CN111598367A

CN111598367A - Driving behavior evaluation method and device, computing equipment and storage medium

Info

Publication number: CN111598367A
Application number: CN201910127179.XA
Authority: CN
Inventors: 申子豪; 燕丽敬; 陈永祥; 郝勇刚
Original assignee: Hangzhou Hikvision Digital Technology Co Ltd
Current assignee: Hangzhou Hikvision Digital Technology Co Ltd
Priority date: 2019-02-20
Filing date: 2019-02-20
Publication date: 2020-08-28
Anticipated expiration: 2039-02-20
Also published as: CN111598367B

Abstract

The application discloses a driving behavior evaluation method and device, computing equipment and a storage medium, and belongs to the technical field of intelligent analysis. According to the driving behavior evaluation method provided by the embodiment of the application, the computing device determines the first weight and the first comprehensive score of each appointed driving behavior according to the maximum frequency and the first frequency by acquiring the maximum frequency and the first frequency of a target driver; determining a second weight for each designated driving behavior based on the degree of importance of the first designated driving behavior relative to the second designated driving behavior; and determining a second comprehensive score of the driving behavior of the target driver according to the first weight, the first comprehensive score and the second weight of each designated driving behavior. According to the method, the second comprehensive score is determined through the first comprehensive score and two different weights, the condition that a few extreme values influence the overall evaluation is avoided, and the accuracy of the driving behavior evaluation is improved.

Description

Driving behavior evaluation method and device, computing equipment and storage medium

Technical Field

The application relates to the technical field of intelligent analysis. In particular to a driving behavior evaluation method, a driving behavior evaluation device, a computing device and a storage medium.

Background

Along with the enhancement of the travel safety awareness of people, people pay more and more attention to the driving behavior of a driver in the driving process, and the driving behavior of the driver directly influences the travel safety of people. Therefore, it is important to reasonably evaluate the driving behavior of the driver. The evaluation of the driving behavior of the driver is not only beneficial to the performance assessment of the driver by the company where the driver is located, but also beneficial to the specification of the driving behavior of the driver, and the driving safety is ensured.

In the related art, the driving behavior of the driver is mainly evaluated through the final score of the driving behavior. The higher the final score of the driving behavior, the more normative the driving behavior of the driver. Wherein, before determining the final score of the driving behavior, a plurality of specified driving behaviors such as smoking, calling, unbelted belts and the like are set, and the corresponding relation between the occurrence frequency of each specified driving behavior and the score is set. Accordingly, the process of determining the final score of the driving behavior may be: determining the occurrence frequency of each specified driving behavior in the driving behaviors of the target driver according to the driving behavior data of the target driver; and determining the grade of each appointed driving behavior corresponding to the target driver from the corresponding relationship between the occurrence frequency and the grade according to the occurrence frequency of each appointed driving behavior, and taking the average value of the grades of each appointed driving behavior corresponding to the target driver as the final grade of the target driver.

However, in the driving behavior evaluation method based on the average value in the related art, the driving behavior evaluation accuracy is low due to the fact that a few extreme values easily affect the overall evaluation.

Disclosure of Invention

The embodiment of the application provides a driving behavior evaluation method and device, a computing device and a storage medium, and can solve the problem of low driving behavior evaluation accuracy. The technical scheme is as follows:

in one aspect, a driving behavior evaluation method is provided, and the method includes:

acquiring the maximum frequency of each specified driving behavior of a target driver in a first historical time and the first frequency of each specified driving behavior in a current evaluation time range;

determining a first weight and a first composite score of each designated driving behavior based on the maximum frequency and the first frequency of occurrence of each designated driving behavior;

determining a second weight of each of a first number of specified driving behaviors, the first number being the number of specified driving behaviors, based on a degree of importance of the first specified driving behavior relative to a second specified driving behavior, the first specified driving behavior and the second specified driving behavior being any two different specified driving behaviors of the first number of specified driving behaviors;

determining a second composite score for the driving behavior of the target driver based on the first weight, the first composite score, and the second weight for each designated driving behavior.

In one possible implementation, the determining the first weight and the first composite score of each designated driving behavior based on the maximum frequency and the first frequency of occurrence of each designated driving behavior includes:

determining a first rating of the each designated driving behavior for a plurality of rating periods within the current rating time range based on the maximum frequency and the first frequency of occurrence of the each designated driving behavior;

determining a first weight of each designated driving behavior based on a first score and a second number of each designated driving behavior in each evaluation period, wherein the second number is the number of the plurality of evaluation periods;

determining a first composite score for each designated driving behavior based on the first score for each designated driving behavior for each evaluation period.

In another possible implementation manner, the determining a first score of the plurality of evaluation periods of each designated driving behavior within the current evaluation time range based on the maximum frequency and the first frequency of occurrence of each designated driving behavior includes:

determining, for each evaluation period for each specified driving behavior, a ratio between a first frequency of occurrence of the specified driving behavior and a maximum frequency;

determining a second score for the designated driving behavior to be deducted within the evaluation period according to the ratio;

and determining a first score of the specified driving behavior in the evaluation period according to the total score and the second score corresponding to the evaluation period.

In another possible implementation manner, the determining a first weight of each designated driving behavior based on the first score and the second number of each designated driving behavior in each evaluation period includes:

determining a first entropy value for each of the designated driving behaviors based on the first score and the second quantity for the each of the designated driving behaviors over the each evaluation period;

determining a first weight for the each specified driving behavior based on the first entropy value and the first quantity for the each specified driving behavior.

In another possible implementation manner, the determining a first entropy value of each of the designated driving behaviors based on the first score and the second number of each of the designated driving behaviors in each of the evaluation periods includes:

performing normalization processing on the first score of each specified driving behavior in each evaluation period to obtain a first score of each specified driving behavior in each evaluation period;

determining a first entropy value of each designated driving behavior based on the first score and the second quantity of each designated driving behavior in each evaluation period by the following formula one;

the formula I is as follows:

wherein i is a number designating a driving behavior, j is a number designating an evaluation period, and H_iA first entropy value for the ith specified driving behavior, s is the second quantity,

b_ija first score for the ith designated driving behavior for the jth evaluation period.

In another possible implementation manner, the determining a first weight of each specified driving behavior based on the first entropy and the first number of each specified driving behavior includes:

determining a weight matrix including a first weight of the each specified driving behavior by the following formula two based on the first entropy value and the first number of the each specified driving behavior;

the formula II is as follows: w ═ ω_i')_1×t

Wherein W is the weight matrix,

ω_i' is the first weight of the ith designated driving behavior, H_iA first entropy value of the driving behavior is specified for the ith, and

t is the first number;

determining a first weight for said each specified driving behavior from said weight matrix.

In another possible implementation, the determining the second weight of each of the first number of specified driving behaviors based on the degree of importance of the first specified driving behavior relative to the second specified driving behavior includes:

constructing a first judgment matrix based on the importance degree of the first specified driving behavior relative to the second specified driving behavior;

determining a matrix order and a maximum eigenvalue of the first judgment matrix based on the first judgment matrix;

determining a random consistency index of the first judgment matrix from a corresponding relation between the matrix order and the random consistency index based on the matrix order;

and determining a second weight of each designated driving behavior in the first number of designated driving behaviors based on the matrix order of the first judgment matrix, the maximum eigenvalue and the random consistency index.

In another possible implementation manner, the determining, based on the matrix order of the first determination matrix, the maximum eigenvalue, and the random consistency index, a second weight of each of the first number of designated driving behaviors includes:

determining consistency proportion of the first judgment matrix based on the matrix order of the first judgment matrix, the maximum eigenvalue and the random consistency index;

when the consistency ratio is smaller than a ratio threshold, determining the first number of first feature vectors corresponding to the maximum feature value;

normalizing the first quantity of first feature vectors to obtain a first quantity of second feature vectors;

determining the value of the first number of second eigenvectors as a second weight of the first number of specified driving behaviors.

In another possible implementation manner, the determining a second composite score of the driving behavior of the target driver based on the first weight, the first composite score and the second weight of each designated driving behavior includes:

determining a comprehensive weight of each designated driving behavior based on the first weight and the second weight of each designated driving behavior;

determining a third score of each designated driving feature in a third number of designated driving features based on the comprehensive weight and the first comprehensive score of each designated driving behavior, wherein the third number is the number of the designated driving features, and each designated driving feature corresponds to a plurality of designated driving behaviors;

determining a third weight for each of the specified driving characteristics based on a degree of importance of the first specified driving characteristic relative to a second specified driving characteristic, the first and second specified driving characteristics being any two different specified driving characteristics of the third number of specified driving characteristics;

determining the second composite score based on the third score and the third weight for each specified driving feature.

In another possible implementation manner, the determining a comprehensive weight of each designated driving behavior based on the first weight and the second weight of each designated driving behavior includes:

for each appointed driving behavior, determining the product of the first weight and the second weight of the appointed driving behavior to obtain a first numerical value;

summing the first numerical values of each designated driving behavior to obtain a second numerical value;

and determining the ratio of the first numerical value to the second numerical value to obtain the comprehensive weight of the specified driving behavior.

In another possible implementation manner, the determining the second composite score based on the third score and the third weight of each specified driving feature includes:

weighting and summing the third score and the third weight of each appointed driving feature to obtain a second comprehensive score; alternatively, the first and second electrodes may be,

and carrying out weighted summation on the third score and the third weight of each appointed driving feature to obtain a third comprehensive score, and fusing the third comprehensive score with a plurality of fourth comprehensive scores of the driving behavior of the target driver in a second historical time to obtain a second comprehensive score.

In another possible implementation manner, the fusing the third composite score with a plurality of fourth composite scores of the driving behavior of the target driver in a second historical time to obtain the second composite score includes:

determining, by a target gaussian mixture model, a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores;

normalizing the first probability and the second probabilities to obtain a fourth weight corresponding to the third composite score and a plurality of fifth weights corresponding to the fourth composite scores, wherein one fourth composite score corresponds to one fifth weight;

and weighting and summing the third composite score, the fourth weights, the plurality of fourth composite scores and the plurality of fifth weights to obtain the second composite score.

In another possible implementation manner, the method further includes:

initializing a plurality of first model parameters of the initial Gaussian mixture model to obtain a plurality of second model parameters;

determining the number of the component models in the initial Gaussian mixture model, and setting a convergence threshold value and the maximum iteration times;

determining a first responsiveness of the first sub-model to the third composite score and the plurality of fourth composite scores based on the plurality of second model parameters;

determining a plurality of third model parameters based on the first responsiveness, the third composite score, and the plurality of fourth composite scores;

when the plurality of third model parameters meet the convergence threshold or reach the maximum iteration number, outputting the plurality of third model parameters to obtain the target Gaussian mixture model;

and when the plurality of third model parameters do not meet the convergence threshold value or the maximum iteration number is not reached, determining a plurality of fourth model parameters based on the plurality of third model parameters, the third comprehensive score and the plurality of fourth comprehensive scores until the plurality of model parameters meet the convergence threshold value or the maximum iteration number is reached, and outputting a plurality of corresponding model parameters when the convergence threshold value is met or the maximum iteration number is reached to obtain the target Gaussian mixture model.

In another possible implementation manner, the determining, by the target gaussian mixture model, a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores includes:

determining the first probability and the plurality of second probabilities by the target gaussian mixture model and the following formula three;

the formula III is as follows: m ═ P (y)_v|θ_k)*2*y_const

Wherein the content of the first and second substances,

m is the first probability or the second probability, phi (y)_v|θ_k) Is the Gaussian distribution density of the kth partial model, y_vIs the third composite score or the fourth composite score, y_constIs a constant, mu_kIs a first parameter, σ_k ²As a second parameter, α_kAnd R is the number of the partial models, and k is less than or equal to R.

In another aspect, there is provided a driving behavior evaluation device including:

the acquisition module is used for acquiring the maximum frequency of each specified driving behavior of the target driver in a first historical time and the first frequency of each specified driving behavior in the current evaluation time range;

the first determination module is used for determining a first weight and a first comprehensive score of each specified driving behavior based on the maximum frequency and the first frequency of occurrence of each specified driving behavior;

a second determination module, configured to determine a second weight of each of a first number of specified driving behaviors based on a degree of importance of the first specified driving behavior relative to a second specified driving behavior, where the first number is the number of specified driving behaviors, and the first specified driving behavior and the second specified driving behavior are any two different specified driving behaviors of the first number of specified driving behaviors;

and the third determination module is used for determining a second comprehensive score of the driving behavior of the target driver based on the first weight, the first comprehensive score and the second weight of each specified driving behavior.

In a possible implementation manner, the first determining module is further configured to determine, based on the maximum frequency and the first frequency of occurrence of each designated driving behavior, a first score of each designated driving behavior for a plurality of evaluation periods within the current evaluation time range; determining a first weight of each designated driving behavior based on a first score and a second number of each designated driving behavior in each evaluation period, wherein the second number is the number of the plurality of evaluation periods; determining a first composite score for each designated driving behavior based on the first score for each designated driving behavior for each evaluation period.

In another possible implementation manner, the first determining module is further configured to determine, for each evaluation period for each specified driving behavior, a ratio between a first frequency and a maximum frequency of occurrence of the specified driving behavior; determining a second score for the designated driving behavior to be deducted within the evaluation period according to the ratio; and determining a first score of the specified driving behavior in the evaluation period according to the total score and the second score corresponding to the evaluation period.

In another possible implementation manner, the first determining module is further configured to determine a first entropy value of each designated driving behavior based on the first score and the second quantity of each designated driving behavior in each evaluation period; determining a first weight for the each specified driving behavior based on the first entropy value and the first quantity for the each specified driving behavior.

In another possible implementation manner, the first determining module is further configured to perform normalization processing based on the first score of each designated driving behavior in each evaluation period, so as to obtain a first score of each designated driving behavior in each evaluation period;

formula (II)Firstly, the method comprises the following steps:

wherein i is a number designating a driving behavior, j is a number designating an evaluation period, H_iA first entropy value for the ith specified driving behavior, s is the second quantity,

In another possible implementation manner, the first determining module is further configured to determine a weight matrix based on the first entropy and the first number of each specified driving behavior, where the weight matrix includes a first weight of each specified driving behavior according to the following formula two;

the formula II is as follows: w ═ ω_i')_1×t

Wherein W is the weight matrix,

t is the first number;

In another possible implementation manner, the second determining module is further configured to construct a first judgment matrix based on the importance degree of the first specified driving behavior relative to the second specified driving behavior; determining a matrix order and a maximum eigenvalue of the first judgment matrix based on the first judgment matrix; determining a random consistency index of the first judgment matrix from a corresponding relation between the matrix order and the random consistency index based on the matrix order; and determining a second weight of each designated driving behavior in the first number of designated driving behaviors based on the matrix order of the first judgment matrix, the maximum eigenvalue and the random consistency index.

In another possible implementation manner, the second determining module is further configured to determine a consistency ratio of the first determination matrix based on a matrix order of the first determination matrix, the maximum eigenvalue, and the random consistency index; when the consistency ratio is smaller than a ratio threshold, determining the first number of first feature vectors corresponding to the maximum feature value; normalizing the first quantity of first feature vectors to obtain a first quantity of second feature vectors; determining the value of the first number of second eigenvectors as a second weight of the first number of specified driving behaviors.

In another possible implementation manner, the third determining module is further configured to determine a comprehensive weight of each designated driving behavior based on the first weight and the second weight of each designated driving behavior; determining a third score of each designated driving feature in a third number of designated driving features based on the comprehensive weight and the first comprehensive score of each designated driving behavior, wherein the third number is the number of the designated driving features, and each designated driving feature corresponds to a plurality of designated driving behaviors; determining a third weight for each of the specified driving characteristics based on a degree of importance of the first specified driving characteristic relative to a second specified driving characteristic, the first and second specified driving characteristics being any two different specified driving characteristics of the third number of specified driving characteristics; determining the second composite score based on the third score and the third weight for each specified driving feature.

In another possible implementation manner, the second determining module is further configured to determine, for each specified driving behavior, a product of the first weight and the second weight of the specified driving behavior to obtain a first numerical value; summing the first numerical values of each designated driving behavior to obtain a second numerical value; and determining the ratio of the first numerical value to the second numerical value to obtain the comprehensive weight of the specified driving behavior.

In another possible implementation manner, the second determining module is further configured to perform weighted summation on the third score and the third weight of each specified driving feature to obtain the second composite score; or carrying out weighted summation on the third score and the third weight of each designated driving feature to obtain a third comprehensive score, and fusing the third comprehensive score with a plurality of fourth comprehensive scores of the driving behavior of the target driver in a second historical time to obtain a second comprehensive score.

In another possible implementation manner, the second determining module is further configured to determine, through a target gaussian mixture model, a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores; normalizing the first probability and the second probabilities to obtain a fourth weight corresponding to the third composite score and a plurality of fifth weights corresponding to the fourth composite scores, wherein one fourth composite score corresponds to one fifth weight; and weighting and summing the third composite score, the fourth weights, the plurality of fourth composite scores and the plurality of fifth weights to obtain the second composite score.

In another possible implementation manner, the apparatus further includes:

the fourth determining module is used for initializing a plurality of first model parameters of the initial Gaussian mixture model to obtain a plurality of second model parameters; determining the number of the component models in the initial Gaussian mixture model, and setting a convergence threshold value and the maximum iteration times; determining a first responsiveness of the first sub-model to the third composite score and the plurality of fourth composite scores based on the plurality of second model parameters; determining a plurality of third model parameters based on the first responsiveness, the third composite score, and the plurality of fourth composite scores; when the plurality of third model parameters meet the convergence threshold or reach the maximum iteration number, outputting the plurality of third model parameters to obtain the target Gaussian mixture model; and when the plurality of third model parameters do not meet the convergence threshold value or the maximum iteration number is not reached, determining a plurality of fourth model parameters based on the plurality of third model parameters, the third comprehensive score and the plurality of fourth comprehensive scores until the plurality of model parameters meet the convergence threshold value or the maximum iteration number is reached, and outputting a plurality of corresponding model parameters when the convergence threshold value is met or the maximum iteration number is reached to obtain the target Gaussian mixture model.

In another possible implementation manner, the second determining module is further configured to determine the first probability and the plurality of second probabilities through the target gaussian mixture model and the following formula three;

the formula III is as follows: m ═ P (y)_v|θ_k)*2*y_const

Wherein the content of the first and second substances,

In another aspect, a computing device is provided, the computing device comprising:

a processor and a memory, the memory having stored therein at least one instruction, at least one program, set of codes, or set of instructions, the instruction, the program, the set of codes, or the set of instructions being loaded and executed by the processor to implement the operations performed in the driving behavior assessment method of any of the first aspects.

In another aspect, a computer-readable storage medium is provided, in which at least one instruction, at least one program, a set of codes, or a set of instructions is stored, which is loaded and executed by a processor to implement the operations performed by any of the above driving behavior evaluation methods.

The technical scheme provided by the embodiment of the application has the following beneficial effects:

according to the driving behavior evaluation method provided by the embodiment of the application, the computing device determines the first weight and the first comprehensive score of each designated driving behavior according to the maximum frequency and the first frequency by acquiring the maximum frequency of each designated driving behavior of the target driver in the first historical time and the first frequency of each designated driving behavior in the current evaluation time range. The computing device also determines a second weight for each of the first number of specified driving behaviors based on a degree of importance of the first specified driving behavior relative to the second specified driving behavior. The computing device determines a second composite score for the driving behavior of the target driver based on the first weight, the first composite score, and the second weight for each designated driving behavior. According to the method, the second comprehensive score is determined through the first comprehensive score and two different weights, the condition that a few extreme values influence the overall evaluation is avoided, and the accuracy of the driving behavior evaluation is improved.

Drawings

Fig. 1 is a schematic diagram of an application scenario of driving behavior evaluation provided in an embodiment of the present application;

FIG. 2 is a flow chart of a driving behavior evaluation method provided in an embodiment of the present application;

FIG. 3 is a flow chart of a driving behavior evaluation method provided in an embodiment of the present application;

fig. 4 is a schematic diagram of determining the number of partial models according to an embodiment of the present disclosure;

FIG. 5 is a schematic diagram of determining a first probability and a plurality of second probabilities provided by an embodiment of the present application;

FIG. 6 is a schematic diagram of another method for determining a first probability and a plurality of second probabilities provided by an embodiment of the present application;

fig. 7 is a schematic structural diagram of a driving behavior evaluation device according to an embodiment of the present application;

fig. 8 is a block diagram of a computing device according to an embodiment of the present disclosure.

Detailed Description

In order to make the technical solutions and advantages of the present application more clear, the following describes the embodiments of the present application in further detail.

An embodiment of the present application provides an application scenario for driving behavior evaluation, and referring to fig. 1, the application scenario includes: a computing device 101 and a plurality of sensors 102. The computing device 101 and the plurality of sensors 102 may be connected by a wireless connection or a wired connection, which is not particularly limited in the embodiments of the present application. The plurality of sensors 102 are disposed in a target vehicle driven by a target driver, and are configured to collect driving behavior data of the target driver during driving, and send the driving behavior data to the computing device 101. The driving behavior data may be driving behavior data within the first history time, may also be driving behavior data within the current evaluation time range, and may also include driving behavior data within the first history time and driving behavior data within the current evaluation time range.

The computing device 101 may be a terminal or a server. When the computing device 101 is a terminal, the plurality of sensors 102 transmit the acquired driving behavior data to the terminal, and the terminal receives the driving behavior data transmitted by the plurality of sensors 102, so that a second comprehensive score of the target driver is finally obtained through the driving behavior data. When the computing device 101 is a server, the application scenario includes: a terminal, a computing device 101 (server), and a plurality of sensors 102. An Application program APP (Application) for driving behavior evaluation is installed on the terminal, and the server is a background server of the APP. The terminal logs in the server through the APP and communicates with the server, and therefore the server determines a second comprehensive score of the target driver through the driving behavior data. The plurality of sensors 102 send collected driving behavior data to the terminal, the terminal runs the APP, the driving behavior data are sent to the server through the APP, the server obtains a second comprehensive score of a target driver through the driving behavior data, and the second comprehensive score is sent to the terminal. In the embodiment of the present application, the computing device 101 is taken as an example for explanation.

The computing device 101 determines, from the received driving behavior data, a maximum frequency of occurrence of each specified driving behavior of the target driver over a first historical time, and a first frequency of occurrence of each specified driving behavior over a current evaluation time range; determining a first weight and a first composite score for each designated driving behavior based on the maximum frequency and the first frequency of occurrence of each designated driving behavior; determining a second weight for each of the first number of designated driving behaviors based on a degree of importance of the first designated driving behavior relative to the second designated driving behavior; a second composite score for the driving behavior of the target driver is determined based on the first weight, the first composite score, and the second weight for each designated driving behavior. The higher the second composite score, the more normative the driving behavior of the target driver. The company where the target driver is located can evaluate the driving behavior of the target driver and evaluate the performance of the target driver according to the second comprehensive score, and rewards the driver with the higher second comprehensive score, so that the driving behavior of the driver can be normalized, and the driving safety is ensured.

The target driver may be a taxi driver, a bus driver, a shared bus company driver, or a car rental company driver. For example, when the target driver is a taxi driver, the company where the target driver is located is a taxi company; when the target driver is a bus driver, the company where the target driver is located is a bus company. In the embodiment of the present application, the target driver and the company where the target driver is located are not particularly limited. The plurality of sensors 102 that collect driving behavior data may be a plurality of different kinds of sensors 102, such as a video sensor 102, a steering wheel sensor 102, a foot pedal sensor 102, an eyelid sensor 102, a car speed sensor 102, and the like. Different kinds of sensors 102 may collect different driving behavior data, e.g., eyelid sensor 102 may monitor the time during which the target driver's eyelid blinks persist, from which computing device 101 determines whether the target driver is driving fatigue; the car speed sensor 102 may monitor the speed of the target vehicle driven by the target driver, from which the computing device 101 determines whether the target driver is speeding. The specified driving behavior refers to a driving behavior that does not comply with safe driving, for example, not giving a courtesy to pedestrians, fatigue driving, smoking, making and receiving calls, speeding, and the like. In the embodiment of the present application, the specified driving behavior is not particularly limited.

The embodiment of the application provides a driving behavior evaluation method, and referring to fig. 2, the method comprises the following steps:

step 201: the maximum frequency of occurrence of each specified driving behavior of the target driver in the first historical time and the first frequency of occurrence of each specified driving behavior in the current evaluation time range are obtained.

Step 202: a first weight and a first composite score for each designated driving behavior are determined based on a maximum frequency and a first frequency of occurrence of each designated driving behavior.

