CN113538940B - Real-time optimal lane selection method suitable for vehicle-road cooperative environment - Google Patents
Real-time optimal lane selection method suitable for vehicle-road cooperative environment Download PDFInfo
- Publication number
- CN113538940B CN113538940B CN202110770608.2A CN202110770608A CN113538940B CN 113538940 B CN113538940 B CN 113538940B CN 202110770608 A CN202110770608 A CN 202110770608A CN 113538940 B CN113538940 B CN 113538940B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- lane
- time
- individual
- selection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/042—Backward inferencing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- 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
-
- 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
Abstract
The invention discloses a real-time optimal lane selection method suitable for a vehicle-road cooperative environment, which specifically comprises the following steps: s1, establishing a traffic model under the vehicle-road cooperative environment as a network evolution game model with state-dependent time lag by using a mixed value logic function, and deriving a strategy evolution equation by using a matrix method; s2, acquiring vehicle data of each lane within limited time by using road data collecting equipment; s3, taking the vehicle speed as income, taking the selection of lanes as a strategy and taking the decision made by all the current vehicles as situation; when information time lag exists, analyzing the influence of different situations on the speed of the vehicle, finding out a strategy for realizing the maximum benefit of the individual vehicle, and further making a strategy updating rule of optimal response; s4, constructing a state transition matrix H by using a matrix method; and analyzing the characteristics, judging whether the traffic system is stable, if so, solving a mean cross point strategy, and finding out a criterion for convergence of the traffic system, namely the optimal path planning.
Description
Technical Field
The invention relates to the field of intelligent traffic systems, in particular to a real-time optimal lane selection method suitable for a vehicle-road cooperative environment.
Background
In the traditional traffic system, the random lane change of vehicles is an important reason for traffic jam and accidents. Most drivers can only rely on self experience and visual judgment to achieve the purpose of lane changing, unpredictable risks exist behind the decision, especially when traffic is congested, blind lane changing further aggravates the traffic jam, and even accidents are caused. With the development of perception and communication technology, a vehicle-road cooperative system provides an effective way for solving the problems, and the design of a reliable lane decision method is the key for the effective operation of the vehicle-road cooperative system.
The game theory provides a research method and an analysis means for solving the problem of competition tendency. Lane changes between vehicles can be considered a non-cooperative gambling activity. The driver in the lane selection process is regarded as a game player in the game, the win-win is realized through the game process, and the participants can make optimal decision by combining with the expected income of the participants under the mutual constraint of the external conditions. The method not only accords with the driving psychology of the driver pursuing the maximum benefit, but also enables the lane changing decision to be more reasonable, so that the acceptability of the driver to the game cooperation strategy is high, the cooperation will is strong, and the method is very suitable for solving the problem of the optimal lane selection of the vehicle by adopting a game theory method. In an actual traffic network, the topological relation among vehicles has an important influence on the evolution of the situation, and especially the influence of the lane selection of the vehicles in the neighborhood on each other is more direct; meanwhile, when the vehicle receives the information of the data acquisition equipment, time lag occurs, so that the judgment of the vehicle on the lane selection of the neighbor vehicle is influenced, the time lag has an inseparable relationship with the situation, and the influence of state-dependent time lag on the optimal lane selection has to be considered.
Disclosure of Invention
The invention aims to provide an optimal lane selection method under a vehicle-road cooperative environment, which is used for guiding a driver to make optimal lane selection, so that the road smoothness is improved and traffic jam is relieved on the premise of ensuring traffic safety, and the method has very important social value and engineering significance.
In order to solve the technical problems, the invention adopts the technical scheme that: the real-time optimal lane selection method suitable for the vehicle-road cooperative environment specifically comprises the following steps:
s1: establishing a traffic model under the vehicle-road cooperative environment as a network evolution game model with state-dependent time lag by using a mixed value logic function, and deriving a strategy evolution equation by using a matrix method;
s2: acquiring historical vehicle data of each lane within limited time by using road data collection equipment;
s3: taking the vehicle speed as the income, taking the selection of lanes as a strategy, and taking the decision made by all the vehicles at present as the situation; when information time lag exists, analyzing the influence of different situations on the speed of the vehicle, further finding out a strategy for enabling the individual vehicle to realize the maximum benefit, and further making a strategy updating rule of optimal response;
s4: constructing a state transition matrix H by using a matrix method; and analyzing the characteristics of the state transition matrix H, judging whether the traffic system is stable, if so, solving a mean transverse point strategy, finding out a criterion for convergence of the traffic system, and further obtaining a stable point, namely the sought optimal path plan.
By adopting the technical scheme, mathematical modeling is firstly carried out according to the real-time traffic network condition; then, according to the income parameter selection of the vehicle owner, the acquisition work of model data is carried out by means of the vehicle-road cooperation technology; then, according to the income parameter selection of the vehicle owner, a strategy updating rule is made; further judging whether the system is stable, namely whether the game can converge to the equilibrium point, and if the system is stable, directly solving an equilibrium point strategy, namely an optimal strategy; if the system is unstable, the participant strategy is intervened through a system controller to realize system stability, then the equilibrium point strategy is solved, and finally an optimal lane selection suggestion is given; the method for modeling the traffic model in the vehicle-road cooperative environment by using the mathematical formula obtains a network evolution game model with state-dependent time lag to simulate the vehicle-road cooperative traffic model so as to solve the problem of optimal lane selection of individual vehicles, and on one hand, helps a vehicle owner to select a scientific optimal lane and maximize the benefit of the vehicle owner; on the other hand, the existence of the global optimal solution is guaranteed, so that the pressure of the whole traffic jam is relieved, and the core of the method is to solve the game problem of the individual vehicles and the adjacent vehicles by utilizing the thought of a game theory, find out the global optimal solution and realize the optimal control.
As a preferred embodiment of the present invention, the method further includes step S5: and if the traffic system is unstable, namely convergence cannot be realized, designing a state feedback controller to perform route optimization control on the individual vehicle so as to ensure that the traffic system is converged.
As a preferred embodiment of the present invention, the step S1 specifically includes:
s11: establishing a traffic model under a vehicle-road cooperative environment into a network evolution game model G with state-dependent time lag by using a mixed value logic networkd；
Assume that there are N vehicles forming a set N ═ {1, 2.., N }, where the set of lanes in which an individual vehicle i may be allowed to travel is Si＝{k1，k2，...，ki}，i＝1，2，...，n，S＝S1×S2×…×SnIs a set of situations, ci: s → a set of real numbers, where i ═ 1, 2., n is the revenue function of individual vehicle i; the topology between vehicles is represented by a connection graph (N, E), where N is the set of points representing the vehicle and E is the set of edges; each side is connected with two vehicles with information interaction relation, and because the influence of adjacent vehicles on each other is the largest, the influence of one-step neighbor vehicles on each other is only considered in lane selection; n is a radical ofiRepresenting a neighborhood of individual vehicle i that contains all neighboring vehicles of vehicle i; game G of vehicle i and vehicle j in its neighborhoodij(ii) a Using mixed-value logic function fiA policy update rule representing an individual vehicle i, where i ═ 1, 2i：S→{0，1，...，λiDenotes the time lag, τ, of the individual vehicle i at the time of receiving the informationiIs a mixed value logic function, the networked vehicle GdThe dynamic model of (2) can be built into the following patterns:
xi(t+1)＝fi(xj(t-τi(x(t)))|j∈Ni，t＝0，1，...，)；
wherein xi(t) is the lane selection for individual vehicle i at time t, x (t) e S is the situation at time t, τi(x (t)) is the state-dependent time lag of the individual vehicle i at time t, ci(t) is the average profit of the neighbor vehicle of the individual vehicle i when playing the game, namely the speed of the vehicle i at the time t, and the formula is as follows:
as a preferred embodiment of the present invention, the road data collection device in step S2 includes a vehicle speed sensor, a roadside sensor and a high-precision map; the step S2 specifically includes: based on the vehicle running speed acquired by the vehicle speed sensor, taking the vehicle running speed as a profit quantification index of the individual vehicle; the lane position of the individual vehicle is positioned based on a positioning system, a road side sensor and a high-precision map of the intelligent vehicle; and (3) providing data support for the optimal decision of the individual vehicle by combining the historical lane selection condition of each individual vehicle in the t steps stored by the intelligent vehicle-road cooperative system and the information time lag of the individual vehicle in the data receiving process.
As a preferred embodiment of the present invention, the step S3 specifically includes:
s31: according to the revenue function c of the individual vehicle ii(t) searching for the time t, selecting the lane with the maximum profit for the neighbor vehicle j, and forming a set Pi(t)：
Wherein i represents a variable of the player, whose value range is {1, 2.., n }, SiIs a set of policies for player i, xi(t) is a strategy variable selected by player i at time t;
s32: suppose Pi(t)＝{s1，s2，...，sr}∈SiIf the lane selection for the vehicle is already the best lane, then the next time the vehicle will remain traveling in that lane; if the lane of the vehicle is not the best lane for the vehicles in its neighborhood, then the set P is selected at the next timei(t) designing a lane with the minimum subscript, namely designing a short potential optimal response strategy updating rule as follows:
as a preferred embodiment of the present invention, the step S4 specifically includes:
s41: identifying the real number set as vector set, converting it into algebraic form z (t +1) ═ Hz (t) and λ ═ max { λ } by using matrix half tensor product method1，λ2，...，λn}; the situation of the lambda +1 step is fused into z (t) by a dimension expansion method, so that the above formula becomes a standard logic dynamic system; the state transition matrix H absorbs the original time delay information, reflects the reachability of the situation transition, reveals the property of the game, calculates and observes and calculates the state transition matrix
S42: judging whether the traffic system is stable or not, wherein the judgment formula is as follows:
wherein Rows(H) Represents the s-th Row, of the matrix H-s(H) Denotes the rows of the matrix H except the s-th row, k ═ k1×k2×...×kn(ii) a If the integer s is existed, the traffic system is stable, and if the integer s is not existed, the traffic system is unstable;
s43: if the traffic system is stable, the traffic game is solved to be converged to an equilibrium point
Will be provided with
Performing a unique decomposition yields:
i.e. when the individual vehicle i selects lane SiAt the same time, its speedThe optimal degree can be realized, and the lane S isiMost unobstructed and stable, with i 1, 2.
As a preferred technical solution of the present invention, the step S5 specifically includes: if the traffic system is unstable, namely the traffic game cannot be converged, intervention is applied to the selection of the individual paths through the system state feedback controller, and lane selection of individual vehicles is controlled, so that the traffic system reaches a stable state; and returning to the step S43 to obtain the optimal lane selection.
As a preferred embodiment of the present invention, the specific method of the system state feedback controller in step S5 to intervene in selecting the individual path includes the following three methods:
s5-1: restricting the vehicle's selection set of lanes, i.e. the set of lanes for which vehicle i can be allowed to be
Is a set SiTo narrow the degree of freedom of the vehicle;
s5-2: the selection of the lane has priority, the original mode of uniformly and mixedly selecting the lane is broken through, and the lane different from the adjacent vehicle is preferentially selected, so that the system achieves dynamic balance;
s5-3: one of the vehicles is directly controlled and subjected to direct lane selection intervention, so that the lane selection intervention influences the behavior of the neighbor vehicles and further influences the selection of the global vehicle, namely:
where t represents a time node, NiRepresents the neighbor set of player i, the value range of player i is {1, 2., n }, u (t) represents the inputtable external control variable, g1Is a artificially designed, fully controllable mixed-value logic function, fiIs the policy of player iThe rules are updated slightly.
As a preferable technical scheme of the invention, the roadside sensor comprises a camera and a millimeter wave radar.
The invention has the beneficial effects that: the optimal lane selection method under the vehicle-road cooperative environment is provided, and is used for guiding a driver to make scientific optimal lane selection, so that the road smoothness is improved on the premise of ensuring traffic safety, traffic jam is relieved, and the method has great social value and engineering significance.
Drawings
FIG. 1 is a schematic overall flow chart of a real-time optimal lane selection method for use in a vehicle-road cooperative environment according to the present invention;
FIG. 2 is a schematic diagram of lane simulation in the real-time optimal lane selection method for the vehicle-road cooperative environment according to the present invention;
FIG. 3 is a schematic diagram of a vehicle topological relation in an embodiment of a real-time optimal lane selection method for use in a vehicle-road coordination environment according to the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the drawings of the embodiments of the present invention.
Example (b): as shown in fig. 1, the real-time optimal lane selection method applied to the vehicle-road cooperative environment specifically includes the following steps:
s1: establishing a traffic model under the vehicle-road cooperative environment as a network evolution game model with state-dependent time lag by using a mixed value logic function, and deriving a strategy evolution equation by using a matrix method;
the step S1 includes the following steps:
s11: establishing a traffic model under a vehicle-road cooperative environment into a network evolution game model G with state-dependent time lag by using a mixed value logic networkd；
Assume that there are N vehicles forming a set N ═ {1, 2.., N }, where the set of lanes in which an individual vehicle i may be allowed to travel is Si＝{k1，k2，...，ki}，i＝1，2，...，n，S＝S1×S2×…×SnIs a set of situations, ci: s → a set of real numbers, where i ═ 1, 2., n is the revenue function of individual vehicle i; the topology between vehicles is represented by a connection graph (N, E), where N is the set of points representing the vehicle and E is the set of edges; each side is connected with two vehicles with information interaction relation, and because the influence of adjacent vehicles on each other is the largest, the influence of one-step neighbor vehicles on each other is only considered in lane selection; n is a radical ofiRepresenting a neighborhood of individual vehicle i that contains all neighboring vehicles of vehicle i; game G of vehicle i and vehicle j in its neighborhoodij(ii) a Using mixed-value logic function fiA policy update rule representing an individual vehicle i, where i ═ 1, 2i：S→{0，1，...，λiDenotes the time lag, τ, of the individual vehicle i at the time of receiving the informationiIs a mixed value logic function, the networked vehicle GdThe dynamic model of (2) can be built into the following patterns:
xi(t+1)＝fi(xj(t-τi(x(t)))|j∈Ni，t＝0，1，...，)；
wherein xi(t) is the lane selection for individual vehicle i at time t, x (t) e S is the situation at time t, τi(x (t)) is the state-dependent time lag of the individual vehicle i at time t, ci(t) is the average profit of the neighbor vehicle of the individual vehicle i when playing the game, namely the speed of the vehicle i at the time t, and the formula is as follows:
s2: acquiring historical vehicle data of each lane within limited time by using road data collection equipment;
the road data collection device in the step S2 includes a vehicle speed sensor, a roadside sensor and a high-precision map; the step S2 specifically includes: based on the vehicle running speed acquired by the vehicle speed sensor, taking the vehicle running speed as a profit quantification index of the individual vehicle; positioning the lane position of the individual vehicle based on a positioning system, a road side sensor and a high-precision map of the intelligent vehicle; the road side sensor comprises a camera and a millimeter wave radar; the historical lane selection condition of each individual vehicle in the t steps stored by the intelligent vehicle-road coordination system is combined with the information time lag of the individual vehicle when receiving data, so that data support is provided for the optimal decision of the individual vehicle;
s3: taking the vehicle speed as the income, taking the selection of lanes as a strategy, and taking the decision made by all the vehicles at present as the situation; when information time lag exists, analyzing the influence of different situations on the speed of the vehicle, further finding out a strategy for enabling the individual vehicle to realize the maximum benefit, and further making a strategy updating rule of optimal response;
the step S3 specifically includes:
s31: according to the revenue function c of the individual vehicle ii(t) searching for the time t, selecting the lane with the maximum profit for the neighbor vehicle j, and forming a set Pi(t)：
Wherein i represents a variable of the player, whose value range is {1, 2.., n }, SiIs a set of policies for player i, xi(t) is a strategy variable selected by player i at time t;
s32: suppose Pi(t)＝{s1，s2，...，sr}∈SiIf the lane selection for the vehicle is already the best lane, then the next time the vehicle will remain traveling in that lane; if the lane of the vehicle is not the best lane for the vehicles in its neighborhood, then the set P is selected at the next timei(t) designing a lane with the minimum subscript, namely designing a short potential optimal response strategy updating rule as follows:
s4: constructing a state transition matrix H by using a matrix method; and analyzing the characteristics of the state transition matrix H, judging whether the traffic system is stable, if so, solving a mean transverse point strategy, finding out a criterion for convergence of the traffic system, and further obtaining a stable point, namely the sought optimal path plan.
The step S4 specifically includes:
s41: identifying the real number set as vector set, converting it into algebraic form z (t +1) ═ Hz (t) and λ ═ max { λ } by using matrix half tensor product method1，λ2，...，λn}., respectively; the situation of the lambda +1 step is fused into z (t) by a dimension expansion method, so that the above formula becomes a standard logic dynamic system; the state transition matrix H absorbs the original time delay information, reflects the reachability of the situation transition, reveals the property of the game, calculates and observes and calculates the state transition matrix
Judging whether the traffic system is stable or not, wherein the judgment formula is as follows:
wherein Rows(H) Represents the s-th Row, of the matrix H-s(H) Denotes the rows of the matrix H except the s-th row, k ═ k1×k2×...×kn(ii) a If the integer s is existed, the traffic system is stable, and if the integer s is not existed, the traffic system is unstable;
s43: if the traffic system is stable, the traffic game is solved to be converged to an equilibrium point
Will be provided with
Performing a unique decomposition yields:
i.e. when the individual vehicle i selects lane siAt the time, the speed of the lane S can be optimizediMost unobstructed and stable, with i 1, 2.
S5: if the traffic system is unstable, namely convergence cannot be achieved, designing a state feedback controller to perform route optimization control on the individual vehicle so that the traffic system is converged;
the step S5 specifically includes: if the traffic system is unstable, namely the traffic game cannot be converged, intervention is applied to the selection of the individual paths through the system state feedback controller, and lane selection of individual vehicles is controlled, so that the traffic system reaches a stable state; returning to the step S43 to obtain the optimal lane selection;
the specific method for the system state feedback controller in step S5 to intervene in the selection of the individual path includes the following three methods:
s5-1: restricting the vehicle's selection set of lanes, i.e. the set of lanes that vehicle i can be allowed to be
Is a set SiTo narrow the degree of freedom of the vehicle;
s5-2: the selection of the lane has priority, the original lane mode of uniformly mixing and selecting is broken through, and the lane different from the adjacent vehicle is preferentially selected, so that the system achieves dynamic balance;
s5-3: one of the vehicles is directly controlled and subjected to direct lane selection intervention, so that the lane selection intervention influences the behavior of the neighbor vehicles and further influences the selection of the global vehicle, namely:
where u (t) represents the external control variables that can be entered，g1Is a fully controllable mixed value logic function which is designed artificially; the matrix half tensor product method adopts the following documents of Yang Zheng, Changxi Li, and Jun-e Feng, Modeling and Dynamics of network evolution with Switched Time Delay, DOI: 10.1109/TCNS.2021.3084548.
FIG. 2 depicts a simple gaming diagram having a network architecture; by adopting the method, without loss of generality, only three vehicles are considered, and a fire truck p is assumed1Private car p2Bus p3The three vehicles are driven with r1，r2，r3On the highway with three lanes, the fire engine can run on any lane, and the private car only allows r to run1，r2Lane, bus allowed to travel r only2，r3Driveways, i.e. S1＝{r1，r2，r3}，S2＝{r1，r2}，S3＝{r2，r3The strategy situation is S ═ S1×S2×S3By calculation, 12(3 × 2 × 2) different situations are found.
FIG. 3 depicts a topology between vehicles where two vehicle representations connected by a line segment have direct information interaction. Let p be1And p2Game notation G for road selection in driving process12，p1And p3Game play G13，p2And p3There is no direct gaming relationship. At this time G12And G13The earnings in the traffic game are quantified as shown in tables 1 and 2;
TABLE 1G12The revenue matrix
| c1\c2 | r1 | r2 |
| r1 | (3，0) | (1，1) |
| r2 | (1，1) | (2，0) |
| r3 | (2，1) | (3，0) |
TABLE 2G13The revenue matrix
| c1\c3 | r2 | r3 |
| r1 | (0，0) | (3，1) |
| r2 | (4，1) | (2，2) |
| r3 | (3，2) | (1，0) |
And selecting the lanes by the vehicle according to the short potential optimal response rule. Traffic congestion can result in information skew. Different time lags can be generated in different situations, and the greater the traffic jam degree is, the greater the time lag is. This situation-dependent time lag is described as follows:
wherein R is1＝{(r1，r2，r3)，(r2，r1，r3)，(r3，r1，r2)}；R3＝{(r2，r2，r2)}；R2＝S\(R1∪R3) (ii) a The evolution equation of the traffic game can be constructed as follows:
by calculation, we can find t 10, when H11＝H10(ii) a And Row472(H10)+Row1257(H10)＝11728(ii) a The result shows that the traffic game can be converged; namely from (r)1，r1，r2，r1，r1，r2，r1，r1，r2) The departing vehicle is finally stabilized in the situation (r)1，r2，r3) (ii) a From (r)2，r1，r3，r2，r1，r3，r2，r1，r3) The obtained vehicle is finally stabilized on the lane (r)2，r1，r3) At the moment, the traffic is smooth.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A real-time optimal lane selection method suitable for a vehicle-road cooperative environment is characterized by specifically comprising the following steps:
s1: establishing a traffic model under the vehicle-road cooperative environment as a network evolution game model with state-dependent time lag by using a mixed value logic function, and deriving a strategy evolution equation by using a matrix method;
s2: acquiring historical vehicle data of each lane within limited time by using road data collection equipment;
s3: taking the vehicle speed as the income, taking the selection of lanes as a strategy, and taking the decision made by all the vehicles at present as the situation; when information time lag exists, analyzing the influence of different situations on the speed of the vehicle, further finding out a strategy for enabling the individual vehicle to realize the maximum benefit, and further making a strategy updating rule of optimal response;
s4: constructing a state transition matrix H by using a matrix method; analyzing the characteristics of the state transition matrix H, judging whether the traffic system is stable, if so, solving a mean transverse point strategy, finding out a criterion of convergence of the traffic system, and further obtaining a stable point, namely the sought optimal path plan;
the step S3 specifically includes:
s31: according to the revenue function c of the individual vehicle ii(t) searching for the time t, selecting the lane with the maximum profit for the neighbor vehicle j, and forming a set Pi(t)：
Wherein i represents a variable of the player, whose value range is {1, 2.., n }, SiIs a playeri policy set, xi(t) is a strategy variable selected by player i at time t;
s32: suppose Pi(t)＝{s1，s2，...，sr}∈SiIf the lane selection for the vehicle is already the best lane, then the next time the vehicle will remain traveling in that lane; if the lane of the vehicle is not the best lane for the vehicles in its neighborhood, then the set P is selected at the next timeiAnd (t) designing a lane with the minimum subscript, namely designing a short potential optimal response strategy updating rule as follows:
the step S4 specifically includes:
s41: identifying a real number set as a vector set, and converting the real number set into an algebraic form z (t +1) ═ Hz (t) by using a matrix half tensor product method; the state transition matrix H absorbs the original time delay information, reflects the reachability of situation transition and reveals the property of the game; since the dimension of one situation is k, the dimension of (lambda +1) situations is kλ+1Thus computing and observing the computation state transition matrix
S42: judging whether the traffic system is stable or not, wherein the judgment formula is as follows:
wherein Rows(H) Represents the s-th Row, of the matrix H-s(H) Denotes the rows of the matrix H except the s-th row, k ═ k1×k2×...×kn(ii) a If the integer s is existed, the traffic system is stable, and if the integer s is not existed, the traffic system is unstable;
s43: if the traffic systemThe system is stable, and the traffic game is solved to obtain a convergence equilibrium point
Will be provided with
Performing a unique decomposition yields:
i.e. when the individual vehicle i selects lane siAt the time, the speed can be optimized, at which time the lane siMost unobstructed and stable, with i 1, 2.
2. The method for selecting the optimal lane in real time under the cooperative vehicle and road environment as claimed in claim 1, further comprising step S5: and if the traffic system is unstable, namely convergence cannot be realized, designing a state feedback controller to perform route optimization control on the individual vehicle so as to ensure that the traffic system is converged.
3. The method for selecting the optimal lane in real time under the cooperative vehicle and road environment as claimed in claim 2, wherein the step S1 includes the following steps:
s11: establishing a traffic model under a vehicle-road cooperative environment into a network evolution game model G with state-dependent time lag by using a mixed value logic networkd；
Assume that there are N vehicles forming a set N ═ {1, 2.., N }, where the set of lanes in which an individual vehicle i may be allowed to travel is Si＝{k1，k2，...，ki}，i＝1，2，...，n，S＝S1×S2×…×SnIs a set of situations, ci: s → a set of real numbers, where i ═ 1, 2., n is the revenue function of individual vehicle i; the topology between vehicles is represented by a connection graph (N, E), where N is the set of points representing the vehicle and E is the set of edges; n is a radical ofiRepresenting individual vehicles iA neighborhood of (c); using mixed-value logic function fiA policy update rule representing an individual vehicle i, where i ═ 1, 2i：S→{0，1，...，λiRepresents the time lag of the individual vehicle i when receiving the information, and the networked vehicle GdThe dynamic model of (2) can be built into the following patterns:
xi(t+1)＝fi(xj(t-τi(x(t)))|j∈Ni，t＝0，1，...，)；
wherein xi(t) is the lane selection for individual vehicle i at time t, x (t) e S is the situation at time t, τi(x (t)) is the state-dependent time lag of the individual vehicle i at time t, ci(t) is the average income of the neighbor vehicle of the individual vehicle i when the game is played, and the formula is as follows:
4. the method for selecting the optimal lane in real time under the cooperative environment of the vehicle and road according to claim 1, wherein the road data collecting device in the step S2 comprises a vehicle speed sensor, a road side sensor and a high precision map; the step S2 specifically includes: based on the vehicle running speed acquired by the vehicle speed sensor, taking the vehicle running speed as a profit quantification index of the individual vehicle; positioning the lane position of the individual vehicle based on a positioning system, a road side sensor and a high-precision map of the intelligent vehicle; and (3) providing data support for the optimal decision of the individual vehicle by combining the historical lane selection condition of each individual vehicle in the t steps stored by the intelligent vehicle-road cooperative system and the information time lag of the individual vehicle in the data receiving process.
5. The method for selecting the optimal lane in real time according to claim 1, wherein the step S5 specifically includes: if the traffic system is unstable, namely the traffic game cannot be converged, intervention is applied to the selection of the individual paths through the system state feedback controller, and lane selection of individual vehicles is controlled, so that the traffic system reaches a stable state; and returning to the step S43 to obtain the optimal lane selection.
6. The method for selecting the optimal lane in real time under the cooperative vehicle and road environment as claimed in claim 5, wherein the specific method of the system status feedback controller in step S5 to intervene in selecting the individual path includes the following three methods:
s5-1: restricting the vehicle's selection set of lanes, i.e. the set of lanes that vehicle i can be allowed to be
Is a set SiTo narrow the degree of freedom of the vehicle;
s5-2: the selection of the lane has priority, the original lane mode of uniformly mixing and selecting is broken through, and the lane different from the adjacent vehicle is preferentially selected, so that the system achieves dynamic balance;
s5-3: one of the vehicles is directly controlled and subjected to direct lane selection intervention, so that the lane selection intervention influences the behaviors of the neighbor vehicles and further influences the selection of the global vehicle, namely:
where t represents a time node, NiRepresents the neighbor set of player i, the value range of player i is {1, 2., n }, u (t) represents the inputtable external control variable, g1Is a artificially designed, fully controllable mixed-value logic function, fiIs the policy update rule for player i.
7. The method for selecting the optimal lane in real time under the cooperative environment of the vehicle and the road according to claim 4, wherein the road side sensor comprises a camera and a millimeter wave radar.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110770608.2A CN113538940B (en) | 2021-07-08 | 2021-07-08 | Real-time optimal lane selection method suitable for vehicle-road cooperative environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110770608.2A CN113538940B (en) | 2021-07-08 | 2021-07-08 | Real-time optimal lane selection method suitable for vehicle-road cooperative environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113538940A CN113538940A (en) | 2021-10-22 |
| CN113538940B true CN113538940B (en) | 2022-05-17 |
Family
ID=78098023
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110770608.2A Active CN113538940B (en) | 2021-07-08 | 2021-07-08 | Real-time optimal lane selection method suitable for vehicle-road cooperative environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113538940B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114413921B (en) * | 2022-01-19 | 2022-12-16 | 广东工业大学 | Excitation mechanism-based internet vehicle running path planning method, system and equipment |
| CN114118633B (en) * | 2022-01-26 | 2022-06-10 | 之江实验室 | Index self-optimization prediction method and device based on precedence relationship |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190202462A1 (en) * | 2017-12-28 | 2019-07-04 | Automotive Research & Testing Center | Method of lane change decision-making and path planning |
| CN110362910A (en) * | 2019-07-05 | 2019-10-22 | 西南交通大学 | Automatic driving vehicle lane-change conflict coordination method for establishing model based on game theory |
| CN111932909A (en) * | 2020-06-09 | 2020-11-13 | 淮阴工学院 | Real-time variable lane dynamic allocation method under intelligent vehicle-road cooperative environment |
| CN111994088A (en) * | 2020-09-02 | 2020-11-27 | 中国科学技术大学 | Driver lane change intention identification method and system based on hybrid strategy game |
| CN112907967A (en) * | 2021-01-29 | 2021-06-04 | 吉林大学 | Intelligent vehicle lane change decision-making method based on incomplete information game |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102019104974A1 (en) * | 2019-02-27 | 2020-08-27 | Zf Automotive Germany Gmbh | Method and system for determining a driving maneuver |
-
2021
- 2021-07-08 CN CN202110770608.2A patent/CN113538940B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190202462A1 (en) * | 2017-12-28 | 2019-07-04 | Automotive Research & Testing Center | Method of lane change decision-making and path planning |
| CN110362910A (en) * | 2019-07-05 | 2019-10-22 | 西南交通大学 | Automatic driving vehicle lane-change conflict coordination method for establishing model based on game theory |
| CN111932909A (en) * | 2020-06-09 | 2020-11-13 | 淮阴工学院 | Real-time variable lane dynamic allocation method under intelligent vehicle-road cooperative environment |
| CN111994088A (en) * | 2020-09-02 | 2020-11-27 | 中国科学技术大学 | Driver lane change intention identification method and system based on hybrid strategy game |
| CN112907967A (en) * | 2021-01-29 | 2021-06-04 | 吉林大学 | Intelligent vehicle lane change decision-making method based on incomplete information game |
Non-Patent Citations (1)
| Title |
|---|
| "基于混合模糊多人多目标非合作博弈的车道选择模型";王晓原等;《自动化学报》;20171130(第43期);第2033-2043页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113538940A (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108847037B (en) | Non-global information oriented urban road network path planning method | |
| CN113538940B (en) | Real-time optimal lane selection method suitable for vehicle-road cooperative environment | |
| CN108510764B (en) | Multi-intersection self-adaptive phase difference coordination control system and method based on Q learning | |
| CN107067759B (en) | ACP-based parallel traffic signal lamp real-time control method | |
| CN103593535A (en) | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration | |
| CN109269516B (en) | Dynamic path induction method based on multi-target Sarsa learning | |
| CN111028504A (en) | Urban expressway intelligent traffic control method and system | |
| CN109115220A (en) | A method of for parking system path planning | |
| Dai et al. | A simulation-based approach to investigate the driver route choice behavior under the connected vehicle environment | |
| Xu et al. | Decentralized time and energy-optimal control of connected and automated vehicles in a roundabout | |
| Pendrith | Distributed reinforcement learning for a traffic engineering application | |
| Lentzakis et al. | Region-based dynamic forecast routing for autonomous vehicles | |
| CN117075473A (en) | Multi-vehicle collaborative decision-making method in man-machine mixed driving environment | |
| Zhang et al. | An optimal vehicle speed planning algorithm for regenerative braking at traffic lights intersections based on reinforcement learning | |
| Wang et al. | Joint traffic signal and connected vehicle control in IoV via deep reinforcement learning | |
| Zhu et al. | Trajectory optimization of CAVs in freeway work zone considering car-following behaviors using online multiagent reinforcement learning | |
| CN115100867A (en) | Urban intelligent traffic simulation method based on digital twins | |
| Wang et al. | Xrouting: Explainable vehicle rerouting for urban road congestion avoidance using deep reinforcement learning | |
| Lu et al. | A Game-Theoretic Approach on Conflict Resolution of Autonomous Vehicles at Unsignalized Intersections | |
| Khodadadi et al. | Improve traffic management in the vehicular ad hoc networks by combining ant colony algorithm and fuzzy system | |
| Guo et al. | Network Multi-scale Urban Traffic Control with Mixed Traffic Flow | |
| Huang et al. | Research on multi-agent traffic signal control system based on VANET information | |
| Ma et al. | Feudal multi-agent reinforcement learning with adaptive network partition for traffic signal control | |
| Wang et al. | Research on intelligent traffic control methods at intersections based on game theory | |
| Yang et al. | Straight-Going Priority in Hierarchical Control Framework for Right-Turning Vehicle Merging Based on Cooperative Game |
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 |