CN116341288A - Heterogeneous traffic epidemic car security field modeling method - Google Patents

Heterogeneous traffic epidemic car security field modeling method Download PDF

Info

Publication number
CN116341288A
CN116341288A CN202310593492.9A CN202310593492A CN116341288A CN 116341288 A CN116341288 A CN 116341288A CN 202310593492 A CN202310593492 A CN 202310593492A CN 116341288 A CN116341288 A CN 116341288A
Authority
CN
China
Prior art keywords
vehicle
field
cav
psychological
environmental
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.)
Granted
Application number
CN202310593492.9A
Other languages
Chinese (zh)
Other versions
CN116341288B (en
Inventor
王薇
马国栋
宋佳
刘娇娇
孙宝凤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN202310593492.9A priority Critical patent/CN116341288B/en
Publication of CN116341288A publication Critical patent/CN116341288A/en
Application granted granted Critical
Publication of CN116341288B publication Critical patent/CN116341288B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • 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)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention belongs to the field of traffic control systems, and particularly relates to a heterogeneous traffic epidemic vehicle safety field modeling method, which comprises the steps of respectively constructing a CAV vehicle field model, an HDV vehicle field model, an environment field model taking a transverse distance as a variable and an environment field model taking a longitudinal distance as a variable, and drawing a driving safety field intensity diagram; and the influence of the individual difference of the HDV driver and the driving environment on the driving safety is considered, the comprehensive index of the psychological visibility of the driver environment is constructed, the psychological acting force of the HDV driver in the mixed driving is deduced by means of the concept of the psychological field, and the vehicle field model of the HDV traditional vehicle is established.

Description

Heterogeneous traffic epidemic car security field modeling method
Technical Field
The invention belongs to the field of traffic control systems, and particularly relates to a heterogeneous traffic epidemic car security field modeling method.
Background
The inherent chaos of highway ramp confluence is considered to be one of the main causes of traffic accidents and congestion. Future internet-connected automatic driving Vehicles (Connected and Autonomous Vehicle, CAV) coexist with manual driving Vehicles (HDV) for a long time, which is called as heterogeneous traffic flow characteristics, and deep exploration of microscopic traffic behavior characteristics and mechanisms such as following and lane changing of the heterogeneous traffic flow is crucial to improving traffic safety and traffic efficiency of a mixed traffic system.
Heterogeneous traffic flow modeling composed of CAV and HDV is mostly based on a vehicle networking traffic flow modeling method based on continuous cellular automaton as disclosed in China patent CN 115601958A; the Chinese patent CN 113204863B discloses a manual-CACC automatic driving vehicle mixed flow simulation method based on cellular automata; a traffic flow cellular automaton modeling method under the environment of internet of vehicles, and the like, disclosed in Chinese patent CN 106652564A. However, cellular automata has poor interpretability for embodying the safety element of vehicle operation. In order to highlight the importance of safety for automatic driving traffic control, a heterogeneous traffic flow driving behavior vehicle field model and an environment field model are built based on a safety potential field theory, and high attention is drawn to the academic community.
The vehicle field modeling problem based on the safety potential field is reported that articles such as safety potential field-based mixed traffic flow motion situation model research and the like model the current traffic safety field heterogeneous traffic flow or indiscriminate, or a degraded CAV vehicle field model is used for replacing the vehicle field of the HDV, the information such as acceleration, position and the like which cannot be observed by the HDV vehicle is ignored, traffic influence factors such as individuality, visual field, trust degree and the like of a driver of the HDV vehicle in the heterogeneous traffic flow are not considered, the influence degree of different types of drivers on the safety field range cannot be shown, and the safety field modeling of two types of vehicles in the heterogeneous traffic flow is not researched and comprehensively analyzed. Meanwhile, because the potential field strength in the vehicle field model is limited by the action range, short-range force is applied to surrounding vehicles within a short distance, the force changes along with the change of factors such as vehicle speed, acceleration, distance and the like, when the distance between a target vehicle and the surrounding vehicles exceeds a certain range, the influence on the field strength of the surrounding vehicles is negligible, but at present, students do not limit the action range of the field force.
The environment field modeling problem based on the safety potential field is that the existing report only adopts the environment field modeling with the transverse distance as a variable, the key influence of the longitudinal distance is not considered, and the intrinsic mechanism of real lane changing in some special scenes, such as the situation that lanes are accelerated in a confluence area, the number of lanes is reduced due to traffic accidents or temporary construction, and the like, cannot be embodied. The concrete steps are as follows: the variable of the existing environment field is the transverse distance between the vehicle and the road boundary as well as the marked line, and the limitation of the environment facility and laws and regulations on the vehicle behavior can be well described when no road condition change occurs. However, when the number of lanes becomes small, it is necessary to forcibly change lanes to adjacent lanes when the vehicle travels to the end of the lane. Simulation experiments show that the number of vehicles which are not successfully changed can be increased by adopting the existing environment field model, and the combined vehicles which are not successfully changed can only stop at the tail end of a lane, so that the chance of waiting for combined lane change by peak traffic flow is avoided, and the vehicles bear larger driving risks and are inconsistent with the actual lane change situation.
Disclosure of Invention
Aiming at the problem of indiscriminate modeling existing in the safety field modeling of the existing heterogeneous traffic popular cars, the invention reconstructs the vehicle field models of the existing CAV and HDV based on the safety potential field theory, and solves the problem that the existing vehicle field models do not fully describe the influence degree of different types of drivers on the safety field range. Aiming at the problem that the lane change environment does not consider the influence of the longitudinal distance of a lane under certain special conditions of the existing heterogeneous traffic flow, the invention establishes the driving environment field taking the longitudinal distance as a variable, truly reflects the influence of the reduction of the number of temporary lanes caused by acceleration lanes, road construction or traffic accidents of a certain special conditions, such as a confluence region, on the running condition of a vehicle, and accords with the actual lane change condition. The technical scheme of the invention is as follows:
a heterogeneous traffic epidemic car security field modeling method comprises the following steps:
step A: construction of CAV vehicle field model
A1, quantifying target vehicle attributes;
a2, determining pseudo distance;
the dangerous degree of the target vehicle also depends on the relative positions of the target vehicle and surrounding potential risk vehicles, and the actual space distance of the potential risk vehicles, which are close to the target vehicle at different angles, is corrected by referring to Euclidean distance to obtain pseudo distance;
a3, coordinate transformation is carried out
Consider potentially risky vehicle roll
Figure SMS_1
When the vehicle runs in the axial direction, the vehicle body can deflect to a certain extent, the vehicle field model also deflects along with the deflection, and in a lane change scene, the formula is as follows:
Figure SMS_2
；
the counterclockwise direction is defined as positive, where:
Figure SMS_3
deflecting the course angle counterclockwise for the field model;
a4 determination of CAV vehicle field model
Describing CAV vehicle field strength as a function of the form of the takawa potential to describe short-range interactions between nuclei
Figure SMS_4
；
A5, calibrating parameters of CAV vehicle field model and restraining action range of the model
Using a differential evolution algorithm, and performing performance parameter calibration on undetermined coefficients in the established CAV vehicle field model by taking a difference value between a force range of a minimized CAV vehicle field and a vehicle head time interval in a strong following state as an objective function, so that field intensity distribution of the undetermined coefficients is more in accordance with a real running state of the vehicle, and taking the strong following vehicle head time interval as a critical action range of field intensity of the CAV vehicle field, so that a field intensity value of the CAV vehicle field can be obtained, and the action range of the CAV vehicle field is restrained;
and (B) step (B): construction of HDV vehicle field model
Step B1: restraining the reach of an HDV vehicle field
The HDV vehicle can keep a stable state when following a car ahead with a 2.7s headway, and the field intensity value of the vehicle is a critical value 0, namely:
Figure SMS_5
；
wherein:
Figure SMS_6
is the standard headway; />
Figure SMS_7
The vehicle speed is the self-vehicle speed;
step B2: construction of driver environmental psychological bearing degree
Adopting a fuzzy theory, selecting typical characteristic factors which reflect the characteristics of a driver: determining the individual experience degree, the environment visibility degree and the psychological trust degree, inputting the characteristic value of the driver environment psychological tolerance assessment model, fuzzifying the characteristic value, obtaining a fuzzy quantity of the driver environment psychological tolerance through logic operation of a fuzzy rule, and converting the fuzzy quantity into an accurate specific numerical value by utilizing defuzzification, namely the driver environment psychological tolerance finally calculated by the driver environment psychological tolerance assessment model;
step B3: determining HDV vehicle field model
Figure SMS_8
；
Wherein:
Figure SMS_9
the field strength is HDV vehicle field strength; />
Figure SMS_10
Is an adjustment coefficient related to the extremum of the function; />
Figure SMS_11
The environmental psychological bearing degree is used for the driver; />
Figure SMS_12
A critical threshold for a safe distance; />
Figure SMS_13
Is a coefficient to be determined in relation to speed; />
Figure SMS_14
Is pseudo distance;
step C: construction of an environmental field model with lateral distance as a variable
An exponential function is selected to construct an environment field taking the transverse distance as a variable, so that the value of the environment field is infinite at the boundary to have a trend of preventing the vehicle from driving away;
step D: construction of an environmental field model with longitudinal distance as a variable
Selecting repulsive potential of potential field theory to construct a vehicle boundary environment field so as to quantitatively represent the influence of reduction of the number of lanes in road condition change on vehicle driving safety, and establishing an environment field along the driving direction of the traffic flow for the lanes;
step E: drawing field intensity diagram of driving safety field
And drawing a CAV vehicle field, an HDV vehicle field, a multi-vehicle superposition vehicle field and an environment field through MATLAB.
In the preferred embodiment of the present invention, in the A1, the target vehicle
Figure SMS_15
Equivalent mass of->
Figure SMS_16
The formula of (2) is:
Figure SMS_17
；
wherein:
Figure SMS_18
for the target vehicle->
Figure SMS_19
Equivalent mass of (a); />
Figure SMS_20
For the target vehicle->
Figure SMS_21
Is the actual mass of (3); />
Figure SMS_22
For the target vehicle->
Figure SMS_23
Is a function of the speed of the machine.
As a preferred aspect of the present invention, in the A2, the pseudo distance
Figure SMS_24
The formula of (2) is:
Figure SMS_25
；
wherein:
Figure SMS_26
and->
Figure SMS_27
Vehicle length and width, respectively, < >>
Figure SMS_28
Is a predetermined coefficient related to the road.
As a preferred aspect of the present invention, in the A4, CAV vehicle field strength
Figure SMS_29
The formula is:
Figure SMS_30
；
Figure SMS_31
；
wherein:
Figure SMS_32
、/>
Figure SMS_33
all are undetermined coefficients; />
Figure SMS_34
The position coordinates of the space where the mass center of the target vehicle is located; />
Figure SMS_35
The current acceleration of the target vehicle; />
Figure SMS_36
For the spatial coordinates of a point around the target vehicle to the centroid of the vehicle +.>
Figure SMS_37
Is included in the plane of the first part;
Figure SMS_38
the value of the original coordinate after deflection is taken.
In the preferred embodiment of the invention, in the A5, the specific method for calibrating the CAV vehicle field model parameter is as follows:
a5.1 screening of vehicle Natural Driving trajectory data set
A5.2 pretreatment of data set
According to screening conditions of the following state, ensuring that front and rear vehicles run on the same lane, setting the distance between the front and rear vehicles within the range of 2-150m, defaulting to vehicle queuing state rejection when the distance is too small, and defaulting to free flow state rejection when the distance is too large; setting a minimum following time of 10s, and defaulting to a state that a stable state cannot be achieved and rejecting when the duration is too large or too small; acquiring a distribution interval of each following vehicle data by arranging and analyzing track data of a data set, and extracting effective following data by analyzing a distribution rule of the distribution interval;
a5.3 obtaining the headway in the Strong heel-and-heel state
A5.4, calibrating performance parameters of undetermined coefficients in the CAV vehicle field model by utilizing effective following data through a differential evolution algorithm; and taking the headway in a strong following state as a critical action range of the CAV vehicle field intensity to obtain a CAV vehicle field intensity value and restraining the action range of the CAV vehicle field.
As a preferred aspect of the present invention, in the B2, the specific construction method of the driver environmental psychological tolerance is as follows:
b2.1 determining environmental psychological bearing index
Determining the individual experience degree, the environmental visibility and the psychological trust degree of the driver as environmental psychological bearing degree indexes;
b2.2 determining the discourse of each index
The driver environmental psychological bearing degree evaluation model is three-input single-output, an individual experience degree domain interval is set to be [0,5], a fuzzy subset is { Low, medium, high }, a domain interval of environmental visibility is [0,3], the fuzzy subset is { Near, middle, far }, a psychological trust degree domain is [ -3,3], and the fuzzy subset is { negative big NB, negative small NS, zero ZO, positive small PS, positive big PB };
b2.3 determining the membership function and membership of each index
Setting fuzzy input as individual experience degree of a driver and environment visibility, selecting a triangular form membership function, selecting a Gaussian form membership function when the fuzzy input is psychological trust degree, and obtaining membership of each index according to the membership function;
b2.4 establishing a driver environmental psychological bearing evaluation model rule table according to the membership degree
And B2.5, determining the relation among three input indexes by a rule table to obtain a bearing degree value in the range of [0,1], namely the environmental psychological bearing degree of the driver.
In the step C, the environmental field model formula using the lateral distance as a variable is as follows:
Figure SMS_39
；
wherein: assuming that the vehicle travel road is a two-lane,
Figure SMS_40
indicating a lane outside road boundary line, too close to which a substantial collision with the vehicle can occur; />
Figure SMS_41
The vehicle running coordinates; />
Figure SMS_42
And->
Figure SMS_43
Risk coefficients of the road marking and the boundary risk field are respectively; />
Figure SMS_44
Is the total width of the lane; />
Figure SMS_45
Is the potential field convergence coefficient.
As a preferred aspect of the present invention, in the step D, the environmental field model formula using the longitudinal distance as a variable is:
Figure SMS_46
；
wherein:
Figure SMS_47
risk coefficients of the road marking and the boundary risk field; />
Figure SMS_48
For vehicle->
Figure SMS_49
Vector distance to the end of the acceleration lane.
In the step E, the specific process of the field intensity diagram of the driving safety field is as follows:
CAV vehicle field, input: vehicle speed
Figure SMS_51
Vehicle acceleration->
Figure SMS_54
Yaw angle>
Figure SMS_57
Vehicle mass->
Figure SMS_50
Vehicle position coordinates and differential evolution methodDefined +.>
Figure SMS_53
、/>
Figure SMS_55
、/>
Figure SMS_58
、/>
Figure SMS_52
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: CAV vehicle field strength>
Figure SMS_56
；
HDV vehicle field, input: adjustment coefficients related to function extremum
Figure SMS_59
Driver mental tolerance->
Figure SMS_60
Vehicle speed->
Figure SMS_61
The method comprises the steps of carrying out a first treatment on the surface of the Vehicle position coordinates; and (3) outputting: HDV vehicle field strength->
Figure SMS_62
；
Multiple vehicle superposition vehicle field, input: speed of two vehicles
Figure SMS_63
、/>
Figure SMS_67
Acceleration of two vehicles->
Figure SMS_71
、/>
Figure SMS_64
Yaw angle +.>
Figure SMS_69
、/>
Figure SMS_72
Two-wheeled vehicle mass->
Figure SMS_74
、/>
Figure SMS_65
Two-position coordinates, marked by differential evolution method +.>
Figure SMS_70
、/>
Figure SMS_73
、/>
Figure SMS_75
、/>
Figure SMS_66
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: field intensity of vehicle after multi-vehicle superposition>
Figure SMS_68
；
An environmental field with lateral distance as a variable, input: vehicle coordinates, total lane width
Figure SMS_76
Risk coefficient of road marking and boundary risk field +.>
Figure SMS_77
、/>
Figure SMS_78
Potential field convergence factor->
Figure SMS_79
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: ambient field strength as a function of lateral distance +.>
Figure SMS_80
；
Environmental field, transfusion, with longitudinal distance as a variableThe method comprises the following steps: risk coefficient of road marking and boundary risk field
Figure SMS_81
Vehicle->
Figure SMS_82
Vector distance to the end of the acceleration lane +.>
Figure SMS_83
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: environmental field with longitudinal distance as a function>
Figure SMS_84
。
The beneficial effects of the invention are as follows:
(1) The invention is based on heterogeneous traffic flow differential modeling concept, which considers the difference of the perception, acting force and the like of CAV vehicles and HDV vehicles, builds a vehicle field model and perfects the action range constraint of the CAV vehicle field; the influence of individual difference of the HDV driver and driving environment on driving safety is considered, the comprehensive index of the psychological visibility of the driver environment is constructed, the psychological acting force of the HDV driver in mixed driving is deduced by means of the concept of psychological fields, and the vehicle field model of the HDV traditional vehicle is established. Combining the determined vehicle field model with an environment field model representing road boundaries, marked lines and other elements to form a safety field model for describing heterogeneous traffic flow running states and safety together;
(2) According to the invention, the environmental field model taking the longitudinal distance as a variable is added for the first time, so that the driving environmental field model is perfected, the influence of the environmental field on the running condition of the vehicle is truly reflected, the environmental field model accords with the actual lane change condition, the situation analysis shows that the vehicle field model can quantitatively analyze the driving risk of the vehicle, the contour line can visualize the field intensity distribution of the vehicle, the driving safety space of the vehicle is represented, and the interpretability of the heterogeneous traffic flow safety field model is enhanced. The dynamic driving risk of the vehicle can be quantitatively described, a following model and a lane changing model based on a safety potential field can be deduced according to a driving safety field, and a theoretical basis is made for researching the movement situation of the vehicle under heterogeneous traffic.
Drawings
Other objects and attainments together with a more complete understanding of the invention will become apparent and appreciated by referring to the following description taken in conjunction with the accompanying drawings. In the drawings:
FIG. 1 is a flow chart of the present invention;
figure 2 is a view of a CAV vehicle safety potential field model in the invention,
Figure SMS_85
；
figure 3 is a view of a CAV vehicle safety potential field model in the invention,
Figure SMS_86
；
figure 4 is a view of a CAV vehicle safety potential field model in the invention,
Figure SMS_87
；
figure 5 is a view of a CAV vehicle safety potential field model in the invention,
Figure SMS_88
；
figure 6 is a diagram of a HDV vehicle safety potential field model in accordance with the present invention,
FIG. 7 is a schematic illustration of a multi-car superimposed-on-fly field intensity contour line in accordance with the present invention;
FIG. 8 is a diagram of a model of the environmental safety potential field of a mainline lane in the present invention;
FIG. 9 is a diagram of an environmental safety potential field model of an acceleration lane in accordance with the present invention.
Detailed Description
The following detailed description is provided to enable those skilled in the art to better understand the technical scheme and advantages of the present invention, and is not intended to limit the scope of the present invention.
Referring to fig. 1: a heterogeneous traffic epidemic car security field modeling method comprises the following steps:
a1, quantifying the target vehicle attribute, wherein the formula is as follows:
Figure SMS_89
；
wherein:
Figure SMS_90
for the target vehicle->
Figure SMS_91
Equivalent mass of (a); />
Figure SMS_92
For the target vehicle->
Figure SMS_93
Is the actual mass of (3); />
Figure SMS_94
For the target vehicle->
Figure SMS_95
Is a speed of (2);
a2 determination of pseudo-ranges
The dangerous degree of the target vehicle also depends on the relative positions of the target vehicle and surrounding potential risk vehicles, and the Euclidean distance is referred to, so that the space actual distance of the potential risk vehicles approaching the target vehicle at different angles is corrected to obtain the pseudo distance
Figure SMS_96
The formula of (2) is:
Figure SMS_97
；
wherein:
Figure SMS_98
a critical threshold for a safe distance; />
Figure SMS_99
Is a coefficient to be determined in relation to speed; />
Figure SMS_100
And->
Figure SMS_101
Vehicle length and width, respectively, < >>
Figure SMS_102
Is a coefficient to be determined related to the road;
a3, coordinate transformation is carried out
Consider potentially risky vehicle roll
Figure SMS_103
When the vehicle runs in the axial direction, the vehicle body can deflect to a certain extent, the vehicle field model also deflects along with the deflection, and in a lane change scene, the formula is as follows:
Figure SMS_104
；
the counterclockwise direction is defined as positive, where:
Figure SMS_105
deflecting the course angle counterclockwise for the field model;
a4 determination of CAV vehicle field model
Describing CAV vehicle field strength as a function of the form of the takawa potential to describe short-range interactions between nuclei
Figure SMS_106
The formula is:
Figure SMS_107
；
Figure SMS_108
；
wherein:
Figure SMS_109
、/>
Figure SMS_110
all are undetermined coefficients; />
Figure SMS_111
Is the centroid of the target vehiclePosition coordinates of the space; />
Figure SMS_112
The current acceleration of the target vehicle; />
Figure SMS_113
For the spatial coordinates of a point around the target vehicle to the centroid of the vehicle +.>
Figure SMS_114
Is included in the plane of the first part;
Figure SMS_115
the value of the original coordinates after deflection is taken;
a5, calibrating parameters of CAV vehicle field model and restraining action range of the model
Since the potential field strength in the vehicle field model is limited by the range of action, only short-range forces are applied to surrounding vehicles within a short distance, and the forces vary with changes in vehicle speed, acceleration, distance, etc., when the target vehicle is beyond a certain range from the surrounding vehicle, the field strength effect on the surrounding vehicles is negligible. Using a differential evolution algorithm, and performing performance parameter calibration on undetermined coefficients in the established CAV vehicle field model by taking a difference value between a force range of a minimized CAV vehicle field and a vehicle head time interval in a strong following state as an objective function, so that field intensity distribution of the undetermined coefficients is more in accordance with a real running state of the vehicle, and taking the strong following vehicle head time interval as a critical action range of field intensity of the CAV vehicle field, so that a field intensity value of the CAV vehicle field can be obtained, and the action range of the CAV vehicle field is restrained;
step B1: restraining the reach of an HDV vehicle field
The HDV vehicle can keep a stable state when following a car ahead with a 2.7s headway, and the field intensity value of the vehicle is a critical value 0, namely:
Figure SMS_116
；
wherein:
Figure SMS_117
is standard toA time interval of the head; />
Figure SMS_118
The vehicle speed is the self-vehicle speed;
step B2: construction of driver environmental psychological bearing degree
Adopting a fuzzy theory, selecting typical characteristic factors which reflect the characteristics of a driver: determining the individual experience degree, the environment visibility degree and the psychological trust degree, inputting the characteristic value of the driver environment psychological tolerance assessment model, fuzzifying the characteristic value, obtaining a fuzzy quantity of the driver environment psychological tolerance through logic operation of a fuzzy rule, and converting the fuzzy quantity into an accurate specific numerical value by utilizing defuzzification, namely the driver environment psychological tolerance finally calculated by the driver environment psychological tolerance assessment model;
step B3: determining HDV vehicle field model
A CAV vehicle field is a physical quantity that varies in space-time, and thus a potential field essentially belongs to a physical field. In physics, due to the law of conservation of momentum, momentum is thought to exist in fields, and physical fields are truly present. And a field is a region of space where each location is subjected to a force, the field lines are typically used to represent the magnitude of the field at a location. Therefore, the physical field generated by the attribute of the people, the vehicles and the roads in the traffic system exists;
road users in traffic systems, i.e. drivers or pedestrians, are more concerned about obstacles or threats in their direct view in the direction of movement, which directly affect their safety judgment and movement-related behavior. When applying the safety field theory to road users, the interacting vehicle fields also generate a "repulsive force" that keeps the two interacting person-vehicle combinations in a safe physical gap between each other. However, this risk-symbolized "force" is not an actual physical force, since it does not follow newton's law, as the force and the reaction force are equal, but a psychological force, the effect of which is only manifested by the behaviour of the interacting road users. Thus, unlike the physical field, the theory of the safety field in the context of road users is a psychological field, which is only manifested by the effect of psychological effects of risk on the road user's behavior;
thus, the "effort" perceived by drivers and pedestrians is inherently linked to their perception. While it would normally repel interacting road users, the presence of a careless road user does not "feel" the risk forces and does not take corresponding action even though the situation may be critical. In fact, such behavior that does not take timely action based on forces in traffic interactions can lead to traffic collisions and accidents;
based on the discussion of the psychological acting force of the road user, unlike the physical potential field generated by the road environment and the CAV vehicle due to the self quality, speed and other attributes, the method fully considers the judgment difference of the individuality of the driver on the driving safety aiming at the modeling of the driver safety field, and establishes the potential field reflecting the psychological reaction of the driver as the vehicle field model of the traditional vehicle by taking the road visibility as an important index influencing the psychological state of the driver. The decision maker of the traditional vehicle generates psychological acting force for changing subjective cognition under the change of external traffic conditions, the field intensity of the traditional vehicle field represents the stress degree of a driver on conflict objects in the external traffic space position, the field intensity is the same as the direction of the psychological acting force, and the larger the driver perceives the external stress degree, the larger the psychological field force is, and the larger the field intensity value is;
the HDV vehicle field strength calculation formula derived from the discussion above is as follows:
Figure SMS_119
；
wherein:
Figure SMS_120
the field strength is HDV vehicle field strength; />
Figure SMS_121
Is an adjustment coefficient related to the extremum of the function; />
Figure SMS_122
The environmental psychological bearing degree is used for the driver;
in the running process of the vehicle, the road marking and the boundaries on two sides can generate a certain constraint effect on the running of the vehicle, and the vehicle is restricted to keep running in the middle of the lane line to maintain the continuity of the vehicle flow, so that the relatively stable head distance and the transverse distance are kept. Therefore, the risk of driving is small when the vehicle is driving in the center position of the lane, and the risk of deviating from the road boundary is greater than that of crossing the marked line. Assuming that the vehicle driving road is a double-lane, representing a road boundary line outside the lane, and being too close, the vehicle can be in substantial collision with the vehicle;
to avoid abrupt changes, an exponential function is selected to construct an environmental field with lateral distance as a variable, so that the value of the environmental field is infinite at the boundary to have a trend of preventing the vehicle from driving away, and the formula is as follows:
Figure SMS_123
；
wherein: assuming that the vehicle travel road is a two-lane,
Figure SMS_124
indicating a lane outside road boundary line, too close to which a substantial collision with the vehicle can occur; />
Figure SMS_125
The vehicle running coordinates; />
Figure SMS_126
And->
Figure SMS_127
Risk coefficients of the road marking and the boundary risk field are respectively; />
Figure SMS_128
Is the total width of the lane; />
Figure SMS_129
Is the potential field convergence coefficient;
aiming at the problem of lane number change, an environment field along the traffic flow driving direction is established for the lanes, a repulsive potential of a potential field theory is selected to construct a vehicle boundary environment field so as to quantitatively represent the influence of lane number reduction on the vehicle driving safety in the road condition change, and the expression is as follows:
Figure SMS_130
；
wherein:
Figure SMS_131
risk coefficients of the road marking and the boundary risk field; />
Figure SMS_132
For vehicle->
Figure SMS_133
Vector distance to the end of the acceleration lane;
drawing a CAV vehicle field, an HDV vehicle field, a multi-vehicle superposition vehicle field and an environment field through MATLAB;
referring to fig. 2-5, cav vehicle field, input: vehicle speed
Figure SMS_135
Vehicle acceleration->
Figure SMS_137
Yaw angle>
Figure SMS_139
Vehicle mass->
Figure SMS_136
Vehicle position coordinates, and ++specified by differential evolution method>
Figure SMS_140
、/>
Figure SMS_141
、/>
Figure SMS_142
、/>
Figure SMS_134
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: CAV vehicle field strength>
Figure SMS_138
；
Referring to fig. 6, hdv vehicle field, input: adjustment coefficients related to function extremum
Figure SMS_143
Psychological tolerance of driver's environment
Figure SMS_144
Vehicle speed->
Figure SMS_145
The method comprises the steps of carrying out a first treatment on the surface of the Vehicle position coordinates; and (3) outputting: HDV vehicle field strength->
Figure SMS_146
；
Referring to fig. 7, a multi-vehicle superposition vehicle field, input: speed of two vehicles
Figure SMS_148
、/>
Figure SMS_153
Acceleration of two vehicles->
Figure SMS_156
、/>
Figure SMS_147
Yaw angle +.>
Figure SMS_154
、/>
Figure SMS_157
Two-wheeled vehicle mass->
Figure SMS_159
、/>
Figure SMS_149
Two-position coordinates, marked by differential evolution method +.>
Figure SMS_151
、/>
Figure SMS_155
、/>
Figure SMS_158
、/>
Figure SMS_150
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: field intensity of vehicle after multi-vehicle superposition>
Figure SMS_152
；
Referring to fig. 8, an ambient field, in which a lateral distance is a variable, is input: vehicle coordinates, total lane width
Figure SMS_160
Risk coefficient of road marking and boundary risk field +.>
Figure SMS_161
、/>
Figure SMS_162
Potential field convergence factor->
Figure SMS_163
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: ambient field strength as a function of lateral distance
Figure SMS_164
；
Referring to fig. 9, an environmental field, which is a longitudinal distance, is input: risk coefficient of road marking and boundary risk field
Figure SMS_165
Vehicle->
Figure SMS_166
Vector distance to the end of the acceleration lane +.>
Figure SMS_167
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: in the longitudinal directionEnvironmental field with distance as variable->
Figure SMS_168
。
Further, the specific method of A5 is as follows:
a5.1 screening of vehicle Natural Driving trajectory data set
A5.2 pretreatment of data set
Statistical information about the lane distribution and speed distribution of the vehicle is obtained, however, there are various running states of the vehicle during running in the area, so the present patent sets screening conditions of the following state to process the data. Firstly, ensuring that front and rear vehicles run on the same lane; secondly, setting the distance between the front vehicle and the rear vehicle within the range of 2-150m, defaulting to vehicle queuing state rejection when the distance is too small, and defaulting to free flow state rejection when the distance is too large; and finally, setting the minimum following time of 10s, and defaulting to be unable to reach a stable state and rejecting when the duration is too large or too small. Acquiring a distribution interval of each following vehicle data by arranging and analyzing track data of a data set, and extracting effective following data by analyzing a distribution rule of the distribution interval;
a5.3 obtaining the headway in the Strong heel-and-heel state
A5.4, calibrating performance parameters of undetermined coefficients in the CAV vehicle field model by utilizing effective following data through a differential evolution algorithm; and taking the headway in a strong following state as a critical action range of the CAV vehicle field intensity to obtain a CAV vehicle field intensity value and restraining the action range of the CAV vehicle field.
Further, in the step B2, the specific construction method of the driver environmental psychological bearing degree is as follows:
b2.1 determining environmental psychological bearing index
Determining the individual experience degree, the environmental visibility and the psychological trust degree of the driver as environmental psychological bearing degree indexes;
b2.2 determining the discourse of each index
The driver environmental psychological bearing degree evaluation model is three-input single-output, an individual experience degree domain interval is set to be [0,5], a fuzzy subset is { Low, medium, high }, a domain interval of environmental visibility is [0,3], the fuzzy subset is { Near, middle, far }, a psychological trust degree domain is [ -3,3], and the fuzzy subset is { negative big NB, negative small NS, zero ZO, positive small PS, positive big PB };
b2.3 determining the membership function and membership of each index
Setting fuzzy input as individual experience degree of a driver and environment visibility, selecting a triangular form membership function, selecting a Gaussian form membership function when the fuzzy input is psychological trust degree, and obtaining membership of each index according to the membership function;
b2.4 establishing a driver environmental psychological bearing evaluation model rule table according to the membership degree
And B2.5, determining the relation among three input indexes by a rule table to obtain a bearing degree value in the range of [0,1], namely the environmental psychological bearing degree of the driver.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (9)

1. The modeling method for the security field of the heterogeneous traffic epidemic car is characterized by comprising the following steps of:
step A: construction of CAV vehicle field model
A1, quantifying target vehicle attributes;
a2, determining pseudo distance;
the dangerous degree of the target vehicle also depends on the relative positions of the target vehicle and surrounding potential risk vehicles, and the actual space distance of the potential risk vehicles, which are close to the target vehicle at different angles, is corrected by referring to Euclidean distance to obtain pseudo distance;
a3, coordinate transformation is carried out
Consider potentially risky vehicle roll
Figure QLYQS_1
When the vehicle runs in the axial direction, the vehicle body can deflect to a certain extent,the vehicle field model should also deflect accordingly, especially in lane change scenarios, the formula is:
Figure QLYQS_2
；
the counterclockwise direction is defined as positive, where:
Figure QLYQS_3
deflecting the course angle counterclockwise for the field model;
a4 determination of CAV vehicle field model
Describing CAV vehicle field strength as a function of the form of the takawa potential to describe short-range interactions between nuclei
Figure QLYQS_4
；
A5, calibrating parameters of CAV vehicle field model and restraining action range of the model
Using a differential evolution algorithm, and performing performance parameter calibration on undetermined coefficients in the established CAV vehicle field model by taking a difference value between a force range of a minimized CAV vehicle field and a vehicle head time interval in a strong following state as an objective function, so that field intensity distribution of the undetermined coefficients is more in accordance with a real running state of the vehicle, and taking the strong following vehicle head time interval as a critical action range of field intensity of the CAV vehicle field, so that a field intensity value of the CAV vehicle field can be obtained, and the action range of the CAV vehicle field is restrained;
and (B) step (B): construction of HDV vehicle field model
Step B1: restraining the reach of an HDV vehicle field
The HDV vehicle can keep a stable state when following a car ahead with a 2.7s headway, and the field intensity value of the vehicle is a critical value 0, namely:
Figure QLYQS_5
；
wherein:
Figure QLYQS_6
is the standard headway; />
Figure QLYQS_7
The vehicle speed is the self-vehicle speed;
step B2: construction of driver environmental psychological bearing degree
Adopting a fuzzy theory, selecting typical characteristic factors which reflect the characteristics of a driver: determining the individual experience degree, the environment visibility degree and the psychological trust degree, inputting the characteristic value of the driver environment psychological tolerance assessment model, fuzzifying the characteristic value, obtaining a fuzzy quantity of the driver environment psychological tolerance through logic operation of a fuzzy rule, and converting the fuzzy quantity into an accurate specific numerical value by utilizing defuzzification, namely the driver environment psychological tolerance finally calculated by the driver environment psychological tolerance assessment model;
step B3: determining HDV vehicle field model
Figure QLYQS_8
；
Wherein:
Figure QLYQS_9
the field strength is HDV vehicle field strength; />
Figure QLYQS_10
Is an adjustment coefficient related to the extremum of the function; />
Figure QLYQS_11
The environmental psychological bearing degree is used for the driver; />
Figure QLYQS_12
A critical threshold for a safe distance; />
Figure QLYQS_13
Is a coefficient to be determined in relation to speed; />
Figure QLYQS_14
Is pseudo distance;
step C: construction of an environmental field model with lateral distance as a variable
An exponential function is selected to construct an environment field taking the transverse distance as a variable, so that the value of the environment field is infinite at the boundary to have a trend of preventing the vehicle from driving away;
step D: construction of an environmental field model with longitudinal distance as a variable
Selecting repulsive potential of potential field theory to construct a vehicle boundary environment field so as to quantitatively represent the influence of reduction of the number of lanes in road condition change on vehicle driving safety, and establishing an environment field along the driving direction of the traffic flow for the lanes;
step E: drawing field intensity diagram of driving safety field
And drawing a CAV vehicle field, an HDV vehicle field, a multi-vehicle superposition vehicle field and an environment field through MATLAB.
2. The heterogeneous transportation epidemic safety field modeling method according to claim 1, wherein in the A1, the target vehicle
Figure QLYQS_15
Equivalent mass of->
Figure QLYQS_16
The formula of (2) is:
Figure QLYQS_17
；
wherein:
Figure QLYQS_18
for the target vehicle->
Figure QLYQS_19
Equivalent mass of (a); />
Figure QLYQS_20
For the target vehicle->
Figure QLYQS_21
Is the actual mass of (3); />
Figure QLYQS_22
For the target vehicle->
Figure QLYQS_23
Is a function of the speed of the machine.
3. The method for modeling security of heterogeneous transportation vehicles according to claim 2, wherein in A2, the pseudo distance is calculated
Figure QLYQS_24
The formula of (2) is:
Figure QLYQS_25
；
wherein:
Figure QLYQS_26
and->
Figure QLYQS_27
Vehicle length and width, respectively, < >>
Figure QLYQS_28
Is a predetermined coefficient related to the road.
4. A method for modeling security of a heterogeneous transportation epidemic vehicle according to claim 3, wherein in A4, CAV vehicle field intensity is
Figure QLYQS_29
The formula is:
Figure QLYQS_30
；
Figure QLYQS_31
；
wherein:
Figure QLYQS_32
、/>
Figure QLYQS_33
all are undetermined coefficients; />
Figure QLYQS_34
The position coordinates of the space where the mass center of the target vehicle is located; />
Figure QLYQS_35
The current acceleration of the target vehicle; />
Figure QLYQS_36
For the spatial coordinates of a point around the target vehicle to the centroid of the vehicle +.>
Figure QLYQS_37
Is included in the plane of the first part; />
Figure QLYQS_38
The value of the original coordinate after deflection is taken.
5. The heterogeneous traffic epidemic car security field modeling method according to claim 4, wherein in the A5, the specific method for calibrating the CAV car field model parameters is as follows:
a5.1 screening of vehicle Natural Driving trajectory data set
A5.2 pretreatment of data set
According to screening conditions of the following state, ensuring that front and rear vehicles run on the same lane, setting the distance between the front and rear vehicles within the range of 2-150m, defaulting to vehicle queuing state rejection when the distance is too small, and defaulting to free flow state rejection when the distance is too large; setting a minimum following time of 10s, and defaulting to a state that a stable state cannot be achieved and rejecting when the duration is too large or too small; acquiring a distribution interval of each following vehicle data by arranging and analyzing track data of a data set, and extracting effective following data by analyzing a distribution rule of the distribution interval;
a5.3 obtaining the headway in the Strong heel-and-heel state
A5.4, calibrating performance parameters of undetermined coefficients in the CAV vehicle field model by utilizing effective following data through a differential evolution algorithm; and taking the headway in a strong following state as a critical action range of the CAV vehicle field intensity to obtain a CAV vehicle field intensity value and restraining the action range of the CAV vehicle field.
6. The heterogeneous traffic epidemic car security field modeling method according to claim 5, wherein the specific construction method of the driver environmental psychological bearing degree in B2 is as follows:
b2.1 determining environmental psychological bearing index
Determining the individual experience degree, the environmental visibility and the psychological trust degree of the driver as environmental psychological bearing degree indexes;
b2.2 determining the discourse of each index
The driver environmental psychological bearing degree evaluation model is three-input single-output, an individual experience degree domain interval is set to be [0,5], a fuzzy subset is { Low, medium, high }, a domain interval of environmental visibility is [0,3], the fuzzy subset is { Near, middle, far }, a psychological trust degree domain is [ -3,3], and the fuzzy subset is { negative big NB, negative small NS, zero ZO, positive small PS, positive big PB };
b2.3 determining the membership function and membership of each index
Setting fuzzy input as individual experience degree of a driver and environment visibility, selecting a triangular form membership function, selecting a Gaussian form membership function when the fuzzy input is psychological trust degree, and obtaining membership of each index according to the membership function;
b2.4 establishing a driver environmental psychological bearing evaluation model rule table according to the membership degree
And B2.5, determining the relation among three input indexes by a rule table to obtain a bearing degree value in the range of [0,1], namely the environmental psychological bearing degree of the driver.
7. The method for modeling security of heterogeneous transportation vehicles according to claim 6, wherein in the step C, the environmental field model formula using the lateral distance as a variable is:
Figure QLYQS_39
；
wherein: assuming that the vehicle travel road is a two-lane,
Figure QLYQS_40
indicating a lane outside road boundary line, too close to which a substantial collision with the vehicle can occur; />
Figure QLYQS_41
The vehicle running coordinates; />
Figure QLYQS_42
And->
Figure QLYQS_43
Risk coefficients of the road marking and the boundary risk field are respectively; />
Figure QLYQS_44
Is the total width of the lane; />
Figure QLYQS_45
Is the potential field convergence coefficient.
8. The method for modeling security of heterogeneous transportation vehicles according to claim 7, wherein in the step D, the environmental field model formula using the longitudinal distance as a variable is:
Figure QLYQS_46
；
wherein:
Figure QLYQS_47
risk coefficients of the road marking and the boundary risk field; />
Figure QLYQS_48
For vehicle->
Figure QLYQS_49
Vector distance to the end of the acceleration lane.
9. The method for modeling a security field of a heterogeneous transportation epidemic vehicle according to claim 8, wherein in the step E, the specific process of the field intensity diagram of the security field of the transportation is as follows:
CAV vehicle field, input: vehicle speed
Figure QLYQS_50
Vehicle acceleration->
Figure QLYQS_53
Yaw angle>
Figure QLYQS_56
Vehicle mass->
Figure QLYQS_51
Vehicle position coordinates, and ++specified by differential evolution method>
Figure QLYQS_55
、/>
Figure QLYQS_57
、/>
Figure QLYQS_58
、/>
Figure QLYQS_52
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: CAV vehicle field strength>
Figure QLYQS_54
；
HDV vehicle field, input: adjustment coefficients related to function extremum
Figure QLYQS_59
Driver mental tolerance->
Figure QLYQS_60
Speed of vehicle
Figure QLYQS_61
The method comprises the steps of carrying out a first treatment on the surface of the Vehicle position coordinates; and (3) outputting: HDV vehicle field strength->
Figure QLYQS_62
；
Multiple vehicle superposition vehicle field, input: speed of two vehicles
Figure QLYQS_64
、/>
Figure QLYQS_71
Acceleration of two vehicles->
Figure QLYQS_73
、/>
Figure QLYQS_65
Yaw angle +.>
Figure QLYQS_70
、/>
Figure QLYQS_74
Two-wheeled vehicle mass->
Figure QLYQS_75
、/>
Figure QLYQS_63
Two-position coordinates and differential evolution methodMarked->
Figure QLYQS_68
、/>
Figure QLYQS_69
、/>
Figure QLYQS_72
、/>
Figure QLYQS_66
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: field intensity of vehicle after multi-vehicle superposition>
Figure QLYQS_67
；
An environmental field with lateral distance as a variable, input: vehicle coordinates, total lane width
Figure QLYQS_76
Risk coefficient of road marking and boundary risk field +.>
Figure QLYQS_77
、/>
Figure QLYQS_78
Potential field convergence factor->
Figure QLYQS_79
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: ambient field strength as a function of lateral distance +.>
Figure QLYQS_80
；
Environmental field with longitudinal distance as variable, input: risk coefficient of road marking and boundary risk field
Figure QLYQS_81
Vehicle->
Figure QLYQS_82
Vector distance to the end of the acceleration lane +.>
Figure QLYQS_83
The method comprises the steps of carrying out a first treatment on the surface of the And (3) outputting: environmental field with longitudinal distance as a function>
Figure QLYQS_84
。
CN202310593492.9A 2023-05-25 2023-05-25 Heterogeneous traffic epidemic car security field modeling method Active CN116341288B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310593492.9A CN116341288B (en) 2023-05-25 2023-05-25 Heterogeneous traffic epidemic car security field modeling method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310593492.9A CN116341288B (en) 2023-05-25 2023-05-25 Heterogeneous traffic epidemic car security field modeling method

Publications (2)

Publication Number Publication Date
CN116341288A true CN116341288A (en) 2023-06-27
CN116341288B CN116341288B (en) 2023-09-05

Family

ID=86882658

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310593492.9A Active CN116341288B (en) 2023-05-25 2023-05-25 Heterogeneous traffic epidemic car security field modeling method

Country Status (1)

Country Link
CN (1) CN116341288B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117125083A (en) * 2023-10-27 2023-11-28 青岛理工大学 Vehicle following behavior risk quantification method considering driving style inclination
CN118629219A (en) * 2024-08-15 2024-09-10 吉林大学 Vehicle lane change decision and track planning interactive coupling method based on security field

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110455552A (en) * 2019-08-08 2019-11-15 上海测迅汽车科技有限公司 Compound scene test method in a kind of intelligent vehicle room
CN112224211A (en) * 2020-10-19 2021-01-15 中交第一公路勘察设计研究院有限公司 Driving simulation system based on multi-autonomous-body traffic flow
CN114771570A (en) * 2022-04-24 2022-07-22 北京京东乾石科技有限公司 Control method and device for automatic driving vehicle
CN114789728A (en) * 2022-05-27 2022-07-26 长沙理工大学 Method for estimating state change of following vehicle by considering transverse interference under two-lane driving
CN114889625A (en) * 2022-06-02 2022-08-12 东南大学 Method for establishing vehicle following model at road curve based on safe potential field theory
CN115169908A (en) * 2022-07-14 2022-10-11 吉林大学 Vehicle driving state safety assessment method based on perceptual risk field in networking environment
CN115238958A (en) * 2022-06-20 2022-10-25 清华大学 Dangerous event chain extraction method and system based on complex traffic scene
CN115503700A (en) * 2022-11-07 2022-12-23 江苏大学 Active collision avoidance method based on multivariate risk fusion potential field and vehicle-mounted electronic equipment
CN115620530A (en) * 2022-09-19 2023-01-17 哈尔滨工业大学 Automatic driving takeover risk assessment model considering driving risk field and construction method thereof
CN115855075A (en) * 2022-11-28 2023-03-28 南湖实验室 Vehicle layered multi-module fusion accurate positioning method under satellite system rejection condition

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110455552A (en) * 2019-08-08 2019-11-15 上海测迅汽车科技有限公司 Compound scene test method in a kind of intelligent vehicle room
CN112224211A (en) * 2020-10-19 2021-01-15 中交第一公路勘察设计研究院有限公司 Driving simulation system based on multi-autonomous-body traffic flow
CN114771570A (en) * 2022-04-24 2022-07-22 北京京东乾石科技有限公司 Control method and device for automatic driving vehicle
CN114789728A (en) * 2022-05-27 2022-07-26 长沙理工大学 Method for estimating state change of following vehicle by considering transverse interference under two-lane driving
CN114889625A (en) * 2022-06-02 2022-08-12 东南大学 Method for establishing vehicle following model at road curve based on safe potential field theory
CN115238958A (en) * 2022-06-20 2022-10-25 清华大学 Dangerous event chain extraction method and system based on complex traffic scene
CN115169908A (en) * 2022-07-14 2022-10-11 吉林大学 Vehicle driving state safety assessment method based on perceptual risk field in networking environment
CN115620530A (en) * 2022-09-19 2023-01-17 哈尔滨工业大学 Automatic driving takeover risk assessment model considering driving risk field and construction method thereof
CN115503700A (en) * 2022-11-07 2022-12-23 江苏大学 Active collision avoidance method based on multivariate risk fusion potential field and vehicle-mounted electronic equipment
CN115855075A (en) * 2022-11-28 2023-03-28 南湖实验室 Vehicle layered multi-module fusion accurate positioning method under satellite system rejection condition

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
QIANGQIANG GUO ET AL.: "Mixed traffic flow of human driven vehicles and automated vehicles on dynamic transportation networks", TRANSPORTATION RESEARCH PART C, pages 1 - 30 *
侯贻栋;赵炜华;魏朗;陈昊;: "驾驶人空间距离判识规律心理学分析", 长安大学学报(自然科学版), no. 03, pages 86 - 90 *
单肖年 等: "智能网联环境下多车道异质交通流建模与仿真", 交通运输系统工程与信息, vol. 22, no. 6, pages 74 - 84 *
吴兵 等: "多前车影响的智能网联车辆纵向控制模型", 交通运输工程学报, vol. 20, no. 2, pages 184 - 194 *
陈康 等: "基于心理场理论的高速公路弯道路段车辆跟驰建模与仿真", 交通信息与安全, vol. 34, no. 6, pages 77 - 82 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117125083A (en) * 2023-10-27 2023-11-28 青岛理工大学 Vehicle following behavior risk quantification method considering driving style inclination
CN117125083B (en) * 2023-10-27 2024-01-05 青岛理工大学 Vehicle following behavior risk quantification method considering driving style inclination
CN118629219A (en) * 2024-08-15 2024-09-10 吉林大学 Vehicle lane change decision and track planning interactive coupling method based on security field

Also Published As

Publication number Publication date
CN116341288B (en) 2023-09-05

Similar Documents

Publication Publication Date Title
CN116341288B (en) Heterogeneous traffic epidemic car security field modeling method
WO2021077725A1 (en) System and method for predicting motion state of surrounding vehicle based on driving intention
CN108313054B (en) The autonomous lane-change decision-making technique of automatic Pilot and device and automatic driving vehicle
Moujahid et al. Machine learning techniques in ADAS: A review
CN110027553B (en) Anti-collision control method based on deep reinforcement learning
Gidado et al. A survey on deep learning for steering angle prediction in autonomous vehicles
Huang et al. A probabilistic risk assessment framework considering lane-changing behavior interaction
CN113946943B (en) Human-vehicle-road micro traffic system modeling and risk identification method and device
Wang et al. Evaluation of lane departure correction systems using a regenerative stochastic driver model
Ni et al. Multivehicle cooperative lane change control strategy for intelligent connected vehicle
Sun et al. Modelling and simulation of highly mixed traffic flow on two-lane two-way urban streets
CN115601954B (en) Lane change judgment method, device, equipment and medium for intelligent networked fleet
Wang et al. Vehicle Trajectory Prediction by Knowledge‐Driven LSTM Network in Urban Environments
Wang et al. Cognitive competence improvement for autonomous vehicles: A lane change identification model for distant preceding vehicles
Tang et al. Driving capability-based transition strategy for cooperative driving: from manual to automatic
CN118238847B (en) Autonomous lane change decision planning method and system adaptive to different driving styles and road surface environments
Sun et al. Driving safety field modeling focused on heterogeneous traffic flows and cooperative control strategy in highway merging zone
Ren et al. Self-learned intelligence for integrated decision and control of automated vehicles at signalized intersections
CN113306558B (en) Lane changing decision method and system based on lane changing interaction intention
Wang et al. A deep reinforcement learning-based approach for autonomous lane-changing velocity control in mixed flow of vehicle group level
Azam et al. Driving behaviour modelling in the context of heterogeneous traffic and poor lane discipline conditions: the state of the art and beyond
Gao et al. Discretionary cut-in driving behavior risk assessment based on naturalistic driving data
Griesbach et al. Prediction of lane change by echo state networks
Wu et al. An integrated decision and motion planning framework for automated driving on highway
Suh et al. Stochastic predictive control based motion planning for lane change decision using a vehicle traffic simulator

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant