CN111874000A - Method for judging safety level of driving behavior and storage medium - Google Patents
Method for judging safety level of driving behavior and storage medium Download PDFInfo
- Publication number
- CN111874000A CN111874000A CN202010710791.2A CN202010710791A CN111874000A CN 111874000 A CN111874000 A CN 111874000A CN 202010710791 A CN202010710791 A CN 202010710791A CN 111874000 A CN111874000 A CN 111874000A
- Authority
- CN
- China
- Prior art keywords
- driving behavior
- data
- score
- safety level
- result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
- B60W2520/105—Longitudinal acceleration
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/12—Lateral speed
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/12—Lateral speed
- B60W2520/125—Lateral acceleration
Abstract
The invention discloses a method for judging safety level of driving behavior and a storage medium, comprising the following steps: s1, acquiring driving behavior data of the target vehicle in a monitoring period; s2, normalizing the acquired driving behavior data; s3, obtaining a clustering result from the normalized data result by a clustering method; s4, training the data set by using a deep learning model for the clustering result, finally giving the probability that each piece of data is 0, and multiplying the probability by 100 to obtain the result which is the calculated safe driving behavior score; and S5, giving a safe driving behavior grade according to the safe driving behavior score. According to the invention, the advantages and disadvantages of the driving behavior data are evaluated, and meanwhile, a deep learning model is introduced, so that the training fitting degree is higher, the training effect is better, and the model result can reflect the real situation.
Description
Technical Field
The invention belongs to the technical field of driving behavior safety judgment, and particularly relates to a method for judging the safety level of driving behavior and a storage medium.
Background
The car networking realizes information sharing through interconnection and intercommunication of cars, cars and people and cars and roads, collects information of the cars, the roads and the environment, processes, calculates, shares and safely releases the information collected by multiple sources on an information network platform, effectively guides and supervises the cars according to different functional requirements, provides professional multimedia and mobile internet application services, realizes extraction and effective utilization of attribute information and static and dynamic information of all the cars on the information network platform, and effectively supervises the running states of all the cars according to different functional requirements and provides comprehensive services.
With the gradual deepening of the car networking application, the installation of an On-Board Diagnostic (OBD) device and a front-mounted integrated device On a car enables more and more vehicles to be included in the range of information full coverage. The automobile is used as a comprehensive information service terminal, on one hand, the operation data of each subsystem of the automobile is continuously sent to a remote server, and on the other hand, various information including news information, audio, video, location-based services and the like is obtained from the Internet.
Among the factors causing traffic accidents issued by domestic and foreign traffic departments, the traffic accidents caused by drivers account for over 70%, and the safety of driving behaviors is related to the safety of individuals and other people, so that intensive research on the safety of driving behaviors is necessary.
At present, when the driving behavior is subjected to safety rating, insurance data and the like are mostly adopted as the lab values of training, and the insurance data cannot define the quality of the driving behavior due to other environmental and working condition factors. Meanwhile, because the risk data is less than the risk data which is not present, the imbalance of the training samples is easily caused, and the input result of the model cannot reflect the real situation.
Therefore, it is necessary to develop a method and a storage medium for determining a safety level of driving behavior.
Disclosure of Invention
The invention aims to provide a method and a storage medium for judging the safety level of driving behaviors, which can reliably reflect the real driving situation.
The invention relates to a method for judging the safety level of driving behaviors, which comprises the following steps:
s1, acquiring driving behavior data of the target vehicle in a monitoring period;
s2, normalizing the acquired driving behavior data;
s3, obtaining a clustering result from the normalized data result by a clustering method;
s4, training the data set by using a deep learning model for the clustering result, finally giving the probability that each piece of data is 0, and multiplying the probability by 100 to obtain the result which is the calculated safe driving behavior score;
and S5, giving a safe driving behavior grade according to the safe driving behavior score.
Further, in step S2, the normalization method is linear function normalization (Min-Max scaling).
Further, in the step S3, the clustering method adopts a k-means clustering method.
Further, in the step S4, the deep learning model is a VGG16 deep learning model.
Further, in the step S5, the safe driving behavior level is divided into A, B, C, D, E five gears, where the score is greater than 80 and less than or equal to 100 for the a gear, the score is greater than 60 and less than or equal to 80 for the B gear, the score is greater than 40 and less than or equal to 60 for the C gear, the score is greater than 20 and less than or equal to 40 for the D gear, and the score is greater than 0 and less than or equal to 20 for the E gear.
Further, the driving behavior data includes one or more of positioning data, speed data, acceleration data, and fuel consumption data collected by various sensors on the target vehicle.
Further, the driving behavior data includes a number of rapid accelerations, a number of rapid decelerations, a number of rapid brakes, a maximum speed, and an average speed.
In the present invention, a storage medium stores a computer readable program, and when the computer readable program is called by a processor, the steps of the method for determining the safety level of driving behavior according to the present invention can be implemented.
The invention has the following advantages: the method is characterized in that a user travel is labeled based on driving behavior data such as more extensive rapid acceleration times, rapid deceleration times, rapid braking times and the like, then the user travel is trained through a deep learning model, a method for calculating safe driving behavior scores is provided, dependence on data such as insurance is avoided, influence of unbalanced insurance data samples is avoided, advantages and disadvantages are evaluated from driving behavior data, and meanwhile, the deep learning model is introduced, so that the training fitting degree is higher, the training effect is better, and the model result can reflect the real driving situation.
Drawings
Fig. 1 is a flowchart in the present embodiment.
Detailed Description
The invention will be further explained with reference to the drawings.
As shown in fig. 1, in this embodiment, a method for determining a safety level of driving behavior includes the following steps:
and S1, acquiring the driving behavior data of the target vehicle in the monitoring period.
The driving behavior data comprises one or more of positioning data, speed data, acceleration data and oil consumption data collected by various sensors on the target vehicle. The data collected by the sensors can reflect the driving state of the vehicle in the driving process, so that the driving decision of the driver such as rapid acceleration, sharp turning, steady speed driving and the like can be reflected.
In general, a vehicle includes various sensors for acquiring data, for example, a GPS sensor for acquiring longitude and latitude, a gyroscope for acquiring angular velocity, an accelerometer for acquiring longitudinal acceleration, lateral acceleration, or vertical acceleration, an oil amount sensor for acquiring instantaneous oil consumption, and the like. The data collected by these sensors can be transmitted to an onboard T-BOX device or OBD device. The vehicle-mounted T-BOX device or the OBD device uploads the data to the Internet of vehicles, so that driving behavior data can be acquired through the Internet of vehicles.
And S2, normalizing the acquired driving behavior data.
The data are mapped to the range of 0-1 for processing, so that the data processing is more convenient and faster, and meanwhile, indexes of different units or orders of magnitude can be compared and weighted conveniently.
In this embodiment, the normalization method is linear function normalization (Min-Max scaling).
And S3, obtaining a clustering result from the normalized data result by a clustering method.
In this embodiment, the clustering method performs clustering by using k-means, and specifically includes:
s31, selecting the number of clusters to be 3 and randomly initializing Gaussian distribution parameters (mean and variance) of each cluster;
s32, given the Gaussian distribution of each cluster, calculating the probability that each data point belongs to each cluster;
s33, calculating gaussian distribution parameters based on the probabilities to maximize the probability of the data point, and calculating the new parameters using the weighting of the probability of the data point, where the weighting is the probability of the data point belonging to the cluster;
s34, repeating the iteration steps S32 and S33 until the change in the iteration reaches a preset value;
s35, marking the cluster label on the left as 1, not processing the middle cluster, and marking the cluster on the right as 0;
and S36, combining the left clusters and the right clusters together to form a new data set, and marking the value unchanged.
And S4, training the data set by using a deep learning model for the clustering result, giving the probability that each piece of data is 0, and multiplying the probability by 100 to obtain the result, namely the calculated safe driving behavior score.
In this embodiment, the deep learning model is a VGG16 deep learning model, and the VGG16 deep learning model is an improved network structure performed on the basis of a deep learning network VGG Net, and includes 13 convolutional layers, 3 fully-connected layers, and 5 pooling layers.
In the embodiment, data after normalization of the number of rapid acceleration, the number of rapid deceleration, the number of rapid braking, the maximum speed and the average speed in the driving behavior data are used as characteristics of a training sample, the data are used as input of an vgg16 deep learning model, a label value is used as a training label and is used as output of a vgg16 deep learning model, and parameter values of a vgg16 deep learning model are trained through iteration; and (4) deeply learning the probability value output by the model according to vgg16, and then multiplying the probability value by 100 to obtain the score of safe driving behavior.
And S5, giving a safe driving behavior grade according to the safe driving behavior score.
In this embodiment, the higher the score is, the higher the driving behavior compliance is, and the lower the risk potential is. According to specific data characteristics, the safe driving behavior grades are divided into A, B, C, D, E five gears, wherein the A gear is obtained when the score is greater than 80 and less than or equal to 100, the B gear is obtained when the score is greater than 60 and less than or equal to 80, the C gear is obtained when the score is greater than 40 and less than or equal to 60, the D gear is obtained when the score is greater than 20 and less than or equal to 40, and the E gear is obtained when the score is greater than 0 and less than or equal to 20. Wherein, the grade A represents the highest grade of the score, and the grade E represents the lowest grade of the score.
In this embodiment, a storage medium stores a computer readable program, and when the computer readable program is called and executed by a processor, the steps of the method for determining the safety level of driving behavior as described in this embodiment can be implemented.
Claims (8)
1. A method for judging the safety level of driving behavior is characterized by comprising the following steps:
s1, acquiring driving behavior data of the target vehicle in a monitoring period;
s2, normalizing the acquired driving behavior data;
s3, obtaining a clustering result from the normalized data result by a clustering method;
s4, training the data set by using a deep learning model for the clustering result, finally giving the probability that each piece of data is 0, and multiplying the probability by 100 to obtain the result which is the calculated safe driving behavior score;
and S5, giving a safe driving behavior grade according to the safe driving behavior score.
2. The method of determining a safety level of driving behavior according to claim 1, characterized in that: in step S2, the normalization method is linear function normalization (Min-Max scaling).
3. The method of determining a safety level of driving behavior according to claim 2, characterized in that: in the step S3, the clustering method adopts a k-means clustering method.
4. The method of determining a safety level of driving behavior according to claim 3, characterized in that: in the step S4, the deep learning model is a VGG16 deep learning model.
5. The method of determining a safety level of driving behavior according to claim 4, characterized in that: in step S5, the safe driving behavior is classified into A, B, C, D, E five gears, where the score is greater than 80 and less than or equal to 100, the score is greater than 60 and less than or equal to 80, the score is greater than 40 and less than or equal to 60, the score is greater than 20 and less than or equal to 40, the score is greater than 0 and less than or equal to 20, and the score is equal to E.
6. The method of judging the safety level of driving behavior according to any one of claims 1 to 5, characterized in that: the driving behavior data comprises one or more of positioning data, speed data, acceleration data and oil consumption data collected by various sensors on the target vehicle.
7. The method of determining a safety level of driving behavior according to claim 6, characterized in that: the driving behavior data comprises the times of sharp acceleration, the times of sharp deceleration, the times of sharp braking, the maximum speed and the average speed.
8. A storage medium having stored therein a computer readable program which, when invoked by a processor, is capable of carrying out the steps of the method of determining a safety level for driving behaviour according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010710791.2A CN111874000A (en) | 2020-07-22 | 2020-07-22 | Method for judging safety level of driving behavior and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010710791.2A CN111874000A (en) | 2020-07-22 | 2020-07-22 | Method for judging safety level of driving behavior and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111874000A true CN111874000A (en) | 2020-11-03 |
Family
ID=73155223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010710791.2A Pending CN111874000A (en) | 2020-07-22 | 2020-07-22 | Method for judging safety level of driving behavior and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111874000A (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050159851A1 (en) * | 2001-01-21 | 2005-07-21 | Volvo Technology Corporation | System and method for real-time recognition of driving patterns |
| CN101633359A (en) * | 2008-07-24 | 2010-01-27 | 通用汽车环球科技运作公司 | Adaptive vehicle control system with driving style recognition |
| CN103661397A (en) * | 2012-09-26 | 2014-03-26 | 福特环球技术公司 | Assessment of driving behavior of a driver of a motor vehicle |
| CN107153916A (en) * | 2017-04-30 | 2017-09-12 | 安徽中科美络信息技术有限公司 | A kind of driving behavior evaluation method clustered based on FCM with BP neural network |
| US20180060970A1 (en) * | 2016-09-01 | 2018-03-01 | International Business Machines Corporation | System and method for context-based driver monitoring |
| CN108280484A (en) * | 2018-01-30 | 2018-07-13 | 辽宁工业大学 | A kind of driver's accelerating performance online classification and discrimination method |
| CN108407814A (en) * | 2018-03-08 | 2018-08-17 | 辽宁工业大学 | A kind of driver characteristics discrimination method |
| CN109987102A (en) * | 2018-01-03 | 2019-07-09 | 田文洪 | A kind of method and apparatus of the High Precision Automatic identification driver's unsafe behaviors based on convolutional neural networks |
| US20190220028A1 (en) * | 2019-03-26 | 2019-07-18 | Intel Corporation | Computer-assisted (ca)/autonomous driving (ad) vehicle inference model creation |
| US20190263417A1 (en) * | 2018-02-28 | 2019-08-29 | CaIAmp Corp. | Systems and methods for driver scoring with machine learning |
| GB2572640A (en) * | 2018-04-06 | 2019-10-09 | Jaguar Land Rover Ltd | Modifying mapping of throttle pedal input to torque demand |
| CN110667593A (en) * | 2019-09-06 | 2020-01-10 | 中国平安财产保险股份有限公司 | Driving reminding method, device and equipment based on deep learning and storage medium |
| US20200184359A1 (en) * | 2016-07-14 | 2020-06-11 | King Fahd University Of Petroleum And Minerals | Method for predicting permeability and oil content in a geological formation |
| CN111310562A (en) * | 2020-01-10 | 2020-06-19 | 中国平安财产保险股份有限公司 | Vehicle driving risk management and control method based on artificial intelligence and related equipment thereof |
-
2020
- 2020-07-22 CN CN202010710791.2A patent/CN111874000A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050159851A1 (en) * | 2001-01-21 | 2005-07-21 | Volvo Technology Corporation | System and method for real-time recognition of driving patterns |
| CN101633359A (en) * | 2008-07-24 | 2010-01-27 | 通用汽车环球科技运作公司 | Adaptive vehicle control system with driving style recognition |
| CN103661397A (en) * | 2012-09-26 | 2014-03-26 | 福特环球技术公司 | Assessment of driving behavior of a driver of a motor vehicle |
| US20200184359A1 (en) * | 2016-07-14 | 2020-06-11 | King Fahd University Of Petroleum And Minerals | Method for predicting permeability and oil content in a geological formation |
| US20180060970A1 (en) * | 2016-09-01 | 2018-03-01 | International Business Machines Corporation | System and method for context-based driver monitoring |
| CN107153916A (en) * | 2017-04-30 | 2017-09-12 | 安徽中科美络信息技术有限公司 | A kind of driving behavior evaluation method clustered based on FCM with BP neural network |
| CN109987102A (en) * | 2018-01-03 | 2019-07-09 | 田文洪 | A kind of method and apparatus of the High Precision Automatic identification driver's unsafe behaviors based on convolutional neural networks |
| CN108280484A (en) * | 2018-01-30 | 2018-07-13 | 辽宁工业大学 | A kind of driver's accelerating performance online classification and discrimination method |
| US20190263417A1 (en) * | 2018-02-28 | 2019-08-29 | CaIAmp Corp. | Systems and methods for driver scoring with machine learning |
| CN108407814A (en) * | 2018-03-08 | 2018-08-17 | 辽宁工业大学 | A kind of driver characteristics discrimination method |
| GB2572640A (en) * | 2018-04-06 | 2019-10-09 | Jaguar Land Rover Ltd | Modifying mapping of throttle pedal input to torque demand |
| US20190220028A1 (en) * | 2019-03-26 | 2019-07-18 | Intel Corporation | Computer-assisted (ca)/autonomous driving (ad) vehicle inference model creation |
| CN110667593A (en) * | 2019-09-06 | 2020-01-10 | 中国平安财产保险股份有限公司 | Driving reminding method, device and equipment based on deep learning and storage medium |
| CN111310562A (en) * | 2020-01-10 | 2020-06-19 | 中国平安财产保险股份有限公司 | Vehicle driving risk management and control method based on artificial intelligence and related equipment thereof |
Non-Patent Citations (2)
| Title |
|---|
| 张志鸿: "基于OBD数据分析的驾驶行为研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
| 陈应显等: "《矿山空间信息集成与数据挖掘》", 30 November 2017, 煤炭工业出版社 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Magana et al. | GAFU: Using a gamification tool to save fuel | |
| CN110615001B (en) | Driving safety reminding method, device and medium based on CAN data | |
| JP6918137B2 (en) | Driving behavior evaluation method, device and computer-readable storage medium | |
| CN108407816B (en) | Method and system for evaluating driving of automobile driver | |
| CN111461185A (en) | Driving behavior analysis method based on improved K-means | |
| CN109840612A (en) | User's driving behavior analysis method and system | |
| CN107953888B (en) | Road type identification method and system | |
| CN111688713B (en) | Driving behavior analysis method and device | |
| Kovaceva et al. | Identification of aggressive driving from naturalistic data in car-following situations | |
| CN109572706B (en) | Driving safety evaluation method and device | |
| CN112200932A (en) | Method and equipment for evaluating energy consumption of heavy-duty diesel vehicle | |
| CN110858312A (en) | Driver driving style classification method based on fuzzy C-means clustering algorithm | |
| CN111563555A (en) | Driver driving behavior analysis method and system | |
| WO2022237418A1 (en) | Longitudinal tracking control method and apparatus, device, and storage medium | |
| Wu et al. | Clustering of several typical behavioral characteristics of commercial vehicle drivers based on GPS data mining: Case study of highways in China | |
| CN113581188A (en) | Commercial vehicle driver driving style identification method based on Internet of vehicles data | |
| CN114332825B (en) | Road terrain distribution identification method and device based on deep learning and storage medium | |
| CN113657432A (en) | Commercial vehicle driving behavior risk level identification method based on Internet of vehicles data | |
| CN113442935B (en) | Method and system for judging poor driving behavior of commercial vehicle | |
| CN115221234A (en) | Method and system for portraying user based on power assembly data | |
| CN111874000A (en) | Method for judging safety level of driving behavior and storage medium | |
| CN114426025B (en) | Driving assistance method, driving assistance device, computer device, and storage medium | |
| CN111047142A (en) | Automobile scoring method and system based on analytic hierarchy process | |
| CN113173175B (en) | Vehicle weight determination method and device | |
| Stokić et al. | Evaluation of driver's eco-driving skills based on fuzzy logic model: A realistic example of vehicle operation in real-world conditions |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201103 |