CN113611119A - Vehicle induction method based on gated recursion unit - Google Patents

Vehicle induction method based on gated recursion unit Download PDF

Info

Publication number
CN113611119A
CN113611119A CN202110914395.6A CN202110914395A CN113611119A CN 113611119 A CN113611119 A CN 113611119A CN 202110914395 A CN202110914395 A CN 202110914395A CN 113611119 A CN113611119 A CN 113611119A
Authority
CN
China
Prior art keywords
time
road network
matrix
node
delta
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110914395.6A
Other languages
Chinese (zh)
Other versions
CN113611119B (en
Inventor
任毅龙
付翔
于海洋
姜涵
吴超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN202110914395.6A priority Critical patent/CN113611119B/en
Publication of CN113611119A publication Critical patent/CN113611119A/en
Application granted granted Critical
Publication of CN113611119B publication Critical patent/CN113611119B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G08G1/0145—Measuring and analyzing of parameters relative to traffic conditions for specific applications for active traffic flow control
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)

Abstract

The patent relates to a vehicle induction method based on a gated recursion unit, which comprises the following steps: step 100, establishing a road network G, collecting traffic states from networked vehicles in the road network, obtaining the average speed of each road section, and obtaining the weight values of the nodes in the matrix and the road sections among the nodes; step 200, based on a gate control recursion unit, predicting the traffic state of a later time period by historical data from the previous moment to the current moment to obtain a predicted value matrix of the road network state; step 300, updating a road network adjacent matrix at the next moment according to the predicted value matrix of the road network state, and generating a vehicle guidance path considering time cost; step 400, updating traffic flow information at intervals, acquiring time cost and an optimal path, comparing the saved time of switching the path with the size of a set threshold value, and adopting different responses according to different relations; and 500, judging whether the vehicle reaches the destination within the time step, and obtaining different responses according to different judgment results.

Description

Vehicle induction method based on gated recursion unit
Technical Field
The invention belongs to the technical field of intelligent traffic control, and particularly relates to a vehicle induction method based on a gating recursion unit.
Background
Along with the rapid development of economy and the improvement of the living standard of people, the diversified travel demands of residents provide challenges for the existing urban road service facilities: on one hand, the road infrastructure with relatively delayed development is difficult to adapt to the greatly increased motor vehicle holding amount, so that the road congestion frequently occurs in the peak period; on the other hand, occasional phenomena such as road construction and traffic accidents also aggravate congestion and cause the congestion to spread to the road network, resulting in economic loss and environmental pollution.
Traffic induction is considered an important means of alleviating congestion. However, the existing traffic guidance method is often poor in effect and can only provide partial reference for vehicles by setting a guidance screen on the road side to show the congestion degree of the road section.
Disclosure of Invention
In view of the foregoing analysis, the present invention aims to provide a vehicle guidance method based on a gated recursion unit, so as to solve the problems in the prior art that only the congestion degree of a road segment is shown in a guidance manner under the condition of traffic congestion, so that the referential performance is low and the effect is poor.
The purpose of the invention is realized as follows:
a vehicle induction method based on a gated recursion unit is provided, and comprises the following steps: step 100, establishing a road network, wherein the road network comprises a road network adjacency matrix, acquiring traffic states from networked vehicles in the road network to obtain an average speed of each road section, and obtaining weight values of nodes in the matrix and road sections among the nodes according to the distance between adjacent nodes and the average speed, wherein the weight values form the road network adjacency matrix; step 200, based on a gated recursion unit, predicting the traffic state from the time t + delta t to the time t + n delta t after a time period t + delta t through historical data from the previous time (t- (delta-1) delta t) to the current time t:
Figure BDA0003204905690000021
wherein t is time, delta is prediction time lag, n is prediction step length, and VtThe traffic speed at the time t, delta t is the size of each event window, GRU is a gating recursion unit, and the traffic state from the time t + delta t to the time t + n delta tThe traffic state matrix of (1), i.e. the prediction matrix of the road network state, is [ V ]t +Δt，Vt+2Δt，...，Vt+nΔt]The traffic is the average speed of each road section, and the algorithm of the gate control recursion unit is as follows:
r(t)＝sigmoid(QrIt+UrOt-Δt)
u(t)＝sigmoid(QuIt+UuOt-Δt)
O’t＝tanh{QhIt+Uh[r(t)⊙Ot-Δt]}
Ot＝[1-u(t)]⊙Ot-Δt+u(t)⊙O’t
where r (t) is reset gate in GRU, u (t) is update gate in GRU, O'tFor the current memory content, OtIs the output state at time t, ItIs the input state at time t, Ot-ΔtOutput state at time t- Δ t, QrUr，QuUuAnd QhUhRespectively obtaining a reset gate and an update gate in the GRU by using historical data in a time period from the current moment to the current moment, and obtaining the current memory content and the output state at the t moment by using the reset gate and the update gate; an example indicates a Hadamard product, which is a matrix operation, and sigmoid () indicates a sigmoid function, which is often used as an activation function of a neural network; tanh () represents a hyperbolic tangent function:
Figure BDA0003204905690000022
and, It＝[Vt-(δ-1)Δt，Vt -(δ-2)Δt，...，Vt]，Ot＝[Vt+Δt，Vt+2Δt，...，Vt+nΔt](ii) a Step 300, according to the predictive value matrix [ V ] of the road network statet +Δt，Vt+2Δt，...，Vt+nΔt]Updating the road network adjacency matrix at the next moment according to the distance between adjacent nodes and the average speed, and generating the vehicle guidance considering the time cost according to the road network adjacency matrixA path; step 400, updating traffic flow information every time at intervals of a period of time delta t, acquiring time cost and an optimal path, comparing the saved time of switching paths and the size of a set threshold according to the updated traffic flow information, and adopting different responses according to different relations; and 500, judging whether the vehicle reaches the destination within the time step, and obtaining different responses according to different judgment results. Compared with a common time sequence model, the GRU has a simple structure, is suitable for constructing a large network suitable for a road network, has higher calculation efficiency, and is favorable for rapidly obtaining the future road network state.
Further, the road network G ═ (N, E, W), where N is a set of nodes, N ═ N (N)1，N2，…，Nm) The characteristic value of the node is P ═ P (P)1，p2，…，pn) E is a link set, and W is a road network adjacency matrix.
Further, obtaining a weight value expression between a node and a node in a road network adjacent matrix according to the road network is as follows:
Figure BDA0003204905690000031
wherein, wijRepresenting a node NiAnd NjWeight value of the inter link set from NiTo NjTime required if NiAnd NjIf there is a road between two nodes, thenijFor a neighboring node Ni、NjThe distance between them; v. ofijNIL is the null value for the average velocity of the distance between two nodes.
Furthermore, a road network adjacent matrix at the t + delta t moment is obtained according to the road network state prediction at the t moment
Figure BDA0003204905690000032
Figure BDA0003204905690000033
Wherein the content of the first and second substances,
Figure BDA0003204905690000034
for road network adjacent matrix
Figure BDA0003204905690000035
Row i and column j.
Further, according to
Figure BDA0003204905690000036
And generating a vehicle guidance path considering the time cost, and calculating to obtain the vehicle time cost:
Figure BDA0003204905690000037
Figure BDA0003204905690000038
for vehicle time cost, for Ni、NjE.g. N, define NiAnd NjIs K ═ N (1, 2.., K) ∈ N,
Figure BDA0003204905690000039
for the time cost of node i to node k,
Figure BDA00032049056900000310
the time cost of node k to node j.
Further, when K is an empty set, i.e., NiTo NjIn the absence of an intermediate point in between,
Figure BDA0003204905690000041
when K is not an empty set,
Figure BDA0003204905690000042
by
Figure BDA0003204905690000043
The recursion is obtained by the following specific process: go throughCheck for each intermediate node k
Figure BDA0003204905690000044
If yes, updating
Figure BDA0003204905690000045
If not, the data is not updated, and the data can be output after the traversal of all the points is finished
Figure BDA0003204905690000046
Minimum value of (3) and optimal path PtThis path is the induced path of the output.
Further, a saving time threshold value Δ T is set, if the time saved by selecting a switching path relative to a non-switching path is greater than the value, the induced path switching is performed, otherwise, the condition may be expressed as:
Figure BDA0003204905690000047
wherein the content of the first and second substances,
Figure BDA0003204905690000048
for the time cost of the optimal path at time t,
Figure BDA0003204905690000049
the time cost from node i' to node j at time t + Δ t.
Compared with the prior art, the invention can quickly and effectively reduce the time cost of vehicle passing and improve the passing efficiency.
Drawings
In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present specification, and other drawings can be obtained by those skilled in the art according to the drawings.
FIG. 1 is a flowchart illustrating steps of a vehicle induction method based on a gated recursion unit according to the present invention;
FIG. 2 is a flowchart of an algorithm of a vehicle induction method based on a gated recursion unit according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
For the purpose of facilitating understanding of the embodiments of the present application, the following description will be made in terms of specific embodiments with reference to the accompanying drawings, which are not intended to limit the embodiments of the present application.
Example 1
Referring to fig. 1 and 2, the present embodiment provides a vehicle induction method based on a gated recursion unit.
As shown in fig. 1, the vehicle induction method based on the gated recursion unit includes the following steps:
step 100, establishing a road network, wherein the road network comprises a road network adjacency matrix, collecting traffic states from network connection vehicles in the road network to obtain an average speed of each road section, and obtaining weight values of nodes in the matrix and road sections among the nodes according to the distance between adjacent nodes and the average speed, wherein the weight values form the road network adjacency matrix.
The concept of complex networks is used to define the road network: let G benThe road network of the individual nodes is,
G＝(N，E，W)
where N is a set of nodes, N ═ N (N)1，N2，…，Nm) The characteristic value of the node is P ═ P (P)1，p2，…，pn) (ii) a E is a set of road segments; w is a road network adjacency matrix.
Node NiAnd NjWeight value w of inter-road sectionijExpressed as:
Figure BDA0003204905690000051
wijrepresenting a node NiAnd NjWeight value of the inter link set from NiTo NjTime required if NiAnd NjIf there is a road between two nodes, thenijFor a neighboring node Ni、NjThe distance between them; v. ofijNIL is the null value for the average velocity of the distance between two nodes.
Step 200, based on a gating recursion unit, predicting the traffic state of a later time period by historical data from the previous time to the current time to obtain a prediction value matrix of the road network state at the current time.
The traffic state of the road network G is defined as V, which in the present embodiment specifically refers to the average speed of each road segment, i.e., V ═ V (V ═ in1，v2，…，vm)。
Predicting the traffic state from the time t + delta t to the time t + n delta t by historical data from the time t- (delta-1) delta t to the time t, namely:
Figure BDA0003204905690000061
where t is time, δ is prediction time lag, VtAnd the traffic speed at the time t, delta t is the size of each event window, and n is the prediction step length. The prediction time lag and the prediction step length can be set by self. Such as: δ is 2, n is 1, i.e. using [ V ]t-Δt，Vt]To [ V ]t+Δt]And (6) performing prediction.
The GRU (gated recursive unit) is a gate-controlled recursive unit, and is a variant of a Long-short-term memory neural network (LSTM), and the GRU keeps the effect of the LSTM, and simultaneously has a simpler structure and higher calculation efficiency.
The GRU operation process is as follows:
r(t)＝sigmoid(QrIt+UrOt-Δt)
u(t)＝sigmoid(QuIt+UuOt-Δt)
O’t＝tanh{QhIt+Uh[r(t)⊙Ot-Δt]}
Ot＝[1-u(t)]⊙Ot-Δt+u(t)⊙O’t
where r (t) is reset gate in GRU, u (t) is update gate in GRU, O'tFor the current memory content, OtIs the output state at time t. I istIs the input state at time t, Ot-ΔtOutput state at time t- Δ t, QrUr，QuUuAnd QhUhThe weight matrices for the reset gate, the update gate and the current memory content, respectively, can be defined by themselves.
An indication of a Hadamard product is a matrix operation. sigmoid () represents a sigmoid function, which is often used as an activation function of a neural network; tanh () represents a hyperbolic tangent function:
Figure BDA0003204905690000062
wherein the content of the first and second substances,
It＝[Vt-(δ-1)Δt，Vt-(δ-2)Δt，...，Vt]
Ot＝[Vt+Δt，Vt+2Δt，...，Vt+TΔt]
the traffic state from the time T +1 to the time T + T obtained by the input state calculation can be represented as Vt+1，Vt+2，...，Vt+T]。
Step 300, according to the predictive value matrix [ V ] of the road network statet+Δt，Vt+Δt，...，Vt+nΔt]Updating the road network adjacency matrix at the next moment according to the distance between adjacent nodes and the average speed, and generating a vehicle guidance path considering time cost according to the road network adjacency matrix。
Suppose that the node where the vehicle is located closest to the node at time t is NiThe destination is node NjIf the induced path is Ni、NjTime cost of the minimum path.
Obtaining a predictive value matrix [ V ] of the road network state at the time tt+1，Vt+2，...，Vt+T]Then, according to the predicted value Vt+1Updating road network adjacency matrices, i.e.
Figure BDA00032049056900000716
The corner mark p indicates that the matrix is a predictor. Thus, the road network adjacent matrix at t + delta t moment is obtained according to the road network state prediction at t moment
Figure BDA0003204905690000071
The road network adjacency matrix can be expressed as:
Figure BDA0003204905690000072
wherein the content of the first and second substances,
Figure BDA0003204905690000073
for road network adjacent matrix
Figure BDA0003204905690000074
Row i and column j.
Based on OtUpdate Wt to
Figure BDA0003204905690000075
The expression for the vehicle time cost is:
Figure BDA0003204905690000076
wherein the content of the first and second substances,
Figure BDA0003204905690000077
for vehicle time cost, for Ni、NjE.g. N, define NiAnd NjK ═ 1, 2,. K ∈ N,
Figure BDA0003204905690000078
for the time cost of node i to node k,
Figure BDA0003204905690000079
the time cost of node k to node j.
When K is an empty set, i.e. NiTo NjIn the absence of an intermediate point in between,
Figure BDA00032049056900000710
when K is not an empty set,
Figure BDA00032049056900000711
by
Figure BDA00032049056900000712
The recursion is obtained by the following specific process: go through each intermediate node k to check
Figure BDA00032049056900000713
If yes, updating
Figure BDA00032049056900000714
If not, the updating is not carried out. After all the points are traversed, the output can be obtained
Figure BDA00032049056900000715
Minimum value of (3) and optimal path PtThis path is the induced path of the output.
Step 400, updating traffic flow information every time at intervals of a period of time delta t, obtaining time cost and an optimal path, comparing the saved time of switching paths and the size of a set threshold according to the updated traffic flow information, and taking different responses according to different relations.
Judging whether the time saving of the switching path is greater than a threshold value, if so, carrying out induced path switching, and if not, not carrying out induced path switching.
Since the traffic flow in the road network changes from moment to moment, it is necessary to update the traffic flow information at intervals of Δ t and to perform a new judgment based on the updated traffic flow information to flexibly adjust the guidance route.
Let t be t + delta t, resampling the state of the road network, and obtaining Wt+ΔtRepeating the actions of step 100-step 300 to obtain
Figure BDA0003204905690000081
And an optimal path Pt+Δt。
Here, in order to reduce the amount of operation for the driver to switch the route, the present embodiment defines a time saving threshold value Δ T, which is determined in magnitude relation to the time saving for switching the route. If the saved time of the switching path is greater than the threshold value, switching the induction path; and if the saved time of the switching path is less than the threshold value, switching the guidance path is not performed.
The above conditions can be expressed as:
Figure BDA0003204905690000082
wherein the content of the first and second substances,
Figure BDA0003204905690000083
for the time cost of the optimal path at time t,
Figure BDA0003204905690000084
the time cost from node i' to node j at time t + Δ t.
In this embodiment Δ T is set to 2 minutes, i.e. if the time saved is below 2 minutes, the induction path is not changed.
And 500, judging whether the vehicle reaches the destination within the time step, and obtaining different responses according to different judgment results.
And judging whether the vehicle reaches the destination within a time step, if so, ending the route guidance, and if not, making t equal to t + delta t, and repeating the steps 100 to 300 until the vehicle can reach the destination within the time step.
The above-mentioned embodiments, objects, technical solutions and advantages of the present application are described in further detail, it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present application, and are not intended to limit the scope of the present application, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present application should be included in the scope of the present application.

Claims (7)

1. A vehicle induction method based on a gated recursion unit is characterized by comprising the following steps:
step 100, establishing a road network, wherein the road network comprises a road network adjacency matrix, acquiring traffic states from networked vehicles in the road network to obtain an average speed of each road section, and obtaining weight values of nodes in the matrix and road sections among the nodes according to the distance between adjacent nodes and the average speed, wherein the weight values form the road network adjacency matrix;
step 200, based on a gated recursion unit, predicting the traffic state from the time t + delta t to the time t + n delta t after a time period t + delta t through historical data from the previous time (t- (delta-1) delta t) to the current time t:
Figure FDA0003204905680000011
wherein t is time, delta is prediction time lag, n is prediction step length, and VtThe traffic speed at the time t, delta t is the size of each event window, GRU is a gating recursion unit, and a traffic state matrix of traffic states from the time t + delta t to the time t + n delta t, namely a prediction matrix of the road network state, is [ V [ [ V ]t+Δt，Vt+2Δt，...，Vt+nΔt]The traffic is the average speed of each road section, and the algorithm of the gated recursion unitComprises the following steps:
r(t)＝sigmoid(QrIt+UrOt-Δt)
u(t)＝sigmoid(QuIt+UuOt-Δt)
O’t＝tanh(QhIt+Uh[r(t)⊙Ot-Δt]}
Ot＝[1-u(t)]⊙Ot-Δt+u(t)⊙O’t
where r (t) is reset gate in GRU, u (t) is update gate in GRU, O'tFor the current memory content, OtIs the output state at time t, ItIs the input state at time t, Ot-ΔtOutput state at time t- Δ t, QrUr，QuUuAnd QhUhRespectively obtaining a reset gate and an update gate in the GRU by using historical data in a time period from the current moment to the current moment, and obtaining the current memory content and the output state at the t moment by using the reset gate and the update gate;
an example indicates a Hadamard product, which is a matrix operation, and sigmoid () indicates a sigmoid function, which is often used as an activation function of a neural network; tanh () represents a hyperbolic tangent function:
Figure FDA0003204905680000012
and the number of the first and second electrodes,
It＝[Vt-(δ-1)Δt，Vt-(δ-2)Δt，...，Vt]
Ot＝[Vt+Δt，Vt+2Δt，...，Vt+nΔt]；
step 300, according to the predictive value matrix [ V ] of the road network statet+Δt，Vt+2Δt，...，Vt+nΔt]Updating a road network adjacent matrix at the next moment according to the distance between adjacent nodes and the average speed, and generating a vehicle guidance path considering the time cost according to the road network adjacent matrix;
step 400, updating traffic flow information every time at intervals of a period of time delta t, acquiring time cost and an optimal path, comparing the saved time of switching paths and the size of a set threshold according to the updated traffic flow information, and adopting different responses according to different relations;
and 500, judging whether the vehicle reaches the destination within the time step, and obtaining different responses according to different judgment results.
2. The gated recursion unit based vehicle induction method of claim 1, wherein the road network G ═ (N, E, W), where N is the set of nodes and N ═ N (N, E, W)1，N2，…，Nm) The characteristic value of the node is P ═ P (P)1，p2，…，pn) E is a link set, and W is a road network adjacency matrix.
3. The vehicle induction method based on the gated recursion unit as claimed in claim 2, wherein the weight value expressions between nodes in the road network adjacent matrix obtained according to the road network are:
Figure FDA0003204905680000021
wherein, wijRepresenting a node NiAnd NjWeight value of the inter link set from NiTo NjTime required if NiAnd NjIf there is a road between two nodes, thenijFor a neighboring node Ni、NjThe distance between them; v. ofijNIL is the null value for the average velocity of the distance between two nodes.
4. The vehicle induction method based on the gated recursion unit as claimed in claim 1, wherein the road network adjacency matrix at t + Δ t is obtained according to the road network state prediction at t
Figure FDA0003204905680000022
Figure FDA0003204905680000031
Wherein the content of the first and second substances,
Figure FDA0003204905680000032
for road network adjacent matrix
Figure FDA0003204905680000033
Row i and column j.
5. The vehicle induction method based on the gated recursion unit as claimed in claim 4, characterized in that, according to
Figure FDA0003204905680000034
And generating a vehicle guidance path considering the time cost, and calculating to obtain the vehicle time cost:
Figure FDA0003204905680000035
Figure FDA0003204905680000036
for vehicle time cost, for Ni、NjE.g. N, define NiAnd NjK ═ 1, 2,. K ∈ N,
Figure FDA0003204905680000037
for the time cost of node i to node k,
Figure FDA0003204905680000038
the time cost of node k to node j.
6. Gating according to claim 5Method for inducing a vehicle with recursive elements, characterized in that when K is an empty set, NiTo NjIn the absence of an intermediate point in between,
Figure FDA0003204905680000039
when K is not an empty set,
Figure FDA00032049056800000310
by
Figure FDA00032049056800000311
The recursion is obtained by the following specific process: go through each intermediate node k to check
Figure FDA00032049056800000312
If yes, updating
Figure FDA00032049056800000313
If not, the data is not updated, and the data can be output after the traversal of all the points is finished
Figure FDA00032049056800000314
Minimum value of (3) and optimal path PtThis path is the induced path of the output.
7. The vehicle guidance method based on the gated recursion unit as claimed in claim 1, wherein a saving time threshold Δ T is set, and if the saving time of selecting the switching path relative to the non-switching path is greater than the value, the guidance path switching is performed, otherwise, the condition is not performed, and the condition can be expressed as:
Figure FDA00032049056800000315
wherein the content of the first and second substances,
Figure FDA00032049056800000316
for the time cost of the optimal path at time t,
Figure FDA00032049056800000317
the time cost from node i' to node j at time t + Δ t.
CN202110914395.6A 2021-08-10 2021-08-10 Vehicle induction method based on gated recursion unit Active CN113611119B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110914395.6A CN113611119B (en) 2021-08-10 2021-08-10 Vehicle induction method based on gated recursion unit

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110914395.6A CN113611119B (en) 2021-08-10 2021-08-10 Vehicle induction method based on gated recursion unit

Publications (2)

Publication Number Publication Date
CN113611119A true CN113611119A (en) 2021-11-05
CN113611119B CN113611119B (en) 2022-06-28

Family

ID=78308023

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110914395.6A Active CN113611119B (en) 2021-08-10 2021-08-10 Vehicle induction method based on gated recursion unit

Country Status (1)

Country Link
CN (1) CN113611119B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104464320A (en) * 2014-12-15 2015-03-25 东南大学 Shortest path induction method based on real road network features and dynamic travel time
CN106781489A (en) * 2016-12-29 2017-05-31 北京航空航天大学 A kind of road network trend prediction method based on recurrent neural network
CN109272157A (en) * 2018-09-12 2019-01-25 中交信息技术国家工程实验室有限公司 A kind of freeway traffic flow parameter prediction method and system based on gate neural network
US20190164418A1 (en) * 2017-11-30 2019-05-30 Volkswagen Ag System and method for predicting and maximizing traffic flow
CN111612243A (en) * 2020-05-18 2020-09-01 湖南大学 Traffic speed prediction method, system and storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104464320A (en) * 2014-12-15 2015-03-25 东南大学 Shortest path induction method based on real road network features and dynamic travel time
CN106781489A (en) * 2016-12-29 2017-05-31 北京航空航天大学 A kind of road network trend prediction method based on recurrent neural network
US20190164418A1 (en) * 2017-11-30 2019-05-30 Volkswagen Ag System and method for predicting and maximizing traffic flow
CN109272157A (en) * 2018-09-12 2019-01-25 中交信息技术国家工程实验室有限公司 A kind of freeway traffic flow parameter prediction method and system based on gate neural network
CN111612243A (en) * 2020-05-18 2020-09-01 湖南大学 Traffic speed prediction method, system and storage medium

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JIANYA YUAN等: "A Novel GRU-RNN Network Model for Dynamic Path Planning of Mobile Robot", 《IEEE ACCESS》 *
刘晓雯: "基于浮动车数据的路网区域交通状态分析", 《中国优秀博硕士学位论文全文数据库(硕士)·工程科技Ⅱ辑》 *
杨帅 等: "基于卡口数据的车辆出行轨迹重构方法研究", 《第十一届中国智能交通年会大会论文集》 *
赵朋 等: "高速公路突发事件救援车辆诱导", 《中国公路学报》 *

Also Published As

Publication number Publication date
CN113611119B (en) 2022-06-28

Similar Documents

Publication Publication Date Title
US20210407284A1 (en) Vehicle traffic flow prediction method with missing data
Dotoli et al. A signal timing plan formulation for urban traffic control
CN110222950B (en) Health index system and evaluation method for urban traffic
CN109959388B (en) Intelligent traffic refined path planning method based on grid expansion model
JP2011138487A (en) Method and system for traffic prediction based on space-time relation
Dakic et al. On the modeling of passenger mobility for stochastic bi-modal urban corridors
CN111710162B (en) Urban road network traffic operation condition monitoring method and system
CN113537626B (en) Method for predicting neural network combined time sequence by aggregating information difference
KR20200132067A (en) Intelligent parking guidance service system and method by deep learning based parking lot usage analysis
CN115565369A (en) Hypergraph-based time-space hypergraph convolution traffic flow prediction method and system
CN110992233A (en) Emergency evacuation method and system for urban gathering event
CN113408189B (en) Urban multipoint circulating emergency evacuation and simulation deduction method based on variable cells
CN113611119B (en) Vehicle induction method based on gated recursion unit
CN105260551B (en) Car networking interior joint distribution character analysis method
CN110648531B (en) Node mobility prediction method based on deep learning in vehicle-mounted self-organizing network
Uchimura et al. The dial‐a‐ride problem in a public transit system
Sinha et al. Sustainable time series model for vehicular traffic trends prediction in metropolitan network
CN114419917B (en) Traffic jam grooming method and system based on single-direction graph
Dell’Olmo et al. A model for real-time traffic coordination using simulation based optimization
CN110599769B (en) Hierarchical ranking method for road importance in urban road network in time intervals
Xu et al. The evaluation of the urban road network based on the complex network
CN113628455A (en) Intersection signal optimization control method considering number of people in vehicle under Internet of vehicles environment
Alkhatib et al. A New System for Road Traffic Optimisation Using the Virtual Traffic Light Technology.
Tian et al. Multi-mode spatial-temporal convolution network for traffic flow forecasting
CN116153084B (en) Vehicle flow direction prediction method, prediction system and urban traffic signal control method

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