CN116432448B - Variable speed limit optimization method based on intelligent network coupling and driver compliance - Google Patents
Variable speed limit optimization method based on intelligent network coupling and driver compliance Download PDFInfo
- Publication number
- CN116432448B CN116432448B CN202310360227.6A CN202310360227A CN116432448B CN 116432448 B CN116432448 B CN 116432448B CN 202310360227 A CN202310360227 A CN 202310360227A CN 116432448 B CN116432448 B CN 116432448B
- Authority
- CN
- China
- Prior art keywords
- speed limit
- compliance
- vehicle
- variable speed
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005457 optimization Methods 0.000 title claims abstract description 31
- 230000008878 coupling Effects 0.000 title claims abstract description 19
- 238000010168 coupling process Methods 0.000 title claims abstract description 19
- 238000005859 coupling reaction Methods 0.000 title claims abstract description 19
- 238000011217 control strategy Methods 0.000 claims abstract description 50
- 238000004088 simulation Methods 0.000 claims abstract description 48
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 31
- 230000002452 interceptive effect Effects 0.000 claims abstract description 5
- 238000001514 detection method Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 38
- 239000002245 particle Substances 0.000 claims description 36
- 230000008859 change Effects 0.000 claims description 26
- 238000012549 training Methods 0.000 claims description 21
- 230000008901 benefit Effects 0.000 claims description 16
- 238000012937 correction Methods 0.000 claims description 10
- 230000001133 acceleration Effects 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000011144 upstream manufacturing Methods 0.000 claims description 8
- 230000006399 behavior Effects 0.000 claims description 7
- 238000013136 deep learning model Methods 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000003064 k means clustering Methods 0.000 claims description 6
- 230000006855 networking Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000003993 interaction Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000007621 cluster analysis Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims description 3
- 239000011521 glass Substances 0.000 claims description 3
- 238000005286 illumination Methods 0.000 claims description 3
- 238000011835 investigation Methods 0.000 claims description 3
- 238000007477 logistic regression Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000002787 reinforcement Effects 0.000 claims description 3
- 230000001276 controlling effect Effects 0.000 description 6
- 230000006978 adaptation Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 230000019771 cognition Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 208000011597 CGF1 Diseases 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000004039 social cognition Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B9/00—Simulators for teaching or training purposes
- G09B9/02—Simulators for teaching or training purposes for teaching control of vehicles or other craft
- G09B9/04—Simulators for teaching or training purposes for teaching control of vehicles or other craft for teaching control of land vehicles
- G09B9/05—Simulators for teaching or training purposes for teaching control of vehicles or other craft for teaching control of land vehicles the view from a vehicle being simulated
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B9/00—Simulators for teaching or training purposes
- G09B9/02—Simulators for teaching or training purposes for teaching control of vehicles or other craft
- G09B9/04—Simulators for teaching or training purposes for teaching control of vehicles or other craft for teaching control of land vehicles
- G09B9/052—Simulators for teaching or training purposes for teaching control of vehicles or other craft for teaching control of land vehicles characterised by provision for recording or measuring trainee's performance
-
- 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)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Hardware Design (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Molecular Biology (AREA)
- Aviation & Aerospace Engineering (AREA)
- Educational Technology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Structural Engineering (AREA)
- Civil Engineering (AREA)
- Architecture (AREA)
- Medical Informatics (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Traffic Control Systems (AREA)
Abstract
The variable speed limit optimization method based on intelligent network coupling and driver compliance comprises the following steps: 1) Designing a three-dimensional road scene of the expressway; 2) Constructing an interactive simulation experiment platform based on the microscopic traffic simulation software and the group driving simulation platform; 3) After the experimental platform is started, the simulated traffic flow data are exported in real time, the current traffic flow risk level and the compliance degree of different drivers for controlling the simulated vehicles are calculated by using a traffic flow risk detection algorithm, and a preset lane-level variable speed limit control instruction is issued by using a lane-level variable speed limit issuing system in a road scene according to a preset lane-level variable speed limit control strategy. The method solves the problem that personalized speed limiting guidance is difficult to carry out for different drivers in an intelligent network environment and a man-machine mixed traffic environment by adjusting the compliance of a background vehicle generated based on simulation in a microscopic traffic simulation platform to a variable speed limiting instruction.
Description
Technical Field
The invention belongs to the technical field of traffic intelligent control, and particularly relates to a variable speed limit optimization method based on intelligent network coupling and driver compliance.
Background
With the rapid development of digital technology and the popularization and application of the digital technology in various industries, the public expects and demands on the intellectualization and the intellectualization of expressway service, how to solve the problems of traffic safety, congestion relief, traffic efficiency improvement, traffic flow regulation and control, automatic driving adaptation and intelligent networking by using modern means is more and more urgent, and industry response and implementation are needed, so that ubiquitous highway users can obtain the intellectualized and intelligent high-level service.
Along with the development of traffic sensing technology, the construction of traffic digital twin technology and parallel simulation system provides a powerful data base and an effective decision basis for variable speed limit control. The digital twin refers to virtual mapping of a physical system, provides a new solution idea and a high-precision calibration means for high-credibility microscopic traffic simulation, and enables high-precision microscopic traffic simulation under complex road conditions to be possible. On the other hand, with the fusion development of the simulated driving technology and the micro driving model, the group simulated driving can realize the behavior interaction among vehicles, so that the fusion of the micro traffic simulation platform and the group driving simulation platform becomes a new traffic simulation means for exploring driving behaviors. The method provides an effective research means for the exploration of the driving behaviors of the driver in the intelligent networking environment and the man-machine mixed traffic environment.
Disclosure of Invention
The invention provides a variable speed limit optimization method based on intelligent network coupling and driver compliance, which is used for solving the defects in the prior art.
The invention is realized by the following technical scheme:
the variable speed limit optimization method based on intelligent network coupling and driver compliance comprises the following steps:
1) Designing a three-dimensional road scene of the expressway, constructing the three-dimensional road scene of the expressway based on a group driving simulation platform, and simultaneously importing the scene into micro-traffic simulation software;
2) Constructing an interactive simulation experiment platform based on the microscopic traffic simulation software and the group driving simulation platform;
3) After the experimental platform is started, the simulated traffic flow data are exported in real time, the current traffic flow risk level and the compliance of different drivers for controlling the simulated vehicles are calculated by using a traffic flow risk detection algorithm, and a preset lane-level variable speed limit control instruction is issued by using a lane-level variable speed limit issuing system in a road scene according to a preset lane-level variable speed limit control strategy;
4) Recruiting a plurality of drivers to carry out virtual simulation group driving by utilizing a driving simulator to wear VR glasses, controlling the running states of simulated vehicles by the drivers, loading the drivers in a unified scene for interaction in the driving process aiming at large-scale group drivers and drivers in different driving styles under different traffic states and scenes, and loading background vehicles generated based on parallel simulation according to different traffic states; acquiring basic information of a driver by using a questionnaire investigation mode, and simultaneously acquiring experimental road section road environment information data, vehicle running state data and traffic state data;
5) Calculating the compliance degree of each driver in the simulated driving, wherein the compliance degree is defined as:
wherein the ratio is i (t) is the compliance of the vehicle at time i,for i speed of the vehicle 250m upstream of the location receiving the variable speed limit information,/i>For i vehicle speed 200m downstream of the location receiving variable speed limit information, VSL i And (t) is a variable speed limit value issued to the i vehicle at the moment t.
Wherein whenWhen the compliance was considered to be 100%;
6) The compliance degree of a background vehicle generated based on simulation in a micro traffic simulation platform to a variable speed limit instruction is adjusted, one part of the compliance degree can be selected to be set to be 100% as an automatic driving vehicle, and the other part of the compliance degree is set to be different compliance degrees according to requirements;
7) The driver and the vehicle are defined as high, medium and low compliance with 100% compliance and 80% compliance as limits; wherein compliance level is defined as:
8) Dividing drivers and vehicles into three data sets according to compliance grades, namely a high compliance data set, a medium compliance data set and a low compliance data set; and carrying out cluster analysis on parameter matrixes of vehicles with different compliance data sets in each time period by adopting a K-means clustering algorithm, wherein each column in the matrix is a time sequence of one parameter, and the parameter matrixes comprise the following parameters: (1) a road environment information data matrix: black/daytime, weather, illumination, current road alignment, gradient, current lane speed limit and distance variable speed limit plate distance; (2) a driver information data matrix: the sex of the driver, the driving age of the driver and the age of the driver; (3) a vehicle operating state data matrix: time, vehicle center longitudinal position, vehicle center lateral position, vehicle longitudinal speed, vehicle longitudinal acceleration, vehicle lateral speed, vehicle lateral acceleration, lane offset, steering wheel angle, and vehicle speed product; (4) a traffic state data matrix: the longitudinal speed of the front vehicle of the lane, the time interval of the vehicle head, the distance between the front vehicles of the adjacent lanes, the longitudinal speed of the front vehicles of the adjacent lanes, the distance between the front vehicles and the rear vehicles of the lane, the traffic saturation, the average speed of the current road section and the current traffic flow; the Euclidean distance is used as an index for measuring the similarity of the data points, the distance between each data point in the sample set and the initial particle is calculated, and the closest principle is adopted to distribute the data points to the closest particles, so that the similar data can be classified into a cluster; dividing the parameter matrix data set into 3 clusters, namely, exciting, resisting and taking care; the drivers are finally classified into 9 categories: high aggressive compliance, high careless compliance, high careful compliance, aggressive compliance, medium-care compliance, small careful compliance, low aggressive compliance, low careless compliance, low careful compliance, constructing a driver personality style matrix;
9) Taking the style of the driver as a dependent variable and a parameter matrix as an independent variable, and training a deep learning model based on SAEs;
training a SAEs-based deep learning model, taking a training set as input, the first layer being trained as an automatic encoder; after the first hidden layer is obtained, the output of the mth hidden layer is used as the input of the (m+1) th hidden layer; the model structure consists of SAEs for extracting short-time traffic flow characteristics and a logistic regression layer for supervised parameter matrix prediction; training a deep network by adopting a Back Propagation (BP) algorithm based on a gradient optimization technology, wherein a greedy layered unsupervised learning algorithm shows advantages due to training each layer of parameters in the deep network from bottom to top in sequence; after the pre-training stage is completed, parameters of the prediction model are adjusted from top to bottom by means of the BP neural network, and finally a driving characteristic type judging model based on a parameter matrix is obtained;
10 The variable speed limit decision module is used for optimizing speed limit instructions for drivers of different driving styles under different roads and traffic scenes respectively, searching for a proper variable speed limit control strategy with higher compliance for the driving styles, using the comprehensive risk level based on individual vehicle safety and traffic flow overall safety as an actual rewarding value, using the driver compliance as a correction factor, establishing lane-level variable speed limit control schemes taking the driver compliance into consideration under the intelligent networking environment and the man-machine mixed traffic environment, forming a control scheme strategy library, and the variable speed limit information release module is used for releasing speed limit information for road users by using RSU (reactive power unit), so as to provide personalized speed limit guidance for different drivers under complex traffic environment.
In the above variable speed limit optimization method based on intelligent network coupling and driver compliance, in the step 2), the running state of the micro traffic flow is controlled by the micro traffic simulation software, the simulated vehicles controlled by the driver are generated by group driving simulation software, and the information of the simulated vehicles is synchronously updated in real time in the micro traffic simulation software and the group driving simulation software, wherein the group driving simulation vehicles can interact with each other, and meanwhile, the group driving simulation vehicles can interact with the simulated vehicles generated by the micro traffic simulation software.
According to the variable speed limit optimization method based on intelligent network coupling and driver compliance, in the step 3), RSUs distributed on the road side are information instruction issuing devices of the variable speed limit system, the information instruction issuing devices are distributed at intervals of 500m, different variable speed limit information is issued for different lanes by each instruction, namely lane-level speed limit instructions on different lanes and upstream and downstream can have different speed limit values, and the speed limit values of the variable speed limit marks are adjusted through real-time change, so that the running state of road traffic flow is changed in real time.
In the step 6), an online simulation module is implanted in a microscopic traffic simulation platform, wherein an automatic driving vehicle adopts an IDM (integrated digital matrix) following and lane changing model to simulate the vehicle position in real time, the intelligent network vehicle utilizes a Q-learning reinforcement learning algorithm to simulate the vehicle position in real time, a manual driving vehicle utilizes a Gipps following and lane changing model to simulate the vehicle position in real time, and three models are loaded into the online simulation module according to different proportions and speed limiting compliance.
In the above-mentioned variable speed limit optimization method based on intelligent network coupling and driver compliance, in the step 8), the parameter value rule of the parameter matrix is as follows: the extraction positions are 200m, 800m, 1400m and 2000m upstream of the position where the vehicle receives the variable speed limit instruction, and 400m and 1000m downstream of the position where the vehicle receives the variable speed limit instruction, the extraction time granularity is 5min, and the extraction time is 5min, 10min, 15min, 20min, 25min and 30min before the vehicle receives the variable speed limit instruction.
In the step 9), aiming at driving styles of different drivers, an improved particle swarm optimization algorithm is adopted to find an optimal variable speed limit control strategy, comprehensive risks are used as actual rewarding values, individual vehicle compliance is used as a correction factor, and finally a variable speed limit control scheme suitable for each driving style is established to form a control scheme strategy library. The method comprises the following steps:
1) Judging the driving style of a driver by using a driving style judging method, initializing the speed limit after the judgment is finished, and initializing the variable speed limit speed change gradient and the speed change period;
2) Initializing and setting a variable speed limiting speed change threshold;
3) Controlling the variable speed limit change according to the driving risk, if Q is more than 0.30, changing the variable speed limit, calculating the driving risk at the end of the control period, if Q is still more than 0.30, traversing, adjusting and optimizing the variable speed limit plate speed change gradient and speed change period of the current variable speed limit strategy until the driving risk benefit E (mu) of the combined control strategy p ) And (3) highest, incorporating the control strategy into a variable speed limit strategy library, thereby obtaining targeted variable speed limit control strategies aiming at different driving styles.
In the step 9), parameters in the parameter matrix are used as input variables and are subjected to data standardization processing, an original data set is established by using the driving style type obtained in the step 8), and a random forest method is adopted to perform dimension reduction processing on data dimensions to obtain a training data set.
The variable speed limit optimization method based on intelligent network coupling and driver compliance as described above, in the step 10), the comprehensive risk level algorithm based on individual vehicle safety and traffic flow overall safety, the method comprises the following steps:
1) Establishing individual vehicle safety risk level judgment and evaluation standards, wherein dangerous behaviors are defined as follows:
TTC i i the collision time of the vehicle relative to the front vehicle at the time t, because the vehicle position acquired by data is the position of the vehicle head, X is the position of the vehicle head i (t) is the position of the head of the vehicle at the moment i, X h (t) is the position of the front h head of the front car at the moment i, l h Is the length of the body of the h car, V i (t) is the instantaneous speed of the i-car at time t, V h (t) is the instantaneous speed of the vehicle at time t;
the individual vehicle security risk level definition method comprises the following steps of:
wherein,
wherein,representing the collision risk of i vehicles relative to the preceding vehicle at the moment t, TTC p TTC representing all individual vehicles at that moment i Is a mean value of (c).
2) By using a K-means clustering methodThe clusters are three categories, high risk level, medium risk level and low risk level, respectively. Using Risk (i) to represent i car relative to the front at time tCollision risk level of the vehicle, namely:
3) And constructing a traffic flow overall safety risk level judgment and evaluation standard, and counting the proportion of vehicles with high, medium and low risk levels in the road section at the moment t. Wherein the overall security risk level of the traffic flow is defined as:
4) Building a comprehensive risk algorithm based on individual vehicle safety and overall traffic flow safety:
wherein α+β=1
5) Aiming at drivers with different driving styles, an improved particle swarm optimization algorithm is adopted to find an optimal variable speed limit control strategy, the comprehensive risk Q is used as an actual rewarding value, and the individual vehicle compliance rate is adopted i And (t) taking the variable speed limit control scheme as a correction factor, and finally establishing a variable speed limit control scheme suitable for each driving style to form a control scheme strategy library.
Wherein the individual vehicle security risk benefits of the combined control strategy are defined as:
wherein E (mu) p ) Representing the risk benefit under the control strategy,the risk is integrated before the control strategy is implemented,the risk is integrated after implementation for the control strategy.
Compliance combining loss for a combined control strategy is defined as:
wherein the ratio is i (t) b Individual vehicle compliance, compactness, prior to implementation of the control strategy i (t) a Individual vehicle compliance after implementation of the control strategy.
6) The variable speed limit information release module utilizes RSUs arranged on the road sides to release speed limit information to road users.
The invention has the advantages that:
1. the method is based on a group driving simulation platform to build a three-dimensional road scene of the expressway; based on the interactive microscopic traffic simulation platform, background traffic is loaded to the three-dimensional road scene, and the problem that personalized speed limiting guidance is difficult to conduct for different drivers in an intelligent network environment and a man-machine mixed traffic environment and the problem of personalized speed limiting guidance under the mixed running condition of vehicles with different compliance degrees are solved by adjusting the compliance of the background vehicles generated based on simulation in the microscopic traffic simulation platform to the variable speed limiting instructions.
2. According to the invention, a variable speed limit decision module is utilized to respectively optimize speed limit instructions for drivers of different driving styles under different roads and traffic scenes, a proper variable speed limit control strategy with higher compliance for the driving styles is searched, an individual traffic safety index is utilized as an actual rewarding value, the driver compliance is utilized as a correction factor, lane-level variable speed limit control schemes considering the driver compliance under an intelligent networking environment and a man-machine mixed traffic environment are established, and a control scheme strategy library is formed. The method solves the problem that the driving risk of part of driving vehicles is increased due to lower compliance of drivers in the traditional variable speed limit control method. Meanwhile, the method has the characteristics of replicable popularization and strong robustness.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it will be obvious that the drawings in the following description are some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art.
FIG. 1 is a schematic flow diagram of the principle of the variable speed limit optimization method of the invention;
FIG. 2 is a schematic diagram of a decision flow of a variable speed limit control module according to the present invention;
FIG. 3 is a schematic diagram of an on-line simulation module according to the present invention;
FIG. 4 is a schematic illustration of the lane-level variable speed limit information distribution of the present invention;
fig. 5 is a flow chart of the particle swarm optimization algorithm of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. 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.
The variable speed limit optimization method based on intelligent network coupling and driver compliance comprises the following steps:
1) Designing a three-dimensional road scene of the expressway, constructing the three-dimensional road scene of the expressway based on a group driving simulation platform, and simultaneously importing the scene into micro-traffic simulation software;
2) Constructing an interactive simulation experiment platform based on the microscopic traffic simulation software and the group driving simulation platform;
3) After the experimental platform is started, the simulated traffic flow data are exported in real time, the current traffic flow risk level and the compliance of different drivers for controlling the simulated vehicles are calculated by using a traffic flow risk detection algorithm, and a preset lane-level variable speed limit control instruction is issued by using a lane-level variable speed limit issuing system in a road scene according to a preset lane-level variable speed limit control strategy;
4) Recruiting a plurality of drivers to carry out virtual simulation group driving by utilizing a driving simulator to wear VR glasses, controlling the running states of simulated vehicles by the drivers, loading the drivers in a unified scene for interaction in the driving process aiming at large-scale group drivers and drivers in different driving styles under different traffic states and scenes, and loading background vehicles generated based on parallel simulation according to different traffic states; acquiring basic information of a driver by using a questionnaire investigation mode, and simultaneously acquiring experimental road section road environment information data, vehicle running state data and traffic state data;
5) Calculating the compliance degree of each driver in the simulated driving, wherein the compliance degree is defined as:
wherein the ratio is i (t) is the compliance of the vehicle at time i,for i speed of the vehicle 250m upstream of the location receiving the variable speed limit information,/i>For i vehicle speed 200m downstream of the location receiving variable speed limit information, VSL i And (t) is a variable speed limit value issued to the i vehicle at the moment t.
Wherein whenWhen the compliance was considered to be 100%;
6) The compliance degree of a background vehicle generated based on simulation in a micro traffic simulation platform to a variable speed limit instruction is adjusted, one part of the compliance degree can be selected to be set to be 100% as an automatic driving vehicle, and the other part of the compliance degree is set to be different compliance degrees according to requirements;
7) The driver and the vehicle are defined as high, medium and low compliance with 100% compliance and 80% compliance as limits; wherein compliance level is defined as:
8) Dividing drivers and vehicles into three data sets according to compliance grades, namely a high compliance data set, a medium compliance data set and a low compliance data set; and carrying out cluster analysis on parameter matrixes of vehicles with different compliance data sets in each time period by adopting a K-means clustering algorithm, wherein each column in the matrix is a time sequence of one parameter, and the parameter matrixes comprise the following parameters: (1) a road environment information data matrix: black/daytime, weather, illumination, current road alignment, gradient, current lane speed limit and distance variable speed limit plate distance; (2) a driver information data matrix: the sex of the driver, the driving age of the driver and the age of the driver; (3) a vehicle operating state data matrix: time, vehicle center longitudinal position, vehicle center lateral position, vehicle longitudinal speed, vehicle longitudinal acceleration, vehicle lateral speed, vehicle lateral acceleration, lane offset, steering wheel angle, and vehicle speed product; (4) a traffic state data matrix: the longitudinal speed of the front vehicle of the lane, the time interval of the vehicle head, the distance between the front vehicles of the adjacent lanes, the longitudinal speed of the front vehicles of the adjacent lanes, the distance between the front vehicles and the rear vehicles of the lane, the traffic saturation, the average speed of the current road section and the current traffic flow; the Euclidean distance is used as an index for measuring the similarity of the data points, the distance between each data point in the sample set and the initial particle is calculated, and the closest principle is adopted to distribute the data points to the closest particles, so that the similar data can be classified into a cluster; dividing the parameter matrix data set into 3 clusters, namely, exciting, resisting and taking care; the drivers are finally classified into 9 categories: high aggressive compliance, high careless compliance, high careful compliance, aggressive compliance, medium-care compliance, small careful compliance, low aggressive compliance, low careless compliance, low careful compliance, constructing a driver personality style matrix;
9) Taking the style of the driver as a dependent variable and a parameter matrix as an independent variable, and training a deep learning model based on SAEs;
training a SAEs-based deep learning model, taking a training set as input, the first layer being trained as an automatic encoder; after the first hidden layer is obtained, the output of the mth hidden layer is used as the input of the (m+1) th hidden layer; the model structure consists of SAEs for extracting short-time traffic flow characteristics and a logistic regression layer for supervised parameter matrix prediction; training a deep network by adopting a Back Propagation (BP) algorithm based on a gradient optimization technology, wherein a greedy layered unsupervised learning algorithm shows advantages due to training each layer of parameters in the deep network from bottom to top in sequence; after the pre-training stage is completed, parameters of the prediction model are adjusted from top to bottom by means of the BP neural network, and finally a driving characteristic type judging model based on a parameter matrix is obtained;
10 The variable speed limit decision module is used for optimizing speed limit instructions for drivers of different driving styles under different roads and traffic scenes respectively, searching for a proper variable speed limit control strategy with higher compliance for the driving styles, using the comprehensive risk level based on individual vehicle safety and traffic flow overall safety as an actual rewarding value, using the driver compliance as a correction factor, establishing lane-level variable speed limit control schemes taking the driver compliance into consideration under the intelligent networking environment and the man-machine mixed traffic environment, forming a control scheme strategy library, and the variable speed limit information release module is used for releasing speed limit information for road users by using RSU (reactive power unit), so as to provide personalized speed limit guidance for different drivers under complex traffic environment.
Preferably, in the step 2), the running state of the micro traffic flow is controlled by the micro traffic simulation software, the simulated vehicles controlled by the driver are generated by group driving simulation software, and the information of the simulated vehicles is synchronously updated in real time in the micro traffic simulation software and the group driving simulation software, wherein the group driving simulation vehicles can interact with each other, and meanwhile, the group driving simulation vehicles can interact with the simulated vehicles generated by the micro traffic simulation software.
According to the variable speed limit optimization method based on intelligent network coupling and driver compliance, in the step 3), RSUs distributed on the road side are information instruction issuing devices of the variable speed limit system, the information instruction issuing devices are distributed at intervals of 500m, different variable speed limit information is issued for different lanes by each instruction, namely lane-level speed limit instructions on different lanes and upstream and downstream can have different speed limit values, and the speed limit values of the variable speed limit marks are adjusted through real-time change, so that the running state of road traffic flow is changed in real time.
Preferably, in the step 6), an online simulation module is implanted in the micro-traffic simulation platform, wherein the automatic driving vehicle adopts an IDM (integrated digital matrix) following and lane changing model to simulate the vehicle position in real time, the intelligent network vehicle utilizes a Q-learning reinforcement learning algorithm to simulate the vehicle position in real time, the manual driving vehicle utilizes a Gipps following and lane changing model to simulate the vehicle position in real time, and the three models are loaded into the online simulation module according to different proportions and speed limiting compliance.
Preferably, the parameter value rule of the parameter matrix is as follows: the extraction positions are 200m, 800m, 1400m and 2000m upstream of the position where the vehicle receives the variable speed limit instruction, and 400m and 1000m downstream of the position where the vehicle receives the variable speed limit instruction, the extraction time granularity is 5min, and the extraction time is 5min, 10min, 15min, 20min, 25min and 30min before the vehicle receives the variable speed limit instruction.
Preferably, aiming at driving styles of different drivers, an improved particle swarm optimization algorithm is adopted to find an optimal variable speed limit control strategy, comprehensive risks are used as actual rewarding values, individual vehicle compliance is used as a correction factor, and finally, a variable speed limit control scheme suitable for each driving style is established to form a control scheme strategy library. The method comprises the following steps:
1) Judging the driving style of a driver by using a driving style judging method, initializing the speed limit after the judgment is finished, and initializing the variable speed limit speed change gradient and the speed change period;
2) Initializing and setting a variable speed limiting speed change threshold;
3) Controlling the variable speed limit change according to the driving risk, if Q is more than 0.30, changing the variable speed limit, calculating the driving risk at the end of the control period, if Q is still more than 0.30, traversing, adjusting and optimizing the variable speed limit plate speed change gradient and speed change period of the current variable speed limit strategy until the driving risk benefit E (mu) of the combined control strategy p ) And (3) highest, incorporating the control strategy into a variable speed limit strategy library, thereby obtaining targeted variable speed limit control strategies aiming at different driving styles.
Preferably, in the step 9), the parameters in the parameter matrix are used as input variables and data standardization processing is performed, the driving style type obtained in the step 8) is used for establishing an original data set, and a random forest method is used for performing dimension reduction processing on the data dimension to obtain a training data set.
Preferably, in the step 10), the comprehensive risk level algorithm based on the individual vehicle safety and the overall traffic flow safety comprises the following steps:
1) Establishing individual vehicle safety risk level judgment and evaluation standards, wherein dangerous behaviors are defined as follows:
TTC i i the collision time of the vehicle relative to the front vehicle at the time t, because the vehicle position acquired by data is the position of the vehicle head, X is the position of the vehicle head i (t) is the position of the head of the vehicle at the moment i, X h (t) is the position of the front h head of the front car at the moment i, l h Is the length of the body of the h car, V i (t) is the instantaneous speed of the i-car at time t, V h (t) is the instantaneous speed of the vehicle at time t;
the individual vehicle security risk level definition method comprises the following steps of:
wherein,
wherein,representing the collision risk of i vehicles relative to the preceding vehicle at the moment t, TTC p TTC representing all individual vehicles at that moment i Is a mean value of (c).
2) By using a K-means clustering methodThe clusters are three categories, high risk level, medium risk level and low risk level, respectively. The collision Risk level of i vehicles relative to the preceding vehicle at time t is represented by Risk (i), namely:
3) And constructing a traffic flow overall safety risk level judgment and evaluation standard, and counting the proportion of vehicles with high, medium and low risk levels in the road section at the moment t. Wherein the overall security risk level of the traffic flow is defined as:
4) Building a comprehensive risk algorithm based on individual vehicle safety and overall traffic flow safety:
wherein α+β=1
5) Aiming at drivers with different driving styles, the driver adoptsSearching for an optimal variable speed limit control strategy by using an improved particle swarm optimization algorithm, and utilizing the comprehensive risk Q as an actual rewarding value to enable individual vehicles to follow the rate i And (t) taking the variable speed limit control scheme as a correction factor, and finally establishing a variable speed limit control scheme suitable for each driving style to form a control scheme strategy library.
Wherein the individual vehicle security risk benefits of the combined control strategy are defined as:
wherein E (mu) p ) Representing the risk benefit under the control strategy,the risk is integrated before the control strategy is implemented,the risk is integrated after implementation for the control strategy.
Compliance combining loss for a combined control strategy is defined as:
wherein the ratio is i (t) b Individual vehicle compliance, compactness, prior to implementation of the control strategy i (t) a Individual vehicle compliance after implementation of the control strategy.
And judging the driving style of the driver by using the driving style judging method, initializing the speed limit after the judgment, and initializing the variable speed limit speed change gradient and the speed change period.
And initializing and setting a variable speed limiting speed change threshold.
And initializing and setting the variable speed limiting guide speed of each vehicle, wherein the initial speed limit is 100km/h.
And initializing and setting a speed change gradient, wherein the speed change gradient is 5km/h.
And initializing and setting a speed change period which is 1min.
And acquiring the driving risk according to the real-time traffic track of the road section, and changing the variable speed limit by adopting initialization setting if the risk level is more than 0.30. Calculating the benefit E (mu) of the combined control strategy at the future time of 5 minutes after the speed limit is changed p ) And loss Risk of combined control strategy p 。
Benefit E (μ) of combining control strategies p ) And loss Risk of combined control strategy p As a target optimization function, as input into the particle swarm optimization algorithm model.
Initializing a particle group, setting the group size as N, and initializing the speed and the position of the particles.
And setting a fitness function, and calculating the fitness of each particle through the fitness function. And respectively storing the optimal positions and the optimal fitness of the particles, wherein the optimal positions and the optimal fitness of the population are respectively stored.
And comparing the fitness of each particle with the fitness Pi of the historical optimal point, and if the fitness of the particle is better than the historical optimal point, taking the current position as the historical optimal position of the particle, and meanwhile, the fitness of the particle also becomes the historical optimal fitness so as to find the individual optimal.
The update of the position and velocity of the particles is based on:
wherein the acceleration factor c 1 Regulating maximum step length of individual optimum position flying, c 2 And adjusting the maximum step length of the global optimal position flight. If the acceleration factor is too large, particles can fly out of the target area rapidly, and if the acceleration factor is too small, particles can be caused to fly out of the target area rapidlyThe particles reach the target area too slowly. Suitable c 1 And c 2 The value can promote the convergence rate of the particle swarm algorithm and avoid sinking into local optimum. c 1 And c 2 Value of [0,4 ]]In between, the present embodiment is set to 2.
Representing the influence of the speed left by the previous generation of the particles on the flying behavior of the particles, the influence can be regarded as the inertia of the particles; second part->The method is that the individual cognition is realized, the particles find out the optimal solution direction found by the individual cognition, and the learning of past experience of the particles is represented; third part->The method belongs to social cognition, and the particles refer to learning to other particles in the population, and represent acceptance of the particles to the overall searching condition of the population.
And comparing the adaptation value corresponding to the historical optimal position of each particle with the population optimal adaptation degree Pg, and if the adaptation value is better, updating the historical optimal position and the historical adaptation of the population so as to find global optimal.
If the condition is met, the vehicle is stopped, otherwise, the vehicle risk is acquired according to the real-time traffic track of the road section, and if the risk level is more than 0.30, the variable speed limit is changed by adopting the initialization setting. Calculating the benefit E (mu) of the combined control strategy at the future time of 5 minutes after the speed limit is changed p ) And loss Risk of combined control strategy p The operation continues here and again. Until the optimal speed limit value V under the condition is obtained t 。
The variable speed limit information release module utilizes RSUs arranged on the road sides to release speed limit information to road users.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. The variable speed limit optimization method based on intelligent network coupling and driver compliance is characterized by comprising the following steps of:
1) Designing a three-dimensional road scene of the expressway, constructing the three-dimensional road scene of the expressway based on a group driving simulation platform, and simultaneously importing the scene into micro-traffic simulation software;
2) Constructing an interactive simulation experiment platform based on the microscopic traffic simulation software and the group driving simulation platform;
3) After the experimental platform is started, the simulated traffic flow data are exported in real time, the current traffic flow risk level and the compliance of different drivers for controlling the simulated vehicles are calculated by using a traffic flow risk detection algorithm, and a preset lane-level variable speed limit control instruction is issued by using a lane-level variable speed limit issuing system in a road scene according to a preset lane-level variable speed limit control strategy;
4) Recruiting a plurality of drivers to carry out virtual simulation group driving by utilizing a driving simulator to wear VR glasses, controlling the running states of simulated vehicles by the drivers, loading the drivers in a unified scene for interaction in the driving process aiming at large-scale group drivers and drivers in different driving styles under different traffic states and scenes, and loading background vehicles generated based on parallel simulation according to different traffic states; acquiring basic information of a driver by using a questionnaire investigation mode, and simultaneously acquiring experimental road section road environment information data, vehicle running state data and traffic state data;
5) Calculating the compliance degree of each driver in the simulated driving, wherein the compliance degree is defined as:
wherein the ratio is i (t) is the compliance of the vehicle at time i,for i speed of the vehicle 250m upstream of the location receiving the variable speed limit information,/i>For i vehicle speed 200m downstream of the location receiving variable speed limit information, VSL i (t) is a variable speed limit value issued to the i vehicle at the moment t;
wherein whenWhen the compliance was considered to be 100%;
6) The compliance degree of a background vehicle generated based on simulation in a micro traffic simulation platform to a variable speed limit instruction is adjusted, one part of the compliance degree can be selected to be set to be 100% as an automatic driving vehicle, and the other part of the compliance degree is set to be different compliance degrees according to requirements;
7) The driver and the vehicle are defined as high, medium and low compliance with 100% compliance and 80% compliance as limits; wherein compliance level is defined as:
8) Dividing drivers and vehicles into three data sets according to compliance grades, namely a high compliance data set, a medium compliance data set and a low compliance data set; and carrying out cluster analysis on parameter matrixes of vehicles with different compliance data sets in each time period by adopting a K-means clustering algorithm, wherein each column in the matrix is a time sequence of one parameter, and the parameter matrixes comprise the following parameters: (1) a road environment information data matrix: black/daytime, weather, illumination, current road alignment, gradient, current lane speed limit and distance variable speed limit plate distance; (2) a driver information data matrix: the sex of the driver, the driving age of the driver and the age of the driver; (3) a vehicle operating state data matrix: time, vehicle center longitudinal position, vehicle center lateral position, vehicle longitudinal speed, vehicle longitudinal acceleration, vehicle lateral speed, vehicle lateral acceleration, lane offset, steering wheel angle, and vehicle speed product; (4) a traffic state data matrix: the longitudinal speed of the front vehicle of the lane, the time interval of the vehicle head, the distance between the front vehicles of the adjacent lanes, the longitudinal speed of the front vehicles of the adjacent lanes, the distance between the front vehicles and the rear vehicles of the lane, the traffic saturation, the average speed of the current road section and the current traffic flow; the Euclidean distance is used as an index for measuring the similarity of the data points, the distance between each data point in the sample set and the initial particle is calculated, and the closest principle is adopted to distribute the data points to the closest particles, so that the similar data can be classified into a cluster; dividing the parameter matrix data set into 3 clusters, namely, exciting, resisting and taking care; the drivers are finally classified into 9 categories: high aggressive compliance, high careless compliance, high careful compliance, aggressive compliance, medium-care compliance, small careful compliance, low aggressive compliance, low careless compliance, low careful compliance, constructing a driver personality style matrix;
9) Taking the style of the driver as a dependent variable and a parameter matrix as an independent variable, and training a deep learning model based on SAEs;
training a SAEs-based deep learning model, taking a training set as input, the first layer being trained as an automatic encoder; after the first hidden layer is obtained, the output of the mth hidden layer is used as the input of the (m+1) th hidden layer; the model structure consists of SAEs for extracting short-time traffic flow characteristics and a logistic regression layer for supervised parameter matrix prediction; training a deep network by adopting a Back Propagation (BP) algorithm based on a gradient optimization technology, wherein a greedy layered unsupervised learning algorithm shows advantages due to training each layer of parameters in the deep network from bottom to top in sequence; after the pre-training stage is completed, parameters of the prediction model are adjusted from top to bottom by means of the BP neural network, and finally a driving characteristic type judging model based on a parameter matrix is obtained;
10 The variable speed limit decision module is used for optimizing speed limit instructions for drivers of different driving styles under different roads and traffic scenes respectively, searching for a proper variable speed limit control strategy with higher compliance for the driving styles, using the comprehensive risk level based on individual vehicle safety and traffic flow overall safety as an actual rewarding value, using the driver compliance as a correction factor, establishing lane-level variable speed limit control schemes taking the driver compliance into consideration under the intelligent networking environment and the man-machine mixed traffic environment, forming a control scheme strategy library, and the variable speed limit information release module is used for releasing speed limit information for road users by using RSU (reactive power unit), so as to provide personalized speed limit guidance for different drivers under complex traffic environment.
2. The variable speed limit optimization method based on intelligent network coupling and driver compliance according to claim 1, wherein in the step 2), the running state of the micro traffic flow is controlled by the micro traffic simulation software, the simulated vehicles controlled by the driver are generated by the group driving simulation software, and the information of the simulated vehicles is synchronously updated in real time in the micro traffic simulation software and the group driving simulation software, wherein the group driving simulation vehicles can interact with each other, and meanwhile the group driving simulation vehicles can interact with the simulated vehicles generated by the micro traffic simulation software.
3. The variable speed limit optimizing method based on intelligent network coupling and driver compliance according to claim 1, wherein in the step 3), RSUs arranged at the road side are information instruction issuing devices of the variable speed limit system, and are arranged at intervals of 500m, each instruction issues different variable speed limit information for different lanes, namely lane-level speed limit instructions on and downstream of different lanes can have different speed limit values, and the speed limit values of the variable speed limit signs are adjusted by changing in real time, so that the running state of road traffic flow is changed in real time.
4. The variable speed limit optimizing method based on intelligent network vehicle and driver compliance according to claim 1, wherein in the step 6), an online simulation module is implanted in a micro-traffic simulation platform, wherein the automatic driving vehicle adopts an IDM following and lane changing model to simulate the vehicle position in real time, the intelligent network vehicle utilizes a Q-learning reinforcement learning algorithm to simulate the vehicle position in real time, the manual driving vehicle utilizes a Gipps following and lane changing model to simulate the vehicle position in real time, and three models are loaded into the online simulation module according to different proportions and speed limit compliance.
5. The variable speed limit optimization method based on intelligent network coupling and driver compliance according to claim 1, wherein the parameter value rule of the parameter matrix is: the extraction positions are 200m, 800m, 1400m and 2000m upstream of the position where the vehicle receives the variable speed limit instruction, and 400m and 1000m downstream of the position where the vehicle receives the variable speed limit instruction, the extraction time granularity is 5min, and the extraction time is 5min, 10min, 15min, 20min, 25min and 30min before the vehicle receives the variable speed limit instruction.
6. The variable speed limit optimization method based on intelligent network coupling and driver compliance according to claim 1, wherein in the step 9), an improved particle swarm optimization algorithm is adopted to find an optimal variable speed limit control strategy according to driving styles of different drivers, comprehensive risks are used as actual rewarding values, individual vehicle compliance is used as a correction factor, and finally a variable speed limit control scheme suitable for each driving style is established to form a control scheme strategy library; the method comprises the following steps:
1) Judging the driving style of a driver by using a driving style judging method, initializing the speed limit after the judgment is finished, and initializing the variable speed limit speed change gradient and the speed change period;
2) Initializing and setting a variable speed limiting speed change threshold;
3) Controlling the variable speed limit change according to the driving risk, if Q is more than 0.30, changing the variable speed limit, and calculating the control period nodeIf Q is still more than 0.30, traversing, adjusting and optimizing the speed change gradient and the speed change period of the variable speed limiting plate of the current variable speed limiting strategy until the running risk benefit E (mu) of the combined control strategy p ) And (3) highest, incorporating the control strategy into a variable speed limit strategy library, thereby obtaining targeted variable speed limit control strategies aiming at different driving styles.
7. The variable speed limit optimization method based on intelligent network coupling and driver compliance according to claim 1, wherein in the step 9), parameters in the parameter matrix are used as input variables and data standardization processing is performed, an original data set is established by using the driving style type obtained in the step 8), and a random forest method is used for performing dimensionality reduction processing on the data dimension to obtain a training data set.
8. The variable speed limit optimization method based on intelligent network coupling and driver compliance according to claim 1, wherein in said step 10), said integrated risk level algorithm based on individual vehicle safety and traffic flow overall safety comprises the steps of:
1) Establishing individual vehicle safety risk level judgment and evaluation standards, wherein dangerous behaviors are defined as follows:
TTC i i the collision time of the vehicle relative to the front vehicle at the time t, because the vehicle position acquired by data is the position of the vehicle head, X is the position of the vehicle head i (t) is the position of the head of the vehicle at the moment i, X h (t) is the position of the front h head of the front car at the moment i, l h Is the length of the body of the h car, V i (t) is the instantaneous speed of the i-car at time t, V h (t) is the instantaneous speed of the vehicle at time t;
the individual vehicle security risk level definition method comprises the following steps of:
wherein,
wherein,representing the collision risk of i vehicles relative to the preceding vehicle at the moment t, TTC p TTC representing all individual vehicles at that moment i Is the average value of (2);
2) By using a K-means clustering methodClustering into three categories, namely a high risk level, a medium risk level and a low risk level; the collision Risk level of i vehicles relative to the preceding vehicle at time t is represented by Risk (i), namely:
3) Constructing a traffic flow overall safety risk level judgment and evaluation standard, and counting the proportion of vehicles with high, medium and low risk levels in a road section at the moment t; wherein the overall security risk level of the traffic flow is defined as:
4) Building a comprehensive risk algorithm based on individual vehicle safety and overall traffic flow safety:
wherein α+β=1
5) Aiming at drivers with different driving styles, an improved particle swarm optimization algorithm is adopted to find an optimal variable speed limit control strategy, the comprehensive risk Q is used as an actual rewarding value, and the individual vehicle compliance rate is adopted i (t) as a correction factor, finally establishing a variable speed limit control scheme suitable for each driving style to form a control scheme strategy library;
wherein the individual vehicle security risk benefits of the combined control strategy are defined as:
wherein E (mu) p ) Representing the risk benefit under the control strategy,comprehensive risk before implementation of the control strategy, +.>Comprehensive risk after implementation for the control strategy;
compliance combining loss for a combined control strategy is defined as:
wherein the ratio is i (t) b Individual vehicle compliance, compactness, prior to implementation of the control strategy i (t) a Individual vehicle compliance after implementation of the control strategy;
6) The variable speed limit information release module utilizes RSUs arranged on the road sides to release speed limit information to road users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310360227.6A CN116432448B (en) | 2023-04-06 | 2023-04-06 | Variable speed limit optimization method based on intelligent network coupling and driver compliance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310360227.6A CN116432448B (en) | 2023-04-06 | 2023-04-06 | Variable speed limit optimization method based on intelligent network coupling and driver compliance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116432448A CN116432448A (en) | 2023-07-14 |
| CN116432448B true CN116432448B (en) | 2024-02-23 |
Family
ID=87084910
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310360227.6A Active CN116432448B (en) | 2023-04-06 | 2023-04-06 | Variable speed limit optimization method based on intelligent network coupling and driver compliance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116432448B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116884220A (en) * | 2023-08-01 | 2023-10-13 | 同济大学 | Global variable speed limit compliance prediction method based on track data |
| CN116959254A (en) * | 2023-08-01 | 2023-10-27 | 同济大学 | Lane-level variable speed-limiting individual compliance prediction method based on time sequence track data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105975705A (en) * | 2016-05-13 | 2016-09-28 | 东南大学 | Cell transmission simulation model improved method aiming at variable speed limit control |
| CN111754777A (en) * | 2020-07-10 | 2020-10-09 | 清华大学 | Microscopic traffic simulation method for unmanned and manned mixed traffic flow |
| CN114067561A (en) * | 2021-10-25 | 2022-02-18 | 东南大学 | Virtual reality testing method for urban expressway vehicle-road cooperative active management and control system |
| CN114118795A (en) * | 2021-11-26 | 2022-03-01 | 同济大学 | Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway |
| CN114627647A (en) * | 2022-03-16 | 2022-06-14 | 重庆大学 | Hybrid traffic flow optimization control method based on combination of variable speed limit and lane change |
| CN115206103A (en) * | 2022-07-18 | 2022-10-18 | 山西省智慧交通研究院有限公司 | Variable speed-limiting control system based on parallel simulation system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220197280A1 (en) * | 2020-12-22 | 2022-06-23 | Uatc, Llc | Systems and Methods for Error Sourcing in Autonomous Vehicle Simulation |
-
2023
- 2023-04-06 CN CN202310360227.6A patent/CN116432448B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105975705A (en) * | 2016-05-13 | 2016-09-28 | 东南大学 | Cell transmission simulation model improved method aiming at variable speed limit control |
| CN111754777A (en) * | 2020-07-10 | 2020-10-09 | 清华大学 | Microscopic traffic simulation method for unmanned and manned mixed traffic flow |
| CN114067561A (en) * | 2021-10-25 | 2022-02-18 | 东南大学 | Virtual reality testing method for urban expressway vehicle-road cooperative active management and control system |
| CN114118795A (en) * | 2021-11-26 | 2022-03-01 | 同济大学 | Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway |
| CN114627647A (en) * | 2022-03-16 | 2022-06-14 | 重庆大学 | Hybrid traffic flow optimization control method based on combination of variable speed limit and lane change |
| CN115206103A (en) * | 2022-07-18 | 2022-10-18 | 山西省智慧交通研究院有限公司 | Variable speed-limiting control system based on parallel simulation system |
Non-Patent Citations (4)
| Title |
|---|
| Richard Atta Boateng,Michael D. Fontaine,and etc,..Driver Response to Variable Speed Limits on I-66 in Northern Virginia.《journal of transportation engineering》.2019,全文. * |
| 吴宏涛 ; 刘力源 ; 孟颖 ; 荣亚鹏 ; 李路凯.动态多特征融合的道路遗洒物威胁度分析方法.《计算机科学》.2020,全文. * |
| 基于运营安全的超长地下道路车速控制方法研究;杨旻皓;中国市政工程(05);全文 * |
| 多形态几何约束的道路障碍物检测新方法;吴宏涛;《山西交通科技》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116432448A (en) | 2023-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116432448B (en) | Variable speed limit optimization method based on intelligent network coupling and driver compliance | |
| Du et al. | Comfortable and energy-efficient speed control of autonomous vehicles on rough pavements using deep reinforcement learning | |
| Kuefler et al. | Imitating driver behavior with generative adversarial networks | |
| CN107229973A (en) | The generation method and device of a kind of tactful network model for Vehicular automatic driving | |
| CN110956851B (en) | Intelligent networking automobile cooperative scheduling lane changing method | |
| CN107813820A (en) | A kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver | |
| Jayawardana et al. | Learning eco-driving strategies at signalized intersections | |
| CN108711285B (en) | Hybrid traffic simulation method based on road intersection | |
| CN116822125A (en) | Personalized interactive lane-level variable speed limit optimization method based on virtual reality | |
| CN110619340B (en) | Method for generating lane change rule of automatic driving automobile | |
| CN114580302A (en) | Decision planning method for automatic driving automobile based on maximum entropy reinforcement learning | |
| CN114030485A (en) | Automatic driving automobile man lane change decision planning method considering attachment coefficient | |
| CN114368387B (en) | Attention mechanism-based driver intention recognition and vehicle track prediction method | |
| CN115257789A (en) | Decision-making method for side anti-collision driving of commercial vehicle in urban low-speed environment | |
| CN114802306A (en) | Intelligent vehicle integrated decision-making system based on man-machine co-driving concept | |
| CN113120003B (en) | Unmanned vehicle motion behavior decision method | |
| CN111488674B (en) | Plane intersection vehicle running track simulation method | |
| Zhang et al. | Driving style classification for vehicle-following with unlabeled naturalistic driving data | |
| CN114148349B (en) | Vehicle personalized following control method based on generation of countermeasure imitation study | |
| CN113276860B (en) | Vehicle control method, device, electronic device, and storage medium | |
| Chen et al. | Platoon separation strategy optimization method based on deep cognition of a driver’s behavior at signalized intersections | |
| CN107657106A (en) | traffic flow simulation method based on genetic algorithm | |
| Hua | A new car-following model considering recurrent neural network | |
| Wang et al. | Driver modeling based on vehicular sensing data | |
| CN116767218B (en) | Forced lane change decision method for unmanned vehicle, computer equipment and medium |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhang Linliang Inventor after: Zhao Yuhong Inventor after: Wang Yubiao Inventor after: Niu Bingqing Inventor after: He Lingling Inventor after: Yan Lianshan Inventor after: Wu Hongtao Inventor after: Liang Lei Inventor after: Wang Yanchao Inventor after: Song Hao Inventor after: Fu Yuqiang Inventor after: Xu Xin Inventor after: Li Xianda Inventor before: Zhang Linliang Inventor before: Wang Yubiao Inventor before: Niu Bingqing Inventor before: He Lingling Inventor before: Wu Hongtao Inventor before: Liang Lei Inventor before: Wang Jianchao Inventor before: Song Wu Inventor before: Fu Yuqiang Inventor before: Xu Xin Inventor before: Li Xianda Inventor before: Zhao Yuhong |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |