CN117312776B - Method and system for collecting, mining and analyzing characteristics of following acceleration scene data - Google Patents
Method and system for collecting, mining and analyzing characteristics of following acceleration scene data Download PDFInfo
- Publication number
- CN117312776B CN117312776B CN202311587866.2A CN202311587866A CN117312776B CN 117312776 B CN117312776 B CN 117312776B CN 202311587866 A CN202311587866 A CN 202311587866A CN 117312776 B CN117312776 B CN 117312776B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- following
- acceleration
- scene
- self
- 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
- 230000001133 acceleration Effects 0.000 title claims abstract description 201
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000005065 mining Methods 0.000 title claims description 10
- 238000004140 cleaning Methods 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000009826 distribution Methods 0.000 claims abstract description 13
- 230000007613 environmental effect Effects 0.000 claims description 39
- 238000000605 extraction Methods 0.000 claims description 20
- 239000012634 fragment Substances 0.000 claims description 13
- 230000000737 periodic effect Effects 0.000 claims description 11
- 238000012937 correction Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 125000004122 cyclic group Chemical group 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims description 2
- 230000006399 behavior Effects 0.000 description 22
- 238000010586 diagram Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 7
- 238000011161 development Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/40—Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/174—Form filling; Merging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/206—Drawing of charts or graphs
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method and a system for acquiring, excavating and analyzing characteristics of following acceleration scene data, wherein the method comprises the following steps: acquiring self-driving information and environment vehicle state information; data cleaning processing is carried out on the driving information of the self-vehicle and the state information of the environment vehicle; judging whether the self-driving behavior meets the heel vehicle constraint condition according to the cleaned data, and marking a following yard Jing Jin which meets the following constraint condition; judging whether the marked following scene meets the heel vehicle acceleration constraint condition, if so, extracting a following vehicle acceleration scene segment meeting the requirement from the current marked following scene; and extracting characteristic parameters of the following acceleration scene segment, generating a characteristic parameter table, checking characteristic parameter distribution conditions through a characteristic parameter statistical chart based on the characteristic parameter table, and obtaining driving behavior characteristics of the following acceleration scene. The scheme provided by the invention can collect and store the own vehicle and scene data, and can ensure the effectiveness and richness of the data.
Description
Technical Field
The invention belongs to the field of scene recognition and data mining in intelligent driving, and particularly relates to a method and a system for acquiring, mining and analyzing following acceleration scene data.
Background
The vehicle following behavior mainly describes the driving behavior of a self vehicle following a front vehicle in a lane and influenced by the front vehicle along with the change of the traffic density of a road under the condition of vehicle queuing running on a single lane. The vehicle following acceleration behavior refers to acceleration reaction of a driver of a rear vehicle caused by the increase of the relative distance between the front vehicle and the rear vehicle due to the fact that the front vehicle suddenly accelerates and drives away under the premise that the vehicle and the front vehicle are in a stable following state, and is one of typical driving behaviors of the driver, and is also an important data support of intelligent driving functions such as starting and stopping of the following vehicle, distance regulation and the like. Extensive research into this driving behavior has been conducted to facilitate the development of related intelligent driving assistance technologies.
The current intelligent driving has entered the data-driven era, and massive driving scene libraries and databases can be obtained through high efficiency and low cost, which are the basis for driving the iteration of the intelligent driving technology. The data about the following acceleration scene in the current natural driving scene library is not abundant, scene boundaries are not clear, the scene false extraction and missing extraction rate is higher, the scene identification extraction efficiency is low, the scene information is not comprehensive, the related characteristic analysis content is insufficient, and the like, the driving behavior characteristics of a driver are difficult to characterize, the deep learning and optimization of an intelligent driving decision rule system are not enough to support, and the requirements of the development of related technologies such as functional safety, humanization, comfort and the like of the existing intelligent driving technology cannot be met.
Disclosure of Invention
The invention aims to provide a method and a system for acquiring, mining and analyzing features of a vehicle-following acceleration scene, which are used for solving the problems of insufficient data set, low scene recognition and extraction efficiency, unclear scene boundary, high scene false extraction and missing extraction rate, incomplete scene information and insufficient related feature analysis content of the vehicle-following acceleration scene.
In order to achieve the purpose of the invention, the technical scheme provided by the invention is as follows:
first aspect
The invention provides a method for acquiring, excavating and analyzing characteristics of following acceleration scene data, which comprises the following steps:
step S1: acquiring self-driving information and environment vehicle state information;
step S2: performing data cleaning processing on the self-vehicle driving information and the environmental vehicle state information to obtain cleaned data;
step S3: judging whether the self-driving behavior meets the heel vehicle constraint condition according to the cleaned data, and marking a following yard Jing Jin which meets the following constraint condition;
step S4: adding a following acceleration constraint condition to the marked following field Jing Shi, judging whether the marked following scene meets the following acceleration constraint condition, and if so, extracting a following acceleration scene segment meeting the requirement from the current marked following scene;
step S5: and extracting the characteristic parameters of all the fragments of the following acceleration scene meeting the requirements, summarizing and generating a characteristic parameter table, checking the characteristic parameter distribution condition through a characteristic parameter statistical chart based on the characteristic parameter table, and acquiring the driving behavior characteristics of the following acceleration scene according to the characteristic parameter distribution condition.
Second aspect
Correspondingly to the method, the invention also provides a system for acquiring, excavating and analyzing the following acceleration scene data, which comprises an information acquisition unit, a data cleaning processing unit, a following scene marking unit, a following acceleration scene fragment extraction unit and a driving behavior characteristic acquisition unit;
the information acquisition unit is used for acquiring the driving information of the self-vehicle and the state information of the environment vehicle;
the data cleaning processing unit is used for performing data cleaning processing on the self-vehicle driving information and the environment vehicle state information to obtain cleaned data;
the following scene marking unit is used for judging whether the self-driving behavior is full of the heel constraint condition according to the data after the cleaning treatment and marking the following scene Jing Jin which meets the following constraint condition;
the following acceleration scene segment extraction unit is used for adding following acceleration constraint conditions to the marked following field Jing Shi, judging whether the marked following scene meets the following acceleration constraint conditions, and if so, extracting following acceleration scene segments meeting the requirements from the current marked following scene;
the driving behavior feature acquisition unit is used for extracting feature parameters of all the following acceleration scene fragments meeting the requirements, summarizing and generating a feature parameter table, checking feature parameter distribution conditions through a feature parameter statistical chart based on the feature parameter table, and acquiring driving behavior features of the following acceleration scene according to the feature parameter distribution conditions.
Compared with the prior art, the invention has the following technical advantages:
(1) The scheme provided by the invention can collect and store the own vehicle and scene data, and can ensure the effectiveness and richness of the data;
(2) The scheme provided by the invention can realize data cleaning and automatic extraction, compared with manual data cleaning and scene extraction, the recognition precision and extraction efficiency of the following acceleration scene can be greatly improved, and the cost of scene extraction can be reduced by greatly reducing the manpower consumption;
(3) The scheme provided by the invention can realize the generation of the characteristic parameter table and the characteristic parameter statistical chart, can intuitively and rapidly acquire the driving behavior characteristics of the following acceleration scene, and has important significance for the development and optimization of intelligent driving technology.
Drawings
FIG. 1 is a schematic flow chart of a method according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an information acquisition unit according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a vehicle position under a curve condition according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a following acceleration scenario provided in an embodiment of the present invention;
FIG. 5 is a schematic diagram of a feature parameter scatter diagram according to an embodiment of the present invention;
in the figure, a 1-millimeter wave radar; 2-a display; 3-a vehicle CAN bus; 4-a functional camera; 5-data storage.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is evident that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, the invention provides a method for acquiring, mining and analyzing characteristics of following acceleration scene data, which comprises the following steps:
step S1: acquiring self-driving information and environment vehicle state information;
the self-propelled vehicle information comprises a self-propelled vehicle length, a self-propelled vehicle width, a self-propelled vehicle speed, a self-propelled vehicle acceleration, a steering wheel corner and a distance from a lane line;
the environmental vehicle status information includes an environmental vehicle ID, an environmental vehicle type, an environmental vehicle length, an environmental vehicle width, and an environmental vehicle-to-vehicle relative distance, an environmental vehicle-to-vehicle relative angle, and an environmental vehicle-to-vehicle relative speed.
Step S2: performing data cleaning processing on the self-vehicle driving information and the environmental vehicle state information to obtain cleaned data; the data cleaning processing comprises data filling and filtering processing;
the data filling is specifically as follows: for fragments with continuous Na values and the quantity reaching a preset high threshold value in the self-driving information and the environment vehicle state information, defaulting to the fact that the system does not recognize that the information is not acquired and replacing the Na values with fixed float values; and (3) defaulting to the fragments of which the Na value does not reach the preset low threshold value in the self-driving information and the environment vehicle state information, namely temporarily losing the data abnormality, and filling and replacing the Na value by a linear interpolation method.
The filtering process specifically comprises the following steps: the noise is removed by a symmetric exponential moving average method.
Because certain noise exists in the self-driving information and the environment vehicle state information, the noise is removed by a filtering method, so that the data is smoother and has stronger usability.
Step S3: judging whether the self-driving behavior meets the heel vehicle constraint condition according to the cleaned data, and marking a following yard Jing Jin which meets the following constraint condition;
the method comprises the following steps of judging whether the driving behavior of the self-vehicle meets the constraint condition of the heel vehicle or not, and specifically comprising the following steps:
step S3.1: judging whether a following target vehicle exists or not, wherein the following target vehicle is specifically as follows:
fig. 3 is a schematic diagram of a vehicle position under a curve condition according to an embodiment of the present invention.
According to the relative distance between the vehicle and the own vehicleRelative angle of ambient vehicle and own vehicle->Calculating the relative lateral distance between the surrounding vehicle and the own vehicle>Longitudinal distance between surrounding vehicle and own vehicle>The calculation method is as follows:
；
relative lateral distance between environmental vehicle and own vehicle in case of curveCorrecting the curve curvature radius to ber，The relative lateral distance between the corrected environmental vehicle and the own vehicle is +.>The correction method is as follows:
；
according to the relative longitudinal distance between the vehicle and the own vehicleAnd the corrected relative lateral distance between the ambient vehicle and the host vehicleDetermining that an environmental vehicle which is positioned in a self-lane and in front of the self-lane is a following target vehicle;
step S3.2: if the following target vehicle exists, judging whether the following target vehicle is full of the heel vehicle constraint conditions, wherein the following constraint conditions are as follows:
；
wherein,longitudinal distance maximum distance constraint value for own vehicle and environmental vehicle, < ->For the width of the following target vehicle, +.>For the width of the lane line +.>Is the length of the front axle center of the car from the center of the right lane line, +.>Is the length of the front axle center of the car from the center of the left lane line, < >>For the auxiliary coefficient of the following scene, if the following scene is full of the heel constraint condition, the following scene is ++>The output value is 1, otherwise the output value is 0.
Step S4: adding a following acceleration constraint condition to the marked following field Jing Shi, judging whether the marked following scene meets the following acceleration constraint condition, and if so, extracting a following acceleration scene segment meeting the requirement from the current marked following scene;
judging whether the marked following scene meets the heel vehicle acceleration constraint condition or not, wherein the judging step comprises the following steps:
step S4.1: on the premise of meeting the following constraint condition, the self-vehicle acceleration time t1 is determined according to the following criteria:
；
wherein,for acceleration of bicycle>The self-vehicle acceleration rate is obtained by deriving the self-vehicle acceleration; />Presetting a threshold value for the acceleration of the vehicle; if and only if the acceleration of the bicycleIs larger than a preset threshold value of the acceleration of the bicycle>And the acceleration rate of the bicycle->If yes, determining the self-vehicle acceleration time t1;
step S4.2: the method comprises the steps of performing periodic cycle judgment on the acceleration of the vehicle from the vehicle acceleration time t1, and determining the starting time t2 of the acceleration of the vehicle when the first vehicle acceleration is judged to be greater than 0; the method comprises the steps of performing periodic cycle judgment on the acceleration of a vehicle from the acceleration time t1 of the vehicle, and determining the moment when the first acceleration of the vehicle is less than 0 as the end moment t3 of a vehicle following acceleration scene;
step S4.3: the starting time t4 of the following acceleration scene is determined, and the criterion is as follows:
；
wherein,for following the car target car acceleration, < >>A threshold value is preset for the acceleration of the following target vehicle,the acceleration rate is the target vehicle acceleration rate for following the vehicle; the method comprises the steps of performing periodic and cyclic judgment on target vehicle acceleration from a vehicle acceleration scene starting time t2, and determining the vehicle acceleration scene starting time t4 when judging that the first vehicle following target vehicle acceleration is larger than a vehicle following target vehicle acceleration preset threshold value and the vehicle following target vehicle acceleration rate is positive;
step S4.4: judging whether the time interval from the start time t4 of the following acceleration scene to the end time t3 of the following acceleration scene meets a preset following scene extraction range threshold value, and if so, judging that the following scene of the current mark is full of the heel vehicle acceleration constraint condition; if the following situations are not met, judging that the following situations of the current mark are not met with the heel vehicle acceleration constraint conditions, continuing to start from the finishing moment t3 of the following situations, and continuing to judge the following situations of the next mark until judging the following situations of all marks.
Wherein, the acceleration of the following target vehicleThe method comprises the following steps of obtaining the relative speed of a target vehicle, sampling the time length, estimating the acceleration condition of the target vehicle according to a linear regression method, and calculating the following formula:
；
；
a linear regression is performed on the time series of existing following target vehicle speed information,for the following target vehicle speed estimation, +.>For the intercept, n is the length of the data sequence, in the regression equation +.>Time series representing speed information of following target car, regression equation +.>Information sequence representing the speed of a target vehicle following a car, +.>Is time-seriesAverage value->For the average value of the speed information, slope +.>Represents the acceleration of the following target vehicle>Is the case in (a).
Step S5: and extracting the characteristic parameters of all the fragments of the following acceleration scene meeting the requirements, summarizing and generating a characteristic parameter table, checking the characteristic parameter distribution condition through a characteristic parameter statistical chart based on the characteristic parameter table, and acquiring the driving behavior characteristics of the following acceleration scene according to the characteristic parameter distribution condition.
Wherein the characteristic parameter statistical graph is a scatter diagram, a histogram, a box diagram or the like.
As shown in fig. 5, a scatter diagram is generated by taking key parameters of acceleration time in 400 car-following acceleration scene segments, such as a car speed ego _v (km/h) and a longitudinal distance obj_d (m) between a car and a target car, and the scatter diagram is fitted to obtain a relation between ego _v (km/h) and obj_d (m) as follows:
；
wherein x represents the speed of the vehicle, y represents the longitudinal distance between the distance of the vehicle and the target vehicle, and R represents the fitting coefficient. The relationship between the speed of the vehicle and the longitudinal distance between the vehicle and the target vehicle can be basically reflected when most drivers accelerate along with the vehicle. The requirements for functional safety, humanization, comfort and other related technology optimization and development of the intelligent driving technology can be met.
The invention also provides a system for acquiring, excavating and analyzing the data of the accelerating scene of the following vehicle, which comprises an information acquisition unit, a data cleaning processing unit, a following vehicle scene marking unit, a following vehicle accelerating scene fragment extraction unit and a driving behavior characteristic acquisition unit;
the information acquisition unit is used for acquiring the driving information of the self-vehicle and the state information of the environment vehicle;
the data cleaning processing unit is used for performing data cleaning processing on the self-vehicle driving information and the environment vehicle state information to obtain cleaned data;
the following scene marking unit is used for judging whether the self-driving behavior is full of the heel constraint condition according to the data after the cleaning treatment and marking the following scene Jing Jin which meets the following constraint condition;
the following acceleration scene segment extraction unit is used for adding following acceleration constraint conditions to the marked following field Jing Shi, judging whether the marked following scene meets the following acceleration constraint conditions, and if so, extracting following acceleration scene segments meeting the requirements from the current marked following scene;
the driving behavior feature acquisition unit is used for extracting feature parameters of all the following acceleration scene fragments meeting the requirements, summarizing and generating a feature parameter table, checking feature parameter distribution conditions through a feature parameter statistical chart based on the feature parameter table, and acquiring driving behavior features of the following acceleration scene according to the feature parameter distribution conditions.
It should be noted that, as shown in fig. 2, the information collecting unit includes a sensor, a collecting vehicle, a data memory, etc., where the sensor includes a millimeter wave radar 1 and a functional camera 4, which are respectively fixed on the collecting vehicle and connected to the data memory 5, and meanwhile, the data memory 5 is also connected to the vehicle CAN bus 3 and the display 2, so as to identify, collect, store and display driving behavior information of the vehicle and the environmental vehicle, including relevant information such as the vehicle speed, the vehicle acceleration, the type of the target vehicle, the speed of the target vehicle, the relative distance of the target vehicle, etc.
The method comprises the following steps of judging whether the driving behavior of the self-vehicle meets the constraint condition of the heel vehicle or not, and specifically comprising the following steps:
step S3.1: judging whether a following target vehicle exists or not, wherein the following target vehicle is specifically as follows:
according to the relative distance between the vehicle and the own vehicleRelative angle of ambient vehicle and own vehicle->Calculating the relative lateral distance between the surrounding vehicle and the own vehicle>Longitudinal distance between surrounding vehicle and own vehicle>The calculation method is as follows:
；
relative lateral distance between environmental vehicle and own vehicle in case of curveCorrecting the curve curvature radius to ber，The relative lateral distance between the corrected environmental vehicle and the own vehicle is +.>The correction method is as follows:
；
according to the relative longitudinal distance between the vehicle and the own vehicleAnd the corrected relative lateral distance between the ambient vehicle and the host vehicleDetermining that an environmental vehicle which is positioned in a self-lane and in front of the self-lane is a following target vehicle;
step S3.2: if the following target vehicle exists, judging whether the following target vehicle is full of the heel vehicle constraint conditions, wherein the following constraint conditions are as follows:
；
wherein,longitudinal distance maximum distance constraint value for own vehicle and environmental vehicle, < ->For the width of the following target vehicle, +.>For the width of the lane line +.>Is the length of the front axle center of the car from the center of the right lane line, +.>Is the length of the front axle center of the car from the center of the left lane line, < >>For the auxiliary coefficient of the following scene, if the following scene is full of the heel constraint condition, the following scene is ++>The output value is 1, otherwise the output value is 0.
Judging whether the marked following scene meets the heel vehicle acceleration constraint condition or not, wherein the judging step comprises the following steps:
step S4.1: on the premise of meeting the following constraint condition, the self-vehicle acceleration time t1 is determined according to the following criteria:
；
wherein, among them,for acceleration of bicycle>The self-vehicle acceleration rate is obtained by deriving the self-vehicle acceleration; />Presetting a threshold value for the acceleration of the vehicle; if and only if the acceleration of the bicycleIs larger than a preset threshold value of the acceleration of the bicycle>And the acceleration rate of the bicycle->If yes, determining the self-vehicle acceleration time t1;
step S4.2: the method comprises the steps of performing periodic cycle judgment on the acceleration of the vehicle from the vehicle acceleration time t1, and determining the starting time t2 of the acceleration of the vehicle when the first vehicle acceleration is judged to be greater than 0; the method comprises the steps of performing periodic cycle judgment on the acceleration of a vehicle from the acceleration time t1 of the vehicle, and determining the moment when the first acceleration of the vehicle is less than 0 as the end moment t3 of a vehicle following acceleration scene;
step S4.3: the starting time t4 of the following acceleration scene is determined, and the criterion is as follows:
；
wherein,for following the car target car acceleration, < >>A threshold value is preset for the acceleration of the following target vehicle,the acceleration rate is the target vehicle acceleration rate for following the vehicle; the method comprises the steps of periodically and circularly judging acceleration of a target vehicle from a starting moment t2 of a self-vehicle acceleration scene, and judging that the acceleration of a first vehicle following target vehicle is larger than a preset threshold value of the acceleration of the vehicle following target vehicle and the vehicle following target is carried outThe vehicle acceleration rate is positive, and the vehicle acceleration rate is determined to be the starting time t4 of the following vehicle acceleration scene;
step S4.4: judging whether the time interval from the start time t4 of the following acceleration scene to the end time t3 of the following acceleration scene meets a preset following scene extraction range threshold value, and if so, judging that the following scene of the current mark is full of the heel vehicle acceleration constraint condition; if the following situations are not met, judging that the following situations of the current mark are not met with the heel vehicle acceleration constraint conditions, continuing to start from the finishing moment t3 of the following situations, and continuing to judge the following situations of the next mark until judging the following situations of all marks.
Finally, it should be noted that: the above-described embodiments are provided for illustration and description of the present invention only and are not intended to limit the invention to the embodiments described. In addition, it will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that many variations and modifications may be made in accordance with the teachings of the present invention, which fall within the scope of the claimed invention.
Claims (8)
1. The method for acquiring, excavating and analyzing the characteristics of the following acceleration scene data is characterized by comprising the following steps:
step S1: acquiring self-driving information and environment vehicle state information;
step S2: performing data cleaning processing on the self-vehicle driving information and the environmental vehicle state information to obtain cleaned data;
step S3: judging whether the self-driving behavior meets the heel vehicle constraint condition according to the cleaned data, and marking a following yard Jing Jin which meets the following constraint condition;
step S4: adding a following acceleration constraint condition to the marked following field Jing Shi, judging whether the marked following scene meets the following acceleration constraint condition, and if so, extracting a following acceleration scene segment meeting the requirement from the current marked following scene;
step S5: extracting characteristic parameters of all the following vehicle acceleration scene fragments meeting the requirements, summarizing and generating a characteristic parameter table, checking characteristic parameter distribution conditions through a characteristic parameter statistical chart based on the characteristic parameter table, and acquiring driving behavior characteristics of the following vehicle acceleration scene according to the characteristic parameter distribution conditions;
in step S3, the step of determining whether the self-driving behavior meets the heel constraint condition specifically includes:
step S3.1: judging whether a following target vehicle exists or not, wherein the following target vehicle is specifically as follows:
according to the relative distance rho between the environmental vehicle and the own vehicle and the relative angle theta between the environmental vehicle and the own vehicle, the relative transverse distance Y between the environmental vehicle and the own vehicle and the relative longitudinal distance X between the environmental vehicle and the own vehicle are calculated, and the calculation method is as follows:
correcting the relative transverse distance Y between the environment vehicle and the vehicle under the condition of a curve, wherein the curvature radius of the curve is r, the left-right negative criterion is followed, and the relative transverse distance Y between the corrected environment vehicle and the vehicle is Y xiu The correction method is as follows:
according to the relative longitudinal distance X between the ambient vehicle and the own vehicle and the relative transverse distance Y between the ambient vehicle and the own vehicle after correction xiu Determining that an environmental vehicle which is positioned in a self-lane and in front of the self-lane is a following target vehicle;
step S3.2: if the following target vehicle exists, judging whether the following target vehicle is full of the heel vehicle constraint conditions, wherein the following constraint conditions are as follows:
wherein X is lim For the longitudinal distance maximum distance constraint value W of the own vehicle and the environment vehicle j For following the width of the target vehicle，W i For the width of lane line L r L is the length of the center of the front axle of the automobile from the center of the right lane line l Is the length k of the center of the front axle of the automobile from the center of the left lane line d For the auxiliary coefficient of the following scene, if the following scene meets the constraint condition of the heel vehicle, k d The output value is 1, otherwise the output value is 0.
2. The method for acquiring, excavating and analyzing features of acceleration scene data according to claim 1, wherein in step S1, the vehicle information includes a vehicle length, a vehicle width, a vehicle speed, a vehicle acceleration, a steering wheel angle, and a distance from a lane line;
the environmental vehicle status information includes an environmental vehicle ID, an environmental vehicle type, an environmental vehicle length, an environmental vehicle width, and an environmental vehicle-to-vehicle relative distance, an environmental vehicle-to-vehicle relative angle, and an environmental vehicle-to-vehicle relative speed.
3. The method for acquiring, mining and analyzing features of a following acceleration scene according to claim 1, wherein in step S2, the data cleaning process includes a data filling and filtering process;
the data filling is specifically as follows: for fragments with continuous Na values and the quantity reaching a preset high threshold value in the self-driving information and the environment vehicle state information, defaulting to the fact that the system does not recognize that the information is not acquired and replacing the Na values with fixed float values; and (3) defaulting to the fragments of which the Na value does not reach the preset low threshold value in the self-driving information and the environment vehicle state information, namely temporarily losing the data abnormality, and filling and replacing the Na value by a linear interpolation method.
4. A method for acquiring, mining and analyzing features of a following acceleration scene according to claim 3, wherein the filtering process specifically comprises: the noise is removed by a symmetric exponential moving average method.
5. The method for acquiring, mining and analyzing features of a following acceleration scene according to claim 1, wherein in step S4, the judging flag is used for judging whether the following scene meets the following acceleration constraint condition, specifically as follows:
step S4.1: on the premise of meeting the following constraint condition, the self-vehicle acceleration time t1 is determined according to the following criteria:
wherein Ego_a is the acceleration of the vehicle, ego_a' is the acceleration rate of the vehicle, and the acceleration rate of the vehicle is obtained by deriving the acceleration rate of the vehicle; ego_a min Presetting a threshold value for the acceleration of the vehicle; if and only if the own vehicle acceleration Ego_a is greater than the own vehicle acceleration preset threshold Ego_a min And the acceleration rate Ego_a' of the vehicle is positive, and the vehicle acceleration time t1 is determined;
step S4.2: the method comprises the steps of performing periodic cycle judgment on the acceleration of the vehicle from the vehicle acceleration time t1, and determining the starting time t2 of the acceleration of the vehicle when the first vehicle acceleration is judged to be greater than 0; the method comprises the steps of performing periodic cycle judgment on the acceleration of a vehicle from the acceleration time t1 of the vehicle, and determining the moment when the first acceleration of the vehicle is less than 0 as the end moment t3 of a vehicle following acceleration scene;
step S4.3: the starting time t4 of the following acceleration scene is determined, and the criterion is as follows:
where obj_a is the following target vehicle acceleration, obj_a min Presetting a threshold value for the acceleration of the following target vehicle, wherein obj_a' is the acceleration rate of the following target vehicle; the method comprises the steps of performing periodic and cyclic judgment on target vehicle acceleration from a vehicle acceleration scene starting time t2, and determining the vehicle acceleration scene starting time t4 when judging that the first vehicle following target vehicle acceleration is larger than a vehicle following target vehicle acceleration preset threshold value and the vehicle following target vehicle acceleration rate is positive;
step S4.4: judging whether the time interval from the start time t4 of the following acceleration scene to the end time t3 of the following acceleration scene meets a preset following scene extraction range threshold value, and if so, judging that the following scene of the current mark is full of the heel vehicle acceleration constraint condition; if the following situations are not met, judging that the following situations of the current mark are not met with the heel vehicle acceleration constraint conditions, continuing to start from the finishing moment t3 of the following situations, and continuing to judge the following situations of the next mark until judging the following situations of all marks.
6. The method for acquiring, mining and analyzing features of a following acceleration scene according to claim 1, wherein in step S5, the feature parameter statistical map is a scatter plot, a histogram and/or a box plot.
7. The system is characterized by comprising an information acquisition unit, a data cleaning processing unit, a following scene marking unit, a following accelerating scene fragment extraction unit and a driving behavior characteristic acquisition unit;
the information acquisition unit is used for acquiring the driving information of the self-vehicle and the state information of the environment vehicle;
the data cleaning processing unit is used for performing data cleaning processing on the self-vehicle driving information and the environment vehicle state information to obtain cleaned data;
the following scene marking unit is used for judging whether the self-driving behavior is full of the heel constraint condition according to the data after the cleaning treatment and marking the following scene Jing Jin which meets the following constraint condition;
the following acceleration scene segment extraction unit is used for adding following acceleration constraint conditions to the marked following field Jing Shi, judging whether the marked following scene meets the following acceleration constraint conditions, and if so, extracting following acceleration scene segments meeting the requirements from the current marked following scene;
the driving behavior feature acquisition unit is used for extracting feature parameters of all the following vehicle acceleration scene fragments meeting the requirements, summarizing and generating a feature parameter table, checking feature parameter distribution conditions through a feature parameter statistical chart based on the feature parameter table, and acquiring driving behavior features of the following vehicle acceleration scene according to the feature parameter distribution conditions;
the method for judging whether the driving behavior of the self-vehicle meets the constraint condition of the heel vehicle comprises the following specific steps:
step S3.1: judging whether a following target vehicle exists or not, wherein the following target vehicle is specifically as follows:
according to the relative distance rho between the environmental vehicle and the own vehicle and the relative angle theta between the environmental vehicle and the own vehicle, the relative transverse distance Y between the environmental vehicle and the own vehicle and the relative longitudinal distance X between the environmental vehicle and the own vehicle are calculated, and the calculation method is as follows:
correcting the relative transverse distance Y between the environment vehicle and the vehicle under the condition of a curve, wherein the curvature radius of the curve is r, the left-right negative criterion is followed, and the relative transverse distance Y between the corrected environment vehicle and the vehicle is Y xiu The correction method is as follows:
according to the relative longitudinal distance X between the ambient vehicle and the own vehicle and the relative transverse distance Y between the ambient vehicle and the own vehicle after correction xiu Determining that an environmental vehicle which is positioned in a self-lane and in front of the self-lane is a following target vehicle;
step S3.2: if the following target vehicle exists, judging whether the following target vehicle is full of the heel vehicle constraint conditions, wherein the following constraint conditions are as follows:
wherein X is lim For the longitudinal distance maximum distance constraint value of the own vehicle and the environmental vehicle,W j for following the width of the target vehicle W i For the width of lane line L r L is the length of the center of the front axle of the automobile from the center of the right lane line l Is the length k of the center of the front axle of the automobile from the center of the left lane line d For the auxiliary coefficient of the following scene, if the following scene meets the constraint condition of the heel vehicle, k d The output value is 1, otherwise the output value is 0.
8. The system for acquiring, mining and analyzing features of a following acceleration scene according to claim 7, wherein the determining whether the marked following scene is full of the following acceleration constraint conditions is as follows:
step S4.1: on the premise of meeting the following constraint condition, the self-vehicle acceleration time t1 is determined according to the following criteria:
wherein Ego_a is the acceleration of the vehicle, ego_a' is the acceleration rate of the vehicle, and the acceleration rate of the vehicle is obtained by deriving the acceleration rate of the vehicle; ego_a min Presetting a threshold value for the acceleration of the vehicle; if and only if the own vehicle acceleration Ego_a is greater than the own vehicle acceleration preset threshold Ego_a min And the acceleration rate Ego_a' of the vehicle is positive, and the vehicle acceleration time t1 is determined;
step S4.2: the method comprises the steps of performing periodic cycle judgment on the acceleration of the vehicle from the vehicle acceleration time t1, and determining the starting time t2 of the acceleration of the vehicle when the first vehicle acceleration is judged to be greater than 0; the method comprises the steps of performing periodic cycle judgment on the acceleration of a vehicle from the acceleration time t1 of the vehicle, and determining the moment when the first acceleration of the vehicle is less than 0 as the end moment t3 of a vehicle following acceleration scene;
step S4.3: the starting time t4 of the following acceleration scene is determined, and the criterion is as follows:
where obj_a is the following target vehicle acceleration, obj_a min Presetting a threshold value for the acceleration of the following target vehicle, wherein obj_a' is the acceleration rate of the following target vehicle; the method comprises the steps of performing periodic and cyclic judgment on target vehicle acceleration from a vehicle acceleration scene starting time t2, and determining the vehicle acceleration scene starting time t4 when judging that the first vehicle following target vehicle acceleration is larger than a vehicle following target vehicle acceleration preset threshold value and the vehicle following target vehicle acceleration rate is positive;
step S4.4: judging whether the time interval from the start time t4 of the following acceleration scene to the end time t3 of the following acceleration scene meets a preset following scene extraction range threshold value, and if so, judging that the following scene of the current mark is full of the heel vehicle acceleration constraint condition; if the following situations are not met, judging that the following situations of the current mark are not met with the heel vehicle acceleration constraint conditions, continuing to start from the finishing moment t3 of the following situations, and continuing to judge the following situations of the next mark until judging the following situations of all marks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311587866.2A CN117312776B (en) | 2023-11-27 | 2023-11-27 | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311587866.2A CN117312776B (en) | 2023-11-27 | 2023-11-27 | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117312776A CN117312776A (en) | 2023-12-29 |
| CN117312776B true CN117312776B (en) | 2024-03-12 |
Family
ID=89288659
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311587866.2A Active CN117312776B (en) | 2023-11-27 | 2023-11-27 | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117312776B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103754224A (en) * | 2014-01-24 | 2014-04-30 | 清华大学 | Vehicle multi-target coordinating lane changing assisting adaptive cruise control method |
| CN109816811A (en) * | 2018-10-31 | 2019-05-28 | 杭州云动智能汽车技术有限公司 | A kind of nature driving data acquisition device |
| CN111710176A (en) * | 2020-06-08 | 2020-09-25 | 同济大学 | Intersection signal-vehicle track cooperative control method under cooperative vehicle and road environment |
| CN112466119A (en) * | 2020-11-26 | 2021-03-09 | 清华大学 | Method and system for predicting vehicle following speed of vehicle by using vehicle-road cooperative data |
| CN113487874A (en) * | 2021-05-27 | 2021-10-08 | 中汽研（天津）汽车工程研究院有限公司 | System and method for collecting, identifying and classifying following behavior scene data |
| CN114492043A (en) * | 2022-01-27 | 2022-05-13 | 吉林大学 | Personalized driver following modeling method considering perception limited characteristics |
| CN114973733A (en) * | 2022-04-29 | 2022-08-30 | 北京交通大学 | Method for optimizing and controlling track of networked automatic vehicle under mixed flow at signal intersection |
| CN115366914A (en) * | 2022-08-03 | 2022-11-22 | 清华大学苏州汽车研究院（相城） | Method, apparatus, and medium for controlling autonomous vehicle based on model predictive control |
| CN115713863A (en) * | 2022-10-31 | 2023-02-24 | 长沙理工大学 | Trajectory planning and signal optimization method for special phase for automatic driving |
-
2023
- 2023-11-27 CN CN202311587866.2A patent/CN117312776B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103754224A (en) * | 2014-01-24 | 2014-04-30 | 清华大学 | Vehicle multi-target coordinating lane changing assisting adaptive cruise control method |
| CN109816811A (en) * | 2018-10-31 | 2019-05-28 | 杭州云动智能汽车技术有限公司 | A kind of nature driving data acquisition device |
| CN111710176A (en) * | 2020-06-08 | 2020-09-25 | 同济大学 | Intersection signal-vehicle track cooperative control method under cooperative vehicle and road environment |
| CN112466119A (en) * | 2020-11-26 | 2021-03-09 | 清华大学 | Method and system for predicting vehicle following speed of vehicle by using vehicle-road cooperative data |
| CN113487874A (en) * | 2021-05-27 | 2021-10-08 | 中汽研（天津）汽车工程研究院有限公司 | System and method for collecting, identifying and classifying following behavior scene data |
| CN114492043A (en) * | 2022-01-27 | 2022-05-13 | 吉林大学 | Personalized driver following modeling method considering perception limited characteristics |
| CN114973733A (en) * | 2022-04-29 | 2022-08-30 | 北京交通大学 | Method for optimizing and controlling track of networked automatic vehicle under mixed flow at signal intersection |
| CN115366914A (en) * | 2022-08-03 | 2022-11-22 | 清华大学苏州汽车研究院（相城） | Method, apparatus, and medium for controlling autonomous vehicle based on model predictive control |
| CN115713863A (en) * | 2022-10-31 | 2023-02-24 | 长沙理工大学 | Trajectory planning and signal optimization method for special phase for automatic driving |
Non-Patent Citations (1)
| Title |
|---|
| "基于深度强化学习的驾驶员跟车模型研究";郭景华 等;《汽车工程》;参见正文第1-2章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117312776A (en) | 2023-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113291308B (en) | Vehicle self-learning lane-changing decision-making system and method considering driving behavior characteristics | |
| CN111599181B (en) | Typical natural driving scene recognition and extraction method for intelligent driving system test | |
| CN110775069B (en) | Vehicle driving mode recognition device and method in mixed driving mode | |
| CN112950811B (en) | New energy automobile region operation risk assessment and early warning system integrating whole automobile safety | |
| CN112668779B (en) | Front vehicle motion state prediction method based on self-adaptive Gaussian process | |
| CN113487874A (en) | System and method for collecting, identifying and classifying following behavior scene data | |
| CN112116100B (en) | Game theory decision method considering driver type | |
| CN112365710B (en) | Intelligent vehicle lane change decision-making method based on probability output model | |
| CN113442935B (en) | Method and system for judging poor driving behavior of commercial vehicle | |
| CN112706785A (en) | Method and device for selecting cognitive target of driving environment of automatic driving vehicle and storage medium | |
| CN117312776B (en) | Method and system for collecting, mining and analyzing characteristics of following acceleration scene data | |
| CN114842432A (en) | Automobile light control method and system based on deep learning | |
| CN116481560B (en) | Vehicle driving mileage calculation method, device, terminal and storage medium | |
| CN110509925B (en) | Method for identifying sharp turn based on Internet of vehicles data | |
| CN114074660A (en) | Predictive cruise fuel-saving control method and device and storage medium | |
| CN112298135B (en) | Vehicle braking distance optimization method | |
| CN112158206B (en) | Intelligent vehicle forced lane change merge point determination method and device | |
| US10906553B2 (en) | Systems and methods for vehicle acceleration event prediction inhibit | |
| CN113386773A (en) | Method and device for judging reliability of visual identification | |
| CN111127055A (en) | User portrait construction method and system based on vehicle body collection bus signals | |
| CN115472040B (en) | Personalized anti-collision early warning method for networked vehicle based on collision probability field | |
| CN117719524B (en) | Driving safety risk identification early warning method, device, terminal and storage medium | |
| CN113158415B (en) | Vehicle track similarity evaluation method based on error analysis | |
| CN117842073B (en) | Target vehicle lane change intention recognition method and device and vehicle | |
| CN114783177A (en) | Safe driving method and system of vehicle with intelligent driving function and working method |
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 |