CN108280482A - Driver's recognition methods based on user behavior, apparatus and system - Google Patents
Driver's recognition methods based on user behavior, apparatus and system Download PDFInfo
- Publication number
- CN108280482A CN108280482A CN201810086755.6A CN201810086755A CN108280482A CN 108280482 A CN108280482 A CN 108280482A CN 201810086755 A CN201810086755 A CN 201810086755A CN 108280482 A CN108280482 A CN 108280482A
- Authority
- CN
- China
- Prior art keywords
- driver
- data
- vehicle
- user
- identification
- 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
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 driver's recognition methods based on user behavior, apparatus and systems, and this approach includes the following steps：The user data of collection vehicle within a preset period of time；Data cleansing is carried out to user data；Using word2vec models, user data is packaged into a feature vector；This feature vector is input to the corresponding driver's identification model of the vehicle, identification obtains corresponding driver's recognition result.After the present invention is by vehicle user data within a preset period of time, using word2vec models, user data is packaged into a feature vector, to which the feature vector of acquisition is input to advance trained driver's identification model, you can it is car owner or non-car owner that identification, which obtains driver, and this method is without increasing additional hardware device, can driver's identification directly be carried out according to the user data of vehicle, application cost is low, convenient, fast, can be widely applied in vehicle intellectualized industry.
Description
Technical field
The present invention relates to Vehicular intelligent technical field, more particularly to based on user behavior driver's recognition methods,
Apparatus and system.
Background technology
Driver's knowledge method for distinguishing is carried out at present generally comprises the methods of biological characteristic method and vehicle operation data analytic approach.
Wherein, biological characteristic method refers to being known to driver by the biological characteristic of camera, fingerprint, vocal print or even iris et al.
Not, its advantage is that recognition accuracy is high, but corresponding implementation cost is high, and needs the additional participation of user, such as by taking the photograph
As the recognition of face method of head, the face characteristic for having prerecorded driver, training pattern is needed hereafter face alignment to be needed to take the photograph
As head can just be identified for a period of time, this scheme the restrictions such as is taken by light, orientation, cost and interaction.Vehicle travels
Data analysis method is to carry out logistic regression analysis, the scheme of progress driver's identification, but this type after collecting user data
Scheme is discrete between data, and the precedence relationship between each data cannot show, finally cannot very accurate basis
The sequence of operation feature of user identifies driver.
Invention content
In order to solve the above technical problems, the object of the present invention is to provide the driver identification sides based on user behavior
Method, apparatus and system.
The technical solution adopted by the present invention to solve the technical problems is：
Driver's recognition methods based on user behavior, includes the following steps：
The user data of collection vehicle within a preset period of time；
Data cleansing is carried out to user data；
Using word2vec models, user data is packaged into a feature vector；
This feature vector is input to the corresponding driver's identification model of the vehicle, identification obtains corresponding driver's identification
As a result.
Further, it for each vehicle, is trained by following steps and obtains the corresponding driver's identification model of vehicle：
Obtain the historical use data of vehicle whithin a period of time；
Classification and data cleansing are carried out to historical use data；
Using word2vec models, historical use data is packaged into multidimensional characteristic vectors collection；
Using multidimensional characteristic vectors collection as input data, it is input to after deep neural network is trained, it will be trained
Deep neural network is as driver's identification model；
The deep neural network is the neural network that multilayer connects entirely, specifically includes input layer, hidden layer and output
Layer, and output layer is binary classifier, is car owner or non-car owner for identifying driver's recognition result.
Further, described the step of using word2vec models, historical use data is packaged into multidimensional characteristic vectors collection,
It specifically includes：
By historical use data sequential sequentially in time；
User data when using each driving is input to as input data in word2vec models, obtains model output
Feature vector；
According to time sequencing, multiple feature vectors that word2vec models are exported generate a multidimensional characteristic vectors collection.
Further, the user data includes vehicle setting data, user's driving behavior data, user's trip preference data
And the interaction data between user and vehicle.
Further, further comprising the steps of：
The case where for driver's recognition result being car owner, it is switched to car owner's Personalized Service Model, is provided for car owner a
Property service；
The case where for driver's recognition result being non-car owner, corresponding warning message is sent to car owner.
Further, the output layer builds binary classifier using softmax functions or logistic regression function.
Further, driver's identification model is trained in a cloud server and is obtained, and driver's recognition methods
Identification step is executed in the cloud server.
Driver identification device based on user behavior, including：
At least one processor；
At least one processor, for storing multiple instruction；
The multiple instruction is loaded by least one processor and realizes the driver based on user behavior
Recognition methods.
Driver identification system based on user behavior, including car-mounted terminal and cloud server, the car-mounted terminal with
Cloud server connects；
The car-mounted terminal is used for：Collection vehicle user data within a preset period of time is simultaneously sent to cloud server；
The cloud server is used for：
Data cleansing is carried out to user data；
Using word2vec models, user data is packaged into a feature vector；
This feature vector is input to the corresponding driver's identification model of the vehicle, identification obtains corresponding driver's identification
As a result.
Further, the car-mounted terminal is additionally operable to collection vehicle historical use data whithin a period of time and is sent to cloud
Hold server；
The cloud server includes model training module, and the model training module is used to correspond to for the training of each vehicle
Driver's identification model, especially by following steps train obtain：
Obtain the historical use data of vehicle whithin a period of time；
Classification and data cleansing are carried out to historical use data；
Using word2vec models, historical use data is packaged into multidimensional characteristic vectors collection；
Using multidimensional characteristic vectors collection as input data, it is input to after deep neural network is trained, it will be trained
Deep neural network is as driver's identification model；
The deep neural network is the neural network that multilayer connects entirely, specifically includes input layer, hidden layer and output
Layer, and output layer is binary classifier, is car owner or non-car owner for identifying driver's recognition result.
The beneficial effects of the invention are as follows：After the present invention is by vehicle user data within a preset period of time, use
User data is packaged into a feature vector by word2vec models, is trained in advance to which the feature vector of acquisition to be input to
Driver's identification model, you can it is car owner or non-car owner that identification, which obtains driver, and this method is set without increasing additional hardware
It is standby, can driver's identification directly be carried out according to the user data of vehicle, application cost is low, convenient, fast.
In addition, the present invention can also provide personalized service to car owner according to driver's recognition result, or tied in identification
It when fruit is non-car owner, sends a warning message to car owner, realizes anti-thefting monitoring.
Description of the drawings
Fig. 1 is the flow chart of driver's recognition methods based on user behavior of the present invention；
Fig. 2 is the schematic diagram classified to historical use data in specific embodiments of the present invention；
Fig. 3 is the structure diagram of the driver identification device based on user behavior of the present invention；
Fig. 4 is the structure diagram of the driver identification system based on user behavior of the present invention.
Specific implementation mode
Embodiment of the method
Referring to Fig.1, a kind of driver's recognition methods based on user behavior is present embodiments provided, is included the following steps：
The user data of S1, collection vehicle within a preset period of time；Preset time period is the user-defined period, can
To be 3 minutes, 5 minutes or 1 hour etc., it is configured according to driving habit；
S2, data cleansing is carried out to user data；This step is used to remove the noise data in user data；
S3, using word2vec models, user data is packaged into a feature vector；
S4, this feature vector is input to the corresponding driver's identification model of the vehicle, identification obtains corresponding driver
Recognition result.
After the present invention is by vehicle user data within a preset period of time, using word2vec models, by user data
It is packaged into a feature vector, to which the feature vector of acquisition is input to advance trained driver's identification model, you can know
Not Huo get driver be car owner or non-car owner, this method is without increasing additional hardware device, directly according to the number of users of vehicle
According to driver's identification is carried out, application cost is low, convenient, fast.
It is further used as preferred embodiment, for each vehicle, is trained by following steps and obtains vehicle correspondence
Driver's identification model：
S01, the historical use data of vehicle whithin a period of time is obtained；Here, can be for a period of time a week,
One month or setting any time period, as long as enough data can be got.In addition, the history obtained in this step is used
User data is that vehicle timing acquiring obtains.
S02, classification and data cleansing are carried out to historical use data；In this step, historical use data is divided
Class mainly classifies historical use data according to different user, to obtain the corresponding user data of each user.Such as
Shown in Fig. 2, in Fig. 2, after historical use data is identified, historical use data is classified as user 1, user 2 and user 3
The data of totally 3 users.To which in subsequent neural metwork training, input data is as unit of user.Each vehicle
Multiple users used may be corresponding with, this programme can ensure each with can be correctly validated per family.
S03, using word2vec models, historical use data is packaged into multidimensional characteristic vectors collection；
S04, using multidimensional characteristic vectors collection as input data, be input to after deep neural network is trained, will train
Good deep neural network is as driver's identification model；
The deep neural network is the neural network that multilayer connects entirely, specifically includes input layer, hidden layer and output
Layer, and output layer is binary classifier, is car owner or non-car owner for identifying driver's recognition result.
After historical use data is packaged into multidimensional characteristic vectors collection by word2vec models, using deep neural network
Training obtains driver's identification model, after by collection vehicle user data within a preset period of time, just accurate,
It efficiently identifies and obtains whether driver is car owner.
It is further used as preferred embodiment, the step S03 is specifically included：
S031, by historical use data sequential sequentially in time；
S032, using each drive when user data as input data, be input in word2vec models, acquisition model
The feature vector of output；
S033, according to time sequencing, multiple feature vectors that word2vec models are exported generate a multidimensional characteristic vectors
Collection.
Correspondingly, step S3 is similar to step S03, specially：Using user data as input data, it is input to
In word2vec models, the feature vector of model output is obtained.By word2vec models by user data be packaged into feature to
After amount, using the feature vector of acquisition as the input data of driver's identification model, so that it may in terms of according to driver's identification model
It calculates and obtains corresponding driver's recognition result.
Word2vec models can accurately compare the number of users of the multiple corresponding different trips to sort according to time series
According to similarity, after historical use data is packaged into multidimensional characteristic vectors collection by word2vec models, input can be facilitated
It is trained to deep neural network, reduces the calculation amount of neural network training process.User data is passed through in step S3
Word2vec model encapsulations can be identified easily and quickly by deep neural network at corresponding feature vector and obtain result.
It is further used as preferred embodiment, the user data includes vehicle setting data, user's driving behavior number
It goes on a journey the interaction data between preference data and user and vehicle according to, user.
Preferably, vehicle setting data include at least one of following data：Seat height, seat angle, steering wheel
Highly, steering wheel reach, the upper and lower of left and right rearview mirror, left and right angle, air-conditioner temperature, air quantity, wind-force, blowing pattern are driven
The person's of sailing weight, system sound volume size.Vehicle is arranged data and is obtained by the onboard system automatic collection of automobile, because user's sets
It is relatively-stationary to set preference, and height of seat, front and back position, door mirror angle etc. all do not need each run and is adjusted,
If these more projects of fixed setting of certain stroke are all adjusted or certain projects, such as height of seat, adjustment
Amplitude is larger, it is likely that thinks to be different driver, therefore, data are arranged in vehicle can be as driver's identification model
Input data source.
Preferably, user's driving behavior data include at least one of following data：It is anxious to accelerate frequency, frequency of bringing to a halt,
Zig zag speed, zig zag amplitude, the average speed of different stage road, honk per stroke number, frequency, per stroke distance light
Lamp access times, frequency.The data of user's driving behavior data medium velocity etc are directly acquired by sensor and are obtained, Qi Tacan
Number is to combine sensor gathered data to calculate to obtain.For example, the anxious calculation for accelerating frequency, can be according to vehicle speed sensor
After the speed of acquisition calculates acquisition acceleration, the derivative of acceleration is further calculated, change rate is remembered more than the case where setting value
Record is used as by the anxious acceleration sum calculated in setting time suddenly to accelerate once and suddenly accelerates frequency.In addition, different stage road
Average speed calculated mainly in combination with running car road, road can also be carried out according to the history running data of automobile
It distinguishes, to calculate the average speed for distinguishing the different roads obtained.And per stroke high beam access times then directly by with
Automobile CAN-bus obtains after being communicated.The speed of same driver, driving habit, such as driving takes a sudden turn, brings to a halt,
The custom suddenly accelerated is also metastable, if certain stroke, the driving behavior of driver is shown and previous driving data
There are relatively large deviation, then current driver's are likely to be different drivers.Therefore, arbitrarily it can select or calculate as needed
As the input data source of driver's identification model after user's driving behavior data.
Preferably, user's trip preference data includes at least one of following data：Travel time section and it is corresponding
It often walks route, often go to destination, trip purpose ground type, daily trip number, number of going together.User is in identical set out
Between, identical place etc. under the same conditions, be trip purpose relatively-stationary.If the unusual travel behaviour of certain stroke trip,
Then current driver's are likely to be different drivers.Therefore user's trip preference data can be used for driver's identification.
Preferably, the interaction data between user and vehicle includes at least one of following data：Data are listened in radio station,
Music data, vehicle control data.Specifically, radio station listen to data specifically include listen to the period, listen to channel, listen in
Hold, music data specifically include types of songs, singer, song, and vehicle control data specifically includes the vehicle manually controlled
Function, the vehicle functions of voice control, airconditioning control data etc..Station channel that same user often listens to, type, broadcasting
Types of songs, the singer liked, and manually or the control function of voice operating vehicle and relatively-stationary, if certain row
There is relatively large deviation in Cheng Zhong, the information interacted with vehicle, then current driver's are likely to be different drivers.Therefore, Yong Huyu
Interaction data between vehicle can also accurately reflect driver conditions.
Comprehensive vehicle setting data, user's driving behavior data, user go on a journey between preference data and user and vehicle
Four kinds of user data of interaction data, driver's identification can be accurately carried out, if four kinds of all different degrees of appearance of data are inclined
Sign from previous behavior, then it may not be car owner that current driver's, which have larger,.
It is further used as preferred embodiment, this method is further comprising the steps of：
The case where for driver's recognition result being car owner, it is switched to car owner's Personalized Service Model, is provided for car owner a
Property service；
The case where for driver's recognition result being non-car owner, corresponding warning message is sent to car owner.
The case where this step can be car owner or non-car owner according to recognition result, it is corresponding to execute different functions, for vehicle
It is main, it is switched to car owner's Personalized Service Model, is provided personalized service for car owner, specific car owner's Personalized Service Model can be with
It is that driver is set in advance, can also be to be carried out automating training acquisition using data according to driver.When identification is tied
Fruit, without car owner again into manual operation, improves drive safety after car owner, directly to provide personalized service for car owner.Separately
Outside, be non-car owner for recognition result the case where, corresponding warning message directly is sent to car owner, the form of warning message can
To be that the information such as short message or wechat push window, the abnormal conditions of host vehicle are reminded in time, realize anti-thefting monitoring.Further
, while sending corresponding warning information to car owner, the real-time position information of vehicle can also be sent simultaneously, convenient for vehicle
Carry out location tracking.
It is further used as preferred embodiment, the output layer is built using softmax functions or logistic regression function
Binary classifier.Binary classifier is built by softmax functions or logistic regression function, can accurately train and obtain driver
Identification model.And it when driver's identification model of structure needs to identify the information except car owner/non-car owner, may be used
Softmax functions build output layer.
It is further used as preferred embodiment, driver's identification model is trained in a cloud server and is obtained, and
Driver's recognition methods executes identification step in the cloud server.Specifically, i.e. the step S1 of this method is in vehicle end
It executes, server executes step S2~S4 beyond the clouds, and this programme carries out driver's identification in this way, it is only necessary to which acquisition is pre-
If the user data in the time is sent to cloud server and is identified, without additionally installing other hardware devices, apply
It is at low cost and convenient, fast.When needing more fresh driver's identification model, it is only necessary to which server is updated i.e. beyond the clouds
Can, it is simple, efficient without being modified to each vehicle.
Device embodiment
With reference to Fig. 3, a kind of driver identification device based on user behavior is present embodiments provided, including：
At least one processor 100；
At least one processor 200, for storing multiple instruction；
The multiple instruction is loaded by least one processor 100 and realizes the driving based on user behavior
Member's recognition methods.
The driver identification device based on user behavior of the present embodiment, what executable the method for the present invention embodiment was provided
Driver's recognition methods based on user behavior, the arbitrary combination implementation steps of executing method embodiment, has this method phase
The function and advantageous effect answered.
System embodiment
With reference to Fig. 4, present embodiments provide a kind of driver identification system based on user behavior, including car-mounted terminal and
Cloud server, the car-mounted terminal are connect with cloud server；
The car-mounted terminal is used for：Collection vehicle user data within a preset period of time is simultaneously sent to cloud server；
The cloud server is used for：
Data cleansing is carried out to user data；
Using word2vec models, user data is packaged into a feature vector；
This feature vector is input to the corresponding driver's identification model of the vehicle, identification obtains corresponding driver's identification
As a result.
It is further used as preferred embodiment, the car-mounted terminal is additionally operable to the history of collection vehicle whithin a period of time
User data is simultaneously sent to cloud server；
The cloud server includes model training module, and the model training module is used to correspond to for the training of each vehicle
Driver's identification model, especially by following steps train obtain：
Obtain the historical use data of vehicle whithin a period of time；
Classification and data cleansing are carried out to historical use data；
Using word2vec models, historical use data is packaged into multidimensional characteristic vectors collection；
Using multidimensional characteristic vectors collection as input data, it is input to after deep neural network is trained, it will be trained
Deep neural network is as driver's identification model；
The deep neural network is the neural network that multilayer connects entirely, specifically includes input layer, hidden layer and output
Layer, and output layer is binary classifier, is car owner or non-car owner for identifying driver's recognition result.
The driver identification system based on user behavior of the present embodiment, what executable the method for the present invention embodiment was provided
Driver's recognition methods based on user behavior, the arbitrary combination implementation steps of executing method embodiment, has this method phase
The function and advantageous effect answered.
It is to be illustrated to the preferable implementation of the present invention, but the invention is not limited to the implementation above
Example, those skilled in the art can also make various equivalent variations or be replaced under the premise of without prejudice to spirit of that invention
It changes, these equivalent modifications or replacement are all contained in the application claim limited range.
Claims (10)
1. driver's recognition methods based on user behavior, which is characterized in that include the following steps：
The user data of collection vehicle within a preset period of time；
Data cleansing is carried out to user data；
Using word2vec models, user data is packaged into a feature vector；
This feature vector is input to the corresponding driver's identification model of the vehicle, identification obtains corresponding driver and identifies knot
Fruit.
2. driver's recognition methods according to claim 1 based on user behavior, which is characterized in that be directed to each car
, it is trained by following steps and obtains the corresponding driver's identification model of vehicle：
Obtain the historical use data of vehicle whithin a period of time；
Classification and data cleansing are carried out to historical use data；
Using word2vec models, historical use data is packaged into multidimensional characteristic vectors collection；
Using multidimensional characteristic vectors collection as input data, it is input to after deep neural network is trained, by trained depth
Neural network is as driver's identification model；
The deep neural network is the neural network that multilayer connects entirely, specifically includes input layer, hidden layer and output layer, and
Output layer is binary classifier, is car owner or non-car owner for identifying driver's recognition result.
3. driver's recognition methods according to claim 2 based on user behavior, which is characterized in that the use
Word2vec models, specifically include the step of historical use data is packaged into multidimensional characteristic vectors collection：
By historical use data sequential sequentially in time；
User data when using each driving is input in word2vec models as input data, obtains the spy of model output
Sign vector；
According to time sequencing, multiple feature vectors that word2vec models are exported generate a multidimensional characteristic vectors collection.
4. driver's recognition methods according to claim 1 based on user behavior, which is characterized in that the user data
The interaction number gone on a journey between preference data and user and vehicle including vehicle setting data, user's driving behavior data, user
According to.
5. driver's recognition methods according to claim 1 based on user behavior, which is characterized in that further include following step
Suddenly：
For driver's recognition result be car owner the case where, be switched to car owner's Personalized Service Model, personalization provided for car owner
Service；
The case where for driver's recognition result being non-car owner, corresponding warning message is sent to car owner.
6. driver's recognition methods according to claim 2 based on user behavior, which is characterized in that the output layer is adopted
Binary classifier is built with softmax functions or logistic regression function.
7. driver's recognition methods according to claim 1 based on user behavior, which is characterized in that the driver knows
Other model is trained in a cloud server and is obtained, and driver's recognition methods executes identification step in the cloud server.
8. the driver identification device based on user behavior, which is characterized in that including：
At least one processor；
At least one processor, for storing multiple instruction；
The multiple instruction is loaded by least one processor and is realized as claimed in any one of claims 1 to 6 based on use
Driver's recognition methods of family behavior.
9. the driver identification system based on user behavior, which is characterized in that including car-mounted terminal and cloud server, the vehicle
Mounted terminal is connect with cloud server；
The car-mounted terminal is used for：Collection vehicle user data within a preset period of time is simultaneously sent to cloud server；
The cloud server is used for：
Data cleansing is carried out to user data；
Using word2vec models, user data is packaged into a feature vector；
This feature vector is input to the corresponding driver's identification model of the vehicle, identification obtains corresponding driver and identifies knot
Fruit.
10. the driver identification system according to claim 9 based on user behavior, which is characterized in that the vehicle-mounted end
End is additionally operable to collection vehicle historical use data whithin a period of time and is sent to cloud server；
The cloud server includes model training module, and the model training module is used to drive for the training of each vehicle is corresponding
The person's of sailing identification model is trained especially by following steps and is obtained：
Obtain the historical use data of vehicle whithin a period of time；
Classification and data cleansing are carried out to historical use data；
Using word2vec models, historical use data is packaged into multidimensional characteristic vectors collection；
Using multidimensional characteristic vectors collection as input data, it is input to after deep neural network is trained, by trained depth
Neural network is as driver's identification model；
The deep neural network is the neural network that multilayer connects entirely, specifically includes input layer, hidden layer and output layer, and
Output layer is binary classifier, is car owner or non-car owner for identifying driver's recognition result.
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 true CN108280482A (en) | 2018-07-13 |
| CN108280482B 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) |
Cited By (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 |
| CN110096499A (en) * | 2019-04-10 | 2019-08-06 | 华南理工大学 | A kind of the user object recognition methods and system of Behavior-based control time series big data |
| 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 |
| 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 |
| US20210027178A1 (en) * | 2019-07-26 | 2021-01-28 | Ricoh Company, Ltd. | Recommendation method and recommendation apparatus based on deep reinforcement learning, and non-transitory computer-readable recording medium |
Citations (8)
| 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 |
| CN105808737A (en) * | 2016-03-10 | 2016-07-27 | 腾讯科技（深圳）有限公司 | Information retrieval method and server |
| 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 |
| US20170330363A1 (en) * | 2016-05-13 | 2017-11-16 | Yahoo Holdings Inc. | Automatic video segment selection method and apparatus |
-
2018
- 2018-01-30 CN CN201810086755.6A patent/CN108280482B/en active Active
Patent Citations (8)
| 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 |
| CN105808737A (en) * | 2016-03-10 | 2016-07-27 | 腾讯科技（深圳）有限公司 | Information retrieval method and server |
| CN105761329A (en) * | 2016-03-16 | 2016-07-13 | 成都信息工程大学 | Method of identifying driver based on driving habits |
| US20170330363A1 (en) * | 2016-05-13 | 2017-11-16 | Yahoo Holdings Inc. | Automatic video segment selection method and apparatus |
| 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 |
Cited By (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 |
| CN110096499A (en) * | 2019-04-10 | 2019-08-06 | 华南理工大学 | A kind of the user object recognition methods and system of Behavior-based control time series big data |
| US20210027178A1 (en) * | 2019-07-26 | 2021-01-28 | Ricoh Company, Ltd. | Recommendation method and recommendation apparatus based on deep reinforcement learning, and non-transitory computer-readable recording medium |
| 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 |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108280482B (en) | 2020-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108280482A (en) | Driver's recognition methods based on user behavior, apparatus and system | |
| US11685386B2 (en) | System and method for determining a change of a customary vehicle driver | |
| US10474151B2 (en) | Method for guiding a vehicle system in a fully automated manner, and motor vehicle | |
| CN108995655B (en) | Method and system for identifying driving intention of driver | |
| CN106128099B (en) | Driver's recognition methods and device | |
| US10540557B2 (en) | Method and apparatus for providing driver information via audio and video metadata extraction | |
| US11494685B2 (en) | Learning system, in-vehicle device, and server | |
| US9731713B2 (en) | Modifying autonomous vehicle driving by recognizing vehicle characteristics | |
| CN107600072A (en) | A kind of acquisition methods and system of the common preference parameter of more passengers | |
| US20170313323A1 (en) | Vehicle mode scheduling with learned user preferences | |
| CN111002982B (en) | Apparatus and method for controlling speed | |
| CN104391504A (en) | Vehicle networking based automatic driving control strategy generation method and device | |
| CN104340144A (en) | Multi-vehicle settings | |
| CN109491284A (en) | Control method for vehicle, device and terminal device based on user's trip habit | |
| KR102209421B1 (en) | Autonomous vehicle and driving control system and method using the same | |
| JP2014081947A (en) | Information distribution device | |
| CN109895696A (en) | Operator alert system and method | |
| CN113173170B (en) | Personalized algorithm based on personnel portrait | |
| CN112389451A (en) | Method, device, medium, and vehicle for providing a personalized driving experience | |
| CN105564429A (en) | Running safety pre-warning method and device | |
| CN109515441B (en) | Vehicle speed control system for intelligent driving vehicle | |
| CN110509928B (en) | Driving assisting method and device | |
| CN109466479B (en) | Vehicle control method, device, terminal equipment and medium | |
| CN115203536A (en) | Method and device for recommending intelligent driving parameters based on driving scene | |
| CN113501004B (en) | Control method and device based on gestures, electronic equipment and storage medium |
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 | ||
| 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. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |