnormal level of driver behavior is 2. If X1, X2 and X5 exist, the abnormal level of driver behavior is 2. If there are X1, X3, X4 and X5, the abnormal level of driver behavior is 2. If there are X1, X2, X3, X4 and X5, the abnormal level of driver behavior is 3. And so on.

In this way, the behavior of the driver in the second time period can be compared and analyzed, the abnormal level of the driver's behavior can be accurately judged, and the accuracy of analyzing the result of the sinkhole accident can be ensured.

It should be noted that the above-mentioned thresholds such as the predetermined depth threshold, the first predetermined number of times threshold, the predetermined difference threshold, the second predetermined number of times threshold, and the predetermined angle threshold can all be set according to parameters such as the vehicle model and the driver's behavior habits, and are not specifically limited. For example, the predetermined depth threshold may be 10, 15, 18, etc., the first predetermined number of times threshold and the second predetermined number of times threshold may be 1, 2, 5, etc., and the predetermined difference threshold may be 50, 80, 100, etc., The predetermined angle threshold may be 250 degrees, 300 degrees, 350 degrees, 400 degrees, or the like.

Referring to FIG. 10, in some embodiments, S14 includes:

S142: Identify the parking posture, the abnormal level of the driver's behavior, and the vehicle fault signal according to the pre-stored vehicle sinkhole model to determine whether the vehicle has a sinkhole accident.

In some embodiments, S142 may be implemented by the determination module 114 . In other words, the judging module 114 is configured to identify the parking posture, the abnormal level of the driver's behavior, and the vehicle fault signal according to the pre-stored vehicle sinkhole model, so as to judge whether the vehicle has a sinkhole accident.

In some embodiments, the processor 104 is configured to identify the parking posture, the abnormal level of the driver's behavior and the vehicle fault signal according to the pre-stored vehicle sinkhole model, so as to determine whether the vehicle has a sinkhole accident.

Specifically, referring to FIG. 11 , the parking posture, the abnormal level of driver behavior and the vehicle fault signal can be identified according to the pre-stored vehicle sinkhole model to determine whether the vehicle has a sinkhole accident. The specific processing method of the vehicle pit model can be selected according to the data type, data volume and other factors, and there is no specific limitation. For example, it can be a counting model, a GBDT algorithm, an SVM algorithm, or a regression algorithm.

The identification process of the parking posture, the abnormal level of the driver's behavior and the vehicle fault signal is developed and stored in the form of a model, so that the development efficiency of the monitoring device 110 can be improved. It is also easy to view and modify in case of failure during subsequent use. After the model development is completed, based on the high model reuse rate, the vehicle sinkhole model can be applied to a variety of vehicle models and/or systems, reducing development costs.

Referring to FIG. 12 , in some embodiments, the vehicle fault signal includes a slip signal, a chassis fault signal, a tire pressure monitoring system fault signal and an electrical system fault signal, and S142 includes:

S1421: Process the feature vector formed by the parking posture, the abnormal level of driver behavior and the vehicle fault signal according to the pre-stored vehicle sinkhole model to obtain the probability of a vehicle sinkhole accident;

S1422: When the probability is greater than a predetermined threshold, it is determined that the vehicle has a sinkhole accident.

In some embodiments, S1421 and S1422 may be implemented by the judgment module 114 . In other words, the judging module 114 is configured to process the feature vector composed of the parking posture, the abnormal level of the driver's behavior and the vehicle fault signal according to the pre-stored vehicle sinkhole model to obtain the probability that the vehicle has a sinkhole accident, and is used to obtain the probability that the vehicle has a sinkhole accident when the probability is greater than It is determined that the vehicle has a sinkhole accident when the predetermined threshold value is used.

In some embodiments, the processor 104 is configured to process the feature vector composed of the parking posture, the abnormal level of the driver's behavior and the vehicle fault signal according to the pre-stored vehicle sinkhole model to obtain the probability of the vehicle having a sinkhole accident, and use It is determined that the vehicle has a pit accident when the probability is greater than a predetermined threshold.

Specifically, the vehicle fault signal includes a wheel slip signal, a chassis fault signal, a tire pressure monitoring system fault signal and an electrical system fault signal. The wheel slip signal can be detected by calculating the speed difference between the front wheel and the rear wheel. When the difference between the two is greater than a predetermined speed difference threshold, it can be considered that the wheel slip phenomenon occurs. Chassis fault signal and tire pressure monitoring system fault signal can be detected by vehicle systems such as anti-lock braking system, electronic stability program, automatic parking system and/or tire pressure monitoring system. When any one or more of the monitoring system fault signals is greater than 0, it may be considered that the vehicle chassis and/or wheels are faulty. The electrical system fault signal can be detected by the motor system, battery management system, light control system, radar system and other vehicle systems. When any one or more fault signals in the electrical system are detected to be greater than 0, it can be considered that the vehicle electrical system is faulty .

Comprehensive parking posture, abnormal level of driver behavior and vehicle fault signal constitute a feature vector, and the probability of vehicle pit accident can be obtained by processing the feature vector. When the probability is greater than the predetermined threshold, it is determined that the vehicle has a sinkhole accident.

In this way, by comparing and analyzing the parking posture, the abnormal level of the driver's behavior, and the vehicle fault signal, it is possible to accurately determine whether the vehicle has a sinkhole accident.

It should be noted that the predetermined speed difference threshold, the predetermined threshold for judging the probability of vehicle sinking, etc., can be set according to parameters such as model, driver behavior, vehicle maintenance record, vehicle service life, etc., and there is no specific limitation. For example, the predetermined speed difference The value threshold may be 1 m/s, 2 m/s, 3 m/s, 5 m/s, etc., and the predetermined threshold for judging the probability of vehicle sinking may be 0.5, 0.7, 0.8, etc.

Referring to FIG. 13, in some embodiments, the monitoring method further includes:

S15: When it is determined that the vehicle has a sinkhole accident, send an alarm signal to the service provider of the vehicle so that the service provider can implement rescue according to the alarm signal.

In some embodiments, S15 may be implemented by the determination module 114 . In other words, the judging module 114 is configured to send an alarm signal to a service provider of the vehicle when it is determined that the vehicle has a sinkhole accident, so that the service provider can implement rescue according to the alarm signal.

In some embodiments, the processor 104 is configured to send an alarm signal to a service provider of the vehicle when it is determined that the vehicle has a sinkhole accident, so that the service provider can implement rescue according to the alarm signal.

Specifically, when it is determined that the vehicle has a sinkhole accident, an alarm signal is sent to the service provider of the vehicle, and the service provider can implement rescue according to the alarm signal, or contact the driver to confirm that the vehicle has a sinkhole accident and implement rescue according to the situation.

In this way, after a vehicle sinkhole accident, providing corresponding rescue assistance can improve the after-sales service of the vehicle and enhance the user experience.

The embodiments of the present application also provide a computer-readable storage medium. One or more non-volatile computer-readable storage media storing a computer program, when the computer program is executed by one or more processors, implements the vehicle monitoring method of any one of the above embodiments.

The embodiments of the present application also provide a vehicle. The vehicle includes a memory and one or more processors, and one or more programs are stored in the memory and configured to be executed by the one or more processors. The program includes instructions for performing the monitoring method for a vehicle sinkhole accident of any one of the above embodiments.

The processor can be used to provide the computing and control capabilities that underpin the operation of the entire vehicle. The memory of the vehicle provides an environment for the execution of computer readable instructions in the memory.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the programs can be stored in one or more non-volatile computer-readable storage media , when the program is executed, it may include the processes of the foregoing method embodiments. The storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), or the like.

The above examples only represent several embodiments of the present application, and the descriptions thereof are relatively specific and detailed, but should not be construed as a limitation on the scope of the patent of the present application. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the scope of protection of the patent of the present application shall be subject to the appended claims.

Claims

A monitoring method for a vehicle sinkhole accident, characterized in that the monitoring method comprises:

Obtain vehicle driving signals;

Judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal;

In the case that the parking posture is abnormal, obtain the driver behavior signal and the vehicle fault signal;

Whether the sinkhole accident occurs to the vehicle is determined according to the parking posture, the driver behavior signal and the vehicle failure signal.
The monitoring method according to claim 1, wherein the acquiring the vehicle driving signal comprises:

The vehicle driving signal of each unit time in the first time period from the first predetermined time before the current time to the current time is acquired.
The monitoring method according to claim 2, wherein the judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal comprises;

The vehicle driving signal is identified according to the pre-stored parking attitude model to judge the parking attitude of the vehicle to determine whether the parking attitude is abnormal.
The monitoring method according to claim 3, wherein the vehicle driving signal includes a real-time three-axis acceleration signal of the vehicle and a speed signal of each wheel of the vehicle, and the vehicle is based on a pre-stored parking attitude model. Identifying the vehicle driving signal to determine the parking posture of the vehicle to determine whether the parking posture is abnormal includes:

