WO2020015574A1

WO2020015574A1 - Driving behavior analysis method, device and apparatus, and storage medium

Info

Publication number: WO2020015574A1
Application number: PCT/CN2019/095533
Authority: WO
Inventors: 吴栋磊; 叶敬福; 沈宇峰
Original assignee: 阿里巴巴集团控股有限公司
Priority date: 2018-07-18
Filing date: 2019-07-11
Publication date: 2020-01-23
Also published as: CN110733509A

Abstract

The present disclosure proposes a driving behavior analysis method, device and apparatus, and a storage medium. The method comprises: acquiring vehicle information and ambient environment information of a vehicle during driving; and analyzing the vehicle information and the ambient environment information on the basis of data modeling and/or artificial intelligence, so as to obtain a first driving behavior analysis result for the driver of the vehicle. Thus, by combining richer data, the present disclosure can realize a comprehensive and detailed analysis of driving behavior, thereby improving the accuracy of the driving behavior analysis result.

Description

Driving behavior analysis method, device, equipment and storage medium

This application claims the priority of a Chinese patent application filed on July 18, 2018 with the application number 201810792927.1 and the invention name "Driving Behavior Analysis Method, Device, Equipment, and Storage Medium", the entire contents of which are incorporated herein by reference. .

Technical field

The present disclosure relates to the field of transportation, and in particular, to a driving behavior analysis method, device, device, and storage medium.

Background technique

With the popularity of automobiles, automobile-related service industries have also gradually emerged. For example, car insurance, taxis, and car-hailing are increasingly appearing in people's daily lives. With more and more cars on the road, the issue of safe driving is becoming more prominent.

In Europe, to encourage safe driving, insurers have launched an insurance business that charges based on driving performance. In 2014, sales of insurance-based policies increased by 42%. In North America, total sales of insurance-based insurance policies are expected to increase from 4.2 million in 2014 to 32.5 million in 2019, with an average compound annual growth rate of 50%.

This business model of providing corresponding services based on the driving behavior exhibited by drivers will become increasingly popular. Therefore, how to accurately analyze the driving behavior of the driver is the main problem currently facing.

Summary of the invention

An object of the present disclosure is to provide a driving behavior analysis scheme capable of accurately analyzing a driving behavior of a driver.

According to a first aspect of the present disclosure, a driving behavior analysis method is provided, which includes: acquiring vehicle information and surrounding environment information during a driving process of the vehicle; and analyzing vehicle information and data based on data modeling and / or artificial intelligence. The surrounding environment information is analyzed to obtain a first driving behavior analysis result for the driver of the vehicle.

Optionally, the vehicle information includes one or more of the following: body information; driving records; positioning information; and information collected by vehicle sensors.

Optionally, the surrounding environment information includes one or more of the following: surrounding vehicle information; surrounding pedestrian information; roadside information; and change information of the environment as the vehicle travels.

Optionally, the roadside information includes one or more of the following information collected based on the drive test unit: intersection information; road information; captured images; traffic signal information; and road identification information.

Optionally, the step of analyzing vehicle information and surrounding environment information based on data modeling and / or artificial intelligence includes: modeling the vehicle information and surrounding environment information, and analyzing the established model to obtain the first A driving behavior analysis result; or analysis of vehicle information and surrounding environment information based on artificial intelligence technology to obtain a first driving behavior analysis result; or modeling based on vehicle information and surrounding environment information, and analyzing the established model, and Based on the artificial intelligence technology, the vehicle information and the surrounding environment information are analyzed. Based on the two analysis results, the first driving behavior analysis result is comprehensively obtained.

Optionally, the step of analyzing the vehicle information and the surrounding environment information based on the artificial intelligence technology includes: using a driving behavior analysis model to analyze the driving behavior of the driver based on the vehicle information and the surrounding environment information, wherein The driving behavior analysis model is obtained by training based on a deep learning algorithm.

Optionally, the driving behavior analysis method further includes analyzing the driving state of the driver to obtain a second driving behavior analysis result for the driving state, and / or, based on the vehicle information and the surrounding environment information, whether the driving behavior is violated The traffic regulations are analyzed to obtain the third driving behavior analysis result; and based on the first driving behavior analysis result, the second driving behavior analysis result, and / or the third driving behavior analysis result, the current driving behavior analysis result is obtained.

Optionally, the driving behavior analysis method further includes: identifying the driver's identity; and obtaining a historical driving behavior analysis result corresponding to the driver based on the recognition result.

Optionally, the driving behavior analysis method further includes: determining a total behavior analysis result of the driver based on a current driving behavior analysis result and a historical driving behavior analysis result.

Optionally, the driving behavior analysis method further includes: re-analyzing the driving behavior in a predetermined time according to the vehicle information and the surrounding environment information to obtain a historical driving behavior analysis calibration result.

Optionally, the driving behavior analysis method further includes: determining a total behavior analysis result of the driver based on a current driving behavior analysis result, a historical driving behavior analysis calibration result, and a historical driving behavior analysis result.

Optionally, the driving behavior analysis method further includes: scoring the driver according to the total behavior analysis result, so as to provide the driver with corresponding services according to the score.

According to a second aspect of the present disclosure, a driving behavior analysis device is further provided, including: a first acquisition module for acquiring vehicle information and surrounding environment information of a vehicle during driving; and a first analysis module for The vehicle information and the surrounding environment information are analyzed based on data modeling and / or artificial intelligence to obtain a first driving behavior analysis result for the driver of the vehicle.

Optionally, the first analysis module models the vehicle information and the surrounding environment information, and analyzes the established model to obtain the first driving behavior analysis result, or the first analysis module analyzes the vehicle information and the surrounding area based on artificial intelligence technology. Environmental information is analyzed to obtain the first driving behavior analysis result, or the first analysis module performs modeling based on vehicle information and surrounding environment information, analyzes the established model, and performs vehicle information and surrounding environment information based on artificial intelligence technology Analysis. Based on the two analysis results, the first driving behavior analysis result is synthesized.

Optionally, based on the vehicle information and the surrounding environment information, a first analysis module analyzes the driving behavior of the driver using a driving behavior analysis model, wherein the driving behavior analysis model is trained based on a deep learning algorithm owned.

Optionally, the driving behavior analysis device further includes: a second analysis module configured to analyze the driving state of the driver to obtain a second driving behavior analysis result for the driving state, and / or a third analysis module for Based on vehicle information and surrounding environment information, analyze whether driving behavior violates traffic regulations to obtain a third driving behavior analysis result; and a first determination module for using the first driving behavior analysis result and the second driving behavior analysis result And / or a third driving behavior analysis result to determine a current driving behavior analysis result.

Optionally, the driving behavior analysis device further includes: an identification module for identifying the driver's identity; and a second acquisition module for acquiring a historical driving behavior analysis result corresponding to the driver based on the identification result.

Optionally, the driving behavior analysis device further includes: a second determination module, configured to determine a total behavior analysis result of the driver based on a current driving behavior analysis result and a historical driving behavior analysis result.

Optionally, the driving behavior analysis device further includes: a calibration module, configured to re-analyze the driving behavior in a predetermined time according to the vehicle information and the surrounding environment information to obtain a historical driving behavior analysis calibration result.

Optionally, the driving behavior analysis device further includes a third determination module for determining a total behavior analysis result of the driver based on a current driving behavior analysis result, a historical driving behavior analysis calibration result, and a historical driving behavior analysis result.

Optionally, the driving behavior analysis device further includes: a scoring module, configured to score the driver according to the total behavior analysis result, so as to provide the driver with corresponding services according to the score.

According to a third aspect of the present disclosure, there is also provided a computing device including: a processor; and a memory, which stores executable code, and when the executable code is executed by the processor, causes the processor to execute as the present disclosure The method mentioned in the first aspect.

According to a fourth aspect of the present disclosure, there is also provided a non-transitory machine-readable storage medium having executable code stored thereon, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute as described herein. Disclose the method mentioned in the first aspect.

By combining richer data, the present disclosure can achieve a comprehensive and detailed analysis of driving behavior, thereby improving the accuracy of the driving behavior analysis result.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other objects, features, and advantages of the present disclosure will become more apparent by describing the exemplary embodiments of the present disclosure in more detail with reference to the accompanying drawings. In the exemplary embodiments of the present disclosure, the same reference numerals generally represent Same parts.

1 is a schematic flowchart illustrating a driving behavior analysis method according to an embodiment of the present disclosure;

2 is a flowchart illustrating a driving behavior analysis according to an embodiment of the present disclosure;

3 is a schematic block diagram showing a structure of a driving behavior analysis device according to an embodiment of the present disclosure;

FIG. 4 is a schematic structural diagram of a computing device that can be used to implement the data processing of the driving behavior analysis method according to an embodiment of the present invention.

detailed description

Hereinafter, preferred embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the preferred embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[Term Explanation]

V2X: Vehicle to Everything, which refers to the exchange of information between the car and the outside world, and is a general term for a series of on-board communication technologies. Generally speaking, V2X mainly includes car-to-car (V2V), car-to-roadside equipment (V2R), car-to-infrastructure (V2I), car-to-pedestrian (V2P), car-to-locomotive (V2M), and car-to-bus ( V2T) and other six categories.

OBU: Abbreviation of On Board and Unit, literal translation is the meaning of car unit. OBU is installed on the vehicle and can be regarded as a microwave device that uses DSRC (Dedicated Short Range and Communication) technology to communicate with RSU.

RSU: Abbreviation for Road Side Unit. Literal translation means roadside unit. It is installed on the roadside and uses DSRC (Dedicated Short Range / Communication) technology to communicate with on-board units (OBU).

[Program overview]

The driver's portrait can reflect the driver's driving behavior habits, and can be regarded as a driver's driving behavior evaluation report. However, at present, when analyzing the driving behavior of a driver to obtain a driver's portrait, the driving data of a single vehicle is generally analyzed. The accuracy of the analysis result is poor, making the obtained driver's portrait rough and credible. Not high enough to make a true assessment of the driver's driving behavior. For example, the existing solutions cannot distinguish between braking due to a failure of a front vehicle and braking due to rough driving.

In view of this, the present disclosure proposes that when analyzing the driving behavior of a driver, not only the data of the vehicle driven by the driver (that is, the vehicle data described below), but also the surroundings during the driving of the vehicle can be referenced at the same time. Environmental information, such as surrounding vehicle information, surrounding pedestrian information, roadside information, and changes in the environment with the vehicle's driving process. Comprehensive and detailed analysis of the driver's driving behavior is combined with richer data to improve the results of driving behavior analysis Accuracy.

Various aspects related to the present disclosure are further described below.

[Analysis Method of Driving Behavior]

FIG. 1 is a schematic flowchart illustrating a driving behavior analysis method according to an embodiment of the present disclosure. The driving behavior analysis method of the present disclosure can be implemented as a driving behavior analysis program. The driving behavior analysis program can be deployed in the vehicle terminal, in the cloud, or in other terminals, such as roadside terminals. When the driving behavior analysis program is not deployed in the vehicle terminal, a data acquisition program may be deployed on the vehicle to collect data required by the driving behavior analysis program.

Referring to FIG. 1, in step S110, vehicle information and surrounding environment information of a vehicle during driving are acquired.

The vehicle mentioned here refers to a vehicle whose driving behavior is to be analyzed by a driver who drives the vehicle. The vehicle information may include, but is not limited to, one or more of body information, driving records, positioning information (such as position navigation information), and information collected by vehicle sensors. Further, the vehicle body information may include information such as a vehicle body size, a state of a vehicle body (such as data related to a state of a lamp, a turn signal, a vehicle failure, a state of a seat belt, an airbag lamp, and the like). The driving record may be information such as time, speed, and location recorded by the driving recorder of the vehicle. The positioning information may be position information determined by a vehicle-based navigation system (such as a GPS navigation system). Vehicle sensors can include a variety of sensors for detecting specific parameters. For example, they can include speed sensors for detecting speed, triangular acceleration, horizontal axis angular velocity, gear position, rotational speed, throttle, and other speed information. An image sensor (such as a camera) taken by a person (such as a driver) may also include other types of sensors, and details are not described herein again.

The surrounding environment information refers to environmental information near the vehicle during driving. As an example, the surrounding environment information may include, but is not limited to, surrounding vehicle information, roadside information, surrounding pedestrian information, road facility information, map information, and environmental change information during vehicle driving. Among them, the roadside information may be information collected based on a roadside unit (RSU), for example, it may include, but is not limited to, intersection information (such as information on the location of an intersection, the direction of an intersection lane, etc.), and road information (such as the type of road, location, direction, (Curvature, road conditions, road weather, road failures, road construction, etc.), captured images, traffic signal information (such as traffic light status and duration information), and road identification information (such as road signs, transit time, lane speed limit, etc.) And more.

The surrounding vehicle information may be vehicle information of other nearby vehicles during the driving process of the vehicle. The vehicle information of other vehicles may also include, but is not limited to, one or more of body information, driving records, positioning information, and information collected by vehicle sensors. item. The surrounding pedestrian information may be information of nearby pedestrians during the driving of the vehicle, such as location information of pedestrians (which may include people riding bicycles and non-motorized vehicles). Among them, the surrounding pedestrian information can be obtained from the images taken by the camera on the road. The road facility information may be road facility information on the road where the vehicle is traveling and surrounding road facilities, such as, but may not be limited to, a signal light, a traffic sign, a road marking, a guardrail, a fence, lighting equipment, a line of sight guide, an anti-glare facility, and the like. The map information may be an electronic map corresponding to a road on which the vehicle travels.

The change information of the environment with the vehicle driving process can be used to characterize the change of the surrounding environment information with the driving process. The change information can include the distance change information between the vehicle and the surrounding vehicles and pedestrians during the driving process, and it can also include the traffic signal lights during the driving process. The state change information may also include change information of a lane in which the vehicle is driving. The information of the change of the environment with the driving process of the vehicle can characterize the driving behavior of the driver to a certain extent. For example, the vehicle is always very close to other vehicles, always overtaking, changing lanes, always grabbing red lights, yellow lights, red lights, and green lights as soon as it starts. When the traffic jams ahead, it runs to the emergency lane. These driving behaviors are all acceptable. Reflected from the changes in the surrounding environment information with the driving process. The change information of the environment with the driving process of the vehicle can be obtained through calculation and analysis of the original collected data. For example, for the position change information of the vehicle and the surrounding vehicles, the video captured by the vehicle camera on the surrounding environment during the driving of the vehicle can be obtained. It can be obtained by analysis, and it can also be obtained by analyzing the positioning information of the vehicle and surrounding vehicles during driving. Regarding the manner of acquiring the change information of the environment as the vehicle travels, this disclosure does not limit it.

Taking the driving behavior analysis program deployed in the vehicle terminal as an example, the surrounding environment information can be collected based on V2X technology. For example, a vehicle may obtain vehicle information of other vehicles in the surrounding area based on V2X technology, and may also obtain information of a roadside unit RSU based on V2X technology, and the obtained information may be passed to a driving behavior analysis program together with its own vehicle information. In addition, the driving behavior analysis program can also be deployed in the cloud or on the roadside terminal. When the driving behavior analysis program is deployed on the other end, the specific data transmission process is also different, and details are not repeated here. In addition, the present disclosure does not limit the specific communication methods for acquiring vehicle data and surrounding environment information. For example, data required for analysis can also be acquired based on 3G, 4G, and other communication methods.

In step S120, the vehicle information and the surrounding environment information are analyzed to obtain a first driving behavior analysis result for the driver of the vehicle.

After the vehicle information and the surrounding environment information are obtained, analysis can be performed based on a variety of ways to obtain a driving behavior analysis result for the driver (for convenience of distinction, it may be referred to as a "first driving behavior analysis result"). For example, analysis can be performed through data modeling, or artificial intelligence (that is, AI technology, such as machine learning technology in AI technology) can be used for analysis, and comprehensive analysis can be performed by combining these two methods.

As an example of the present disclosure, modeling may be performed according to vehicle information and surrounding environment information, and the established model may be analyzed to obtain a first driving behavior analysis result. That is, based on the analysis of the driving behavior of the vehicle, the analysis of the surrounding environment information (such as the surrounding vehicle information) can be used to make the analysis result more accurate. For example, when simply analyzing the driving behavior of a vehicle slamming the brake, the behavior will be considered as dangerous driving, and the present disclosure can further analyze the cause of the driving behavior of sudden braking in combination with the surrounding environment information, such as in When a vehicle in front fails or a pedestrian enters in front of it, the sudden braking of the vehicle should not be considered as dangerous driving. For another example, in combination with the surrounding environment information, you can also analyze a variety of other results that cannot be obtained based on the vehicle's own information alone, such as whether a sufficient safety distance is maintained when the vehicle is overtaking, whether the vehicle starts at high speed as soon as the green light is on, etc. .

As another example of the present disclosure, vehicle information and surrounding environment information may also be analyzed based on artificial intelligence technology to obtain a first driving behavior analysis result. For example, a driving behavior analysis model based on a machine learning technology (such as deep learning technology) can be used to analyze vehicle information and surrounding environment information. Specifically, the vehicle information and the surrounding environment information can be used as sample features to determine the driving behavior analysis result (such as the analysis result determined based on the modeling method) as a sample mark, and then use a machine learning algorithm (such as a deep learning algorithm) to perform Model training to get driving behavior analysis models. Therefore, the present disclosure can use deep learning to perform data analysis in the case of a large amount of training data, so that the factors of the data in various aspects can be more fully mined, and a more accurate AI model (that is, driving behavior) can be obtained. Analysis model).

In addition, if conditions permit, these two analysis methods can be used simultaneously to obtain the first driving behavior analysis result. For example, the vehicle information and the surrounding environment information can be used for modeling, the established model can be analyzed, and the vehicle information and the surrounding environment information can be analyzed based on the machine learning technology. The first driving behavior analysis result can be comprehensively obtained according to the two types of analysis results. For example, corresponding weights can be set for different analysis results, and the first driving behavior analysis result can be obtained by weighted summation.

So far, the basic implementation flow of the driving behavior analysis method of the present disclosure has been described with reference to FIG. 1. It can be seen that when analyzing the driving behavior of the driver, the present disclosure fully combines the surrounding environment information, so that the analysis result can be made more accurate.

Further, in combination with vehicle information and surrounding environment information, a variety of other analysis methods can be used to further analyze driving behavior. The obtained analysis results can be used to summarize the first driving behavior analysis result to obtain the current driving behavior analysis. result.

As an example of the present disclosure, the driving state of the driver may be analyzed to obtain a driving behavior analysis result for the driving state (for convenience of distinction, it may be referred to as a “second driving behavior analysis result”). Specifically, an abnormal driving state such as a driver's fatigue state, distraction detection, and drunk driving can be identified through image detection based on the in-vehicle camera, and then a corresponding second driving behavior analysis result can be obtained based on the identified driving state. .

As another example of the present disclosure, it is also possible to analyze whether driving behavior violates traffic regulations based on vehicle information and surrounding environment information to obtain a driving behavior analysis result of a traffic regulation violation (for convenience of distinction, it may be referred to as "third Driving behavior analysis results "). For example, according to the speed, location, and driving direction of the vehicle during driving, it can be compared with road identification information and / or traffic signal information to determine whether the vehicle violates traffic regulations. For example, the current lane can be calculated based on the vehicle position, speed, and heading angle combined with map information and roadside RSU road information. The speed, heading angle, lane information, and vehicle status are used to match the definition of the roadside RSU. Identification information (road signs, transit time, lane speed limit, etc.), other information (such as traffic lights, temporary control information, etc.) and digital traffic regulations to check for illegal driving. Among them, digital traffic regulations can be sourced from the cloud or stored locally, and are not limited to on-board systems.

After obtaining the second driving behavior analysis result and / or the third driving behavior analysis result, the obtained driving behavior analysis result (the second driving behavior analysis result and / or the third driving behavior analysis result) and the first driving behavior result may be obtained. Summarize to get the current driving behavior analysis results. Among them, the current driving behavior analysis result can be regarded as the driving behavior analysis result obtained from the current analysis. For example, the driving behavior analysis method of the present disclosure may be used to analyze the driving behavior of the driver at a predetermined time interval to obtain the current driving behavior analysis result. The driving behavior analysis of the present disclosure may also be used after the driver completes one driving behavior. Methods The driving behavior was analyzed to obtain the current driving behavior analysis result.

Further, before analyzing the driving behavior of the driver of the vehicle, the identity of the driver may be identified first, and a historical driving behavior analysis result corresponding to the driver may be obtained based on the recognition result to facilitate data aggregation. For example, the image information captured by the camera in the car can be identified to identify the driver's identity and obtain the corresponding historical driving behavior analysis result. The historical driving behavior analysis result may be a result obtained by analyzing the driving behavior analysis method based on the present disclosure, or may be a historical driving behavior analysis result in a driving portrait of a driver. Based on the current driving behavior analysis results and historical driving behavior analysis results, the driver's total behavior analysis results can be obtained. For example, the historical driving behavior analysis result may be updated based on the current driving behavior analysis result to obtain the driver's total behavior analysis result. For another example, different weights may be set for the current driving behavior analysis result and the historical driving behavior analysis result, and comprehensive processing may be performed by means of weighted summation or weighted average to obtain the total behavior analysis result. Among them, the overall behavior analysis result can be used as a driver portrait to evaluate the driving behavior of the driver.

The inventors of the present disclosure have also found that, for certain driving behaviors, it is not possible to make a good analysis and judgment based on the currently obtained data. For example, for driving behaviors of other vehicles, the degree of danger should exceed that of overtaking, but when the first or second overtaking is based only on the currently acquired data, it may be determined to be a dangerous overtaking behavior with a high degree of danger. Lower than each other. Therefore, the present disclosure proposes that the driving behavior can be re-analyzed based on vehicle information and surrounding environment information to obtain a historical driving behavior analysis calibration result. The historical driving behavior analysis calibration result can be used to calibrate the previous driving behavior analysis result, thereby further improving the accuracy of the driving behavior analysis result. As an example, the driver's total behavior analysis result can be determined based on the current driving behavior analysis result, historical driving behavior analysis calibration result, and historical driving behavior analysis result. For example, the weighted average method can be used to obtain the total behavior analysis result.

Based on the results of the overall behavior analysis, drivers can be scored in order to provide drivers with corresponding services, such as insurance services, based on the score. As an example, insurance companies, car manufacturers, software service providers, and other businesses can obtain the results of their driving behavior analysis under the authorization of users (that is, drivers) for further development and application. For example, insurance companies can use them to determine auto insurance. Rates can be used by car manufacturers to improve vehicles based on the results of driving behavior analysis to improve the user's driving experience.

[Application example]

FIG. 2 is a flowchart illustrating a driving behavior analysis according to an embodiment of the present disclosure. Among them, the driving behavior analysis program can be regarded as a program capable of executing the driving behavior analysis method of the present disclosure. It can be deployed on the vehicle, or it can be deployed on the server side, roadside RSU, traffic police mobile phone and other ends.

Taking the driving behavior analysis program deployed on a vehicle as an example, vehicles can communicate with each other through V2X, and roadside RSUs can also communicate with vehicles through V2X. Among them, a private protocol may be used instead of the V2X protocol, or other communication methods may be used instead of the V2X communication.

When analyzing the driving behavior of the vehicle, the vehicle can obtain the vehicle information of other vehicles and the information of the roadside RSU through V2X. The vehicle information may include body information, position navigation, driving records, sensor acquisition data, driving cameras, and other information. The host vehicle can pass the vehicle information of the host vehicle, the obtained vehicle information of the surrounding vehicles, and the roadside RSU information to the driving behavior analysis program, and the driving behavior analysis program executes the driving behavior analysis method of the present disclosure to the driver ’s Analysis of driving behavior. Among them, other sensors can also be used to replace some of the data provided by V2X. For example, millimeter-wave radar can be used to determine the relative position, speed, and direction angle of surrounding vehicles.