Step 203: determining a second weight of each of the first number of specified driving behaviors based on a degree of importance of the first specified driving behavior relative to the second specified driving behavior, the first number being the number of specified driving behaviors, the first specified driving behavior and the second specified driving behavior being any two different specified driving behaviors of the first number of specified driving behaviors.

Step 204: a second composite score for the driving behavior of the target driver is determined based on the first weight, the first composite score, and the second weight for each designated driving behavior.

In one possible implementation, determining the first weight and the first composite score for each designated driving behavior based on the maximum frequency and the first frequency of occurrence of each designated driving behavior includes:

determining a first score of a plurality of evaluation periods of each designated driving behavior within a current evaluation time range based on the maximum frequency and the first frequency of occurrence of each designated driving behavior;

determining a first weight of each designated driving behavior based on a first score and a second number of each designated driving behavior in each evaluation period, wherein the second number is the number of a plurality of evaluation periods;

a first composite score for each designated driving behavior is determined based on the first score for each designated driving behavior for each evaluation period.

In another possible implementation manner, determining a first score of each designated driving behavior for a plurality of evaluation periods within a current evaluation time range based on the maximum frequency and the first frequency of occurrence of each designated driving behavior includes:

determining a ratio between a first frequency of occurrence of the designated driving behavior and a maximum frequency for each designated driving behavior at each evaluation period;

and determining a first score of the designated driving behavior in the evaluation period according to the total score and the second score corresponding to the evaluation period.

In another possible implementation, determining the first weight of each designated driving behavior based on the first score and the second number of each designated driving behavior in each evaluation period includes:

determining a first entropy value for each designated driving behavior based on the first score and the second number for each designated driving behavior for each evaluation period;

a first weight for each designated driving behavior is determined based on the first entropy and the first quantity for each designated driving behavior.

In another possible implementation, determining a first entropy value for each of the designated driving behaviors based on the first score and the second number for each of the designated driving behaviors for each evaluation period includes:

performing normalization processing on the first score of each designated driving behavior in each evaluation period to obtain a first score of each designated driving behavior in each evaluation period;

determining a first entropy value of each designated driving behavior based on the first score and the second quantity of each designated driving behavior in each evaluation period through the following formula I;

the formula I is as follows:

wherein i is a number designating a driving behavior, j is a number designating an evaluation period, H_iA first entropy value for the ith specified driving behavior, s a second quantity,

In another possible implementation, determining a first weight for each specified driving behavior based on the first entropy and the first quantity for each specified driving behavior includes:

determining a weight matrix based on the first entropy and the first quantity of each designated driving behavior by the following formula two, wherein the weight matrix comprises a first weight of each designated driving behavior;

the formula II is as follows: w ═ ω_i')_1×t

Wherein, W is a weight matrix,

t is a first number;

a first weight for each specified driving behavior is determined from the weight matrix.

In another possible implementation, determining the second weight for each of the first number of specified driving behaviors based on the degree of importance of the first specified driving behavior relative to the second specified driving behavior includes:

and determining a second weight of each designated driving behavior in the first number of designated driving behaviors based on the matrix order, the maximum eigenvalue and the random consistency index of the first judgment matrix.

In another possible implementation manner, determining the second weight of each of the first number of designated driving behaviors based on the matrix order, the maximum eigenvalue, and the random consistency index of the first determination matrix includes:

determining consistency proportion of the first judgment matrix based on the matrix order, the maximum eigenvalue and the random consistency index of the first judgment matrix;

when the consistency ratio is smaller than a ratio threshold, determining a first number of first feature vectors corresponding to the maximum feature value;

the values of the first number of second eigenvectors are determined as second weights for the first number of specified driving behaviors.

In another possible implementation manner, determining a second composite score of the driving behavior of the target driver based on the first weight, the first composite score and the second weight of each designated driving behavior includes:

determining a third weight for each of the specified driving characteristics based on the degree of importance of the first specified driving characteristic relative to the second specified driving characteristic, the first specified driving characteristic and the second specified driving characteristic being any two different specified driving characteristics of a third number of specified driving characteristics;

a second composite score is determined based on the third score and the third weight for each designated driving feature.

In another possible implementation, determining a composite weight for each designated driving behavior based on the first weight and the second weight for each designated driving behavior includes:

for each appointed driving behavior, determining the product of a first weight and a second weight of the appointed driving behavior to obtain a first numerical value;

and determining the ratio of the first value to the second value to obtain the comprehensive weight of the appointed driving behavior.

In another possible implementation, determining a second composite score based on the third score and the third weight for each specified driving feature includes:

weighting and summing the third score and the third weight of each designated driving feature to obtain a second comprehensive score; alternatively, the first and second electrodes may be,

and carrying out weighted summation on the third score and the third weight of each appointed driving feature to obtain a third comprehensive score, and fusing the third comprehensive score and a plurality of fourth comprehensive scores of the driving behaviors of the target driver in the second historical time to obtain a second comprehensive score.

In another possible implementation manner, the fusing the third composite score with a plurality of fourth composite scores of the driving behavior of the target driver in the second historical time to obtain a second composite score includes:

determining a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores through a target Gaussian mixture model;

normalizing the first probability and the plurality of second probabilities to respectively obtain a fourth weight corresponding to the third comprehensive score and a plurality of fifth weights corresponding to the plurality of fourth comprehensive scores, wherein one fourth comprehensive score corresponds to one fifth weight;

and carrying out weighted summation on the third comprehensive score, the fourth weight, the plurality of fourth comprehensive scores and the plurality of fifth weights to obtain a second comprehensive score.

In another possible implementation manner, the method further includes:

determining the number of component models in the initial Gaussian mixture model, and setting a convergence threshold value and the maximum iteration times;

when the plurality of third model parameters meet the convergence threshold value or reach the maximum iteration times, outputting the plurality of third model parameters to obtain a target Gaussian mixture model;

and when the plurality of third model parameters do not meet the convergence threshold value or do not reach the maximum iteration times, determining a plurality of fourth model parameters based on the plurality of third model parameters, the third comprehensive score and the plurality of fourth comprehensive scores until the plurality of model parameters meet the convergence threshold value or reach the maximum iteration times, and outputting the plurality of corresponding model parameters when the plurality of model parameters meet the convergence threshold value or reach the maximum iteration times to obtain the target Gaussian mixture model.

In another possible implementation manner, determining, by the target gaussian mixture model, a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores includes:

determining a first probability and a plurality of second probabilities through a target Gaussian mixture model and the following formula III;

the formula III is as follows: m ═ P (y)_v|θ_k)*2*y_const

Wherein the content of the first and second substances,

The embodiment of the application provides a driving behavior evaluation method, and referring to fig. 3, the method comprises the following steps:

step 301: the computing device obtains a maximum frequency of occurrence of each specified driving behavior of the target driver over a first historical time, and a first frequency of occurrence of each specified driving behavior over a current evaluation time range.

The method comprises the steps that a plurality of sensors are installed in a target vehicle driven by a target driver, the plurality of sensors can collect driving behavior data of the target driver in the driving process and send the driving behavior data to a computing device, the computing device receives the driving behavior data sent by the plurality of sensors, and the maximum frequency of occurrence of each specified driving behavior of the target driver in a first historical time and the first frequency of occurrence of each specified driving behavior in a current evaluation time range are obtained from the driving behavior data.

The specified driving behavior refers to a driving behavior that does not comply with safe driving, for example, not giving a courtesy to pedestrians, fatigue driving, smoking, making and receiving calls, speeding, and the like. In the embodiment of the present application, the specified driving behavior is not particularly limited. Also, the number of specified driving behaviors may be set and increased as needed. The current evaluation time range may be the current day, the current two days, or the current three days, and the first historical time may be the previous month, the previous three months, or the previous six months prior to the current evaluation time range. In the embodiment of the present application, the current evaluation time range and the first history time are not particularly limited.

The maximum frequency may be the maximum frequency of occurrence of each specified driving behavior in the first history time within one evaluation period or several evaluation periods. The first frequency may be a frequency with which each specified driving behavior within the current evaluation time range occurs within one evaluation period or several evaluation periods. The evaluation period may be set according to time or mileage, and when the evaluation period is set according to time, one evaluation period may be one hour, two hours, or three hours. When the evaluation period is set according to the mileage, one evaluation period may be one kilometer, two kilometers, or three kilometers. In the examples of the present application, the evaluation period is not particularly limited. For example, if the driving behavior is specified as smoking, one evaluation period is one hour, the current evaluation time range is the current day, the first historical time is three months before the current day, the maximum frequency is the maximum frequency occurring within one evaluation period, and the first frequency is the first frequency occurring within one evaluation period, then the maximum frequency of smoking occurring is the maximum frequency of smoking by the target driver within the first three months before the current day within one hour, for example, the maximum frequency is 5. The first frequency of occurrence of a smoke is the number of times the target driver has smoked within one hour of the day, e.g. the first frequency is 1.

It should be noted that, the driving behavior data of one driver in the first historical time is limited, so the maximum frequency may also be the maximum frequency of each designated driving behavior of other drivers of the company where the target driver is located in the first historical time, or the maximum frequency of each designated driving behavior of the drivers of other driving companies in the first historical time, thereby expanding the sources of the driving behavior data and improving the accuracy of the driving behavior evaluation.

For example, the working hours of the target driver's working day is 6:00 to 19:00, where 10:00 to 11:00 are lunch and rest time, 16:00 to 17:00 are dinner and rest time, and both lunch and rest time and dinner and rest time are not involved in the evaluation, and one evaluation period is one hour. The current evaluation time range is the current day, and the first historical time is the first three months before the current day. The first evaluation period is 6:00 to 7:00 and the last evaluation period is 18:00 to 19: 00. The current time is 9:30, 6:00 to 7:00 are the first evaluation period from the current day to the current time, 7:00 to 8:00 are the second evaluation period from the current day to the current time, and 8:00 to 9:00 are the last evaluation period from the current day to the current time. The designated driving behaviors are 106 forward collision alarms, 107 left lane departure alarms, 108 right lane departure alarms, 109 too-close distance alarms, 110 pedestrian collision alarms, 111 vehicle collisions, 112 vehicle rollover, 113 non-courtesy pedestrians, 114 fatigue driving, 115 calls, 116 smoking, 117 distraction warnings, 118 driver exceptions, 119 rapid acceleration warnings, 120 rapid deceleration warnings, 121 rapid left turn warnings, 122 rapid right turn warnings, 123 no-fastening safety belts, 124 overspeed, 125 overtime driving, 126 deviation routes, 127 stay overtime and the like. Where 106, 107, 108, etc. are numbers designating driving behaviors, respectively. Wherein the specified driving characteristics 102 comprise eight specified driving behaviors between 106 and 113; the specified driving characteristics 103 include five specified driving behaviors between 114 and 118; the designated driving characteristics 104 include seven designated driving behaviors between 119-125; the specified driving characteristics 105 comprise two specified driving behaviors between 126 and 127.

It should be noted that the designated driving characteristics 105 are reserved driving characteristics, and the designated driving behaviors included in the designated driving characteristics 105 can be realized by adding the designated driving behaviors according to needs and adopting an indefinite array structure.

For these three evaluation periods, the maximum frequency and the first frequency, respectively, of occurrences of each specified driving behavior in each evaluation period are seen in table 1.

TABLE 1

It should be noted that, after the computing device executes step 301, step 302 may be directly executed; or after the step 301 is executed, the step 303 is executed first, and then the step 302 is executed; alternatively, the computing device performs step 303 prior to performing

steps

301 and 302.

Step 302: the computing device determines a first weight and a first composite score for each designated driving behavior based on the maximum frequency and the first frequency of occurrence of each designated driving behavior.

The first weight is a weight of each specified driving behavior determined by the computing device according to the entropy method. The first composite score is an average score of each designated driving behavior from the first evaluation period to the current evaluation period within the current evaluation time range, and the first composite score can reflect the degree of goodness and badness of each designated driving behavior from the first evaluation period to the current evaluation period.

This step can be realized by the following steps (1) to (3), including:

(1) the computing device determines a first score for a plurality of evaluation periods for each designated driving behavior within a current evaluation time range based on the maximum frequency and the first frequency of occurrence of each designated driving behavior.

The step (1) may be realized by the following steps (1-1) to (1-3), including:

(1-1) for each specified driving behavior in each evaluation period, the computing device determines a ratio between a first frequency of occurrence of the specified driving behavior and a maximum frequency.

For example, freq for the first frequency_ijExpressed as freq for maximum frequency_threshiTo indicate that the ratio can be used

And (4) showing.

