CN109229108B - Driving behavior safety evaluation method based on driving fingerprints - Google Patents
Driving behavior safety evaluation method based on driving fingerprints Download PDFInfo
- Publication number
- CN109229108B CN109229108B CN201810890209.8A CN201810890209A CN109229108B CN 109229108 B CN109229108 B CN 109229108B CN 201810890209 A CN201810890209 A CN 201810890209A CN 109229108 B CN109229108 B CN 109229108B
- Authority
- CN
- China
- Prior art keywords
- driving
- data
- driving behavior
- fingerprint
- fingerprints
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
- B60W40/09—Driving style or behaviour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a driving behavior safety evaluation method based on driving fingerprints, which comprises the following steps: s1, acquiring driving behavior data through a sensor arranged on the vehicle; s2, calculating characteristic indexes representing individual differentiated driving behaviors according to the driving behavior data; s3, selecting a training set, inputting the characteristic indexes into a machine learning classifier for training, and calculating through the machine learning classifier to obtain a driving fingerprint; s4, taking the driving behavior data to be evaluated as a test set, and comparing the difference degree between the driving behavior data and the driving fingerprint in the normal driving state; s5, calculating importance weight of each parameter in the driving fingerprint to driving safety; and S6, calculating the parameters of the difference degree through a principal component analysis method, and comprehensively evaluating the safety of the driving behavior through the obtained factor score. The method can fully consider the influence of individual difference factors and obtain accurate, comprehensive and reliable driver safety evaluation results.
Description
Technical Field
The invention relates to the field of driving safety evaluation, in particular to a driving behavior safety evaluation method based on driving fingerprints.
Background
In recent years, with the development of vehicle-mounted sensor technology, research in the field of traffic safety is deepened gradually, and many automobile manufacturers are actively developing personalized vehicle auxiliary driving systems to improve the reliability and adaptability of the whole system. The basis of such systems is the identification and evaluation of the individual driving characteristics of the driver. Meanwhile, through research on individual differences of driving behaviors in recent years, compared with the traditional individual difference-free model, the individual difference-free model of the personalized driving behavior model (a driving fatigue detection model, a lane keeping model, a vehicle speed control prediction model and the like) which is constructed by reasonably considering the individual difference factors is higher in accuracy and more adaptable to the individual characteristics of the driver.
However, the existing driving safety evaluation method and system do not consider the influence of individual difference factors, and the individual method or system only classifies the drivers into different driving styles in a classification mode. However, due to the existence of individual differences, there are significant differences in driving characteristics even among drivers belonging to the same driving style.
Hearing the 'driving fingerprint' we will first think of the fingerprint. Biologically, each person has a unique fingerprint, and the unique identity of each person is marked by using different 'lines'. Similarly, from the driving behavior perspective, the driving fingerprint refers to the unique driving operation characteristic that each person shows in the driving process, and the characterization index of each driving fingerprint is the unique driving fingerprint line of the driver.
In the application of the driving fingerprints at the present stage, the identity of a driver is identified by utilizing a classifier algorithm through comparing differences of driving fingerprint parameters of different drivers. And the method is not applied to the driving behavior safety evaluation by using the driving fingerprint.
The invention comprehensively considers the influence of individual difference factors of the driver on the evaluation of the driver, and creatively evaluates the driving behavior from the perspective of individual driving state, thereby greatly improving the reliability and accuracy of the evaluation result. Meanwhile, according to the driving fingerprint characteristics of the driver obtained by machine learning, the identity of the driver can be accurately identified and the current driving state can be comprehensively mastered, so that a safety strategy can be provided in a targeted manner.
Disclosure of Invention
The invention aims to solve the technical problems that an existing driving safety evaluation system in the internet of vehicles has low accuracy and poor reliability of evaluation results due to individual difference factors of drivers, and the evaluation results are not convincing and fair due to the fact that a traditional evaluation index method cannot adapt to individual driving habit characteristics of each driver to be evaluated. The invention provides a driver safety evaluation method based on individual driving fingerprint characteristics of a driver, so as to solve the problems.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the invention provides a driving behavior safety evaluation method based on driving fingerprints, which comprises the following steps:
s1, collecting driving behavior data including driving operation data and vehicle running state data through a sensor arranged on the vehicle;
s2, calculating characteristic indexes representing individual differentiated driving behaviors according to the driving behavior data, wherein the characteristic indexes comprise statistical characteristics, morphological characteristics and frequency characteristics;
s3, selecting continuous historical driving behavior data with a certain duration as a training set, inputting the characteristic indexes into a machine learning classifier for training, screening out driving style mutation data through the machine learning classifier, and calculating the distribution characteristics and the variation characteristics of the characteristic indexes of a driver in a normal driving state through a statistical method to serve as driving fingerprints, wherein the driving fingerprints comprise a plurality of parameters;
s4, taking the driving behavior data to be evaluated as a test set, comparing the difference degree between the driving behavior data in the test set and the driving fingerprint of the driving behavior data of the driver in the normal driving state, and representing the difference degree by utilizing a quantized difference index;
s5, calculating importance weight of each parameter in the driving fingerprint obtained in the step S3 on driving safety by a sensitivity analysis or importance analysis method;
and S6, calculating the parameters of the difference degree through a principal component analysis method by combining the difference degree between the driving fingerprints obtained in the step S4 and the importance degree weight of each parameter of the driving fingerprints obtained in the step S5, and comprehensively evaluating the safety of the driving behaviors through the obtained factor scores.
Further, step S2 of the present invention further includes a method for preprocessing driving behavior data:
the preprocessing method comprises the steps of unifying sampling frequency and wavelet noise reduction, and removing interference caused by short-time driving style change of a driver to the process of learning the normal driving state of the driver.
Further, step S2 of the present invention further includes: and screening the obtained characteristic indexes by a questionnaire survey method, and reserving the characteristic indexes with high correlation degree with the driving safety.
Further, the driving fingerprint in step S3 of the present invention includes a plurality of parameters:
the statistical features include: minimum, maximum, median, kurtosis, skewness;
the morphological characteristics comprise: data centralization trend, departure trend and distribution form;
the central tendency analysis comprises: statistical indexes of mean, mean and mode, which represent the centralized trend of the data;
the analysis of the deviation trend includes: the statistical indexes of the total distance, the quarter difference, the average difference, the variance and the standard deviation are used for researching the dispersion degree of the data;
four differential Qd: is the average of the differences between the upper quartile QU and the lower quartile QL, and the calculation formula is:
Qd＝(QU-QL)/2
the quartile difference reflects the discrete degree of the middle 50% of data, and the smaller the numerical value is, the more concentrated the middle data is; the larger the value, the more dispersed the data in the middle.
Further, the machine learning classifier in step S3 of the present invention includes an artificial neural network, a support vector machine and a random forest.
Further, the method for calculating the characteristic index representing the individual differentiated driving behavior in step S2 of the present invention specifically includes:
collecting historical driving behavior data, including driving operation data and vehicle running state data; unifying sampling frequency of each parameter by utilizing a secondary interpolation or extraction method, and filtering data by wavelet noise reduction; morphological characteristics and statistical characteristics of all parameters are obtained through calculation and are used as individual differentiated driving behavior characterization indexes.
The comprehensive evaluation result obtained in step S6 includes: the driving safety degree of the driver is judged visually through different grades according to different grades.
The invention has the following beneficial effects: the driving behavior safety evaluation method based on the driving fingerprints is based on the driving control parameters acquired by the vehicle-mounted sensor and the vehicle running state parameters, the vehicle-mounted sensor is integrated in the vehicle, so that modification is not needed, meanwhile, the driving fingerprints of the normal driving state of the driver can be continuously corrected based on historical driving data, the safety evaluation process cannot influence driving, and the driving behavior safety evaluation method has high driver adaptability.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a general flow diagram of the present invention;
FIG. 2 is a driving data preprocessing schematic of the present invention;
FIG. 3 is a driving fingerprint training schematic of the present invention;
fig. 4 is a driving behavior safety assessment schematic diagram based on driving fingerprints according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The specific implementation mode of the invention provides a driving safety evaluation method based on driving fingerprints, and as shown in fig. 1, a general flow chart of the invention is provided. The main contents of the method are concentrated in three parts: the method comprises the steps of driving data preprocessing, a driving fingerprint training process and a driving behavior safety evaluation process based on the driving fingerprint.
1. Driving data preprocessing process
The principle steps of the part are as shown in fig. 2, firstly, continuous normal driving data with a period of time is selected from data collected by a vehicle-mounted sensor to be used as a training set and is imported into MATLAB data analysis software; then unifying the sampling frequency of various different parameters by utilizing a secondary difference value or extraction method; wavelet denoising is then performed to ensure the usability of data while filtering the influence caused by environmental factors or changes in driving style in a short time.
2. Training process for driving fingerprints
The principle steps of the part are shown in figure 3.
1) Calculating driving fingerprint index
And respectively calculating the statistical characteristics, morphological characteristics and the like of each data parameter as driving fingerprint indexes based on the data acquired by the work, thereby obtaining a driving fingerprint system capable of representing the driving behavior characteristics of the driver.
The statistical features include: minimum, maximum, median, kurtosis, skewness, and the like.
The morphological characteristics comprise: trends in data sets, trends in deviations, distribution patterns, etc.
The central tendency analysis can use statistical indexes such as average, mean and mode to represent the central tendency (positive bias distribution, negative bias distribution, etc.) of the data;
the dispersion degree of data is researched mainly by statistical indexes such as full range, quarter difference, mean difference, variance and standard deviation in the analysis of the trend in the distance.
Four differential (Q)d): it is the average of the differences between the upper quartile (QU, i.e. at 75%) and the lower quartile (QL, i.e. at 25%). The calculation formula is as follows:
Qd＝(QU-QL)/2
the quartile difference reflects the discrete degree of the middle 50% of data, and the smaller the numerical value is, the more concentrated the middle data is; the larger the value, the more dispersed the data in the middle. The quartering potential difference is not affected by the extrema.
Because the driving fingerprint characterization index system has more parameters, the invention aims to research the relation between the driving fingerprint and the driving safety, so the invention screens the index system by adopting questionnaire survey and other modes according to the degree of correlation (sensitivity analysis or importance analysis) between the fingerprint index and the driving safety to finally obtain a driving fingerprint index set facing the driving safety and uses the driving fingerprint index set as a training set for the next step of individual driver driving fingerprint training.
Sensitivity analysis principle formula:
sensitivity is 100% of true positive people/(true positive people + false negative people)
The importance analysis is essentially a plurality of calculation modes of the sensitivity of the multivariate function, and the following formula is taken as an example:
wherein, PiRepresenting the reliability of the object I, IiRepresenting importance of object i, system reliabilityFunction R (P)1，P2，…，Pn) Is about a parameter P1，P2，…，PnN-ary function of (1).
2) Calculating individual driving fingerprint characteristics of driver to be evaluated
After the driving fingerprint index system is built, training is carried out through a supervised machine learning algorithm (such as a support vector machine) based on historical driving data, and therefore driving style mutation data caused by environmental reasons, emergencies and the like are screened out by utilizing a trained classifier. Therefore, the distribution rules and the variation characteristics of different driving fingerprint parameters of the driver in the normal driving process are obtained. Namely, the individual driving fingerprint of the driver is obtained through machine learning training. The obtained driving fingerprint is the driving behavior expression characteristic of the driver in the normal driving state.
3. Driving behavior safety assessment process based on driving fingerprints
The principle steps of this section are shown in fig. 4.
The driving behavior data to be evaluated of the driver in a period of time is selected, the driving fingerprint characteristics to be tested of the current driver are obtained through calculation according to the principle, the driving fingerprint characteristics are compared with the driving fingerprint characteristics obtained by the first part in the normal driving state of the driver, and the difference degree of each driving fingerprint index between the driving fingerprint characteristics and the driving fingerprint characteristics is quantified by using certain parameters (variation coefficient, standard deviation, average difference and the like). And finally, parameters representing the difference degree of each driving fingerprint index are comprehensively represented, and the driving behavior safety level of the test object is evaluated by using a principal component analysis method.
In order to make the invention more clear, a preferred implementation procedure of the invention is now described as follows:
the method comprises the steps of firstly collecting historical driving data such as driving control based on a vehicle-mounted sensor, vehicle motion state and the like, then unifying sampling frequency of each parameter by utilizing a secondary interpolation or extraction method, and filtering the data by wavelet noise reduction. And finally, acquiring morphological characteristics and statistical characteristics of all parameters through calculation to serve as individual differentiated driving behavior characterization indexes.
The specific content of the quadratic interpolation method is as follows:
linear interpolation is often used in methods where the value of the function f at two points is known to be approximated to obtain the value of the other point, where p represents a linear interpolation polynomial:
wherein: (x) a variation function representing data; x is the number of0Representing the interpolated previous data point; x is the number of1Representing the interpolated data point.
And taking the indexes obtained by the preprocessing as a training set for training a classifier, wherein the classifier which is trained (the classifier can be realized by utilizing a scinit-lean machine learning library of Python) is used for screening out the driving style mutation data caused by environmental reasons, emergencies and the like. And then, obtaining the change characteristics (such as an off-center trend, a centralized trend, a distribution form and the like) of each characterization index in the normal driving state by utilizing statistical analysis based on the classified data, and taking the change characteristics as the characteristics of the researched driver in the normal driving state, namely the driving fingerprint.
And taking the data to be evaluated as a test set, comparing the data with the driving fingerprint of the driver in the normal driving state obtained by the calculation, and quantifying the difference degree of the data and the driving fingerprint through parameters such as a variation coefficient, a standard deviation, an average value and the like.
And finally, calculating the parameters representing the difference degree by using a principal component analysis method, and carrying out comprehensive evaluation according to the factor score, wherein the principal component analysis method can be realized by using SPSS data analysis software and MATLAB programming.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (7)
1. A driving behavior safety evaluation method based on driving fingerprints is characterized by comprising the following steps:
s1, collecting driving behavior data including driving operation data and vehicle running state data through a sensor arranged on the vehicle;
s2, calculating characteristic indexes representing individual differentiated driving behaviors according to the driving behavior data, wherein the characteristic indexes comprise statistical characteristics, morphological characteristics and frequency characteristics;
s3, selecting continuous historical driving behavior data with a certain duration as a training set, inputting the characteristic indexes into a machine learning classifier for training, screening out driving style mutation data through the machine learning classifier, and calculating the distribution characteristics and the variation characteristics of the characteristic indexes of a driver in a normal driving state through a statistical method to serve as driving fingerprints, wherein the driving fingerprints comprise a plurality of parameters;
s4, taking the driving behavior data to be evaluated as a test set, comparing the difference degree between the driving behavior data in the test set and the driving fingerprint of the driving behavior data of the driver in the normal driving state, and representing the difference degree by utilizing a quantized difference index;
s5, calculating importance weight of each parameter in the driving fingerprint obtained in the step S3 on driving safety by a sensitivity analysis or importance analysis method;
and S6, calculating the parameters of the difference degree through a principal component analysis method by combining the difference degree between the driving fingerprints obtained in the step S4 and the importance degree weight of each parameter of the driving fingerprints obtained in the step S5, and comprehensively evaluating the safety of the driving behaviors through the obtained factor scores.
2. The driving behavior safety evaluation method based on the driving fingerprint as claimed in claim 1, wherein step S2 further comprises a method for preprocessing the driving behavior data:
the preprocessing method comprises the steps of unifying sampling frequency and wavelet noise reduction, and removing interference caused by short-time driving style change of a driver to the process of learning the normal driving state of the driver.
3. The driving behavior safety evaluation method based on the driving fingerprint according to claim 1, wherein step S2 further comprises: and screening the obtained characteristic indexes by a questionnaire survey method, and reserving the characteristic indexes with high correlation degree with the driving safety.
4. The driving behavior safety evaluation method based on driving fingerprints according to claim 1, wherein the driving fingerprint in step S3 includes a plurality of parameters:
the statistical features include: minimum, maximum, median, kurtosis, skewness;
the morphological characteristics comprise: data centralization trend, departure trend and distribution form;
the central tendency analysis comprises: statistical indexes of mean, mean and mode, which represent the centralized trend of the data;
the analysis of the deviation trend includes: the statistical indexes of the total distance, the quarter difference, the average difference, the variance and the standard deviation are used for researching the dispersion degree of the data;
four differential Qd: is the average of the differences between the upper quartile QU and the lower quartile QL, and the calculation formula is:
Qd＝(QU-QL)/2
the quartile difference reflects the discrete degree of the middle 50% of data, and the smaller the numerical value is, the more concentrated the middle data is; the larger the value, the more dispersed the data in the middle.
5. The driving behavior safety evaluation method based on the driving fingerprint as claimed in claim 1, wherein the machine learning classifier in step S3 comprises an artificial neural network, a support vector machine and a random forest.
6. The driving behavior safety evaluation method based on the driving fingerprint according to claim 1, wherein the method for calculating the characteristic index representing the individual differentiated driving behavior in step S2 specifically comprises:
collecting historical driving behavior data, including driving operation data and vehicle running state data; unifying sampling frequency of each parameter by utilizing a secondary interpolation or extraction method, and filtering data by wavelet noise reduction; morphological characteristics and statistical characteristics of all parameters are obtained through calculation and are used as individual differentiated driving behavior characterization indexes.
7. The driving fingerprint-based driving behavior safety evaluation method according to claim 1, wherein the comprehensive evaluation result obtained in step S6 includes: and grading according to different thresholds according to the obtained factors, and visually judging the driving safety degree of the driver through different grades.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810890209.8A CN109229108B (en) | 2018-08-07 | 2018-08-07 | Driving behavior safety evaluation method based on driving fingerprints |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810890209.8A CN109229108B (en) | 2018-08-07 | 2018-08-07 | Driving behavior safety evaluation method based on driving fingerprints |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109229108A CN109229108A (en) | 2019-01-18 |
| CN109229108B true CN109229108B (en) | 2020-05-05 |
Family
ID=65070479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810890209.8A Active CN109229108B (en) | 2018-08-07 | 2018-08-07 | Driving behavior safety evaluation method based on driving fingerprints |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109229108B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598367B (en) * | 2019-02-20 | 2023-04-07 | 杭州海康威视数字技术股份有限公司 | Driving behavior evaluation method and device, computing equipment and storage medium |
| CN110623680A (en) * | 2019-08-30 | 2019-12-31 | 中国民用航空飞行学院 | Method for testing psychological health of civil aviation flight trainees |
| CN110533261A (en) * | 2019-09-06 | 2019-12-03 | 江苏物联网研究发展中心 | A kind of driving behavior analysis optimization system of transport truck |
| CN111038485B (en) * | 2019-12-30 | 2021-02-19 | 山东大学 | Hybrid electric vehicle control method and system based on driving style recognition |
| CN112651443A (en) * | 2020-12-28 | 2021-04-13 | 华北科技学院 | Driving style identification model evaluation method, device, medium and equipment based on machine learning |
| CN113641423B (en) * | 2021-08-31 | 2023-07-07 | 青岛海信传媒网络技术有限公司 | Display device and system starting method |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007122362A (en) * | 2005-10-27 | 2007-05-17 | Toyota Motor Corp | State estimation method using neural network and state estimation apparatus using neural network |
| CN104527647B (en) * | 2014-12-15 | 2017-01-25 | 清华大学 | Monitoring and evaluation method of driving behavior risk degrees |
| CN105730450B (en) * | 2016-01-29 | 2018-01-02 | 北京荣之联科技股份有限公司 | Driving behavior analysis method and evaluation system based on vehicle-mounted data |
| CN106218641B (en) * | 2016-08-04 | 2018-10-09 | 武汉理工大学 | A kind of vehicle new hand driver automatic identification and safe early warning method |
| JP6720785B2 (en) * | 2016-09-05 | 2020-07-08 | 日産自動車株式会社 | Driving support method and driving support device |
| CN106585635B (en) * | 2016-11-25 | 2019-05-10 | 东软集团股份有限公司 | Driving behavior methods of marking and device |
| KR20180068173A (en) * | 2016-12-13 | 2018-06-21 | 주식회사 티노스 | Safety driving supporting method based on driving style information |
| DE102017200149B4 (en) * | 2017-01-09 | 2022-06-15 | Ford Motor Company | Method for learning and checking driving behavior of a vehicle driver and vehicle |
| CN108108766B (en) * | 2017-12-28 | 2021-10-29 | 东南大学 | Driving behavior identification method and system based on multi-sensor data fusion |
| CN108216252B (en) * | 2017-12-29 | 2019-12-20 | 中车工业研究院有限公司 | Subway driver vehicle-mounted driving behavior analysis method, vehicle-mounted terminal and system |
-
2018
- 2018-08-07 CN CN201810890209.8A patent/CN109229108B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109229108A (en) | 2019-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109229108B (en) | Driving behavior safety evaluation method based on driving fingerprints | |
| US7444311B2 (en) | System and method for real-time recognition of driving patterns | |
| CN109977808A (en) | A kind of wafer surface defects mode detection and analysis method | |
| Das et al. | Differentiating alcohol-induced driving behavior using steering wheel signals | |
| Bravo-Moncayo et al. | A machine learning approach for traffic-noise annoyance assessment | |
| CN113486822B (en) | Surrounding vehicle track prediction method and system based on driving intention | |
| EP3498559B1 (en) | Method for recognizing the driving style of a driver of a land vehicle, and corresponding apparatus | |
| Zylius | Investigation of route-independent aggressive and safe driving features obtained from accelerometer signals | |
| CN110287975B (en) | Flotation dosing abnormity detection method based on NSST morphological characteristics and depth KELM | |
| CN106250801A (en) | Based on Face datection and the fatigue detection method of human eye state identification | |
| CN111460950B (en) | Cognitive distraction method based on head-eye evidence fusion in natural driving conversation behavior | |
| CN103455820A (en) | Method and system for detecting and tracking vehicle based on machine vision technology | |
| Khademi et al. | Fuzzy rules for fuzzy $\overline {X} $ and $ R $ control charts | |
| CN111738337A (en) | Driver distraction state detection and identification method in mixed traffic environment | |
| CN106340007A (en) | Image processing-based automobile body paint film defect detection and identification method | |
| CN114926825A (en) | Vehicle driving behavior detection method based on space-time feature fusion | |
| CN114997218A (en) | Recognition and detection method for polygonal abrasion of wheels of railway vehicle | |
| Banerjee et al. | MyDrive: Drive behavior analytics method and platform | |
| CN111967308A (en) | Online road surface unevenness identification method and system | |
| CN113642114B (en) | Personified random following driving behavior modeling method capable of making mistakes | |
| US20220012546A1 (en) | Measurement of the sensitivity of classifiers based on interacting faults | |
| CN108073978A (en) | A kind of constructive method of the ultra-deep learning model of artificial intelligence | |
| CN113240034A (en) | Depth decision fusion method based on entropy method and D-S evidence theory | |
| Asaddulloh et al. | Machine Learning Techniques to Predict Rain Tomorrow for Automated Plant Watering System | |
| CN111723869A (en) | Special personnel-oriented intelligent behavior risk early warning method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |