CN110211381A - Traffic route distribution method and device - Google Patents
Traffic route distribution method and device Download PDFInfo
- Publication number
- CN110211381A CN110211381A CN201910483977.6A CN201910483977A CN110211381A CN 110211381 A CN110211381 A CN 110211381A CN 201910483977 A CN201910483977 A CN 201910483977A CN 110211381 A CN110211381 A CN 110211381A
- Authority
- CN
- China
- Prior art keywords
- traffic
- time
- matrix
- node
- travel
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 18
- 239000011159 matrix material Substances 0.000 claims description 119
- 238000010586 diagram Methods 0.000 claims description 50
- 230000008859 change Effects 0.000 claims description 37
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000007774 longterm Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 11
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 16
- 238000012545 processing Methods 0.000 description 7
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 3
- 230000006698 induction Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 230000001939 inductive effect Effects 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/096805—Systems involving transmission of navigation instructions to the vehicle where the transmitted instructions are used to compute a route
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/096833—Systems involving transmission of navigation instructions to the vehicle where different aspects are considered when computing the route
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Mathematical Physics (AREA)
- Traffic Control Systems (AREA)
Abstract
A kind of traffic route distribution method and device that the disclosure provides, it is related to technical field of transportation, the traffic route distribution method and device that the disclosure provides, by obtaining the traffic data in the period instantly in real time, and traffic data is input in the traffic prediction model that training obtains in advance, obtain predicted travel time, after obtaining predicted travel time, according to the logic chart network and predicted travel time with time dimension constructed in advance, Trip Costs desired value is calculated, and then recommended route scheme is calculated according to Trip Costs desired value, thus, improve the current smoothness of traffic.
Description
Technical Field
The disclosure relates to the technical field of traffic, in particular to a traffic route distribution method and device.
Background
Existing traffic inducement schemes typically provide inducement suggestions to users based on geographic information and historical road travel times. According to research, when a large number of users receive the same inducing information, aggregation and over-excitation reaction phenomena are easy to occur, so that congestion of one road section is transferred to another road section, even an oscillation phenomenon of one road is generated in severe cases, and the overall traffic efficiency is reduced.
Disclosure of Invention
In view of the above, the present disclosure provides a traffic route allocation method and apparatus.
The present disclosure provides a traffic route allocation method, the method comprising:
and acquiring the traffic data in the current time period in real time.
And inputting the traffic data into a traffic prediction model obtained by pre-training to obtain the predicted travel time.
And calculating to obtain the expected travel cost value according to a pre-constructed logic diagram network with a time dimension and the predicted travel time.
And calculating to obtain a recommended route scheme according to the travel cost expected value.
Further, the traffic prediction model is obtained by training through the following steps:
and extracting a travel time matrix, a distribution probability matrix, a weather characteristic matrix and an event characteristic matrix in a preset time period from the historical road travel time data set.
And obtaining a long-term traffic change trend characteristic matrix, a simultaneous traffic characteristic matrix and an instant traffic change trend characteristic matrix according to the travel time matrix.
And inputting the long-term traffic change trend characteristic matrix, the simultaneous traffic characteristic matrix, the characteristic matrix of the instant traffic change trend, the distribution probability matrix, the weather characteristic matrix and the event characteristic matrix into a deep convolutional neural network for training to obtain prediction output data.
And constructing a loss function based on the prediction output data, and adjusting model parameters according to the loss function, wherein the adjusted parameters are used as parameters of the traffic prediction model.
Further, the step of obtaining a long-term traffic change trend feature matrix, a simultaneous traffic characteristic feature matrix and an instant traffic change trend feature matrix according to the travel time matrix includes:
according to the travel time matrix, obtaining a long-time traffic change trend characteristic matrix, a simultaneous time section traffic characteristic matrix and an instant traffic change trend characteristic matrix through the following formulas:
wherein,represents tiA travel time matrix of the time of day;representing a long-term traffic change trend characteristic matrix;representing a traffic characteristic feature matrix of the same time period;a feature matrix representing an instant traffic change trend; rt represents the time window length; wt represents the length of one week(ii) a st, mt denotes the length of the sampling interval, st<mt; ρ, θ represents the update weight ratio.
Further, the formula of the loss function is:
wherein theta represents a current traffic prediction model parameter; j (theta) is a loss value of the traffic prediction model when the parameter is theta; x is an input value of the traffic prediction model; n is the number of samples; i is a number from 1 to n; y is(i)The real output value corresponding to the ith input value; h is(θ)(x(i)) And outputting data for the prediction of the trained traffic prediction model when the parameter is theta and the input is x.
Further, the logical graph network with the time dimension is obtained by the following steps:
the method comprises the steps of extracting the logic relation among roads in the traffic network from geographic information, setting a demand center, and constructing a logic graph network by taking a traffic flow trend bifurcation point as a node and taking a traffic flow walking route as an edge.
And expanding the logic diagram network on a time dimension to obtain the logic diagram network with the time dimension.
Further, the step of calculating the expected travel cost value according to the pre-constructed logic diagram network with the time dimension and the predicted travel time includes:
according to a pre-constructed logic diagram network with a time dimension and the predicted travel time, calculating to obtain a travel cost expected value through the following formula:
wherein，Is shown at tnAnd the expected travel cost value of the travel from the node i to the terminal d at the time point.
i, j, k represent nodes in the logical graph network with the time dimension.
D represents an end point, D is equal to D, and D is a set of demand center points in the logic diagram network with the time dimension.
Is shown at tnAt that time, the predicted travel time for traveling from node i to node j.
A (j) represents the set of end points of all edges starting from node j.
Is shown at tnThe probability that the time reaches the end point d selects from node j to node k.
Is shown at tnThe time-to-end d selects the desired travel time required from node j to node k.
Further, the step of calculating a recommended route plan according to the expected travel cost value includes:
and calculating to obtain the recommended flow ratio of each road according to the travel cost expected value of each road.
And obtaining a flow ratio suggestion matrix according to the recommended flow ratio of each road.
And obtaining a recommended route scheme according to the flow ratio suggestion matrix.
Further, the step of calculating the recommended flow rate ratio of each road according to the expected travel cost value of each road includes:
according to the expected trip cost value of each road, calculating the recommended flow ratio of each road through the following formula:
wherein,represents tnAnd in the process that the time point goes out from the node i to the end point d, the recommended flow ratio from the node i to the node j is obtained.
i, j represent nodes in the logical graph network with the time dimension.
A (i) represents the set of end points of all edges starting from node i.
Is shown at tnThe time, the relevant adaptive parameter of the edge (i, j);
is shown at tnAnd the expected travel cost value of the travel from the node i to the terminal d at the moment.
Is at tnAnd at the moment, taking the node i as a starting point, all edges are opened to the expected value of the average travel cost of the point d.
Further, the step of obtaining the flow rate matching suggestion matrix according to the recommended flow rate matching of each road includes:
according to the recommended flow ratio of each road, obtaining a flow ratio suggestion matrix through the following formula:
wherein,represents tnAnd in the process that the time point goes out from the node i to the end point d, the recommended flow ratio from the node i to the node j is obtained.
N is an edge in the logic diagram network with the time dimension; d is a set of demand center points in the logic diagram network with the time dimension.
The disclosure provides a traffic route distribution device, which comprises an acquisition module, a calculation module and a recommendation module.
The acquisition module is used for acquiring the traffic data in the current time period in real time.
The calculation module is used for inputting the traffic data into a traffic prediction model obtained by pre-training to obtain the predicted travel time.
The calculation module is further used for calculating to obtain the expected travel cost value according to the pre-constructed logic diagram network with the time dimension and the predicted travel time.
And the recommending module calculates a recommended route scheme according to the travel cost expected value.
According to the traffic route distribution method and device, the traffic data in the current time period are obtained in real time, the traffic data are input into a traffic prediction model obtained through pre-training to obtain the predicted travel time, after the predicted travel time is obtained, a travel cost expected value is obtained through calculation according to a pre-constructed logic diagram network with time dimension and the predicted travel time, and then a recommended route scheme is obtained through calculation according to the travel cost expected value, so that the traffic smoothness is improved.
The traffic route distribution method and the traffic route distribution device have the advantages that traffic data in the current time period are obtained in real time, the traffic data are input into a traffic prediction model obtained through pre-training to obtain predicted travel time, therefore, when a distribution scheme is given to a local traffic network with excessively concentrated traffic, the impending change of the traffic network and the influence caused by a control strategy are pre-judged, after the pre-judgment result, namely the predicted travel time, a travel cost expected value is calculated according to a pre-constructed logic diagram network with time dimension and the predicted travel time, a recommended route scheme is calculated according to the travel cost expected value, and therefore, the next strategy is selected on the pre-judged result, and the accuracy of traffic distribution and the smoothness of traffic passage are improved.
In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
To more clearly illustrate the technical solutions of the present disclosure, the drawings needed for the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present disclosure, and therefore should not be considered as limiting the scope, and those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
Fig. 1 is a block diagram of an electronic device provided in the present disclosure.
Fig. 2 is a schematic flow chart of a traffic route allocation method provided in the present disclosure.
Fig. 3 is another schematic flow chart of a traffic route allocation method provided by the present disclosure.
Fig. 4 is a schematic diagram of a training process of the traffic prediction model provided by the present disclosure.
Fig. 5 is a schematic flow chart of a traffic route allocation method provided by the present disclosure.
Fig. 6 is a schematic diagram of a basic structure of a logic diagram network provided by the present disclosure.
Fig. 7 is a schematic flow chart of a traffic route allocation method provided by the present disclosure.
Fig. 8 is a block schematic diagram of a traffic line distribution apparatus provided by the present disclosure.
Icon: 100-an electronic device; 10-a traffic line distribution device; 11-an acquisition module; 12-a calculation module; 13-a recommendation module; 20-a memory; 30-a processor.
Detailed Description
The technical solutions in the present disclosure will be described clearly and completely with reference to the accompanying drawings in the present disclosure, and it is to be understood that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The components of the present disclosure, as generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure, presented in the figures, is not intended to limit the scope of the claimed disclosure, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the disclosure without making creative efforts, shall fall within the protection scope of the disclosure.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
Existing traffic inducement schemes typically provide inducement suggestions to users based on geographic information and historical road travel times. According to research, when a large number of users receive the same inducing information, aggregation and overstimulation reaction phenomena are easy to occur, so that congestion of one road section is transferred to another road section, and even oscillation phenomena used by one road can occur in severe cases.
At present, some new schemes can provide a guidance strategy for avoiding congestion, but the determination of whether congestion is carried out based on real-time traffic information, and most of the schemes only consider the guidance for a single traffic body. The traffic guidance mode generally has high dependence on the real-time performance and stability of information acquisition, the higher the real-time information updating frequency is, the higher the congestion avoidance degree is, therefore, a user may need to frequently change a driving route to achieve the purpose of reducing the driving time, and in practical application, such behaviors sometimes reduce the overall efficiency of traffic passage instead, and increase the accident rate.
Based on the above research, the present disclosure provides a traffic route allocation method.
The traffic route allocation method provided by the present disclosure is applied to the electronic device 100 shown in fig. 1, and the electronic device 100 executes the traffic route allocation method provided by the present disclosure. In the present disclosure, the electronic device 100 may be, but is not limited to, an electronic device 100 having a data processing capability, such as a Personal Computer (PC), a notebook Computer, a Personal Digital Assistant (PDA), or a server.
The electronic device 100 includes a traffic line assignment device 10, a memory 20, and a processor 30; the various elements of the memory 20 and processor 30 are electrically connected to each other, either directly or indirectly, to enable data transfer or interaction. For example, the components may be directly electrically connected to each other via one or more communication buses or signal lines. The traffic line distribution device 10 includes at least one software function module which can be stored in the memory 20 in the form of software or Firmware (Firmware), and the processor 30 executes various function applications and data processing by running the software programs and modules stored in the memory 20.
The Memory 20 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 30 may be an integrated circuit chip having signal processing capabilities. The processor 30 may be a general-purpose processor including a Central Processing Unit (CPU), a Network Processor (NP), and the like.
Referring to fig. 2, fig. 2 is a schematic flow chart of a traffic route allocation method according to the present disclosure. The detailed flow of the traffic route allocation method shown in fig. 2 is described in detail below.
Step S10: and acquiring the traffic data in the current time period in real time.
The traffic data includes traffic flow data, road data, vehicle driving direction data, and the like in the current time period.
Step S20: and inputting the traffic data into a traffic prediction model obtained by pre-training to obtain the predicted travel time.
Referring to fig. 3, the traffic prediction model is trained from step S50 to step S53.
Step S50: and extracting a travel time matrix, a distribution probability matrix, a weather characteristic matrix and an event characteristic matrix in a preset time period from the historical road travel time data set.
It should be noted that, in the present disclosure, before extracting the travel time matrix, the distribution probability matrix, the weather feature matrix, and the event feature matrix within the preset time period from the historical road travel time dataset, the historical road travel time dataset is first subjected to the dimension reduction processing, that is, the sparse data in the historical road travel time dataset is subjected to the smoothing processing. After dimension reduction processing is carried out on the historical road data set, data characteristic vectors such as travel time TD, weather WD and event EVED in a preset time period are extracted from the historical road data set, and a travel time matrix, a weather characteristic matrix and an event characteristic matrix in the preset time period are obtained according to the extracted data characteristic vectors.
Wherein TD ═ TD (TD)ij)N×T，WD＝(wdij)N×T，EVED＝(evedij)N×T，tdij、wdij、evedijThe method comprises the steps of respectively representing travel time, weather characteristics and event characteristics of a road i at the moment j, N representing the number of edges in a logic diagram network, and T representing the intercepting time length of traffic characteristics, namely a preset time period.
In the present disclosure, the distribution probability matrix used to train the traffic prediction model is obtained by iterative computation of the average travel time in the historical road travel time dataset. The calculation formula is as follows:
where d represents the destination and i, j, k represents a node in the traffic network.
Is the probability of choosing from node j to node k to reach the end point d in the (n-1) th iteration.
The minimum expected travel cost value required to select from node i to node k when reaching the end point d in the (n-1) th iteration.
tijRepresenting the average travel time from node i to node j.
A (j) represents the set of end points of all edges starting from node j.
The minimum expected travel cost value required to select from node i to node j when reaching the end point d in the nth iteration.
And representing the average expected travel cost value of all edges starting from the node i to the point d in the nth iteration.
Tau is an adaptive parameter and is set according to the actual traffic condition.
Representing the probability of choosing from node i to node j to reach the end point d in the nth iteration.
According to the formula, the probability of selecting different node trips when each terminal point is reached can be calculated, and the distribution probability matrix is obtained according to the calculated probability value.
And after a travel time matrix, a distribution probability matrix, a weather characteristic matrix and an event characteristic matrix in a preset time period are obtained. Step S51 is executed.
Step S51: and obtaining a long-term traffic change trend characteristic matrix, a simultaneous traffic characteristic matrix and an instant traffic change trend characteristic matrix according to the travel time matrix.
The characteristic matrix of the long-term traffic change trend, the characteristic matrix of the traffic characteristic of the same time period and the characteristic matrix of the instant traffic change trend can be obtained by the following formulas according to the travel time matrix, wherein the characteristic of the traffic data which is changed periodically is always aimed at:
wherein,a travel time matrix representing time ti;representing a long-term traffic change trend characteristic matrix;representing a traffic characteristic feature matrix of the same time period;a feature matrix representing an instant traffic change trend; rt represents the time window length; wt represents a one week time period; st, mt denotes the length of the sampling interval, st<mt; ρ, θ represents the update weight ratio.
After the long-term traffic change tendency feature matrix, the simultaneous-period traffic characteristic feature matrix, and the feature matrix of the instant traffic change tendency are obtained according to the travel time matrix, step S52 is executed.
Step S52: and inputting the long-term traffic change trend characteristic matrix, the simultaneous traffic characteristic matrix, the characteristic matrix of the instant traffic change trend, the distribution probability matrix, the weather characteristic matrix and the event characteristic matrix into a deep convolutional neural network for training to obtain prediction output data.
Step S53: and constructing a loss function based on the prediction output data, and adjusting model parameters according to the loss function, wherein the adjusted parameters are used as parameters of the traffic prediction model.
Referring to fig. 4, after obtaining the long-term traffic change trend feature matrix, the simultaneous traffic characteristic feature matrix, the immediate traffic change trend feature matrix, the distribution probability matrix, the weather feature matrix, and the event feature matrix, the long-term traffic change trend feature matrix, the simultaneous traffic characteristic feature matrix, the immediate traffic change trend feature matrix, the distribution probability matrix, the weather feature matrix, and the event feature matrix are input into the RCNN neural network as input values, and are trained to obtain predicted output data.
After obtaining the predicted output data, constructing a loss function based on the predicted output data, wherein the formula of the loss function is as follows:
wherein theta represents a current traffic prediction model parameter; j (theta) is a loss value of the traffic prediction model when the parameter is theta; n is the number of samples; i is a number from 1 to n; y is(i)The real output value corresponding to the ith input value; x is the number of(i)Input values for a traffic prediction model; h is(θ)(x(i)) And outputting data for the prediction of the trained traffic prediction model when the parameter is theta and the input is x.
And after the loss function is constructed, the loss function is converged, when the loss function is converged, the error between the actual output value corresponding to the input value and the predicted output data is minimum, and then the model parameter at the moment is used as the parameter of the traffic prediction model obtained by training.
Optionally, in the present disclosure, the trained traffic prediction model is updated along with the update of the distribution probability matrix, so as to improve the accuracy of the prediction output data.
The distribution probability matrix calculation formula for updating the traffic prediction model is as follows;
wherein,distributing a probability matrix for ti moment; λ is an update weight ratio; t is an update time interval; n represents the number of edges in the logical graph network; d is a demand central point in the logic diagram network; pd(i, j) denotes arrivalThe end point d selects the probability from node i to node j.
Referring back to fig. 2, after the traffic data is input into the traffic prediction model trained in advance to obtain the predicted travel time, step S30 is executed.
Step S30: and calculating to obtain the expected travel cost value according to a pre-constructed logic diagram network with a time dimension and the predicted travel time.
Referring to fig. 5, the logical graph network with time dimension is obtained through steps S60 to S61.
Step S60: the method comprises the steps of extracting the logic relation among roads in the traffic network from geographic information, setting a demand center, and constructing a logic graph network by taking a traffic flow trend bifurcation point as a node and taking a traffic flow walking route as an edge.
The basic structure of the graph network is a final carrier of all traffic information analysis results, and is an important basis for calculating the optimal traffic distribution scheme for the traffic network by using a graph search algorithm. The traditional graph network construction mode which takes the intersection as a node and the lane as the side can not well reflect the real state of the traffic network, and even if vehicles on the same road select different traveling directions, the traveling speeds have larger differences due to the influence of various factors such as signal lamp setting, lane setting, traffic flow in the directions and the like.
The method comprises the steps of firstly extracting the logic relation among roads in the traffic network from geographic information, setting main people flow dense areas, large parking lots and the like as demand central points for vehicle generation and reception, and then constructing a logic diagram network by taking traffic flow trend bifurcation points as nodes and traffic flow traveling routes as sides. Compared with the traditional graph network construction mode, the real situation of the traffic network can be more accurately evaluated.
As shown in fig. 6, taking a typical four-way intersection as an example, the present disclosure provides a basic structure diagram of a logic diagram network, in fig. 6, A, B, C, D is a node, and a traffic flow traveling route (solid curve) is an edge.
Step S61: and expanding the logic diagram network on a time dimension to obtain the logic diagram network with the time dimension.
After the logic diagram network is constructed, the logic diagram network is expanded on the time dimension, namely the time dimension is added to the logic diagram network, and the logic diagram network with the time dimension is obtained.
Further, according to a pre-constructed logic diagram network with a time dimension and the predicted travel time, calculating to obtain a travel cost expected value through the following formula:
wherein,is shown at tnAnd the expected travel cost value of the travel from the node i to the terminal d at the time point.
i, j, k represent nodes in the logical graph network with the time dimension.
D represents an end point, D is equal to D, and D is a set of demand center points in the logic diagram network with the time dimension.
Is shown at tnAt that time, the predicted travel time for traveling from node i to node j.
A (j) represents the set of end points of all edges starting from node j.
Is shown at tnIs timed toThe probability that end point d chooses from node j to node k is reached.
Is shown at tnThe time-to-end d selects the desired travel time required from node j to node k.
According to the formula, the expected travel cost value of each road in the logic diagram network with the time dimension can be calculated.
Referring back to fig. 2, after calculating the expected travel cost value of each road in the logical graph network with time dimension, step S40 is executed.
Step S40: and calculating to obtain a recommended route scheme according to the travel cost expected value.
Referring to fig. 7, the step of calculating a recommended route according to the expected travel cost value includes steps S41 to S43.
Step S41: and calculating to obtain the recommended flow ratio of each road according to the travel cost expected value of each road.
The recommended flow rate ratio of each road is calculated according to the expected trip cost value of each road through the following formula:
wherein,represents tnAnd in the process that the time point goes out from the node i to the end point d, the recommended flow ratio from the node i to the node j is obtained.
i, j represent nodes in the logical graph network with the time dimension.
A (i) represents the set of end points of all edges starting from node i.
Is shown at tnThe time, the associated adaptive parameter of the edge (i, j).
Is shown at tnAnd the expected travel cost value of the travel from the node i to the terminal d at the moment.
Is at tnAnd at the moment, taking the node i as a starting point, all edges are opened to the expected value of the average travel cost of the point d.
The self-adaptive regulation can be carried out according to the proportion of vehicles receiving induction in the traffic system, the traffic conditions in the area, the evaluation result of the distribution scheme and the like, so as to avoid negative phenomena caused by improper traffic induction, such as aggregation, over-excitation reaction and the like. When in useWhen the vehicle selection strategy is relatively small, the vehicle selection strategy in the area is relatively concentrated and is suitable for the condition that the traffic flow in the area is less or the proportion of vehicles receiving induction is less, and when the vehicle selection strategy is relatively small, the vehicle selection strategy in the area is relatively concentrated, and the condition that the traffic flow in the area is less or the proportion of the vehiclesIf the value is too large, the behavior of the vehicle may become randomly uncontrollable.
Step S42: and obtaining a flow ratio suggestion matrix according to the recommended flow ratio of each road.
Step S43: and obtaining a recommended route scheme according to the flow ratio suggestion matrix.
The flow ratio suggestion matrix is obtained through the following formula according to the recommended flow ratio of each road:
wherein,represents tnAnd in the process that the time point goes out from the node i to the end point d, the recommended flow ratio from the node i to the node j is obtained.
N is an edge in the logic diagram network with the time dimension; d is a set of demand center points in the logic diagram network with the time dimension.
After the traffic matching suggestion matrix is obtained, a proper road can be selected for going out according to the traffic matching suggestion matrix, so that the road section with traffic jam is avoided, and the traffic passing efficiency is improved.
On the basis, please refer to fig. 8 in combination, the present disclosure further provides a traffic line distribution device 10. The traffic route distribution device 10 provided by the present disclosure includes an acquisition module 11, a calculation module 12, and a recommendation module 13.
The acquiring module 11 is configured to acquire traffic data in the current time period in real time.
The calculation module 12 is configured to input the traffic data into a traffic prediction model obtained through pre-training, so as to obtain a predicted travel time.
The calculation module 12 is further configured to calculate an expected trip cost value according to a pre-constructed logic diagram network with a time dimension and the predicted trip time.
And the recommending module 13 calculates a recommended route scheme according to the travel cost expected value.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific operation of the traffic route distribution apparatus 10 described above may refer to the corresponding process in the foregoing method, and will not be described in too much detail herein.
According to the traffic route distribution method and device, the traffic data in the current time period are obtained in real time, the traffic data are input into a traffic prediction model obtained through pre-training to obtain the predicted travel time, after the predicted travel time is obtained, a travel cost expected value is obtained through calculation according to a pre-constructed logic diagram network with time dimension and the predicted travel time, and then a recommended route scheme is obtained through calculation according to the travel cost expected value, so that the traffic smoothness is improved.
The traffic route distribution method and the traffic route distribution device have the advantages that traffic data in the current time period are obtained in real time, the traffic data are input into a traffic prediction model obtained through pre-training, the predicted travel time is obtained, therefore, when a distribution scheme is given to a local traffic network with excessively concentrated traffic, the upcoming change in the traffic network and the influence caused by a control strategy are pre-judged, the pre-judgment result is obtained, after the pre-judgment result, namely the predicted travel time, a travel cost expected value is obtained through calculation according to a pre-constructed logic diagram network with time dimension and the predicted travel time, a recommended route scheme is obtained through calculation according to the travel cost, therefore, the next strategy is selected on the pre-judgment result, and the accuracy of traffic distribution and the smoothness of traffic passage are improved.
In the several embodiments provided in the present disclosure, it should be understood that the disclosed apparatus and method may be implemented in other manners. The apparatus and method embodiments described above are illustrative only, as the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present disclosure may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, an electronic device, or a network device) to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The foregoing is illustrative of only alternative embodiments of the present disclosure and is not intended to limit the disclosure, which may be modified and varied by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Claims (10)
1. A method of traffic lane assignment, the method comprising:
acquiring traffic data in the current time period in real time;
inputting the traffic data into a traffic prediction model obtained by pre-training to obtain predicted travel time;
calculating to obtain a travel cost expected value according to a pre-constructed logic diagram network with a time dimension and the predicted travel time;
and calculating to obtain a recommended route scheme according to the travel cost expected value.
2. The traffic route distribution method according to claim 1, wherein the traffic prediction model is trained by:
extracting a travel time matrix, a distribution probability matrix, a weather characteristic matrix and an event characteristic matrix in a preset time period from a historical road travel time data set;
obtaining a long-term traffic change trend characteristic matrix, a simultaneous traffic characteristic matrix and a characteristic matrix of the instant traffic change trend according to the travel time matrix;
inputting the long-term traffic change trend characteristic matrix, the simultaneous traffic characteristic matrix, the characteristic matrix of the instant traffic change trend, the distribution probability matrix, the weather characteristic matrix and the event characteristic matrix into a deep convolutional neural network for training to obtain prediction output data;
and constructing a loss function based on the prediction output data, and adjusting model parameters according to the loss function, wherein the adjusted parameters are used as parameters of the traffic prediction model.
3. The traffic route distribution method according to claim 2, wherein the step of obtaining the long-term traffic change tendency feature matrix, the simultaneous-period traffic characteristic feature matrix, and the feature matrix of the instant traffic change tendency from the travel time matrix comprises:
according to the travel time matrix, obtaining a long-time traffic change trend characteristic matrix, a simultaneous time section traffic characteristic matrix and an instant traffic change trend characteristic matrix through the following formulas:
wherein,represents tiA travel time matrix of the time of day;representing a long-term traffic change trend characteristic matrix;representing a traffic characteristic feature matrix of the same time period;a feature matrix representing an instant traffic change trend; rt represents the time window length; wt represents a one week time period; st, mt denotes the length of the sampling interval, st<mt; ρ, θ represents the update weight ratio.
4. A traffic route allocation method according to claim 3, characterized in that the formula of said loss function is:
wherein theta represents a current traffic prediction model parameter; j (theta) is a loss value of the traffic prediction model when the parameter is theta; x is an input value of the traffic prediction model; n is the number of samples; i is a number from 1 to n; y is(i)The real output value corresponding to the ith input value; h is(θ)(x(i)) Predicting model parameters for post-training trafficThe predicted output data when θ is input and x is input.
5. The traffic route allocation method according to claim 1, wherein the logical graph network with time dimension is obtained by:
extracting the logic relation among roads in the traffic network from the geographic information, setting a demand center, and constructing a logic graph network by taking the traffic flow trend bifurcation point as a node and taking a traffic flow walking route as an edge;
and expanding the logic diagram network on a time dimension to obtain the logic diagram network with the time dimension.
6. The traffic route distribution method according to claim 1, wherein the step of calculating the expected travel cost value according to the pre-constructed logic diagram network with time dimension and the predicted travel time comprises:
according to a pre-constructed logic diagram network with a time dimension and the predicted travel time, calculating to obtain a travel cost expected value through the following formula:
wherein,is shown at tnThe travel cost expected value of the time point from the node i to the terminal d;
i, j, k represents a node in the logical graph network with the time dimension;
d represents an end point, D belongs to D, and D is a set of required central points in the logic diagram network with the time dimension;
is shown at tnTime, the predicted travel time to travel from node i to node j;
a (j) represents a set of end points of all edges starting from node j;
is shown at tnThe probability of selecting the node j to the node k when the time reaches the end point d;
is shown at tnThe time-to-end d selects the desired travel time required from node j to node k.
7. The traffic route distribution method according to claim 6, wherein the step of calculating a recommended route plan according to the expected travel cost value includes:
calculating to obtain a recommended flow ratio of each road according to the travel cost expected value of each road;
obtaining a flow ratio suggestion matrix according to the recommended flow ratio of each road;
and obtaining a recommended route scheme according to the flow ratio suggestion matrix.
8. The traffic route distribution method according to claim 7, wherein the step of calculating the recommended traffic ratio of each road according to the expected travel cost value of each road comprises:
according to the expected trip cost value of each road, calculating the recommended flow ratio of each road through the following formula:
wherein,represents tnIn the process that the time point goes out from the node i to the end point d, the recommended flow ratio from the node i to the node j is calculated;
i, j represents nodes in the logic diagram network with the time dimension;
a (i) represents the set of end points of all edges starting from the node i;
is shown at tnThe time, the relevant adaptive parameter of the edge (i, j);
is shown at tnThe travel cost expected value from the node i to the terminal d at the moment;
is at tnAnd at the moment, taking the node i as a starting point, all edges are opened to the expected value of the average travel cost of the point d.
9. The traffic route allocation method according to claim 7, wherein the step of obtaining the traffic proportion suggestion matrix according to the recommended traffic proportion of each road comprises:
according to the recommended flow ratio of each road, obtaining a flow ratio suggestion matrix through the following formula:
wherein,represents tnIn the process that the time point goes out from the node i to the end point d, the recommended flow ratio from the node i to the node j is calculated;
n is an edge in the logic diagram network with the time dimension; d is a set of demand center points in the logic diagram network with the time dimension.
10. The traffic route distribution device is characterized by comprising an acquisition module, a calculation module and a recommendation module;
the acquisition module is used for acquiring traffic data in the current time period in real time;
the calculation module is used for inputting the traffic data into a traffic prediction model obtained by pre-training to obtain predicted travel time;
the calculation module is further used for calculating to obtain a travel cost expected value according to a pre-constructed logic diagram network with a time dimension and the predicted travel time;
and the recommending module calculates a recommended route scheme according to the travel cost expected value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910483977.6A CN110211381B (en) | 2019-06-05 | 2019-06-05 | Traffic route distribution method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910483977.6A CN110211381B (en) | 2019-06-05 | 2019-06-05 | Traffic route distribution method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110211381A true CN110211381A (en) | 2019-09-06 |
| CN110211381B CN110211381B (en) | 2021-02-09 |
Family
ID=67790782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910483977.6A Active CN110211381B (en) | 2019-06-05 | 2019-06-05 | Traffic route distribution method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110211381B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111489553A (en) * | 2020-04-26 | 2020-08-04 | 百度在线网络技术（北京）有限公司 | Route planning method, device, equipment and computer storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101438334A (en) * | 2006-03-03 | 2009-05-20 | 因瑞克斯有限公司 | Dynamic time series prediction of future traffic conditions |
| US20130304367A1 (en) * | 2004-12-22 | 2013-11-14 | Hntb Holdings Ltd. | Optimizing traffic predictions and enhancing notifications |
| CN104169986A (en) * | 2012-02-17 | 2014-11-26 | 宝马股份公司 | Method for model construction for a travel-time database |
| CN104851309A (en) * | 2015-06-01 | 2015-08-19 | 南京邮电大学 | Traffic guidance strategy realization method |
| CN108389419A (en) * | 2018-03-02 | 2018-08-10 | 辽宁工业大学 | A kind of Dynamic Route Guidance Method of Vehicle |
| CN108847021A (en) * | 2018-06-07 | 2018-11-20 | 华南理工大学 | A kind of road network method for predicting considering heterogeneous user |
-
2019
- 2019-06-05 CN CN201910483977.6A patent/CN110211381B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130304367A1 (en) * | 2004-12-22 | 2013-11-14 | Hntb Holdings Ltd. | Optimizing traffic predictions and enhancing notifications |
| CN101438334A (en) * | 2006-03-03 | 2009-05-20 | 因瑞克斯有限公司 | Dynamic time series prediction of future traffic conditions |
| CN104169986A (en) * | 2012-02-17 | 2014-11-26 | 宝马股份公司 | Method for model construction for a travel-time database |
| CN104851309A (en) * | 2015-06-01 | 2015-08-19 | 南京邮电大学 | Traffic guidance strategy realization method |
| CN108389419A (en) * | 2018-03-02 | 2018-08-10 | 辽宁工业大学 | A kind of Dynamic Route Guidance Method of Vehicle |
| CN108847021A (en) * | 2018-06-07 | 2018-11-20 | 华南理工大学 | A kind of road network method for predicting considering heterogeneous user |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111489553A (en) * | 2020-04-26 | 2020-08-04 | 百度在线网络技术（北京）有限公司 | Route planning method, device, equipment and computer storage medium |
| CN111489553B (en) * | 2020-04-26 | 2022-02-25 | 百度在线网络技术（北京）有限公司 | Route planning method, device, equipment and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110211381B (en) | 2021-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11842271B2 (en) | Multi-scale multi-granularity spatial-temporal traffic volume prediction | |
| CN110361024B (en) | Method and system for dynamic lane-level vehicle navigation with vehicle group identification | |
| AU2018307710B2 (en) | Dynamically determining origin and destination locations for a network system | |
| CN107945507B (en) | Travel time prediction method and device | |
| CN107944605B (en) | Dynamic traffic path planning method based on data prediction | |
| US8229666B2 (en) | Generating and using pattern keys in navigation systems to predict user destinations | |
| CN108072381B (en) | Path planning method and device | |
| Chen et al. | Efficient traffic speed forecasting based on massive heterogenous historical data | |
| WO2015072349A1 (en) | Method for predicting destinations during travel | |
| Liu et al. | Holiday passenger flow forecasting based on the modified least-square support vector machine for the metro system | |
| DE102006030271A1 (en) | Navigation system and program for controlling the system | |
| CN113758494B (en) | Navigation path planning method, device, equipment and storage medium | |
| CN111310828B (en) | Target detection model fine-tuning method and device for ADAS scene | |
| US20180045527A1 (en) | Systems and Methods for Predicting Vehicle Fuel Consumption | |
| CN111860763A (en) | Model training method, parameter prediction method, model training device, parameter prediction device, electronic equipment and storage medium | |
| De Nunzio et al. | Bi-objective eco-routing in large urban road networks | |
| CN110211381B (en) | Traffic route distribution method and device | |
| Toledo et al. | Simulation-based optimization of HOT lane tolls | |
| Crişan et al. | Computational intelligence for solving difficult transportation problems | |
| US20210064428A1 (en) | Resource optimization with simultaneous trust region modeling | |
| CN109034494B (en) | Bus dispatching method | |
| CN113643535B (en) | Road traffic prediction method, device, equipment and medium based on smart city | |
| CN114766022A (en) | Vehicle dynamics and powertrain control system and method using multi-time domain optimization | |
| EP4310450A1 (en) | Systems and methods for public transit arrival time prediction | |
| CN114383617B (en) | Navigation route calculation method, device and equipment based on map neural network and storage medium |
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 |