CN108022425B

CN108022425B - Traffic condition prediction method and device and computer equipment

Info

Publication number: CN108022425B
Application number: CN201711396601.9A
Authority: CN
Inventors: 任若愚
Original assignee: Neusoft Corp
Current assignee: Neusoft Corp
Priority date: 2017-12-21
Filing date: 2017-12-21
Publication date: 2020-05-05
Anticipated expiration: 2037-12-21
Also published as: CN108022425A

Abstract

The invention provides a traffic condition prediction method, a traffic condition prediction device and computer equipment, wherein the method comprises the following steps: the method comprises the steps of obtaining congestion information of N time periods before a current time period aiming at a target cell, predicting first congestion information of the current time period according to the congestion information of the N time periods, determining a target time range corresponding to the current time period, obtaining historical congestion information in the target time range, obtaining second congestion information in the current time period according to the historical congestion information, obtaining an influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time period, obtaining third congestion information of the current time period according to the influence value, obtaining similar cells of the target cell, obtaining fourth congestion information in the current time period by using the similar cells, and obtaining target congestion information of the target cell according to at least two of the four obtained congestion information. By the method, the calculation complexity can be reduced, and the universality can be improved.

Description

Traffic condition prediction method and device and computer equipment

Technical Field

The present invention relates to the field of traffic technologies, and in particular, to a method, an apparatus, and a computer device for predicting traffic conditions.

Background

With the development of social economy and the improvement of the living standard quality of people, the number of private cars is more and more, and the problem of urban traffic jam caused by the private cars is more and more serious. Therefore, the method and the system have important significance for predicting the areas with the possibility of traffic jam, finding the traffic road conditions of the areas in time and shunting, and improving the urban service capacity and the travel efficiency of people.

The existing urban traffic road condition prediction method usually only considers the information of a specific road section in a city and only considers the peripheral condition of a target road section, and is lack of universality and large in limitation. In addition, the existing method generally only considers the influence of the vehicle speed on the congestion, and the analysis is carried out according to the road section when the prediction is carried out, so that the calculation complexity is high.

Disclosure of Invention

The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.

Therefore, a first objective of the present invention is to provide a traffic condition prediction method, which divides an urban area into a plurality of cells, and predicts a congestion condition of a target cell by considering various factors, so as to reduce the computational complexity, improve the universality, and solve the technical problems of high computational complexity and poor universality in the prior art.

A second object of the present invention is to provide a traffic condition prediction device.

A third object of the invention is to propose a computer device.

A fourth object of the invention is to propose a non-transitory computer-readable storage medium.

A fifth object of the invention is to propose a computer program product.

To achieve the above object, an embodiment of a first aspect of the present invention provides a traffic condition prediction method, including:

acquiring congestion information of N time periods before the current time period of a target cell aiming at any target cell in a city, and predicting first congestion information of the current time period according to the congestion information of the N time periods; wherein N is an integer and N is more than or equal to 1;

determining a target time range corresponding to the current time period, acquiring historical congestion information of the target cell in the target time range, and acquiring second congestion information in the current time period according to the historical congestion information;

acquiring an influence value of each adjacent cell adjacent to the target cell on the target cell in a previous time period, and acquiring third congestion information of the current time period according to the influence value;

similar cells similar to the target cell are inquired and obtained from the city, and fourth congestion information in the current time period is obtained by using the similar cells;

acquiring target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and/or the fourth congestion information; the target congestion information is used for representing the congestion state of the target cell in the current time slot.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the predicting, according to the congestion information of the N time slots, first congestion information of the current time slot includes:

and performing weighted calculation on the congestion information of the N time periods by using a preset linear model to obtain the first congestion information.

As another optional implementation manner of the embodiment of the first aspect of the present invention, after the obtaining the first congestion information, the method further includes:

calculating a variance between the historical congestion information and the first congestion information in the current time period;

when the variance is not the minimum, adjusting the weight in the linear model to obtain the first congestion information again until the variance is the minimum, and taking the first congestion information corresponding to the minimum variance as the final first congestion information.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the acquiring, according to the historical congestion information, second congestion information in the current time period includes:

inquiring and acquiring the historical congestion information matched with the time range according to the time range to which the current time period belongs;

acquiring factors influencing the current traffic jam state;

and acquiring an adjustment amount matched with the factors, and adjusting the historical congestion information by using the adjustment amount to obtain the second congestion information.

As another optional implementation manner of the embodiment of the first aspect of the present invention, after determining the target time range corresponding to the current time period, the method further includes:

acquiring congestion information before the target cell belonging to each time range according to each time range;

and performing statistical analysis on the previous congestion information to obtain the historical congestion information in the time range.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the acquiring an influence value of each neighboring cell adjacent to the target cell on the target cell in a previous time period, and acquiring third congestion information of the current time period according to the influence value includes:

for any one boundary of the target cell, determining the adjacent cell sharing the boundary with the target cell; wherein one of said borders corresponds to one of said adjacent cells;

acquiring a first number of vehicles and a first average vehicle speed which enter the adjacent cell from the target cell through the boundary in the previous time period, and acquiring a first sub-influence value of the adjacent cell on the boundary according to the first number of vehicles and the first average vehicle speed; acquiring a second number of vehicles and a second average vehicle speed which enter the target cell from the adjacent cell through the boundary in the previous time period, and acquiring a second sub-influence value of the adjacent cell on the boundary according to the second number of vehicles and the second average vehicle speed;

acquiring a third vehicle number and a third average vehicle speed of an adjacent cell entering the adjacent cell from the adjacent cell through another boundary of the adjacent cell in the previous time period, and acquiring a third sub-influence value of the adjacent cell on the boundary according to the third vehicle number and the third average vehicle speed; wherein another boundary of the neighboring cell is a boundary parallel to the boundary of the target cell;

acquiring a fourth number of vehicles and a fourth average vehicle speed which enter another adjacent cell from the target cell through another boundary of the target cell in the previous time period, and acquiring a fourth sub-influence value of the adjacent cell on the boundary according to the fourth number of vehicles and the fourth average vehicle speed; wherein another boundary of the target cell is a boundary parallel to the boundary of the target cell; the other neighboring cell shares the other boundary with the target cell;

obtaining the influence value of a single adjacent cell according to the first sub-influence value, the second sub-influence value, the third sub-influence value and the fourth sub-influence value;

and performing weighted calculation on the influence value of each adjacent cell to obtain the third congestion information.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the influence value of a single adjacent cell is obtained according to the first sub-influence value, the second sub-influence value, the third sub-influence value, and the fourth sub-influence value, and a specific calculation formula is as follows:

wherein C represents the impact value, F₁～F₄The first, second, third and fourth sub-influence values are represented respectively, the α and β are constants, and the η is an adjustment amount.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the querying and obtaining similar cells similar to the target cell from the city includes:

identifying cells having a similar congestion state transition tendency as the target cell as candidate cells from remaining cells of the city other than the target cell;

and calculating the similarity between the target cell and each candidate cell, and extracting the similar cells from all the candidate cells according to the similarity.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the calculating a similarity between the target cell and each candidate cell includes:

forming a matrix using the congestion information for the N time segments for the target cell and the congestion information for the N time segments for each candidate cell;

calculating the average value of each element value in the matrix;

and calculating the similarity between the target cell and each candidate cell according to the average value and the matrix.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the extracting the similar cells from all the candidate cells according to the similarity includes:

and selecting the candidate cell with the similarity exceeding a preset first threshold value as the similar cell.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the obtaining, by using the similar cell, fourth congestion information in the current time period includes:

for each similar cell, predicting the first congestion information of the similar cell in the current time slot by using the congestion information of the N time slots of the similar cell;

multiplying the first congestion information of the similar cells by the similarity corresponding to the similar cells to obtain a first numerical value of the similar cells;

adding the first numerical values of the similar cells to obtain a second numerical value, adding the similarity corresponding to the similar cells to obtain a third numerical value, and making a ratio of the second numerical value to the third numerical value to obtain the fourth congestion information.

acquiring historical congestion information of the similar cells in the target time range aiming at each similar cell;

multiplying the historical congestion information of the similar cells by the similarity corresponding to the similar cells to obtain a fourth numerical value of the similar cells;

adding the fourth numerical values of the similar cells to obtain a fifth numerical value, adding the similarity corresponding to the similar cells to obtain a sixth numerical value, and making a ratio of the fifth numerical value to the sixth numerical value to obtain the fourth congestion information.

