CN110034968A - Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations - Google Patents
Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations Download PDFInfo
- Publication number
- CN110034968A CN110034968A CN201910183238.5A CN201910183238A CN110034968A CN 110034968 A CN110034968 A CN 110034968A CN 201910183238 A CN201910183238 A CN 201910183238A CN 110034968 A CN110034968 A CN 110034968A
- Authority
- CN
- China
- Prior art keywords
- correlation
- sensor
- abnormality detection
- vehicle
- edge calculations
- 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.)
- Pending
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
- B60Q9/00—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
- B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
- B60R16/023—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements for transmission of signals between vehicle parts or subsystems
- B60R16/0231—Circuits relating to the driving or the functioning of the vehicle
- B60R16/0232—Circuits relating to the driving or the functioning of the vehicle for measuring vehicle parameters and indicating critical, abnormal or dangerous conditions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/40—Bus networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/06—Generation of reports
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0811—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking connectivity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/40—Bus networks
- H04L2012/40208—Bus networks characterized by the use of a particular bus standard
- H04L2012/40215—Controller Area Network CAN
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Environmental & Geological Engineering (AREA)
- Human Computer Interaction (AREA)
- Traffic Control Systems (AREA)
Abstract
Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations.The invention proposes a kind of automobile method for detecting abnormality based on edge calculations, the exception of automobile is detected using the correlation between different sensors.This method forms sensor variable pair by analyzing the correlation between interior different sensors, by relevant sensor, and passes through ring structure for these sensor variables to combining, thus one sensor correlation ring of composition.Sensor correlation ring is detected again, judges vehicle condition from the result of detection.And entire vehicle abnormality detection system is then embedded in an edge device, because final detection effect is only fed back to Cloud Server by edge device, so that vehicle abnormality detection system is more efficient, and can protect vehicle user privacy.
Description
Technical field
The present invention relates to intelligent automobile, especially a kind of multisensor based on edge calculations for intelligent automobile melts
Close vehicle safety method for detecting abnormality.
Background technique
Hyundai Motor develops towards more and more intelligentized direction, is no longer only responsible for the mechanical equipment advanced, and becomes
Using modern multiple technologies, gathers multi -function in integral whole, meet the integrated system of a variety of demands of different user.And these skills
The realization of art and the offer of function are then complicated computer systems on Hyundai Motor.And these computer systems are then
Need by hundreds of sensors on Hyundai Motor and independent electronic control unit (Electronic Control Unit,
ECU) a series of task is completed.Hyundai Motor acquires the different data in environment by the sensor of these types multiplicity,
It transfers to ECU to be analyzed and handled again and is broadcasted in CAN bus, Hyundai Motor executes phase further according to corresponding message
The movement answered.And these electronic control units connect each other, can have corresponding because other sensors experience variation
Reaction.In existing intelligent automobile platform, there are various system vulnerabilities, attacker can be in several ways to vehicle
It is attacked, the attack of even long-range physical contact, not only by the sensor of attack part but also can attack CAN
Bus, to achieve the purpose that control automobile, automotive safety is seriously threatened attacker.
Popularizing for intelligent automobile, brings great convenience to people's lives, and people are then given in the appearance of intelligent automobile
Trip bring revolutionary change.However, have in recent years many different model automobiles repeatedly by the event of hacker attack not
In the media, this also results in people for the worry of intelligent automobile safety and fear for disconnected exposure.So automotive safety at present according to
Old is one of significant challenge of pilotless automobile faced, and how to provide a safer driving environment is intelligent vapour
The inevitable problem of vehicle.There are many researchs to focus on automotive safety at present, a variety of solutions is proposed to automotive safety
Certainly scheme.And the first line of defence that abnormality detection is defendd as automobile, with greater need for obtaining enough attention.
Summary of the invention
The present invention be in order to extremely occur when find and handle exception in time, thus reduce abnormal bring risk and
Harm, provides a kind of automobile abnormality detection safety method based on edge calculations, for automotive safety loophole, by sensing
Device correlation is analyzed, to quickly judge whether vehicle is abnormal, it is different to guarantee to calculate equipment by built-in edge
Normal quick response.To reach when vehicle is abnormal or is under attack, which can timely and effectively be detected
Automobile exception and issue the user with alarm out.
Technical solution of the invention is as follows:
A kind of Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations, which is characterized in that including such as
Lower step:
Data acquisition: it is read by board Controller Area Network (Control Area Network, CAN) bus protocol
Equipment reads data by onboard diagnostic system (On-Board Diagnostic, OBD) interface;
Correlation analysis: using Pearson correlation coefficient as the measurement index of correlation, measure each sensing data it
Between correlation, and select the high sensor pair of correlation, construct the model of abnormality detection;
Abnormality detection: the correlation by calculating respective sensor pair is used as the Judging index of abnormality detection
Testing result is submitted: passing through the different submissions for submitting strategy to carry out result under normal and abnormal two states.
The present invention utilizes edge calculations technology, and entire invention system is embedded in the edge device of network.Edge device
Using its specific position in a network, to reach the quick response for automobile abnormality detection, save bandwidth resources, protect
Privacy purpose is protected, meets vehicle and detects the needs of abnormal real-time, quickly.
Data acquisition module of the invention reads in real time and decodes multiple sensing datas by OBD interface, and analysis is different
The correlation of data between sensor pair, then by comparing calculating to correlation under normal circumstances with respective sensor.
If being judged to normally, periodically send and report to cloud server system, to verify Cloud Server in threshold range
Normal connection between edge device;If exceeding corresponding threshold value, it is judged to exception, and different in real time to driver's progress
Often alarm, to achieve the purpose that fast and effective abnormality detection.
The invention has the advantages that opposite known technology, can in the case where not influencing the communication resource of CAN bus,
Using Multi-sensor Fusion data, the exception that vehicle generates is detected real-time, quickly and is sounded an alarm.
Detailed description of the invention
The present invention is based on the flow charts of the Multi-sensor Fusion vehicle safety method for detecting abnormality of edge calculations by Fig. 1
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples.
As seen from Figure 1, the detection method is mainly by four steps, respectively data acquisition, correlation analysis, abnormal inspection
It surveys and testing result is submitted.Their effect difference is as follows:
Data acquisition: vehicle abnormality detection system using proprietary CAN bus agreement read equipment by OBD interface with
CAN is connected, and thus monitors and cache the message broadcasted on CAN for subsequent analysis.
Correlation analysis: vehicle abnormality detection system selects suitable sensor to become by the data of pre-acquired early period
Amount is to construct correlation ring, i.e. abnormality detection model for target vehicle.
Abnormality detection: vehicle abnormality detection system calculates each knot on correlation ring using the data acquired in real time
Correlation between point place sensor variable pair, and compares with correlation before this, by the size of correlation variation come
Judge whether automobile has occurred exception.
As a result it submits: being detected once abnormal, will allow and edge calculations device alerts driver and feed back to result
Cloud Server, if vehicle is normal always, it is anti-that result submits module then to carry out state to Cloud Server with the fixed period
Feedback, and without other operations.
Embodiment:
1, data acquire: vehicle abnormality detection system is designed to passively read data in CAN bus and passes through OBD interface
And analysis and abnormality detection are carried out in the local of edge device.And CAN cannot read disappearing in vehicle abnormality detection system
Breath, vehicle abnormality detection system cannot be written data to CAN bus, take such mode that can vehicle abnormality be detected
System is independently of CAN bus.So vehicle abnormality detection system will not influence the normal operation of CAN bus, will not use
Any calculating and the communication resource in CAN bus.Simultaneously as the independence with CAN bus, so ensure that total even if CAN
Line is invaded by attacker, also can support vehicles abnormality detection system normally detect exception, and find that vehicle person under attack attacks
It hits.Before carrying out abnormality detection, vehicle abnormality detection system can acquire some sensing datas in advance, and they are sent
Correlation ring is established to correlating module for abnormality detection.And in normal driving process, the number that acquires in real time
The exception for being used to detect generation according to abnormality detection module can be sent to.
2, submodule correlation analysis: is established to the computational submodule of correlation and correlation ring comprising sensor variable
Block.Correlation calculations submodule utilizes Pearson correlation coefficient such as formula (1)) sensor is calculated as correlation judging quota
Correlation between:
Wherein, n is the length of the sample sequence X and Y after resampling, and δ X and δ Y are the standard deviations of sample sequence X and Y.
In general, the absolute value of Corr is closer from 1, (linear) correlation between two correlated variables is stronger.In order to obtain more
Correlation between more sensors, and computation burden is reduced, it is directly related that vehicle abnormality detection system finds out those first
Sensor variable pair, that is, the sensor variable of the same physical quantity is measured, for example, the pass between car speed and GPS velocity
System;Then, vehicle abnormality detection system can be handled sensing data according to some theoretical knowledges and experience, such as will
Wheel velocity carries out differential, to generate correlation with acceleration.Vehicle abnormality detection system can calculate these biographies selected
The Pearson correlation coefficient of sensor variable pair.
Correlation ring setting up submodule is by forming ring topology to according to the sequence of correlation for different sensors variable
Structure, so that the computation complexity of abnormality detection be greatly reduced.Vehicle abnormality detection system selects those related and pairs of biographies
They are simultaneously organized into cyclic structure by sensor.Cyclic structure has the advantage that.Firstly, only those biographies in correlation ring
Correlation between sensor variable just needs to calculate, and can save a large amount of computing resource in this way, reduces computation burden.Secondly,
It, can will be more under the restrictive condition of the resource of limited limited time because the computation complexity of correlation ring is lower
Sensor is brought into computing system, so as to guarantee the validity and accuracy of detection.Finally, using cyclic structure, energy
Enough guarantee that each node is involved in in the calculating of two correlations, can similarly guarantee that vehicle abnormality detection system is different
The accuracy often detected.Simplest ring (each node is needed to participate in correlation calculations twice) only includes three nodes,
In order to construct a more complicated ring, a kind of intuitive method be exactly be that all relevant sensor nodes construct a figures
Then structure looks for the ring that a correlation between long enough and variable is greater than a certain threshold value.
3, abnormality detection module: vehicle abnormality detection system calculates n to the correlation between sensor node, and will meter
It calculates result to be compared with relative coefficient before, the state of comprehensive n node, to obtain last as a result, n is
The number of node in correlation ring, as a result describes whether system has occurred exception.In order to reduce noise to relative coefficient
The error of calculating and the sensitivity for guaranteeing detection, vehicle abnormality detection system use in the abnormality detection stage of each node
Sliding window method that is, in each time window has 1000 samples and is used to calculate the correlation in each period
The value of property coefficient, the step number of time window are 100, after each period, have 100 new sample entry time windows
Mouthful, the sample exit time window in 100 Geju City, to achieve the effect that sliding window.Vehicle abnormality detection system has used it
The intermediate data of preceding calculating simplifies calculating process, to can also reduce the use of memory space.And these intermediate variables exist
It is real-time update in memory, and when stroke terminates each time, these data can be worked as in the last one normal stroke
It in middle deposit storage medium, and is used when new stroke starts, can guarantee the continuity detected every time in this way,
And normal data between having used, so as to enhance the accuracy of detection.Then vehicle abnormality detection system can calculate newly
Time window in correlation ring the Pearson correlation coefficient of the sample of all the sensors variable pair and with before be
Number compares, and adaptively judges whether there is abnormal generation.
The method that the coefficient in new time window is calculated using existing intermediate data that we introduce is as follows:
When intermediate data has been extracted, we can utilize newest n2A sample data and the intermediate data of extraction come
Calculate all n1+n2The total relative coefficient of a data.Calculating n1+n2After the average value of a sample data, they
Standard deviation δ can be expressed as follows:
It can obtain,
In formula (3),
And becauseFor constant, so
AndNew n can be passed through2A sample data is calculated,For constant
It is available, so thus the standard deviation δ of all sample datas formula can be obtained.
And it calculates newlyWhen, molecule is divided into following two parts and is calculated:
Section 2 can use rear n in formula (7)2A sample data is directly calculated, but first item needs benefit
It is calculated with formula (8);In formula (8), due toAndFor constant, so
The first item of formula (8) can pass through preceding n1?It is acquired using formula (1).To sum up, all n1+n2A sample
ThisValue can acquire.
After obtaining the correlation results of number of sensors pair of i-th of time window, when following two condition
(11) when (12) meet simultaneously, vehicle abnormality detection system can then determine that exception occurs in automobile.
Corri≤Corri-1 (11)
CorriAnd Corri-1It is the relative coefficient of certain a pair of sensors in i-th and (i-1)-th time window, ε is vehicle
Abnormality detection system determines whether vehicle abnormal threshold value occurs.And this threshold value influence whether abnormality detection accuracy rate and
False positive value, vehicle abnormality detection sensitivity can be adjusted by this threshold value.This threshold value deciding means based on ratio come
The mode of detection vehicle abnormality state can be avoided the shadow of correlation order of magnitude between each node in correlation ring itself
It rings, so as to more accurately detect exception.Meanwhile new relative coefficient can be compared with secondary new relative coefficient,
To ensure that the real time and dynamic of vehicle abnormality detection, avoiding influences the standard of abnormality detection because environmental difference is larger
True rate and robustness.
4, result submits module: vehicle abnormality detection system obtains from abnormality detection module as a result, and being sent to
To the communication module of edge device.If vehicle is in normal condition, edge calculations equipment can be will test just with the fixed period
Normal result is sent to Cloud Server without making any other operation.Once noting abnormalities, vehicle abnormality detection system vehicle is different
Normal detection system just sounds an alarm immediately reminds driver and result is fed back to Cloud Server.In this module, edge calculations
Equipment can save a large amount of bandwidth and energy, and thus, it is possible to accelerate exception because only transmitting those required data
The reaction speed of processing reduces the extent of injury continued to cause.In addition to this, vehicle abnormality detection system can be by from cloud
The reply confirmation message (ACK) that server end periodically returns verifies the connectivity of edge calculations equipment and Cloud Server.
Claims (5)
1. a kind of Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations, which is characterized in that including as follows
Step:
Step 1, data acquire: reading equipment by board Controller Area Network bus protocol and read by onboard diagnostic system interface
Access evidence;
Step 2, correlation analysis: using Pearson correlation coefficient as the measurement index of correlation, each sensing data is measured
Between correlation, and select the high sensor pair of correlation, construct the model of abnormality detection;
Step 3, abnormality detection: the correlation by calculating respective sensor pair is used as the Judging index of abnormality detection
Step 4, testing result is submitted: passing through the different submissions for submitting strategy to carry out result under normal and abnormal two states.
2. the Multi-sensor Fusion vehicle safety method for detecting abnormality according to claim 1 based on edge calculations, special
Sign is that the data acquisition, which refers to, obtains CAN bus frame by collecting the signal in CAN bus, and then decoding obtains
Data on each sensor.
3. the Multi-sensor Fusion vehicle safety method for detecting abnormality according to claim 1 based on edge calculations, special
Sign is that the correlation analysis refers to that data utilize Pearson correlation coefficient by right on each sensor obtained to step 1
Correlation calculations are carried out, then carry out tissue according to the power of correlation, and these sensor variables are formed into correlation ring, phase
The every two adjacent node of Guan Xinghuan all has correlation.
4. the Multi-sensor Fusion vehicle safety method for detecting abnormality according to claim 1 based on edge calculations, special
Sign is, the abnormality detection, which refers to, substitutes into the correlation obtained in step 2 for data on each sensor that step 1 obtains
In the respective sensor variable of ring, it is calculated the Pearson correlation coefficient of respective sensor, and by existing sensing data
The Pearson correlation coefficient that obtained Pearson correlation coefficient and previous time window obtains substitutes into following formula,
Corri≤Corri-1
In above formula, CorriAnd Corri-1It is the relative coefficient of certain a pair of sensors in i-th and (i-1)-th time window, ε is
Vehicle abnormality detection system determines whether vehicle abnormal threshold value occurs, if meeting above formula, determines that vehicle is abnormal, vehicle
Abnormality detection sensitivity is adjusted by parameter ε.
5. the Multi-sensor Fusion vehicle safety method for detecting abnormality according to claim 1 based on edge calculations, special
Sign is that the testing result submission refers to that vehicle obtained in step 3 is belonged to normal or abnormal result to be taken to cloud
The system at business device end carries out result submission, when testing result is normal, then periodically submits testing result, works as testing result
When being abnormal, then testing result is submitted in real time, so that user and system can carry out exception to abnormity early warning at the first time
Processing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910183238.5A CN110034968A (en) | 2019-03-12 | 2019-03-12 | Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910183238.5A CN110034968A (en) | 2019-03-12 | 2019-03-12 | Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110034968A true CN110034968A (en) | 2019-07-19 |
Family
ID=67235827
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910183238.5A Pending CN110034968A (en) | 2019-03-12 | 2019-03-12 | Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110034968A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112394703A (en) * | 2019-08-14 | 2021-02-23 | 中车时代电动汽车股份有限公司 | Vehicle fault management system |
CN113132307A (en) * | 2019-12-31 | 2021-07-16 | 厦门雅迅网络股份有限公司 | Method, device and storage medium for detecting illegal behaviors suffered by CAN network in vehicle |
CN113296952A (en) * | 2021-06-01 | 2021-08-24 | 南京大学 | System and method for performing edge calculation by adopting high-order differential of analog sensor |
CN113422720A (en) * | 2021-06-22 | 2021-09-21 | 河北卓智电子技术有限公司 | Anomaly detection method based on edge computing gateway |
CN114694368A (en) * | 2020-12-28 | 2022-07-01 | 比亚迪股份有限公司 | Vehicle management and control system |
CN115176444A (en) * | 2020-02-11 | 2022-10-11 | 大陆汽车科技有限公司 | Intrusion and anomaly detection method based on edge calculation |
CN117879974A (en) * | 2024-03-11 | 2024-04-12 | 西昌学院 | Network security protection method based on edge calculation |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106647724A (en) * | 2017-02-15 | 2017-05-10 | 北京航空航天大学 | T-BOX information security detection and protection method based on vehicle anomaly data monitoring |
CN109150847A (en) * | 2018-07-27 | 2019-01-04 | 北京梆梆安全科技有限公司 | A kind of method and apparatus for the network intrusions risk detecting vehicle |
-
2019
- 2019-03-12 CN CN201910183238.5A patent/CN110034968A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106647724A (en) * | 2017-02-15 | 2017-05-10 | 北京航空航天大学 | T-BOX information security detection and protection method based on vehicle anomaly data monitoring |
CN109150847A (en) * | 2018-07-27 | 2019-01-04 | 北京梆梆安全科技有限公司 | A kind of method and apparatus for the network intrusions risk detecting vehicle |
Non-Patent Citations (1)
Title |
---|
ZICHANG WANG ET AL: "Detecting Vehicle Anomaly by Sensor Consistency: An Edge Computing Based Mechanism", 《IEEE XPLORE-2018 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM)》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112394703A (en) * | 2019-08-14 | 2021-02-23 | 中车时代电动汽车股份有限公司 | Vehicle fault management system |
CN113132307A (en) * | 2019-12-31 | 2021-07-16 | 厦门雅迅网络股份有限公司 | Method, device and storage medium for detecting illegal behaviors suffered by CAN network in vehicle |
CN113132307B (en) * | 2019-12-31 | 2023-09-01 | 厦门雅迅网络股份有限公司 | Method, device and storage medium for detecting illegal actions suffered by CAN network in vehicle |
CN115176444A (en) * | 2020-02-11 | 2022-10-11 | 大陆汽车科技有限公司 | Intrusion and anomaly detection method based on edge calculation |
CN114694368A (en) * | 2020-12-28 | 2022-07-01 | 比亚迪股份有限公司 | Vehicle management and control system |
CN113296952A (en) * | 2021-06-01 | 2021-08-24 | 南京大学 | System and method for performing edge calculation by adopting high-order differential of analog sensor |
CN113422720A (en) * | 2021-06-22 | 2021-09-21 | 河北卓智电子技术有限公司 | Anomaly detection method based on edge computing gateway |
CN113422720B (en) * | 2021-06-22 | 2023-04-25 | 河北卓智电子技术有限公司 | Anomaly detection method based on edge computing gateway |
CN117879974A (en) * | 2024-03-11 | 2024-04-12 | 西昌学院 | Network security protection method based on edge calculation |
CN117879974B (en) * | 2024-03-11 | 2024-05-14 | 西昌学院 | Network security protection method based on edge calculation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110034968A (en) | Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations | |
CN106767852B (en) | A kind of method, apparatus and equipment generating detection target information | |
CN108981781B (en) | Predictive analysis system and method for analyzing and detecting machine sensor faults | |
US8781688B2 (en) | Method and system for combining sensor data | |
CN1940591B (en) | System and method of target tracking using sensor fusion | |
CN110834642B (en) | Vehicle deviation identification method and device, vehicle and storage medium | |
US20170243478A1 (en) | Vehicle detection and tracking based on wheels using radar and vision | |
Peng et al. | Intelligent method for identifying driving risk based on V2V multisource big data | |
CN111580116A (en) | Method for evaluating target detection performance of vehicle-mounted system and electronic equipment | |
CN111612378A (en) | Potential collision risk prediction method and device and computer equipment | |
CN116611620B (en) | Smart city safety collaborative management information system | |
US6292738B1 (en) | Method for adaptive detection of engine misfire | |
CN103914889B (en) | A kind of system and method that the collection behavior of car speed is carried out to abnormality detection | |
CN115451968A (en) | Multi-sensor synchronous fusion method and device for intelligent driving | |
CN109263649B (en) | Vehicle, object recognition method and object recognition system thereof in automatic driving mode | |
CN105698915B (en) | A kind of optical fiber Recognition of Vibration Sources method and device, system | |
CN106796126A (en) | Method and system for providing the dynamic error value of dynamic measurement in real time | |
Dietrich et al. | Detecting external measurement disturbances based on statistical analysis for smart sensors | |
CN111881874A (en) | Parking space identification method, equipment and system | |
CN115980718B (en) | Lens dirt detection method and device, electronic equipment and readable storage medium | |
JP6555132B2 (en) | Moving object detection device | |
CN110411499A (en) | The appraisal procedure and assessment system of sensor detection recognition capability | |
CN110660217B (en) | Method and device for detecting information security | |
CN112781556A (en) | Well lid transaction monitoring method and device based on multi-data fusion filtering | |
JP2020091704A (en) | Analysis system, onboard device, management device, analysis method, and analysis program |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190719 |