Traffic flow basic graph construction and parameter estimation method fusing data of different time intervals

Abstract

The application discloses a traffic flow basic graph construction and parameter estimation method for fusing data of different time intervals, and comprises the following steps: establishing the correlation between the traffic flow of a short time interval and the traffic flow of a long time interval; determining the probability distribution function of the traffic density of the short time interval by using the traffic flow density of different time intervals, the traffic flow of the long time interval and the flow variance; constructing the traffic flow basic graph by using the traffic flow speed of the short time interval, the traffic flow of the long time interval, the traffic density of different time intervals and the variance of the traffic flow of the short time interval; and estimating the parameters of the traffic flow basic graph model by using the Bayesian hierarchical model and the updating method, so that the model fitting goodness is best after multiple iterations and updating. The application can effectively reduce the error of the parameter estimation value of the traffic flow basic graph, is helpful to estimate and predict the traffic state of a road section under the condition that complete traffic data is lacked, and has application value for relieving the urban road traffic congestion.

Description

Technical Field

[0001] This invention relates to a method for constructing a basic traffic flow map and estimating parameters by integrating data from different time intervals, belonging to the technical field of intersection between data information processing technology and urban road traffic control technology. Background Technology

[0002] With the rapid development of intelligent transportation systems, traffic data collection technologies are evolving rapidly, and data sources are becoming increasingly diversified. This brings new opportunities and challenges to the study of basic traffic flow graphs. Against this backdrop, in-depth research into the theory of basic traffic flow graphs and the establishment of accurate and reliable traffic flow models are of significant theoretical and practical value for improving road network operational efficiency and optimizing traffic management strategies.

[0003] Traditional traffic flow basic map modeling methods are mostly based on a single data source, building the model through data aggregation at fixed time intervals (usually 1-5 minutes). This data primarily comes from fixed detectors, such as loop detectors, microwave radar, millimeter-wave radar, gantries, and video. However, while fixed detectors can provide high-precision point-based traffic flow data, their high equipment cost makes it difficult to achieve full road network coverage. Therefore, relying on fixed detectors to obtain real-time, full-area road network traffic data is impractical.

[0004] The development of intelligent transportation systems in cities has expanded the sources of traffic information. For road sections lacking high-precision traffic data, cost-effective methods for acquiring traffic information include integrating multi-source data. However, data from different sources often differ significantly in terms of temporal resolution and measurement accuracy. For example, a traffic speed detection system provides traffic speed data at 2-minute intervals, representing high temporal resolution speed data recorded over a short time period, while a traffic flow monitoring system provides traffic flow data at 30-minute intervals, representing low temporal resolution flow data recorded over a long time period.

[0005] Existing methods for fusion of traffic data at different time resolutions are relatively simplistic. Most studies address time resolution mismatch by aggregating high-time-resolution data over low-time-resolution intervals. However, averaging high-time-resolution data over low-time-resolution intervals results in the loss of dynamic features related to traffic state changes. Specifically, when high-time-resolution data exhibits significant changes, the aggregated data from low-resolution intervals deviates significantly from the actual data, leading to large biases in model parameter estimation. Therefore, effectively fusing multi-source, heterogeneous traffic data with different time resolutions to establish a more accurate and reliable basic traffic flow model has become a critical scientific problem urgently needing to be solved in the field of traffic engineering. Summary of the Invention

[0006] Purpose of the Invention: The purpose of this invention is to provide a method for constructing and estimating the parameters of a basic traffic flow map that integrates data from different time intervals. This method establishes the correlation between unobservable short-time interval data and observable long-time interval data, considers the system bias caused by data aggregation, constructs a basic traffic flow map that integrates data from different time intervals, and estimates the parameters of the basic traffic flow map. This effectively reduces the error in the estimated values ​​of the basic traffic flow map parameters and helps to estimate and predict the traffic conditions of road segments in the absence of complete traffic data. It has application value in alleviating urban road traffic congestion.

[0007] Technical solution: The above objectives are achieved through the following technical solution:

[0008] In a first aspect, the present invention provides a method for constructing and estimating parameters of a basic traffic flow map by fusing data from different time intervals, comprising the following steps:

[0009] Establish the traffic flow f in the first time interval and the traffic flow in the second time interval. The correlation is such that the first time interval is less than the second time interval, and the traffic flow f within the first time interval follows the relationship of... The mean, The normal distribution of variance is defined as follows: Where N is a normal distribution, It is the variance of traffic flow across multiple first time intervals within the second time interval;

[0010] Determine the distribution function of the traffic density k during the first time interval, where the traffic density k during the first time interval follows a distribution function of k. The mean, The normal distribution of variance is defined as follows: in These are traffic flow densities at different time intervals, consisting of the traffic flow velocity u in the first time interval and the traffic flow rate in the second time interval. The calculation yields the following expression: It is the variance of the traffic density of multiple first time intervals within the second time interval, which is composed of the traffic flow density of different time intervals. Traffic flow during the second time interval and the variance of traffic flow The calculation yields the following expression:

[0011] The basic traffic flow graph is constructed, and its expression is:

[0012]

[0013] Where F(·) is the basic traffic flow graphical model, ω1,…,ω nThese are model parameters. It is the traffic density k within the first time interval and the traffic density at different time intervals. The deviation between them, ε kc It is a deviation from the basic traffic flow diagram model;

[0014] A Bayesian hierarchical model and update method are used to analyze the basic traffic flow graph parameters ω1,…,ω. n Make an estimate.

[0015] Furthermore, when using a Bayesian hierarchical model and update method to estimate the parameters of the basic traffic flow map, we assume that the traffic flow velocity u follows an expected value of μ. u variance is The normal distribution of is expressed as:

[0016]

[0017] in

[0018] Traffic density k in the first time interval and traffic density in different time intervals Deviation between Obedience expectation is variance is The normal distribution of is expressed as:

[0019]

[0020] variance It follows an inverse gamma distribution IG, and its expression is:

[0021]

[0022] Where α k and λ k It is a parameter of the inverse gamma distribution.

[0023] Furthermore, the traffic density k within the first time interval and the traffic density at different time intervals... Deviation between Expectations The expression is:

[0024]

[0025] in It is F in The second partial derivative at point .

[0026] Traffic density k in the first time interval and traffic density in different time intervals Deviation between variance The expression is:

[0027]

[0028] in It is F in The first-order partial derivative at that point.

[0029] Furthermore, when estimating the parameters of the basic traffic flow graph model, the posterior function... The expression is:

[0030]

[0031] Where p(u|ω1,…,ω) n ,α k ,λ k ) is the probability of u, derived from the normal distribution. get; yes The probability from the normal distribution get; yes The probability of the inverse gamma distribution IG(α) k ,λ k ) obtain; p(ω1)…p(ω n )p(α k )p(λ k ) is ω1,…,ω n ,α k ,λ k The prior distribution of .

[0032] Furthermore, the prior distributions of the parameters in the basic traffic flow graphical model adopt a normal distribution N(0,10000) and a uniform distribution U(0,10000), with expressions ω1~N(0,10000),…,ω n ~N(0,10000),α k ~U(0,10000),λ k ~U(0,10000).

[0033] Preferably, the Defined as:

[0034]

[0035] Where u f and k o These are the parameters of the traffic flow model: free-flow velocity and traffic density at maximum capacity.

[0036] Preferably, the first time interval is defined as a 2-minute time interval, and the second time interval is defined as a time interval of 30 minutes, 60 minutes, or 120 minutes.

[0037] In a second aspect, the present invention provides a computer system, including a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor, wherein the computer program / instructions, when executed by the processor, implement the steps of the aforementioned method for constructing a basic traffic flow map and estimating parameters by fusing multi-time-resolution data.

[0038] Thirdly, the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the aforementioned method for constructing a basic traffic flow graph and estimating parameters by fusing multi-time-resolution data.

[0039] Beneficial Effects: This invention proposes a method for constructing a basic traffic flow map and estimating its parameters by integrating data from different time intervals. A probability distribution function is used to establish the relationship between unobservable short-interval traffic flow and observable long-interval traffic flow, as well as the relationship between short-interval traffic density and traffic density across different time intervals. A basic traffic flow map is constructed using short-interval traffic flow velocity, long-interval traffic flow, traffic density across different time intervals, and the variance of short-interval traffic volume. A Bayesian hierarchical model and update method are used to estimate the parameters of the basic traffic flow map, and multiple iterations are performed to achieve the best fit. Experiments show that the method proposed in this invention can effectively reduce the bias of the estimated parameters of the basic traffic flow map, helping to estimate and predict the traffic conditions of road segments in the absence of complete traffic data, and has application value in alleviating urban road traffic congestion. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of a traffic flow basic graph construction method that integrates data from different time intervals according to an embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram of the traffic flow basic graph parameter estimation framework based on Bayesian hierarchical structure of the present invention. Detailed Implementation

[0042] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0043] like Figure 1 As shown in the figure, an embodiment of the present invention discloses a method for constructing a basic traffic flow map and estimating parameters by fusing data from different time intervals, comprising the following steps:

[0044] The first step involves constructing a basic traffic flow map based on data from different time intervals, considering the first time interval (short-term traffic flow speed), the second time interval (long-term traffic flow), traffic density across different time intervals, and the variance of traffic flow. This specifically includes:

[0045] (1) Regarding the traffic flow f within a short time interval t and the traffic flow within a long time interval T Establish a probability distribution function. The traffic flow f within a short time interval t follows a probability distribution function. The mean, The variance is a normal distribution N, i.e. in It is the variance of traffic flow over M short time intervals within a long time interval T.

[0046] (2) Determine the probability distribution function of traffic density k within the short time interval t. The traffic density k within the short time interval t is... The mean, The variance is a normal distribution N, i.e. in It represents traffic flow density at different time intervals, consisting of traffic flow velocity u in short time intervals and traffic flow volume in long time intervals. The calculation shows that, It is the variance of traffic density over M short time intervals within a long time interval T, which is determined by the traffic flow density at different time intervals. Traffic flow over a long time interval T and the variance of traffic flow The calculation shows that,

[0047] (3) Use the traffic flow velocity u within a short time interval t and the traffic flow rate within a long time interval T. Traffic density at different time intervals and the variance of traffic volume over short time intervals The basic traffic flow graph is constructed, and its expression is:

[0048]

[0049] Where F(·) is the basic traffic flow graphical model, ω1,…,ω n These are model parameters. It is the traffic density k within a short time interval and the traffic density at different time intervals. The deviation between them, ε kc It is a deviation in the basic traffic flow diagram model.

[0050] In this embodiment, the traffic density k within a short time interval and the traffic density at different time intervals are... Deviation between It follows a normal distribution N, and its expression is:

[0051]

[0052] in and These are the traffic density k within a short time interval and the traffic density at different time intervals, respectively. Deviation between Expectation and variance.

[0053] Determine the traffic density k within a short time interval and the traffic density at different time intervals. Deviation between Expectations The expression is:

[0054]

[0055] in It is F in The second partial derivative at point .

[0056] Determine the traffic density k within a short time interval and the traffic density at different time intervals. Deviation between variance The expression is:

[0057]

[0058] in It is F in The first-order partial derivative at that point.

[0059] Deviation ε of the basic traffic flow diagram model kc It follows a normal distribution N, and its expression is:

[0060]

[0061] in The deviation ε of the basic traffic flow diagram model kc The variance of follows an inverse gamma distribution IG, and its expression is:

[0062]

[0063] Where α k and λ k It is a parameter of the inverse gamma distribution.

[0064] For example, in this embodiment, the short time interval t is defined as a 2-minute time interval, and the long time interval T is defined as a time interval of 30 minutes, 60 minutes, and 120 minutes.

[0065] The traffic flow model F(·) for constructing the basic traffic flow graph is defined as:

[0066]

[0067] Where u f It is the free-flow velocity, k o It is the traffic density when the traffic capacity is at its maximum.

[0068] The basic traffic flow graph model is expressed as follows:

[0069]

[0070] in It is the traffic density k within a short time interval and the traffic density at different time intervals. The deviation between them has a distribution function as follows:

[0071] expect The expression is:

[0072]

[0073] in It is F in The second partial derivative at point .

[0074] variance The expression is:

[0075]

[0076] in It is F in The first-order partial derivative at that point.

[0077] The second step involves using a Bayesian hierarchical model and update method to estimate the parameters of the basic traffic flow map.

[0078] Traffic flow velocity u follows the expected value μ u variance is The normal distribution N is given by the expression:

[0079]

[0080] in

[0081] Figure 2This embodiment illustrates the framework for estimating basic traffic flow graph parameters based on a Bayesian hierarchical structure.

[0082] According to Bayesian theory, the posterior function The expression is: where p(u|u f ,k o ,α k ,λ k ) is the probability of u, derived from the normal distribution. get; yes The probability from the normal distribution get; yes The probability of the inverse gamma distribution IG(α) k ,λ k ) obtain; p(u f )p(k o )p(α k )p(λ k ) is u f ,k o ,α k ,λ k The prior distribution of .

[0083] Furthermore, the prior distributions of the parameters in the aforementioned basic traffic flow graphical model adopt a normal distribution N(0,10000) and a uniform distribution U(0,10000), and their expressions are u f ~N(0,10000),k o ~N(0,10000),α k ~U(0,10000),λ k ~U(0,10000).

[0084] The Bayesian update method is used to update the basic traffic flow graph parameters u. f ,k o Estimate the fit and iterate repeatedly to achieve the best fit.

[0085] To demonstrate the accuracy of this invention, traffic flow data from major road sections in Hong Kong, China, were selected for case analysis. Basic traffic flow maps based on data from different time intervals were constructed, and parameter estimation was performed on the basic map model. The results are shown in Table 1. To compare with methods for constructing basic traffic flow maps based on multi-time-resolution data, parameter estimation results for basic traffic flow maps constructed using average data over long time intervals are also presented in Table 1. The comparison results show that the proposed method for constructing and estimating basic traffic flow maps by integrating data from different time intervals can effectively reduce the bias in the estimated parameters of the basic traffic flow maps.

[0086] Table 1. Estimation results of basic traffic flow map parameters

[0087]

[0088] The present invention discloses a computer system including a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor. When the computer program / instructions are executed by the processor, they implement the steps of a traffic flow basic map construction and parameter estimation method that integrates data from different time intervals as described in the foregoing embodiments.

[0089] The present invention discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of a traffic flow basic map construction and parameter estimation method that integrates data from different time intervals as described in the foregoing embodiments.

[0090] The program / instruction code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program / instruction code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program / instruction code causes the steps of the methods of the present invention to be performed. The program / instruction code can be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a standalone software package, or entirely on a remote machine or server. All aspects not detailed in this invention are well-known to those skilled in the art.

Claims

1. A method for constructing a basic traffic flow map and estimating parameters by fusing multi-temporal resolution data, characterized in that: Includes the following steps: Establish traffic flow within the first time interval Traffic flow during the second time interval The correlation relationship; where the first time interval is shorter than the second time interval, and the traffic flow within the first time interval... Obey The mean, The variance follows a normal distribution. It is the variance of traffic flow across multiple first time intervals within the second time interval; Determine the traffic density within the first time interval The distribution function; traffic density in the first time interval. Obey The mean, The variance follows a normal distribution. It represents the traffic flow density at different time intervals, derived from the traffic flow velocity within the first time interval. Traffic flow during the second time interval Calculated; It is the variance of the traffic density of multiple first time intervals within the second time interval, which is composed of the traffic flow density of different time intervals. Traffic flow during the second time interval and the variance of traffic flow in multiple first time intervals within the second time interval. Calculated; The basic traffic flow graph is constructed, and its expression is: ;in It is a basic traffic flow graph model. These are model parameters. Traffic density within the first time interval Traffic density at different time intervals The deviation between them It is a deviation from the basic traffic flow diagram model; The model parameters in the basic traffic flow graph model are updated using a Bayesian hierarchical model and update method. The estimation includes: setting the traffic flow speed within the first time interval. Obeying expectations variance is The normal distribution ;deviation Obeying expectations variance is Normal distribution; variance Follows an inverse gamma distribution , ;in and It is a parameter of the inverse gamma distribution.

2. The method for constructing a basic traffic flow map and estimating parameters by fusing multi-time resolution data according to claim 1, characterized in that: expect The expression is: ;in yes exist The second partial derivative at point .

3. The method for constructing a basic traffic flow map and estimating parameters by fusing multi-time resolution data according to claim 1, characterized in that: Traffic density in the first time interval Traffic density at different time intervals Deviation between The expression for the variance is: ;in yes exist The first-order partial derivative at that point.

4. The method for constructing a basic traffic flow map and estimating parameters by fusing multi-time resolution data according to claim 1, characterized in that: When estimating model parameters in the basic traffic flow graph model, the posterior function The expression is: in yes The probability from the normal distribution get; yes The probability from the normal distribution get; yes The probability from the inverse gamma distribution get; yes The prior distribution of .

5. The method for constructing a basic traffic flow map and estimating parameters by fusing multi-time resolution data according to claim 4, characterized in that: The prior distribution of the model parameters in the basic traffic flow graph model is a normal distribution. and uniform distribution Its expression ,…, , , .

6. The method for constructing a basic traffic flow map and estimating parameters by fusing multi-time resolution data according to claim 1, characterized in that: The first time interval is defined as a 2-minute time interval, and the second time interval is defined as a time interval of 30 minutes, 60 minutes, or 120 minutes.

7. A computer system comprising a memory, a processor, and computer programs / instructions stored in the memory and executable on the processor, characterized in that: When the computer program / instructions are executed by the processor, they implement the steps of the traffic flow basic graph construction and parameter estimation method according to any one of claims 1-6, which integrates multi-time-resolution data.

8. A computer program product comprising a computer program / instructions, characterized in that: When the computer program / instructions are executed by the processor, they implement the steps of the traffic flow basic graph construction and parameter estimation method according to any one of claims 1-6, which integrates multi-time-resolution data.

Images