As another optional implementation manner of the embodiment of the first aspect of the present invention, before acquiring, for any cell in a city, congestion information of N time slots before a current time slot of the target cell, the method further includes:

and carrying out cell division on the city.

As another optional implementation manner of the embodiment of the first aspect of the present invention, after the obtaining of the target congestion information of the target cell, the method further includes:

acquiring the current position and the running path of the current vehicle; wherein the driving path is planned according to a departure place and a destination of the current vehicle;

judging whether the current vehicle reaches the destination or not according to the current position and the running path;

if the judgment result is negative, extracting a first cell which is not driven by the current vehicle from the driving path;

acquiring the target congestion information of the first cell;

and predicting the driving time required by the current vehicle to drive to the first cell of the destination according to the target congestion information of the first cell.

As another optional implementation manner of the embodiment of the first aspect of the present invention, the predicting, according to the target congestion information of the first cell, a travel time period required for the current vehicle to travel to the first cell where the destination is located includes:

acquiring a vehicle running speed corresponding to the target congestion information for each first cell;

determining the running time of the current vehicle in the first cell according to the running speed of the vehicle and the running length of the current vehicle in the first cell;

and adding the running time of all the first cells to obtain the running time.

The traffic condition prediction method of the embodiment of the invention comprises the steps of obtaining congestion information of N time periods before a current time period of a target cell by aiming at any target cell in a city, predicting first congestion information of the current time period according to the congestion information of the N time periods, determining a target time range corresponding to the current time period, obtaining historical congestion information in the target time range, obtaining second congestion information in the current time period according to the historical congestion information, obtaining an influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time period, obtaining third congestion information of the current time period according to the influence value, obtaining similar cells of the target cell, obtaining fourth congestion information in the current time period by using the similar cells, and obtaining congestion information of the target cell according to the first congestion information, the second congestion information, the congestion information and the congestion information in the current time period, The third congestion information and/or the fourth congestion information obtain target congestion information of the target cell. The congestion condition of the target area is predicted from multiple aspects such as time, space, the target area and the similar area, so that the factors considered in the process of predicting the traffic condition are more comprehensive, the universality can be improved, the calculation complexity is reduced, and the technical problems of high calculation complexity and poor universality in the prior art are solved.

To achieve the above object, a second embodiment of the present invention provides a traffic condition prediction apparatus, including:

the time congestion prediction module is used for acquiring congestion information of N time periods before the current time period of a target cell aiming at any target cell in a city, and predicting first congestion information of the current time period according to the congestion information of the N time periods; wherein N is an integer and N is more than or equal to 1;

the spatial congestion prediction module is used for determining a target time range corresponding to the current time period, acquiring historical congestion information of the target cell in the target time range, and acquiring second congestion information in the current time period according to the historical congestion information;

the local correlation prediction module is used for acquiring an influence value of each adjacent cell adjacent to the target cell on the target cell in a previous time period and acquiring third congestion information of the current time period according to the influence value;

the global correlation prediction module is used for inquiring and acquiring similar cells similar to the target cell from the city and acquiring fourth congestion information in the current time period by using the similar cells;

the acquisition module is used for acquiring the target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and the fourth congestion information; the target congestion information is used for representing the congestion state of the target cell in the current time slot.

The traffic condition prediction device of the embodiment of the invention obtains congestion information of N time slots before the current time slot of a target cell by aiming at any target cell in a city, predicts first congestion information of the current time slot according to the congestion information of the N time slots, determines a target time range corresponding to the current time slot, obtains historical congestion information in the target time range, obtains second congestion information in the current time slot according to the historical congestion information, obtains an influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time slot, obtains third congestion information of the current time slot according to the influence value, obtains a similar cell of the target cell, obtains fourth congestion information in the current time slot by using the similar cell, and obtains congestion information of the target cell according to the first congestion information, the second congestion information, the congestion information and the congestion information of the target cell in the current time slot, The third congestion information and/or the fourth congestion information obtain target congestion information of the target cell. The congestion condition of the target area is predicted from multiple aspects such as time, space, the target area and the similar area, so that the factors considered in the process of predicting the traffic condition are more comprehensive, the universality can be improved, the calculation complexity is reduced, and the technical problems of high calculation complexity and poor universality in the prior art are solved.

To achieve the above object, a third embodiment of the present invention provides a computer device, including: a processor and a memory; wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the traffic condition prediction method according to the embodiment of the first aspect.

In order to achieve the above object, a fourth aspect of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a traffic condition prediction method as described in the first aspect of the present invention.

In order to achieve the above object, a fifth embodiment of the present invention provides a computer program product, wherein when the instructions of the computer program product are executed by a processor, the method for predicting traffic conditions according to the first embodiment is implemented.

Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.

Drawings

The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

fig. 1 is a schematic flow chart illustrating a traffic condition prediction method according to an embodiment of the present invention;

fig. 2 is a schematic flow chart illustrating a traffic condition prediction method according to a second embodiment of the present invention;

FIG. 3 is a schematic diagram of cell division in a city;

fig. 4 is a schematic flow chart of a traffic condition prediction method according to a third embodiment of the present invention;

fig. 5 is a schematic flow chart of a traffic condition prediction method according to a fourth embodiment of the present invention;

FIG. 6 is a schematic diagram of the effect of neighboring cells on a target cell;

fig. 7 is a schematic flow chart illustrating a traffic condition prediction method according to a fifth embodiment of the present invention;

FIG. 8(a) is a schematic diagram of analyzing traffic conditions based on road segments;

FIG. 8(b) is a schematic diagram of analyzing traffic conditions based on cells;

fig. 9 is a schematic structural diagram of a traffic condition prediction apparatus according to an embodiment of the present invention;

fig. 10 is a schematic structural diagram of a traffic condition prediction apparatus according to a second embodiment of the present invention;

fig. 11 is a schematic structural diagram of a traffic condition prediction apparatus according to a third embodiment of the present invention;

fig. 12 is a schematic structural diagram of a traffic condition prediction apparatus according to a fourth embodiment of the present invention;

fig. 13 is a schematic structural diagram of a traffic condition prediction apparatus according to a fifth embodiment of the present invention; and

fig. 14 is a schematic structural diagram of a computer device according to an embodiment of the present invention.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.

A traffic condition prediction method, apparatus, and computer device according to an embodiment of the present invention are described below with reference to the accompanying drawings.

Fig. 1 is a flowchart illustrating a traffic condition prediction method according to an embodiment of the present invention.

As shown in fig. 1, the traffic condition prediction method includes the steps of:

step 101, aiming at any target cell in a city, obtaining congestion information of N time slices before the current time slice of the target cell, and predicting first congestion information of the current time slice according to the congestion information of the N time slices; wherein N is an integer and N is not less than 1.

In real life, except for some abnormal situations, such as a major car accident, sudden traffic control, etc., the change of the traffic route of a region is often slow, and the change of the traffic state of a region has some precursor, such as more and more crowded, or more and more smooth. Therefore, in this embodiment, the traffic information in the current time period may be estimated according to the traffic information in a time period or a plurality of time periods before the current time period.

Therefore, in this embodiment, when the road condition information of the city needs to be predicted, the city is divided into the cells, then any one of all the cells in the city is used as the target cell, and the congestion state inside the target cell is analyzed according to the subsequent steps.

First, congestion information of N time slots before the current time slot of the target cell is acquired. And then, the congestion information of the current time slot is predicted by using the acquired congestion information of the previous N time slots and is used as the first congestion information.

Step 102, determining a target time range corresponding to the current time period, acquiring historical congestion information of the target cell in the target time range, and acquiring second congestion information in the current time period according to the historical congestion information.

The target time range is the same time range as the current time period in each period in the previous n periods, one period is one day, and n is a positive integer not less than 1. For example, if the current time period is 7: 00-8: 00 in the morning, the target time range may be 7: 00-8: 00 in the morning of the previous day, or may be 7: 00-8: 00 in the morning of each day in the previous week.

In some specific time periods, the traffic information in the same area is not changed much, and the traffic information of the target cells in the same time period in one stage is similar, for example, in the morning, the evening and the peak, the same road section is in a congestion state in the same time period every day. Therefore, for the same area, the historical traffic information in the same time period has higher referential property for predicting the traffic information in the current time period of the area.

Therefore, in this embodiment, a target time range corresponding to the current time period may be determined according to the current time period, historical congestion information of the target cell in the target time range may be obtained, and then congestion information in the current time period may be obtained according to the historical congestion information and may be used as the second congestion information.

In this embodiment, the congestion state corresponding to the target time range of the previous day may be used as the historical congestion information of the current time period. Optionally, the congestion state corresponding to the target time range of multiple consecutive days is counted, and the counted congestion state of the target time range is used as the historical congestion information of the current time period.

And 103, acquiring an influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time period, and acquiring third congestion information of the current time period according to the influence value.

In the cells divided in cities, except the cells at the boundary, each cell in the city has four adjacent city cells, and the inflow or outflow of traffic flow between the target cell and the four adjacent city cells has great influence on the road condition information of the target cell. For example, if the traffic flow of four adjacent city cells entering the target cell is more than the traffic flow of the target cell, and the average vehicle speed is slow, it is inevitable that the target city cell is congested in a previous time period, and otherwise, the target city cell becomes smooth.

Therefore, in this embodiment, the influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time period may be acquired, and the congestion information in the current time period may be acquired according to the influence value and used as the third congestion information.

It should be noted that, for each cell at the boundary, there are also four adjacent cells, and only the four cells include cells of adjacent cities, and the traffic information of the cell at the boundary is also affected by traffic conditions of the four adjacent cells. Therefore, when the target cell is a cell at the boundary, the congestion information of the current time slot may also be acquired according to the influence value of the adjacent four cells on the target cell in the previous time slot. Alternatively, city cells at the boundary may be directly ignored for disregard, since traffic flow at the city boundary is typically small.

It should be noted that, a specific implementation process for obtaining the third congestion information of the current time period according to the influence value will be given in the following content, and for avoiding repeated description, detailed description is not provided here.

And 104, inquiring and acquiring similar cells similar to the target cell from the city, and acquiring fourth congestion information in the current time period by using the similar cells.

In a city, although there is no direct connection between certain areas and traffic does not flow from one place to another far away in a short period of time, there may be greater similarities between these different areas, such as holiday parks, weekend commercial squares, weekend night courts, and the like. The areas are not related from the geographical point of view, but have many similarities from the point of view of their properties and contents, and accordingly, it can be inferred that the traffic states of the areas also have some degrees of correlation (positive correlation), so in this embodiment, the similar cells similar to the target cell can be obtained by searching all cells divided in the city through traversing the global, and the fourth congestion information of the target cell in the current time period is obtained by using the similar cells.

Generally, congestion states may include, for example, five levels of clear, substantially clear, light congestion, medium congestion, and heavy congestion. The congestion state is dependent on the congestion information, which is a discrete value from which the congestion state can be characterized.

In this embodiment, the vehicle speeds of the cells are collected and the average vehicle speed is calculated, and the congestion information is obtained according to the average vehicle speed and the traffic flow density in the cells, and the calculation formula of the congestion information is shown in formula (1).

Wherein, sensitivity_gridRepresenting the traffic density of the cell, Velocity_girdRepresenting the average vehicle speed in the cell, ξ being the value used for normalization, P_gridIndicating congestion information.

In the embodiment, the congestion information can be normalized to be between (0-1) by normalization. The interval (0-1) is divided in advance, for example, the interval is divided according to the interval of 0.2, 5 intervals (0-0.2), [ 0.2-0.4), [ 0.4-0.6), [ 0.6-0.8) and [ 0.8-1.0) are formed, wherein, (0-0.2) represents unblocked, [ 0.2-04) represents basically unblocked, [ 0.4-06) represents light congestion, [ 0.6-0.8) represents medium congestion, and [ 0.8-1.0) represents serious congestion. That is, each block section may correspond to a congestion state.

And 105, acquiring target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and/or the fourth congestion information.

The target congestion information is used for representing the congestion state of the target cell in the current time slot.

In this embodiment, after the first congestion information, the second congestion information, the third congestion information, and the fourth congestion information in the current time slot of the target cell are acquired, a weighted calculation may be performed according to at least two congestion information of the acquired four congestion information, and a calculation result may be used as the target congestion information of the target cell. For example, taking the example that the target congestion information is calculated by a weighted summation according to the congestion information of the current time slot under the four influence factors, the target congestion information may be calculated by the following formula (2).

S_i,t＝α*Y+β*S_i,t,n+γ*C_i+μ*S_g,t+θ(2)

Wherein Y is first congestion information of the target cell, S_i,t,nSecond congestion information for the ith cell (target cell), C_iThird Congestion information for the ith cell, S_g,tα, β, gamma and mu are weights of the first to fourth congestion information, respectively, and theta is a constant value and can be preset for adjusting the congestion information.

For example, if the calculated discrete value corresponding to the target congestion information is 0.7, it can be determined that the congestion state corresponding to the target congestion information is moderate congestion. And when the calculated target congestion information is 0.3, determining that the congestion state corresponding to the target congestion information is basically smooth.

The traffic condition prediction method of the embodiment obtains the congestion information of N time slots before the current time slot of the target cell by aiming at any target cell in the city, predicting first congestion information of the current time slot according to the congestion information of the N time slots, determining a target time range corresponding to the current time slot, acquiring historical congestion information in the target time range, and acquiring second congestion information in the current time period according to the historical congestion information, acquiring the influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time period, acquiring third congestion information of the current time slot according to the influence value, acquiring similar cells of the target cells, acquiring fourth congestion information of the current time slot by using the similar cells, and acquiring the target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and/or the fourth congestion information. The congestion condition of the target area is predicted from multiple aspects such as time, space, the target area and the similar area, so that the factors considered in the process of predicting the traffic condition are more comprehensive, the universality can be improved, the calculation complexity is reduced, and the technical problems of high calculation complexity and poor universality in the prior art are solved.

Fig. 2 is a flowchart illustrating a traffic condition prediction method according to a second embodiment of the present invention.

As shown in fig. 2, the traffic condition prediction method includes the steps of:

step 201, dividing the city into cells.

In this embodiment, for a city for which road condition prediction is required, a longitude and latitude boundary value of the city may be obtained, a rectangle is established by using the obtained boundary value, grid division is performed inside the rectangle, and the city is divided into a plurality of cells with equal size and consistent specification. As shown by the black rectangular frame in fig. 3, the black rectangular frame in fig. 3 is one of the divided cells.

202, acquiring congestion information of N time slots before the current time slot of a target cell aiming at any target cell in a city, and predicting first congestion information of the current time slot according to the congestion information of the N time slots; wherein N is an integer and N is not less than 1.

The congestion or the clear traffic in an area is gradually formed, for example, the traffic flow speed in the area is slower and slower, so in this embodiment, the congestion information of the current time slot may be predicted according to the congestion information of N consecutive time slots before the current time slot of the target cell.

Specifically, when the first congestion information of the current time slot is predicted according to the congestion information of the N time slots, the first congestion information may be obtained by performing weighted calculation on the congestion information of the N time slots by using a preset linear model.

Before linear calculation is performed by using the linear model, a window value of linear regression may be set, where the window value is the same as the number of the acquired time periods, that is, the window value is N. Assuming that the congestion information of the kth time slot is to be predicted, the congestion information of the previous k-N, k-N-1, k-N-2, …, k-1 time slot needs to be acquired. The congestion information of the first N time periods is used as input and input into a preset linear model, and first congestion information of the current time period can be obtained.

Suppose that the first congestion information of the current time slot of the target cell is recorded as Y, and the congestion information of the first N time slots adjacent to the current time slot is respectively recorded as Y₁，Y₂，…，Y_N-1，Y_NThe first congestion information may be represented by equation (3).

Wherein, w_iIndicates the ith congestion information Y_iThe weight of (c). According to the actual situation, the congestion information of the time slot closer to the current time slot has a greater influence on the congestion information of the current time slot, and therefore, w may be initially set_iBeing a sequence of increasing numbers, Y_NHas the largest weight value of Y₁The weight value of (3) is smallest.

Further, in a possible implementation manner of the embodiment of the present invention, the acquired first congestion information may be further optimized. Specifically, the variance between the historical congestion information and the first congestion information in the current time period may be calculated, when the variance does not reach the minimum, the weight in the linear model is adjusted to obtain the first congestion information again, the variance between the new first congestion information and the historical congestion information in the current time period is calculated again until the variance reaches a preset minimum value, and the corresponding first congestion information when the variance is minimum is used as the final first congestion information.

When the N pieces of historical congestion information are obtained, the variance between each piece of historical congestion information and the first congestion information can be calculated respectively, the obtained N variances are summed, if the result of the sum of the variances does not reach the preset minimum value, the weight in the linear model is adjusted until the result of the sum of the variances reaches the minimum value, and the corresponding first congestion information when the result of the sum of the N variances is the minimum value is used as the final first congestion information.

By optimizing the acquired first congestion information according to the historical congestion information of the current time period, the predicted first congestion information can better accord with the real road condition state, and the accuracy of prediction of the first congestion information is improved.

Step 203, determining a target time range corresponding to the current time period, acquiring historical congestion information of the target cell in the target time range, and acquiring second congestion information in the current time period according to the historical congestion information.

And 204, acquiring an influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time period, and acquiring third congestion information of the current time period according to the influence value.

And step 205, searching and acquiring similar cells similar to the target cell from the city, and acquiring fourth congestion information in the current time period by using the similar cells.

And step 206, acquiring target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and/or the fourth congestion information.

It should be noted that, in the present embodiment, the description of step 203 to step 206 may refer to the description of step 102 to step 105 in the foregoing embodiment, and is not repeated here.

Step 207, acquiring the current position and the running path of the current vehicle.

In this embodiment, the current position of the current vehicle may be obtained according to a driving recorder, a GPS navigator, an intelligent device terminal, and other devices installed in the current vehicle. Further, according to the departure place and the destination inputted by the user on the navigator or the like, a driving route can be recommended to the user, and the user can drive along the recommended route.

And step 208, judging whether the current vehicle reaches the destination or not according to the current position of the current vehicle and the running path of the current vehicle.

The position of the vehicle can be tracked during the travel of the vehicle. If the current position of the current vehicle is not at the destination, determining that the current vehicle still runs on the running path and does not reach the destination, and executing step 209; and if the current position of the current vehicle is at the destination, determining that the current vehicle has reached the destination, and ending.

In step 209, a first cell in which the vehicle is not currently running is extracted from the running route.

The first cell may be each cell that the current vehicle does not travel through, among all cells in which the travel path is located. For example, the driving route includes five cells, and from the departure point to the destination, the number of each cell is 1,2,3,4, and 5, respectively, and the target cell where the current vehicle is located is the cell with the number of 3, and then the first cell is the cell with the number of 4 and the cell with the number of 5.

In this embodiment, after the driving route of the current vehicle is acquired, the cells through which the driving route passes may be acquired according to the acquired driving route information, and the first cell in which the current vehicle does not drive may be extracted from the driving route according to the cell in which the current vehicle is located.

And step 210, acquiring the target congestion information of the first cell.

After the first cells are determined, the target congestion information of the first cells may be acquired for each first cell by the same method as the method for acquiring the target congestion information of the target cell.

And step 211, predicting the driving time required by the current vehicle to drive to the first cell where the destination is located according to the target congestion information of the first cell.

In this embodiment, after the target congestion information of the first cell is acquired, the travel time period required when the current vehicle travels to the first cell where the destination is located may be predicted according to the target congestion information of the first cell.

Specifically, the vehicle running speed corresponding to the target congestion information may be acquired for each first cell, for example, different congestion states and corresponding average vehicle speeds in the congestion states may be stored in advance, and a corresponding relationship between the target congestion information and the congestion states may be stored, after the target congestion information of the first cell is acquired, the congestion state of the first cell may be determined by querying the corresponding relationship, and further, when the first cell is determined to be in the congestion state, the average vehicle speed of the vehicle in the first cell may be determined, and the vehicle speed may be used as the vehicle running speed corresponding to the target congestion information. And determining the running time of the current vehicle in the first cell according to the running speed of the vehicle and the running length of the current vehicle in the first cell, and adding the running time of all the first cells to obtain the running time.

According to the traffic condition prediction method, the congestion condition of the target area is predicted from multiple aspects such as time (first congestion information), space (second congestion information), adjacent area (third congestion information) and similar area (fourth congestion information), so that the factors considered in the prediction of the traffic condition are more comprehensive, the universality can be improved, and the calculation complexity can be reduced. The method comprises the steps of obtaining a running path of a current vehicle, extracting a first cell which is not run by the current vehicle from the running path, obtaining target congestion information of the first cell, predicting the running time required when the current vehicle runs to the first cell where a destination is located according to the target congestion information of the first cell, predicting the time of a user reaching the destination, facilitating the schedule arrangement of the user according to the time of reaching the destination, and improving user experience.

In order to more clearly describe a specific implementation process of acquiring the second congestion information in the current time period according to the historical congestion information in the above embodiment, another traffic condition prediction method is provided in the embodiment of the present invention, and fig. 4 is a flowchart of a traffic condition prediction method provided in a third embodiment of the present invention.

As shown in fig. 4, based on the embodiment shown in fig. 1, step 102 may include the following steps:

step 301, according to the time range to which the current time period belongs, historical congestion information matched with the time range is inquired and obtained.

In this embodiment, a day may be taken as a period, and 24 hours of the day may be divided into a plurality of time ranges, for example, a time range including a duration of two hours may be divided into 12 time ranges; alternatively, a time range includes a time period of three hours, a day is divided into 8 time ranges, and the like. When the road condition information of the target cell is predicted, the time range of the current time period can be determined according to the current time period, and then the historical congestion information matched with the time range can be inquired and obtained according to the time range of the current time period.

The historical congestion information matched with the time range and obtained by query may be historical congestion information of the time range in the previous period (one day), or may be historical congestion information obtained after statistical analysis is performed on the historical congestion information of the time range in the previous several periods.

When only the historical congestion information of the previous period is acquired, the historical congestion information of the time range in the previous period can be directly acquired. For example, if the time range of the current time period is 7: 00-9: 00, the congestion information of the target cell can be acquired as the historical congestion information acquired by query within the time range of 7: 00-9: 00 in the previous day.

When historical congestion information of a few previous periods is acquired, congestion information before a target cell belonging to the time range can be acquired in the time range in each period, statistical analysis is performed on the congestion information before, and the historical congestion information in the time range is acquired. For example, assuming that historical congestion information of previous three periods is obtained, and the time range of the current time period is 7: 00-9: 00, the congestion information of the target cell within a time range of 7: 00-9: 00 in each day in three days before the current time period is obtained, and statistical analysis is performed on the obtained three congestion information, for example, by means of averaging, the obtained average value is used as the historical congestion information obtained by query.

Step 302, obtaining the current factors influencing the traffic jam state.

The factors influencing the traffic jam state can be one or more of weather, traffic control, festivals and the like.

In this embodiment, factors such as weather and festivals that affect the traffic congestion state when the target cell is in the current time period can be obtained.

And 303, acquiring an adjustment amount matched with the factor, and adjusting the historical congestion information by using the adjustment amount to obtain second congestion information.

In the embodiment, the influence of the influence factors on the congestion information in the previous congestion information can be analyzed according to different influence factors to obtain the influence degrees corresponding to different influence factors; or, the corresponding influence strength can be preset according to different influence factors according to experience. Therefore, after the factors influencing the traffic jam state at present are obtained, the influence strength matched with the factors can be obtained as an adjustment amount, and the historical jam information is adjusted by using the adjustment amount to obtain second jam information. The second congestion information of the target cell at the current time period may be represented by formula (4).

S_i,t,n＝W_i,t+T_i,t,n(4)

Wherein S is_i,t,nThe congestion information of the ith cell in the nth time slot on the nth day is represented, namely second congestion information; w_i,tRepresenting historical congestion information of the ith cell in a time range of the tth time period; t is_i,t,nAnd the adjustment quantity of the affected factors of the ith cell at the nth time period on the nth day is shown.

According to the traffic condition prediction method, the historical congestion information matched with the time range is obtained through inquiry according to the time range to which the current time period belongs, the factors influencing the traffic congestion state at present are obtained, the adjustment amount matched with the factors is obtained, and the historical congestion information is adjusted by the adjustment amount to obtain the second congestion information. By considering the influence of the historical congestion information and the traffic congestion state influence factors in the same time period, the authenticity and the accuracy of the acquired second congestion information can be improved.

In order to more clearly describe a specific implementation process of obtaining the third congestion information of the current time period according to the influence value of the adjacent cell in the foregoing embodiment, another traffic condition prediction method is provided in the embodiment of the present invention, and fig. 5 is a flowchart of a traffic condition prediction method provided in the fourth embodiment of the present invention.

As shown in fig. 5, step 103 may include the following steps based on the embodiment shown in fig. 1:

step 401, determining an adjacent cell sharing a boundary with a target cell aiming at any boundary of the target cell; wherein one boundary corresponds to one adjacent cell.

A city is divided into a plurality of unit grids, edges are used as boundaries among the unit grids, each unit grid is provided with four edges, each edge of each unit grid corresponds to an adjacent unit grid, and the adjacent unit grid of the unit grid at the edge position which is provided with at least one edge is another unit grid of the city. Therefore, in the present embodiment, for any one boundary of the target cell, the neighboring cell sharing the boundary with the target cell may be determined.

Step 402, acquiring a first number of vehicles and a first average vehicle speed which enter an adjacent cell from a target cell through a boundary in a previous time period, and acquiring a first sub-influence value of the adjacent cell on the boundary according to the first number of vehicles and the first average vehicle speed; and acquiring a second vehicle number and a second average vehicle speed which enter the target cell from the adjacent cell through the boundary in the previous time period, and acquiring a second sub-influence value of the adjacent cell on the boundary according to the second vehicle number and the second average vehicle speed.

Each cell in the city has four boundaries corresponding to traffic flows in four directions, and each direction has a certain number of vehicles and an average flow speed, so that the congestion condition on each boundary can be respectively obtained for the same cell and the congestion condition in each direction.

For each boundary of the target cell, the traffic flow may flow from the adjacent cell into the target cell through the boundary, or may flow from the target cell into the adjacent cell through the boundary. The traffic flow enters the adjacent cells from the target cell to play a role in smoothing the traffic condition of the target cell; and the traffic flow enters the target cell from the adjacent cell, and the traffic jam effect is realized on the traffic condition of the target cell.

In this embodiment, for any one boundary of the target cell, a first number of vehicles passing through the boundary from the target cell to enter an adjacent cell in a previous time period and a first average vehicle speed may be obtained, and a first sub-influence value of the adjacent cell on the boundary may be obtained according to the first number of vehicles and the first average vehicle speed, where the first sub-influence value may be referred to as an escape force. Meanwhile, a second number of vehicles and a second average speed which enter the target cell from the adjacent cell through the boundary in the previous time period are obtained, and a second sub-influence value of the adjacent cell on the boundary is obtained according to the second number of vehicles and the second average speed, wherein the second sub-influence value can be called as driving force.

FIG. 6 is a schematic diagram illustrating the effect of neighboring cells on a target cell. In fig. 6, a is a target cell, and B1, B2, B3, and B4 are four adjacent cells of the target cell. For the boundary shared between the target cell a and the adjacent cell B1, F1 in fig. 6 represents a first sub-influence value, and F2 is a second sub-influence value.

Step 403, acquiring a third number of vehicles and a third average vehicle speed, which enter an adjacent cell of the adjacent cell from the adjacent cell through another boundary of the adjacent cell in the previous time period, and acquiring a third sub-influence value of the adjacent cell on the boundary according to the third number of vehicles and the third average vehicle speed; wherein the other boundary of the neighboring cell is a boundary parallel to the boundary of the target cell.

In addition to the first and second sub-influence values having a direct influence on the target cell, the inflow and outflow of traffic between adjacent cells and adjacent cells of adjacent cells has an indirect influence on the traffic situation of the target cell.

Therefore, in this embodiment, the influence value of another boundary of the adjacent cell, which is parallel to the shared boundary of the adjacent cell and the target cell, on the boundary may also be obtained. Specifically, a third vehicle number and a third average speed that enter an adjacent cell of the adjacent cell from the adjacent cell through another boundary of the adjacent cell in a previous time period may be obtained, and a third sub-influence value of the adjacent cell on the boundary may be obtained according to the third vehicle number and the third average speed.

Taking fig. 6 as an example, the boundary shared by the cell B1 and the cell D2 is another boundary of the adjacent cell B1, and F3 in fig. 6 indicates the third sub-influence value.

And step 404, acquiring a fourth number of vehicles and a fourth average vehicle speed which enter another adjacent cell from the target cell through another boundary of the target cell in the previous time period, and acquiring a fourth sub-influence value of the adjacent cell on the boundary according to the fourth number of vehicles and the fourth average vehicle speed.

Wherein the other boundary of the target cell is a boundary parallel to the boundary of the target cell; another adjacent cell shares another boundary with the target cell.

Taking fig. 6 as an example, when the currently determined adjacent cell is B1, the other boundary of the target cell is a boundary shared between the target cell a and the adjacent cell B4, a fourth number of vehicles and a fourth average vehicle speed that have passed through the other boundary of the target cell and entered the other adjacent cell in a previous time period from the target cell a to the adjacent cell B4 through the other boundary of the target cell a are obtained, that is, a fourth number of vehicles and a fourth average vehicle speed that have passed through the boundary shared between the target cell a and the adjacent cell B4 and entered the adjacent cell B4 in the previous time period are obtained, and a fourth sub-influence value of the adjacent cell on the boundary obtained according to the fourth number of vehicles and the fourth average vehicle speed is F4 in fig. 6.

In this embodiment, the third sub-influence value and the fourth sub-influence value may be referred to as traction force, and are also escape forces of adjacent cells and the target cell at parallel boundaries of the shared boundary, which act as driving or blocking for the first sub-influence value and the second sub-influence value, and indirectly influence the traffic condition of the target cell. For the first sub-influence value, if the traffic flow speed represented by the third sub-influence value is higher, the first sub-influence value is driven, and the congestion condition of the target cell is favorably relieved; and if the traffic speed represented by the third sub-influence value is slower, the first sub-influence value is prevented, and the smoothness of the target cell is not facilitated.

And step 405, obtaining the influence value of a single adjacent cell according to the first sub-influence value, the second sub-influence value, the third sub-influence value and the fourth sub-influence value.

After a first sub-influence value, a second sub-influence value, a third sub-influence value and a fourth sub-influence value of the adjacent cell on each boundary of the target cell are obtained, the influence value of a single adjacent cell on the target cell can be obtained according to the first sub-influence value, the second sub-influence value, the third sub-influence value and the fourth sub-influence value.

Specifically, the influence value of a single adjacent cell on the target cell may be expressed by equation (5) as follows.

Wherein C represents an influence value, F₁～F₄The first, second, third and fourth sub-influence values are respectively represented, α and β are constants, and η is an adjustment quantity, which represents the influence of other factors on the influence value.

And 406, performing weighted calculation on the influence value of each adjacent cell to obtain third congestion information.

The target cell has four adjacent cells, and after the influence value on the target cell is obtained for each adjacent cell, the influence value of each adjacent cell can be weighted and calculated to obtain third congestion information.

Taking FIG. 6 as an example, the third congestion information for target cell A (denoted by C)_AExpressed) can be calculated by the following formula (6).

C_A＝λ₁·C_B1→A+λ₂·C_B2→A+λ₃·C_B3→A+λ₄·C_B4→A(6)

Wherein λ is_i(i ═ 1,2,3,4) denotes the weight of the influence value of each adjacent cell, C_B1→ARepresents the influence value, C, of the neighboring cell B1 on the target cell A_B2→ARepresenting adjacent cell B2 pairs of targetsInfluence value of cell A, C_B3→ARepresents the influence value, C, of the neighboring cell B3 on the target cell A_B4→ARepresenting the value of the influence of the neighboring cell B4 on the target cell a.

According to the traffic condition prediction method, the influence value of each adjacent cell on the target cell is obtained, the influence value of each adjacent cell is weighted to obtain the third congestion information, the influence of the adjacent region on the traffic condition of the target region is considered, and the traffic condition is predicted more comprehensively.

In order to describe more clearly the specific implementation process of querying and acquiring similar cells similar to the target cell from the city and acquiring fourth congestion information in the current time period by using the similar cells in the above embodiment, another traffic condition prediction method is proposed in the embodiment of the present invention, and fig. 7 is a schematic flow chart of the traffic condition prediction method provided in the fifth embodiment of the present invention.

As shown in fig. 7, step 104 may include the following steps based on the embodiment shown in fig. 1:

in step 501, cells having similar congestion state transition tendencies as the target cells are identified as candidate cells from the remaining cells of the city except the target cells.

The similar congestion state change trend refers to that two cells rise or fall simultaneously in the same continuous time period, and the overall change curve keeps basically consistent.

Specifically, the congestion state sequence of the target cell in the last several continuous time periods is obtained, and the congestion state sequence of the target cell is formed. And if the congestion information which represents the congestion state in the two congestion state sequences keeps the same change trend, the target cell and the remaining cells can be determined to be in similar congestion states. That is, the congestion state of the target cell shows an ascending trend in the latest continuous time period, and the congestion states of the remaining cells also show an ascending trend; and in the latest continuous time period, the congestion state of the target cell shows a descending trend, and the congestion states of the rest cells also show a descending trend, namely the congestion state change of the target cell and the congestion state change of the rest cells are basically consistent.

In this embodiment, a cell similar to the congestion state transition tendency of the target cell may be identified as a candidate cell from all cells in the city except for the target cell.

And 502, calculating the similarity between the target cell and each candidate cell, and extracting similar cells from all the candidate cells according to the similarity.

It cannot be determined that two cells are similar according to similar congestion state transformation trends, for example, the time series of the congestion state of one cell is {0,0,0,0,0,0,1}, and the time series of the congestion state of the other cell is {5,5,5,5,5,5,6}, where the time series of the congestion states of the two cells change smoothly, but the congestion degrees differ greatly. Therefore, in this embodiment, the candidate cells may be filtered, and the similarity between the target cell and each candidate cell is calculated, so as to extract similar cells from all candidate cells according to the similarity.

Specifically, the congestion information of the N time slots of the target cell and the congestion information of the N time slots of each candidate cell may be used to form a matrix, as shown in equation (7).

Wherein, grid_i+a(a ═ 0,1, …, m) denotes a cell number, t_j-b(b-0, 1, …, N) denotes the number of N time periods and the current time period,

the expression number is grid_i+aIs numbered t_j-bCongestion information within a time period of (a).

And then, calculating to obtain an average value according to the values of all elements in the matrix, and calculating the similarity between the target cell and each candidate cell according to the average value and the matrix. The similarity between two cells can be calculated by formula (8).

Wherein i represents an i-th candidate cell, g represents a target cell, D_i,gRepresenting the similarity between two cells, i and g, U represents the constructed matrix,

represents the average value, R, of the values of each element in the matrix_u,iRepresenting the corresponding value of cell i in the matrix, R_u,gRepresenting the corresponding value of the target cell g in the matrix.

And finally, selecting the candidate cells with the similarity exceeding a preset first threshold value as similar cells. Wherein the first threshold is preset.

And step 503, acquiring fourth congestion information in the current time period by using the similar cells.

As a possible implementation manner, when the similar cells are used to acquire the fourth congestion information in the current time period, the congestion information of the similar cells in the N time periods may be used to predict the first congestion information of the similar cells in the current time period for each similar cell; multiplying the first congestion information of the similar cells by the corresponding similarity of the similar cells to obtain a first numerical value of the similar cells; and adding the first numerical values of the similar cells to obtain a second numerical value, adding the similarity corresponding to the similar cells to obtain a third numerical value, and making a ratio of the second numerical value to the third numerical value to obtain fourth congestion information.

The above-described process of acquiring the fourth congestion information may be represented by formula (9).

Wherein S is_g,tFourth congestion information representing the target cell g at the current time period t; s_i,tRepresenting the first congestion information of the ith similar cell in the current time period t, m representing the total number of similar cells, D_i,gRepresenting the similarity between the ith similar cell and the target cell g.

As a possible implementation manner, when the similar cells are used to obtain the fourth congestion information in the current time period, the historical congestion information of the similar cells in the target time range may be obtained for each similar cell; multiplying the historical congestion information of the similar cells by the corresponding similarity of the similar cells to obtain a fourth numerical value of the similar cells; and adding the fourth numerical values of the similar cells to obtain a fifth numerical value, adding the similarity corresponding to the similar cells to obtain a sixth numerical value, and making a ratio of the fifth numerical value to the sixth numerical value to obtain fourth congestion information.

According to the traffic condition prediction method, the candidate cells are identified according to the transformation trend of the congestion condition, the similar cells of the target cell are extracted from the candidate cells by calculating the similarity, the fourth congestion information in the current time period is obtained by using the similar cells, the influence factors of the congestion information can be considered from the whole situation, the influence of the surrounding environment is considered, and the consideration factors are more comprehensive.

Fig. 8(a) is a schematic view illustrating traffic conditions analyzed based on a section, and fig. 8(b) is a schematic view illustrating traffic conditions analyzed based on a cell. As can be seen from fig. 8(a), when analyzing traffic conditions based on the section of road, the vehicle needs to pass 5 roads and 4 crossings from a to B. As can be seen from fig. 8(B), when the traffic condition is analyzed based on the cells, the vehicle passes through only 3 cells from a to B. Therefore, compared with the road section-based analysis method, the cell-based traffic condition prediction method has the advantages that the calculation complexity is reduced, and the prediction efficiency is improved.

In order to implement the above embodiment, the present invention further provides a traffic condition prediction device.

Fig. 9 is a schematic structural diagram of a traffic condition prediction apparatus according to an embodiment of the present invention.

As shown in fig. 9, the traffic condition prediction device 90 includes: temporal congestion prediction module 910, spatial congestion prediction module 920, local correlation prediction module 930, global correlation prediction module 940, and acquisition module 950. Wherein the content of the first and second substances,

the time congestion prediction module 910 is configured to, for any target cell in a city, obtain congestion information of N time slots before a current time slot of the target cell, and predict first congestion information of the current time slot according to the congestion information of the N time slots; wherein N is an integer and N is not less than 1.

Specifically, the temporal congestion prediction module 910 may perform weighted calculation on the congestion information of N time slots by using a preset linear model to obtain the first congestion information.

Further, in a possible implementation manner of the embodiment of the present invention, after the time congestion prediction module 910 obtains the first congestion information, a variance between the historical congestion information and the first congestion information in the current time period may also be calculated; and when the variance does not reach the minimum, adjusting the weight in the linear model, and obtaining the first congestion information again until the variance is minimum, wherein the first congestion information corresponding to the minimum variance is used as the final first congestion information.

The spatial congestion prediction module 920 is configured to determine a target time range corresponding to a current time period, obtain historical congestion information of a target cell in the target time range, and obtain second congestion information in the current time period according to the historical congestion information.

The local correlation prediction module 930 is configured to obtain an influence value of each neighboring cell adjacent to the target cell on the target cell in a previous time period, and obtain third congestion information of the current time period according to the influence value.

And the global correlation prediction module 940 is configured to query and acquire similar cells similar to the target cell from the city, and acquire fourth congestion information in the current time period by using the similar cells.

An obtaining module 950, configured to obtain target congestion information of a target cell according to the first congestion information, the second congestion information, the third congestion information, and/or the fourth congestion information; the target congestion information is used for representing the congestion state of the target cell in the current time slot.

Further, in a possible implementation manner of the embodiment of the present invention, as shown in fig. 10, on the basis of the embodiment shown in fig. 9, the traffic condition prediction apparatus 90 further includes:

a dividing module 900, configured to divide a city into cells.

The duration prediction module 960 is configured to obtain a travel path of a current vehicle, extract a first cell in which the current vehicle does not travel from the travel path, obtain target congestion information of the first cell, and predict a travel duration required when the current vehicle travels to the first cell in which the destination is located according to the target congestion information of the first cell.

Specifically, the duration prediction module 960 may obtain a vehicle running speed corresponding to the target congestion information for each first cell when the running duration required when the current vehicle runs to the first cell where the destination is located is predicted according to the target congestion information of the first cell; determining the running time of the current vehicle in the first cell according to the running speed of the vehicle and the running length of the current vehicle in the first cell; and adding the running time of all the first cells to obtain the running time.

The method comprises the steps of obtaining a running path of a current vehicle, extracting a first cell which is not run by the current vehicle from the running path, obtaining target congestion information of the first cell, predicting the running time required when the current vehicle runs to the first cell where a destination is located according to the target congestion information of the first cell, predicting the time of a user reaching the destination, facilitating the schedule arrangement of the user according to the time of reaching the destination, and improving user experience.

Further, in a possible implementation manner of the embodiment of the present invention, as shown in fig. 11, on the basis of the embodiment shown in fig. 9, the spatial congestion prediction module 920 may include:

and the query obtaining unit 921 is configured to query and obtain historical congestion information matched with the time range according to the time range to which the current time period belongs.

Alternatively, when only the historical congestion information of the previous cycle (1 day) is acquired, the query acquisition unit 921 may directly acquire the historical congestion information of the time range in the previous cycle.

Alternatively, when acquiring historical congestion information of several previous cycles, the query acquiring unit 921 may acquire congestion information of a target cell belonging to the time range for the time range in each cycle, perform statistical analysis on the congestion information, and acquire historical congestion information of the time range.

And the obtaining factor unit 922 is configured to obtain a current factor affecting the traffic congestion state.

And an adjusting unit 923, configured to obtain an adjustment amount matched with the factor, and adjust the historical congestion information by using the adjustment amount to obtain second congestion information.

By considering the influence of the historical congestion information and the traffic congestion state influence factors in the same time period, the authenticity and the accuracy of the acquired second congestion information can be improved.

Further, in a possible implementation manner of the embodiment of the present invention, as shown in fig. 12, on the basis of the embodiment shown in fig. 9, the local correlation prediction module 930 may include:

a determining unit 931 configured to determine, for any one boundary of the target cell, an adjacent cell that shares a boundary with the target cell; wherein one boundary corresponds to one adjacent cell.

The obtaining sub-influence value unit 932 is configured to obtain a first number of vehicles entering an adjacent cell from the target cell through the boundary in a previous time period and a first average vehicle speed, and obtain a first sub-influence value of the adjacent cell on the boundary according to the first number of vehicles and the first average vehicle speed; acquiring a second number of vehicles entering the target cell from the adjacent cell through the boundary in the previous time period and a second average vehicle speed, and acquiring a second sub-influence value of the adjacent cell on the boundary according to the second number of vehicles and the second average vehicle speed; acquiring a third vehicle number and a third average vehicle speed of an adjacent cell entering the adjacent cell from the adjacent cell through another boundary of the adjacent cell in the previous time period, and acquiring a third sub-influence value of the adjacent cell on the boundary according to the third vehicle number and the third average vehicle speed; wherein the other boundary of the adjacent cell is a boundary parallel to the boundary of the target cell; acquiring a fourth number of vehicles and a fourth average vehicle speed which enter another adjacent cell from the target cell through another boundary of the target cell in the previous time period, and acquiring a fourth sub-influence value of the adjacent cell on the boundary according to the fourth number of vehicles and the fourth average vehicle speed; wherein the other boundary of the target cell is a boundary parallel to the boundary of the target cell; another adjacent cell shares another boundary with the target cell.

And an influence value calculation unit 933, configured to obtain an influence value of a single adjacent cell according to the first sub-influence value, the second sub-influence value, the third sub-influence value, and the fourth sub-influence value.

Specifically, the influence value calculation unit 933 may calculate the influence value of a single adjacent cell according to the above equation (5).

And the weighting calculation module 934 is configured to perform weighting calculation on the influence value of each adjacent cell to obtain third congestion information.

Further, in a possible implementation manner of the embodiment of the present invention, as shown in fig. 13, on the basis of the embodiment shown in fig. 9, the global correlation prediction module 940 may include:

an identifying unit 941, configured to identify, as a candidate cell, a cell having a similar congestion state transition tendency to the target cell from the remaining cells of the city except for the target cell.

An extracting module 942 is configured to calculate a similarity between the target cell and each candidate cell, and extract similar cells from all candidate cells according to the similarity.

Specifically, when the extracting module 942 is used to calculate the similarity between the target cell and each candidate cell, it may first use the congestion information of the target cell in N time periods and the congestion information of each candidate cell in N time periods to form a matrix, calculate an average value of values of each element in the matrix, and calculate the similarity between the target cell and each candidate cell according to the average value and the matrix. Further, the extracting module 942 may select a candidate cell with a similarity degree exceeding a preset first threshold as the similar cell.

An obtaining unit 943 is configured to obtain the fourth congestion information in the current time period by using the similar cells.

As one possible implementation, the obtaining unit 943 may predict, for each similar cell, the first congestion information of the current time period of the similar cell by using the congestion information of the N time periods of the similar cell; multiplying the first congestion information of the similar cells by the corresponding similarity of the similar cells to obtain a first numerical value of the similar cells; and adding the first numerical values of the similar cells to obtain a second numerical value, adding the similarity corresponding to the similar cells to obtain a third numerical value, and making a ratio of the second numerical value to the third numerical value to obtain fourth congestion information.

As a possible implementation manner, the obtaining unit 943 may obtain, for each similar cell, historical congestion information of the similar cell within the target time range; multiplying the historical congestion information of the similar cells by the corresponding similarity of the similar cells to obtain a fourth numerical value of the similar cells; and adding the fourth numerical values of the similar cells to obtain a fifth numerical value, adding the similarity corresponding to the similar cells to obtain a sixth numerical value, and making a ratio of the fifth numerical value to the sixth numerical value to obtain fourth congestion information.

It should be noted that the explanation of the embodiment of the traffic condition prediction method is also applicable to the traffic condition prediction apparatus of the embodiment, and the implementation principle is similar, and is not repeated here.

The traffic condition prediction apparatus of the present embodiment obtains the congestion information of N time slots before the current time slot of a target cell by aiming at any target cell in a city, predicting first congestion information of the current time slot according to the congestion information of the N time slots, determining a target time range corresponding to the current time slot, acquiring historical congestion information in the target time range, and acquiring second congestion information in the current time period according to the historical congestion information, acquiring the influence value of each adjacent cell adjacent to the target cell on the target cell in the previous time period, acquiring third congestion information of the current time slot according to the influence value, acquiring similar cells of the target cells, acquiring fourth congestion information of the current time slot by using the similar cells, and acquiring the target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and/or the fourth congestion information. The congestion condition of the target area is predicted from multiple aspects such as time, space, the target area and the similar area, so that the factors considered in the process of predicting the traffic condition are more comprehensive, the universality can be improved, the calculation complexity is reduced, and the technical problems of high calculation complexity and poor universality in the prior art are solved.

In order to implement the above embodiments, the present invention further provides a computer device.

Fig. 14 is a schematic structural diagram of a computer device according to an embodiment of the present invention. As shown in fig. 14, the computer apparatus 140 includes: a processor 141 and a memory 142. The processor 141 reads the executable program code stored in the memory 142 to run a program corresponding to the executable program code, so as to implement the traffic condition prediction method according to the foregoing embodiment.

In order to implement the above embodiments, the present invention also proposes a non-transitory computer-readable storage medium on which a computer program is stored, which when executed by a processor implements the traffic condition prediction method as described in the foregoing embodiments.

In order to implement the above embodiments, the present invention further provides a computer program product, wherein instructions in the computer program product, when executed by a processor, implement the traffic condition prediction method according to the foregoing embodiments.

In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.

Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.

The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.

It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.

In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.

Claims

1. A traffic condition prediction method, comprising:

acquiring target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and/or the fourth congestion information; the target congestion information is used for representing the congestion state of the target cell in the current time slot;

the acquiring an influence value of each adjacent cell adjacent to the target cell on the target cell in a previous time period, and acquiring third congestion information of the current time period according to the influence value includes:

2. The method of claim 1, wherein predicting first congestion information for the current time segment from the congestion information for the N time segments comprises:

3. The method of claim 2, wherein after obtaining the first congestion information, further comprising:

4. The method of claim 1, wherein obtaining second congestion information for the current time period based on the historical congestion information comprises:

acquiring factors influencing the current traffic jam state;

5. The method of claim 4, wherein after determining the target time range corresponding to the current time period, further comprising:

6. The method according to claim 1, wherein the influence value of a single adjacent cell is obtained according to the first sub-influence value, the second sub-influence value, the third sub-influence value, and the fourth sub-influence value, and a specific calculation formula is as follows:

7. The method of claim 1, wherein the querying from the city for similar cells to the target cell comprises:

8. The method of claim 7, wherein calculating the similarity between the target cell and each candidate cell comprises:

calculating the average value of each element value in the matrix;

calculating the similarity between the target cell and each candidate cell according to the average value and the matrix, wherein the calculation formula of the similarity between the target cell and each candidate cell is as follows:

wherein i represents the ith candidate cell, g represents the target cell, D_i,gRepresenting the degree of similarity between the candidate cell i and the target cell g, U representing the matrix formed,

represents the average value, R, of the values of each element in the matrix_u,iRepresenting the corresponding value, R, of the candidate cell i in the matrix_u,gRepresenting the corresponding value of the target cell g in the matrix.

9. The method of claim 8, wherein the extracting the similar cells from all candidate cells according to the similarity comprises:

10. The method according to claim 9, wherein the obtaining of the fourth congestion information in the current time period using the similar cell comprises:

11. The method according to claim 9, wherein the obtaining of the fourth congestion information in the current time period using the similar cell comprises:

12. The method according to any one of claims 1 to 11, wherein before acquiring congestion information of N time slots before the current time slot of the target cell for any one cell in the city, the method further comprises:

and carrying out cell division on the city.

13. The method according to any one of claims 1-11, wherein after obtaining the target congestion information of the target cell, further comprising:

acquiring the target congestion information of the first cell;

14. The method according to claim 13, wherein the predicting a travel time period required for the current vehicle to travel to a first cell where a destination is located according to the target congestion information of the first cell comprises:

and adding the running time of all the first cells to obtain the running time.

15. A traffic condition prediction apparatus, comprising:

the acquisition module is used for acquiring the target congestion information of the target cell according to the first congestion information, the second congestion information, the third congestion information and the fourth congestion information; the target congestion information is used for representing the congestion state of the target cell in the current time slot;

wherein the local correlation prediction module is specifically configured to:

16. A computer device comprising a processor and a memory;

wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory for implementing the traffic condition prediction method according to any one of claims 1 to 14.

17. A non-transitory computer-readable storage medium having stored thereon a computer program, which when executed by a processor implements a traffic condition prediction method according to any one of claims 1-14.