CN112991745B - Traffic flow dynamic cooperative allocation method under distributed framework - Google Patents
Traffic flow dynamic cooperative allocation method under distributed framework Download PDFInfo
- Publication number
- CN112991745B CN112991745B CN202110480692.4A CN202110480692A CN112991745B CN 112991745 B CN112991745 B CN 112991745B CN 202110480692 A CN202110480692 A CN 202110480692A CN 112991745 B CN112991745 B CN 112991745B
- Authority
- CN
- China
- Prior art keywords
- road
- traffic
- time
- road section
- flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 150
- 230000008569 process Effects 0.000 claims abstract description 30
- 230000007246 mechanism Effects 0.000 claims abstract description 17
- 230000005540 biological transmission Effects 0.000 claims description 12
- 238000011068 loading method Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 230000002441 reversible effect Effects 0.000 claims description 4
- 239000007983 Tris buffer Substances 0.000 claims 1
- 230000002776 aggregation Effects 0.000 claims 1
- 238000004220 aggregation Methods 0.000 claims 1
- 238000012216 screening Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 44
- 238000004422 calculation algorithm Methods 0.000 description 27
- 230000003068 static effect Effects 0.000 description 14
- 238000004364 calculation method Methods 0.000 description 7
- 238000005457 optimization Methods 0.000 description 7
- 238000004088 simulation Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000005094 computer simulation Methods 0.000 description 3
- 238000010845 search algorithm Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000005295 random walk Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a traffic flow dynamic cooperative allocation method under a distributed framework, which is characterized in that a plurality of virtual local sub-planning centers are established by taking an intersection as a secondary allocation unit, and each sub-planning center manages a plurality of road sections connected with the intersection. Each sub-planning center replans the traffic route for the vehicle in the local area according to a specific rule and selects the next traffic road section. The flow distribution process of the sub-planning centers can be synchronously carried out, and the flow distribution result is finally fed back to the previous level, so that the traffic flow distribution efficiency is greatly improved; meanwhile, a dynamic cooperative distribution method for the traffic flow facing the urban road network is provided by distributing the traffic flow to the road sections by means of a bidding mechanism in the market behavior. The method comprises the steps of firstly analyzing and screening congested road intersections on the basis of road impedance analysis, taking the congested road as a bidder in each congested intersection intelligent agent, taking the unblocked road as a bidder, and allocating vehicles to the bidding road with the highest bid price by adopting a bidding mechanism.
Description
Technical Field
The invention belongs to the field of traffic engineering, and particularly relates to a dynamic cooperative distribution method for traffic flow under a distributed framework.
Background
The dynamic traffic flow distribution theory is the core of the research on traffic network flow, and an efficient routing algorithm is a key problem to be solved in the fields of geographic information, traffic and even the whole network science. The dynamic traffic distribution is to distribute time-varying traffic travel demands to a road network according to a predetermined rule. Since the first time the concept of dynamic traffic distribution was proposed, scholars at home and abroad in the traffic field developed four types of research methods for dynamic traffic distribution: a mathematical programming method, an optimal control method, a variational inequality method and a computer simulation method. The mathematical programming method is that a nonlinear multi-target mathematical programming model which accords with the dynamic Wardrop system optimization or user optimization principle is constructed to distribute traffic flow to road sections; although the method can guarantee an optimal solution, the complex mathematical constraints, the inefficient solving algorithm and the FIFO rules limit the method to be used as a verification means for a small-scale simple network. The optimal control method utilizes an optimal control theory to convert dynamic flow distribution into a continuous optimal control problem, and determines a traffic flow distribution state by controlling an optimal value condition; such models are usually converted into time-discrete integer programming problems for solving, but at present, mature and efficient solving algorithms are still lacking. The variational inequality method divides dynamic traffic distribution into two steps of network loading and network distribution, and decomposes an original problem into a sub-linear programming problem to be solved. The computer simulation method can simulate the traffic flow signal behavior in each iterative distribution process, but the convergence and the precision of the result cannot be analyzed from the perspective of the solution of the computer simulation method. The traffic flow dynamic allocation methods enrich the theoretical basis of traffic allocation, but the constructed models are complex, the solving algorithm is low in efficiency, and the assumed conditions are ideal, so that the methods are difficult to be applied to large-scale urban road network environments with complex structures.
Meanwhile, the prior art also has the following defects: (1) the algorithm efficiency caused by the centralized global traffic flow distribution framework is low. In a centralized global flow distribution framework, in order to determine an optimal path from a current position to a target point of a vehicle, the impedance of all road sections needs to be measured according to the real-time position of traffic flow, and the algorithm time complexity is O (n × m × t), wherein n is the number of vehicles; m is the number of road sections; and t is the re-planning times. The undifferentiated calculation of all traffic flows and all road segments not only results in a sharp drop in traffic flow distribution efficiency, but traffic flow distribution solutions may also be generated completely consistent solutions to a large extent with the existence of coincidence with the previous time step. Meanwhile, the global cooperative architecture also causes the problems of overlarge calculation pressure load and low robustness.
(2) The problem of low overall traffic efficiency of a road network caused by unreasonable traffic flow distribution rules. In the traffic flow distribution problem, the path selection of a user can be regarded as a composite assignment process of multiple vehicles and multiple feasible road segments, any road allows multiple vehicles to pass through, and a single vehicle can also take multiple road segments as candidate solutions. Therefore, dynamic traffic distribution is a typical combinatorial optimization problem, and the space of the joint strategy is huge due to the numerous complex road sections, the multiple long-step vehicles and the continuous change of road bearing rate. Given m road segment nodes, n vehicle tasks and t dynamic flow allocations, the number of candidate solutions for the dynamic flow allocation problem will reach t × (m/2) × n-1) | at most. The danger of explosion of such solution candidate combinations is a main cause of low traffic capacity of the road network of the distribution scheme.
Disclosure of Invention
The invention aims to provide a dynamic cooperative distribution method of traffic flow under a distributed framework, which can improve traffic flow distribution efficiency.
The invention provides a traffic flow dynamic cooperative allocation method under a distributed framework, which comprises the following steps:
s1, planning an initial route scheme for all vehicles;
s2, counting the current road network vehicles, and when the current road network vehicles are larger than a preset value, carrying out the following steps; otherwise, counting the road section flow and outputting the cooperative distribution results of all vehicles, and ending the current traffic flow dynamic cooperative distribution process;
s3, counting the traffic density of all road sections at the current moment;
s4, judging the road congestion condition according to the congested traffic density threshold; if the road section is not congested, performing step S5, otherwise, performing step S6;
s5, extracting the route scheme FL of the previous iteration, and performing the step S8;
s6, extracting a vehicle set T of a re-planned path on a congested road section;
s7, replanning a route scheme FN of vehicles in the vehicle set T of the route according to a bidding mechanism;
s8, inputting a new traffic demand, and updating the vehicle position according to the route scheme FN of the vehicles in the vehicle set T of the re-planned path in the step S7 or the route scheme FL of the last iteration in the step S5;
s9, completing the steps, counting the road section flow and outputting the cooperative distribution results of all vehicles;
in the bidding mechanism, each bidding road section node acquires k optimal shortest paths as candidate flow loading strategies from a current road section to a target point, and k paths have a selected probability according to the passing impedance of the paths; probability of selection of jth path of OD vs. u
Is composed of
Wherein, OD is a beginning point and an end point,θdiscrete parameters for measuring the degree of the sensing error of the trip vehicle;
the bid price of the bid section of the jth path of u is OD; j. the design is a squareuA set of k candidate paths between OD pairs u;
based on the path selection probability, the Bidder Bidder allocates an optimal path for each vehicle task and submits a task scheme BT to the Tenderer TendereAnd its corresponding path impedance quote Ve。
The traffic density of step S3 refers to the traffic volume existing on a road of a unit length at a certain time, and the traffic density of a road section r at a certain time is calculated as:
in the formula: l isrFor the length of the section of road r,
the number of vehicles in the ith cluster on the road section r;
when the traffic density of the road section at the current moment is not more than 0.9 time of the traffic density of the road section at the current moment when the road section is congested, the vehicles on the road section do not change the route scheme, and the vehicles continue to run according to the route scheme of the previous time step within the time step; otherwise, the vehicles on the road section replans the routing scheme according to the bidding mechanism.
The bidding mechanism comprises the following steps:
(1) calculating the road section impedance;
(2) and acquiring a flow loading strategy and calculating the probability of selecting the path.
Step (1), calculating the road section impedance specifically comprises the following steps:
1) determining the impedance of each road section in a road network;
2) calculating the road section passing time in a traffic jam state;
3) and constructing a road section impedance matrix based on the road section passing time to be used as a quotation for calculating the flow distribution scheme.
Step 1), the road section impedance adopts the vehicle running time as the travel cost, and the road section passing time without congestion is calculated by a BPR function:
Time r =Time free [1+α(q i /Q i ) β ]
wherein,Time r for sections of road in non-congested conditionsr i Normal transit time of;Time free for road sectionsr i The transit time of the free stream;q i for road sectionsr i The traffic flow of (2);Q i for road sectionsr i (ii) a traffic capacity;αandβin order to be the coefficient of retardation,αis a reference value of 0.223,βreference value of (2.037).
Step 2), the road section passing time under the traffic jam state is specifically as follows:
Time jam =
wherein,Time jam for road sections in congestionr i The transit time of (c);L i for road sectionsr i Length of (d);
for road sectionsr i The critical speed of (c);ρ jam the traffic density is the traffic density when the road section is congested at the current moment;ρ ifor road sectionsr i The traffic density of (2).
Step 3), the bidding price of the bidding road section comprises road section passing time, intersection delay time and time cost of time estimation noise:
wherein,Nrthe number of uncongested segments traversed for the route;Njamthe number of congested road segments traversed by the route;Nisthe number of intersections traversed by the path;
normal transit time for each non-congested road segment;
transit time for each congested road segment;Time δ is time noise;
delay time of the intersection, including vehicle queuing delay and signal lamp delay time;
wherein,N Ve the number of vehicles in line in the intersection;
is a linear coefficient;εand controlling parameters for the delay time of the signal lamp.
Step S8, calculating the route section according to the vehicle route scheme in time stepr i-1 Inflow into the road sectionr i And updating the vehicle position information;
the traffic propagation relationship among the road sections is as follows:
yi(x)=qi(x)tr=min{ ni-1(x), Qi(x), w[Ni(x)- ni(x)]/v}
wherein q isi(x) For the section r at the moment xiTraffic inflow rate of; tr is a time step; n isi-1(x) Is a section of road riTraffic volume at time x-1; qi(x) For the section r at time xiThe maximum inflow of the road section, i.e. the traffic capacity of the road section; n is a radical ofi(x) For the section r at time xiThe maximum bearing capacity of the system, namely the traffic volume of the congestion critical point; n isi(x) Is a section of road riTraffic volume at time x; v is the free traffic flow velocity; w is the reverse propagation speed in the case of traffic congestion; thus, the maximum number of vehicles passing through the link is calculated as yi(x) (ii) a The transmission of the traffic flow in the adjacent road sections in the urban road network is realized through the FIFO principle and the formula, and the vehicle positions are dynamically updated in the vehicle distribution process.
The invention provides a traffic flow dynamic cooperative allocation method under a distributed framework, which sets a plurality of virtual local sub-planning centers by taking an intersection as a secondary allocation unit, and each sub-planning center manages a plurality of road sections connected with the intersection. Each sub-planning center replans the traffic route for the vehicle in the local area according to a specific rule and selects the next traffic road section. The flow distribution process of the sub-planning centers can be synchronously carried out, and the flow distribution result is finally fed back to the previous level, so that the traffic flow distribution efficiency is greatly improved; meanwhile, a dynamic cooperative distribution method for the traffic flow facing the urban road network is provided by distributing the traffic flow to the road sections by means of a bidding mechanism in the market behavior. The method comprises the steps of firstly analyzing and screening congested road intersections on the basis of road impedance analysis, taking the congested road as a bidder in each congested intersection intelligent agent, taking the unblocked road as a bidder, and allocating vehicles to the bidding road with the highest bid price by adopting a bidding mechanism.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention.
FIG. 2 is a schematic diagram of the relationship between urban traffic flow and density according to the method of the present invention.
Fig. 3 is a schematic diagram of cooperative allocation based on a contract network according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a network structure and initial parameters according to an embodiment of the present invention.
Fig. 5a is a schematic diagram of SSP flow allocation results in the embodiment of the present invention, fig. 5b is a schematic diagram of DSP method flow allocation results in the embodiment of the present invention, fig. 5c is a schematic diagram of STSP method flow allocation results in the embodiment of the present invention, fig. 5d is a schematic diagram of DTSP method flow allocation results in the embodiment of the present invention, fig. 5e is a schematic diagram of SL method flow allocation results in the embodiment of the present invention, fig. 5f is a schematic diagram of DL method flow allocation results in the embodiment of the present invention, fig. 5g is a schematic diagram of DCA method flow allocation results in the embodiment of the present invention, and fig. 5h is a schematic diagram of section ID descriptions in the embodiment of the present invention.
Fig. 6 is a schematic view of a traffic flow transmission process of different flow distribution methods according to an embodiment of the present invention.
Fig. 7 is a schematic diagram illustrating an influence of an initial traffic flow on a road network traffic efficiency according to an embodiment of the present invention.
Fig. 8a is a schematic diagram of road segment saturation of a DCA method according to an embodiment of the present invention, fig. 8b is a schematic diagram of road segment saturation of a DTSP method according to an embodiment of the present invention, fig. 8c is a schematic diagram of road segment saturation of a DSP method according to an embodiment of the present invention, fig. 8d is a schematic diagram of road segment saturation of a DL method according to an embodiment of the present invention, fig. 8e is a schematic diagram of road segment saturation of an STSP method according to an embodiment of the present invention, fig. 8f is a schematic diagram of road segment saturation of an SL method according to an embodiment of the present invention, and fig. 8g is a schematic diagram of road segment saturation of an SSP method according to an embodiment of the present invention.
Fig. 9 is a schematic diagram illustrating a comparison of the path impedances before and after dynamic adjustment according to an embodiment of the present invention.
Detailed Description
FIG. 1 is a schematic flow chart of the method of the present invention: the invention provides a traffic flow dynamic cooperative allocation method under a distributed framework, which comprises the following steps:
s1, planning an initial route scheme F for all vehicles through A-star algorithmori(ii) a The a-Star algorithm is the most effective direct search method for solving the shortest path in the static road network.
S2, counting the current road network vehicles Nv，Nv>0, treating by the following steps; otherwise, counting the road section flow and outputting all vehicle distribution results, and ending the current traffic flow dynamic cooperative distribution process;
s3, counting the traffic density rho of all road sections at the current momenta；
S4, according to the traffic density threshold rho of the congestionjamJudging the congestion condition of the road section; rhojamThe traffic density is the traffic density when the road section is congested; no congestion in the road section, i.e. [ rho ]a <0.9*ρjamIf not, performing step S5, otherwise, performing step S6;
s5, extracting the route scheme FL of the previous iteration, and performing the step S8;
s6, extracting a vehicle set T of a re-planned path on a congested road section;
s7, replanning a route scheme FN of vehicles in the vehicle set T of the route according to a bidding mechanism;
s8, inputting a new traffic demand, and updating the vehicle position according to the route scheme FN of the vehicles in the vehicle set T of the re-planned path in the step S7 or the route scheme FL of the last iteration in the step S5;
and S9, completing the steps, counting the road section flow and outputting the cooperative distribution result of all vehicles.
The traffic density in step S3 refers to the amount of traffic existing on a road of a unit length at a certain time, and is generally expressed as the ratio of the number of vehicles on a link to the length of the link. The traffic density for a road segment r at a time can be calculated as:
wherein L isrFor the length of the section of road r,
the number of vehicles in the ith cluster on the road section r;
fig. 2 is a schematic diagram of the relationship between the urban traffic flow and the traffic density according to the method of the present invention, and a triangular function or a trapezoidal function relationship is mostly adopted between the urban traffic flow and the traffic density. Wherein q is traffic flow; q. q.smaxIs road section saturation flow; v is the free traffic flow velocity; and w is the backward propagation speed in the case of traffic congestion. When the instantaneous density of the current road section is not more than 0.9 time of the congestion traffic density of the current road section, rhoa≤0.9*ρjamThe vehicle on the road section does not change the route scheme, and the vehicle continues to run according to the route scheme of the previous time step within the time step; otherwise, when the road section is jammed, the vehicles on the road section replan the route scheme according to the bidding mechanism.
The collaborative planning method based on the bid-attracting mechanism provided by the invention reforms a route scheme for the traffic flow of the congested road section, and fig. 3 is a collaborative allocation schematic diagram based on a contract network according to the embodiment of the invention. Once traffic congestion occurs on segment 2, a path is re-planned within this intersection for the affected vehicles on segment 1. By constructing an intersection agent where the road section 2 is located, the road section 1 is used as a sponsor to issue vehicle tasks, and the road sections 3, 4 and 5 with task execution conditions are used as bidder competitive vehicle tasks.
The collaborative planning method first determines the sender r shown in fig. 3iAnd bidder RSiAnd the publisher issues the bidding document (including task ID, task information and weight level) of the task T; calculating k optimal paths and vehicle passing impedance thereof on the basis of capturing the traffic demand of the task T by each bidding section; determining the flow passing through each path and the vehicle task by combining a Logit loading model; each bidder RSiDetermine and direct the sender riFeeding back a bidding task set GT and a corresponding quotation set VT; thus, the publisher ri selects the final vehicle allocation plan
。
The specific implementation mode is as follows:
let T = { T = }jI j =1,2, …, n is the vehicle task that needs to be assigned, where n is the number of tasks; p = { PkAnd | k =1,2,. and m } is a set of intersection sub-planning centers, wherein m is the number of intersections. Rk = {riI =1,2, …, mk is intersected sub-planning center pkManaged set of road segment nodes, mkThe number of road segment nodes. RSiFor executable road section riA set of potential road segment nodes for the upper vehicle mission. GT = { BTe| e =1,2, …, bm } is the set of bid tasks, BT, fed back by all bidderseThe set of bid tasks fed back for the e-th bidder, bm is the number of bidders,
and is
。
Set of bid prices, V, fed back for bidderseThe set of bid prices fed back to bid task set BTe for the e-th bidder.
Representing bid winning scheme
The set of bid tasks in (1) of (1),
。
firstly, the cooperative flow distribution algorithm determines a sender riAnd bidder RSiAnd the bid amount of the task T is issued by the publisher. Then the sub-planning center pkCalculating the impedance of all road sections in the road network and constructing a road section impedance matrix TnetAnd thus each oneThe bidding road section captures traffic OD requirements of the task T and then carries out the bidding according to the road section impedance matrix TnetK optimal paths and their impedances are calculated. And determining the flow passing through each path and the vehicle task by combining a Logit loading model. Through the steps above, each bidder can determine and direct riA set of feedback bid tasks GT and a corresponding set of bids VT. r isiSelection of final vehicle allocation scheme by bid-winning determination algorithm presented herein
. If there are individual tasks that cannot be performed by this scenario, then this portion of the vehicle still adopts the latest scenario for the last time step
。
(1) Calculating the road section impedance:
in order to determine the optimal path of the vehicle task from the current position to the target point, the impedance of each road segment in the road network needs to be determined first. The generalized road section impedance can be a centralized representation of various factors such as time required for passing through a road section, driving comfort, travel distance, driving cost, steering times and the like, and can be regarded as comprehensive cost of vehicle operation. The link impedances can have different meanings depending on different actual requirements and algorithm purposes.
1) In this embodiment, the link impedance adopts the vehicle running time as the travel cost, the free passage is that the link travel time and the travel time of the congested link have a great difference, and the link travel time without congestion is calculated by the BPR function:
Time r =Time free [1+α(qi/Qi)β]
wherein,Time r for sections of road in non-congested conditionsr i Normal transit time of;Time free for road sectionsr i The transit time of the free stream;q i for road sectionsr i The traffic flow of (2);Q i for road sectionsr i (ii) a traffic capacity;αandβin order to be the coefficient of retardation,αis a reference value of 0.223,βreference value of (2.037).
2) When the road section is in traffic jam, the road section passing time shows a completely different rule from the free passing, and the road section passing time in the traffic jam state is given according to a Greenberg logarithmic model:
wherein,Time jam for road sections in congestionr i The transit time of (c);L i for road sectionsr i Length of (d);
for road sectionsr i The critical speed of (c);ρ jam the traffic density is the traffic density when the road section is congested at the current moment;ρ i for road sectionsr i The traffic density of (2).
3) Constructing a road section impedance matrix based on the road section passing time in the traffic jam state, and using the road section impedance matrix as a quotation for calculating a flow distribution scheme; the bid price of the bid road section comprises three parts of road section passing time, intersection delay time and time estimation noise under the traffic jam state:
wherein Nr is the number of uncongested segments that the route passes through; njam is the number of congested road sections passed by the path; nis is the number of intersections passed by the path;
normal transit time for each non-congested road segment；
Transit time for each congested road segment; time δ is the temporal noise;
delay time of the intersection, including vehicle queuing delay and signal lamp delay time; to simplify the calculation, in this embodiment:
wherein,N Ve the number of vehicles in line in the intersection;
is a linear coefficient; epsilon is a signal lamp delay time control parameter, and is taken as 30s in the embodiment.
(2) Acquiring a flow loading strategy and calculating the probability of selecting a path;
acquiring k optimal shortest paths as candidate flow loading strategies from a current road section to a target point by each bidding road section node, wherein k paths have a selected probability according to the traffic impedance of the paths, and the vehicle task tends to select a path with smaller impedance; in this embodiment, assuming that the random perception error of the traveling vehicle follows a Gumbel distribution, the probability of selecting the jth path of the OD (origin-destination) pair u
Comprises the following steps:
wherein, OD is a beginning-end point, and theta is a discrete parameter for measuring the perception error degree of the trip vehicle;
the bid price of the bid section of the jth path of u is OD; j. the design is a squareuA set of k candidate paths between OD pairs u;
based on the path selection probability, Bidder (road segment node) allocates optimal path for each vehicle task and submits task scheme BT to Tenderer TerdereAnd its corresponding path impedance quote Ve。
In a specific embodiment, the method specifically includes a local search algorithm based on a floating bid selection mechanism:
the winner decision problem of the road section nodes in the intersection sub-planning center is a combination optimization problem. During the negotiation and allocation process of the road section contract network, all bidders Bidders return bid sets to the Tenderer Tenderer. The bidding scenarios fed back by different bidding segments may contain the same vehicle task, thereby causing task conflicts. The bidding road section Tenderrer needs to preferably select a globally optimal vehicle route scheme for avoiding task conflict according to the feedback scheme. Therefore, an integer programming model is firstly constructed to describe the preferred targets and constraints, and then a local search algorithm based on a floating target selection mechanism is designed to efficiently solve the model.
Defining: and (4) conflict bidding: let BTaAnd BTbThe bidding task sets fed back by the a-th and b-th bidders respectively,
and is
. If BTaAnd BTbAt least one identical task exists in both task sets, i.e.
While BTaAnd BTbIs the set of conflicting tasks. Otherwise, the two task sets are referred to as consistent task sets. And constructing a conflict bid matrix Mcon (symmetric matrix) according to the pairwise conflict relationship.
The Winner Decision Problem (WDP) is to select a subset from the candidate setThe sum of bid quotations in the feasible solution is maximized (small) as the feasible solution. Herein a subset of the bid task set GT for which there is no conflict may be selected as a feasible solution C, with the goal of equation (6) being to minimize the sum of the bid offers for the feasible solution C. The solution may be expressed as a boolean set X = { xe | e =1,2, …, bm }, where xe =1 represents the set of bid tasks BTeAnd if so, bm is the number of the bidding sections. Thus, the objective function can be expressed as:
the constraint conditions are as follows: (a)
;(b)
. In the formula: ve represents the quote for the e-th bid section. Operational character
Is defined as: if xi=1, xj=1, and BTiAnd BTjConflict, then xi 
xj=1, otherwise, xi 
xjAnd =0. Constraint (b) indicates that each task can only be selected once, i.e. no conflict can occur between the selected tasks.
Aiming at the problem solved by the optimal scheme of the road section node contract network, a local search algorithm (FLS) based on a floating bid selection mechanism is provided. The algorithm utilizes the probability parameters and the floating label selecting machine to control random walk, thereby enhancing the diversity of the solution and greatly improving the accuracy of the solution. And a priority search strategy is used for preventing repeated retrieval of a candidate solution set space, so that the convergence speed of the optimal solution is improved.
The FLS algorithm is a process of multiple iterations and stepwise optimization. The number of iterations y may be determined manually or until an optimal solution is found. In the searching process, if the algorithm searches all candidate bidding solution sets in each iteration, the convergence rate is necessarily slowed down. In practice, the prior search weights of the candidate solution sets are different. Therefore, in order to accelerate the search speed, the algorithm designs a preferential search bid set QB。QBIs a set of bids compatible with the current optimal solution set. To improve the solution efficiency, at the beginning of each iteration, the algorithm searches the bid set QB preferentially and increases Q according to the conflict bid matrix MconBThe bidding vehicle with the lowest bid in the candidate solution C. And further, obtaining a temporary bidding task set TemB according to the candidate solution C: TemB = GT-C.
In addition, in order to avoid the continuous optimization process to cause the algorithm to fall into local optimization, the algorithm designs a probability parameter
And a floating interval of quoted prices
. The probability parameter is used to control the probability of selecting the lowest bid in each iteration, and the bid float interval represents the gap between the current bid and the lowest bid to determine the bid range of the candidate bidding solution. FLS algorithm with probability
Executing a minimum offer VminGreedy search, in turn, with minimum price VminIs floating interval
Selection of a bid task set F from a provisional bid task set TemB for a bid scopeBI.e. task set FBThe bid quote and minimum quote VminIn the floating interval
Within. The algorithm then proceeds from set FBMedium random selection bidding task BcanAnd adding the solution to the candidate solution C. Where the function random (x) represents the random selection of elements from the set x. Furthermore, the FLS algorithm is probabilistic
Randomly selecting one bidding task B from temporary task set TemBcanAdding the optimal solution C.
The algorithm updates the candidate solution C by judging the conflict relationship of the bidding task. Let VC = { Vc =g| g =1,2, …, mc } is the corresponding bid offer in bid task set C; VB = { VBh| h =1,2, …, mb } is the optimal solution CbestThe bid price corresponding to the task in (1). If it is not
The global optimal solution C is updatedbest. Finally, the FLS algorithm is based on the conflict matrix MconUpdate QB。
Step S8 is embodied in that the route r can be calculated according to the vehicle route plan within the time stepi-1Flows into the section riAnd updating the vehicle position information;
when the traffic flow and the traffic density follow the relationship shown in fig. 2, the link instantaneous flow q is:
wherein,
in order to obtain the traffic density,
the traffic density is the traffic density when the road section is congested; qmax is the road segment saturated streamAn amount; v is the free traffic flow velocity; w is the reverse propagation speed in the case of traffic congestion;
thus, within a time step tr, by the route section ri-1Flows into the section riThe traffic volume is as follows:
wherein y isi(x) For the section r at the moment xiTraffic inflow of (2); q. q.si(x) For the section r at the moment xiTraffic inflow rate of; tr is a time step;
for the section r at the moment xi-1The traffic density of (c);
for the section r at the moment xiMaximum traffic inflow rate of;
for the section r at time xiThe traffic density of (2).
The relationship of traffic density and traffic flow as shown in fig. 2 can be obtained:
wherein n isi(x) Is a section of road riTraffic volume at time x;
for the section r at time xiThe traffic density of (c); tr is a time step; v is the free traffic flow velocity;
therefore, the traffic propagation relationship between road sections is:
yi(x)=qi(x)tr=min{ ni-1(x), Qi(x), w[Ni(x)- ni(x)]/v}
wherein Q isi(x) For the section r at time xiThe maximum inflow of the road section, i.e. the traffic capacity of the road section; n is a radical ofi(x) The maximum bearing capacity of the road section ri at the time x, namely the traffic volume of the congestion critical point; v is the free traffic flow velocity; tr is a time step; w is the reverse propagation speed in the case of traffic congestion; therefore, the number of vehicles whose maximum passage of the link is calculated as yi(x) In the embodiment, the transmission of the traffic flow in the adjacent road sections in the urban road network is realized through the FIFO principle and the formula, and the vehicle positions are dynamically updated in the vehicle distribution process; in particular embodiments, the specific application of the present invention in traffic flow distribution is specifically illustrated in conjunction with an Nguyen network:
set simulation parameters as shown in table 1:
TABLE 1 simulation of Nguyen network traffic distribution parameters
Description of simulation scene: the Nguyen network was originally proposed as a classic traffic research case and is hereafter mostly found in traffic-related domestic and foreign research results due to its close topology to the real road network. As shown in fig. 4, which is a schematic diagram of a network structure and initial parameters according to an embodiment of the present invention, the Nguyen network structure includes 13 nodes and 38 bidirectional road segments, in order to simulate distribution of real traffic flow as much as possible, in this experiment, an initial flow is randomly set in each road segment, and four nodes on the edge of the network are used as travel endpoints.
In order to verify the effectiveness of the dynamic cooperative flow allocation method (DCA) of the present invention, the following 6 classical unbalanced traffic flow allocation methods including 3 dynamic flow allocation strategies and 3 static flow allocation methods are compared in this embodiment. The static shortest path method (SSP) plans paths for all vehicles according to the A-shortest path algorithm, and once the path scheme determines that the routing scheme is not changed in the vehicle running process. To prevent road congestion caused by the unique shortest path, a static Top-K shortest path method (STSP) first determines K optimal paths between ODs based on a shortest path algorithm and selects one of them as a path scheme based on a given probability. Actually, when a traveler selects a route, a certain deviation exists between the perceived driving impedance of each route and the actual route impedance, and the deviation has different probability distribution modes, and in the discrete selection model, a random utility model based on static Logit loading (SL) assumes that the random perceived error of the traveling vehicle obeys Gumbel distribution, and has been widely used in route planning and traffic distribution applications. In addition, a dynamic shortest path method (DSP), a dynamic Top-K shortest path method (DTSP), and a dynamic Logit loading method (DL) are dynamic flow distribution schemes developed according to the above three classical static route schemes, and they execute respective path selection processes within discrete time steps to achieve a dynamic flow distribution effect.
Comparing the flow distribution scheme with the flow space distribution, as shown in fig. 5, which is a schematic diagram of the Nguyen network flow distribution result according to the embodiment of the present invention, the flow distribution results of each method and traffic flow show obvious differences in the road network space. Fig. 5(a) is a schematic diagram of SSP flow allocation results in the embodiment of the present invention, fig. 5(b) is a schematic diagram of DSP method flow allocation results in the embodiment of the present invention, fig. 5(c) is a schematic diagram of STSP method flow allocation results in the embodiment of the present invention, fig. 5(d) is a schematic diagram of DTSP method flow allocation results in the embodiment of the present invention, fig. 5(e) is a schematic diagram of SL method flow allocation results in the embodiment of the present invention, fig. 5(f) is a schematic diagram of DL method flow allocation results in the embodiment of the present invention, fig. 5(g) is a schematic diagram of DCA method flow allocation results in the embodiment of the present invention, and fig. 5(h) is a schematic diagram of link ID descriptions in the embodiment of the present invention. Because the traffic flow does not need to be reconfigured, and the vehicle travels along the specified route, the road section statistical flow of the three static flow distribution methods is obviously smaller than the flow distribution result of the dynamic method. And because the traffic flow path is dynamically changed, the actual passing distance of the jammed vehicles bypassing is increased, and the flow of the road section in the flow distribution result of the dynamic method is larger. The flow distribution results of the SSP, STSP and SL methods have similarity, the flow rate of the road section which is positioned near 4 end points and points to the end points is obviously larger than that of other paths, particularly the flow rate of 21 and 25 road sections leading to the end point 1 presents the highest value, and the marginal road section basically does not play a transmission role. The DSP method in the dynamic method only takes the shortest path because of each re-planning process, but still ignores the utility of other paths compared to the static method, although transferring part of the traffic to other segments ( segments 12, 20, 22). The DTSP method and the DL method further transfer the traffic flow to the global road network by the probability selection process of the path, fully utilize the transportation capacity of the most sections as possible, but the global adjustment of each time step not only needs to spend a great amount of calculation cost, but also the path adjustment benefit of the free-passing vehicles is not great and the transmission pressure of the relevant sections (the sections 20, 22, 27 and 34) is increased, and further the path adjustment space of the congested vehicles is occupied. Compared with the prior art, the traffic distribution result of the DCA method is more balanced, the overall road section transport capacity of the road network can be fully coordinated, and the road network traffic efficiency is improved.
The traffic flow transmission process of different methods also has obvious difference in the traffic flow transmission process in the traffic flow distribution process. As shown in fig. 6, which is a schematic diagram of a traffic flow transmission process of different traffic distribution methods according to an embodiment of the present invention, in a traffic flow simulation transmission scenario of 1560 initial vehicles, all the methods can rapidly transport vehicles near an end point in a road network to the end point in an initial period (within 80 × 30 s), but in a middle period of the traffic flow transmission, because a traffic jam situation is generated due to an ineffective path adjustment, the transport efficiency of a static method starts to decrease, and the DTSP, DL and DCA can still maintain high transport efficiency. In the simulation scenario, the three dynamic methods DTSP, DL and DCA respectively complete all flow tasks at the time points 228 × 30s, 192 × 30s and 123 × 30s, which is much higher than the other methods. In addition, all the methods repeatedly calculate the same data for 5 times to obtain an error interval (shaded area) in the graph, and the result shows that all the methods have relatively stable traffic flow transportation results, and the results have repeated invariance because the probability selection problem is not considered by the two shortest path methods SSP and DSP.
In order to verify the stability and effectiveness of the method under different initial flow rates, the present embodiment compares the road network traffic efficiency of 10 groups of different initial traffic flows. Fig. 7 is a schematic diagram illustrating an influence of an initial traffic flow on a road network traffic efficiency according to an embodiment of the present invention, and as the initial vehicles increase, road network traffic time of all traffic distribution methods shows a steady and continuous ascending trend, and a road network traffic efficiency of a result obtained by a dynamic traffic distribution method is significantly better than that obtained by a static method. In the static flow distribution scheme, the shortest path method SSP presents the most stable time growth situation, and the global traffic time of the road network is linearly related to the initial vehicle number. Although the road network traffic time of the distribution result of the SL method is locally unstable, the overall road network traffic time still shows a rising rule. The result of the DCA method in all cases shows the best road network transportation efficiency.
In order to further reveal the difference between the traffic efficiency of the distribution process of different methods, the embodiment detects the change process of the transportation capacity utilization rate of the road section in the distribution process from the angle of the local area road network. Road segment transport capacity is defined as the saturation of road segment flow
The ratio of the number of vehicles on the road segment at the current time to the road segment capacity is shown as follows.
In the formula:
representing road sectionsr i Is on the firsttrThe number of vehicles in the secondary path re-planning;C i representing road sectionsr i Single transport capacity of (a).
The node 6 in the Nguyen network is used as the center of the network, and is a traffic flow gathering point and a regulation center which are globally important in the network. The process that a large amount of traffic flow is gathered to 6 nodes and diverges outward can represent the flow distribution process of the corresponding method, so the embodiment selects only 4 outward-diverging Road segments (Road 7, Road11, Road22 and Road 25) adjacent to the node 6 as the detection objects of the flow saturation change process. As a result, fig. 8 is a schematic diagram illustrating a comparison of the road section saturation change process according to the embodiment of the present invention, fig. 8a is a schematic diagram illustrating the road section saturation of the DCA method according to the embodiment of the present invention, and fig. 8b is a schematic diagram illustrating the road section saturation of the DTSP method according to the embodiment of the present inventionFig. 8c is a schematic diagram of a road section saturation by a DSP method according to an embodiment of the present invention, fig. 8d is a schematic diagram of a road section saturation by a DL method according to an embodiment of the present invention, fig. 8e is a schematic diagram of a road section saturation by an STSP method according to an embodiment of the present invention, fig. 8f is a schematic diagram of a road section saturation by an SL method according to an embodiment of the present invention, and fig. 8g is a schematic diagram of a road section saturation by an SSP method according to an embodiment of the present invention. The large initial flow in the section 7 results in the section 7 being near flow saturation (ρ) at the beginning of the distributionjam= 0.15), 4 dynamic flow distribution methods rapidly direct traffic to three segments, 11, 22, and 25, to equalize segment utilization. The DCA method gives full play to the transportation capacity of 4 road sections, and the 4 road sections have relatively balanced flow saturation, so that the flow task converged by the node is quickly decomposed and completed. The DTSP and DL methods segments 11 and 22 disperse the flow pressure of the segment 7, but the relative idle of the segment 25 results in a longer processing time. In addition, the 3 static methods do not fully exploit the transportation capacity of the road segments 11, 22 and 25, and long congestion of the road segment 7 also results in low efficiency of the whole road network.
And comparing the road section impedance of the flow distribution scheme, and greatly improving the overall traffic efficiency of the road network by the DCA method through the continuous path adjustment process of the vehicles affected by the congested road section. In this embodiment, the DCA method performs more than 3000 times of path re-planning operations on the relevant vehicle in the flow distribution process, impedance pairs before and after each path adjustment are shown in fig. 9, and fig. 9 is a schematic diagram comparing impedance of paths before and after dynamic adjustment according to an embodiment of the present invention. The path impedance before and after adjustment has a significant correlation. The dynamic flow distribution early-stage and middle-stage path impedance is relatively stable, and the path regulation road section selection solution set is reduced due to the fact that the traffic flow reaches the position near the terminal point in the flow distribution later stage, so that the overall impedance is obviously increased. In a whole, the impedance after path re-planning is averagely reduced by 32.71%, and the transmission efficiency of the road network is greatly improved.
In addition, the present embodiment compares various algorithm run times for different vehicle size scenarios. The results are shown in table 2:
TABLE 2 relationship between traffic flow size and algorithm run time (time: s)
All flow distribution methods are in positive correlation with the initial traffic flow scale, but the dynamic method occupies a large amount of computing resources due to the fact that the dynamic path re-planning of each time step occupies a large amount of computing resources, and therefore the dynamic method is greatly different from the static flow distribution method in computing efficiency. The static shortest path method SSP has a second-level calculation efficiency, while the calculation time of the dynamic Logit loading method DL is measured in hours. Compared with other dynamic flow distribution methods, the DCA method has a huge time cost advantage due to the fact that the congested road sections are judged, and the operation efficiency is improved by about 100% compared with the DSP method which is better in performance in the dynamic method. By combining the flow distribution process and the flow distribution result, the DCA method compresses the calculation cost to the maximum extent under the condition of greatly improving the traffic efficiency of the road network.
Claims (8)
1. A traffic flow dynamic cooperative allocation method under a distributed framework is characterized by comprising the following steps:
s1, planning an initial route scheme for all vehicles;
s2, counting the current road network vehicles, and when the current road network vehicles are larger than a preset value, carrying out the following steps; otherwise, counting the road section flow and outputting the cooperative distribution results of all vehicles, and ending the current traffic flow dynamic cooperative distribution process;
s3, counting the traffic density of all road sections at the current moment;
s4, judging the road congestion condition according to the congested traffic density threshold; if the road section is not congested, performing step S5, otherwise, performing step S6;
s5, extracting the route scheme of the last iterationFLAnd proceeds to step S8;
s6, extracting a vehicle set of a re-planned path on a congested road sectionT；
S7, replanning vehicle set of path according to bidding mechanismTRoute scheme for interior vehiclesFN；
S8, inputting new traffic demand and re-inputting according to the step S7Vehicle aggregation for planning a pathTRoute scheme for interior vehiclesFNOr the route plan of the last iteration of step S5FLUpdating the vehicle position;
s9, completing the steps, counting the road section flow and outputting the cooperative distribution results of all vehicles;
wherein, in the bidding mechanism, each bidding section node is acquiredkTaking the optimal shortest path as a candidate flow loading strategy from the current road section to the target point, and according to the passing impedance of the pathkThe probability that all paths are selected; OD pairuTo (1) ajSelected probability of a strip path
Is composed of
Wherein, OD is a beginning point and an end point,θdiscrete parameters for measuring the degree of the sensing error of the trip vehicle;
is OD pairuTo (1) ajThe bid price of the bid section of the path;J u is OD pairuBetweenkA set of candidate paths;
bidder selecting probability based on above pathBidderAllocating optimal path for each vehicle task and providing tender party with optimal pathTendererSubmission task schemeBT e And corresponding path impedance quotationV e 。
2. The distributed framework-based dynamic cooperative distribution method for traffic flow according to claim 1, wherein the traffic density in step S3 refers to the amount of traffic existing on a road with a unit length at a certain time, and the road section at a certain timerThe traffic density of (c) is calculated as:
wherein,L r for road sectionsrThe length of (a) of (b),
for road sectionsrTo go toiNumber of vehicles in a cluster;
when the traffic density of the road section at the current moment is not more than 0.9 time of the traffic density of the road section at the current moment when the road section is congested, the vehicles on the road section do not change the route scheme, and the vehicles continue to run according to the route scheme of the previous time step within the time step; otherwise, the vehicles on the road section replan the route scheme according to the bidding mechanism.
3. The dynamic cooperative distribution method for traffic flow under distributed framework according to claim 2, characterized in that the bidding mechanism comprises the following steps:
(1) calculating the road section impedance;
(2) and acquiring a flow loading strategy and calculating the probability of selecting the path.
4. The dynamic cooperative distribution method for traffic flow under distributed framework according to claim 3, wherein in the step (1), the step of calculating the road section impedance specifically comprises the following steps:
1) determining the impedance of each road section in a road network;
2) calculating the road section passing time in a traffic jam state;
3) and constructing a road section impedance matrix based on the road section passing time to be used as a quotation for calculating the flow distribution scheme.
5. The traffic flow dynamic cooperative allocation method under the distributed framework according to claim 4, characterized in that in step 1), the road section impedance adopts the vehicle running time as the travel cost, and the road section passing time without congestion is calculated by the BPR function:
Time r =Time free [1+α(q i /Q i ) β ]
wherein,Time r for sections of road in non-congested conditionsr i Normal transit time of;Time free for road sectionsr i The transit time of the free stream;q i for road sectionsr i The traffic flow of (2);Q i for road sectionsr i (ii) a traffic capacity;αandβin order to be the coefficient of retardation,αis a reference value of 0.223,βreference value of (2.037).
6. The dynamic cooperative distribution method for traffic flow under the distributed framework according to claim 5, wherein in step 2), the road section passing time under the traffic congestion state is specifically:
wherein,Time jam for road sections in congestionr i The transit time of (c);L i for road sectionsr i Length of (d);
for road sectionsr i The critical speed of (c);ρ jam the traffic density is the traffic density when the road section is congested at the current moment;ρ i for road sectionsr i The traffic density of (2).
7. The dynamic cooperative distribution method for traffic flow under the distributed framework according to claim 6, characterized in that in step 3), the bid price of the bid section includes section passing time, intersection delay time and time cost of time estimation noise:
wherein,Nrthe number of uncongested segments traversed for the route;Njamthe number of congested road segments traversed by the route;Nisthe number of intersections traversed by the path;
normal transit time for each non-congested road segment;
transit time for each congested road segment;Time δ is time noise;
delay time of the intersection, including vehicle queuing delay and signal lamp delay time;
wherein,N Ve the number of vehicles in line in the intersection;
is a linear coefficient;εand controlling parameters for the delay time of the signal lamp.
8. The distributed framework-based dynamic cooperative distribution method for traffic flow according to any one of claims 1 to 7, wherein the step S8 is to calculate a route section according to a vehicle route scheme within a time stepr i-1Inflow into the road sectionr i And updating the vehicle position information;
the traffic propagation relationship among the road sections is as follows:
y i (x)=q i (x)tr=min{ n i-1(x), Q i (x), w[N i (x)- n i (x)]/v}
wherein,q i (x) Is at the same timexTime road sectionr i Traffic inflow rate of;tris the time step;n i-1(x) For road sectionsr i In thatx-traffic volume at time 1;Q i (x) Is composed ofxTime road sectionr i The maximum inflow of the road section, i.e. the traffic capacity of the road section;N i (x) Is composed ofxTime road sectionr i The maximum bearing capacity of the system, and the traffic volume of a congestion critical point;n i (x) For road sectionsr i In thatxThe amount of traffic at a moment;vis the free traffic flow velocity;wthe speed of reverse propagation in the case of traffic congestion; thus, the maximum number of vehicles passing through the link is calculated asy i (x) (ii) a The transmission of the traffic flow in the adjacent road sections in the urban road network is realized through the FIFO principle and the formula, and the vehicle positions are dynamically updated in the vehicle distribution process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110480692.4A CN112991745B (en) | 2021-04-30 | 2021-04-30 | Traffic flow dynamic cooperative allocation method under distributed framework |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110480692.4A CN112991745B (en) | 2021-04-30 | 2021-04-30 | Traffic flow dynamic cooperative allocation method under distributed framework |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112991745A CN112991745A (en) | 2021-06-18 |
| CN112991745B true CN112991745B (en) | 2021-08-03 |
Family
ID=76336882
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110480692.4A Active CN112991745B (en) | 2021-04-30 | 2021-04-30 | Traffic flow dynamic cooperative allocation method under distributed framework |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112991745B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114023093B (en) * | 2021-08-29 | 2024-03-15 | 北京工业大学 | Multi-vehicle dynamic evacuation method driven smoothly in short distance |
| CN113935525B (en) * | 2021-10-11 | 2024-08-23 | 上海交通大学 | Dynamic traffic flow distribution method based on POI influence |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS63161749A (en) * | 1986-12-25 | 1988-07-05 | Nec Corp | Flow assigning system for hybrid link system |
| CN101571404A (en) * | 2009-06-11 | 2009-11-04 | 东南大学 | Shortest path auction algorithm considering intersection turning |
| WO2015015235A1 (en) * | 2013-07-29 | 2015-02-05 | Ren Jinxiang | Dynamic travel guidance system for motorists |
| CN104851309B (en) * | 2015-06-01 | 2017-07-14 | 南京邮电大学 | A kind of implementation method of traffic guidance strategy |
| CN108847037B (en) * | 2018-06-27 | 2020-11-17 | 华中师范大学 | Non-global information oriented urban road network path planning method |
| CN109409773B (en) * | 2018-11-14 | 2022-04-08 | 中南大学 | Dynamic planning method for earth observation resources based on contract network mechanism |
| CN109686082B (en) * | 2018-12-07 | 2020-08-07 | 西安电子科技大学 | Urban traffic monitoring system based on edge computing nodes and deployment method |
| CN110379161B (en) * | 2019-07-18 | 2021-02-02 | 中南大学 | Urban road network traffic flow distribution method |
| CN110930704B (en) * | 2019-11-27 | 2021-11-05 | 连云港杰瑞电子有限公司 | Traffic flow state statistical analysis method based on edge calculation |
-
2021
- 2021-04-30 CN CN202110480692.4A patent/CN112991745B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112991745A (en) | 2021-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108847037B (en) | Non-global information oriented urban road network path planning method | |
| CN108470444B (en) | A kind of city area-traffic big data analysis System and method for based on genetic algorithm optimization | |
| CN108510764B (en) | Multi-intersection self-adaptive phase difference coordination control system and method based on Q learning | |
| CN109785619B (en) | Regional traffic signal coordination optimization control system and control method thereof | |
| CN112991745B (en) | Traffic flow dynamic cooperative allocation method under distributed framework | |
| Sabar et al. | A memetic algorithm for real world multi-intersection traffic signal optimisation problems | |
| Shang et al. | A multi-population cooperative coevolutionary algorithm for multi-objective capacitated arc routing problem | |
| Sun et al. | Dynamic path planning algorithms with load balancing based on data prediction for smart transportation systems | |
| CN110515380B (en) | Shortest path planning method based on turning weight constraint | |
| CN110472792B (en) | Logistics distribution vehicle route optimization method based on discrete bat algorithm | |
| Tian et al. | IQGA: A route selection method based on quantum genetic algorithm-toward urban traffic management under big data environment | |
| CN111966495B (en) | Data processing method and device | |
| CN109919532A (en) | Logistics node determination method and device | |
| Pham et al. | Learning coordinated traffic light control | |
| CN110119399B (en) | Business process optimization method based on machine learning | |
| CN112884229B (en) | Large-scale public place people flow guiding path planning method based on differential evolution algorithm | |
| Sayyah et al. | Solving the vehicle routing problem with simultaneous pickup and delivery by an effective ant colony optimization | |
| Lin et al. | Niching Pareto ant colony optimization algorithm for bi-objective pathfinding problem | |
| CN112330001A (en) | Logistics distribution vehicle route optimization method based on discrete bat algorithm | |
| Ye et al. | Fairlight: Fairness-aware autonomous traffic signal control with hierarchical action space | |
| CN115048576A (en) | Flexible recommendation method for airport passenger group travel mode | |
| CN113034899B (en) | Signal-controlled intersection coordination optimization control method and system | |
| Meng et al. | Applying improved K-means algorithm into official service vehicle networking environment and research | |
| Xiang et al. | A weighted mean velocity feedback strategy in intelligent two-route traffic systems | |
| CN113191029A (en) | Traffic simulation method, program, and medium based on cluster computing |
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 |