CN112319479B - Vehicle longitudinal driving safety distance estimation method based on vehicle trust - Google Patents

Vehicle longitudinal driving safety distance estimation method based on vehicle trust Download PDF

Info

Publication number
CN112319479B
CN112319479B CN202011130600.1A CN202011130600A CN112319479B CN 112319479 B CN112319479 B CN 112319479B CN 202011130600 A CN202011130600 A CN 202011130600A CN 112319479 B CN112319479 B CN 112319479B
Authority
CN
China
Prior art keywords
vehicle
trust
safe
driver
safety distance
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
Application number
CN202011130600.1A
Other languages
Chinese (zh)
Other versions
CN112319479A (en
Inventor
孙文财
温雪
李世武
郭梦竹
于淼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN202011130600.1A priority Critical patent/CN112319479B/en
Publication of CN112319479A publication Critical patent/CN112319479A/en
Application granted granted Critical
Publication of CN112319479B publication Critical patent/CN112319479B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0019—Control system elements or transfer functions
    • B60W2050/0028—Mathematical models, e.g. for simulation
    • B60W2050/0031—Mathematical model of the vehicle
    • B60W2050/0034—Multiple-track, 2D vehicle model, e.g. four-wheel model

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

A vehicle longitudinal driving safety distance estimation method based on vehicle trust belongs to the technical field of automobile active safety, a traditional vehicle safety distance model is optimized by introducing a vehicle trust concept, and the vehicle trust directly reflects the control of a driver on a vehicle in the vehicle driving process and indirectly reflects the behaviors of keeping the safe distance, whether safe overtaking can be performed and the like. The invention solves the problems that the calculation result is not accurate enough, the safety distance can not be ensured to meet the expectation of the driver in real time and the practicability is not strong in the prior art, provides the definition of the vehicle trust degree, namely the vehicle performance which can be exerted by the driver, and improves the accuracy of the driver in judging the safety distance.

Description

Vehicle longitudinal driving safety distance estimation method based on vehicle trust
Technical Field
The invention belongs to the technical field of automobile active safety, and particularly relates to a vehicle safety distance estimation method based on vehicle trust degree in a following scene.
Background
The coupling relation of the human-vehicle roads is embodied on the traffic flow characteristics. If the single lane capacity is highly dependent on the driver's choice of the following distance, which is a result of human-vehicle interaction, the driver's choice of the following distance directly affects the vehicle's capacity.
Under the networking environment, the information that the driver can learn increases, and under the environment of known vehicle information around, the main reason of the accident of knocking into the back is that the driver is not enough to the performance judgement of the vehicle that drives, and vehicle trust is too high or low, can make the driver unreasonable to the selection of car following distance, easily takes place the accident or influences the efficiency of passing.
The safe vehicle distance model at the present stage mainly analyzes the collected parameter data through a basic statistical method, and because the data collection of the limit driving working condition is not carried out, the vehicle performance which can be exerted by a driver can not be accurately described, the specificity of the driver can not be embodied, and only the safe vehicle distance model which embodies the specificity of the driver is more in line with the expectation of the driver, thereby providing the most effective and accurate safe distance information for the driver.
Therefore, there is a need in the art for a new solution to solve the above problems.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the vehicle longitudinal driving safety distance estimation method based on the vehicle confidence is provided to solve the problems that in the prior art, specific parameters of a driver cannot be dynamically changed, the coupling relation between various tires and a road surface is not considered, and the safety distance does not meet the expectation of the driver.
A vehicle longitudinal driving safety distance estimation method based on vehicle confidence is characterized by comprising the following steps: comprises the following steps which are sequentially carried out,
step one, data extraction
Acquiring natural driving data of a driver through a CAN bus; extracting straight line, right-angle bend, S bend and U-shaped bend section data in the driving data, and judging the driving road surface condition according to the wheel speed signal and ABS data;
step two, data analysis
Determining the weight of each road section by an analytic hierarchy process; the dimension reduction is carried out through principal component analysis to obtain principal component weight coefficients which are respectively absolute values F of the maximum braking deceleration in the deceleration stage1Coefficient of road surface adhesion F2Time taken to complete a road section F3；
Step three, establishing a vehicle trust level module
Obtaining vehicle trust degree comprehensive scores of the driver under various road conditions according to the principal component weight coefficients obtained in the step two; carrying out primary classification on the vehicle trust level by adopting K-means clustering analysis by taking the comprehensive score as a division index; training a network relation between the principal component and the vehicle confidence by using a BP neural network algorithm to obtain a vehicle confidence module;
step four, establishing a safe vehicle distance model based on vehicle trust
Optimizing and improving the safe vehicle distance model through the vehicle trust degree module obtained in the third step by adopting the existing safe vehicle distance model to obtain a safe vehicle distance model based on the vehicle trust degree;
step five, updating the running vehicle in real time
Continuously updating the collected data in the continuous operation of the vehicle, and repeating the first step to the fourth step to obtain a real-time updated safe vehicle distance model based on the vehicle trust level.
The vehicle trust level is expressed by a safe vehicle distance value.
The maximum slip rate of the running road surface condition passing through the road surface in the step one
Figure GDA0003147874560000021
In the expression, s is the maximum road surface slip rate, v is the vehicle speed, and rw is the ABS wheel speed signal.
The consistency index of various road section weights in the step two
Figure GDA0003147874560000022
Consistency ratio
Figure GDA0003147874560000023
Where n is the index number, CI is the consistency index, CR is the consistency ratio, where RI is the random consistency index.
The step three BP neural network algorithm adopts an input layer as a main component, and the absolute value | a of the maximum braking deceleration in the deceleration stagemaxI, a road surface adhesion coefficient mu, the time t for completing a road section, the output layer is the vehicle confidence degree delta, and the network relation delta is obtained after training as f (| a)max|，μ，t)。
And analyzing the acquired parameter data by the existing safe vehicle distance model in the step four through a basic statistical method to obtain the safe vehicle distance model.
Through the design scheme, the invention can bring the following beneficial effects: a vehicle longitudinal driving safety distance estimation method based on vehicle trust solves the problems that in the prior art, calculation results are not accurate enough, the safety distance can not be guaranteed to meet the expectation of a driver in real time, and the practicability is not strong.
Drawings
The invention is further described with reference to the following figures and detailed description:
FIG. 1 is a schematic overall flow chart of a vehicle longitudinal driving safety distance estimation method based on vehicle confidence level.
Fig. 2 is a detailed flow diagram of a vehicle longitudinal driving safety distance estimation method based on vehicle confidence level according to the present invention.
FIG. 3 is a schematic diagram of a BP neural network training method for estimating the longitudinal driving safety distance of a vehicle based on vehicle confidence.
Detailed Description
A method for estimating the safe distance of vehicle longitudinal driving based on vehicle confidence level, as shown in FIG. 1 and FIG. 2, the implementation process comprises the following steps,
1001, collecting natural driving data through a CAN bus;
specifically, the natural driving data of the driver is acquired through the CAN bus, and the acquired vehicle driving data includes but is not limited to: vehicle speed, throttle opening, brake pressure, steering angle, etc.
Step 1002, extracting four road section data;
specifically, four road sections capable of reflecting vehicle confidence are selected: the method comprises the steps of straight line, right-angle bend, S bend and U-shaped bend, and four road data are extracted from natural driving data by specifying parameter ranges through three parameters of steering angle, braking pressure and throttle opening collected by a CAN bus.
Setting the range of the three parameters according to the vehicle speed, the steering angle, the brake pressure and the opening degree of a throttle valve, which are acquired by a CAN bus, and extracting four road sections of a straight line, a right-angle bend, an S bend and a U-shaped bend;
under the condition of high speed (v is more than or equal to 80km/h and less than or equal to 120 km/h):
Figure GDA0003147874560000041
under the condition of medium speed (v is more than or equal to 60km/h and less than or equal to 80 km/h):
Figure GDA0003147874560000042
under the condition of low speed (v is more than or equal to 20km/h and less than or equal to 60 km/h):
Figure GDA0003147874560000043
step 1003, judging road surface conditions;
specifically, the maximum slip rate of the current road surface CAN be obtained through the vehicle speed acquired by the CAN bus and the wheel speed signal of the ABS
Figure GDA0003147874560000044
Wherein v is the vehicle speed, rw is the wheel speed signal of ABS, and the road surface conditions can be judged according to the relationship between the slip rate and the road surface adhesion coefficient;
1004, performing weighting on the four road sections through a chromatography analysis method;
specifically, a hierarchical analysis matrix is established for four road sections; secondly, according to the influence degrees of the four road sections on the vehicle trust level, a pair comparison array is constructed by using a 1-9 comparison ruler; finally, calculating weight vector and carrying out consistency check, and carrying out consistency ratio
Figure GDA0003147874560000051
Where n is the index number, CI is the consistency index, CR is the consistency ratio, where RI is the random consistency index, which can be obtained by the following table:
n 1 2 3 4 5 6 7 8 9 10 11
RI 0 0 0.58 0.90 1.12 1.24 1.32 1.41 1.45 1.49 1.51
obtaining the weight value omega (omega) of the straight line, the right angle bend, the S bend and the U-shaped bend1,ω2,ω3,ω4) Further obtaining a weighted database;
step 1005, reducing the dimension of the data by a principal component analysis method;
specifically, the raw data is normalized to obtain the standard deviation and the average value of each parameter, and the formula used for the processing is
Figure GDA0003147874560000052
Wherein x isnM is a variable description of each sample, for the number of samples to be analyzed,
Figure GDA0003147874560000053
is the mean value of the samples, ynIs the standard deviation of the sample. Judging the correlation among the parameters; solving a matrix characteristic root and a characteristic vector; and calculating the principal component contribution rate and the accumulated contribution rate to obtain the database after dimension reduction. Selecting the first 2 principal components including F from the data parameters after dimensionality reduction1(absolute value of maximum braking deceleration in deceleration phase), F2(road surface adhesion coefficient), F3(time required for completing a road section), solving the weight coefficient of each index according to the principal component initial load factor, and specifically, dividing the initial load factor by the square value of the initial characteristic value, and obtaining a scoring model of the vehicle trust degree according to the weight coefficient: f ═ aF1+bF2+cF3Wherein, a, b and c are weight coefficients of each index obtained.
Step 1006, classifying the vehicle confidence levels by using K-means clustering;
specifically, 1-5 classification numbers are respectively explored by taking the obtained comprehensive trust score as an index, the internal overall difference is observed, and when the clustering data is larger than 3, the internal overall difference is not greatly reduced, so that the vehicle trust is graded into 3 types: high, medium, low.
Step 1007, training the relationship between each characteristic parameter and the vehicle confidence by using a BP neural network;
specifically, firstly, labeling an input layer and an output layer, dividing the score of the output layer into 1-10 scores, and setting the input layer of the neural network as the absolute value | a of the maximum braking deceleration in the deceleration stagemaxI, road surface adhesion coefficient mu, time t for completing a road section, and vehicle as output layerThe degree of trust δ. After training, the network relation delta f (| a) can be obtainedmaxL, μ, t). The network relationship is shown in fig. 3.
Step 1008, establishing a longitudinal safe vehicle distance model based on vehicle trust;
specifically, a classical safe vehicle distance model is selected, a vehicle trust degree module obtained through neural network training is introduced into the classical safe vehicle distance model, and the safe vehicle distance model based on the vehicle trust degree is constructed.
The operation process of the specific model is as follows:
(1) front vehicle running at constant speed
Figure GDA0003147874560000061
(2) Deceleration running of front vehicle
Figure GDA0003147874560000062
(3) Acceleration of front vehicle
Figure GDA0003147874560000063
Wherein the safe distance between the two vehicles is d, and the distance traveled by the front vehicle is d1The distance traveled by the rear vehicle is d2The minimum distance between two vehicles after stopping is d0(ii) a Reaction time t of driver1T 'is brake clearance elimination time'2(ii) a The deceleration of the preceding vehicle deceleration running is
Figure GDA0003147874560000064
Initial braking speed of
Figure GDA0003147874560000065
The deceleration of the bicycle is a0Initial braking velocity v0The vehicle confidence is δ ═ f (| a)maxL, mu, t), the brake force increase time is t2", the preceding vehicle is running at an accelerated speedAcceleration of at2The initial braking speed of the front vehicle is vt0。
And (4) continuously updating the collected data along with the continuous operation of the vehicle, repeating the steps and continuously updating the calculation result.
The invention provides a vehicle trust level-based safe distance estimation method for longitudinal vehicle running, which introduces a concept of vehicle trust level, defines the trust level of a driver on a vehicle in the driving process, directly reflects the control of the driver on the vehicle in the driving process, and indirectly reflects the behaviors of keeping the safe distance, whether safe overtaking and the like. And the vehicle trust degree is evaluated, so that a driver can more clearly recognize the vehicle and the road environment, the vehicle performance which can be exerted by the driver is determined, the proper safety distance which is required to be kept with a front vehicle is further determined, and a reliable basis is provided for the accuracy of a safety distance model in the internet environment. In addition, the data collected by the invention is vehicle operation data, and the collection method is mature, for example, a controller local area network becomes the factory standard configuration of most automobiles. Therefore, the invention has strong practicability and generalization and is also suitable for scenes such as lane changing, overtaking and the like.
The above description is only an example of the method of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention.

Claims (6)

1. A vehicle longitudinal driving safety distance estimation method based on vehicle confidence is characterized by comprising the following steps: comprises the following steps which are sequentially carried out,
step one, data extraction
Acquiring natural driving data of a driver through a CAN bus; extracting straight line, right-angle bend, S bend and U-shaped bend section data in the driving data, and judging the driving road surface condition according to the wheel speed signal and ABS data;
step two, data analysis
Determination of each by analytic hierarchy processA weight of the seed road segment; the dimension reduction is carried out through principal component analysis to obtain principal component weight coefficients which are respectively absolute values F of the maximum braking deceleration in the deceleration stage1Coefficient of road surface adhesion F2Time taken to complete a road section F3；
Step three, establishing a vehicle trust level module
Obtaining vehicle trust degree comprehensive scores of the driver under various road conditions according to the principal component weight coefficients obtained in the step two; carrying out primary classification on the vehicle trust level by adopting K-means clustering analysis by taking the comprehensive score as a division index; training a network relation between the principal component and the vehicle confidence by using a BP neural network algorithm to obtain a vehicle confidence module;
step four, establishing a safe vehicle distance model based on vehicle trust
Optimizing and improving the safe vehicle distance model through the vehicle trust degree module obtained in the third step by adopting the existing safe vehicle distance model to obtain a safe vehicle distance model based on the vehicle trust degree;
step five, updating the running vehicle in real time
Continuously updating the collected data in the continuous operation of the vehicle, and repeating the first step to the fourth step to obtain a real-time updated safe vehicle distance model based on the vehicle trust;
the vehicle trust level is the vehicle performance exerted by the driver, and the accuracy of the driver in judging the safe distance is improved.
2. The vehicle longitudinal driving safety distance estimation method based on the vehicle credibility as claimed in claim 1, wherein: the vehicle trust level is expressed by a safe vehicle distance value.
3. The vehicle longitudinal driving safety distance estimation method based on the vehicle credibility as claimed in claim 1, wherein: the maximum slip rate of the running road surface condition passing through the road surface in the step one
Figure FDA0003147874550000011
In the expression, s is the maximum road surface slip rate, v is the vehicle speed, and rw is the ABS wheel speed signal.
4. The vehicle longitudinal driving safety distance estimation method based on the vehicle credibility as claimed in claim 1, wherein: the consistency index of various road section weights in the step two
Figure FDA0003147874550000021
Consistency ratio
Figure FDA0003147874550000022
Where n is the index number, CI is the consistency index, CR is the consistency ratio, where RI is the random consistency index.
5. The vehicle longitudinal driving safety distance estimation method based on the vehicle credibility as claimed in claim 1, wherein: the step three BP neural network algorithm adopts an input layer as a main component, and the absolute value | a of the maximum braking deceleration in the deceleration stagemaxI, a road surface adhesion coefficient mu, the time t for completing a road section, the output layer is the vehicle confidence degree delta, and the network relation delta is obtained after training as f (| a)max|，μ，t)。
6. The vehicle longitudinal driving safety distance estimation method based on the vehicle credibility as claimed in claim 1, wherein: and analyzing the acquired parameter data by the existing safe vehicle distance model in the step four through a basic statistical method to obtain the safe vehicle distance model.
CN202011130600.1A 2020-10-21 2020-10-21 Vehicle longitudinal driving safety distance estimation method based on vehicle trust Active CN112319479B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011130600.1A CN112319479B (en) 2020-10-21 2020-10-21 Vehicle longitudinal driving safety distance estimation method based on vehicle trust

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011130600.1A CN112319479B (en) 2020-10-21 2020-10-21 Vehicle longitudinal driving safety distance estimation method based on vehicle trust

Publications (2)

Publication Number Publication Date
CN112319479A CN112319479A (en) 2021-02-05
CN112319479B true CN112319479B (en) 2021-08-24

Family

ID=74311806

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011130600.1A Active CN112319479B (en) 2020-10-21 2020-10-21 Vehicle longitudinal driving safety distance estimation method based on vehicle trust

Country Status (1)

Country Link
CN (1) CN112319479B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114913714A (en) * 2022-05-05 2022-08-16 中国第一汽车股份有限公司 Method for determining a safe distance of a vehicle

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100209885A1 (en) * 2009-02-18 2010-08-19 Gm Global Technology Operations, Inc. Vehicle stability enhancement control adaptation to driving skill based on lane change maneuver
CN108891418B (en) * 2018-07-10 2020-04-14 湖南大学 Driver confidence-based adaptive cruise control device and method
CN109624987B (en) * 2019-01-30 2020-06-30 长安大学 Method for calculating longitudinal running safety distance of vehicle in wet and slippery road surface state
US11314258B2 (en) * 2019-12-27 2022-04-26 Intel Corporation Safety system for a vehicle

Also Published As

Publication number Publication date
CN112319479A (en) 2021-02-05

Similar Documents

Publication Publication Date Title
CN111238825B (en) Intelligent driving automatic emergency braking performance testing method for combined test pavement
CN109552338B (en) Evaluation method and system for ecological driving behavior of pure electric vehicle
CN105034986B (en) A kind of driver's steering characteristic on-line identification method and device
CN112046489B (en) Driving style identification algorithm based on factor analysis and machine learning
CN113401123B (en) Automobile prediction cruise parameter self-tuning control system fusing driving mode information
CN111563555A (en) Driver driving behavior analysis method and system
CN114169444B (en) Driving style classification method considering risk potential field distribution under vehicle following working condition
CN112319479B (en) Vehicle longitudinal driving safety distance estimation method based on vehicle trust
CN113581188A (en) Commercial vehicle driver driving style identification method based on Internet of vehicles data
CN111783943A (en) Driver braking strength prediction method based on LSTM neural network
CN115774942A (en) Driving style identification model modeling and statistical method based on Internet of vehicles real vehicle data and SVM
CN113222385A (en) Method for constructing and evaluating driving condition of electric automobile
CN111340074B (en) Driver braking intention identification method based on improved RBF neural network
CN113761715A (en) Method for establishing personalized vehicle following model based on Gaussian mixture and hidden Markov
CN117021959A (en) Method for acquiring coasting recovery torque based on driving style identification
CN114817371A (en) XGboost-based vehicle driving condition identification method
CN114162132B (en) Driving mode identification method based on subjective and objective evaluation
CN116373881A (en) In-transit driving style identification method based on driving behavior semantic understanding
CN113743466B (en) Road type identification method and system based on decision tree
CN116975671A (en) L3-level automatic driving vehicle driver trust evaluation method and system
CN109910865B (en) Vehicle early warning braking method based on Internet of things
CN114613131B (en) Safety margin-based personalized forward collision early warning method
CN112528568B (en) Road feel simulation method based on K-Means and BP neural network
CN115009274A (en) Intelligent networking automobile risk assessment method and personalized decision-making method
CN113591780A (en) Method and system for identifying driving risk of driver

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