As an example, the host vehicle may collect information of surrounding vehicles and roadside RSUs through V2X, and output the information to a driving behavior analysis program. Among them, the information of surrounding vehicles includes: body size, location time (current time, GPS position), body status (data about lights, turn signals, vehicle failure, seat belt status, airbag lights, etc.), speed Direction (heading angle, speed, triangular acceleration, horizontal axis angular speed, gear, rotation speed, throttle and other speed-first data) and other data; roadside RSU information includes: intersection information (intersection location, traffic light status and duration, intersection Lane direction, etc.), road information (road type, location, direction, curvature, road conditions, road weather, road failure, road construction, etc.), identification information (road signs, transit time, lane speed limit, etc.), other information (including Temporary control, etc.); all the above information types can be adjusted according to circumstances;

As shown in Figure 2, the driving behavior analysis program can integrate multiple functions such as data modeling, deep learning, driving record analysis, image detection, status detection, classification labeling, and comprehensive adjustment to comprehensively and accurately analyze driving behavior. .

The driving behavior analysis program can be analyzed as follows.

1. Output driving analysis to the driving behavior analysis program, perform face recognition, determine the identity of the driver, and obtain the historical driving behavior analysis result A, which is convenient for subsequent summary analysis of the data to avoid multiple drivers in a car. Causes data mixing problems.

2. According to the in-vehicle camera, through image detection, identify the driver's fatigue state, distraction detection, etc., and obtain the behavior analysis result of the user's driving state, and record it as result B.

3. Calculate the current lane according to the position, speed, and heading angle of the vehicle in combination with the road information of the roadside RSU, and use the speed, heading angle, lane information, and body status to match the definition of the roadside RSU. Information (road signs, transit time, lane speed limits, etc.), other information (including temporary controls, etc.) and digital traffic regulations to check for illegal driving; recorded as result C; where digital traffic regulations can come from the cloud, It can also be at the local end, not limited to on-board systems.

4. Based on the collected surrounding vehicle data and roadside RSU information, it can combine its own body size, location time (current time, GPS position), body status (lights, turn lights, vehicle failure, seat belt status, Data such as airbag lights and other conditions), speed direction (head angle, speed, triangular acceleration, horizontal axis angular speed, gear, rotational speed, throttle and other speed-off data) are analyzed for data analysis to generate driving behavior analysis Result D. There are two types of analysis methods. If conditions permit, you can try both of them for deployment, and finally synthesize the results.

4.1 Through traditional modeling and analysis, in the analysis of traditional driving behavior data and the analysis of surrounding vehicle factors, the analysis result will be more accurate; for example, the traditional behavior of slamming the brake in traditional bicycle analysis is considered to be a dangerous driving This kind of behavior, but by combining the surrounding vehicle factors, you can further confirm the cause, and the sudden braking when the front vehicle fails should not be considered dangerous driving; in addition, combining the surrounding vehicle factors, you can also analyze what the bicycle cannot analyze. As a result, whether a sufficient safety distance is maintained when the vehicle passes.

4.2 Analysis by means of AI. In the case of a large amount of training data, it is suitable to use deep learning for data analysis; in this respect, the mining of all aspects of the data will be more sufficient, and more accurate analysis results will be obtained.

5. Analysis of surrounding vehicle data and roadside RSU information combined with driving records, re-analysis of driving behavior within a certain time threshold, and perform a certain calibration in conjunction with the current situation to generate a historical record behavior analysis calibration result E; for example The dangerous behavior of the mutual passing vehicle is considered to be a dangerous overtaking when passing for the first time, and the danger is far lower than that of the mutual passing vehicle.

6. The driving behavior analysis program comprehensively summarizes the appeal results B, C, and D, and obtains the analysis result F of the current driving. The historical analysis result A, the calibration result E, and the current analysis result F are weighted to generate a total behavior. The analysis result is the portrait of the driver and user; in the process, driving ratings can also be digitized using safe driving and dangerous driving, and the ratings can be directly applied for commercialization.

Specifically, the data during driving can be saved in the driving record to form a new driving record. The generated analysis results and driving records (if allowed) can be saved under the cloud corresponding to the user's identity for authorized application calls and further analysis. Insurance companies, car manufacturers, and software service providers can obtain this driving behavior analysis result for further development and application under the authorization of users. For example, insurance companies can use it to determine auto insurance rates, and auto manufacturers can use it to Improve driving experience.

[Driving Behavior Analysis Device]

The present disclosure can also be implemented as a driving behavior analysis device. FIG. 3 is a schematic block diagram showing a structure of a driving behavior analysis device according to an embodiment of the present disclosure. The functional modules of the driving behavior analysis device may be implemented by hardware, software, or a combination of hardware and software that implements the principles of the present invention. Those skilled in the art can understand that the functional modules described in FIG. 3 can be combined or divided into sub-modules, thereby realizing the principle of the above invention. Therefore, the description herein may support any possible combination, or division, or further definition of the functional modules described herein.

The following briefly describes the functional modules that the driving behavior analysis device can have and the operations that can be performed by each functional module. For the details involved, refer to the description of the driving planning method above, which is not repeated here.

Referring to FIG. 3, the driving behavior analysis apparatus 300 includes a first acquisition module 310 and a first analysis module 320. The first acquiring module 310 is configured to acquire vehicle information and surrounding environment information of a vehicle during a driving process. The vehicle information may include one or more of the following: body information; driving records; positioning information; and information collected by vehicle sensors. The roadside information may include one or more of the following information collected based on the drive test unit: intersection information; road information; captured images; traffic signal information; and road identification information. The surrounding environment information may include one or more of the following: surrounding vehicle information; surrounding pedestrian information; roadside information; and information about changes in the environment as the vehicle travels.

The first analysis module 320 may analyze vehicle information and surrounding environment information based on data modeling and / or artificial intelligence to obtain a first driving behavior analysis result for a driver of the vehicle.

As an example, the first analysis module 320 may model the vehicle information and the surrounding environment information, and analyze the established model to obtain the first driving behavior analysis result, or the first analysis module 320 may also be based on artificial intelligence technology ( (If it can be machine learning technology) analyze vehicle information and surrounding environment information to obtain the first driving behavior analysis result, or the first analysis module 320 may also perform modeling based on the vehicle information and surrounding environment information, Analysis, and based on artificial intelligence technology (such as machine learning technology) to analyze vehicle information and surrounding environment information, according to the two analysis results, comprehensively obtain the first driving behavior analysis result. The first analysis module 320 may analyze vehicle information and surrounding environment information using a driving behavior analysis model established based on machine learning technology. As an example, the driving behavior analysis model may be trained based on a deep learning algorithm.

As shown in FIG. 3, the driving behavior analysis device 300 may optionally further include a second analysis module 330 and / or a third analysis module 340 and a first determination module 350 shown by dashed boxes in the figure.

The second analysis module 330 is configured to analyze the driving state of the driver to obtain a second driving behavior analysis result for the driving state. The third analysis module 340 is configured to analyze whether driving behavior violates traffic regulations based on vehicle information and surrounding environment information, so as to obtain a third driving behavior analysis result. The first determining module 350 is configured to determine a current driving behavior analysis result based on a first driving behavior analysis result, a second driving behavior analysis result, and / or a third driving behavior analysis result.

As shown in FIG. 3, the driving behavior analysis device 300 may optionally further include a recognition module 360 and a second acquisition module 365 shown by dashed boxes in the figure. The identification module 360 is configured to identify the identity of the driver. The second acquisition module 365 is configured to acquire a historical driving behavior analysis result corresponding to the driver based on the recognition result.

As shown in FIG. 3, the driving behavior analysis device 300 may optionally further include a second determination module 370 shown by a dashed box in the figure. The second determination module 370 is configured to determine a total behavior analysis result of the driver based on a current driving behavior analysis result and a historical driving behavior analysis result.

As shown in FIG. 3, the driving behavior analysis device 300 may optionally further include a calibration module 380 shown by a dashed box in the figure. The calibration module 380 is configured to re-analyze the driving behavior in the previous predetermined time according to the vehicle information and the surrounding environment information to obtain a historical driving behavior analysis calibration result.

As shown in FIG. 3, the driving behavior analysis device 300 may optionally further include a third determination module 390 shown by a dashed box in the figure. The third determination module 390 is configured to determine a total behavior analysis result of the driver based on a current driving behavior analysis result, a historical driving behavior analysis calibration result, and a historical driving behavior analysis result.

As shown in FIG. 3, the driving behavior analysis device 300 may optionally further include a scoring module 395 shown by a dashed box in the figure. The scoring module 395 is configured to score the driver according to the total behavior analysis result, so as to provide the driver with corresponding services according to the score.

[Computing Equipment]

Referring to FIG. 4, the computing device 400 includes a memory 410 and a processor 420.

The processor 420 may be a multi-core processor, or may include multiple processors. In some embodiments, the processor 420 may include a general-purpose main processor and one or more special co-processors, such as a graphics processor (GPU), a digital signal processor (DSP), and the like. In some embodiments, the processor 420 may be implemented using customized circuits, such as Application Specific Integrated Circuits (ASICs) or Field Programmable Gate Arrays (FPGAs).

The memory 410 may include various types of storage units, such as a system memory, a read-only memory (ROM), and a permanent storage device. The ROM may store static data or instructions required by the processor 420 or other modules of the computer. The persistent storage device may be a readable and writable storage device. Persistent storage devices can be non-volatile storage devices that do not lose stored instructions and data even when the computer is powered off. In some embodiments, the permanent storage device uses a mass storage device (eg, magnetic or optical disk, flash memory) as the permanent storage device. In other embodiments, the permanent storage device may be a removable storage device (for example, a floppy disk, an optical drive). The system memory can be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory can store some or all of the instructions and data required by the processor while it is running. In addition, the memory 410 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory). Magnetic disks and / or optical disks may also be used. In some embodiments, the memory 410 may include a readable and / or writable removable storage device, such as a compact disc (CD), a read-only digital versatile disc (eg, DVD-ROM, dual-layer DVD-ROM), Read-only Blu-ray discs, ultra-density discs, flash memory cards (such as SD cards, SD cards, Micro-SD cards, etc.), magnetic floppy disks, etc. Computer-readable storage media does not contain carrier waves and transient electronic signals transmitted by wireless or wire.

The memory 410 stores executable code. When the executable code is executed by the processor 420, the processor 420 can be caused to execute the driving behavior analysis method mentioned above.

The driving behavior analysis method, device, device, and storage medium according to the present invention have been described in detail above with reference to the accompanying drawings.

In addition, the method according to the present invention may also be implemented as a computer program or computer program product including computer program code instructions for performing the above steps defined in the above method of the present invention.

Alternatively, the present invention may also be implemented as a non-transitory machine-readable storage medium (or computer-readable storage medium or machine-readable storage medium) on which executable code (or computer program or computer instruction code) is stored. ), When the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or computing device, server, etc.), causing the processor to perform each step of the above-mentioned method according to the present invention .

Those skilled in the art will also appreciate that the various exemplary logic blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or combinations of both.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems and methods according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, which contains one or more components for implementing a specified logical function Executable instructions. It should also be noted that in some alternative implementations, the functions labeled in the blocks may also occur in a different order than those labeled in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified function or operation. , Or it can be implemented with a combination of dedicated hardware and computer instructions.

The embodiments of the present invention have been described above, the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the embodiments described. The terminology used herein is chosen to best explain the principles of the embodiments, practical applications or improvements to the technology in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

A driving behavior analysis method, comprising:

Obtain vehicle information and surrounding environment information while the vehicle is driving; and

The vehicle information and the surrounding environment information are analyzed based on data modeling and / or artificial intelligence to obtain a first driving behavior analysis result for a driver of the vehicle.
The driving behavior analysis method according to claim 1, wherein the vehicle information includes one or more of the following:

Body information

Driving record

Location information; and

Information collected by vehicle sensors.
The driving behavior analysis method according to claim 1, wherein the surrounding environment information includes one or more of the following:

Information on surrounding vehicles;

Information on surrounding pedestrians;

Roadside information; and

Information about changes in the environment as the vehicle travels.
The driving behavior analysis method according to claim 3, wherein the roadside information comprises one or more of the following information collected based on a drive test unit:

Intersection information

Road information

Take an image

Traffic light information; and

Road sign information.
The driving behavior analysis method according to claim 1, wherein the step of analyzing the vehicle information and the surrounding environment information based on the data modeling and / or artificial intelligence method comprises:

Modeling according to the vehicle information and the surrounding environment information, and analyzing the established model to obtain the first driving behavior analysis result; or

Analyze the vehicle information and the surrounding environment information based on artificial intelligence technology to obtain the first driving behavior analysis result; or

Modeling according to the vehicle information and the surrounding environment information, analyzing the established model, and analyzing the vehicle information and the surrounding environment information based on artificial intelligence technology. The results of the first driving behavior analysis are described.
The driving behavior analysis method according to claim 5, wherein the step of analyzing the vehicle information and the surrounding environment information based on artificial intelligence technology comprises:

Based on the vehicle information and the surrounding environment information, a driving behavior analysis model is used to analyze the driving behavior of the driver, wherein the driving behavior analysis model is obtained by training based on a deep learning algorithm.
The driving behavior analysis method according to claim 1, further comprising:

Analyze the driving state of the driver to obtain a second driving behavior analysis result for the driving state, and / or,

Analyze whether driving behavior violates traffic regulations based on the vehicle information and the surrounding environment information to obtain a third driving behavior analysis result; and

Based on the first driving behavior analysis result, the second driving behavior analysis result, and / or the third driving behavior analysis result, a current driving behavior analysis result is obtained.
The driving behavior analysis method according to claim 7, further comprising:

Identifying the identity of the driver; and

An analysis result of historical driving behavior corresponding to the driver is obtained based on the recognition result.
The driving behavior analysis method according to claim 8, further comprising:

Based on the current driving behavior analysis result and the historical driving behavior analysis result, a total behavior analysis result of the driver is determined.
The driving behavior analysis method according to claim 8, further comprising:

According to the vehicle information and the surrounding environment information, re-analyze the driving behavior in the previous predetermined time to obtain a historical driving behavior analysis and calibration result.
The driving behavior analysis method according to claim 10, further comprising:

Based on the current driving behavior analysis result, the historical driving behavior analysis calibration result, and the historical driving behavior analysis result, a total behavior analysis result of the driver is determined.
The driving behavior analysis method according to claim 11, further comprising:

The driver is scored according to the overall behavior analysis result, so as to provide the driver with a corresponding service according to the score.
A driving behavior analysis device, comprising:

A first acquisition module for acquiring vehicle information and surrounding environment information of the vehicle during driving; and

A first analysis module is configured to analyze the vehicle information and the surrounding environment information based on data modeling and / or artificial intelligence to obtain a first driving behavior analysis result for a driver of the vehicle.
A computing device includes:

Processor; and

The memory stores executable code thereon, and when the executable code is executed by the processor, causes the processor to execute the method according to any one of claims 1-12.
A non-transitory machine-readable storage medium having executable code stored thereon, when the executable code is executed by a processor of an electronic device, causing the processor to execute any one of claims 1 to 12 The method described.