(1-2) the computing device determines a second score by which the specified driving behavior is deducted during the evaluation period based on the ratio.

In the embodiment of the present application, the scoring may be performed in a tenth system, a hundredth system or a thousandth system. For example, in the embodiment of the present application, when the score may be a percentage, the second score for the designated driving behavior to be deducted in the evaluation period may be used

And (4) showing. When the score is in the form of a dial, a second score indicating that the specified driving behavior is to be cancelled within the evaluation period may be used

And (4) showing.

(1-3) the computing device determines a first score of the designated driving behavior in the evaluation period according to the total score and the second score corresponding to the evaluation period.

The first score is the difference between the total score and the second score. For example, if the score is a percentage, the first score can be represented by the following formula four;

the formula four is as follows:

wherein, Score_ijScoring, freq, of the first index of the ith evaluation index in the jth evaluation period_ijFrequency one, freq_threshiIs the maximum frequency.

For example, if the first frequency of the 109 too-near distance alarm in the second evaluation period (7:00-8:00) is 2 and the maximum frequency is 8, the first score of the 109 too-near distance alarm in the second evaluation period is:

the first frequency of the 117 distraction prompt in the third evaluation period (8:00-9:00) is 1, and the maximum frequency is 8, then the first score of the 117 distraction prompt in the third evaluation period is:

the first frequency of the 121 sharp left turn warning in the first evaluation period (6:00-7:00) is 1, and the maximum frequency is 5, then the first score of the 121 sharp left turn warning in the first evaluation period is:

TABLE 2

The first scores for the remaining specified driving behaviors over the respective evaluation periods can be seen in table 2 above.

(2) The computing device determines a first weight for each designated driving behavior based on the first score and the second amount for each designated driving behavior for each evaluation period.

Wherein the second number is a number of the plurality of evaluation cycles. This step can be realized by the following steps (2-1) to (2-2), including:

(2-1) the computing device performing normalization processing based on the first score of each designated driving behavior in each evaluation period to obtain a first score of each designated driving behavior in each evaluation period; a first entropy value for each of the specified driving behaviors is determined based on the first score and the second quantity for each of the specified driving behaviors over each evaluation period.

In step (2-1), the computing device may construct a second determination matrix based on the first scores and the second numbers of the plurality of specified driving behaviors included in each of the specified driving characteristics. For example, the computing device constructs a second decision matrix R based on the specified driving behavior that the specified driving characteristics 102 include;

and the computing equipment performs range normalization processing on the second judgment matrix to obtain a third judgment matrix. For example, after the calculation device performs the range normalization on the second determination matrix R, a third determination matrix B is obtained, and each element in the third determination matrix B is a first score obtained after the range normalization is performed on the first score of each designated driving behavior in the second determination matrix R.

The computing device determines a first entropy value of each designated driving behavior based on a first score and a second quantity of each designated driving behavior in each evaluation period through the following formula one;

the formula I is as follows:

For example, the calculation device determines the first entropy value of each specified driving behavior by performing the calculation by the above formula one for each element in the third determination matrix B. One element in the third judgment matrix B corresponds to a specified driving behavior. The first element in the first row corresponds to a first designated driving behavior 106 in a first evaluation period, the second element in the first row corresponds to a second designated driving behavior 107 in the first evaluation period, the first element in the second row corresponds to a first designated driving behavior 106 in the second evaluation period, and the first element in the third row corresponds to a first designated driving behavior 106 in a third evaluation period. And the value of one element in the third judgment matrix corresponds to a value of the first score of the appointed driving behavior after range normalization processing.

The calculation device obtains H ═ 10.9460.9460.9660.946110.960 according to formula one]Wherein, in the step (A),

(2-2) the computing device determines a first weight for each specified driving behavior based on the first entropy and the first quantity for each specified driving behavior.

In the step, the computing device determines a weight matrix based on the first entropy and the first quantity of each designated driving behavior through the following formula two; a first weight for each specified driving behavior is determined from the weight matrix.

The formula II is as follows: w ═ ω_i')_1×t

Wherein, W is a weight matrix,

t is a first number.

And (3) the computing equipment brings the first entropy value and the first quantity of each specified driving behavior obtained in the step (1) into a formula II to obtain a weight matrix.

The number of the weight matrixes can be multiple, and one weight matrix can comprise one or more elements, wherein one element corresponds to one specified driving behavior.

For example, the computing device substitutes H ═ 10.9460.9460.9660.946110.960 into equation two, resulting in a weight matrix W ═ 00.2280.2280.1460.228000.169 specifying the driving behavior that the driving characteristics 102 include. Similarly, the computing device obtains a weight matrix W ═ 0.250.250.250.250 of the specified driving behavior included in the specified driving characteristics 103 according to the above method; the weight matrix of the specified driving behavior included in the specified driving characteristics 104 is W ═ 0.1970.1700.2660.19700.1700; the weight matrix of the specified driving behavior included in the specified driving characteristics 105 is W ═ 0.4810.519.

Wherein the step of the computing device determining the first weight of each specified driving behavior from the weight matrix may be:

one element in the weight matrix corresponds to a specified driving behavior, a first element in the weight matrix corresponds to a first specified driving behavior in a specified driving characteristic, and a second element corresponds to a second specified driving behavior in the specified driving characteristic. The values of the elements in the weight matrix are the first weights of the corresponding specified driving behavior. For example, the designated driving characteristics 102 include a first element in the weight matrix of the designated driving behavior corresponding to the first designated driving behavior 106, the value of the first element is a first weight of the designated driving behavior 106, and the first weight is 0; the second element corresponds to a second designated driving behavior 107, the value of the second element being a first weight of the designated driving behavior 107, the first weight being 0.228.

(3) The computing device determines a first composite score for each designated driving behavior based on the first score for each designated driving behavior for each evaluation period.

The first composite score is an average score of each designated driving behavior from the first evaluation period to the current evaluation period within the current evaluation time range, and the first composite score can reflect the degree of goodness and badness of each designated driving behavior from the first evaluation period to the current evaluation period. The first composite score for each designated driving behavior is an average of the first scores for that designated driving behavior over each evaluation period. For example, the designated driving characteristics 102 include designated driving behaviors of 106 forward collision warning, 107 left lane departure warning, 108 right lane departure warning, 109 vehicle distance too close warning, 110 pedestrian collision warning, 111 vehicle collision, 112 vehicle rollover, 113 no-courts pedestrian, respectively.

Then 106 the first composite score of the forward collision warning is:

the first composite score for a 107 left lane departure warning is:

108 the first composite score for the right lane departure warning is:

109 the first composite score for the warning of too close a distance is:

the first composite score of the 110 pedestrian collision avoidance warning is:

111 the first composite score for a vehicle collision is:

112 the first composite score for vehicle rollover is:

113 the first composite score for not giving the pedestrian a gift is:

i.e. the first composite scores of the specified driving behavior comprised by the specified driving characteristics 102 are score respectively₁₀₂＝[100 86.7 88.9 87.586.7 100 100 95.8]. Similarly, according to the above method, the first composite scores of the specified driving behaviors included in the specified driving characteristics 103 are determined to be score respectively₁₀₃＝[93.3 88.9 93.3 83.3 100](ii) a The first composite scores of the specified driving behaviors included in the specified driving characteristics 104 are score respectively₁₀₄＝[91.7 75 86.7 94.4 100 80 100](ii) a A first composite score specifying a driving behavior comprised by the driving characteristics 105Are respectively score₁₀₅＝[83.3 79.2]。

Step 303: the computing device determines a second weight for each of the first number of specified driving behaviors based on a degree of importance of the first specified driving behavior relative to the second specified driving behavior.

The second weight is obtained by the computing device according to an analytic hierarchy process, the first number is the number of the specified driving behaviors, and the first specified driving behavior and the second specified driving behavior are any two different specified driving behaviors in the first number of the specified driving behaviors.

This step can be realized by the following steps (1) to (4), including:

(1) the computing device constructs a first determination matrix based on a degree of importance of the first specified driving behavior relative to the second specified driving behavior.

The first and second designated driving behaviors are any two different designated driving behaviors of the first number of designated driving behaviors. The computing device may construct a first determination matrix corresponding to one specified driving characteristic based on any two different specified driving behaviors of a plurality of specified driving behaviors included in the specified driving characteristic, wherein the degree of importance of one specified driving behavior relative to another specified driving behavior.

The first decision matrix is denoted by A, then

Wherein n is the number of a plurality of specified driving behaviors contained in the specified driving characteristics, and the matrix element a_ij＞0，

When i is j, a_ij＝1。

Matrix element a_ijSee table 3 for the calibration method.

TABLE 3

Factor i to factor j	Quantized value
		Of equal importance	1
Of slight importance	3
		Of greater importance	5
Of strong importance	7
		Of extreme importance	9
Intermediate values of two adjacent judgments	2,4,6,8

In this step, the computing device scores a plurality of specified driving behaviors included in one specified driving feature according to table 3 to obtain a scoring result of each specified driving behavior, and constructs a first judgment matrix according to the scoring result. For example, the computing device obtains the scoring results of a plurality of specified driving behaviors contained in the specified driving characteristics 102 as: 106 forward collision alarm-9, 107 left lane departure alarm-6, 108 right lane departure alarm-6, 109 vehicle distance over-close alarm-4, 110 pedestrian collision alarm-8, 111 vehicle collision-9, 112 vehicle rollover-9, 113 no courtesy pedestrian-7.

The calculating device measures the importance degree of the two specified driving behaviors by using the ratio of the scores of the two specified driving behaviors, and a first judgment matrix is constructed by:

similarly, the evaluation results of the plurality of specified driving behaviors included in the specified driving characteristics 103 obtained by the computing device are respectively: 114 fatigue driving-9, 115 call receiving-7, 116 smoking-8, 117 distraction reminding-5 and 118 driver abnormity-9, and the first judgment matrix of five specified driving behaviors contained in the specified driving characteristics 103 is constructed as follows:

the calculation device obtains the scoring results of the plurality of specified driving behaviors contained in the specified driving characteristics 104 as follows: 119 sharp acceleration early warning-7, 120 sharp deceleration early warning-8, 121 sharp left turn early warning-7, 122 sharp right turn early warning-7, 123 no fastening of a safety belt-6, 124 overspeed-9 and 125 overtime driving-5, wherein a first judgment matrix of seven specified driving behaviors contained in the specified driving characteristics 104 constructed by the computing equipment is as follows:

the calculation device obtains the scoring results of the plurality of specified driving behaviors contained in the specified driving characteristics 105 as follows: 126 deviate from the route-5, 127 stay timeout-4, the computing device constructs a first decision matrix of two specified driving behaviors contained in the specified driving characteristics 105 as:

(2) the computing device determines a matrix order and a maximum eigenvalue of the first decision matrix based on the first decision matrix.

The matrix order is the number of rows and columns of the matrix, and in the embodiment of the application, the number of rows and columns of the first judgment matrix is equal. After the computing device constructs the first judgment matrix, the matrix order of the first judgment matrix can be directly determined. The computing device may determine a maximum eigenvalue of the first decision matrix from the eigenpolynomial of the matrix.

(3) The computing device determines a random consistency index of the first decision matrix from a correspondence of matrix orders and random consistency indices based on matrix orders.

The calculation device may pre-establish and store a correspondence between the matrix order and the random consistency index, and determine the random consistency index from the correspondence according to the matrix order. For example, the matrix order and the random consistency index corresponding relationship established by the computing device are shown in table 4.

TABLE 4

For the specified driving characteristics 102, the computing device constructs a first judgment matrix of a plurality of specified driving behaviors contained in the specified driving characteristics 102, and determines a random consistency index of the first judgment matrix corresponding to the specified driving characteristics 102 according to the corresponding relation. The designated driving characteristics 102 include eight designated driving behaviors, the matrix order of the constructed first judgment matrix is 8, and the random consistency index of the first judgment matrix with the matrix order of 8 is determined to be 1.41 according to the corresponding relationship between the matrix order and the random consistency index.

(4) The computing device determines a second weight for each of the first number of designated driving behaviors based on a matrix order, a maximum eigenvalue, and a random consistency index of the first decision matrix.

The step (4) may be realized by the following steps (4-1) to (4-4), including:

(4-1) the calculation device determines a consistency ratio of the first determination matrix based on the matrix order, the maximum eigenvalue, and the random consistency index of the first determination matrix.

In one possible implementation, the computing device may determine a consistency ratio of the first determination matrix based on a matrix order, a maximum eigenvalue, and a random consistency index of the first determination matrix by the following formula five;

the formula five is as follows:

wherein CR is a consistency ratio and RI isThe index of the consistency of the machine is,

λ_maxn is the matrix order for the maximum eigenvalue.

(4-2) when the consistency ratio is less than the ratio threshold, the computing device determines a first number of first feature vectors corresponding to the largest feature value.

The scale threshold may be set and changed as needed, for example, the scale threshold may be 0.08, 0.1, or 0.11, etc. In the examples of the present application, the comparative example threshold value is not particularly limited. When the consistency proportion is smaller than the proportion threshold value, the computing equipment determines that the first judgment matrix passes consistency check; and (3) when the consistency ratio is not less than the ratio threshold, the computing equipment determines that the first judgment matrix does not pass the consistency check, and the step (1) is executed again to adjust and correct the first judgment matrix. For example, in the present embodiment, the ratio threshold may be 0.1. That is, when the consistency ratio is less than 0.1, the computing device determines a first number of first feature vectors corresponding to the largest feature.

And (4-3) the computing equipment performs normalization processing on the first quantity of first feature vectors to obtain a first quantity of second feature vectors.

In this step, the computing device normalizes the first feature vector to obtain a second feature vector, which reflects the relative importance among the plurality of designated driving behaviors.

(4-4) the computing device determines the numerical values of the first number of second eigenvectors as second weights of the first number of specified driving behaviors.

The feature vector is directional, and the computing device removes the direction of the second feature vector and determines the value of the second feature vector as the second weight. For example, for the specified driving characteristics 102, if the computing device determines that the consistency ratio of the first judgment matrix of the specified driving characteristics is smaller than the ratio threshold, the first judgment matrix of the specified driving characteristics passes the consistency check, and the computing device determines that the second weight of the specified driving behavior included in the specified driving characteristics is: w₁₀₂'＝[0.155 0.1030.103 0.069 0.138 0.155 0.155 0.121](ii) a For the specified driving characteristics 103, the computing device determines that the consistency ratio of the first judgment matrix of the specified driving characteristics is less than the ratio threshold value, the first judgment matrix of the specified driving characteristics passes the consistency check, and the computing device determines that the second weight of the specified driving behaviors contained in the specified driving characteristics is W₁₀₃'＝[0.237 0.184 0.211 0.132 0.237](ii) a For the specified driving characteristics 104, the computing device determines that the consistency ratio of the first judgment matrix of the specified driving characteristics is less than the ratio threshold value, the first judgment matrix of the specified driving characteristics passes the consistency check, and the computing device determines that the second weight of the specified driving behaviors contained in the specified driving characteristics is W₁₀₄'＝[0.1430.163 0.143 0.143 0.122 0.184 0.102](ii) a For the specified driving characteristics 105, the computing device determines that the consistency proportion of the first judgment matrix of the specified driving characteristics is less than the proportion threshold value, the first judgment matrix of the specified driving characteristics passes the consistency check, and the computing device determines that the second weight of the specified driving behaviors contained in the specified driving characteristics is W₁₀₅'＝[0.556 0.444]。

Step 304: the computing device determines a composite weight for each designated driving behavior based on the first weight and the second weight for each designated driving behavior.

The calculation equipment determines the comprehensive weight of each designated driving behavior based on the first weight and the second weight, the comprehensive weight is determined after the calculation equipment combines the first weight and the second weight, and the importance and the actual condition of the problem of each designated driving behavior can be objectively and comprehensively reflected.

This step can be realized by the following steps (1) to (3), including:

(1) for each designated driving behavior, the computing device determines a product of a first weight and a second weight of the designated driving behavior, resulting in a first numerical value.

In this step, for each specified driving behavior, the computing device may determine a first numerical value by the following formula six;

formula six: q₁＝ω_i'ω_i”

Wherein Q is₁Is the first value, ω, of the specified driving behavior_i' is a first weight, ω, of the specified driving behavior_i"is the second weight for the specified driving behavior.

(2) The computing device sums the first values for each designated driving behavior to obtain a second value.

The computing device determines a plurality of specified driving behaviors contained in each specified driving feature, and sums first numerical values of the specified driving behaviors to obtain a second numerical value. For a plurality of specified driving behaviors contained in each specified driving characteristic, the computing device may determine a second numerical value of the specified driving characteristic by the following formula seven;

the formula seven:

wherein Q is₂Is a second value, t is a first quantity, ω_i' first weight, ω, for the ith designated driving behavior_i"is the second weight of the ith designated driving behavior.

For example, if the specified driving characteristics 102 include eight specified driving behaviors, the computing device sums the first values of the eight specified driving behaviors to obtain a second value of the specified driving characteristics 102; the specified driving characteristics 103 comprise five specified driving behaviors, the computing device sums the first values for the five specified driving behaviors to obtain a second value for the specified driving characteristics 103.

(3) The computing device determines a ratio of the first value to the second value to obtain a composite weight for the specified driving behavior.

The computing device uses the ratio of the first value to the second value as the composite weight for the specified driving behavior. The computing device may determine a composite weight for the specified driving behavior by the following formula eight;

the formula eight:

wherein Q is the integrated weight of the specified driving behavior, Q₁、Q₂A first value and a second value, respectively.

For the eight specified driving behaviors contained in the specified driving characteristics 102, respectively: 106 forward collision warning, 107 left lane departure warning, 108 right lane departure warning, 109 vehicle distance over-approach warning, 110 pedestrian collision warning, 111 vehicle collision, 112 vehicle rollover, 113 non-courtesy pedestrian.

The calculation device determines, by formula eight, that the comprehensive weights of the specified driving behaviors contained in the specified driving characteristics 102 are respectively:

that is, the comprehensive weights of the designated driving behaviors included in the designated driving characteristics 102 are respectively: u shape₁₀₂＝[0 0.2160.216 0.092 0.288 0 0 0.187]。

Similarly, the computing device determines, by formula eight, that the comprehensive weights of the specified driving behaviors contained in the specified driving characteristics 103 are respectively: u shape₁₀₃＝[0.310 0.241 0.276 0.172 0]。

The integrated weights of the designated driving behaviors included in the designated driving characteristics 104 are respectively U₁₀₄＝[0.184 0.1810.247 0.184 0 0.204 0]。

The integrated weights of the specified driving behaviors included in the specified driving characteristics 105 are respectively U₁₀₅＝[0.537 0.463]。

The computing device determines the values of the first weight, the second weight, and the composite weight of the specified driving behavior encompassed by each specified driving characteristic, as can be seen in table 5.

TABLE 5

Step 305: the computing device determines a third score for each of a third number of specified driving characteristics based on the composite weight for each of the specified driving behaviors and the first composite score.

For each specified driving feature, the computing device determines each specified driving behavior contained in the specified driving feature, and performs weighted summation on the comprehensive weight of each specified driving behavior contained in the specified driving feature and the first comprehensive score to obtain a third score of the specified driving feature.

For example, for a specified driving characteristic 102, the computing device determines that the total weight of each of eight specified driving behaviors included in the specified driving characteristic is respectively U₁₀₂＝[0 0.216 0.216 0.092 0.288 0 00.187]The first composite scores are score respectively₁₀₂＝[100 86.7 88.9 87.5 86.7 100 100 95.8]The computing device weights and sums the composite weight of each designated driving behavior with the first composite Score to obtain a third Score, of the designated driving characteristics₁₀₂'(100×0+86.7×0.216+88.9×0.216+87.5×0.092+86.7×0.288+100×0+100×0+；95.8×0.187)＝88.9 minutes

For the specified driving characteristics 103, the computing device determines that the comprehensive weight of each specified driving behavior in the five specified driving behaviors contained in the specified driving characteristics is respectively U₁₀₃＝[0.310 0.241 0.276 0.172 0]The first composite scores are score respectively₁₀₃＝[93.3 88.9 93.3 83.3 100]The computing device weights and sums the first composite Score and the composite weight of each designated driving behavior to obtain a third Score of the designated driving behavior, Score₁₀₃' (93.3 × 0.31+88.9 × 0.241+93.3 × 0.276+83.3 × 0.172.172 +100 × 0) ═ 90.4 points;

for the specified driving characteristics 104, the computing device determines that the comprehensive weight of each specified driving behavior in seven specified driving behaviors contained in the specified driving characteristics is respectively U₁₀₄＝[0.184 0.181 0.247 0.184 0 0.204 0]The first composite scores are score respectively₁₀₄＝[91.7 75 86.7 94.4 100 80 100]The computing device weights and sums the first composite Score and the composite weight of each designated driving behavior to obtain a third Score of the designated driving behavior, Score₁₀₄(91.7 × 0.184+75 × 0.181+86.7 × 0.247+94.4 × 0.184.184 +100 × 0+80 × 0.204.204 +100 × 0) ═ 85.6 minutes;

for a given driving characteristic 105, the computing device determines that the overall weight of each of two given driving behaviors included in the given driving characteristic is respectively U₁₀₅＝[0.537 0.463]The first composite scores are score respectively₁₀₅＝[83.3 79.2]The computing device weights and sums the first composite Score and the composite weight of each designated driving behavior to obtain a third Score of the designated driving behavior, Score₁₀₅' (83.3 × 0.537+79.2 × 0.463) ═ 81.4 points the third score for each specified driving characteristic can be seen in table 6.

TABLE 6

Step 306: the computing device determines a third weight for each specified driving feature based on the degree of importance of the first specified driving feature relative to the second specified driving feature.

The third weight is a weight obtained by the computing device according to an analytic hierarchy process. The first designated driving characteristic and the second designated driving characteristic are any two different designated driving characteristics of a third number of designated driving characteristics.

The step of the computing device determining the third weight for each specified driving feature in this step may be realized by the following steps (1) to (4), including:

(1) the computing device constructs a fourth determination matrix based on the degree of importance of the first specified driving characteristic relative to the second specified driving characteristic.

(2) The computing device determines a matrix order and a maximum eigenvalue of a fourth decision matrix based on the fourth decision matrix.

(3) The calculation device determines a random consistency index of the fourth determination matrix from a correspondence between the matrix order and the random consistency index based on the matrix order.

(4) The calculation device determines a third weight of each of the third number of designated driving characteristics based on the matrix order, the maximum eigenvalue, and the random consistency index of the fourth determination matrix.

Steps (1) to (4) are similar to steps (1) to (4) in step 303, respectively, and are not described again here.

For example, for the specified driving characteristics 102, 103, 104, and 105, the computing device scores the plurality of specified driving characteristics according to table 3, obtains a score result of each of the specified driving characteristics, and constructs a fourth determination matrix according to the score result. The calculation device obtains the scoring results of the plurality of specified driving characteristics as follows: driving characteristics 102-8 are specified, driving characteristics 103-5 are specified, driving characteristics 104-7 are specified, and driving characteristics 105-2 are specified.

The fourth decision matrix constructed by the computing device is

The computing device determines a matrix order, a maximum eigenvalue, and a random consistency index of the fourth decision matrix, and determines based on the matrix order, the maximum eigenvalue, and the random consistency indexA third weight is assigned to each of the designated driving characteristics. If the consistency ratio of the fourth judgment matrix is smaller than the ratio threshold, the computing device determines that the fourth judgment matrix passes the consistency test, and the computing device determines that the third weight of each specified driving behavior is K ═ 0.3640.2270.3180.091]Wherein the third weight for the driving characteristic 102 is assigned to 0.364, the third weight for the driving characteristic 103 is assigned to 0.227, the third weight for the driving characteristic 104 is assigned to 0.318, and the third weight for the driving characteristic 105 is assigned to 0.091.

It should be noted that the first weight is a weight determined by the computing device according to an entropy method and is an objective weight, and the second weight is a weight determined by the computing device according to an analytic hierarchy method and is a subjective weight. The comprehensive weight is a weight obtained by combining the first weight and the second weight, and the comprehensive weight comprises the first weight, so that when the computing equipment determines the third weight of each specified driving feature, the computing equipment only needs to determine the third weight according to an analytic hierarchy process and does not need to determine according to an entropy method. If the computing device still determines the weights according to the entropy method, there may be cases where the first scores of the specified driving behavior are reused for each evaluation period.

Step 307: the computing device determines a second composite score based on the third score and the third weight for each specified driving feature.

The second comprehensive score is a comprehensive score of the driving behavior of the target driver, the computing device evaluates the driving behavior of the target driver through the second comprehensive score, the second comprehensive score can reflect the overall performance of the driving behavior of the target driver, and the higher the second comprehensive score is, the more standard the driving behavior of the target driver is. This step can be implemented in the following two implementations.

In a first implementation, the computing device performs weighted summation on the third score and the third weight of each designated driving feature to obtain a second composite score.

In this implementation, the computing device may determine the second composite score by the following formula nine;

the formula is nine:

wherein Y is the second composite Score, Score_xA third score, K, corresponding to the x-th designated driving characteristic_xA third weight corresponding to the xth specified driving characteristic, and m is a third number.

For example, the computing device determines that the third scores corresponding to the specified driving characteristics are Score [ 88.990.485.681.4 ], the third weights are K [ 0.3640.2270.3180.091 ], and the second composite Score Y is 88.9 × 0.364+90.4 × 0.227+85.6 × 0.318+81.4 × 0.091-87.5.

In a second implementation manner, the computing device performs weighted summation on the third score and the third weight of each designated driving feature to obtain a third comprehensive score, and fuses the third comprehensive score and a plurality of fourth comprehensive scores of the driving behavior of the target driver in a second historical time to obtain a second comprehensive score.

In a second implementation, the computing device may be implemented by the following steps (1) to (2), including:

(1) and the computing equipment performs weighted summation on the third score and the third weight of each appointed driving feature to obtain a third comprehensive score.

The step of obtaining the third composite score by the computing device in this step is similar to the step of obtaining the second composite score by the computing device in the first implementation manner, and is not described herein again.

(2) And the computing equipment fuses the third comprehensive score and a plurality of fourth comprehensive scores of the driving behavior of the target driver in the second historical time to obtain a second comprehensive score.

The second historical time may be set and changed as needed, and in the embodiment of the present application, the second historical time is not specifically limited. For example, the second historical time may be the first 29 days before the current day, the first 30 days before the current day, the first 20 days before the current day, or the like. The fourth composite score is similar to the third composite score, and the computing device may obtain a plurality of fourth composite scores by weighted summation of the third score and the third weight for each of the specified driving characteristics in a plurality of evaluation periods included each day during the second historical time.

The step (2) may be realized by the following steps (2-1) to (2-4) including:

(2-1) the computing device obtains a target Gaussian mixture model.

The computing equipment can train the initial Gaussian mixture model by itself to obtain a target Gaussian mixture model; the computing device may also obtain a target gaussian mixture model trained by other devices. In the embodiments of the present application, this is not particularly limited. When the computing device trains the initial gaussian mixture model by itself to obtain the target gaussian mixture model, the step of determining the target gaussian mixture model by the computing device may be:

(2-1-1) the computing device initializes a plurality of first model parameters of the initial Gaussian mixture model to obtain a plurality of second model parameters.

The plurality of first model parameters are respectively a first parameter, a second parameter and a third parameter, wherein the first parameter refers to a mean value, the second parameter refers to a variance, and the third parameter refers to a coefficient. The computing device initializes the plurality of first model parameters to obtain a plurality of second model parameters.

(2-1-2) the computing equipment determines the number of component models in the initial Gaussian mixture model, and sets a convergence threshold value and the maximum iteration number.

The computing device may determine a number of partial models from a scatter plot of the third composite score and the plurality of fourth composite scores. And the computing equipment marks the third comprehensive score and a plurality of fourth comprehensive scores in the second historical time in one-dimensional coordinates respectively, and determines the number of the sub-models according to the number of clusters formed by the third comprehensive score and the fourth comprehensive scores. For example, the second historical time is the first 29 days prior to the current day, the plurality of fourth composite scores are fourth composite scores for each of the 29 days, and the plurality of fourth composite scores obtained by the computing device are 89.4, 92.7, 90.5, 93.5, 92.6, 95.5, 61.8, 93.0, 92.7, 93.0, 88.4, 92.6, 93.0, 89.5, 63.5, 90.5, 93.3, 92.6, 88.5, 92.4, 93.9, 92.0, 93.6, 88.5, 60.5, 93.5, 88.9, 65.5, 75.7, respectively. Referring to fig. 4, the computing device marks the third composite score and the plurality of fourth composite scores in the one-dimensional coordinates, respectively, and it can be seen that the third composite score and the plurality of fourth composite scores approximately form three clusters, and the computing device determines that the number of the partial models is three.

The convergence threshold may be set and changed as needed, for example, the convergence threshold may be 0.000001, 0.000002, 0.00001, or the like. The maximum number of iterations may also be set and changed as needed, for example, the maximum number of iterations may be 30, 40, or 50, etc. In the embodiment of the present application, neither the convergence threshold nor the maximum number of iterations is specifically limited.

(2-1-3) the computing device determines a first responsiveness of the first partial model to the third composite score and the plurality of fourth composite scores based on the plurality of second model parameters.

The computing device may determine a plurality of first responsibilities of the first scoring model to the third composite score and the plurality of fourth composite scores by equation ten below;

formula ten:

wherein the content of the first and second substances,

first responsiveness to the vth third composite score or the th composite score for the kth partial model, α_kIs the third parameter of the kth model, phi (y)_v|θ_k) Is the Gaussian distribution density of the kth partial model, R is the number of partial models, and k is 1 in the step.

(2-1-4) the computing device determines a plurality of third model parameters based on the first responsiveness, the third composite score, and the plurality of fourth composite scores.

The computing device may determine a plurality of third model parameters by the following equations eleven to thirteen;

formula eleven:

equation twelve:

formula thirteen:

wherein the content of the first and second substances,

first responsiveness, y, to the kth partial model to the vth third composite score or the vth fourth composite score_vIs the vth third composite score or the fourth composite score,

is the first parameter, σ, of the kth partial model_k ²As a second parameter of the kth partial model, α_kAnd N is the sum of the number of the third comprehensive scores and the number of the plurality of fourth comprehensive scores.

The first parameter mu_kIs a mean value, a second parameter σ_k ²As variance, the third parameter α_kAre coefficients.

(2-1-5) when the plurality of third model parameters meet the convergence threshold or reach the maximum iteration number, outputting the plurality of third model parameters by the computing equipment to obtain the target Gaussian mixture model.

In this step, the plurality of third model parameters output by the computing device are corresponding model parameters when the convergence threshold is met or the maximum iteration number is reached.

(2-1-6) when the plurality of third model parameters do not meet the convergence threshold value or the maximum iteration number is not reached, determining a plurality of fourth model parameters based on the plurality of third model parameters, the third comprehensive score and the plurality of fourth comprehensive scores until the plurality of model parameters meet the convergence threshold value or the maximum iteration number is reached, and outputting the corresponding plurality of model parameters by the computing equipment when the convergence threshold value or the maximum iteration number is reached to obtain the target Gaussian mixture model.

When the plurality of third model parameters do not meet the convergence threshold or do not reach the maximum iteration number, the computing equipment continues to determine the model parameters through the formulas from ten to thirteen until the convergence threshold is met or the maximum iteration number is reached, and outputs the plurality of third model parameters as the plurality of corresponding model parameters when the convergence threshold is met or the maximum iteration number is reached.

(2-2) the computing device determines a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores through the target gaussian mixture model.

The computing equipment determines a first probability and a plurality of second probabilities through a target Gaussian mixture model and the following formula III;

the formula III is as follows: m ═ P (y)_v|θ_k)*2*y_const

Wherein the content of the first and second substances,

It should be noted that the formula three is obtained by transforming, by the computing device, the following formula fourteen:

wherein phi (y)_v|θ_k) Is the gaussian distribution density of the kth partial model. In the formula fourteen, integral operation is performed, and see fig. 5. However, the process of the integral operation is complex, and the calculation efficiency is affected. Therefore, in the present application, the formula fourteen is transformed to obtain the formula three, and the length of the formula three is P (y)_v|θ_k) And a width of 2 × y_constIs substituted for the function P (y)_v|θ_k) From y_v-y_constTo y_v+y_constIntegral of (a), P (y)_v|θ_k) And 2 × y_constAre all constants, see fig. 6.

When the first probability and the plurality of second probabilities obtained by the computing device through the Gaussian mixture model are calculated, the lower probability is given to the third comprehensive score or the fourth comprehensive score with the lower occurrence frequency, and therefore the influence of the third comprehensive score or the fourth comprehensive score with the lower occurrence frequency on the second comprehensive score is reduced.

And (2-3) the computing equipment normalizes the first probability and the plurality of second probabilities to respectively obtain a fourth weight corresponding to the third comprehensive score and a plurality of fifth weights corresponding to the plurality of fourth comprehensive scores.

A fourth comprehensive score corresponds to a fifth weight, and a formula III can ensure that the ratio of the first probability or the second probabilities to the total probability is unchanged before and after transformation, so that complex integral operation is avoided and the calculation efficiency is improved under the condition of ensuring the accuracy. Under the condition that the ratio of the first probability or the second probability to the total probability is unchanged, the computing device normalizes the first probability and the plurality of second probabilities to obtain a fourth weight and a plurality of fifth weights which are unchanged. In this step, the process of normalizing the first probability and the plurality of second probabilities by the computing device is similar to the process of normalizing the first feature vector by the computing device in step 303, and is not described herein again.

For example, the values of the plurality of fifth weights determined by the computing device may be seen in table 7.

TABLE 7

Fourth composite score	Fifth weight	Fourth composite score	Fifth weight
				89.4	0.02306	90.5	0.01391
92.7	0.06374	93.3	0.07059
				90.5	0.01391	92.6	0.05890
93.5	0.06338	88.5	0.00414
				92.6	0.05890	92.4	0.04770
95.5	0.00051	93.9	0.04131
				61.8	0.00157	92.0	0.02538
93.0	0.07265	93.6	0.05848
				92.7	0.06374	88.5	0.00414
93.0	0.07265	60.5	0.00141
				88.4	0.00310	93.5	0.06339
92.6	0.05890	88.9	0.01107
				93.0	0.07265	65.5	0.00196
89.5	0.02485	75.7	0.00178
				63.5	0.00177	/	/

The third composite score was 87.5 points and the fourth weight determined by the computing device to be 0.00050.

And (2-4) the computing equipment carries out weighted summation on the third comprehensive score, the fourth weights, the plurality of fourth comprehensive scores and the plurality of fifth weights to obtain a second comprehensive score.

In this step, the computing device performs weighted summation on the fourth weight corresponding to the third composite score and the fifth weights corresponding to the fourth composite scores to obtain a second composite score. The computing device evaluates the driving behavior of the target driver according to the level of the second composite score.

The point to be described is that the computing device determines the second comprehensive score according to the third comprehensive score in the current evaluation time range and the plurality of fourth comprehensive scores in the second historical time, so that the overall driving behavior of the target driver can be reflected more comprehensively and objectively, the situation that the overall evaluation is influenced by a few extreme values is avoided, and the accuracy of the driving behavior evaluation is improved.

In a second implementation manner, for the third composite score and the data in table 7, the second composite score obtained by the computing device according to the method is 92.4 points. If the average value of the third composite score and the plurality of fourth composite scores is calculated, the result is 87.3 points. It can be seen that: the result is lower than the second composite score. The reason for this is that the fourth composite score among table 7 was low, 61.8, 63.5, 60.5, 65.5 and 75.7 points, respectively. These several values pull down the overall calculation. Therefore, with the average method, a few low scores will lower the evaluation result, and a few high scores will also raise the evaluation result, thereby causing the evaluation result not to reflect the overall driving behavior of the target driver.

An embodiment of the present application provides a driving behavior evaluation device, see fig. 7, the device including:

the obtaining module 701 is configured to obtain a maximum frequency of occurrence of each specified driving behavior of the target driver in a first historical time, and a first frequency of occurrence of each specified driving behavior in a current evaluation time range;

a first determining module 702, configured to determine a first weight and a first composite score of each designated driving behavior based on a maximum frequency and a first frequency of occurrence of each designated driving behavior;

a second determining module 703, configured to determine, based on an importance degree of the first specified driving behavior relative to the second specified driving behavior, a second weight of each specified driving behavior in the first number of specified driving behaviors, where the first number is the number of specified driving behaviors, and the first specified driving behavior and the second specified driving behavior are any two different specified driving behaviors in the first number of specified driving behaviors;

a third determining module 704 for determining a second composite score of the driving behavior of the target driver based on the first weight, the first composite score and the second weight of each designated driving behavior.

In a possible implementation manner, the first determining module 702 is further configured to determine, based on the maximum frequency and the first frequency of occurrence of each designated driving behavior, a first score of a plurality of evaluation periods of each designated driving behavior within the current evaluation time range; determining a first weight of each designated driving behavior based on a first score and a second number of each designated driving behavior in each evaluation period, wherein the second number is the number of a plurality of evaluation periods; a first composite score for each designated driving behavior is determined based on the first score for each designated driving behavior for each evaluation period.

In another possible implementation manner, the first determining module 702 is further configured to determine, for each evaluation period for each specified driving behavior, a ratio between a first frequency of occurrence of the specified driving behavior and a maximum frequency; determining a second score for the designated driving behavior to be deducted within the evaluation period according to the ratio; and determining a first score of the designated driving behavior in the evaluation period according to the total score and the second score corresponding to the evaluation period.

In another possible implementation manner, the first determining module 702 is further configured to determine a first entropy value of each specified driving behavior based on the first score and the second number of each specified driving behavior in each evaluation period; a first weight for each designated driving behavior is determined based on the first entropy and the first quantity for each designated driving behavior.

In another possible implementation manner, the first determining module 702 is further configured to perform normalization processing based on the first score of each designated driving behavior in each evaluation period, so as to obtain a first score of each designated driving behavior in each evaluation period; determining a first entropy value of each designated driving behavior based on the first score and the second quantity of each designated driving behavior in each evaluation period through the following formula I;

the formula I is as follows:

In another possible implementation manner, the first determining module 702 is further configured to determine a weight matrix based on the first entropy and the first number of each specified driving behavior, where the weight matrix includes a first weight of each specified driving behavior according to the following formula two;

the formula II is as follows: w ═ ω_i')_1×t

Wherein, W is a weight matrix,

t is a first number;

In another possible implementation manner, the second determining module 703 is further configured to construct a first determining matrix based on the importance degree of the first specified driving behavior relative to the second specified driving behavior; determining a matrix order and a maximum eigenvalue of the first judgment matrix based on the first judgment matrix; determining a random consistency index of the first judgment matrix from a corresponding relation between the matrix order and the random consistency index based on the matrix order; and determining a second weight of each designated driving behavior in the first number of designated driving behaviors based on the matrix order, the maximum eigenvalue and the random consistency index of the first judgment matrix.

In another possible implementation manner, the second determining module 703 is further configured to determine a consistency ratio of the first determination matrix based on a matrix order, a maximum eigenvalue, and a random consistency index of the first determination matrix; when the consistency ratio is smaller than a ratio threshold, determining a first number of first feature vectors corresponding to the maximum feature value; normalizing the first quantity of first feature vectors to obtain a first quantity of second feature vectors; the values of the first number of second eigenvectors are determined as second weights for the first number of specified driving behaviors.

In another possible implementation manner, the third determining module 704 is further configured to determine a comprehensive weight of each designated driving behavior based on the first weight and the second weight of each designated driving behavior; determining a third score of each designated driving feature in a third number of designated driving features based on the comprehensive weight and the first comprehensive score of each designated driving behavior, wherein the third number is the number of the designated driving features, and each designated driving feature corresponds to a plurality of designated driving behaviors; determining a third weight for each of the specified driving characteristics based on the degree of importance of the first specified driving characteristic relative to the second specified driving characteristic, the first specified driving characteristic and the second specified driving characteristic being any two different specified driving characteristics of a third number of specified driving characteristics; a second composite score is determined based on the third score and the third weight for each designated driving feature.

In another possible implementation manner, the second determining module 703 is further configured to determine, for each specified driving behavior, a product of a first weight and a second weight of the specified driving behavior, so as to obtain a first numerical value; summing the first numerical values of each designated driving behavior to obtain a second numerical value; and determining the ratio of the first value to the second value to obtain the comprehensive weight of the appointed driving behavior.

In another possible implementation manner, the second determining module 703 is further configured to perform weighted summation on the third score and the third weight of each specified driving feature to obtain a second composite score; or carrying out weighted summation on the third score and the third weight of each designated driving feature to obtain a third comprehensive score, and fusing the third comprehensive score and a plurality of fourth comprehensive scores of the driving behavior of the target driver in the second historical time to obtain a second comprehensive score.

In another possible implementation manner, the second determining module 703 is further configured to determine, through a target gaussian mixture model, a first probability corresponding to a third composite score and a plurality of second probabilities corresponding to a plurality of fourth composite scores; normalizing the first probability and the plurality of second probabilities to respectively obtain a fourth weight corresponding to the third comprehensive score and a plurality of fifth weights corresponding to the plurality of fourth comprehensive scores, wherein one fourth comprehensive score corresponds to one fifth weight; and carrying out weighted summation on the third comprehensive score, the fourth weight, the plurality of fourth comprehensive scores and the plurality of fifth weights to obtain a second comprehensive score.

In another possible implementation manner, the apparatus further includes:

the fourth determining module is used for initializing a plurality of first model parameters of the initial Gaussian mixture model to obtain a plurality of second model parameters; determining the number of component models in the initial Gaussian mixture model, and setting a convergence threshold value and the maximum iteration times; determining a first responsiveness of the first sub-model to the third composite score and the plurality of fourth composite scores based on the plurality of second model parameters; determining a plurality of third model parameters based on the first responsiveness, the third composite score, and the plurality of fourth composite scores; when the plurality of third model parameters meet the convergence threshold value or reach the maximum iteration times, outputting the plurality of third model parameters to obtain a target Gaussian mixture model; and when the plurality of third model parameters do not meet the convergence threshold value or do not reach the maximum iteration times, determining a plurality of fourth model parameters based on the plurality of third model parameters, the third comprehensive score and the plurality of fourth comprehensive scores until the plurality of model parameters meet the convergence threshold value or reach the maximum iteration times, and outputting the plurality of corresponding model parameters when the plurality of model parameters meet the convergence threshold value or reach the maximum iteration times to obtain the target Gaussian mixture model.

In another possible implementation manner, the second determining module 703 is further configured to determine the first probability and the plurality of second probabilities through a target gaussian mixture model and the following formula three;

the formula III is as follows: m ═ P (y)_v|θ_k)*2*y_const

Wherein the content of the first and second substances,

m is the first probability or the second probabilityTwo probabilities, phi (y)_v|θ_k) Is the Gaussian distribution density of the kth partial model, y_vIs the third composite score or the fourth composite score, y_constIs a constant, mu_kIs a first parameter, σ_k ²As a second parameter, α_kAnd R is the number of the partial models, and k is less than or equal to R.

According to the driving behavior evaluation device provided by the embodiment of the application, the computing equipment obtains the maximum frequency of each appointed driving behavior of the target driver in the first historical time and the first frequency of each appointed driving behavior in the current evaluation time range, and determines the first weight and the first comprehensive score of each appointed driving behavior according to the maximum frequency and the first frequency. The computing device also determines a second weight for each of the first number of specified driving behaviors based on a degree of importance of the first specified driving behavior relative to the second specified driving behavior. The computing device determines a second composite score for the driving behavior of the target driver based on the first weight, the first composite score, and the second weight for each designated driving behavior. The device determines the second comprehensive score through the first comprehensive score and two different weights, avoids the condition that a few extreme values influence the overall evaluation, and improves the accuracy of the driving behavior evaluation.

It should be noted that: the driving behavior evaluation device provided in the above embodiment is only illustrated by dividing the functional modules when evaluating the driving behavior, and in practical applications, the functions may be distributed by different functional modules as needed, that is, the internal structure of the device may be divided into different functional modules to complete all or part of the functions described above. In addition, the driving behavior evaluation device provided by the above embodiment and the driving behavior evaluation method embodiment belong to the same concept, and specific implementation processes thereof are detailed in the method embodiment and are not described herein again.

Fig. 8 is a block diagram of a computing device 800 according to an embodiment of the present disclosure. For example, the computing device 800 may be used to execute the driving behavior evaluation methods provided in the various embodiments described above. Referring to fig. 8, the computing device 800 includes: a processor 801 and a memory 802.

The processor 801 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and so forth. The processor 801 may be implemented in at least one hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 801 may also include a main processor and a coprocessor, where the main processor is a processor for processing data in an awake state, and is also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 801 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the display screen. In some embodiments, the processor 801 may further include an AI (Artificial Intelligence) processor for processing computing operations related to machine learning.

Memory 802 may include one or more computer-readable storage media, which may be non-transitory. Memory 802 may also include high speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in memory 802 is used to store at least one instruction for execution by processor 801 to implement the driving behavior assessment methods provided by method embodiments herein.

In some embodiments, the computing device 800 may also optionally include: a peripheral interface 803 and at least one peripheral. The processor 801, memory 802 and peripheral interface 803 may be connected by bus or signal lines. Various peripheral devices may be connected to peripheral interface 803 by a bus, signal line, or circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 804, a display 805, a camera 806, an audio circuit 807, a positioning component 808, and a power supply 809.

The peripheral interface 803 may be used to connect at least one peripheral related to I/O (Input/Output) to the processor 801 and the memory 802. In some embodiments, the processor 801, memory 802, and peripheral interface 803 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 801, the memory 802, and the peripheral interface 803 may be implemented on separate chips or circuit boards, which are not limited by this embodiment.

The Radio Frequency circuit 804 is used for receiving and transmitting RF (Radio Frequency) signals, also called electromagnetic signals. The radio frequency circuitry 804 communicates with communication networks and other communication devices via electromagnetic signals. The rf circuit 804 converts an electrical signal into an electromagnetic signal to be transmitted, or converts a received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 804 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a subscriber identity module card, and so forth. The radio frequency circuitry 804 may communicate with other computing devices via at least one wireless communication protocol. The wireless communication protocols include, but are not limited to: the world wide web, metropolitan area networks, intranets, generations of mobile communication networks (2G, 3G, 4G, and 5G), Wireless local area networks, and/or WiFi (Wireless Fidelity) networks. In some embodiments, the radio frequency circuit 804 may further include NFC (Near Field Communication) related circuits, which are not limited in this application.

The display screen 805 is used to display a UI (User Interface). The UI may include graphics, text, icons, video, and any combination thereof. When the display 805 is a touch display, the display 805 also has the ability to capture touch signals on or above the surface of the display 805. The touch signal may be input to the processor 801 as a control signal for processing. At this point, the display 805 may also be used to provide virtual buttons and/or a virtual keyboard, also referred to as soft buttons and/or a soft keyboard. In some embodiments, the display 805 may be one, providing the front panel of the computing device 800; in other embodiments, the display 805 may be at least two, each disposed on a different surface of the computing device 800 or in a folded design; in still other embodiments, the display 805 may be a flexible display disposed on a curved surface or on a folded surface of the computing device 800. Even further, the display 805 may be arranged in a non-rectangular irregular pattern, i.e., a shaped screen. The Display 805 can be made of LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode), and other materials.

The camera assembly 806 is used to capture images or video. Optionally, camera assembly 806 includes a front camera and a rear camera. Typically, the front facing camera is disposed on the front panel of the computing device and the rear facing camera is disposed on the back of the computing device. In some embodiments, the number of the rear cameras is at least two, and each rear camera is any one of a main camera, a depth-of-field camera, a wide-angle camera and a telephoto camera, so that the main camera and the depth-of-field camera are fused to realize a background blurring function, and the main camera and the wide-angle camera are fused to realize panoramic shooting and VR (Virtual Reality) shooting functions or other fusion shooting functions. In some embodiments, camera assembly 806 may also include a flash. The flash lamp can be a monochrome temperature flash lamp or a bicolor temperature flash lamp. The double-color-temperature flash lamp is a combination of a warm-light flash lamp and a cold-light flash lamp, and can be used for light compensation at different color temperatures.

The audio circuit 807 may include a microphone and a speaker. The microphone is used for collecting sound waves of a user and the environment, converting the sound waves into electric signals, and inputting the electric signals to the processor 801 for processing or inputting the electric signals to the radio frequency circuit 804 to realize voice communication. The microphones may be multiple and placed at different locations on the computing device 800 for stereo capture or noise reduction purposes. The microphone may also be an array microphone or an omni-directional pick-up microphone. The speaker is used to convert electrical signals from the processor 801 or the radio frequency circuit 804 into sound waves. The loudspeaker can be a traditional film loudspeaker or a piezoelectric ceramic loudspeaker. When the speaker is a piezoelectric ceramic speaker, the speaker can be used for purposes such as converting an electric signal into a sound wave audible to a human being, or converting an electric signal into a sound wave inaudible to a human being to measure a distance. In some embodiments, the audio circuitry 807 may also include a headphone jack.

The positioning component 808 is operable to locate a current geographic location of the computing device 800 to implement navigation or LBS (location based Service). The positioning component 808 may be a positioning component based on the GPS of the united states, the beidou system of china, or the galileo system of the european union.

The power supply 809 is used to power the various components in the computing device 800. The power supply 809 can be ac, dc, disposable or rechargeable. When the power supply 809 includes a rechargeable battery, the rechargeable battery may be a wired rechargeable battery or a wireless rechargeable battery. The wired rechargeable battery is a battery charged through a wired line, and the wireless rechargeable battery is a battery charged through a wireless coil. The rechargeable battery may also be used to support fast charge technology.

In some embodiments, computing device 800 also includes one or more sensors 810. The one or more sensors 810 include, but are not limited to: acceleration sensor 811, gyro sensor 812, pressure sensor 813, fingerprint sensor 814, optical sensor 815 and proximity sensor 816.

The acceleration sensor 811 may detect acceleration magnitudes on three coordinate axes of a coordinate system established with the computing device 800. For example, the acceleration sensor 811 may be used to detect the components of the gravitational acceleration in three coordinate axes. The processor 801 may control the touch screen 805 to display the user interface in a landscape view or a portrait view according to the gravitational acceleration signal collected by the acceleration sensor 811. The acceleration sensor 811 may also be used for acquisition of motion data of a game or a user.

The gyro sensor 812 may detect a body direction and a rotation angle of the computing device 800, and the gyro sensor 812 may cooperate with the acceleration sensor 811 to acquire a 3D motion of the user with respect to the computing device 800. From the data collected by the gyro sensor 812, the processor 801 may implement the following functions: motion sensing (such as changing the UI according to a user's tilting operation), image stabilization at the time of photographing, game control, and inertial navigation.

Pressure sensors 813 may be disposed on the side bezel of computing device 800 and/or underneath touch display 805. When the pressure sensor 813 is disposed on the side frame of the computing device 800, the holding signal of the user to the computing device 800 can be detected, and the processor 801 performs left-right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 813. When the pressure sensor 813 is disposed at a lower layer of the touch display screen 805, the processor 801 controls the operability control on the UI interface according to the pressure operation of the user on the touch display screen 805. The operability control comprises at least one of a button control, a scroll bar control, an icon control and a menu control.

The fingerprint sensor 814 is used for collecting a fingerprint of the user, and the processor 801 identifies the identity of the user according to the fingerprint collected by the fingerprint sensor 814, or the fingerprint sensor 814 identifies the identity of the user according to the collected fingerprint. Upon identifying that the user's identity is a trusted identity, the processor 801 authorizes the user to perform relevant sensitive operations including unlocking a screen, viewing encrypted information, downloading software, paying for and changing settings, etc. Fingerprint sensor 814 may be disposed on the front, back, or side of computing device 800. When a physical key or vendor Logo is provided on computing device 800, fingerprint sensor 814 may be integrated with the physical key or vendor Logo.

The optical sensor 815 is used to collect the ambient light intensity. In one embodiment, the processor 801 may control the display brightness of the touch screen 805 based on the ambient light intensity collected by the optical sensor 815. Specifically, when the ambient light intensity is high, the display brightness of the touch display screen 805 is increased; when the ambient light intensity is low, the display brightness of the touch display 805 is turned down. In another embodiment, the processor 801 may also dynamically adjust the shooting parameters of the camera assembly 806 based on the ambient light intensity collected by the optical sensor 815.

A proximity sensor 816, also known as a distance sensor, is typically disposed on the front panel of the computing device 800. The proximity sensor 816 is used to capture the distance between the user and the front of the computing device 800. In one embodiment, the processor 801 controls the touch display 805 to switch from a bright screen state to a dark screen state when the proximity sensor 816 detects that the distance between the user and the front face of the computing device 800 is gradually decreased; when the proximity sensor 816 detects that the distance between the user and the front of the computing device 800 is gradually increasing, the touch display 805 is controlled by the processor 801 to switch from a breath-screen state to a light-screen state.

Those skilled in the art will appreciate that the configuration illustrated in FIG. 8 does not constitute a limitation of computing device 800, and may include more or fewer components than those illustrated, or may combine certain components, or may employ a different arrangement of components.

The embodiment of the present application further provides a computer-readable storage medium, which is applied to a terminal, and the computer-readable storage medium stores at least one instruction, at least one program, a code set, or a set of instructions, where the instruction, the program, the code set, or the set of instructions are loaded and executed by a processor to implement the operations performed by the computing device in the driving behavior evaluation method according to the foregoing embodiment.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only for facilitating the understanding of the technical solutions of the present application by those skilled in the art, and is not intended to limit the present application. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A driving behavior evaluation method, characterized in that the method comprises:

2. The method of claim 1, wherein determining the first weight and the first composite score for each designated driving behavior based on the maximum frequency and the first frequency of occurrence of each designated driving behavior comprises:

3. The method of claim 2, wherein determining a first score for a plurality of evaluation periods of the each designated driving behavior within the current evaluation time range based on the maximum frequency and the first frequency of occurrence of the each designated driving behavior comprises:

4. The method of claim 2, wherein determining the first weight for each designated driving behavior based on the first score and the second quantity for each evaluation period for the each designated driving behavior comprises:

5. The method of claim 4, wherein determining the first entropy value for each of the designated driving behaviors based on the first score and the second quantity for the each of the designated driving behaviors over the each evaluation period comprises:

the formula I is as follows:

6. The method of claim 4, wherein determining the first weight for each of the specified driving behaviors based on the first entropy value and the first quantity for each of the specified driving behaviors comprises:

the formula II is as follows: w ═ W'_i)_1×t

Wherein W is the weight matrix,

ω′_ia first weight, H, for the ith designated driving behavior_iA first entropy value of the driving behavior is specified for the ith, and

t is the first number;

7. The method of claim 1, wherein determining the second weight for each of the first number of specified driving behaviors based on the degree of importance of the first specified driving behavior relative to the second specified driving behavior comprises:

8. The method of claim 7, wherein determining the second weight for each of the first number of designated driving behaviors based on the matrix order of the first determination matrix, the maximum eigenvalue, and the stochastic consistency index comprises:

9. The method of claim 1, wherein determining a second composite score for the driving behavior of the target driver based on the first weight, the first composite score, and the second weight for each designated driving behavior comprises:

10. The method of claim 9, wherein determining the composite weight for each designated driving behavior based on the first weight and the second weight for each designated driving behavior comprises:

11. The method of claim 9, wherein determining the second composite score based on the third score and the third weight for each specified driving feature comprises:

12. The method of claim 11, wherein said fusing the third composite score with a plurality of fourth composite scores for driving behavior of the target driver over a second historical time to obtain the second composite score comprises:

13. The method of claim 12, further comprising:

14. The method of claim 12, wherein determining, by the target gaussian mixture model, a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores comprises:

the formula III is as follows: m ═ P (y)_v|θ_k)*2*y_const

Wherein the content of the first and second substances,

15. A driving behavior evaluation device characterized by comprising:

16. The apparatus of claim 15, wherein the first determining module is further configured to determine a first score for a plurality of evaluation periods of the each designated driving behavior within the current evaluation time range based on a maximum frequency and a first frequency of occurrence of the each designated driving behavior; determining a first weight of each designated driving behavior based on a first score and a second number of each designated driving behavior in each evaluation period, wherein the second number is the number of the plurality of evaluation periods; determining a first composite score for each designated driving behavior based on the first score for each designated driving behavior for each evaluation period.

17. The apparatus of claim 16, wherein the first determining module is further configured to determine, for each evaluation period for each specified driving behavior, a ratio between a first frequency and a maximum frequency of occurrence of the specified driving behavior; determining a second score for the designated driving behavior to be deducted within the evaluation period according to the ratio; and determining a first score of the specified driving behavior in the evaluation period according to the total score and the second score corresponding to the evaluation period.

18. The apparatus of claim 16, wherein the first determining module is further configured to determine a first entropy value for each of the designated driving behaviors based on the first score and the second quantity for the each of the designated driving behaviors over the each evaluation period; determining a first weight for the each specified driving behavior based on the first entropy value and the first quantity for the each specified driving behavior.

19. The apparatus of claim 18, wherein the first determining module is further configured to perform normalization processing based on the first score of each driving behavior specified in each evaluation period to obtain a first score of each driving behavior specified in each evaluation period;

the formula I is as follows:

20. The apparatus of claim 18, wherein the first determining module is further configured to determine a weight matrix based on the first entropy value and the first quantity of each of the designated driving behaviors, the weight matrix comprising a first weight of each of the designated driving behaviors, by the following formula two;

the formula II is as follows: w ═ W'_i)_1×t

Wherein W is the weight matrix,

t is the first number;

21. The apparatus of claim 15, wherein the second determining module is further configured to construct a first decision matrix based on the degree of importance of the first designated driving behavior relative to the second designated driving behavior; determining a matrix order and a maximum eigenvalue of the first judgment matrix based on the first judgment matrix; determining a random consistency index of the first judgment matrix from a corresponding relation between the matrix order and the random consistency index based on the matrix order; and determining a second weight of each designated driving behavior in the first number of designated driving behaviors based on the matrix order of the first judgment matrix, the maximum eigenvalue and the random consistency index.

22. The apparatus of claim 21, wherein the second determining module is further configured to determine a consistency ratio of the first decision matrix based on a matrix order of the first decision matrix, the maximum eigenvalue, and the random consistency index; when the consistency ratio is smaller than a ratio threshold, determining the first number of first feature vectors corresponding to the maximum feature value; normalizing the first quantity of first feature vectors to obtain a first quantity of second feature vectors; determining the value of the first number of second eigenvectors as a second weight of the first number of specified driving behaviors.

23. The apparatus of claim 15, wherein the third determining module is further configured to determine a composite weight for each of the designated driving behaviors based on the first weight and the second weight for each of the designated driving behaviors; determining a third score of each designated driving feature in a third number of designated driving features based on the comprehensive weight and the first comprehensive score of each designated driving behavior, wherein the third number is the number of the designated driving features, and each designated driving feature corresponds to a plurality of designated driving behaviors; determining a third weight for each of the specified driving characteristics based on a degree of importance of the first specified driving characteristic relative to a second specified driving characteristic, the first and second specified driving characteristics being any two different specified driving characteristics of the third number of specified driving characteristics; determining the second composite score based on the third score and the third weight for each specified driving feature.

24. The apparatus of claim 23, wherein the second determining module is further configured to determine, for each designated driving behavior, a product of a first weight of the designated driving behavior and the second weight to obtain a first value; summing the first numerical values of each designated driving behavior to obtain a second numerical value; and determining the ratio of the first numerical value to the second numerical value to obtain the comprehensive weight of the specified driving behavior.

25. The apparatus of claim 23, wherein the second determining module is further configured to perform a weighted summation of the third score and the third weight of each designated driving feature to obtain the second composite score; or carrying out weighted summation on the third score and the third weight of each designated driving feature to obtain a third comprehensive score, and fusing the third comprehensive score with a plurality of fourth comprehensive scores of the driving behavior of the target driver in a second historical time to obtain a second comprehensive score.

26. The apparatus of claim 25, wherein the second determining module is further configured to determine a first probability corresponding to the third composite score and a plurality of second probabilities corresponding to the plurality of fourth composite scores through a target gaussian mixture model; normalizing the first probability and the second probabilities to obtain a fourth weight corresponding to the third composite score and a plurality of fifth weights corresponding to the fourth composite scores, wherein one fourth composite score corresponds to one fifth weight; and weighting and summing the third composite score, the fourth weights, the plurality of fourth composite scores and the plurality of fifth weights to obtain the second composite score.

27. The apparatus of claim 26, further comprising:

28. The apparatus of claim 26, wherein the second determining module is further configured to determine the first probability and the plurality of second probabilities by using the target gaussian mixture model and the following formula three;

the formula III is as follows: m ═ P (y)_v|θ_k)*2*y_const

Wherein the content of the first and second substances,

29. A computing device, wherein the computing device comprises:

a processor and a memory, the memory having stored therein at least one instruction, at least one program, set of codes, or set of instructions, the instruction, the program, the set of codes, or the set of instructions being loaded and executed by the processor to carry out the operations performed in the driving behavior assessment method of any of claims 1-14.

30. A computer-readable storage medium having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by a processor to carry out the operations performed in the driving behavior assessment method according to any one of claims 1-14.