CN111267863A - Driver driving type identification method and device, storage medium and terminal equipment - Google Patents
Driver driving type identification method and device, storage medium and terminal equipment Download PDFInfo
- Publication number
- CN111267863A CN111267863A CN201811476975.6A CN201811476975A CN111267863A CN 111267863 A CN111267863 A CN 111267863A CN 201811476975 A CN201811476975 A CN 201811476975A CN 111267863 A CN111267863 A CN 111267863A
- Authority
- CN
- China
- Prior art keywords
- driving type
- driver
- driving
- identification
- scheme
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
-
- 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)
Abstract
The invention discloses a method and a device for identifying the driving type of a driver, a storage medium and a terminal device, wherein the method comprises the following steps: acquiring initial data of driving type identification of a driver; acquiring an initial driving type recognition result of the driver according to at least two preset recognition schemes of a first recognition scheme, a second recognition scheme, a third recognition scheme and a fourth recognition scheme; the first identification scheme is used for identifying the driving type according to dynamic information of the vehicle, the second identification scheme is used for identifying the driving type according to operation information of a driver under a preset working condition, the third identification scheme is used for identifying the driving type according to a pre-trained machine learning network, and the fourth identification scheme is used for identifying the driving type according to a pre-trained deep learning network; and acquiring the current driving type of the driver according to the initial data and the initial driving type identification result. The method and the device can improve the accuracy and the real-time performance of the driver driving type identification.
Description
Technical Field
The invention relates to the technical field of driver driving type identification, in particular to a driver driving type identification method and device, a computer readable storage medium and a terminal device.
Background
The prior art provides a method for identifying the driving type of a driver, which uses a vehicle signal-based identification method and a specific operating condition-based identification method. The identification method based on the vehicle signals adopts fewer vehicle signals to identify the driving type of the driver, the parameters required to be acquired are simple, the requirement on hardware of a processor is not high, and the identification result with higher accuracy is difficult to obtain; the identification method of the operation behavior based on the specific working condition adds the operation input of the driver as one of the judgment conditions, needs to set the identification working condition, and can identify the driving type of the driver according to the corresponding operation behavior only when the specific working condition is triggered, so that better real-time performance is difficult to obtain.
Disclosure of Invention
The technical problem to be solved by the embodiments of the present invention is to provide a method and an apparatus for identifying a driver driving type, a computer-readable storage medium, and a terminal device, which can improve the accuracy and real-time performance of identifying the driver driving type.
In order to solve the above technical problem, an embodiment of the present invention provides a method for identifying a driving type of a driver, including:
acquiring initial data of driving type identification of a driver; wherein the initial data includes driving type recognition results and driving behavior data of a history corresponding to the driver;
acquiring an initial driving type recognition result of the driver according to at least two preset recognition schemes of a first recognition scheme, a second recognition scheme, a third recognition scheme and a fourth recognition scheme; the method comprises the steps that a first identification scheme is used for identifying a driving type according to dynamic information of a vehicle, a second identification scheme is used for identifying the driving type according to operation information of a driver under a preset working condition, a third identification scheme is used for identifying the driving type according to a pre-trained machine learning network, and a fourth identification scheme is used for identifying the driving type according to a pre-trained deep learning network;
and acquiring the current driving type of the driver according to the initial data and the initial driving type identification result.
Further, before the obtaining initial data of driving type identification of the driver, the method further comprises:
identifying the identity of the driver;
the acquiring of the initial data of the driving type identification of the driver specifically includes:
and acquiring initial data of driving type identification corresponding to the identity of the driver according to a preset driving behavior database.
Further, the acquiring an initial driving type recognition result of the driver according to at least one of a first recognition scheme, a second recognition scheme, a third recognition scheme and a fourth recognition scheme, which are preset, specifically includes:
when a first preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme and the second recognition scheme;
when a second preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme and the third recognition scheme;
and when a third preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme, the third recognition scheme and the fourth recognition scheme.
Further, the method acquires the initial driving type recognition result of the driver according to the first recognition scheme and the second recognition scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
taking the first driving type and the second driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type and the second driving type according to the initial data respectively;
and acquiring the current driving type of the driver according to the corrected first driving type and the corrected second driving type.
Further, the respectively correcting the first driving style and the second driving style according to the initial data specifically includes:
wherein, y1Representing said modified first driving style, y1(k1) Representing said first driving style, y1(0)、y1(1)、···、y1(k1-1) corresponding, k, from said initial data1Is the number of identification times, p, corresponding to the first identification scheme1Is the weight coefficient corresponding to the first identification scheme, 0<p1<1;y2Indicating said modified second driving style, y2(k2) Representing said second driving style, y2(0)、y2(1)、···、y2(k2-1) corresponding, k, from said initial data2The number of identification times, p, corresponding to the second identification scheme2Is the weight coefficient corresponding to the second identification scheme, 0<p2<1;
The obtaining the current driving style of the driver according to the modified first driving style and the modified second driving style specifically includes:
according to the formula Y ═ Y1·q1+y2·q2Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2And the comprehensive weight coefficient is corresponding to the second identification scheme.
Further, the air conditioner is provided with a fan,b represents the triggering times of the preset working condition; l represents the current mileage of the vehicle.
Further, the method acquires the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme and the third recognition scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
taking the first driving type, the second driving type and the third driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type, the second driving type and the third driving type according to the initial data;
and acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type and the corrected third driving type.
Further, the acquiring the current driving style of the driver according to the modified first driving style, the modified second driving style and the modified third driving style specifically includes:
according to the formula Y ═ Y1·q1+y2·q2+y3·q3Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, Y1Representing said modified first driving style, y2Indicating said modified second driving style, y3Representing said modified third driving style, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3And the comprehensive weight coefficient is corresponding to the third identification scheme.
Further, the air conditioner is provided with a fan,q2=0.2×1.01b,b represents the triggering times of the preset working condition; l represents the current mileage of the vehicle.
Further, the method acquires the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme, the third recognition scheme and the fourth recognition scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
acquiring a fourth driving type of the driver according to a cloud server; the cloud server acquires the fourth driving type through a pre-trained deep learning network according to second operation information of the driver on the vehicle and second dynamic response information correspondingly generated by the vehicle in a second preset time period or a second preset driving mileage;
taking the first driving type, the second driving type, the third driving type and the fourth driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type, the second driving type, the third driving type and the fourth driving type according to the initial data;
and acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type, the corrected third driving type and the corrected fourth driving type.
Further, the acquiring the current driving style of the driver according to the modified first driving style, the modified second driving style, the modified third driving style and the modified fourth driving style specifically includes:
according to the formula Y ═ Y2·q2+y3·q3+y4·q4Acquiring the current driving type of the driver; wherein Y represents the driverCurrent driving type of y2Indicating said modified second driving style, y3Indicating said modified third driving style, y4Representing said modified fourth driving style, q2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3Is the comprehensive weight coefficient, q, corresponding to the third identification scheme4The comprehensive weight coefficient corresponding to the fourth identification scheme;
and checking the current driving type of the driver according to the corrected first driving type.
Further, q is2=0.15×1.01b,q3=0.5-0.15×1.01b,q40.5; and b represents the triggering times of the preset working condition.
Further, the checking the current driving style of the driver according to the corrected first driving style specifically includes:
comparing the modified first driving style with the current driving style of the driver;
and when the deviation between the corrected first driving type and the current driving type of the driver exceeds a preset threshold value, sending fault reminding information to the cloud server.
In order to solve the above technical problem, an embodiment of the present invention further provides a device for identifying a driving type of a driver, including:
the initial data acquisition module is used for acquiring initial data for identifying the driving type of the driver; wherein the initial data includes driving type recognition results and driving behavior data of a history corresponding to the driver;
the initial driving type recognition module is used for acquiring an initial driving type recognition result of the driver according to at least two recognition schemes of a first preset recognition scheme, a second preset recognition scheme, a third preset recognition scheme and a fourth preset recognition scheme; the method comprises the steps that a first identification scheme is used for identifying a driving type according to dynamic information of a vehicle, a second identification scheme is used for identifying the driving type according to operation information of a driver under a preset working condition, a third identification scheme is used for identifying the driving type according to a pre-trained machine learning network, and a fourth identification scheme is used for identifying the driving type according to a pre-trained deep learning network; and the number of the first and second groups,
and the current driving type identification module is used for acquiring the current driving type of the driver according to the initial data and the initial driving type identification result.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program; wherein the computer program, when running, controls the device on which the computer-readable storage medium is located to perform any of the above-described driver driving type identification methods.
The embodiment of the invention also provides a terminal device, which comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein the processor realizes the driver driving type identification method in any one of the above items when executing the computer program.
Compared with the prior art, the embodiment of the invention provides a method and a device for identifying the driving type of a driver, a computer-readable storage medium and a terminal device, the initial driving type recognition result of the driver is obtained by adopting at least two of four recognition schemes of recognizing the driving type according to the dynamic information of the vehicle, recognizing the driving type according to the operation information of the driver under the preset working condition, recognizing the driving type according to the pre-trained machine learning network and recognizing the driving type according to the pre-trained deep learning network, the current driving type of the driver is identified by combining the acquired initial data of the driving type identification and the initial driving type identification result, so that the problems of low accuracy and poor real-time performance of the identification result caused by adopting a single identification scheme in the prior art can be solved, and the accuracy and the real-time performance of the driving type identification of the driver are improved.
Drawings
FIG. 1 is a flow chart of a preferred embodiment of a driver driving type identification method provided by the present invention;
FIG. 2 is a detailed flowchart of a preferred embodiment of step S12 of a method for identifying a driver' S driving style according to the present invention;
FIG. 3 is a block diagram of a preferred embodiment of a driver driving type recognition apparatus according to the present invention;
fig. 4 is a block diagram of a preferred embodiment of a terminal device provided in the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without any inventive step, are within the scope of the present invention.
Referring to fig. 1, it is a flowchart of a preferred embodiment of a method for identifying a driving type of a driver according to the present invention, and the method includes steps S11 to S13:
step S11, acquiring initial data of driving type identification of the driver; wherein the initial data includes driving type recognition results and driving behavior data of a history corresponding to the driver;
step S12, acquiring an initial driving type recognition result of the driver according to at least two preset recognition schemes of a first recognition scheme, a second recognition scheme, a third recognition scheme and a fourth recognition scheme; the method comprises the steps that a first identification scheme is used for identifying a driving type according to dynamic information of a vehicle, a second identification scheme is used for identifying the driving type according to operation information of a driver under a preset working condition, a third identification scheme is used for identifying the driving type according to a pre-trained machine learning network, and a fourth identification scheme is used for identifying the driving type according to a pre-trained deep learning network;
and step S13, acquiring the current driving type of the driver according to the initial data and the initial driving type recognition result.
Specifically, four driving type recognition schemes are preset, wherein the first recognition scheme is used for recognizing the driving type according to dynamic information of a vehicle, the second recognition scheme is used for recognizing the driving type according to operation information of a driver under a preset working condition, the third recognition scheme is used for recognizing the driving type according to a pre-trained machine learning network, the fourth recognition scheme is used for recognizing the driving type according to a pre-trained deep learning network, and each driving type recognition result of the driver and corresponding driving behavior data are recorded and stored in a cloud server or a vehicle end; when the driving type of the driver needs to be identified, initial data for identifying the driving type of the driver is obtained from a cloud server or a vehicle end, the initial data comprises historical driving type identification results and driving behavior data, at least two identification schemes of four driving type identification schemes are adopted to obtain the initial driving type identification results of the driver, and therefore comprehensive weighting is carried out according to the obtained initial data and the initial driving type identification results, and the current driving type of the driver is obtained.
It should be noted that the driving types of the drivers may be divided according to the driving style and the driving ability, wherein the driving types may be divided into 3 driving types (e.g., aggressive, medium-sized, and robust) or 5 driving types (e.g., aggressive, more aggressive, medium-sized, more robust, and robust), the driving types may be divided into 3 driving types (e.g., professional, general, and novice) or 5 driving types (e.g., professional, more professional, general, more novice, and novice) according to the driving ability, and the two classification methods are combined to obtain a driving type matrix of 3 × 3 or 5 × 5, where each element in the matrix has a corresponding driving style attribute and driving ability attribute, such as that the driving type of a certain driver is a more aggressive general driver, or a robust professional driver, and the like.
According to the method for identifying the driving type of the driver, provided by the embodiment of the invention, the initial driving type identification result of the driver is obtained by adopting at least two identification schemes of the driving type identification scheme according to the dynamic information of the vehicle, the driving type identification scheme according to the operation information of the driver under the preset working condition, the driving type identification scheme according to the pre-trained machine learning network and the driving type identification scheme according to the pre-trained deep learning network, and the comprehensive weighting processing is carried out by combining the obtained initial data of the driving type identification scheme and the initial driving type identification result, so that the current driving type of the driver is identified, the problems of low accuracy and poor real-time performance of the identification result caused by adopting a single identification scheme in the prior art can be solved, and the accuracy and the real-time performance of the driving type identification of the driver are improved.
In another preferred embodiment, before the obtaining initial data of driving type identification of the driver, the method further comprises:
identifying the identity of the driver;
the acquiring of the initial data of the driving type identification of the driver specifically includes:
and acquiring initial data of driving type identification corresponding to the identity of the driver according to a preset driving behavior database.
It can be understood that, since there may be a plurality of drivers driving the vehicle, when identifying the driving type of the driver, the identity of the driver needs to be confirmed first, and after confirming the identity of the driver, the initial data of the driving type identification corresponding to the identity of the driver can be obtained from the cloud server or the driving behavior database preset at the vehicle end.
In specific implementation, after a driver enters a vehicle, the method for confirming the identity of the driver comprises a manual selection method and an automatic selection method; in the manual selection method, a driver selects information such as a name or a number corresponding to the driver in a vehicle through a key, a knob, a touch screen and other human-computer interaction parts, so that the identity of the driver is confirmed; the automatic selection method comprises the steps that a camera in a vehicle captures the face of a driver, the identity of the driver is identified through facial features or iris features and the like, or the identity of the driver is confirmed through fingerprint feature collection and matching; after the identity of the driver is recognized, the latest driving type recognition result and the corresponding driving behavior data are called from a driving behavior database at the cloud end or the vehicle end and serve as initial data of the current driver type recognition.
Referring to fig. 2, it is a specific flowchart of a preferred embodiment of step S12 of a method for identifying a driving type of a driver according to the present invention, where the obtaining of an initial driving type identification result of the driver according to at least two of a first identification scheme, a second identification scheme, a third identification scheme, and a fourth identification scheme includes steps S1201 to S1203:
step S1201, when a first preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme and the second recognition scheme;
step S1202, when a second preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme and the third recognition scheme;
and step S1203, when a third preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme, the third recognition scheme and the fourth recognition scheme.
Specifically, when at least two of four driving type identification schemes are selected to obtain an initial driving type identification result of a driver, different identification schemes need to be selected according to different conditions, and when a first preset condition is met, a first identification scheme and a second identification scheme are selected, wherein the first preset condition is that no good vehicle end processor and no good memory exist; when a second preset condition is met, selecting a first identification scheme, a second identification scheme and a third identification scheme, wherein the second preset condition is that a good vehicle-end processor and a good memory are available, but a cloud-end processor and a cloud-end memory are not available; and when a third preset condition is met, selecting a first identification scheme, a second identification scheme, a third identification scheme and a fourth identification scheme, wherein the third preset condition comprises that the vehicle-end processor and the memory are good, and the cloud-end processor and the memory are good.
In yet another preferred embodiment, the method obtains the initial driving-type recognition result of the driver according to the first recognition scheme and the second recognition scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
taking the first driving type and the second driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type and the second driving type according to the initial data respectively;
and acquiring the current driving type of the driver according to the corrected first driving type and the corrected second driving type.
Specifically, with reference to the above embodiment, when a first preset condition is met, that is, when there is no good vehicle-side processor or memory, a first driving type of the driver is obtained according to the collected dynamic information of the vehicle, and a second driving type of the driver is obtained according to the operation information of the driver on the vehicle under a preset working condition and the dynamic response information correspondingly generated by the vehicle, where the obtained first driving type and the obtained second driving type are corresponding initial driving type identification results; and respectively correcting the first driving type and the second driving type according to the acquired initial data, so that comprehensive weighting is performed according to the corrected first driving type and the corrected second driving type, and the current driving type of the driver is obtained.
In specific implementation, longitudinal acceleration information of the vehicle is collected, mathematical statistical analysis is performed according to characteristic information such as longitudinal acceleration change rate, peak occurrence frequency, peak absolute value size, peak-to-valley absolute difference and the like under a certain driving mileage of the vehicle, statistical parameters of dynamic characteristics of the vehicle are obtained, such as unit mileage longitudinal acceleration peak occurrence frequency, peak slope probability distribution, longitudinal acceleration variance and the like, and therefore the first driving type of the driver is obtained according to the statistical parameters.
In general, the more proficient the driver or the smoother the driver with a driving style tending to be stable, the smaller the change of the generated longitudinal acceleration, and the relatively smaller the number of peaks and absolute values correspondingly appeared; the more the operation of a novice driver or a driver with a driving style tending to be aggressive, the more the change of the generated longitudinal acceleration is, the more the longitudinal acceleration is changed, the longitudinal acceleration peak value is easy to appear, the shape of the peak value is sharp, and the absolute value is relatively large.
It should be noted that the longitudinal acceleration information of the vehicle can be directly obtained by the vehicle acceleration sensor or obtained by differentiating the vehicle speed information, and the identification algorithm is relatively simple, and the requirements on the calculated speed of the vehicle-side processor and the data memory are not high.
When the vehicle is identified to enter the preset working condition, acquiring operation action information of a driver on the vehicle under the current working condition, including a steering wheel angle, an accelerator pedal opening, master cylinder pressure, gear information and the like, and dynamic response information correspondingly generated by the vehicle, including longitudinal acceleration, lateral acceleration, yaw velocity and the like, extracting characteristic parameters, correspondingly setting a multi-stage threshold value of vehicle dynamic response information change under different working conditions, comparing the extracted characteristic parameters with the multi-stage threshold value step by step, and finding a threshold value point with the minimum deviation degree as an identification result of the driving type of the driver, namely obtaining a second driving type; for example, the confidence of the driver's excitement when classified by the driving style is from 0 to 100%, and can be divided into 3 thresholds, which are 0 to 33%, 33 to 66%, and 66 to 100%, respectively, and if the confidence of the currently recognized excitement of the driver is 56%, it corresponds to the 2 nd level, i.e., a medium-sized driver; if the confidence of the aggressiveness of the currently identified driver is 80%, it corresponds to a level 3, i.e., aggressive, driver.
Under the general condition, the more skilled drivers or the drivers with driving styles tending to be steady can better control the vehicle state, so that the vehicle state under the specific working conditions changes smoothly, for example, when the driver drives on a curve, the steering wheel operation of the skilled drivers can be operated in place quickly, the vehicle speed is controlled through an accelerator to control insufficient or excessive steering, the correction of the steering wheel is less, the change of the longitudinal acceleration and the lateral acceleration of the vehicle caused by the change is uniform, and the posture of the vehicle body shakes less; however, the more the novice driver or the more aggressive the driving style, the worse the behavior of the vehicle state controlled by the driver, the larger the fluctuation of the vehicle state under the specific working condition, for example, when driving on a curve, the novice driver often has difficulty in controlling the proper vehicle speed to pass through the curve, and needs to use the steering wheel operation to correct the yaw angle and yaw rate of the vehicle, because the coordination of the vehicle speed and the steering wheel operation is not good, the fluctuation of the longitudinal and lateral acceleration of the vehicle is larger, and the larger the vehicle posture fluctuation is caused.
It should be noted that, under the acceleration condition, the opening degree of an accelerator pedal and the longitudinal acceleration of a vehicle are main judgment bases for identifying the type of a driver, and gear information is an auxiliary judgment base; under the braking working condition, the pressure of a brake master cylinder and the longitudinal acceleration of a vehicle are main judgment basis for identifying the type of a driver, and gear information is auxiliary judgment basis; under the steering working condition, the steering wheel angle, the vehicle speed, the lateral acceleration and the yaw velocity are main judgment bases for identifying the type of a driver, and the opening degree of an accelerator pedal, the pressure of a brake master cylinder and gear information are auxiliary judgment bases; under the working condition of constant-speed cruising, the opening degree of an accelerator pedal and the speed of the vehicle are main judgment bases for identifying the type of a driver.
For setting a multi-level threshold value of the vehicle speed dynamic response signal change under different working conditions, taking the driving style of the constant-speed cruising working condition as an example for explanation, if a threshold value vector [0.8, 0.9, 1.0, 1.1, 1.2] is set to be 5 levels of threshold values, which respectively correspond to 5 different driving styles [ robust, more robust, medium-sized, more aggressive, aggressive ], the ratio of the vehicle speed to the current road speed limit is 1.04 on the assumption that the current vehicle speed is 83km/h, the current road speed limit is 80km/h, the vector obtained by comparing with the threshold values of all levels is [0.24, 0.14, 0.04, -0.06, -0.16], and the threshold value with the minimum deviation is a level-3 threshold value, the driving style type of the driver at the current moment is judged to be 'medium'.
As a preferable scheme, the respectively correcting the first driving style and the second driving style according to the initial data specifically includes:
wherein, y1Representing said modified first driving style, y1(k1) Representing said first driving style, y1(0)、y1(1)、···、y1(k1-1) corresponding, k, from said initial data1Is the number of identification times, p, corresponding to the first identification scheme1Is the weight coefficient corresponding to the first identification scheme, 0<p1<1;y2Indicating said modified second driving style, y2(k2) Representing said second driving style, y2(0)、y2(1)、···、y2(k2-1) corresponding, k, from said initial data2The number of identification times, p, corresponding to the second identification scheme2Is the weight coefficient corresponding to the second identification scheme, 0<p2<1;
The obtaining the current driving style of the driver according to the modified first driving style and the modified second driving style specifically includes:
according to the formula Y ═ Y1·q1+y2·q2Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2And the comprehensive weight coefficient is corresponding to the second identification scheme.
Specifically, after the first driving type and the second driving type are obtained, the first driving type needs to be further corrected and updated according to the driving type recognition result and the driving behavior data of the historical record of the driver corresponding to the first recognition scheme, the second driving type needs to be corrected and updated according to the driving type recognition result and the driving behavior data of the historical record of the driver corresponding to the second recognition scheme, the corrected first driving type is used as the currently recognized first driving type (and is correspondingly stored in the driving behavior database at the cloud server or the vehicle end), the corrected second driving type is used as the currently recognized second driving type (and is correspondingly stored in the driving behavior database at the cloud server or the vehicle end), and therefore Y is obtained according to the formula Y1·q1+y2·q2And carrying out weighted integration to obtain the current driving type of the driver.
In the embodiment, the first driving type is modified, and the kth recognition scheme is assumed to be adopted currently1Secondary driving type recognition, the corresponding driving type recognition result of the history includes y1(0)、y1(1)、···、y1(k1-1), wherein y1(0) For reference driving type recognition result obtained from preset initial data, y1(1)、···、y1(k1-1) are each the corresponding corrected first driving style obtained, k1The recognition result of the sub-driving type recognition needs to be the reference driving type recognition result y1(0) And all k1Secondary driving type recognition result y1(1)、···、y1(k1) As parameters, and each parameter is assigned a corresponding weight coefficient (p)1Values may be set and adjusted appropriately at initialization and thereafter remain unchanged), the resulting modified first driving style beingThe same applies to the correction of the second driving style, and is not described here again.
It should be noted that, because the update cycles of the driving type identifications corresponding to different identification schemes are different, the identification times corresponding to different identification schemes may be different in the same driving type identification.
In addition, the corrected driving type identification result is obtained by weighting based on the reference driving type identification result and all the currently obtained driving type identification results, the more recent driving type identification result is assigned with higher weight, and the more recent driving type identification result is assigned with lower weight, so that the possible oscillation of the driving type identification result can be reduced, the influence of the historical driving type identification result can be gradually eliminated, and the latest driving type identification result is taken as the standard.
As a preferred embodiment, it is possible to,b represents the triggering times of the preset working condition; l represents the current mileage of the vehicle.
It should be noted that whenAnd then, taking the comprehensive weight coefficient corresponding to the second identification scheme as 1, and taking the comprehensive weight coefficient corresponding to the first identification scheme as 0.
In yet another preferred embodiment, the method obtains the initial driving style recognition result of the driver according to the first recognition scheme, the second recognition scheme, and the third recognition scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
taking the first driving type, the second driving type and the third driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type, the second driving type and the third driving type according to the initial data;
and acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type and the corrected third driving type.
Specifically, in this embodiment, the method for acquiring the first driving style and the second driving style of the driver according to the first identification scheme and the second identification scheme is the same as that in the above embodiment, and is not described again here; for acquiring a third driving type of the driver according to a third identification scheme, acquiring first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage, wherein the first operation information comprises characteristic parameters such as a steering wheel angle, an accelerator pedal opening degree, a brake master cylinder pressure, a vehicle speed, a longitudinal acceleration, a lateral acceleration and a yaw angular velocity, and the like, and inputting the acquired characteristic parameters into a machine learning network trained in advance to acquire the third driving type of the driver; and respectively correcting the first driving type, the second driving type and the third driving type according to the acquired initial data, so as to comprehensively weight according to the corrected first driving type, the corrected second driving type and the corrected third driving type, and obtain the current driving type of the driver.
It should be noted that, when a driver drives a vehicle, a controller at a vehicle end collects operation information of the driver and dynamic response information of the vehicle, segments data according to mileage and time, extracts statistical characteristic parameters from each segment of data, and inputs the data as an initial training sample into a plurality of machine learning networks in the vehicle to perform forward calculation at the same time, where the result of an input end is a confidence coefficient of matching of each machine learning network under current input, and when an input value is higher, that is, a confidence coefficient is higher, it indicates that the driving type of the current driver is closer to the driving type of the initial training sample trained by the machine learning network, that is, it is recognized that the driving type of the current driver is more consistent with the driving type provided by the initial training sample corresponding to the machine learning network.
As a preferable scheme, the acquiring the current driving style of the driver according to the modified first driving style, the modified second driving style and the modified third driving style specifically includes:
according to the formula Y ═ Y1·q1+y2·q2+y3·q3Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, Y1Representing said modified first driving style, y2Indicating said modified second driving style, y3Representing said modified third driving style, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3And the comprehensive weight coefficient is corresponding to the third identification scheme.
In particular, according to the formulaModifying the first driving style according to the formulaModifying the second driving style according to the formulaCorrecting the third driving style, and updating the correction in detailThe process is the same as the theory of the above embodiment, and is not described again here; after obtaining the corrected first driving style y1Corrected second driving style y2And a modified third driving style y3(and correspondingly storing the driving behavior into a cloud server or a driving behavior database at the vehicle end) and then according to the formula Y, changing the formula Y into Y1·q1+y2·q2+y3·q3And performing weighted integration to obtain the current driving type of the driver.
As a preferred embodiment, it is possible to,q2=0.2×1.01b,b represents the triggering times of the preset working condition; l represents the current mileage of the vehicle.
It should be noted that whenWhen the first identification scheme is used, the comprehensive weight coefficient corresponding to the first identification scheme is taken as 0.3, and the comprehensive weight coefficient corresponding to the scheme is not updated; when q is2=0.2×1.01bAnd when the weight value is more than or equal to 0.3, taking the comprehensive weight value coefficient corresponding to the second identification scheme as 0.3, and not updating the comprehensive weight value coefficient corresponding to the scheme.
In yet another preferred embodiment, the method obtains the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme, the third recognition scheme, and the fourth recognition scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
acquiring a fourth driving type of the driver according to a cloud server; the cloud server acquires the fourth driving type through a pre-trained deep learning network according to second operation information of the driver on the vehicle and second dynamic response information correspondingly generated by the vehicle in a second preset time period or a second preset driving mileage;
taking the first driving type, the second driving type, the third driving type and the fourth driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type, the second driving type, the third driving type and the fourth driving type according to the initial data;
and acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type, the corrected third driving type and the corrected fourth driving type.
In this embodiment, the method for obtaining the first driving style, the second driving style and the third driving style of the driver according to the first identification scheme, the second identification scheme and the third identification scheme respectively is the same as the above embodiment, and is not repeated herein; for the fourth driving type of the driver obtained according to the fourth recognition scheme, when the driver drives the vehicle, after a certain driving range or time is met, the controller at the vehicle end collects operation information of the driver and vehicle dynamic response information including data of a steering wheel angle, an accelerator opening degree, a brake master cylinder pressure, a vehicle speed, a longitudinal acceleration, a lateral acceleration, a yaw angular velocity and the like, uploads the collected data to a cloud server through a communication network or other channels, the data are stored by a cloud storage, characteristic parameters of the information are extracted by the cloud server, the characteristic parameters are input into a plurality of pre-trained deep learning networks (such as neural networks) which are registered at the cloud end and have relatively stable typical or information data, and output results of different types of neural networks, namely the fourth driving type, are obtained through forward calculation, including driving ability and driving style; and respectively correcting the first driving type, the second driving type, the third driving type and the fourth driving type according to the acquired initial data, so as to comprehensively weight according to the corrected first driving type, the corrected second driving type, the corrected third driving type and the corrected fourth driving type, and obtain the current driving type of the driver.
As a preferable scheme, the acquiring the current driving style of the driver according to the modified first driving style, the modified second driving style, the modified third driving style and the modified fourth driving style specifically includes:
according to the formula Y ═ Y2·q2+y3·q3+y4·q4Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, Y2Indicating said modified second driving style, y3Indicating said modified third driving style, y4Representing said modified fourth driving style, q2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3Is the comprehensive weight coefficient, q, corresponding to the third identification scheme4The comprehensive weight coefficient corresponding to the fourth identification scheme;
and checking the current driving type of the driver according to the corrected first driving type.
In particular, according to the formulaModifying the first driving style according to the formulaModifying the second driving style according to the formulaCorrecting the third driving style according to the formulaCorrecting the fourth driving type, wherein the detailed correction and update process is the same as the theory of the embodiment, and is not described again; after obtaining the corrected first driving style y1Corrected second driving style y2And the corrected third driving style y3And a fourth driving style y after correction4(and correspondingly storing the driving behavior into a cloud server or a driving behavior database at the vehicle end) and then according to the formula Y, changing the formula Y into Y2·q2+y3·q3+y4·q4And performing weighted integration to obtain the current driving type of the driver, and checking the obtained current driving type of the driver according to the corrected first driving type.
Preferably, q is2=0.15×1.01b,q3=0.5-0.15×1.01b,q40.5; and b represents the triggering times of the preset working condition.
When q is equal to2=0.15×1.01bAnd when the value is more than or equal to 0.3, taking the comprehensive weight coefficient corresponding to the second identification scheme as 0.3, taking the comprehensive weight coefficient corresponding to the third identification scheme as 0.2, and not updating the comprehensive weight coefficient corresponding to the identification scheme.
Preferably, the checking the current driving style of the driver according to the corrected first driving style specifically includes:
comparing the modified first driving style with the current driving style of the driver;
and when the deviation between the corrected first driving type and the current driving type of the driver exceeds a preset threshold value, sending fault reminding information to the cloud server.
It should be noted that the modified first driving type corresponding to the first identification scheme is used as a self-checking signal of other identification schemes, for example, when the current driving mileage of the vehicle exceeds 50km, the modified first driving type is compared with the current driving type of the driver, if the deviation exceeds a preset threshold, the fault reminding information is sent to the cloud server for prompting whether a signal receiving and transmitting unit at the vehicle end and a data storage unit at the cloud end have a fault, and if the identification result indicates that the fault exists, the maintenance reminding information is sent to a background operator; and if the identification result is non-fault, calling the travel mileage data with the deviation exceeding the threshold value of the identification result, checking other information such as a route, weather, traffic conditions, the mental fatigue degree of a driver and the like, matching the travel mileage data with a small deviation, extracting main difference items and sending the main difference items to a deep learning network operation engineer, and determining whether the difference points are used as input nodes of the deep learning network or are ignored by the engineer.
In other embodiments, after the driving type recognition result of a certain driver is obtained according to the comprehensive weighting scheme, the driving type recognition results of the driver stored at the vehicle end and the cloud end are updated, the driving process is continuously updated until the vehicle is completely static and kept for a period of time, and the latest driving type recognition results at the vehicle end and the cloud end are read as the reference values for subsequent driving type recognition and updating after the vehicle is started next time.
For identifying the driving type according to the dynamic information of the vehicle, the parameters needing to be obtained are simple, the requirement on hardware of a processor is not high, but the identification result with higher accuracy is difficult to obtain; for identifying the driving type according to the operation information of the driver under the preset working condition, better instantaneity is difficult to obtain; for recognizing the driving types according to the pre-trained machine learning network, the machine learning level is shallow due to the influence of the arithmetic capability of the processor, and the nuances among different driving types are difficult to distinguish; for recognizing the driving type according to the pre-trained deep learning network, massive data are needed for training, and when the operation mileage of a driver of a corresponding vehicle is insufficient, a good recognition result cannot be obtained; therefore, each identification scheme has certain limitation when being used independently; according to the method for identifying the driving type of the driver, provided by the embodiment of the invention, different identification schemes are combined under different conditions, the driving type of the driver is identified by using a comprehensive weighting mode, the limitation caused by the adoption of a single identification scheme is avoided, and the accuracy and the real-time performance of identifying the driving type of the driver are improved.
The embodiment of the present invention further provides a device for identifying a driver driving type, which can implement all processes of the method for identifying a driver driving type described in any one of the above embodiments, and the functions and technical effects of the modules and units in the device are respectively the same as those of the method for identifying a driver driving type described in the above embodiment, and are not described herein again.
Referring to fig. 3, it is a block diagram of a preferred embodiment of a driving type recognition device for a driver according to the present invention, the device includes:
an initial data acquisition module 11, configured to acquire initial data for identifying a driving type of a driver; wherein the initial data includes driving type recognition results and driving behavior data of a history corresponding to the driver;
the initial driving type recognition module 12 is configured to obtain an initial driving type recognition result of the driver according to at least two recognition schemes of a first preset recognition scheme, a second preset recognition scheme, a third preset recognition scheme and a fourth preset recognition scheme; the method comprises the steps that a first identification scheme is used for identifying a driving type according to dynamic information of a vehicle, a second identification scheme is used for identifying the driving type according to operation information of a driver under a preset working condition, a third identification scheme is used for identifying the driving type according to a pre-trained machine learning network, and a fourth identification scheme is used for identifying the driving type according to a pre-trained deep learning network; and the number of the first and second groups,
and a current driving type identification module 13, configured to obtain a current driving type of the driver according to the initial data and the initial driving type identification result.
Preferably, the apparatus further comprises:
the identity recognition module is used for recognizing the identity of the driver;
the initial data obtaining module specifically includes:
and the initial data acquisition unit is used for acquiring the driving type identification initial data corresponding to the identity of the driver according to a preset driving behavior database.
Preferably, the initial driving type identification module specifically includes:
the first initial driving type recognition unit is used for acquiring the initial driving type recognition result of the driver according to the first recognition scheme and the second recognition scheme when a first preset condition is met;
the second initial driving type recognition unit is used for acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme and the third recognition scheme when a second preset condition is met; and the number of the first and second groups,
and the third initial driving type recognition unit is used for acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme, the third recognition scheme and the fourth recognition scheme when a third preset condition is met.
Preferably, the first initial driving type identification unit is specifically configured to:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
taking the first driving type and the second driving type as the initial driving type recognition result;
the current driving type identification module specifically includes:
the first driving type correction unit is used for correcting the first driving type and the second driving type according to the initial data; and the number of the first and second groups,
and the first current driving type identification unit is used for acquiring the current driving type of the driver according to the corrected first driving type and the corrected second driving type.
Preferably, the first driving style correction unit is specifically configured to:
wherein, y1Representing said modified first driving style, y1(k1) Representing said first driving style, y1(0)、y1(1)、···、y1(k1-1) corresponding, k, from said initial data1Is the number of identification times, p, corresponding to the first identification scheme1Is the weight coefficient corresponding to the first identification scheme, 0<p1<1;y2Indicating said modified second driving style, y2(k2) Representing said second driving style, y2(0)、y2(1)、···、y2(k2-1) corresponding, k, from said initial data2The number of identification times, p, corresponding to the second identification scheme2Is the weight coefficient corresponding to the second identification scheme, 0<p2<1;
The first current driving type identification unit is specifically configured to:
according to the formula Y ═ Y1·q1+y2·q2Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2And the comprehensive weight coefficient is corresponding to the second identification scheme.
Preferably, the first and second electrodes are formed of a metal,b represents the triggering times of the preset working condition; l represents the current mileage of the vehicle.
Preferably, the second initial driving type identification unit is specifically configured to:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
taking the first driving type, the second driving type and the third driving type as the initial driving type recognition result;
the current driving type identification module specifically includes:
the second driving type correction unit is used for correcting the first driving type, the second driving type and the third driving type according to the initial data; and the number of the first and second groups,
and the second current driving type identification unit is used for acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type and the corrected third driving type.
Preferably, the second current driving type identifying unit is specifically configured to:
according to the formula Y ═ Y1·q1+y2·q2+y3·q3Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, Y1Representing said modified first driving style, y2Indicating said modified second driving style, y3Representing said modified third driving style, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3And the comprehensive weight coefficient is corresponding to the third identification scheme.
Preferably, the first and second electrodes are formed of a metal,q2=0.2×1.01b,b represents the triggering times of the preset working condition; l represents the current mileage of the vehicle.
Preferably, the third initial driving type identification unit is specifically configured to:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
acquiring a fourth driving type of the driver according to a cloud server; the cloud server acquires the fourth driving type through a pre-trained deep learning network according to second operation information of the driver on the vehicle and second dynamic response information correspondingly generated by the vehicle in a second preset time period or a second preset driving mileage;
taking the first driving type, the second driving type, the third driving type and the fourth driving type as the initial driving type recognition result;
the current driving type identification module specifically includes:
a third driving style revision unit for revising the first driving style, the second driving style, the third driving style and the fourth driving style respectively according to the initial data; and the number of the first and second groups,
and the third current driving type identification unit is used for acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type, the corrected third driving type and the corrected fourth driving type.
Preferably, the third current driving type identifying unit specifically includes:
a third current driving type identification subunit for identifying a current driving type according to the formula Y ═ Y2·q2+y3·q3+y4·q4Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, Y2Indicating said modified second driving style, y3Indicating said modified third driving style, y4Representing said modified fourth driving style, q2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3Is the comprehensive weight coefficient, q, corresponding to the third identification scheme4The comprehensive weight coefficient corresponding to the fourth identification scheme; and the number of the first and second groups,
and the current driving type checking subunit is used for checking the current driving type of the driver according to the corrected first driving type.
Preferably, q is2=0.15×1.01b,q3=0.5-0.15×1.01b,q40.5; and b represents the triggering times of the preset working condition.
Preferably, the current driving type checking subunit is specifically configured to:
comparing the modified first driving style with the current driving style of the driver;
and when the deviation between the corrected first driving type and the current driving type of the driver exceeds a preset threshold value, sending fault reminding information to the cloud server.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program; wherein the computer program, when running, controls the device on which the computer-readable storage medium is located to execute the method for identifying a driving type of a driver according to any of the above embodiments.
An embodiment of the present invention further provides a terminal device, as shown in fig. 4, which is a block diagram of a preferred embodiment of the terminal device provided in the present invention, the terminal device includes a processor 10, a memory 20, and a computer program stored in the memory 20 and configured to be executed by the processor 10, and the processor 10, when executing the computer program, implements the method for identifying the driving type of the driver according to any of the embodiments.
Preferably, the computer program can be divided into one or more modules/units (e.g. computer program 1, computer program 2,) which are stored in the memory 20 and executed by the processor 10 to accomplish the present invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used for describing the execution process of the computer program in the terminal device.
The Processor 10 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, etc., the general purpose Processor may be a microprocessor, or the Processor 10 may be any conventional Processor, the Processor 10 is a control center of the terminal device, and various interfaces and lines are used to connect various parts of the terminal device.
The memory 20 mainly includes a program storage area that may store an operating system, an application program required for at least one function, and the like, and a data storage area that may store related data and the like. In addition, the memory 20 may be a high speed random access memory, may also be a non-volatile memory, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card), and the like, or the memory 20 may also be other volatile solid state memory devices.
It should be noted that the terminal device may include, but is not limited to, a processor and a memory, and those skilled in the art will understand that the structural block diagram of fig. 4 is only an example of the terminal device and does not constitute a limitation to the terminal device, and may include more or less components than those shown, or combine some components, or different components.
In summary, the embodiments of the present invention provide a method and an apparatus for identifying a driving type of a driver, a computer-readable storage medium, and a terminal device, the initial driving type recognition result of the driver is obtained by adopting at least two of four recognition schemes of recognizing the driving type according to the dynamic information of the vehicle, recognizing the driving type according to the operation information of the driver under the preset working condition, recognizing the driving type according to the pre-trained machine learning network and recognizing the driving type according to the pre-trained deep learning network, and the obtained initial data of the driving type identification and the initial driving type identification result are combined to carry out comprehensive weighting processing, so that the current driving type of the driver is identified, the problems of low accuracy and poor real-time performance of the identification result caused by the adoption of a single identification scheme in the prior art can be solved, and the accuracy and the real-time performance of the driving type identification of the driver are improved.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (16)
1. A driver driving type recognition method, characterized by comprising:
acquiring initial data of driving type identification of a driver; wherein the initial data includes driving type recognition results and driving behavior data of a history corresponding to the driver;
acquiring an initial driving type recognition result of the driver according to at least two preset recognition schemes of a first recognition scheme, a second recognition scheme, a third recognition scheme and a fourth recognition scheme; the method comprises the steps that a first identification scheme is used for identifying a driving type according to dynamic information of a vehicle, a second identification scheme is used for identifying the driving type according to operation information of a driver under a preset working condition, a third identification scheme is used for identifying the driving type according to a pre-trained machine learning network, and a fourth identification scheme is used for identifying the driving type according to a pre-trained deep learning network;
and acquiring the current driving type of the driver according to the initial data and the initial driving type identification result.
2. The driver driving style identification method according to claim 1, characterized in that, before the acquiring initial data of the driving style identification of the driver, the method further comprises:
identifying the identity of the driver;
the acquiring of the initial data of the driving type identification of the driver specifically includes:
and acquiring initial data of driving type identification corresponding to the identity of the driver according to a preset driving behavior database.
3. The method for identifying the driving style of the driver as claimed in claim 1 or 2, wherein the obtaining of the initial driving style identification result of the driver according to at least two identification schemes of a first identification scheme, a second identification scheme, a third identification scheme and a fourth identification scheme which are preset specifically comprises:
when a first preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme and the second recognition scheme;
when a second preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme and the third recognition scheme;
when a third preset condition is met, acquiring the initial driving type recognition result of the driver according to the first recognition scheme, the second recognition scheme, the third recognition scheme and the fourth recognition scheme;
wherein the first preset condition is that no good vehicle end processor and no good memory exist; the second preset condition is that a good vehicle end processor and a good memory are provided, but a cloud end processor and a good memory are not provided; the third preset condition is that the vehicle-mounted terminal processor and the memory are good, and the cloud-mounted processor and the memory are good.
4. The driver driving style identification method according to claim 3, characterized in that the method acquires the initial driving style identification result of the driver according to the first identification scheme and the second identification scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
taking the first driving type and the second driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type and the second driving type according to the initial data respectively;
and acquiring the current driving type of the driver according to the corrected first driving type and the corrected second driving type.
5. The method for identifying the driving style of the driver as claimed in claim 4, wherein the modifying the first driving style and the second driving style respectively according to the initial data specifically comprises:
wherein, y1Representing said modified first driving style, y1(k1) Representing said first driving style, y1(0)、y1(1)、…、y1(k1-1) corresponding, k, from said initial data1Is the number of identification times, p, corresponding to the first identification scheme1Is the weight coefficient corresponding to the first identification scheme, 0<p1<1;y2Indicating said modified second driving style, y2(k2) Representing said second driving style, y2(0)、y2(1)、…、y2(k2-1) corresponding, k, from said initial data2The number of identification times, p, corresponding to the second identification scheme2Is the weight coefficient corresponding to the second identification scheme, 0<p2<1;
The obtaining the current driving style of the driver according to the modified first driving style and the modified second driving style specifically includes:
according to the formula Y ═ Y1·q1+y2·q2Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2And the comprehensive weight coefficient is corresponding to the second identification scheme.
7. The driver driving style identification method according to claim 3, characterized in that the method acquires the initial driving style identification result of the driver according to the first identification scheme, the second identification scheme, and the third identification scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
taking the first driving type, the second driving type and the third driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type, the second driving type and the third driving type according to the initial data;
and acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type and the corrected third driving type.
8. The method for identifying the driving style of the driver as claimed in claim 7, wherein the obtaining the current driving style of the driver according to the modified first driving style, the modified second driving style and the modified third driving style specifically comprises:
according to the formula Y ═ Y1·q1+y2·q2+y3·q3Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, Y1Representing said modified first driving style, y2Indicating said modified second driving style, y3Representing said modified third driving style, q1Is the comprehensive weight coefficient, q, corresponding to the first identification scheme2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3And the comprehensive weight coefficient is corresponding to the third identification scheme.
10. The driver driving style identification method according to claim 3, characterized in that the method acquires the initial driving style identification result of the driver from the first identification scheme, the second identification scheme, the third identification scheme, and the fourth identification scheme by:
acquiring a first driving type of the driver according to dynamic information of a vehicle; wherein the dynamic information comprises longitudinal acceleration information;
acquiring a second driving type of the driver according to operation information of the driver on the vehicle under a preset working condition and dynamic response information correspondingly generated by the vehicle;
based on a pre-trained machine learning network, acquiring a third driving type of the driver according to first operation information of the driver on the vehicle and first dynamic response information correspondingly generated by the vehicle within a first preset time period or a first preset driving mileage;
acquiring a fourth driving type of the driver according to a cloud server; the cloud server acquires the fourth driving type through a pre-trained deep learning network according to second operation information of the driver on the vehicle and second dynamic response information correspondingly generated by the vehicle in a second preset time period or a second preset driving mileage;
taking the first driving type, the second driving type, the third driving type and the fourth driving type as the initial driving type recognition result;
then, the obtaining the current driving style of the driver according to the initial data and the initial driving style identification result specifically includes:
correcting the first driving type, the second driving type, the third driving type and the fourth driving type according to the initial data;
and acquiring the current driving type of the driver according to the corrected first driving type, the corrected second driving type, the corrected third driving type and the corrected fourth driving type.
11. The method for identifying the driving style of the driver as claimed in claim 10, wherein the obtaining the current driving style of the driver according to the modified first driving style, the modified second driving style, the modified third driving style and the modified fourth driving style specifically comprises:
according to the formula Y ═ Y2·q2+y3·q3+y4·q4Acquiring the current driving type of the driver; wherein Y represents the current driving style of the driver, Y2Indicating said modified second driving style, y3Indicating said modified third driving style, y4Representing said modified fourth driving style, q2Is the comprehensive weight coefficient, q, corresponding to the second identification scheme3Is the comprehensive weight coefficient, q, corresponding to the third identification scheme4The comprehensive weight coefficient corresponding to the fourth identification scheme;
and checking the current driving type of the driver according to the corrected first driving type.
12. The driver driving type recognition method according to claim 11, wherein q is2=0.15×1.01b,q3=0.5-0.15×1.01b,q40.5; and b represents the triggering times of the preset working condition.
13. The method for identifying the driving style of the driver as claimed in claim 11, wherein the checking the current driving style of the driver according to the modified first driving style specifically comprises:
comparing the modified first driving style with the current driving style of the driver;
and when the deviation between the corrected first driving type and the current driving type of the driver exceeds a preset threshold value, sending fault reminding information to the cloud server.
14. A driver driving type recognition apparatus, characterized by comprising:
the initial data acquisition module is used for acquiring initial data for identifying the driving type of the driver; wherein the initial data includes driving type recognition results and driving behavior data of a history corresponding to the driver;
the initial driving type recognition module is used for acquiring an initial driving type recognition result of the driver according to at least two recognition schemes of a first preset recognition scheme, a second preset recognition scheme, a third preset recognition scheme and a fourth preset recognition scheme; the method comprises the steps that a first identification scheme is used for identifying a driving type according to dynamic information of a vehicle, a second identification scheme is used for identifying the driving type according to operation information of a driver under a preset working condition, a third identification scheme is used for identifying the driving type according to a pre-trained machine learning network, and a fourth identification scheme is used for identifying the driving type according to a pre-trained deep learning network; and the number of the first and second groups,
and the current driving type identification module is used for acquiring the current driving type of the driver according to the initial data and the initial driving type identification result.
15. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored computer program; wherein the computer program, when executed, controls an apparatus on which the computer-readable storage medium is located to perform the driver driving type identification method according to any one of claims 1 to 13.
16. A terminal device, characterized by comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the driver driving type identification method according to any one of claims 1 to 13 when executing the computer program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811476975.6A CN111267863B (en) | 2018-12-04 | 2018-12-04 | Driver driving type identification method and device, storage medium and terminal equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811476975.6A CN111267863B (en) | 2018-12-04 | 2018-12-04 | Driver driving type identification method and device, storage medium and terminal equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111267863A true CN111267863A (en) | 2020-06-12 |
CN111267863B CN111267863B (en) | 2021-03-19 |
Family
ID=70993516
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811476975.6A Active CN111267863B (en) | 2018-12-04 | 2018-12-04 | Driver driving type identification method and device, storage medium and terminal equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111267863B (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111806431A (en) * | 2020-06-30 | 2020-10-23 | 中国第一汽车股份有限公司 | Parking control method and device, computer equipment and storage medium |
CN112053610A (en) * | 2020-10-29 | 2020-12-08 | 延安大学 | VR virtual driving training and examination method based on deep learning |
CN112109722A (en) * | 2020-09-28 | 2020-12-22 | 安徽江淮汽车集团股份有限公司 | Intelligent driving auxiliary control method and system |
CN112232254A (en) * | 2020-10-26 | 2021-01-15 | 清华大学 | Pedestrian risk assessment method considering pedestrian acceleration rate |
CN112477872A (en) * | 2020-11-26 | 2021-03-12 | 中国第一汽车股份有限公司 | Parameter calibration method, device, equipment and storage medium |
CN112693458A (en) * | 2021-01-15 | 2021-04-23 | 一汽解放汽车有限公司 | Cruise control method and device, vehicle and storage medium |
CN112706777A (en) * | 2020-12-28 | 2021-04-27 | 东软睿驰汽车技术(沈阳)有限公司 | Method and device for adjusting driving behaviors of user under vehicle working conditions |
CN113815625A (en) * | 2020-06-19 | 2021-12-21 | 广州汽车集团股份有限公司 | Vehicle auxiliary driving control method and device and intelligent steering wheel |
CN113859246A (en) * | 2020-06-30 | 2021-12-31 | 广州汽车集团股份有限公司 | Vehicle control method and device |
CN113859247A (en) * | 2020-06-30 | 2021-12-31 | 比亚迪股份有限公司 | Vehicle user identification method and device, vehicle machine and storage medium |
CN114103845A (en) * | 2022-01-25 | 2022-03-01 | 星河智联汽车科技有限公司 | Vehicle central control screen operator identity recognition method and device and vehicle |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101633358A (en) * | 2008-07-24 | 2010-01-27 | 通用汽车环球科技运作公司 | Adaptive vehicle control system with integrated driving style recognition |
US8280601B2 (en) * | 2008-07-24 | 2012-10-02 | GM Global Technology Operations LLC | Adaptive vehicle control system with integrated maneuver-based driving style recognition |
DE102012024649A1 (en) * | 2012-12-17 | 2014-06-18 | Valeo Schalter Und Sensoren Gmbh | Method for determining driving behavior of motor vehicle driver by driver assistance system of motor vehicle, involves determining behavior determination parameter for determining driving behavior depending on environment information |
CN107235044A (en) * | 2017-05-31 | 2017-10-10 | 北京航空航天大学 | It is a kind of to be realized based on many sensing datas to road traffic scene and the restoring method of driver driving behavior |
CN108482187A (en) * | 2018-04-26 | 2018-09-04 | 浙江吉利汽车研究院有限公司 | The control method for vehicle and system of identity-based identification |
-
2018
- 2018-12-04 CN CN201811476975.6A patent/CN111267863B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101633358A (en) * | 2008-07-24 | 2010-01-27 | 通用汽车环球科技运作公司 | Adaptive vehicle control system with integrated driving style recognition |
US8280601B2 (en) * | 2008-07-24 | 2012-10-02 | GM Global Technology Operations LLC | Adaptive vehicle control system with integrated maneuver-based driving style recognition |
DE102012024649A1 (en) * | 2012-12-17 | 2014-06-18 | Valeo Schalter Und Sensoren Gmbh | Method for determining driving behavior of motor vehicle driver by driver assistance system of motor vehicle, involves determining behavior determination parameter for determining driving behavior depending on environment information |
CN107235044A (en) * | 2017-05-31 | 2017-10-10 | 北京航空航天大学 | It is a kind of to be realized based on many sensing datas to road traffic scene and the restoring method of driver driving behavior |
CN108482187A (en) * | 2018-04-26 | 2018-09-04 | 浙江吉利汽车研究院有限公司 | The control method for vehicle and system of identity-based identification |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113815625A (en) * | 2020-06-19 | 2021-12-21 | 广州汽车集团股份有限公司 | Vehicle auxiliary driving control method and device and intelligent steering wheel |
CN113815625B (en) * | 2020-06-19 | 2024-01-19 | 广州汽车集团股份有限公司 | Vehicle auxiliary driving control method and device and intelligent steering wheel |
CN111806431B (en) * | 2020-06-30 | 2022-03-11 | 中国第一汽车股份有限公司 | Parking control method and device, computer equipment and storage medium |
CN111806431A (en) * | 2020-06-30 | 2020-10-23 | 中国第一汽车股份有限公司 | Parking control method and device, computer equipment and storage medium |
CN113859246B (en) * | 2020-06-30 | 2023-09-08 | 广州汽车集团股份有限公司 | Vehicle control method and device |
CN113859247A (en) * | 2020-06-30 | 2021-12-31 | 比亚迪股份有限公司 | Vehicle user identification method and device, vehicle machine and storage medium |
CN113859246A (en) * | 2020-06-30 | 2021-12-31 | 广州汽车集团股份有限公司 | Vehicle control method and device |
CN112109722A (en) * | 2020-09-28 | 2020-12-22 | 安徽江淮汽车集团股份有限公司 | Intelligent driving auxiliary control method and system |
CN112232254B (en) * | 2020-10-26 | 2021-04-30 | 清华大学 | Pedestrian risk assessment method considering pedestrian acceleration rate |
CN112232254A (en) * | 2020-10-26 | 2021-01-15 | 清华大学 | Pedestrian risk assessment method considering pedestrian acceleration rate |
CN112053610A (en) * | 2020-10-29 | 2020-12-08 | 延安大学 | VR virtual driving training and examination method based on deep learning |
CN112477872B (en) * | 2020-11-26 | 2022-05-27 | 中国第一汽车股份有限公司 | Parameter calibration method, device, equipment and storage medium |
CN112477872A (en) * | 2020-11-26 | 2021-03-12 | 中国第一汽车股份有限公司 | Parameter calibration method, device, equipment and storage medium |
CN112706777A (en) * | 2020-12-28 | 2021-04-27 | 东软睿驰汽车技术(沈阳)有限公司 | Method and device for adjusting driving behaviors of user under vehicle working conditions |
CN112706777B (en) * | 2020-12-28 | 2022-05-10 | 东软睿驰汽车技术(沈阳)有限公司 | Method and device for adjusting driving behaviors of user under vehicle working conditions |
CN112693458A (en) * | 2021-01-15 | 2021-04-23 | 一汽解放汽车有限公司 | Cruise control method and device, vehicle and storage medium |
CN114103845A (en) * | 2022-01-25 | 2022-03-01 | 星河智联汽车科技有限公司 | Vehicle central control screen operator identity recognition method and device and vehicle |
CN114103845B (en) * | 2022-01-25 | 2022-04-15 | 星河智联汽车科技有限公司 | Vehicle central control screen operator identity recognition method and device and vehicle |
Also Published As
Publication number | Publication date |
---|---|
CN111267863B (en) | 2021-03-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111267863B (en) | Driver driving type identification method and device, storage medium and terminal equipment | |
CN107492251B (en) | Driver identity recognition and driving state monitoring method based on machine learning and deep learning | |
CN108995653B (en) | Method and system for identifying driving style of driver | |
US9663112B2 (en) | Adaptive driver identification fusion | |
US11494685B2 (en) | Learning system, in-vehicle device, and server | |
CN110569783B (en) | Method and system for identifying lane changing intention of driver | |
CN110949386B (en) | Vehicle adaptive cruise control system and method capable of recognizing driving tendency | |
CN110386144B (en) | GHMM/GGAP-RBF hybrid model for identifying driver braking intention and identification method | |
US10913455B2 (en) | Method for the improved detection of objects by a driver assistance system | |
Jeong et al. | Real-time driver identification using vehicular big data and deep learning | |
US20220230536A1 (en) | Method and device for analyzing a sensor data stream and method for guiding a vehicle | |
CN112466118A (en) | Vehicle driving behavior recognition method, system, electronic device and storage medium | |
CN107358678B (en) | Driving event detection and identification method | |
CN114996109B (en) | User behavior recognition method, device, equipment and storage medium | |
CN109727334B (en) | Method and device for identifying terrain where vehicle is located and vehicle | |
CN114162132B (en) | Driving mode identification method based on subjective and objective evaluation | |
JP2013016037A (en) | Traveling scene recognition model generation device, driving support device, and program | |
CN112215118B (en) | Method and device for automatically extracting data features of vehicle sensor and computer-readable storage medium | |
JP6365035B2 (en) | Traffic object detection device | |
CN111368784B (en) | Target identification method, device, computer equipment and storage medium | |
JP7120258B2 (en) | Action recognition device and action recognition method | |
CN114103966A (en) | Control method, device and system for driving assistance | |
CN114155607A (en) | Driving risk monitoring and alarming method, device, equipment and storage medium | |
CN113221843A (en) | Driving style classification method based on empirical mode decomposition characteristics | |
CN111860312A (en) | Driving environment adjusting method and device |
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 |