Determine whether the vehicle is in a parking state according to the speed signal;

When the vehicle is in the parking state, determining the attitude angle of the vehicle per unit time in the first time period according to the three-axis acceleration signal;

The attitude angle is processed according to the parking attitude model to obtain the parking attitude of the vehicle.
The monitoring method according to claim 1, wherein, in the case that the parking posture is abnormal, acquiring the driver behavior signal and the vehicle fault signal comprises;

In the case that the parking posture is abnormal, obtain the driver behavior signal of each unit time in the second time period from the second predetermined time before the current time to the third predetermined time after the current time;

Acquire a vehicle fault signal per unit time in the second time period.
The monitoring method according to claim 5, wherein the determining whether the vehicle has the sinkhole accident according to the parking posture, the driver's behavior signal and the vehicle fault signal comprises:

The driver's behavior signal is identified according to a pre-stored abnormal driver behavior model to determine the abnormal level of the driver's behavior.
The monitoring method according to claim 6, wherein the driver behavior signal includes a vehicle lock signal, a double flashing light signal, a gear signal, an accelerator pedal signal, a main driving door signal and a steering wheel angle signal, and the Identifying the driver behavior signal according to the pre-stored abnormal driver behavior model to determine the abnormal level of driver behavior includes:

Determine whether there is a first abnormal behavior of unlocking the vehicle within a predetermined period of time after getting off the vehicle according to the main driving door signal and the vehicle locking signal;

Determine whether there is a second abnormal behavior of turning on the double flashing lights according to the gear signal and the double flashing light signal;

judging whether there is a third abnormal behavior of slamming the accelerator pedal according to the accelerator pedal signal;

Determine whether there is a fourth abnormal behavior of repeatedly operating the steering wheel according to the steering wheel angle signal;

According to the main driving door signal and the steering wheel angle signal, determine whether there is a fifth abnormal behavior of the steering wheel not returning to the right position after getting off the car;

The abnormal level of the driver's behavior is judged according to a preset rating rule, the first abnormal behavior, the second abnormal behavior, the third abnormal behavior, the fourth abnormal behavior and the fifth abnormal behavior .
The monitoring method according to claim 6, wherein the determining whether the vehicle has the sinkhole accident according to the parking posture, the driver's behavior signal and the vehicle fault signal comprises:

The parking posture, the abnormal level of the driver's behavior and the vehicle fault signal are identified according to a pre-stored vehicle sinkhole model to determine whether the vehicle has the sinkhole accident.
The monitoring method according to claim 8, wherein the vehicle fault signal comprises a skid signal, a chassis fault signal, a tire pressure monitoring system fault signal and an electrical system fault signal, and the vehicle fault signal is determined according to a pre-stored vehicle sinkhole model. Identifying the parking posture, the abnormal level of the driver's behavior, and the vehicle fault signal to determine whether the vehicle has the sinkhole accident includes:

According to the pre-stored vehicle sinkhole model, the feature vector formed by the parking posture, the abnormal level of the driver's behavior and the vehicle fault signal is processed to obtain the probability of the vehicle having the sinkhole accident;

It is determined that the sinkhole accident occurs to the vehicle when the probability is greater than a predetermined threshold.
The monitoring method according to claim 1, wherein the monitoring method further comprises:

When it is determined that the vehicle has the sinkhole accident, an alarm signal is sent to a service provider of the vehicle so that the service provider can implement rescue according to the alarm signal.
A monitoring device for a vehicle, characterized in that the monitoring device comprises:

an acquisition module, the acquisition module is used to acquire the vehicle driving signal;

a judging module, which is used for judging the parking posture of the vehicle according to the vehicle driving signal to determine whether the parking posture is abnormal;

The obtaining module is further configured to obtain the driver behavior signal and the vehicle fault signal when the parking posture is abnormal;

The judging module is further configured to judge whether the vehicle has the sinkhole accident according to the parking posture, the driver's behavior signal and the vehicle fault signal.
A server, comprising a memory and a processor, wherein the memory stores a computer program, wherein the processor is configured to implement the monitoring of the vehicle sinkhole accident according to any one of claims 1-10 when executing the computer program method.
One or more non-volatile computer-readable storage media storing a computer program that, when executed by one or more processors, implements the vehicle sinkhole accident of any one of claims 1-10 monitoring method.