CN113961258A - Method and system for recognizing abnormal automobile state based on time sequence transfer and storage medium - Google Patents
Method and system for recognizing abnormal automobile state based on time sequence transfer and storage medium Download PDFInfo
- Publication number
- CN113961258A CN113961258A CN202111278209.0A CN202111278209A CN113961258A CN 113961258 A CN113961258 A CN 113961258A CN 202111278209 A CN202111278209 A CN 202111278209A CN 113961258 A CN113961258 A CN 113961258A
- Authority
- CN
- China
- Prior art keywords
- state
- vehicle
- time sequence
- awakening
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/4401—Bootstrapping
- G06F9/4418—Suspend and resume; Hibernate and awake
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Computer Security & Cryptography (AREA)
- Traffic Control Systems (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Emergency Alarm Devices (AREA)
Abstract
The invention requests to protect an automobile state abnormity identification method, an identification system and a storage medium based on time sequence transfer, which belong to the field of automobile electronic systems, utilize vehicle signal data and based on human control logic of vehicles to construct a time sequence state table containing states of awakening normal, awakening abnormal, sleeping, data loss, starting an engine and not sleeping after fortification, extract the vehicle time sequence state, and arrange a related active awakening source, and construct a vehicle time sequence state transfer table related to state transfer based on the vehicle signal table and the time sequence state; and extracting the abnormal state of the vehicle by using the active awakening source according to the time sequence state table, constructing a network abnormal data table, and remotely identifying the abnormal condition of the vehicle, the starting time and the ending time of the fault according to the network abnormal data table to realize remote detection and monitoring of the vehicle abnormality in the parking state.
Description
The invention belongs to the field of automobile electronic systems, and particularly relates to an automobile state abnormity identification technology based on time sequence state transition.
Background
Automobile intellectualization has become an automobile development trend, automobile cloud cooperation is the basis for realizing automobile intellectualization, and more directions can be mined in automobile internet data along with the continuous deepening of the cognition of internet automobiles. The vehicle network abnormal knowledge is an important direction for researching vehicle internet data, and the vehicle network abnormal knowledge can cause the problems of vehicle power loss, unstable data transmission and the like.
Currently, the external Can device is usually relied on to detect the network state of the automobile, and the method has the following defects: the time period for acquiring vehicle Can data is long; can equipment costs additional costs; the batch detection is difficult, so that the network abnormity analysis cannot be carried out in batch. The automobile networking data is generally transmitted to a cloud server through an automobile built-in network data transmission module. The car information includes heartbeat data when the car is asleep and various information when the car is awake. The information when the vehicle wakes up includes: door switch, lock switch, vehicle switch, engine start and stop, and vehicle speed, vehicle network signal strength, and battery voltage. The problems of vehicle power loss, unstable transmission data and the like caused by abnormal vehicle network are solved, and the identification of the abnormal vehicle network is an important research direction in the field of automobile electronic systems.
Publication No.: CN113110967A, chinese patent application entitled "vehicle abnormal dormancy wakeup warning method, apparatus, device and readable storage medium", provides a vehicle abnormal dormancy wakeup warning method. The method comprises the following steps: after a vehicle power supply gear is in an OFF gear, detecting whether a vehicle gateway is in normal dormancy; if the vehicle gateway does not sleep normally, performing abnormal sleep reminding; if the vehicle gateway is in normal dormancy, detecting whether the vehicle gateway is abnormally awakened; and when the vehicle gateway is abnormally awakened, performing abnormal awakening reminding. After the vehicle is powered off, the abnormal detection of vehicle dormancy and awakening is realized based on monitoring the gateway state, and abnormal reminding is carried out when abnormal dormancy or abnormal awakening exists, so that related personnel can find and process the abnormal state of the vehicle in time. Publication No.: CN112835735A, entitled "method, system and vehicle for solving abnormal dormancy and abnormal awakening of vehicle", discloses a method for solving abnormal dormancy and abnormal awakening of vehicle, judging whether the vehicle is in normal dormancy state, if so, the vehicle is in normal dormancy; if not, recording abnormal vehicle dormancy data; after the vehicle is in a normal sleeping state, judging whether the vehicle is in an abnormal awakening state, and if so, recording abnormal awakening data of the vehicle; if not, ending the judgment program; and the vehicle abnormal sleeping data and the vehicle abnormal awakening data are stored in the local DID. According to the method, the fault point can be quickly positioned, the resource investment of the user is reduced, and the satisfaction degree of the user is improved. Publication No.: the chinese patent application CN109581998A entitled "a cloud diagnosis system for dealing with vehicle faults" provides a cloud diagnosis system for dealing with vehicle faults, which comprises a vehicle-mounted terminal, a cloud server and a mobile client, and all the components are communicated through a 4G network; the vehicle-mounted terminal is used for acquiring vehicle information, transmitting acquired data to the cloud server through the 4G network, storing the acquired data into the TF card and attaching a timestamp; the cloud server is used for logging in, logging out, receiving data, sending data, storing the data and sending the data to the mobile client; the mobile client is used for receiving data sent by the cloud server, and the user receives information and operates the information at the mobile client.
The above-mentioned publications do not consider the remote monitoring of "no sleep", "abnormal awakening" and "data loss" of the automobile after the automobile is locked in real time, and for the power shortage problem of the automobile which is not used for a long time, the automobile which has a fault in the field cannot provide remote rescue in time.
Disclosure of Invention
Aiming at the problems that the automobile state during the automobile locking period is not monitored in real time and the remote early warning cannot be provided for a user in real time in the prior art, the state signals of components related to human operation after the automobile is locked are extracted, the automobile signal table with the timestamp is used as input, the time sequence state table is constructed through the time sequence state transfer condition, then the awakening abnormity, the non-sleep after the defense, the data loss automobile network abnormity is extracted from the time sequence state table, and the detection and monitoring of the automobile abnormal state are realized.
According to the technical scheme for solving the technical problems, under the condition that hardware is not added, a time sequence state table containing vehicle states of awakening normal, awakening abnormal, sleeping, data loss, starting an engine and not sleeping after fortification is constructed by using data uploaded to a cloud end by a vehicle end and relying on normal control logic of a vehicle, and abnormal information of awakening abnormal, not sleeping after fortification and data loss is extracted to realize detection and monitoring of abnormal states of the vehicle. Specifically, a vehicle state anomaly detection and identification method based on time sequence state transition is provided, which comprises the following steps: analyzing the acquired original vehicle data in a data preprocessing stage and generating a vehicle information table; extracting and defining a vehicle time sequence state, listing a related active awakening source, and constructing a vehicle time sequence state transfer table related to state transfer based on a vehicle signal table and the time sequence state; and traversing the time sequence state table, extracting the abnormal state of the vehicle from the time sequence state table, constructing a network abnormal data table, and remotely identifying the abnormal condition of the vehicle according to the network abnormal data table.
Further preferably, the header of the vehicle information table includes an event occurrence status field and a continuous signal field, wherein the status field includes: timestamp, left front door, right front door, left back door, right back door, left front lock, right front lock, left back lock, right back lock, engine start-stop, power supply gear, continuous signal field includes: vehicle speed, battery voltage.
Further preferably, the current time sequence state of the vehicle is determined according to the power consumption condition of the storage battery in the automobile dormant state. The method comprises the following steps: when the storage battery is detected to be in a high power consumption state as an awakening state, the storage battery is detected to be in a low power consumption state as a sleeping state, when the power consumption abnormity of the storage battery is increased after fortification is abnormal, active awakening is not performed within a specified time, the power consumption abnormity of the storage battery is increased as an awakening abnormity, data frames are detected, and the time difference between the two frames of data before and after the data frames is greater than a specified value, so that the data loss state is realized.
Further preferably, the vehicle signal data is acquired from the vehicle signal table in the order of the time stamps from small to large, the state transition condition judgment is performed based on the current vehicle time sequence state and the current vehicle signal, if the state transition condition is satisfied, the state transition is performed according to the time sequence state transition condition table, the current state in the time sequence state table is updated, and otherwise, the next frame data in the vehicle signal table is continuously acquired. And outputting a time sequence state table until the vehicle signal table is traversed, and finishing the state transition.
Preferably, the vehicle information table is called to obtain first frame data, the time sequence state transition condition table is inquired and traversed, if the transition condition is met, state transition is carried out according to the time sequence state transition condition table, the current state after transition is added to the last row of the time sequence state transition table, the time sequence state table is updated, otherwise, the vehicle information table is traversed continuously to obtain next frame data as current data, the state represented by the last row of data in the time sequence state table is obtained as the current state, whether the transition condition is met is judged until the vehicle information table is traversed, and the time sequence state table is output.
The invention also provides a vehicle state abnormity detection and identification system based on time sequence state transition, which comprises the following components: the system comprises a data preprocessing unit, a vehicle information table, a time sequence state acquisition unit, a vehicle state transition condition table and a time sequence logic control unit, wherein the data preprocessing unit analyzes acquired original data and generates the vehicle information table; the time sequence state acquisition unit acquires and defines the time sequence states of normal, dormant, non-dormant after fortification and data loss of the vehicle, lists related active awakening sources, and constructs a vehicle time sequence state transfer table related to state transfer based on a vehicle signal table and the time sequence state; the time sequence logic control unit traverses the time sequence state table, extracts states of awakening abnormity, data loss and non-sleep after fortification from the time sequence state table, constructs a network abnormity data table, and the cloud identification system remotely identifies the vehicle abnormity condition according to the network abnormity data table.
The present invention further provides a computer-readable storage medium, which includes at least one instruction, at least one program, and a set of code codes stored in the computer-readable storage medium, so that the computer performs the above-mentioned identification method of the present invention
The method utilizes the active awakening source to extract the abnormal state of the vehicle based on the vehicle state information, reduces the dependence on diagnostic equipment, conforms to the intelligent automobile development concept, and can diagnose the network abnormality of the vehicle at a more abstract level. Meanwhile, a large amount of vehicle state information data of the cloud can be combined, so that not only can the high calculation power of the cloud be utilized to carry out abnormity diagnosis on each vehicle, the problem of vehicle faults can be solved for clients in a remote mode in a targeted mode, but also the abnormal state of a vehicle network can be subjected to batch statistical analysis, and later-stage design is improved directionally. And constructing a time sequence state table containing vehicle states of normal awakening, abnormal awakening, data loss, engine starting and non-sleeping after fortification, extracting the abnormality containing the information of abnormal awakening, non-sleeping after fortification and data loss through the time sequence state table, and realizing detection and monitoring of the abnormal vehicle states.
According to the invention, the abnormal state of the vehicle is deduced based on the vehicle information with the timestamp and the man-vehicle operation relationship, so that the dependence on hardware is reduced, and the abnormality of the vehicle can be deduced in a more abstract level. The network state of each vehicle can be monitored by using cloud data, and if abnormity occurs, the abnormal information is immediately pushed to the terminal, so that a user is quickly reminded.
Drawings
Fig. 1 is a schematic diagram of a vehicle signal table, the table head of which comprises: the system comprises a timestamp, a sleep state signal, a left front door switch signal, a right front door switch signal, a left rear door switch signal, a right rear door switch signal, a left front lock switch signal, a right front lock switch signal, a left rear lock switch signal, a right rear lock switch signal, a fortification state signal, an engine start-stop state signal and a power supply gear position signal. The state information of the left ellipsis part in the schematic table is all 0;
fig. 2 is a time sequence state representation intention, and the header comprises: vehicle state, jump time, whether abnormal or not;
fig. 3 is a vehicle network abnormality representation intention, and the header includes: fault type, starting time, ending time and single fault time (min);
fig. 4 is a state transition flowchart, in which a current vehicle timing state (a state corresponding to a last line in a current timing state table) is combined with a current vehicle signal, and a state transition condition is determined, if the state transition condition is satisfied, the timing state transition is performed, the current state is updated, and the current state is added in the timing state table, otherwise, the timing state table is not changed, the vehicle signal table is continuously traversed to obtain next frame data until the vehicle signal table is traversed, the timing state table is output, and the timing state transition process is ended.
Detailed Description
In order to facilitate understanding of the invention, embodiments of the invention are described in further detail below with reference to the accompanying drawings and specific examples.
The invention determines the active wake-up source and the timing state based mainly on generating a vehicle signal table related to human-to-vehicle operation. And defining the time sequence states of normal awakening, abnormal awakening, dormancy, data loss, no-break dormancy after fortification and the like. And constructing a time sequence state table. And constructing a time sequence state table according to the definition of the vehicle signal table and the time sequence state and the state transition relation. And constructing a vehicle network abnormity table. And extracting abnormal states such as abnormal awakening, data loss and no sleep after fortification from the time sequence state table, and constructing a vehicle network abnormal table.
A vehicle state abnormity detection and identification method based on time sequence state transition comprises the following steps: analyzing the acquired original vehicle data in a data preprocessing stage and generating a vehicle information table; extracting and defining a vehicle time sequence state, listing a related active awakening source, and constructing a vehicle time sequence state transfer table related to state transfer based on a vehicle signal table and the time sequence state; and traversing the time sequence state table, extracting the abnormal state of the vehicle from the time sequence state table, constructing a network abnormal data table, and remotely identifying the abnormal condition of the vehicle according to the network abnormal data table. And acquiring vehicle signal data from the vehicle signal table in the sequence of the timestamps from small to large, judging a state transition condition based on the current vehicle time sequence state and the current vehicle signal, if the state transition condition is satisfied, performing state transition according to the time sequence state transition condition table, updating the current state in the time sequence state table, and otherwise, continuously acquiring next frame data of the vehicle signal table. And outputting a time sequence state table until the vehicle signal table is traversed, and finishing the state transition.
First, a vehicle signal table related to human-to-vehicle operation is generated, and as shown in fig. 1, a vehicle signal table is constructed, that is, the vehicle signal table includes time-stamped vehicle state information, and the header thereof includes: the system comprises a timestamp, a sleep state signal, a left front door switch signal, a right front door switch signal, a left rear door switch signal, a right rear door switch signal, a left front lock switch signal, a right front lock switch signal, a left rear lock switch signal, a right rear lock switch signal, a fortification state signal, an engine start-stop state signal and a power supply gear position signal.
Messages reflecting the states of all parts of the vehicle can be periodically generated and cached in a vehicle memory, and the message information can be utilized by a vehicle controller and also can be transmitted to a cloud end through a network transmission module for data analysis. The method includes extracting messages related to human-to-vehicle operations from vehicle messages collected from a vehicle memory or a cloud to generate a vehicle signal table, where the signals generally include, but are not limited to: the system comprises a sleep state signal, a left front door switch signal, a right front door switch signal, a left rear door switch signal, a right rear door switch signal, a left front lock switch signal, a right front lock switch signal, a left rear lock switch signal, a right rear lock switch signal, a defense state signal, an engine start-stop state signal and a power supply gear position signal, wherein each signal is provided with a timestamp of the occurrence moment of the signal. Thus, the header of the vehicle signal table contains the above-named signal and a timestamp of its time of occurrence. And finally, due to the disorder of the messages in the cache, the analyzed vehicle signal table needs to be sequenced according to the time stamps.
As shown in fig. 2, the time sequence state table records the time sequence state of the vehicle in time sequence, and the table head comprises: the method comprises the following steps of vehicle state, jumping time and whether the vehicle state is abnormal, wherein the vehicle state content comprises: waking up normally, waking up abnormally, sleeping, losing data, and not sleeping after fortification.
Determining an active wake-up source and timing states, comprising:
I. active wake-up source: i.e. actions related to human operations such as door opening, lock closing, trunk opening, lock opening, disarming, engine starting, etc.
Wake-up state: activating an automobile network, wherein a sleep signal is 0;
sleep state: the automobile network sleeps, and the sleep signal is 1;
and IV, not sleeping after fortification: after the vehicle receives the fortification command, the vehicle does not sleep for a period of time under the condition of no active awakening source, and is in an abnormal state;
v. wake exception: the vehicle is in a sub-state of an awakening state, and no active awakening source exists in a small time range at the awakening moment, which indicates that the vehicle is not awakened by a person and is in an abnormal state;
wake up normal: the vehicle is in a sub-state of an awakening state, and an active awakening source is arranged in a short time range at the awakening moment, which indicates that the vehicle is awakened by a person and is in a normal state;
vii, data loss: the time difference between the two frames of data is greater than the specified value, so that the periodic vehicle signal data is considered to be lost, and the state is an abnormal transition state.
As shown in fig. 3, the vehicle network abnormality table, i.e., the abnormality state extracted from the time-series state table, includes at its head: fault type, start time, end time, single fault time (min), wherein the fault type includes: abnormal arousal, data loss and no sleep after fortification.
Obtaining a vehicle network abnormity table based on the abnormity state in the time sequence state table: and traversing the time sequence state table to extract the abnormal state, and writing the fault type, the starting time, the ending time and the single fault time into the vehicle network abnormal table. Wherein the fault types include: the method comprises the steps of awakening abnormity, losing data and not sleeping after fortification, wherein the fault starting time is the jump time corresponding to the fault state in the time sequence state table, the fault ending time is the jump time corresponding to the next state of the fault state, and the single fault time is the time difference between the fault ending time and the fault starting time.
And constructing a time sequence state table as shown in fig. 4, acquiring vehicle signal data from the vehicle signal table in the order of the smaller timestamp to the larger timestamp, judging a state transition condition based on the current vehicle time sequence state and the current vehicle signal, if the state transition condition is satisfied, performing state transition according to the time sequence state transition condition, updating the current state in the time sequence state table, and if not, continuously traversing the vehicle signal table to acquire the next frame of data. And outputting a time sequence state table until the vehicle signal table is traversed, and finishing the state transition.
TABLE 1 State transition conditions
As shown in table 1, a time sequence state transition condition is constructed according to the real-time monitoring of the time sequence state table: when the sleep signal is 0 in the sleep transition state or the data loss transition state. Triggering state transition and transferring to an awakening state; when a sleep signal is detected to be 1 in the wake-up transition state or the data loss transition state, triggering state transition and transitioning to the sleep state; in the awakening state, no active awakening source occurs within a specified time to trigger state transition to be abnormally awakened; detecting that an active awakening source jumps to normal awakening within a specified time in the awakening state; awakening the normal state, detecting that a fortification signal appears, and triggering state transition when a sleep signal is kept to be 0 in preset time, and transferring to fortification and then not sleeping; an active awakening source triggering state transfer instruction appears in a non-sleep state after fortification, and the state is transferred to an awakening normal state; and the timestamp difference corresponding to the next frame data and the current frame data is greater than the specified value to trigger a state transition instruction and transition to a data loss state. Wherein, the active wake-up source can be a human-to-vehicle operation, including: door opening, door opening and closing, lock opening, lock closing, trunk opening, lock opening, fortification removing, engine starting and the like. Particularly, in the time range before and after the current timestamp, the vehicle signal jump caused by the operation of people and vehicles can be found in the vehicle signal table, namely, the active awakening source appears under the current timestamp, otherwise, no active awakening source appears.
Further, based on the abnormal state in the time sequence state table, the fault type, the fault starting time, the fault ending time and the single fault time (min) are extracted and written into the vehicle network abnormal table. Wherein the fault types include: the method comprises the steps of awakening abnormity, losing data and not sleeping after fortification, wherein the fault starting time is the jump time corresponding to the fault state in a state table, the fault ending time is the jump time corresponding to the next state of the fault state, and the single fault time is the time difference between the fault ending time and the fault starting time.
A time sequence logic table is constructed based on a vehicle state table and a time sequence state, the process relates to the transition of each state, the state transition conditions refer to table 1, and the specific steps are as follows:
and I, starting. Acquiring first frame data from a vehicle information table, and taking 'start' as an initialization state to enter a state transition II;
and II, performing state transition. As shown in fig. 4, the state transition condition corresponding to the sequence number in table 1 is looked up based on the state transition condition sequence number corresponding to the current time sequence state, if the state transition condition is satisfied, the state transition is performed, the update of the current state is completed, and the name of the current state, the timestamp of the current state jump, whether the current state is an abnormal state or not are added to the end of the time sequence state table, otherwise, the original state is maintained to enter into iii;
if the data of the vehicle information table is not traversed, acquiring next frame data in the vehicle information table as current data, acquiring the state represented by the last row of data in the time sequence state table as the current state, entering a second state, and entering a fourth state;
and IV, ending. And outputting the time sequence state table.
Illustrated in the example figures herein:
1. initializing a state table: "vehicle state: starting and jumping time: NULL, whether abnormal: 0 ";
2. and acquiring the 1 st frame data from the vehicle information table, and entering state flow conversion. The first frame data sleep signal is 1, and the state transition condition 2 corresponding to the initial transition state is satisfied, so the current state is updated to the sleep state, and the following is added in the time sequence transition table: "vehicle state: sleeping and jumping time: 1602076928, whether abnormal: 0 ";
3. and acquiring 2 nd frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is a sleep state, and the corresponding state transition conditions 1 and 3 are not satisfied, so the sleep state is maintained;
4. and acquiring the 3 rd frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is a sleep state, the sleep signal is 0, the corresponding state transition bar 1 is established, the current time sequence state jumps to an awakening state, an active awakening source with door opening action appears near the corresponding time (60 s before and after taking in an example), the current time sequence state is updated to an awakening normal state, and the current time sequence state is added in a time sequence transition state table: "vehicle state: waking up normal and jumping time: 1602076968, whether abnormal: 0 ";
5. and acquiring the 4 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is an awakening normal state, and the corresponding state transition conditions 2, 3 and 6 are not established, so that the awakening normal state is maintained;
6. and acquiring 5 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is an awakening normal state, and the corresponding state transition conditions 2, 3 and 6 are not established, so that the awakening normal state is maintained;
7. and acquiring the 6 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is a wake-up normal state, and the corresponding state transition condition 2 is satisfied, so that the current time sequence state jumps to a sleep state, and the following is added in a time sequence transition state table: "vehicle state: sleeping and jumping time: 1602077028, whether abnormal: 0 ";
8. and acquiring 7 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is a sleep state, the time difference with the next frame data is 10000s, and the corresponding state transition condition 3 is satisfied, so that the data loss state is jumped to, and the following is added in the time sequence transition table: "vehicle state: data loss and jump time: 1602077048, whether abnormal: 1.
9. and acquiring 8 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is a data loss state, the sleep signal is 1, and the corresponding state transition condition 2 is satisfied, so the state jumps to the sleep state, and the following is added in the time sequence transition state table: "vehicle state: sleeping and jumping time: 1602087048, whether abnormal: 0.
10. and acquiring 9 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is a sleep state, and the corresponding state transition conditions 1 and 3 are not satisfied, so the sleep state is maintained;
11. and acquiring 10 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is a sleep state, a sleep signal is changed into 0, a corresponding state transition condition 1 is satisfied, the state is transitioned to an awakening state, no active awakening source is arranged near the corresponding time, the state is updated to an awakening abnormal state, and the following is added in a time sequence transition state table: "vehicle state: awakening exception and jumping time: 1602087088, whether abnormal: 1 ";
12. and acquiring 11 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is an awakening abnormal state, and the corresponding state transition conditions 2, 3 and 6 are not satisfied, so that the awakening abnormal state is maintained;
13. and acquiring 12 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is an awakening abnormal state, and the corresponding state transition conditions 2, 3 and 6 are not satisfied, so that the awakening abnormal state is maintained;
14. and acquiring 13 th frame data from the vehicle signal table, and entering state flow conversion. The current time sequence state is an awakening abnormal state, a sleep signal is 1, a corresponding state transition condition 2 is met, the current time sequence state is transitioned to the sleep state, and the following is added in a time sequence transition state table: "vehicle state: sleeping and jumping time: 1602087148, whether abnormal: 0 ";
15. and finishing traversing of the vehicle signal table, and outputting a time sequence state table.
Claims (10)
1. A vehicle state abnormity detection and identification method based on time sequence state transition is characterized by comprising the following steps: in the data preprocessing stage, a man-vehicle operation related message is extracted from a vehicle memory or a cloud-side acquired vehicle message to generate a vehicle signal table; extracting a vehicle time sequence state, listing a related active awakening source, and constructing a vehicle time sequence state transfer table related to state transfer based on a vehicle signal table and the time sequence state; and traversing the time sequence state table, extracting the abnormal state of the vehicle by using the active awakening source according to the time sequence state table, constructing a network abnormal data table, and remotely identifying the abnormal condition of the vehicle, the starting time and the ending time of the fault according to the network abnormal data table.
2. The method of claim 1, wherein the vehicle signal jump caused by human-vehicle operation is consistent with the vehicle signal table in the time range before and after the current timestamp, and the active wake-up source is determined to be present under the current timestamp.
3. The method according to claim 1 or 2, characterized in that when the sleep signal is 0 in the sleep transition state or the data loss transition state, the state transition is triggered and transferred to the wake-up state; when a sleep signal is 1 in an awakening transition state or a data loss transition state, triggering state transition and transitioning to a sleep state; detecting that no active awakening source exists within preset time in an awakening state, and jumping to an awakening abnormity; detecting that an active awakening source jumps to normal awakening within preset time in the awakening state; awakening a fortification signal detected in a normal transition state, triggering state transition when a sleep signal is kept to be 0 within preset time and transferring to fortification and then not sleeping; detecting an active awakening source in a non-sleep state after fortification, triggering a state transfer instruction and transferring to an awakening normal state; and triggering a state transition instruction and transitioning to a data loss state when the timestamp difference corresponding to the next frame data and the current frame data is greater than a preset value.
4. The method according to claim 1 or 2, wherein the fault starting time is a jump time corresponding to a fault state in the time sequence state table, the fault ending time is a jump time corresponding to a next state of the fault state, and the single fault time is a time difference between the fault ending time and the fault starting time.
5. The method according to claim 1 or 2, characterized in that the vehicle signal data are obtained from the vehicle signal table in the order of the time stamps from small to large, the state transition condition judgment is carried out based on the current vehicle time sequence state and the current vehicle signal, if the state transition condition is satisfied, the state transition is carried out according to the time sequence state transition condition table, and the current state in the time sequence state table is updated, otherwise, the next frame data of the vehicle signal table is continuously obtained until the vehicle signal table is traversed, the time sequence state table is output, and the state transition is finished.
6. The method of claim 5, wherein the vehicle information table is invoked to obtain first frame data, the time sequence state transition condition table is queried and traversed, if the transition condition is satisfied, state transition is performed according to the time sequence state transition condition table, the current state after transition is added to the last row of the time sequence state transition table, the time sequence state table is updated, otherwise, the vehicle information table is continuously traversed to obtain next frame data as current data, the state represented by the last row of data in the time sequence state table is obtained as the current state, whether the transition condition is satisfied is judged, until the vehicle information table is traversed, and the time sequence state table is output.
7. A vehicle state abnormality detection and identification system based on time-series state transition is characterized by comprising: the system comprises a data preprocessing unit, a vehicle information table, a time sequence state acquisition unit, a vehicle state transition condition table and a time sequence logic control unit, wherein the data preprocessing unit analyzes acquired messages related to human-vehicle operation to generate a vehicle signal table; the time sequence state acquisition unit acquires the time sequence states of normal, dormant, non-dormant after fortification and data loss of the vehicle, lists related active awakening sources, and constructs a vehicle time sequence state transfer table related to state transfer based on a vehicle signal table and the time sequence state; the time sequence logic control unit traverses the time sequence state table, an active awakening source is used for extracting awakening abnormity, data loss and non-sleep state after fortification in the time sequence state table, a network abnormity data table is constructed, and the cloud end identification system remotely identifies vehicle abnormity conditions, fault starting time and fault ending time according to the network abnormity data table.
8. The system of claim 7, wherein when the sleep signal is 0 in the sleep transition state or the data loss transition state, the state transition is triggered and transits to the wake-up state; when a sleep signal is 1 in an awakening transition state or a data loss transition state, triggering state transition and transitioning to a sleep state; detecting that no active awakening source exists within preset time in an awakening state, and jumping to an awakening abnormity; detecting that an active awakening source jumps to normal awakening within preset time in the awakening state; awakening a fortification signal detected in a normal transition state, triggering state transition when a sleep signal is kept to be 0 within preset time and transferring to fortification and then not sleeping; detecting an active awakening source in a non-sleep state after fortification, triggering a state transfer instruction and transferring to an awakening normal state; and triggering a state transition instruction and transitioning to a data loss state when the timestamp difference corresponding to the next frame data and the current frame data is greater than a preset value.
9. The system of claim 7 or 8, wherein constructing a vehicle timing state transition table associated with a state transition comprises: and acquiring vehicle signal data from the vehicle signal table in the sequence of the timestamps from small to large, judging a state transition condition based on the current vehicle time sequence state and the current vehicle signal, if the state transition condition is satisfied, performing state transition according to the time sequence state transition condition table, updating the current state in the time sequence state table, and otherwise, continuously acquiring next frame data of the vehicle signal table until the vehicle signal table is traversed, outputting the time sequence state table, and finishing the state transition.
10. A computer-readable storage medium storing at least one instruction, at least one program, and a set of code codes instructions for causing a computer to perform the vehicle state abnormality detection recognition method according to claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111278209.0A CN113961258B (en) | 2021-10-30 | 2021-10-30 | Automobile state anomaly identification method, system and storage medium based on time sequence transfer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111278209.0A CN113961258B (en) | 2021-10-30 | 2021-10-30 | Automobile state anomaly identification method, system and storage medium based on time sequence transfer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113961258A true CN113961258A (en) | 2022-01-21 |
| CN113961258B CN113961258B (en) | 2023-06-30 |
Family
ID=79468483
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111278209.0A Active CN113961258B (en) | 2021-10-30 | 2021-10-30 | Automobile state anomaly identification method, system and storage medium based on time sequence transfer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113961258B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114546509A (en) * | 2022-02-28 | 2022-05-27 | 石家庄通合电子科技股份有限公司 | Method, device, terminal and storage medium for realizing self-awakening function |
| CN114564242A (en) * | 2022-02-28 | 2022-05-31 | 石家庄通合电子科技股份有限公司 | Method, device, terminal and storage medium for realizing self-awakening function |
| CN115277746A (en) * | 2022-06-24 | 2022-11-01 | 重庆长安汽车股份有限公司 | Vehicle battery abnormal state real-time monitoring method and system and readable storage medium |
| WO2024109320A1 (en) * | 2022-11-23 | 2024-05-30 | 深蓝汽车科技有限公司 | Method and apparatus for monitoring abnormal wake-up of vehicle, and server and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103197162A (en) * | 2012-01-04 | 2013-07-10 | 中国人民解放军第二炮兵工程学院 | Multiple switch parameter fault diagnosis method based on template matching |
| CN105808923A (en) * | 2016-02-29 | 2016-07-27 | 北京航空航天大学 | Anomaly detection method and device of data sequence |
| CN110503743A (en) * | 2019-08-12 | 2019-11-26 | 浙江吉利汽车研究院有限公司 | A kind of method for detecting abnormality of automatic driving vehicle, device and equipment |
| CN110515781A (en) * | 2019-07-03 | 2019-11-29 | 北京交通大学 | A kind of complication system status monitoring and method for diagnosing faults |
| CN110658807A (en) * | 2019-10-16 | 2020-01-07 | 上海仁童电子科技有限公司 | Vehicle fault diagnosis method, device and system |
| CN111683855A (en) * | 2019-04-15 | 2020-09-18 | 深圳市大疆创新科技有限公司 | Alarm control method and equipment for automatic driving vehicle and automatic driving vehicle |
| CN112835735A (en) * | 2021-01-04 | 2021-05-25 | 宝能(西安)汽车研究院有限公司 | Method and system for solving abnormal sleeping and abnormal awakening of vehicle and vehicle |
| CN113015135A (en) * | 2021-03-29 | 2021-06-22 | 重庆长安汽车股份有限公司 | Dormancy awakening method for remote monitoring vehicle-end controller based on network management |
| CN113110967A (en) * | 2021-03-23 | 2021-07-13 | 东风汽车集团股份有限公司 | Early warning method, device and equipment for abnormal dormancy awakening of vehicle and readable storage medium |
-
2021
- 2021-10-30 CN CN202111278209.0A patent/CN113961258B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103197162A (en) * | 2012-01-04 | 2013-07-10 | 中国人民解放军第二炮兵工程学院 | Multiple switch parameter fault diagnosis method based on template matching |
| CN105808923A (en) * | 2016-02-29 | 2016-07-27 | 北京航空航天大学 | Anomaly detection method and device of data sequence |
| CN111683855A (en) * | 2019-04-15 | 2020-09-18 | 深圳市大疆创新科技有限公司 | Alarm control method and equipment for automatic driving vehicle and automatic driving vehicle |
| CN110515781A (en) * | 2019-07-03 | 2019-11-29 | 北京交通大学 | A kind of complication system status monitoring and method for diagnosing faults |
| CN110503743A (en) * | 2019-08-12 | 2019-11-26 | 浙江吉利汽车研究院有限公司 | A kind of method for detecting abnormality of automatic driving vehicle, device and equipment |
| CN110658807A (en) * | 2019-10-16 | 2020-01-07 | 上海仁童电子科技有限公司 | Vehicle fault diagnosis method, device and system |
| CN112835735A (en) * | 2021-01-04 | 2021-05-25 | 宝能(西安)汽车研究院有限公司 | Method and system for solving abnormal sleeping and abnormal awakening of vehicle and vehicle |
| CN113110967A (en) * | 2021-03-23 | 2021-07-13 | 东风汽车集团股份有限公司 | Early warning method, device and equipment for abnormal dormancy awakening of vehicle and readable storage medium |
| CN113015135A (en) * | 2021-03-29 | 2021-06-22 | 重庆长安汽车股份有限公司 | Dormancy awakening method for remote monitoring vehicle-end controller based on network management |
Non-Patent Citations (3)
| Title |
|---|
| 孙志华: ""基于OSGi的汽车远程诊断平台的开发"", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, pages 035 - 928 * |
| 陈鹏远: ""智能网联汽车CAN总线异常检测技术研究"", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, pages 035 - 207 * |
| 马云林等: ""某车型主机初次上电的策略研究"", 《重庆汽车工程学会2019年论文汇编 》, pages 62 - 65 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114546509A (en) * | 2022-02-28 | 2022-05-27 | 石家庄通合电子科技股份有限公司 | Method, device, terminal and storage medium for realizing self-awakening function |
| CN114564242A (en) * | 2022-02-28 | 2022-05-31 | 石家庄通合电子科技股份有限公司 | Method, device, terminal and storage medium for realizing self-awakening function |
| CN114546509B (en) * | 2022-02-28 | 2023-10-10 | 石家庄通合电子科技股份有限公司 | Method, device, terminal and storage medium for realizing self-wake-up function |
| CN115277746A (en) * | 2022-06-24 | 2022-11-01 | 重庆长安汽车股份有限公司 | Vehicle battery abnormal state real-time monitoring method and system and readable storage medium |
| WO2024109320A1 (en) * | 2022-11-23 | 2024-05-30 | 深蓝汽车科技有限公司 | Method and apparatus for monitoring abnormal wake-up of vehicle, and server and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113961258B (en) | 2023-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113961258B (en) | Automobile state anomaly identification method, system and storage medium based on time sequence transfer | |
| CN112380045B (en) | Vehicle abnormality detection method, device, equipment and storage medium | |
| CN106371424B (en) | A kind of vehicle dormancy method and system | |
| US10212663B2 (en) | Power management system and method for vehicle locating unit | |
| CN110758293B (en) | Power management method and device for vehicle-mounted intelligent equipment and readable storage medium | |
| CN111619473A (en) | Automobile static power supply management system and management method | |
| CN207937961U (en) | The dormancy awakening system of remote information processor | |
| CN112306044A (en) | Method and system for monitoring non-dormancy abnormity of bus network of whole vehicle | |
| CN102749892A (en) | Network management method for electronic control system in parking braking of automobile | |
| WO2022142938A1 (en) | Method, apparatus and device for detecting vehicle start-up, and computer storage medium | |
| CN113923137A (en) | Whole vehicle bus network abnormity monitoring method and system | |
| CN113763592B (en) | Vehicle feed detection method, system, readable storage medium and computer device | |
| CN114844769A (en) | Vehicle feed risk monitoring method | |
| CN111929600B (en) | Storage battery diagnosis monitoring method, monitoring system, vehicle and vehicle networking cloud platform | |
| CN114368353A (en) | System for waking up electronic plug-in device (EPID) and related method thereof | |
| CN113377086A (en) | Vehicle power shortage early warning method | |
| CN112389352A (en) | Finished automobile static current management system and method | |
| CN115277746A (en) | Vehicle battery abnormal state real-time monitoring method and system and readable storage medium | |
| CN106325985A (en) | Timer adjusting method and system and mobile terminal | |
| CN107357178B (en) | Control system and control method for vehicle | |
| CN115932587A (en) | Sentinel mode battery state monitoring method and device based on power domain | |
| CN115000533A (en) | Low-voltage energy storage battery management system and control method | |
| CN111555947A (en) | MCU low-power consumption dormancy awakening system | |
| CN113923695A (en) | Awakening fault detection method and device and message sending method and device | |
| CN115016429B (en) | Automatic reset method and system for whole vehicle controller network not to sleep |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |