CN106651210B - Driver comprehensive quality evaluation method based on CAN data - Google Patents
Driver comprehensive quality evaluation method based on CAN data Download PDFInfo
- Publication number
- CN106651210B CN106651210B CN201611269823.XA CN201611269823A CN106651210B CN 106651210 B CN106651210 B CN 106651210B CN 201611269823 A CN201611269823 A CN 201611269823A CN 106651210 B CN106651210 B CN 106651210B
- Authority
- CN
- China
- Prior art keywords
- value
- steering wheel
- speed
- acceleration
- driver
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
Abstract
The invention relates to a driver comprehensive quality evaluation method based on CAN data, which comprises the following steps: one, obtainCalculating instantaneous oil consumption, acceleration and the ratio of the engine rotating speed to the engine speed by taking the longitude latitude, the engine rotating speed, the vehicle speed, the steering wheel rotating angle, the oil consumption, the accelerator angle and the brake pedal state; secondly, determining evaluation factors: steering wheel angle entropy value H(θ)Steering wheel angular velocity VθMotor vehicle speed entropy value H(v)Absolute value of acceleration | a |, intensity of acceleration VaThe method comprises the steps of obtaining a positive acceleration a, a ratio phi of an engine rotating speed to a motor vehicle speed, establishing a weight α occupied by a single factor to driver comprehensive quality evaluation by adopting a fuzzy analytic hierarchy process, establishing a required weight vector Q, establishing membership functions as few as possible, obtaining a single-factor fuzzy judgment matrix R, obtaining a judgment vector Q R, analyzing and solving to obtain a judgment set for evaluating the driver comprehensive quality.
Description
Technical Field
The invention relates to the field of intelligent transportation, in particular to a driver comprehensive quality evaluation method based on CAN data.
Background
Along with the progress of the economy, the automobile is gradually popularized to the home of common people. However, traffic accidents are also rising continuously, and the main reason is that the contradiction among people, vehicles and roads is increasingly excited. According to the statistics of the european and american countries, about 85% of traffic accidents are related to drivers among causes of traffic accidents. In the domestic investigation statistics of traffic accidents, 80% -90% of accidents are caused by human factors. Therefore, the method has important significance for researching and evaluating the behaviors of the motor vehicle drivers, pertinently dissuading and educating the drivers and ensuring the traffic safety.
Traditional traffic detection system mainly adopts the suspension type sensor of technologies such as radar, ultrasonic wave, infrared ray, microwave, audio frequency and video image, and the information that this kind of detecting system gathered is accurate, but too loaded down with trivial details, and equipment cost is too high. There are also research institutions carrying driving behavior research foundation laboratories, namely driving simulators, which can collect diversified data, but the data authenticity is not good enough. With the progress of science and technology, a method for detecting vehicles by utilizing vehicle-mounted data acquisition equipment is expected to replace the traditional method.
The information of various types needed by the behavior of the driver comprises psychological and physiological information of the driver, the state of the motor vehicle, the control of the driver, the traffic environment and the like. Different evaluation indexes can be proposed according to different information, such as: the fuel consumption of the automobile is reflected through the acceleration, and the fatigue state of a driver is reflected through the steering wheel corner entropy and the eye movement. In order to fully utilize the collected data, more evaluation indexes need to be mined to provide a more sufficient basis for analyzing the behavior of the driver, however, based on the data indexes, the comprehensive quality of the driver cannot be quantitatively evaluated. In order to solve the problem, a plurality of scholars experts at home and abroad successfully apply the fuzzy logic theory to a model for analyzing the behavior of the driver, and the analysis result has high fitting degree with the actual situation.
Disclosure of Invention
The invention aims to provide a driver comprehensive quality evaluation method based on CAN data, which CAN quantitatively evaluate the driver comprehensive quality based on CAN data analysis.
The technical scheme of the invention is as follows:
a driver comprehensive quality evaluation method based on CAN data comprises the following steps:
the method comprises the steps of collecting driving data of a motor vehicle through a CAN bus, obtaining longitude and latitude, engine rotating speed, vehicle speed, steering wheel turning angle, oil consumption, accelerator angle and brake pedal state, and calculating instantaneous oil consumption, acceleration and the ratio of the engine rotating speed to the engine speed.
Determining evaluation factors according to three indexes of driving safety, riding comfort and oil consumption of the motor vehicle: steering wheel angle entropy value H(θ)Steering wheel angular velocity VθThe motor-drivenEntropy of vehicle speed H(v)Absolute value of acceleration | a |, intensity of acceleration VaPositive acceleration a, ratio phi of engine speed to motor vehicle speed.
Thirdly, weights α occupied by the single factors for comprehensive quality evaluation of the driver are established by adopting a fuzzy analytic hierarchy process, and required weight vectors Q are established as (α 1, α 2, α 3, α 4, α 5, α 6, α 7).
And fourthly, considering the difference value of the extreme difference and the mean value of the single factor value, classifying different factors by combining the discrete degree of value distribution after data normalization, establishing membership function as few as possible, and substituting the single factor value into the membership function to obtain a single factor fuzzy judgment matrix R:
and fifthly, multiplying the weight vector Q by the single-factor fuzzy judgment matrix R to obtain a judgment vector, namely Q R, and analyzing and solving by combining a maximum membership principle to obtain a judgment set for evaluating the comprehensive quality of the driver.
Wherein the steering wheel angle entropy value H(θ)Representing the fatigue or drunk driving state of a driver; the steering wheel turning angle rate VθRepresenting the speed of a driver rotating a steering wheel; the motor vehicle speed entropy value H(v)Representing the degree of disorder of the speed of the vehicle in the running process; the absolute value of the acceleration | a | represents the speed change speed of the vehicle in the running process; the acceleration intensity VaCharacterizing a rate of change of vehicle acceleration; the positive acceleration a represents the speed of the change of the accelerated running speed of the vehicle; the ratio phi of the engine speed to the motor vehicle speed represents the traction of the motor vehicle.
Further, define e(n)The actual value theta of the steering wheel angle at the moment n(n)Predicted value theta with time np(n)Absolute value of difference: e.g. of the type(n)=|θ(n)-θp(n)I, calculating steering wheel corner entropy value H by redefining nine intervals(θ)The calculation process is as follows:
step i, according to three time points of n-1, n-2 and n-3Predicting the value of the n-time rotation angle to the value of the disc rotation angle, wherein the time interval is 100ms, and calculating thetap(n):
Step ii, calculating the absolute value e of the deviation value of the steering wheel angle value and the predicted value at the moment n(n)Calculate e(n):
e(n)=|θ(n)-θp(n)| ②。
Step iii, determining α ≦ 0.3 such that P {0 ≦ e(n)< α } -, 90%, divide the deviation value e(n)The following nine corresponding intervals:
step iv, calculating a deviation value e(n)Probability P of falling in each interval distributioniFinally, calculating the steering wheel angle entropy according to formula ③(θ):
Further, define v(n)For the actual value V of vehicle speed at time n(n)Predicted value V corresponding to time np(n)Absolute difference
The value: v is(n)=|V(n)-Vp(n)Defining nine intervals to calculate speed entropy value H of square vehicle(v)The calculation process is as follows:
step i, predicting the vehicle speed value at the time n according to the vehicle speed values at the time n-1, n-2 and n-3, wherein the time interval is 100ms, and calculating Vp(n):
Step ii, calculating the absolute value e of the deviation value between the vehicle speed value and the predicted value at the moment n(n)Calculate e(n):
v(n)=|V(n)-Vp(n)| ⑤。
Step iii, determining α to 0.3 such that P {0 ≦ v(n)< α } -, 90%, dividing the deviation value v(n)The following nine corresponding intervals:
step iv, calculating a deviation value v(n)Probability P of falling in each interval distributioniFinally, calculating the steering wheel angle entropy according to formula ③(v):
Further, according to the empirical value, a judgment matrix A is constructed:
wherein, a1Indicating the amount of oil consumption, a2Denotes security, a3Indicating comfort, factor a(ij)Representing the evaluation of the driver's overall quality, factor aiCompared with ajThe more important the value is, the more important the characteristic vector of the matrix A is obtained and normalized to obtain the weight vector Q1If the consistency test is passed, Q is1Weight vectors of three indexes are (0.091,0.691, 0.217); otherwise, reconstructing the judgment matrix A.
The following are obtained by an analytic hierarchy process:
for oil consumption a1The one-factor weight vector is: qa1=(0.667,0.333,0,0,0,0,0)。
For security a2The one-factor weight vector is Qa2=(0,0,0.464,0.095,0.130,0.182,0.130)。
For comfort a3The one-factor weight vector is Qa3=(0,0,0.073,0.072,0.318,0.318,0.218)。
And (3) calculating the weight of the single factor to the comprehensive quality of the driver:
further, the evaluation factor is obtained by eliminating abnormal data according to a 3 principle, or each factor value X is obtained by performing the following normalization processing on the collected driving data:
wherein x ismaxIs the maximum value of the sample, xminFor the minimum value of the sample to be,is the sample mean.
Further, two different membership functions are established according to the discrete degree of the distribution of the factor values X after normalization processing, and the entropy value H of the steering wheel corner is(θ)Motor vehicle speed entropy value H(v)And the ratio φ of engine speed to vehicle speed, each establish a membership function ⑧ as follows:
for steering wheel angular velocity VθAbsolute value of acceleration | a |, intensity of acceleration VaAnd positive acceleration a are each established as the following membership function ⑨:
wherein: r is1Indicates that the degree of membership is "good"; r is2Representing that the degree of membership is "good"; r is3Representing that the degree of membership is "general"; r is4Representing that the degree of membership belongs to "difference"; r is5Indicating that the degree of membership is "poor".
The invention evaluates the comprehensive quality of the driver and considers three indexes of riding safety, riding comfort and motor vehicle oil consumption. For riding safety and riding comfort, the invention mainly uses the steering wheel corner entropy value H(θ)Steering wheel angular velocity VθMotor vehicle speed entropy value H(v)Absolute value of acceleration | a |, intensity of acceleration (rate of change of acceleration) VaFive factors are analyzed, and the weight occupied by each evaluation factor is different according to different indexes. For the fuel consumption of the motor vehicle, two factors of positive acceleration a, the ratio phi of the engine rotating speed to the motor vehicle speed and new factors which are not considered by other models are mainly considered, so that the comprehensive quality of a driver is quantitatively evaluated, and the evaluation accuracy is improved. And improves the steering wheel angle entropy value proposed by a japanese scholars Nakayama, so that the whole evaluation method is simpler and more effective.
Drawings
FIG. 1 is a schematic diagram of the present invention;
FIG. 2 is a GPS trajectory in CAN data of the present invention;
FIG. 3 is a graph showing the recorded change of the steering angle and the vehicle speed of the motor vehicle according to the driving path of FIG. 1;
FIG. 4 is a line graph showing the instantaneous fuel consumption according to the acceleration in the present invention;
FIG. 5 is a line graph showing the instantaneous fuel consumption according to the ratio of the engine speed to the vehicle speed in the present invention;
FIG. 6 is a flow chart of calculating an entropy of a steering wheel angle in accordance with the present invention;
FIG. 7 is a flow chart of the driver comprehensive quality evaluation model building and solving in the invention.
Detailed Description
The invention is further described below with reference to the figures and examples.
The invention is characterized in that new evaluation index factors and evaluation models of the comprehensive quality of the drivers of some motor vehicles are provided based on CAN data, and data acquisition is necessary. In order to solve the problem that other sensor devices are not added, more types of data are obtained as much as possible, and further more evaluation indexes are provided to perfect a fuzzy evaluation model, the invention takes a CAN bus protocol as a basis, uses a data acquisition module Openxc-vi, carries an automobile test platform under Windows 10 environment on Microsoft Surface 4 computer, and acquires CAN data, wherein the CAN data acquisition comprises the following steps: steering wheel angle, longitude and latitude, engine speed, vehicle speed and other 20 items of data.
The invention builds up the test platform at first, has debugged the operational environment, please a driver to drive and carry on the data acquisition, this system can gather more than 20 data, withdraw the required longitude and latitude of this invention from it, engine speed, speed of a motor vehicle, steering wheel angle, oil consumption, throttle angle, brake pedal state multinomial data, calculate instantaneous oil consumption, acceleration, and engine speed and speed ratio. Before the collected data is used, partial data needs to be preprocessed, abnormal data is eliminated by adopting a 3 principle, in order to prove the effectiveness and the practicability of the data, partial data is taken and drawn as shown in figure 3, and obviously, the running track of the automobile on the Google map shown in figure 2 is perfectly matched with the steering wheel corner and the speed of the automobile shown in figure 3.
In the research of vehicle energy conservation and emission reduction, except for objective conditions such as road environment and vehicle performance, the behavior of a driver is undoubtedly a main factor influencing oil consumption. The invention extracts the data of oil consumption, speed, engine speed and the like collected by the detection system to analyze the oil consumption condition of the vehicle, processes the data and can obtain the instantaneous oil consumption, the acceleration and the ratio of the engine speed to the speed. The invention obtains the relation between the acceleration a and the oil consumption by analyzing the data of thousands of sampling points, and draws a line graph. As shown in FIG. 4, when the vehicle is in a decelerating state (a ≦ 0 m/s)2) When the vehicle is in an accelerating state (a is more than 0 and less than 1.3 m/s)2) When the vehicle acceleration reaches a certain value (a is more than or equal to 1.3 m/s), the oil consumption is in an ascending trend and gradually increases rapidly2) In time, the oil consumption gradually stabilizes. The relationship between the ratio phi of the rotation speed and the oil consumption is shown in FIG. 5, when the ratio is gradually increased (0 < phi < 160), namely the traction force of the vehicle is gradually increased, the oil consumption is in a obviously rising trend, and when the ratio reaches a certain degree (phi > 160), the oil consumption is in a stable state。
The invention determines seven evaluation factors according to three indexes of driving safety, riding comfort and oil consumption of the motor vehicle: steering wheel angle entropy value H(θ)Steering wheel angular velocity VθMotor vehicle speed entropy value H(v)Absolute value of acceleration | a |, intensity of acceleration VaPositive acceleration a, ratio phi of engine speed to motor vehicle speed. Wherein the steering wheel angle entropy value H(θ)Representing the fatigue or drunk driving state of a driver; the steering wheel turning angle rate VθRepresenting the speed of a driver rotating a steering wheel; the motor vehicle speed entropy value H(v)Representing the degree of disorder of the speed of the vehicle in the running process; the absolute value of the acceleration | a | represents the speed change speed of the vehicle in the running process; the acceleration intensity VaCharacterizing a rate of change of vehicle acceleration; the positive acceleration a represents the speed of the change of the accelerated running speed of the vehicle; the ratio phi of the engine speed to the motor vehicle speed represents the traction of the motor vehicle.
The invention adopts a fuzzy analytic hierarchy process to establish the weight α occupied by a single factor to the comprehensive quality evaluation of the driver, and establishes the required weight vector Q as (α 1, α 2, α 3, α 4, α 5, α 6, α 7).
The method considers the difference value of the extreme difference and the mean value of a single evaluation factor value, and classifies different evaluation factors by combining the discrete degree of the value distribution after data normalization. Considering that 7 evaluation factors are too many to establish one membership function, dividing the dispersion degree of the evaluation factor value distribution, establishing two membership functions to establish few membership functions as far as possible, substituting a single factor value into the membership function to obtain a single factor fuzzy judgment matrix R,
and multiplying the weight vector Q by the single-factor fuzzy judgment matrix R to obtain a judgment vector, namely Q R, analyzing and solving by combining a maximum membership principle to obtain a judgment set for evaluating the comprehensive quality of the driver, and realizing qualitative evaluation according to quantitative analysis, wherein the logical principle of the whole method is shown in figure 1.
Aiming at two indexes of driving safety and riding comfort, the entropy value H of the improved steering wheel corner is obtained from the 5 related main evaluation factors(θ)Entropy of vehicle speed H(v)The specific implementation process is shown in fig. 6:
step i, predicting the value of the steering wheel angle at n time according to the values of the steering wheel angles at n-1, n-2 and n-3, wherein the interval of each time is 100ms, and calculating thetap(n):
Step ii, calculating the absolute value e of the deviation value of the steering wheel angle value and the predicted value at the moment n(n)Calculate e(n):
e(n)=|θ(n)-θp(n)| ②。
Step iii, the distribution of the sample point deviation values is basically obeyed to the right half of the normal distribution, and one α is needed to be found so that P {0 ≦ e(n)(α) } 90%, and through analysis and calculation, the deviation value e is divided into α ═ 0.3(n)The following nine corresponding intervals:
step iv, calculating a deviation value e(n)Probability P of falling in each interval distributioniFinally, calculating the steering wheel angle entropy according to formula ③(θ):
Similarly, entropy H for vehicle speed(v)The solving process is the same as above, and v is defined(n)For the actual value V of vehicle speed at time n(n)And time nPredicted value V ofp(n)Absolute value of difference: v is(n)=|V(n)-Vp(n)Defining nine intervals to calculate speed entropy value H of square vehicle(v)The calculation process is as follows:
step i, predicting the vehicle speed value at the time n according to the vehicle speed values at the time n-1, n-2 and n-3, wherein the time interval is 100ms, and calculating Vp(n):
Step ii, calculating the absolute value e of the deviation value between the vehicle speed value and the predicted value at the moment n(n)Calculate e(n):
v(n)=|V(n)-Vp(n)| ⑤。
Step iii, determining α to 0.3 such that P {0 ≦ v(n)< α } -, 90%, dividing the deviation value v(n)The following nine corresponding intervals:
step iv, calculating a deviation value v(n)Probability P of falling in each interval distributioniFinally, calculating the steering wheel angle entropy according to formula ③(v):
The influence between internal factors of each layer is not very large, and if the complexity is multiplied by adopting a network analysis rule, the method adopts a fuzzy analytic hierarchy process (F-AHP) to establish a model for judging the comprehensive quality of the driver, and comprises the following specific steps of:
firstly, the comprehensive quality of a driver is evaluated, the weights occupied by different indexes are different, and the invention adopts an analytic hierarchy process to calculate the weights of 3 indexes of the driving safety, the riding comfort and the oil consumption of the motor vehicle. First, the 1-9 Scale method of Santy is used to knotConstructing a judgment matrix A (factor a) according to suggestions given by experts, experienced drivers, and the like(ij)Representing factor aiCompared with ajImportance of) as follows:
wherein, a1Indicating the amount of oil consumption, a2Denotes security, a3Indicating comfort, factor a(ij)Representing the evaluation of the driver's overall quality, factor aiCompared with ajThe more important the value is, the more important the characteristic vector of the matrix A is obtained and normalized to obtain the weight vector Q1If the consistency test is passed, Q is1Weight vectors of three indexes are (0.091,0.691, 0.217); otherwise, reconstructing the judgment matrix A.
Then, the weights of the evaluation factors to different evaluation indexes are calculated, the calculation process is similar to the previous step, and the oil consumption a is calculated by an analytic hierarchy process1The one-factor weight vector is:
Qa1=(0.667,0.333,0,0,0,0,0),
for security a2The one-factor weight vector is Qa2=(0,0,0.464,0.095,0.130,0.182,0.130),
For comfort a3The one-factor weight vector is Qa3=(0,0,0.073,0.072,0.318,0.318,0.218)。
And finally, solving the weight of the single factor to the comprehensive quality of the driver:
the invention mainly has two value processing modes for the evaluation factor value, one mode is to eliminate abnormal data by utilizing a 3 principle for the collected data, and the other mode is to obtain each factor value X after carrying out the following normalization processing on the data in order to avoid the situation that a large number covers a decimal number:
wherein x ismaxIs the maximum value of the sample, xminFor the minimum value of the sample to be,is the sample mean.
In order to make the evaluation grade more definite, different membership function is established according to the discrete degree of X distribution after normalization processing, and the membership function is divided into two sine functions reasonably through analysis. Establishing two different membership functions according to the discrete degree of the distribution of the factor values X after normalization processing, and establishing a steering wheel corner entropy value H for a steering wheel(θ)Motor vehicle speed entropy value H(v)And the ratio φ of engine speed to vehicle speed, each establish a membership function ⑧ as follows:
for steering wheel angular velocity VθAbsolute value of acceleration | a |, intensity of acceleration VaAnd positive acceleration a are each established as the following membership function ⑨:
wherein: r is1Indicates that the degree of membership is "good"; r is2Representing that the degree of membership is "good"; r is3Representing that the degree of membership is "general"; r is4Representing that the degree of membership belongs to "difference"; r is5Indicating that the degree of membership is "poor".
The invention adopts the data analysis and calculation of the driver A, substitutes each factor value into the corresponding membership function, such as the ratio phi in the data sample,xmax=24.491,xmin16.477, substituting formula ⑦ to obtain X0.677, and substituting X into formula ⑧ to obtain row vector phi (0,0,0.6767,0.3233,0) in the one-factor fuzzy matrix(θ)、H(v)、Vθ、VaAnd | a | form a single-factor fuzzy judgment matrix R as follows:
and multiplying the single-factor weight vector Q by R to obtain a final membership vector of the comprehensive quality grade evaluation:
=Q*R=[0.1634,0.5523,0.2607,0.0197,0]
according to the maximum membership rule, 0.5523 is the largest in the set of judgments, and the rank thereof is "good", so that the comprehensive quality of the driver a is rated as "good".
In order to prove the practical accuracy of the model, a famous and elegant model driver B is invited to drive to walk through a road section similar to the road section A, and the model is used for solving a grade rating membership vector of the driver B:
=Q*R=[0.4300,0.3435,0.1607,0.0610,0]
according to the maximum membership rule, 0.4300 is the largest in the evaluation set, and the grade is 'good', so that the comprehensive complaints of the B driver are rated as 'good'.
In the process of solving the model, the quality grade of each factor of the driver can be obtained from a single evaluation factor fuzzy judgment matrix, and if simple processing is carried out, the membership degree vectors of evaluation grades of three indexes, namely driving safety, riding comfort and motor vehicle oil consumption, can be respectively obtained, so that the quality grade evaluation is carried out on each index of the driver.
Claims (6)
1. A driver comprehensive quality evaluation method based on CAN data is characterized by comprising the following steps:
acquiring driving data of a motor vehicle through a CAN bus, acquiring longitude and latitude, engine rotation speed, vehicle speed, steering wheel rotation angle, oil consumption, accelerator angle and brake pedal state, and calculating instantaneous oil consumption, acceleration and the ratio of the engine rotation speed to the engine speed;
determining evaluation factors according to three indexes of driving safety, riding comfort and oil consumption of the motor vehicle: steering wheel angle entropy value H(θ)Steering wheel angular velocity VθMotor vehicle speed entropy value H(v)Absolute value of acceleration | a |, intensity of acceleration VaPositive acceleration a, the ratio phi of the engine speed to the motor vehicle speed;
thirdly, establishing α weight of a single factor to comprehensive driver quality evaluation by adopting a fuzzy analytic hierarchy process, and establishing a required weight vector Q (α 1, α 2, α 3, α 4, α 5, α 6 and α 7);
fourthly, considering the difference value of the extreme difference and the mean value of the single factor value, classifying different factors by combining the discrete degree of the value distribution after data normalization, establishing membership function as few as possible, substituting the single factor value into the membership function to obtain a single factor fuzzy judgment matrix R,
multiplying the weight vector Q by the single-factor fuzzy judgment matrix R to obtain a judgment vector, namely Q R, and analyzing and solving by combining a maximum membership principle to obtain a judgment set for evaluating the comprehensive quality of the driver;
wherein the steering wheel angle entropy value H(θ)Representing the fatigue or drunk driving state of a driver;
the steering wheel turning angle rate VθRepresenting the speed of a driver rotating a steering wheel;
the motor vehicle speed entropy value H(v)Representing the degree of disorder of the speed of the vehicle in the running process;
the absolute value of the acceleration | a | represents the speed change speed of the vehicle in the running process;
the acceleration intensity VaCharacterizing a rate of change of vehicle acceleration;
the positive acceleration a represents the speed of the change of the accelerated running speed of the vehicle;
the ratio phi of the engine speed to the motor vehicle speed represents the traction of the motor vehicle.
2. The CAN data-based driver comprehensive quality evaluation method according to claim 1, characterized in that: definition e(n)The actual value theta of the steering wheel angle at the moment n(n)Predicted value at time nAbsolute value of difference: e.g. of the type(n)=|θ(n)-θp(n)I, calculating steering wheel corner entropy value H by redefining nine intervals(θ)The calculation process is as follows:
step i, predicting the value of the steering wheel angle at n time according to the values of the steering wheel angles at n-1, n-2 and n-3, wherein the interval of each time is 100ms, and calculating
Step ii, calculating the absolute value e of the deviation value of the steering wheel angle value and the predicted value at the moment n(n)Calculate e(n):
e(n)=|θ(n)-θp(n)| ②;
Step iii, determining α ≦ 0.3 such that P {0 ≦ e(n)< α } -, 90%, divide the deviation value e(n)The following nine corresponding intervals:
step iv, calculating a deviation value e(n)Probability P of falling in each interval distributioniFinally according to formula ③Calculating steering wheel angle entropy(θ):
3. The CAN data-based driver comprehensive quality evaluation method according to claim 1 or 2, characterized in that: definition of v(n)For the actual value V of vehicle speed at time n(n)Predicted value at time nAbsolute value of difference: v is(n)=|V(n)-Vp(n)Defining nine intervals to calculate speed entropy value H of square vehicle(v)The calculation process is as follows:
step i, predicting the vehicle speed value at the time n according to the vehicle speed values at the time n-1, n-2 and n-3, wherein the time interval is 100ms, and calculating
Step ii, calculating the absolute value e of the deviation value between the vehicle speed value and the predicted value at the moment n(n)Calculate e(n):
v(n)=|V(n)-Vp(n)| ⑤;
Step iii, determining α to 0.3 such that P {0 ≦ v(n)< α } -, 90%, dividing the deviation value v(n)The following nine corresponding intervals:
step iv, calculating a deviation value v(n)Probability P of falling in each interval distributioniFinally according to formula ③Calculating steering wheel angle entropy value H(v):
4. The CAN data-based driver comprehensive quality evaluation method according to claim 3, characterized in that: constructing a judgment matrix A according to empirical values:
wherein, a1Indicating the amount of oil consumption, a2Denotes security, a3Indicating comfort, factor a(ij)Representing the evaluation of the driver's overall quality, factor aiCompared with ajThe more important the value is, the more important the characteristic vector of the matrix A is obtained and normalized to obtain the weight vector Q1If the consistency test is passed, Q is1Weight vectors of three indexes are (0.091,0.691, 0.217); otherwise, reconstructing a judgment matrix A;
the oil consumption a is obtained by an analytic hierarchy process1The one-factor weight vector is: qa1=(0.667,0.333,0,0,0,0,0);
For security a2The one-factor weight vector is Qa2=(0,0,0.464,0.095,0.130,0.182,0.130);
For comfort a3The one-factor weight vector is Qa3=(0,0,0.073,0.072,0.318,0.318,0.218);
And (3) calculating the weight of the single factor to the comprehensive quality of the driver:
5. the CAN data-based driver comprehensive quality assessment method according to any one of claims 1, 2, and 4, characterized in that: the evaluation factors are obtained by eliminating abnormal data according to a principle 3, or all the factor values X are obtained by carrying out normalization processing on the collected driving data as follows:
6. The CAN-data-based driver comprehensive quality evaluation method according to claim 5, wherein: establishing two different membership functions according to the discrete degree of the distribution of the factor values X after normalization processing, and establishing a steering wheel corner entropy value H for a steering wheel(θ)Motor vehicle speed entropy value H(v)And the ratio φ of engine speed to vehicle speed, each establish a membership function ⑧ as follows:
for steering wheel angular velocity VθAbsolute value of acceleration | a |, intensity of acceleration VaAnd positive acceleration a are each established as the following membership function ⑨:
wherein: r is1Indicates that the degree of membership is "good"; r is2Representing that the degree of membership is "good"; r is3Representing that the degree of membership is "general"; r is4Representing that the degree of membership belongs to "difference"; r is5Indicating that the degree of membership is "poor".
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611269823.XA CN106651210B (en) | 2016-12-30 | 2016-12-30 | Driver comprehensive quality evaluation method based on CAN data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611269823.XA CN106651210B (en) | 2016-12-30 | 2016-12-30 | Driver comprehensive quality evaluation method based on CAN data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106651210A CN106651210A (en) | 2017-05-10 |
CN106651210B true CN106651210B (en) | 2020-08-04 |
Family
ID=58838260
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611269823.XA Active CN106651210B (en) | 2016-12-30 | 2016-12-30 | Driver comprehensive quality evaluation method based on CAN data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106651210B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107566402B (en) * | 2017-10-13 | 2020-08-04 | 成都信息工程大学 | SOEKS-based intrusion detection method and implementation of vehicle-mounted electronic information system |
CN108614547B (en) * | 2018-06-14 | 2022-05-24 | 上海大学 | Industrial control protocol security assessment method based on reduction factor |
CN109118034B (en) * | 2018-06-14 | 2021-11-23 | 江苏大学 | Three-dimensional quantitative comprehensive evaluation method for driving behaviors |
CN108876165B (en) * | 2018-06-27 | 2021-08-27 | 上海天方夜谭网络科技有限公司 | Driver safety monitoring learning system |
CN111354456A (en) * | 2018-12-21 | 2020-06-30 | 广东美的白色家电技术创新中心有限公司 | Traditional Chinese medicine constitution type detection method and device |
CN110782110B (en) * | 2019-01-29 | 2023-07-18 | 长城汽车股份有限公司 | Method and device for evaluating driving condition of driver and machine-readable storage medium |
CN109858568B (en) * | 2019-03-06 | 2022-11-04 | 长安大学 | Method for classifying operation modes of driver in curve |
CN109934511A (en) * | 2019-03-26 | 2019-06-25 | 中国联合网络通信集团有限公司 | The evaluation method and system of bus comfort level |
CN110491126B (en) * | 2019-08-25 | 2021-11-16 | 安徽深蓝大健康智能科技有限公司 | Non-motor vehicle intersection driving control method and system based on Internet of things technology |
CN113642832A (en) * | 2021-06-29 | 2021-11-12 | 桂林电子科技大学 | Method and system for evaluating driving behavior of commercial vehicle |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20080005792A (en) * | 2006-07-10 | 2008-01-15 | (주)컨피테크 | System and its method for providing information of driver's driving pattern |
JP4371081B2 (en) * | 2005-06-14 | 2009-11-25 | トヨタ自動車株式会社 | Driving evaluation device |
CN104504531A (en) * | 2014-12-30 | 2015-04-08 | 大连楼兰科技股份有限公司 | Vehicle driving risk assessment method |
CN104657623A (en) * | 2015-03-17 | 2015-05-27 | 天津市市政工程设计研究院 | Road alignment safe checking method based on driving simulation technology |
CN104978492A (en) * | 2015-07-09 | 2015-10-14 | 彩虹无线(北京)新技术有限公司 | Safety driving evaluation method based on telematics data flow |
CN106022561A (en) * | 2016-05-05 | 2016-10-12 | 广州星唯信息科技有限公司 | Driving comprehensive evaluation method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015195275A2 (en) * | 2014-05-23 | 2015-12-23 | Michael Thurston | Method for optimizing asset value based on driver acceleration and braking behavior |
-
2016
- 2016-12-30 CN CN201611269823.XA patent/CN106651210B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4371081B2 (en) * | 2005-06-14 | 2009-11-25 | トヨタ自動車株式会社 | Driving evaluation device |
KR20080005792A (en) * | 2006-07-10 | 2008-01-15 | (주)컨피테크 | System and its method for providing information of driver's driving pattern |
CN104504531A (en) * | 2014-12-30 | 2015-04-08 | 大连楼兰科技股份有限公司 | Vehicle driving risk assessment method |
CN104657623A (en) * | 2015-03-17 | 2015-05-27 | 天津市市政工程设计研究院 | Road alignment safe checking method based on driving simulation technology |
CN104978492A (en) * | 2015-07-09 | 2015-10-14 | 彩虹无线(北京)新技术有限公司 | Safety driving evaluation method based on telematics data flow |
CN106022561A (en) * | 2016-05-05 | 2016-10-12 | 广州星唯信息科技有限公司 | Driving comprehensive evaluation method |
Also Published As
Publication number | Publication date |
---|---|
CN106651210A (en) | 2017-05-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106651210B (en) | Driver comprehensive quality evaluation method based on CAN data | |
EP3533681B1 (en) | Method for detecting safety of driving behavior, apparatus and storage medium | |
Chen et al. | A graphical modeling method for individual driving behavior and its application in driving safety analysis using GPS data | |
Araújo et al. | Driving coach: A smartphone application to evaluate driving efficient patterns | |
CN107492251A (en) | It is a kind of to be identified and driving condition supervision method based on the driver identity of machine learning and deep learning | |
CN111461185A (en) | Driving behavior analysis method based on improved K-means | |
CN111428960B (en) | Intelligent vehicle driving automatic evaluation method fusing multi-source vehicle-mounted sensor information | |
CN102717765A (en) | Fatigue driving detection method and anti-fatigue driving auxiliary device | |
CN109552338A (en) | A kind of pure electric automobile ecology driving behavior appraisal procedure and system | |
CN111563555A (en) | Driver driving behavior analysis method and system | |
CN102789690B (en) | Illegal vehicle identifying method and system | |
CN110304068A (en) | Acquisition method, device, equipment and the storage medium of running car environmental information | |
CN108647836A (en) | A kind of driver's Energy efficiency evaluation method and system | |
CN113487197A (en) | TOPSIS (technique for order preference by similarity to Ideal solution) evaluation model-based vehicle networking information screening method | |
CN114169444B (en) | Driving style classification method considering risk potential field distribution under vehicle following working condition | |
CN108961680B (en) | Performance detection system and method of drunk driving and fatigue driving discrimination system | |
CN114332825A (en) | Road terrain distribution identification method and device based on deep learning and storage medium | |
CN110610326A (en) | Driving management system based on driving data | |
CN115221234A (en) | Method and system for portraying user based on power assembly data | |
Yarlagadda et al. | Heterogeneity in the driver behavior: an exploratory study using real-time driving data | |
CN106651133B (en) | Method for achieving user driving behavior scoring based on ADAS by using mobile phone | |
Chen et al. | The impact of truck proportion on traffic safety using surrogate safety measures in China | |
CN110867075A (en) | Method for evaluating influence of road speed meter on reaction behavior of driver under rainy condition | |
Rosyid et al. | Comparison of deep learning models in pothole avoidance for self-driving car | |
CN111932829B (en) | Fatigue driving prevention facility utility testing method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |