CN108280482B - Driver identification method, device and system based on user behaviors - Google Patents
Driver identification method, device and system based on user behaviors Download PDFInfo
- Publication number
- CN108280482B CN108280482B CN201810086755.6A CN201810086755A CN108280482B CN 108280482 B CN108280482 B CN 108280482B CN 201810086755 A CN201810086755 A CN 201810086755A CN 108280482 B CN108280482 B CN 108280482B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- user data
- model
- data
- driver
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a driver identification method, a device and a system based on user behaviors, wherein the method comprises the following steps: collecting user data of a vehicle in a preset time period; carrying out data cleaning on user data; packaging user data into a feature vector by adopting a word2vec model; and inputting the characteristic vector into a driver identification model corresponding to the vehicle, and identifying to obtain a corresponding driver identification result. According to the method, after the user data of the vehicle in the preset time period is obtained, the word2vec model is adopted, the user data is packaged into the feature vector, and the obtained feature vector is input into the pre-trained driver identification model, so that the driver can be identified and obtained as the owner or the non-owner.
Description
Technical Field
The invention relates to the technical field of vehicle intellectualization, in particular to a driver identification method, device and system based on user behaviors.
Background
The current methods for identifying drivers generally include biometric methods and vehicle driving data analysis methods. The method has the advantages of high identification accuracy, high corresponding implementation cost and additional participation of a user, for example, the method needs to record the facial features of the driver in advance and train a model by a face identification method of the camera, and then the driver can be identified by aligning the face to the camera for a period of time. The vehicle driving data analysis method is a scheme for identifying a driver by performing logistic regression analysis after user data are collected, but in the scheme, data are discrete, the sequence relation among the data cannot be shown, and finally the driver cannot be accurately identified according to the operation sequence characteristics of the user.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a driver identification method, a driver identification device and a driver identification system based on user behaviors.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the driver identification method based on the user behavior comprises the following steps:
collecting user data of a vehicle in a preset time period;
carrying out data cleaning on user data;
packaging user data into a feature vector by adopting a word2vec model;
and inputting the characteristic vector into a driver identification model corresponding to the vehicle, and identifying to obtain a corresponding driver identification result.
Further, for each vehicle, training and obtaining a driver identification model corresponding to the vehicle through the following steps:
acquiring historical user data of a vehicle within a period of time;
classifying and cleaning historical user data;
packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
inputting a multi-dimensional feature vector set as input data into a deep neural network for training, and then taking the trained deep neural network as a driver recognition model;
the deep neural network is a multilayer fully-connected neural network and specifically comprises an input layer, a hidden layer and an output layer, wherein the output layer is a binary classifier and is used for identifying whether a driver recognition result is an owner or a non-owner.
Further, the step of encapsulating the historical user data into a multidimensional feature vector set by using a word2vec model specifically includes:
arranging the historical user data in sequence according to the time sequence;
inputting user data in each driving as input data into a word2vec model to obtain a feature vector output by the model;
and generating a multi-dimensional feature vector set from a plurality of feature vectors output by the word2vec model according to the time sequence.
Further, the user data includes vehicle setting data, user driving behavior data, user trip preference data, and interaction data between the user and the vehicle.
Further, the method also comprises the following steps:
switching to an owner personalized service model to provide personalized service for the owner of the vehicle under the condition that the driver identification result is the owner of the vehicle;
and sending corresponding alarm information to the vehicle owner aiming at the condition that the driver identification result is not the vehicle owner.
Further, the output layer adopts a softmax function or a logistic regression function to construct a binary classifier.
Further, the driver identification model is obtained by training in a cloud server, and the driver identification method executes the identification step in the cloud server.
Driver identification apparatus based on user behavior, comprising:
at least one processor;
at least one memory to store a plurality of instructions;
the plurality of instructions are loaded by the at least one processor and implement the user behavior based driver identification method.
The driver identification system based on the user behaviors comprises a vehicle-mounted terminal and a cloud server, wherein the vehicle-mounted terminal is connected with the cloud server;
the vehicle-mounted terminal is used for: collecting user data of a vehicle in a preset time period and sending the user data to a cloud server;
the cloud server is used for:
carrying out data cleaning on user data;
packaging user data into a feature vector by adopting a word2vec model;
and inputting the characteristic vector into a driver identification model corresponding to the vehicle, and identifying to obtain a corresponding driver identification result.
Further, the vehicle-mounted terminal is also used for collecting historical user data of the vehicle in a period of time and sending the historical user data to the cloud server;
the cloud server comprises a model training module, wherein the model training module is used for training a corresponding driver identification model for each vehicle, and the model training module is obtained by training the following steps:
acquiring historical user data of a vehicle within a period of time;
classifying and cleaning historical user data;
packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
inputting a multi-dimensional feature vector set as input data into a deep neural network for training, and then taking the trained deep neural network as a driver recognition model;
the deep neural network is a multilayer fully-connected neural network and specifically comprises an input layer, a hidden layer and an output layer, wherein the output layer is a binary classifier and is used for identifying whether a driver recognition result is an owner or a non-owner.
The invention has the beneficial effects that: according to the method, after the user data of the vehicle in the preset time period is obtained, the word2vec model is adopted, the user data is packaged into the feature vector, and the obtained feature vector is input into the pre-trained driver identification model, so that the driver can be identified and obtained as the owner or the non-owner.
In addition, the invention can also provide personalized service for the vehicle owner according to the driver identification result, or send alarm information to the vehicle owner when the identification result is not the vehicle owner, thereby realizing anti-theft monitoring.
Drawings
FIG. 1 is a flow chart of a driver identification method based on user behavior of the present invention;
FIG. 2 is a schematic illustration of classification of historical user data in an embodiment of the present invention;
FIG. 3 is a block diagram of the structure of the driver recognition apparatus based on user behavior according to the present invention;
fig. 4 is a block diagram of the structure of the driver recognition system based on user behavior of the present invention.
Detailed Description
Method embodiment
Referring to fig. 1, the present embodiment provides a driver identification method based on user behavior, including the following steps:
s1, collecting user data of the vehicle in a preset time period; the preset time period is a user-defined time period which can be 3 minutes, 5 minutes or 1 hour and is set according to driving habits;
s2, carrying out data cleaning on the user data; the step is used for removing noise data in the user data;
s3, packaging user data into a feature vector by adopting a word2vec model;
and S4, inputting the characteristic vector into a driver identification model corresponding to the vehicle, and identifying to obtain a corresponding driver identification result.
According to the method, after the user data of the vehicle in the preset time period is obtained, the word2vec model is adopted, the user data is packaged into the feature vector, and the obtained feature vector is input into the pre-trained driver identification model, so that the driver can be identified and obtained as the owner or the non-owner.
Further as a preferred embodiment, for each vehicle, a driver identification model corresponding to the vehicle is obtained through training by the following steps:
s01, acquiring historical user data of the vehicle in a period of time; here, the period of time may be one week, one month, or any set period of time as long as sufficient data can be acquired. In addition, the historical user data acquired in the step is acquired by the vehicle timing acquisition.
S02, classifying historical user data and cleaning the data; in this step, the historical user data is classified, mainly the historical user data is classified according to different users, so as to obtain the user data corresponding to each user. As shown in fig. 2, after the historical user data is identified, the historical user data is classified into data of 3 users, namely, user 1, user 2 and user 3. So that in subsequent neural network training, the input data is in units of users. Each vehicle may correspond to a plurality of users, and the scheme can ensure that each user can be correctly identified.
S03, packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
s04, inputting the multi-dimensional feature vector set as input data into a deep neural network for training, and then taking the trained deep neural network as a driver recognition model;
the deep neural network is a multilayer fully-connected neural network and specifically comprises an input layer, a hidden layer and an output layer, wherein the output layer is a binary classifier and is used for identifying whether a driver recognition result is an owner or a non-owner.
After historical user data are packaged into a multi-dimensional feature vector set through a word2vec model, a driver recognition model is obtained through deep neural network training, and therefore whether a driver is an owner or not can be accurately and efficiently recognized and obtained after user data of a vehicle in a preset time period are collected.
Further as a preferred embodiment, the step S03 specifically includes:
s031, arrange the historical user data according to the time sequence in order;
s032, inputting user data in each driving as input data into a word2vec model to obtain a feature vector output by the model;
s033, generating a multi-dimensional feature vector set from a plurality of feature vectors output by the word2vec model according to the time sequence.
Correspondingly, step S3 is similar to step S03, and specifically includes: and inputting the user data serving as input data into the word2vec model to obtain the feature vector output by the model. After user data are packaged into feature vectors through the word2vec model, the obtained feature vectors are used as input data of the driver identification model, and then the corresponding driver identification result can be obtained through calculation according to the driver identification model.
The word2vec model can accurately compare the similarity of a plurality of user data corresponding to different strokes and sorted according to the time sequence, historical user data can be conveniently input into the deep neural network for training after being packaged into a multi-dimensional feature vector set through the word2vec model, and the calculated amount in the neural network training process is reduced. In step S3, the user data is packaged into corresponding feature vectors through a word2vec model, and results can be conveniently and quickly obtained through deep neural network identification.
Further as a preferred embodiment, the user data includes vehicle setting data, user driving behavior data, user trip preference data, and interaction data between the user and the vehicle.
Preferably, the vehicle setting data includes at least one of: the height of the seat, the inclination of the seat, the height of the steering wheel, the extending range of the steering wheel, the up-down and left-right angles of the left rearview mirror and the right rearview mirror, the temperature of an air conditioner, the air volume, the wind power and the blowing mode, the weight of a driver and the volume of a system. The vehicle setting data is acquired automatically by an on-board system of the automobile, because the setting preference of a user is relatively fixed, the height, the front and rear positions, the angle of a rearview mirror and the like of a seat do not need to be adjusted every time of a stroke, if more fixed items are adjusted in a certain stroke or the adjustment range of some items such as the height of the seat is larger, the items are probably considered to be different drivers, and therefore the vehicle setting data can be used as an input data source of a driver identification model.
Preferably, the user driving behavior data comprises at least one of the following data: the emergency acceleration frequency, the emergency braking frequency, the emergency turning speed, the emergency turning amplitude, the average speed of roads with different levels, the horn times and frequency of each stroke, and the high beam use times and frequency of each stroke. The data such as speed in the user driving behavior data are directly acquired through the sensor, and other parameters are obtained through calculation by combining the data acquired by the sensor. For example, the rapid acceleration frequency may be calculated by obtaining an acceleration from a vehicle speed acquired by a vehicle speed sensor, calculating a derivative of the acceleration, recording a case where the rate of change exceeds a set value as a rapid acceleration once, and calculating the total number of rapid accelerations within a set time as the rapid acceleration frequency. In addition, the average speeds of the roads of different levels are mainly calculated by combining the driving roads of the automobiles, and the roads can be distinguished according to the historical driving data of the automobiles, so that the obtained average speeds of the different roads are calculated and distinguished. And the using times of the high beam in each stroke are acquired after the high beam is directly communicated with the automobile CAN bus. The driving habits of the same driver, such as the driving speed, sharp turns, sharp brakes and sharp accelerations, are relatively stable, and if the driving behavior of the driver shows a large deviation from the past driving data in a certain trip, the current driver may be a different driver. Therefore, the driving behavior data of the user can be arbitrarily selected or calculated as an input data source of the driver recognition model according to needs.
Preferably, the user travel preference data comprises at least one of the following data: the travel time period, the corresponding frequent travel route, the frequent travel destination, the travel destination type, the travel times per day and the number of the co-workers. The travel destination of the user is relatively fixed under the same conditions of the same departure time, the same place and the like. If abnormal travel behaviors are caused when a certain trip is taken, the current driver is possibly a different driver. The user travel preference data may also be used for driver identification.
Preferably, the interaction data between the user and the vehicle comprises at least one of the following data: radio station listening data, music playing data, vehicle control data. Specifically, the radio listening data specifically includes listening time periods, listening channels, and listening contents, the music playing data specifically includes song types, singers, and songs, and the vehicle control data specifically includes manually controlled vehicle functions, voice controlled vehicle functions, air conditioner control data, and the like. The radio channel, the type of the played song, the favorite singer and the control function of the manual or voice-operated vehicle which are frequently listened to by the same user are relatively fixed, and if the information interacted with the vehicle has large deviation in a certain journey, the current driver may be different drivers. Therefore, the interaction data between the user and the vehicle can accurately reflect the condition of the driver.
The method has the advantages that the vehicle setting data, the user driving behavior data, the user trip preference data and the interaction data between the user and the vehicle are integrated, driver identification can be accurately carried out, and if the four data show different degrees of deviation from past behaviors, the current driver is probably not the owner of the vehicle.
Further as a preferred embodiment, the method further comprises the steps of:
switching to an owner personalized service model to provide personalized service for the owner of the vehicle under the condition that the driver identification result is the owner of the vehicle;
and sending corresponding alarm information to the vehicle owner aiming at the condition that the driver identification result is not the vehicle owner.
In the step, different functions can be correspondingly executed according to the condition that the recognition result is that the vehicle owner is the vehicle owner or not, the vehicle owner is switched to the vehicle owner personalized service model, personalized service is provided for the vehicle owner, and the specific vehicle owner personalized service model can be set by a driver in advance or obtained by automatic training according to the use data of the driver. When the recognition result is the owner, the personalized service is directly provided for the owner, the owner does not need to perform manual operation, and the driving safety is improved. In addition, corresponding alarm information is directly sent to the vehicle owner under the condition that the identification result is not the vehicle owner, the alarm information can be in the form of short messages or micro messages and the like, and a window is pushed, so that the abnormal condition of the vehicle owner is timely reminded, and the anti-theft monitoring is realized. Furthermore, when corresponding warning information is sent to the vehicle owner, real-time position information of the vehicle can be sent at the same time, and the vehicle can be conveniently located and tracked.
Further as a preferred embodiment, the output layer adopts a softmax function or a logistic regression function to construct a binary classifier. And constructing a binary classifier through a softmax function or a logistic regression function, and accurately training to obtain a driver identification model. And when the constructed driver identification model needs to identify information other than the owner/non-owner, the output layer can be constructed by using the softmax function.
Further, as a preferred embodiment, the driver identification model is obtained by training in a cloud server, and the driver identification method executes the identification step in the cloud server. Specifically, step S1 of the method is executed at the vehicle end, and steps S2-S4 are executed at the cloud server. When the driver identification model needs to be updated, the driver identification model only needs to be updated at the cloud server, and each vehicle does not need to be changed, so that the method is simple and efficient.
Device embodiment
Referring to fig. 3, the present embodiment provides a driver recognition apparatus based on user behavior, including:
at least one processor 100;
at least one memory 200 for storing a plurality of instructions;
the plurality of instructions are loaded by the at least one processor 100 and implement the user behavior based driver identification method.
The driver recognition device based on the user behaviors can execute the driver recognition method based on the user behaviors provided by the method embodiment of the invention, can execute any combination of the implementation steps of the method embodiment, and has corresponding functions and beneficial effects of the method.
System embodiment
Referring to fig. 4, the embodiment provides a driver identification system based on user behavior, which includes a vehicle-mounted terminal and a cloud server, where the vehicle-mounted terminal is connected with the cloud server;
the vehicle-mounted terminal is used for: collecting user data of a vehicle in a preset time period and sending the user data to a cloud server;
the cloud server is used for:
carrying out data cleaning on user data;
packaging user data into a feature vector by adopting a word2vec model;
and inputting the characteristic vector into a driver identification model corresponding to the vehicle, and identifying to obtain a corresponding driver identification result.
Further as a preferred embodiment, the vehicle-mounted terminal is further configured to collect historical user data of the vehicle within a period of time and send the historical user data to the cloud server;
the cloud server comprises a model training module, wherein the model training module is used for training a corresponding driver identification model for each vehicle, and the model training module is obtained by training the following steps:
acquiring historical user data of a vehicle within a period of time;
classifying and cleaning historical user data;
packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
inputting a multi-dimensional feature vector set as input data into a deep neural network for training, and then taking the trained deep neural network as a driver recognition model;
the deep neural network is a multilayer fully-connected neural network and specifically comprises an input layer, a hidden layer and an output layer, wherein the output layer is a binary classifier and is used for identifying whether a driver recognition result is an owner or a non-owner.
The driver recognition system based on the user behaviors can execute the driver recognition method based on the user behaviors provided by the method embodiment of the invention, can execute any combination of the implementation steps of the method embodiment, and has corresponding functions and beneficial effects of the method.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (9)
1. The driver identification method based on the user behaviors is characterized by comprising the following steps of:
collecting user data of a vehicle in a preset time period;
carrying out data cleaning on user data;
packaging user data into a feature vector by adopting a word2vec model;
inputting the characteristic vector into a pre-trained driver recognition model corresponding to the vehicle, and recognizing to obtain a corresponding driver recognition result;
the training process of the driver recognition model comprises the following steps: packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
the step of packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model specifically comprises the following steps:
arranging the historical user data in sequence according to the time sequence;
inputting user data in each driving as input data into a word2vec model to obtain a feature vector output by the model;
and generating a multi-dimensional feature vector set from a plurality of feature vectors output by the word2vec model according to the time sequence.
2. The method according to claim 1, wherein for each vehicle, a driver recognition model corresponding to the vehicle is obtained by training:
acquiring historical user data of a vehicle within a period of time;
classifying and cleaning historical user data;
packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
inputting a multi-dimensional feature vector set as input data into a deep neural network for training, and then taking the trained deep neural network as a driver recognition model;
the deep neural network is a multilayer fully-connected neural network and specifically comprises an input layer, a hidden layer and an output layer, wherein the output layer is a binary classifier and is used for identifying whether a driver recognition result is an owner or a non-owner.
3. The user behavior-based driver recognition method according to claim 1, wherein the user data includes vehicle setting data, user driving behavior data, user travel preference data, and interaction data between the user and the vehicle.
4. The user behavior based driver identification method according to claim 1, further comprising the steps of:
switching to an owner personalized service model to provide personalized service for the owner of the vehicle under the condition that the driver identification result is the owner of the vehicle;
and sending corresponding alarm information to the vehicle owner aiming at the condition that the driver identification result is not the vehicle owner.
5. The user behavior-based driver identification method according to claim 2, wherein the output layer employs a softmax function or a logistic regression function to construct a binary classifier.
6. The user behavior based driver recognition method of claim 1, wherein the driver recognition model is trained on a cloud server, and the driver recognition method performs the recognition step on the cloud server.
7. Driver recognition apparatus based on user behavior, characterized by comprising:
at least one processor;
at least one memory to store a plurality of instructions;
the plurality of instructions are loaded by the at least one processor and implement the user behavior based driver recognition method of any of claims 1-6.
8. The driver identification system based on the user behaviors is characterized by comprising a vehicle-mounted terminal and a cloud server, wherein the vehicle-mounted terminal is connected with the cloud server;
the vehicle-mounted terminal is used for: collecting user data of a vehicle in a preset time period and sending the user data to a cloud server;
the cloud server is used for:
carrying out data cleaning on user data;
packaging user data into a feature vector by adopting a word2vec model;
inputting the characteristic vector into a pre-trained driver recognition model corresponding to the vehicle, and recognizing to obtain a corresponding driver recognition result;
the training process of the driver recognition model comprises the following steps: packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
the step of packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model specifically comprises the following steps:
arranging the historical user data in sequence according to the time sequence;
inputting user data in each driving as input data into a word2vec model to obtain a feature vector output by the model;
and generating a multi-dimensional feature vector set from a plurality of feature vectors output by the word2vec model according to the time sequence.
9. The user behavior based driver identification system of claim 8, wherein the vehicle terminal is further configured to collect historical user data of the vehicle over a period of time and send the historical user data to the cloud server;
the cloud server comprises a model training module, wherein the model training module is used for training a corresponding driver identification model for each vehicle, and the model training module is obtained by training the following steps:
acquiring historical user data of a vehicle within a period of time;
classifying and cleaning historical user data;
packaging historical user data into a multi-dimensional feature vector set by adopting a word2vec model;
inputting a multi-dimensional feature vector set as input data into a deep neural network for training, and then taking the trained deep neural network as a driver recognition model;
the deep neural network is a multilayer fully-connected neural network and specifically comprises an input layer, a hidden layer and an output layer, wherein the output layer is a binary classifier and is used for identifying whether a driver recognition result is an owner or a non-owner.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810086755.6A CN108280482B (en) | 2018-01-30 | 2018-01-30 | Driver identification method, device and system based on user behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810086755.6A CN108280482B (en) | 2018-01-30 | 2018-01-30 | Driver identification method, device and system based on user behaviors |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108280482A CN108280482A (en) | 2018-07-13 |
| CN108280482B true CN108280482B (en) | 2020-10-16 |
Family
ID=62805649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810086755.6A Active CN108280482B (en) | 2018-01-30 | 2018-01-30 | Driver identification method, device and system based on user behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108280482B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948729A (en) * | 2019-03-28 | 2019-06-28 | 北京三快在线科技有限公司 | Driver identification recognition methods and device, electronic equipment |
| CN110096499B (en) * | 2019-04-10 | 2021-08-10 | 华南理工大学 | User object identification method and system based on behavior time series big data |
| CN112307214A (en) * | 2019-07-26 | 2021-02-02 | 株式会社理光 | Deep reinforcement learning-based recommendation method and recommendation device |
| CN110969844A (en) * | 2019-11-19 | 2020-04-07 | 惠州市德赛西威汽车电子股份有限公司 | Method for calculating driving behavior similarity based on driving data and application |
| CN110901582A (en) * | 2019-11-19 | 2020-03-24 | 惠州市德赛西威汽车电子股份有限公司 | Stolen vehicle tracking method based on driving behavior similarity |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8527146B1 (en) * | 2012-01-30 | 2013-09-03 | Google Inc. | Systems and methods for updating vehicle behavior and settings based on the locations of vehicle passengers |
| EP2891589A2 (en) * | 2014-01-06 | 2015-07-08 | Harman International Industries, Incorporated | Automatic driver identification |
| CN105761329A (en) * | 2016-03-16 | 2016-07-13 | 成都信息工程大学 | Method of identifying driver based on driving habits |
| CN106128099A (en) * | 2016-07-01 | 2016-11-16 | 斑马信息科技有限公司 | Driver's recognition methods and device |
| CN106649819A (en) * | 2016-12-29 | 2017-05-10 | 北京奇虎科技有限公司 | Method and device for extracting entity words and hypernyms |
| CN107215307A (en) * | 2017-05-24 | 2017-09-29 | 清华大学深圳研究生院 | Driver identity recognition methods and system based on vehicle sensors correction data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105808737B (en) * | 2016-03-10 | 2021-04-06 | 腾讯科技（深圳）有限公司 | Information retrieval method and server |
| US9911223B2 (en) * | 2016-05-13 | 2018-03-06 | Yahoo Holdings, Inc. | Automatic video segment selection method and apparatus |
-
2018
- 2018-01-30 CN CN201810086755.6A patent/CN108280482B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8527146B1 (en) * | 2012-01-30 | 2013-09-03 | Google Inc. | Systems and methods for updating vehicle behavior and settings based on the locations of vehicle passengers |
| EP2891589A2 (en) * | 2014-01-06 | 2015-07-08 | Harman International Industries, Incorporated | Automatic driver identification |
| CN105761329A (en) * | 2016-03-16 | 2016-07-13 | 成都信息工程大学 | Method of identifying driver based on driving habits |
| CN106128099A (en) * | 2016-07-01 | 2016-11-16 | 斑马信息科技有限公司 | Driver's recognition methods and device |
| CN106649819A (en) * | 2016-12-29 | 2017-05-10 | 北京奇虎科技有限公司 | Method and device for extracting entity words and hypernyms |
| CN107215307A (en) * | 2017-05-24 | 2017-09-29 | 清华大学深圳研究生院 | Driver identity recognition methods and system based on vehicle sensors correction data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108280482A (en) | 2018-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108280482B (en) | Driver identification method, device and system based on user behaviors | |
| US11685386B2 (en) | System and method for determining a change of a customary vehicle driver | |
| CN107444402B (en) | Vehicle mode scheduling with learning user preferences | |
| US11021166B2 (en) | Systems and methods for driver scoring with machine learning | |
| US11945448B2 (en) | Vehicle telematics based driving assessment | |
| US10565460B1 (en) | Apparatuses, systems and methods for classifying digital images | |
| CN107368069B (en) | Automatic driving control strategy generation method and device based on Internet of vehicles | |
| US20180174457A1 (en) | Method and system using machine learning to determine an automotive driver's emotional state | |
| EP3496969A1 (en) | Method and system for providing information via collected and stored metadata using inferred attentional model | |
| CN107042824A (en) | System and method for detecting the accident in vehicle | |
| EP3343535A1 (en) | Information processing device, information processing method, and program | |
| AU2017382381A1 (en) | Method and system for providing artificial intelligence analytic (AIA) services using operator fingerprints and cloud data | |
| CN105844211A (en) | System and method for classifying a road surface | |
| CN105966405A (en) | Driver distraction detection system | |
| CN107226025A (en) | A kind of system and method for real-time monitoring driving behavior | |
| CN105766008B (en) | Communication system | |
| CN110654394A (en) | Driving control system and method and vehicle | |
| CN110556113A (en) | Vehicle control method based on voiceprint recognition and cloud server | |
| EP2052376B1 (en) | Adaptive interface providing apparatus and method | |
| CN105564429A (en) | Running safety pre-warning method and device | |
| CN114091547A (en) | Selectively enabling or disabling vehicle features based on driver classification | |
| CN111971218A (en) | Driver contour analysis and recognition | |
| Abdelrahman et al. | Driver behavior classification in crash and near-crash events using 100-CAR naturalistic data set | |
| CN107506698A (en) | The method of public transportation vehicle anti-fatigue-driving management based on Internet of Things | |
| CN109515441B (en) | Vehicle speed control system for intelligent driving vehicle |
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 | ||
| CB02 | Change of applicant information |
Address after: 510000 No.8 Songgang street, Cencun, Tianhe District, Guangzhou City, Guangdong Province Applicant after: GUANGZHOU XPENG AUTOMOBILE TECHNOLOGY Co.,Ltd. Address before: 510000 nine Guangdong, Guangzhou 333, Jianshe Road 245, Guangzhou, China Applicant before: GUANGZHOU XPENG AUTOMOBILE TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |