CN116959244A - Vehicle network channel congestion control method and system based on regional danger - Google Patents
Vehicle network channel congestion control method and system based on regional danger Download PDFInfo
- Publication number
- CN116959244A CN116959244A CN202310770589.2A CN202310770589A CN116959244A CN 116959244 A CN116959244 A CN 116959244A CN 202310770589 A CN202310770589 A CN 202310770589A CN 116959244 A CN116959244 A CN 116959244A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- risk
- representing
- road
- grid
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 81
- 230000006870 function Effects 0.000 claims abstract description 65
- 230000007613 environmental effect Effects 0.000 claims abstract description 11
- 238000013528 artificial neural network Methods 0.000 claims abstract description 8
- 230000015654 memory Effects 0.000 claims abstract description 5
- 230000005540 biological transmission Effects 0.000 claims description 51
- 239000002245 particle Substances 0.000 claims description 30
- 238000012502 risk assessment Methods 0.000 claims description 18
- 238000004891 communication Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 230000002776 aggregation Effects 0.000 claims description 6
- 238000004220 aggregation Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000035945 sensitivity Effects 0.000 claims description 4
- 238000013459 approach Methods 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 230000036961 partial effect Effects 0.000 claims description 3
- 238000006116 polymerization reaction Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 abstract description 6
- 230000006399 behavior Effects 0.000 description 9
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000007787 long-term memory Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 101001093748 Homo sapiens Phosphatidylinositol N-acetylglucosaminyltransferase subunit P Proteins 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000013058 risk prediction model Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
- G08G1/0141—Measuring and analyzing of parameters relative to traffic conditions for specific applications for traffic information dissemination
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/0289—Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
-
- 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
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention belongs to the technical field of intelligent traffic systems, and particularly discloses a vehicle-mounted network channel congestion control method and system based on regional danger, wherein the system evaluates the running risk of a single vehicle by using a combined fuzzy reasoning method based on driver factors, vehicle factors and environmental factors which influence running safety; estimating the driving risk of the road area, and using a convolution long-short-term memory neural network to predict the driving risk of the road area in a short time; constructing a non-cooperative game model, and incorporating the driving risk and the safety information of the road area where the vehicle is located into the utility function of the non-cooperative game model; and solving the utility function to obtain an equalization strategy for controlling the channel congestion of the vehicle-mounted network. By adopting the technical scheme, the influence of interaction of the driver factors, the vehicle factors and the environmental factors on the running risk can be comprehensively and effectively estimated, and a more comprehensive and accurate running risk estimation result is obtained.
Description
Technical Field
The invention belongs to the technical field of intelligent traffic systems, and relates to a vehicle network channel congestion control method and system based on regional danger.
Background
With the continuous development of economy and the update progress of the automobile industry, the concept of an intelligent transportation system (Intelligent Traffic System, ITS) has been developed, and as one of the key technologies of the intelligent transportation system in the future, C-V2X (Cellular-Vehicle to everything, C-V2X) communication can be used for the message interaction between a vehicle and the surrounding environment, which covers the fields of security, infotainment, automatic driving and the like. Admittedly, the resource allocation strategy used by C-V2X has a relatively low probability of congestion compared to dedicated short range communication technologies (Dedicated Short Range Communications, DSRC). However, as the contention for available resources increases, C-V2X also suffers from reduced performance in dense network scenarios.
Compared with the traditional wireless network, the internet of vehicles needs to meet the communication requirement between vehicles in a high-speed mobile environment. However, communication resources in the internet of vehicles have obvious limitations, so that reasonable utilization of the resources is particularly critical. In order to realize effective allocation of resources, the transmission frequency of the safety message can be adjusted according to the running risk of the vehicle. For high-risk vehicles, increasing the transmission frequency is helpful to ensure real-time transmission of key safety information, thereby improving road safety. Correspondingly, for the common vehicles with lower risks, the reduction of the transmission frequency can save channel resources and relieve network congestion.
The current running risk assessment method is mainly divided into two types, namely one type of running risk assessment based on a single factor and the other type of running risk assessment based on multiple factors. The vehicle risk assessment based on a single factor is mainly carried out on a certain factor in drivers, vehicles and environments, and is helpful for identifying and preventing potential risks to a certain extent. However, with the development of internet of vehicles technology and the continuous progress of traffic data acquisition technology in the field of "driver-vehicle-environment", researchers are gradually turning to a multi-factor-based driving risk assessment method. The method comprehensively considers various factors, and can comprehensively and effectively evaluate the influence of interaction of driver factors, vehicle factors and environmental factors on the driving risk. However, the prior art does not dynamically formulate a strategy for congestion control according to the running state of the vehicle, which may cause traffic safety hidden trouble in a congestion environment. Nor do the various factors affecting the risk of driving consider that are mostly characterized by non-linearities, deformations and uncertainties, the interrelationship between these factors and the difficult-to-quantify nature make direct analysis difficult.
Disclosure of Invention
The invention aims to provide a vehicle network channel congestion control method and system based on regional risk, which can comprehensively and effectively evaluate the influence of interaction of driver factors, vehicle factors and environmental factors on the running risk and obtain a more comprehensive and accurate running risk evaluation result.
In order to achieve the above purpose, the basic scheme of the invention is as follows: a vehicle network channel congestion control method based on regional danger degree comprises the following steps:
based on driver factors, vehicle factors and environmental factors affecting driving safety, evaluating the driving risk of a single vehicle by using a combined fuzzy reasoning method;
estimating the driving risk of the road area, and using a convolution long-short-term memory neural network to predict the driving risk of the road area in a short time;
constructing a non-cooperative game model, and incorporating the driving risk and the safety information of the road area where the vehicle is located into the utility function of the non-cooperative game model;
and solving the utility function to obtain the equalization strategy of the transmission frequency control of the vehicle-mounted network channel congestion control.
The working principle and the beneficial effects of the basic scheme are as follows: according to the technical scheme, the driving risk of the vehicle is firstly estimated, the driving risk of the road area is estimated according to the estimated driving risk degree, the driving risk of the short-time road area is predicted, a non-cooperative game model is built according to the driving risk degree and solved, the balance strategy of the vehicle is obtained, and the vehicle self-adaptively adjusts the safety message transmission frequency of the vehicle based on the obtained balance strategy. The vehicle risk assessment method based on the combined fuzzy reasoning can consider factors influencing the vehicle safety as much as possible, so that a more comprehensive and accurate vehicle risk assessment result is obtained. In the method, the transmission frequency of the driving risk and the safety information is considered in the utility function, and the optimal transmission frequency is set for vehicles with different communication demands so as to relieve channel congestion and save communication resources.
Further, the method for evaluating the running risk of the single vehicle is as follows:
a traditional fuzzy reasoning model is selected for fuzzy reasoning, a driving risk assessment model influenced by a driver factor belongs to a double-input multi-rule reasoning model, and the processing procedure of the reasoning model is as follows:
wherein ,Xn Is the input 1, Y under the nth rule n Is input 2, Z under the nth rule n Is the reasoning result obtained by the nth rule, X * Is the 'U' aggregation of input parameter 1 in all rules, Y * Is the 'U' aggregation of input parameter 2 in all rules;
each rule is first processed:
and then the results obtained by each reasoningPerforming U-shaped polymerization to obtain a final fuzzy result Z * The method comprises the following steps:
deblurring by gravity center method to obtain Z * Assuming that the driving risk degree accuracy value affected by the obtained driver factor is a, a is:
wherein N represents the number of points in the domain, A i Mu as the i-th point Z* (A i ) Representing the ambiguity solution Z * Membership functions of (2);
obtaining a driving risk degree accurate value D influenced by driver factors 1 Similarly, the driving risk degree D of the vehicle factors and the environment factors is calculated respectively 2 、D 3 ；
At D 1 ,D 2 ,D 3 On the basis of (1) carrying out fuzzy reasoning and outputting a corresponding total driving risk degree D 0 Triangle membership functions are used:
controlling the sensitivity degree of the triangle membership function to input change by adjusting parameters a, b and c; x is an independent variable; a, b, c represent the vertex positions of triangles respectively;
and (3) reasoning by using a traditional fuzzy reasoning method, and obtaining the running risk degree of the vehicle after defuzzification.
And the factors influencing the driving safety are analyzed, and based on the factors, the driving risk is evaluated by using a combined fuzzy reasoning method, so that the operation is simple.
Further, the method for evaluating the driving risk of the road area comprises the following steps:
dividing a road into grids, carrying out road area driving risk assessment by taking the grids as units, and determining surrounding grids of the current road grid;
the communication range of C-V2X is 150m, but in an actual road scene, when the mesh approaches the boundary, the mesh range to be considered needs to be adjusted accordingly according to the actual situation, so for each road mesh (i, j), the surrounding mesh is defined as:
extending in the length direction from min (i-30, 0) to max (i+30, 199); a grid extending in the width direction from min (j-50, 0) to max (j+50, 3) region;
for each road grid, calculating a risk value through a weighted average value, and representing the influence of vehicles with different distances on the current road grid by adopting a Euclidean distance weighting method;
let the current vehicle's grid be i and the neighboring surrounding grids be j 1 ,j 2 ,…,j k The corresponding driving risk values are R1, R2, … and Rk; the running risk assessment result of the current grid i is:
R i ＝(ω 1 R 1 +ω 2 R 2 +…+ω n R n )/(ω 1 +ω 2 +…+ω n )
wherein ,ωi Representing the current grid i to the adjacent grid j 1 ,j 2 ,…,j k The inverse of the euclidean distance, i.e.:
...
where d (i, j) represents the Euclidean distance between grid i and grid j, namely:
wherein ,(xi ,y i ) Represents the coordinates of grid i, (x) j ,y j ) Representing coordinates of grid j;
obtaining the running risk degree of the road grid where the vehicle is located by using a matrix X t To represent the traffic risk degree distribution condition of the road area at the moment t:
wherein ,representing the driving risk degree of the ith row and jth column road grids at the moment t.
And carrying out running risk assessment to ensure that the running risk condition of the road area is accurately reflected, thereby providing guarantee for running safety.
Further, the specific method for predicting the driving risk in a short time in the road area is as follows:
predicting the driving risk degree of n times in the future by adopting the driving risk degree of m times in the past, which is expressed as follows:
X t,t+n ＝Γ(X t-m+1,t )
wherein ,Xt,t+n The travel risk from time t to time t+n is represented by Γ, the prediction method is represented by t, the time X is represented by t t-m+1,t The running risk degree from the time t-m+1 to the time t is shown.
The road traffic environment has the characteristics of diversity, uncertainty and the like, and the running risk at the current moment can not fully reflect the running risk at the next moment, so that the result is inaccurate when the self-adaptive adjustment parameters are used for congestion control. To overcome these problems, short-term prediction of road area driving risk is required to make up for the lack of real-time assessment.
Further, the method for constructing the non-cooperative game model comprises the following steps:
the congestion control model based on the travelling risk game is as follows:
G＝{V；r 1 ,r 2 ,…,r n ；u 1 ,u 2 ,…,u n }
where v= {1,2, …, n } represents the set of participants, i.e. vehicles, participating in the game, r= { R 1 ,r 2 ,…r n Policy set representing the frequency of transmissions of vehicles participating in a game, r i Representing the strategy of the vehicle i, the transmission frequency range of the vehicle i participating in the game is 0<r min ≤r i ≤r max The method comprises the steps of carrying out a first treatment on the surface of the The transmission frequency in C-V2X ranges from [1,50]I.e. 1.ltoreq.ri.ltoreq.50;
u i utility function representing vehicle i participating in gaming:
u i ＝(r i ,r -i ),
wherein ,r-i ＝(r 1 ,…,r i-1 ,r i+1 ,…,r n ) Representing the transmission frequency strategy of other participating game vehicles except the vehicle i, wherein i is a natural number;
considering that each participant needs to select an optimal strategy to achieve the maximization of own interests according to own interests and targets, the optimal transmission frequency of the vehicle i is determined by the maximization u i ＝(r i ,r -i ) The resulting problem is expressed as:
r i * ＝arg max u i (r i ,r -i )
r i ∈[r min ,r max ]
the final utility function is:
wherein ,ri Representing the transmission frequency of vehicle i, d i Representing the driving risk degree of the vehicle i, alpha i Representing utility factor, beta i Is a constant coefficient, and alpha i >>β i ，λ i In order to take the cost factor into account,and to thisThe strategy of his vehicle is related, N being the total number of vehicles.
And (3) carrying out self-adaptive adjustment parameters according to the driving risk degree of the road network where each vehicle is located so as to ensure driving safety, and establishing a non-cooperative game model suitable for the Internet of vehicles so as to describe the competition relationship and interaction between vehicles, thereby being convenient to use.
Further, the method for proving the existence of Nash equilibrium of the non-cooperative game model is as follows:
for any vehicle i, its strategy is discrete and is in interval 0<r min ≤r i ≤r max Internally, R is thus non-empty, closed and bounded;
utility function u for each vehicle i (r i ,r -i ) Is continuous on self R, for u i The first order bias derivative is calculated by:
and continuing to obtain a second-order partial derivative:
for any r in the interval i All have 0<r min ≤r i ≤r max, and therefore, the utility function of the vehicle is a continuous concave function in the strategy space, so that a Nash equilibrium solution exists in the non-cooperative game model;
the only evidence of nash equilibrium was used: it is assumed that there are two different nash equalizations (r i1 ,r -i1) and (ri2 ,r -i2), and ri1 Not equal to r i2 Since each participant will choose its own optimal transmission frequency, it is achieved that:
u i (r i1 ,r -i1 )>u i (r i2 ,r -i1 )
u i (r i2 ,r -i2 )>u i (r i1 ,r -i2 )
adding the two formulas: u (u) i (r i1 ,r -i1 )+u i (r i2 ,r -i2 )>u i (r i2 ,r -i1 )+u i (r i1 ,r -i2 ) According to a given utility function u i (r i1 ,r -i1 ) The method can obtain:
since the participants under Nash equalization will all choose the transmission frequency that maximizes their utility, r i1 and ri2 The condition that the derivative of the corresponding utility function is equal to 0 is satisfied, namely:
dividing the two formulas to obtain: 1+beta i ·r i1 ＝1+β i ·r i2 Obtaining r i1 ＝r i2 And r is equal to i1 ≠r i2 Contradiction, the nash equilibrium solution of the gaming model is therefore unique.
The operation is simple, and the use is facilitated.
Further, the method for solving the utility function by using the mixed particle swarm optimization algorithm comprises the following steps:
regarding each vehicle as a particle, the transmission frequency of the safety message is taken as the speed of the particle, one solution of the problem is represented by the position of the particle, and the fitness function is represented by the utility function u of the vehicle i i Definition:
f(r i )＝u i (r i ,r -i )
wherein the fitness function f (r i ) By a strategy r of the vehicle i i As a function of the argument; r is (r) i Representing the strategy of vehicle i, r -i ＝(r 1 ,…,r i-1 ,r i+1 ,…,r n ) Representing the transmission frequency strategy of other participating gaming vehicles except vehicle i;
calculating and comparing the fitness value f of the particle position, carrying out iterative updating, and solving the optimal solution of the problem;
the position and the speed of the particles are continuously adjusted, the currently searched individual optimal value and the group optimal value are recorded at the same time, the particle swarm is continuously optimized in the searching range of the particle swarm, and finally the particle swarm is converged to the optimal position gbest with the maximum fitness value f.
And solving through a mixed particle swarm optimization algorithm based on the driving risk of the road grid area where the vehicle is located and the transmission frequency of the safety message to obtain an equalization strategy, and adaptively adjusting the transmission frequency of the safety message of the vehicle based on the obtained equalization strategy by the vehicle.
The invention also provides a vehicle-mounted network channel congestion control system based on the regional risk, which comprises a data acquisition unit and a processing unit, wherein the data acquisition unit is used for acquiring data of driver factors, vehicle factors and environmental factors which influence driving safety, the output end of the data acquisition unit is connected with the input end of the processing unit, and the processing unit executes the method of the invention to control the vehicle-mounted network channel congestion.
The system is used for obtaining the balance strategy of the vehicle, and is beneficial to use.
Drawings
Fig. 1 is a schematic flow chart of a method for controlling channel congestion of a vehicle network based on regional risk;
FIG. 2 is a combined fuzzy inference schematic diagram of the vehicle network channel congestion control method based on regional danger;
fig. 3 is a diagram of a driving risk prediction network structure of the vehicle network channel congestion control method based on regional risk.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
In the description of the present invention, it should be understood that the terms "longitudinal," "transverse," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate describing the present invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention.
In the description of the present invention, unless otherwise specified and defined, it should be noted that the terms "mounted," "connected," and "coupled" are to be construed broadly, and may be, for example, mechanical or electrical, or may be in communication with each other between two elements, directly or indirectly through intermediaries, as would be understood by those skilled in the art, in view of the specific meaning of the terms described above.
The invention discloses a vehicle network channel congestion control method based on regional risk, which is mainly characterized in that a C-V2X congestion control strategy is researched, the running risk of a vehicle is firstly estimated, the running risk of the road region is estimated according to the estimated running risk, the running risk of the road region is estimated, the running risk of the short-time road region is predicted, a non-cooperative game model is built according to the running risk and solved, the balance strategy of the vehicle is obtained, and the vehicle self-adaptively adjusts the safety message transmission frequency of the vehicle based on the obtained balance strategy.
As shown in fig. 1, the method for controlling channel congestion of the vehicle network includes the following steps:
as shown in fig. 2, based on the driver factor, the vehicle factor and the environmental factor affecting the driving safety, a combined fuzzy reasoning method (first, the combined reasoning is performed, then the road area is divided into grids with equal size, and the driving risk of the road area is evaluated by using a weighting method) is used for evaluating the driving risk of the single vehicle; because of the mutual influence among vehicles and the complex and changeable traffic flow, various factors influencing driving safety need to be analyzed, a fuzzy reasoning model is established based on the factors, and combined fuzzy reasoning is carried out based on the models.
Estimating the driving risk of the road area, and using a convolution long-short-term memory neural network (ConvLSTM) to predict the driving risk of the road area in a short time;
constructing a non-cooperative game model, and incorporating the driving risk and the safety information of the road area where the vehicle is located into the utility function of the non-cooperative game model;
and solving the utility function to obtain an equalization strategy of transmission frequency control of the vehicle-mounted network channel congestion control, wherein the output gbest is the global optimal position.
In a preferred scheme of the invention, considering the complexity of influencing factors, the method for evaluating the running risk of a single vehicle by adopting a two-stage fuzzy inference model is as follows:
the first stage includes three fuzzy inference modules for evaluating driving risk levels corresponding to driver factors, vehicle factors, and environmental factors, respectively.
Taking a driver factor as an example, selecting the continuous driving time and driving behavior of the driver as input variables, performing fuzzy reasoning by using an IF-THEN fuzzy rule according to a knowledge base, and defuzzifying to obtain an output language variable of the driving risk degree of the driver factor.
The "driver duration DT" is used as an input language variable, the fuzzy set is divided into three fuzzy language values of normal (S), long (M) and long (B), that is, the fuzzy set of the input language variable "driver duration DT" is { S, M, B }, and the driver duration can be obtained by the vehicle-mounted device. The universe is denoted as [0,4], the longer the duration of driving, the larger the value mapped to the universe.
The "driver driving behavior DB" is used as another input linguistic variable, and the fuzzy set is divided into three fuzzy linguistic values of normal (S), bad (M) and dangerous (B), namely, the fuzzy set is { S, M, B }, which respectively represent normal driving behavior, bad driving behavior (the driver gets closer to the vehicle, changes lanes frequently, accelerates rapidly frequently, etc.), and dangerous driving behavior (does not follow traffic rules, distracted driving, etc.). The driving behavior of the driver can be obtained through the vehicle-mounted device, the smart phone application program and the like, the domain is defined as [0,3], and the more dangerous the driving behavior is, the larger the value mapped to the domain is.
A traditional fuzzy reasoning model is selected for fuzzy reasoning, a driving risk assessment model influenced by a driver factor belongs to a double-input multi-rule reasoning model, and the processing procedure of the reasoning model is as follows:
wherein ,Xn Is the input 1, Y under the nth rule n Is input 2, Z under the nth rule n Is the reasoning result obtained by the nth rule, X * Is the 'U' aggregation of input parameter 1 in all rules, Y * Is the 'U' aggregation of input parameter 2 in all rules;
each rule is first processed:
and then the results obtained by each reasoningPerforming U-shaped polymerization to obtain a final fuzzy result Z * The method comprises the following steps:
heavy useDefuzzification by heart method to obtain Z * Assuming that the driving risk degree accuracy value affected by the obtained driver factor is a, a is:
wherein N represents the number of points in the domain, A i Mu as the i-th point Z* (A i ) Representing the ambiguity solution Z * Membership functions of (2);
obtaining a driving risk degree accurate value D influenced by driver factors 1 (A is general meaning, the accurate values D1, D2 and D3 of the driving risk degree are brought into different values which are influenced by different factors, and the driving risk degree D of the vehicle factors and the environment factors are calculated respectively in the same way 2 、D 3 ；
The second stage is at D 1 ,D 2 ,D 3 On the basis of (1) carrying out fuzzy reasoning and outputting a corresponding total driving risk degree D 0 Triangle membership functions are used:
controlling the sensitivity degree of the triangle membership function to input change by adjusting parameters a, b and c; x is an independent variable; a, b, c represent the vertex positions of triangles respectively; these parameters determine the shape and width of the triangle; if the distance between a, b and c is smaller, i.e. the triangle is narrower, the function is more sensitive to input changes; on the contrary, if the triangle is wider, it is less sensitive to input changes, and a larger input value change is needed to cause a significant change in the membership function.
The fuzzy inference system is obtained by carrying out Cartesian product on membership sets and fuzzy language variables, and each input variable has three membership sets. Thus, the fuzzy rule number of the fuzzy inference system is the product of the membership set numbers of two input language variables, namely 9. The rules formulated are shown in table 1. And (3) reasoning by using a traditional fuzzy reasoning method (Mamdani), and defuzzifying a fuzzy output result by using a maximum value method, a gravity center method, a center method or a weighted average method to obtain the running risk degree D0 of the vehicle.
TABLE 1 driver factor driving risk assessment inference rule table
In a preferred scheme of the invention, the method for evaluating the driving risk of the road area comprises the following steps:
the running state and behavior of the current vehicle may affect the running state and behavior of other vehicles, which may also affect the current vehicle. Therefore, after the running risk evaluation result of the single vehicle is obtained, the running risk of the road area needs to be comprehensively evaluated. Dividing a road into grids, carrying out road area driving risk assessment by taking the grids as units, and determining surrounding grids of the current road grid;
the communication range of C-V2X is 150m, but in an actual road scene, when the mesh approaches the boundary, the mesh range to be considered needs to be adjusted accordingly according to the actual situation, so for each road mesh (i, j), the surrounding mesh is defined as:
extending in the length direction from min (i-30, 0) to max (i+30, 199); a grid extending in the width direction from min (j-50, 0) to max (j+50, 3) region;
for each road grid, calculating a risk value through a weighted average value, and representing the influence of vehicles with different distances on the current road grid by adopting a Euclidean distance weighting method;
let the current vehicle's grid be i and the neighboring surrounding grids be j 1 ,j 2 ,…,j k The corresponding driving risk values are R1, R2, … and Rk; the running risk assessment result of the current grid i is:
R i ＝(ω 1 R 1 +ω 2 R 2 +…+ω n R n )/(ω 1 +ω 2 +…+ω n )
wherein ,ωi Representing the current grid i to the adjacent grid j 1 ,j 2 ,…,j k The inverse of the euclidean distance, i.e.:
...
where d (i, j) represents the Euclidean distance between grid i and grid j, namely:
wherein ,(xi ,y i ) Represents the coordinates of grid i, (x) j ,y j ) Representing coordinates of grid j;
obtaining the running risk degree of the road grid where the vehicle is located by using a matrix X t To represent the traffic risk degree distribution condition of the road area at the moment t:
wherein ,representing the driving risk degree of the ith row and jth column road grids at the moment t.
In a preferred scheme of the invention, the specific method for predicting the driving risk in short time in the road area comprises the following steps:
the road traffic environment has the characteristics of diversity, uncertainty and the like, and the running risk at the current moment can not fully reflect the running risk at the next moment, so that the result is inaccurate when the self-adaptive adjustment parameters are used for congestion control. To overcome these problems, short-term prediction of road area driving risk is required to make up for the lack of real-time assessment.
The traffic flow changes are not only time-dependent but also spatially-dependent, so that the distribution changes of the driving risk degree also have a time-space correlation.
Predicting the driving risk degree of n times in the future by adopting the driving risk degree of m times in the past, which is expressed as follows:
X t,t+n ＝Γ(X t-m+1,t )
wherein ,Xt,t+n The travel risk from time t to time t+n is represented by Γ, the prediction method is represented by t, the time X is represented by t t-m+1,t Representing the driving risk degree from the time t-m+1 to the time t;
in order to more accurately capture the distribution change of the driving risk degree in the space-time dimension, the scheme adopts a convolution long-short-term memory neural network (ConvLSTM) to conduct short-time prediction. ConvLSTM integrates the advantages of a convolutional neural network and a long-term and short-term memory network, can process time sequence data and space data at the same time, and accords with the characteristics of road traffic.
Unlike conventional LSTMs, convLSTM uses convolution to perform the computation rather than matrix multiplication as used in LSTMs. Second, the input of the ConvLSTM model is the input of the 3D tensor, not the 2D input in LSTM. Short-term prediction can be performed by using the existing convolutional long-term memory neural network, for example, the convLstm formula of the convolutional long-term memory neural network is as follows:
wherein ,it 、f t 、o t 、X t 、H t 、C t All represent 3D tensors, "* "represents a convolution operation; sigma is a sigma function, and by which is the same or sign.
The road area traveling risk prediction model of the invention stacks a plurality of ConvLSTM layers, and consists of a coding network and a prediction network, as shown in figure 3. Both the coding network and the prediction network are composed of multiple ConvLSTM layers, which are used to process spatio-temporal sequence data. The encoding network compresses the input sequence to generate a hidden state tensor, and the tensor carries information of driving risks. And the prediction network expands the hidden state tensor into an output sequence, and the prediction is realized by gradually predicting the value of each position in the sequence. To ensure continuity and consistency of the neural network, both the initial state of the predictive network and the cell output are replicated from the final state of the encoding network.
Finally, all states in the prediction network can be connected and the final prediction result is generated through the convolution layer. By the method, the model can effectively process the time-space correlation and realize accurate prediction of the road driving risk degree.
In a preferred scheme of the invention, the method for constructing the non-cooperative game model comprises the following steps:
according to the structural elements of the game theory, the congestion control model based on the travelling risk game is as follows:
G＝{V；r 1 ,r 2 ,…,r n ；u 1 ,u 2 ,…,u n }
where v= {1,2, …, n } represents the set of participants, i.e. vehicles, participating in the game, r= { R 1 ,r 2 ,…r n Policy set representing the frequency of transmissions of vehicles participating in a game, r i Representing the strategy of the vehicle i, the transmission frequency range of the vehicle i participating in the game is 0<r min ≤r i ≤r max The method comprises the steps of carrying out a first treatment on the surface of the The transmission frequency in C-V2X ranges from [1,50]I.e. 1.ltoreq.ri.ltoreq.50;
u i utility function representing vehicle i participating in gaming:
u i ＝(r i ,r -i )
wherein ,r-i ＝(r 1 ,…,r i-1 ,r i+1 ,…,r n ) Representing the transmission frequency strategy of other participating game vehicles except the vehicle i, wherein i is a natural number;
considering that each participant needs to select an optimal strategy to achieve the maximization of own interests according to own interests and targets, the optimal transmission frequency of the vehicle i is determined by the maximization u i ＝(r i ,r -i ) The resulting problem is expressed as:
r i * ＝arg max u i (r i ,r -i )
r i ∈[r min ,r max ]
the final utility function is:
wherein ,ri Representing the transmission frequency of vehicle i (i.e. the strategy of vehicle i), d i Representing the driving risk degree of the vehicle i, alpha i Representing utility factor, beta i Is a constant coefficient, and alpha i >>β i ，λ i In order to take the cost factor into account,n is the total number of vehicles, related to the strategy of the other vehicles.
The first term of the utility function is a benefit function representing the benefit of the vehicle i in the game, and the transmission frequency r i Relatedly, with r i The utility value will also increase. Wherein the utility factor alpha i Indicating the gain sensitivity of the ith vehicle to the transmission frequency, and the coefficient beta i Indicating the rate of increase of its revenue. The second term of the utility function represents the cost function of the vehicle i and the driving risk degree d i Related to the following. The cost function is proportional to the transmission frequency of the vehicle and the driving risk degree d i Inversely proportional. This means that the greater the driving risk of the vehicle, the easier it is to obtain a higher transmission frequency.
The utility function can balance the running risk and income of the vehicle, and the vehicle with high risk has higher transmission frequency due to lower punishment force so as to ensure that safety information can be timely and reliably conveyed. For a common vehicle, the higher punishment force enables the common vehicle to obtain lower transmission frequency, so that the common vehicle is beneficial to saving communication resources and relieving the problem of channel congestion.
In a preferred scheme of the invention, the method for proving the existence of Nash equilibrium of the non-cooperative game model is as follows:
for any vehicle i, its strategy is discrete and is in interval 0<r min ≤r i ≤r max Internally, R is thus non-empty, closed and bounded;
utility function u for each vehicle i (r i ,r -i ) Is continuous on self R, for u i The first order bias derivative is calculated by:
and continuing to obtain a second-order partial derivative:
for any r in the interval i All have 0<r min ≤r i ≤r max, and therefore, the utility function of the vehicle is a continuous concave function in the strategy space, so that a Nash equilibrium solution exists in the non-cooperative game model;
the only evidence of nash equilibrium was used: it is assumed that there are two different nash equalizations (r i1 ,r -i1) and (ri2 ,r -i2), and ri1 Not equal to r i2 Since each participant will choose its own optimal transmission frequency, it is achieved that:
u i (r i1 ,r -i1 )>u i (r i2 ,r -i1 )
u i (r i2 ,r -i2 )>u i (r i1 ,r -i2 )
adding the two formulas: u (u) i (r i1 ,r -i1 )+u i (r i2 ,r -i2 )>u i (r i2 ,r -i1 )+u i (r i1 ,r -i2 ) According to a given utility function u i (r i1 ,r -i1 ) The method can obtain:
since the participants under Nash equalization will all choose the transmission frequency that maximizes their utility, r i1 and ri2 The condition that the derivative of the corresponding utility function is equal to 0 is satisfied, namely:
/>
dividing the two formulas to obtain: 1+beta i ·r i1 ＝1+β i ·r i2 Obtaining r i1 ＝r i2 And r is equal to i1 ≠r i2 Contradiction, the nash equilibrium solution of the gaming model is therefore unique.
In a preferred embodiment of the present invention, the method for solving the utility function by using the hybrid particle swarm optimization algorithm (HPSO) comprises:
regarding each vehicle as a particle, the transmission frequency of the safety message is taken as the speed of the particle, one solution of the problem is represented by the position of the particle, and the fitness function is represented by the utility function u of the vehicle i i Definition:
f(r i )＝u i (r i ,r -i )
wherein the fitness function f (r i ) By a strategy r of the vehicle i i As a function of the argument; r is (r) i Representing the strategy of vehicle i, r -i ＝(r 1 ,…,r i-1 ,r i+1 ,…,r n ) Representing the transmission frequency strategy of other participating gaming vehicles except vehicle i;
calculating and comparing the fitness value f of the particle position, carrying out iterative updating, and solving the optimal solution of the problem;
the position and the speed of the particles are continuously adjusted, the currently searched individual optimal value and the group optimal value are recorded at the same time, the particle swarm is continuously optimized in the searching range of the particle swarm, and finally the particle swarm is converged to the optimal position gbest with the maximum fitness value f.
According to the technical scheme, a non-cooperative game model suitable for the Internet of vehicles is established at first to describe the competition relationship and interaction between vehicles. Based on the running risk of the road grid area where the vehicle is located and the transmission frequency of the safety message, a utility function of the vehicle is designed, and the uniqueness and the existence of Nash equilibrium of the model are proved. Finally, solving through a mixed particle swarm optimization (Hybrid Particle Swarm Optimization, HPSO) algorithm to obtain an equalization strategy, wherein the equalization strategy can be applied to adaptive adjustment of the vehicle transmission frequency.
The invention also provides a vehicle network channel congestion control system based on the regional danger degree, which comprises a data acquisition unit and a processing unit, wherein the data acquisition unit (such as a camera, a temperature sensor, a speed sensor and the like) is used for acquiring data of driver factors, vehicle factors and environmental factors which influence driving safety, the output end of the data acquisition unit is electrically connected with the input end of the processing unit, and the processing unit executes the method disclosed by the invention to control the vehicle network channel congestion. The system is used for obtaining the balance strategy of the vehicle, and is beneficial to use.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
Claims (8)
1. The vehicle network channel congestion control method based on the regional danger degree is characterized by comprising the following steps of:
based on driver factors, vehicle factors and environmental factors affecting driving safety, evaluating the driving risk of a single vehicle by using a combined fuzzy reasoning method;
estimating the driving risk of the road area, and using a convolution long-short-term memory neural network to predict the driving risk of the road area in a short time;
constructing a non-cooperative game model, and incorporating the driving risk and the safety information of the road area where the vehicle is located into the utility function of the non-cooperative game model;
and solving the utility function to obtain the equalization strategy of the transmission frequency control of the vehicle-mounted network channel congestion control.
2. The method for controlling channel congestion of an on-board network based on regional risk according to claim 1, wherein the method for evaluating the running risk of a single vehicle is as follows:
a traditional fuzzy reasoning model is selected for fuzzy reasoning, a driving risk assessment model influenced by a driver factor belongs to a double-input multi-rule reasoning model, and the processing procedure of the reasoning model is as follows:
wherein ,Xn Is the input 1, Y under the nth rule n Is input 2, Z under the nth rule n Is the reasoning result obtained by the nth rule, X * Is the 'U' aggregation of input parameter 1 in all rules, Y * Is the 'U' aggregation of input parameter 2 in all rules;
each rule is first processed:
and then the results obtained by each reasoningPerforming U-shaped polymerization to obtain a final fuzzy result Z * The method comprises the following steps:
deblurring by gravity center method to obtain Z * Assuming that the driving risk degree accuracy value affected by the obtained driver factor is a, a is:
wherein N represents the number of points in the domain, A i As the point of the i-th point,representing the ambiguity solution Z * Membership functions of (2);
obtaining a driving risk degree accurate value D influenced by driver factors 1 Similarly, calculate respectivelyVehicle risk degree D of vehicle factor and environmental factor 2 、D 3 ；
At D 1 ,D 2 ,D 3 On the basis of (1) carrying out fuzzy reasoning and outputting a corresponding total driving risk degree D 0 Triangle membership functions are used:
controlling the sensitivity degree of the triangle membership function to input change by adjusting parameters a, b and c; x is an independent variable; a, b, c represent the vertex positions of triangles respectively;
and (3) reasoning by using a traditional fuzzy reasoning method, and obtaining the running risk degree of the vehicle after defuzzification.
3. The method for controlling channel congestion of an on-board network based on regional risk according to claim 1, wherein the method for evaluating the running risk of the road region is as follows:
dividing a road into grids, carrying out road area driving risk assessment by taking the grids as units, and determining surrounding grids of the current road grid;
the communication range of C-V2X is 150m, but in an actual road scene, when the mesh approaches the boundary, the mesh range to be considered needs to be adjusted accordingly according to the actual situation, so for each road mesh (i, j), the surrounding mesh is defined as:
extending in the length direction from min (i-30, 0) to max (i+30, 199); a grid extending in the width direction from min (j-50, 0) to max (j+50, 3) region;
for each road grid, calculating a risk value through a weighted average value, and representing the influence of vehicles with different distances on the current road grid by adopting a Euclidean distance weighting method;
let the current vehicle's grid be i and the neighboring surrounding grids be j 1 ,j 2 ,…,j k The corresponding driving risk values are R1, R2, … and Rk; then the current grid i is drivingThe risk assessment results were:
R i ＝(ω 1 R 1 +ω 2 R 2 +…+ω n R n )/(ω 1 +ω 2 +…+ω n )
wherein ,ωi Representing the current grid i to the adjacent grid j 1 ,j 2 ,…,j k The inverse of the euclidean distance, i.e.:
...
where d (i, j) represents the Euclidean distance between grid i and grid j, namely:
wherein ,(xi ,y i ) Represents the coordinates of grid i, (x) j ,y j ) Representing coordinates of grid j;
obtaining the running risk degree of the road grid where the vehicle is located by using a matrix X t To represent the traffic risk degree distribution condition of the road area at the moment t:
wherein ,representing the driving risk degree of the ith row and jth column road grids at the moment t.
4. The method for controlling channel congestion of an on-board network based on regional risk according to claim 1, wherein the specific method for predicting the traffic risk for the road region in short time is as follows:
predicting the driving risk degree of n times in the future by adopting the driving risk degree of m times in the past, which is expressed as follows:
X t,t+n ＝Γ(X t-m+1,t )
wherein ,Xt,t+n The travel risk from time t to time t+n is represented by Γ, the prediction method is represented by t, the time X is represented by t t-m+1,t The running risk degree from the time t-m+1 to the time t is shown.
5. The method for controlling channel congestion of an on-board network based on regional danger according to claim 1, wherein the method for constructing the non-cooperative game model is as follows:
the congestion control model based on the travelling risk game is as follows:
G＝{V；r 1 ,r 2 ,…,r n ；u 1 ,u 2 ,…,u n }
where v= {1,2, …, n } represents the set of participants, i.e. vehicles, participating in the game, r= { R 1 ,r 2 ,…r n Policy set representing the frequency of transmissions of vehicles participating in a game, r i Representing the strategy of the vehicle i, the transmission frequency range of the vehicle i participating in the game is 0<r min ≤r i ≤r max The method comprises the steps of carrying out a first treatment on the surface of the The transmission frequency in C-V2X ranges from [1,50]I.e. 1.ltoreq.ri.ltoreq.50;
u i utility function representing vehicle i participating in gaming:
u i ＝(r i ,r -i) wherein ,r-i ＝(r 1 ,…,r i-1 ,r i+1 ,…,r n ) Representing the transmission frequency strategy of other participating game vehicles except the vehicle i, wherein i is a natural number;
considering that each participant needs to select an optimal strategy to achieve the maximization of own interests according to own interests and targets, the optimal transmission frequency of the vehicle i is determined by the maximization u i ＝(r i ,r -i ) The resulting problem is expressed as:
r i * ＝arg max u i (r i ,r -i )
r i ∈[r min ,r max ]
the final utility function is:
wherein ,ri Representing the transmission frequency of vehicle i, d i Representing the driving risk degree of the vehicle i, alpha i Representing utility factor, beta i Is a constant coefficient, and alpha i >>β i ，λ i In order to take the cost factor into account,n is the total number of vehicles, related to the strategy of the other vehicles.
6. The method for controlling channel congestion in an on-board network based on regional risk of claim 5, further comprising the step of proving existence of nash equalization in a non-cooperative game model as follows:
for any vehicle i, its strategy is discrete and is in interval 0<r min ≤r i ≤r max Internally, R is thus non-empty, closed and bounded;
utility function u for each vehicle i (r i ,r -i ) Is continuous on self R, for u i The first order bias derivative is calculated by:
and continuing to obtain a second-order partial derivative:
for any r in the interval i All have 0<r min ≤r i ≤r max, and therefore, the utility function of the vehicle is a continuous concave function in the strategy space, so that a Nash equilibrium solution exists in the non-cooperative game model;
the only evidence of nash equilibrium was used: it is assumed that there are two different nash equalizations (r i1 ,r -i1) and (ri2 ,r -i2), and ri1 Not equal to r i2 Since each participant will choose its own optimal transmission frequency, it is achieved that:
u i (r i1 ,r -i1 )>u i (r i2 ,r -i1 )
u i (r i2 ,r -i2 )>u i (r i1 ,r -i2 )
adding the two formulas: u (u) i (r i1 ,r -i1 )+u i (r i2 ,r -i2 )>u i (r i2 ,r -i1 )+u i (r i1 ,r -i2 ) According to a given utility function u i (r i1 ,r -i1 ) The method can obtain:
since the participants under Nash equalization will all choose the transmission frequency that maximizes their utility, r i1 and ri2 The condition that the derivative of the corresponding utility function is equal to 0 is satisfied, namely:
dividing the two formulas to obtain: 1+beta i ·r i1 ＝1+β i ·r i2 Obtaining r i1 ＝r i2 And r is equal to i1 ≠r i2 Contradiction, the nash equilibrium solution of the gaming model is therefore unique.
7. The method for controlling channel congestion of an on-board network based on regional danger according to claim 1, wherein the method for solving utility functions by using a hybrid particle swarm optimization algorithm is as follows:
regarding each vehicle as a particle, the transmission frequency of the safety message is taken as the speed of the particle, one solution of the problem is represented by the position of the particle, and the fitness function is represented by the utility function u of the vehicle i i Definition:
f(r i )＝u i (r i ,r -i )
wherein the fitness function f (r i ) By a strategy r of the vehicle i i As a function of the argument; r is (r) i Representing the strategy of vehicle i, r -i ＝(r 1 ,…,r i-1 ,r i+1 ,…,r n ) Representing the transmission frequency strategy of other participating gaming vehicles except vehicle i;
calculating and comparing the fitness value f of the particle position, carrying out iterative updating, and solving the optimal solution of the problem;
the position and the speed of the particles are continuously adjusted, the currently searched individual optimal value and the group optimal value are recorded at the same time, the particle swarm is continuously optimized in the searching range of the particle swarm, and finally the particle swarm is converged to the optimal position gbest with the maximum fitness value f.
8. The vehicle network channel congestion control system based on the regional risk is characterized by comprising a data acquisition unit and a processing unit, wherein the data acquisition unit is used for acquiring data of driver factors, vehicle factors and environment factors which influence driving safety, the output end of the data acquisition unit is connected with the input end of the processing unit, and the processing unit executes the method of one of claims 1-7 to control the vehicle network channel congestion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310770589.2A CN116959244A (en) | 2023-06-28 | 2023-06-28 | Vehicle network channel congestion control method and system based on regional danger |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310770589.2A CN116959244A (en) | 2023-06-28 | 2023-06-28 | Vehicle network channel congestion control method and system based on regional danger |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116959244A true CN116959244A (en) | 2023-10-27 |
Family
ID=88447014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310770589.2A Pending CN116959244A (en) | 2023-06-28 | 2023-06-28 | Vehicle network channel congestion control method and system based on regional danger |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116959244A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117657140A (en) * | 2024-02-01 | 2024-03-08 | 中印云端(深圳)科技有限公司 | Dual-drive control system based on new energy automobile |
-
2023
- 2023-06-28 CN CN202310770589.2A patent/CN116959244A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117657140A (en) * | 2024-02-01 | 2024-03-08 | 中印云端(深圳)科技有限公司 | Dual-drive control system based on new energy automobile |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | Deep reinforcement learning for mobile edge caching: Review, new features, and open issues | |
| WO2021103625A1 (en) | Short-term vehicle speed condition real-time prediction method based on interaction between vehicle ahead and current vehicle | |
| CN112265546B (en) | Networked automobile speed prediction method based on time-space sequence information | |
| CN113538910B (en) | Self-adaptive full-chain urban area network signal control optimization method | |
| CN112668799A (en) | Intelligent energy management method and storage medium for PHEV (Power electric vehicle) based on big driving data | |
| CN114495499B (en) | Multi-target intelligent internet vehicle cooperative optimization control method | |
| CN101282243A (en) | Method for recognizing distributed amalgamation of wireless sensor network | |
| CN116959244A (en) | Vehicle network channel congestion control method and system based on regional danger | |
| CN112950963B (en) | Self-adaptive signal control optimization method for main branch intersection of city | |
| Jamil et al. | Adaptive traffic signal control system using composite reward architecture based deep reinforcement learning | |
| CN107045785A (en) | A kind of method of the short-term traffic flow forecast based on grey ELM neutral nets | |
| CN113886095A (en) | Container memory elastic expansion method based on combination of fuzzy reasoning and reinforcement learning | |
| CN114710410B (en) | Network data acquisition efficiency optimization method and system based on deep reinforcement learning | |
| CN116782296A (en) | Digital twinning-based internet-of-vehicles edge computing and unloading multi-objective decision method | |
| CN110930696B (en) | AI navigation-based intelligent city traffic management operation method and system | |
| CN114444922A (en) | Hybrid traffic efficiency evaluation method under group intelligent control | |
| Chen et al. | Traffic signal optimization control method based on adaptive weighted averaged double deep Q network | |
| Zhang et al. | Multi-objective deep reinforcement learning approach for adaptive traffic signal control system with concurrent optimization of safety, efficiency, and decarbonization at intersections | |
| CN112508408B (en) | Mapping model construction method of radio resource management index under edge calculation | |
| Anirudh et al. | Intelligent Traffic Control System using Deep Reinforcement Learning | |
| CN112750298A (en) | Truck formation dynamic resource allocation method based on SMDP and DRL | |
| CN110021168B (en) | Grading decision method for realizing real-time intelligent traffic management under Internet of vehicles | |
| Wu et al. | Collaborative caching relay algorithm based on recursive deep reinforcement learning in mobile vehicle edge network | |
| Zhu et al. | Edge collaborative caching solution based on improved NSGA II algorithm in Internet of Vehicles | |
| CN113344071A (en) | Intrusion detection algorithm based on depth strategy gradient |
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 |