A method for dynamic estimation of highway traffic state based on ETC data

Abstract

This invention relates to a method for dynamic estimation of highway traffic conditions based on ETC data. The method includes the following steps: taking the road segment between adjacent ETC gantries as the research unit based on the spatial location of the highway ETC gantries; acquiring current and historical ETC data for consecutive passage through adjacent gantries A-B of the highway; determining the sampling period; cleaning the historical data of the research road segment; statistically analyzing the free-flow time t1 of the research road segment at historical times; calculating the current travel time t2 of the research road segment, and the travel time ratio TTI = t2 / t1, and determining the traffic condition. The data foundation of this invention is based on the widely existing ETC gantry system on highways, which has the advantages of easy implementation and promotion; it considers the impact of weather conditions, service areas in the road segment, and highway entrance and exit ramps on traffic flow data statistics, eliminating interference from outliers, thereby improving the accuracy of traffic condition identification.

Description

Technical Field

[0001] This invention relates to the field of traffic engineering technology, specifically a method for dynamic estimation of highway traffic conditions based on ETC data. Background Technology

[0002] In recent years, with the development of the highway network and the advancement of urbanization, the mileage and scale of the national highway network have increased significantly, leading to frequent traffic congestion due to increased traffic demand. ETC gantry systems are dedicated systems built along highway sections, possessing functions such as segmented toll calculation and vehicle image recognition. For highway transportation networks, ETC gantry systems are widespread and can provide reliable continuous data. Therefore, effectively utilizing ETC data to identify traffic conditions and provide a basis for highway management measures is crucial for improving highway service levels and driving safety.

[0003] Currently, most research on highway traffic state estimation relies on multi-source data collection using video sensors, loop detectors, etc., and then utilizes traffic theory or machine learning algorithms to assess traffic flow. For example, Chinese invention patent CN116665456A, a method for assessing traffic state by combining high-dimensional index dimensionality reduction, estimates traffic state by constructing a high-dimensional index set and then reducing its dimensionality. However, it fails to adequately consider the current deployment of highway data collection equipment; different types of sensors exist at various points on highways, and their data recording formats differ, making data fusion difficult. Given the current widespread and continuous presence of ETC gantry systems on highways, traffic state identification based on ETC data warrants further research. For instance, Chinese invention patent CN116386324A, an online method for identifying traffic segment congestion based on ETC data, determines the congestion value of a study segment based on average vehicle speed and traffic volume. However, it fails to consider the impact of service areas and highway entrance / exit ramps on traffic volume and average speed calculations within the study segment. Current research based on ETC data rarely considers the impact of weather conditions, vehicles entering and exiting service areas, and vehicles using highway ramps on traffic condition estimation, resulting in low accuracy of the research results.

[0004] Given the aforementioned shortcomings of existing research, there is an urgent need for a dynamic estimation method for highway traffic conditions based on ETC data, which should consider the impact of weather conditions, highway service areas, and highway entrance and exit ramps on traffic flow, thereby improving the accuracy of traffic condition identification. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a dynamic estimation method for highway traffic conditions based on ETC data. Based on this method, the traffic conditions between any two adjacent gantries on a highway at any given time can be calculated. The spatial location between two adjacent gantries is used as the study segment unit. ETC data of the studied highway segment is collected to construct the data foundation. The free-flow travel time of this segment is defined based on the mean of the 95% confidence interval of travel time without congestion under the same weather conditions in historical periods. By comparing the current travel time with the free-flow travel time, the current traffic conditions are estimated. Furthermore, the method considers the interference of weather conditions, service areas, and highway entrance / exit ramps on traffic flow, improving the accuracy of the research results and providing a basis for highway management and decision-making.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for dynamic estimation of highway traffic status based on ETC data, the method comprising the following steps:

[0008] S1. Road segments are divided according to the spatial location of the ETC gantries on the highway, and the road segment between two adjacent gantries is used as the research unit for traffic status identification.

[0009] S2, obtains the current time period ETC data of consecutively passing through adjacent gantries AB of the highway through the ETC system, as well as the historical ETC data of the same road segment under the same weather conditions at the same time.

[0010] S3 uses a preset unit time period as the sampling period to analyze the distribution trend of vehicle travel time under free flow conditions at historical moments of the road segment. The data is cleaned in combination with whether the road segment includes highway entrance / exit ramps and service areas, so as to build the data foundation for estimating the free flow travel time of the road segment.

[0011] S4, calculate the time taken for vehicles passing through gantry A to reach gantry B within n preset time units during the same weather period in historical time when no congestion occurred. Define the average t1 of the 95% confidence interval of the travel time of vehicles passing through two adjacent gantries as the free-flow time to pass through the two gantries.

[0012] S5. Calculate the average value t2 of the 95% confidence interval of the travel time of vehicles continuously passing through gantry AB within the preset time period at the current moment, calculate the travel time ratio TTI = t2 / t1, and judge the current traffic state by the obtained TTI value of the current time period. Iterate continuously to realize the dynamic estimation of the traffic state of the research road segment in the current time period.

[0013] As a further technical solution of the present invention, in step S1, the road segment between two adjacent gantries is selected as the unit road segment for traffic status identification. The elements to be counted include "mileage between the two gantries", "whether there is a service area between the two gantries", and "whether there are entrance / exit ramps between the two gantries".

[0014] As a further technical solution of the present invention, in step S2, the collected ETC data elements include "weather conditions", "highway name", "gantry number", "vehicle passage time", "driving direction", "vehicle image" and "identified license plate number".

[0015] As a further technical solution of the present invention, in step S3, for vehicles passing through the road segment between the two gantries, the traffic conditions of vehicles in the road segment are divided into four situations according to whether the road segment includes entrance / exit ramps and service areas: continuously passing through gantries AB, entering the service area after passing through gantry A, exiting the highway from the exit ramp after passing through gantry A, and entering the highway from the entrance ramp and passing through gantry B; for vehicles entering or exiting the highway in the middle of segment AB, only one gantry will record the vehicle's passage and remove it; secondly, according to the distribution map of the passage time, vehicles entering the service area are judged and removed, and only vehicles continuously passing through gantries AB are retained as the data basis for estimating the free-flow time of the road segment.

[0016] As a further technical solution of the present invention, in steps S4 to S5, the average of the 95% confidence interval of the passage time of vehicles continuously passing through gantry AB in n preset time periods under the same weather conditions in historical time is selected as the free-flow time t1 of the study road segment; the average of the 95% confidence interval of the passage time of vehicles continuously passing through AB in the preset time period of the current study period is selected as the passage time t2 of the current traffic state, and then the travel time ratio TTI of the current period of the study road segment is calculated as TTI=t2 / t1. This is used as an indicator to estimate the current traffic state of the study road segment, and the dynamic estimation of the traffic state of the study road segment is realized through continuous iterative updates.

[0017] Furthermore, in step S5, if the estimated current travel time t2 is less than the free-flow time t1, then TTI is defined as 1; if the estimated current travel time t2 is greater than twice the free-flow time t1, then TTI is defined as 2. The specific traffic state level classification based on TTI can be determined according to the actual traffic flow characteristics and road speed limits.

[0018] Compared with the prior art, the beneficial effects of the present invention are:

[0019] 1. This invention utilizes the existing continuous ETC gantry data on highways to accurately calculate changes in road traffic conditions without requiring additional equipment on the highway network. It fully considers the current status of highway data collection equipment and data elements, making it more universal and applicable.

[0020] 2. This invention takes into account the interference of weather conditions, service areas, and highway entrance and exit ramps on traffic flow. It removes interfering data based on the data collected by the gantry system, thereby improving the accuracy of traffic condition estimation results.

[0021] 3. The judgment method of the present invention fully considers the real historical traffic conditions of the research road segment, which can more scientifically estimate the current road traffic conditions, and achieve dynamic estimation of traffic conditions through continuous iteration. Attached Figure Description

[0022] Figure 1 This is a flowchart of a method for dynamic estimation of highway traffic conditions based on ETC data.

[0023] Figure 2 This is a schematic diagram of the road segment in the research unit of this invention.

[0024] Figure 3 This is the construction process of the free-flowing time data basis in the embodiment.

[0025] Figure 4 This is the process for determining vehicles entering the service area in this embodiment.

[0026] Figure 5 This is used to determine vehicles that continuously pass through the AB gantry in the example.

[0027] Figure 6 The example shows the trend of traffic status (TTI) during the current time period. Detailed Implementation

[0028] The technical solution of this patent will be further described in detail below with reference to specific embodiments.

[0029] like Figures 1 to 6 As shown, a method for dynamic estimation of highway traffic status based on ETC data is described, the method comprising the following steps:

[0030] S1. Road segments are divided according to the spatial location of the ETC gantries on the highway, and the road segment between two adjacent gantries is used as the research unit for traffic status identification.

[0031] S2, obtains the current time period ETC data of consecutively passing through adjacent gantries AB of the highway through the ETC system, as well as the historical ETC data of the same road segment under the same weather conditions at the same time.

[0032] S3 uses a preset unit duration (30 min) as the sampling period to analyze the distribution trend of vehicle travel time under free flow conditions at historical moments of the road segment. The data is cleaned in combination with whether the road segment includes highway entrance / exit ramps and service areas, so as to build the data foundation for estimating the free flow travel time of the road segment.

[0033] S4, calculate the time taken for vehicles passing through gantry A to reach gantry B within n preset time units (30 min) during the period when no congestion occurred under the same weather conditions in historical time. Define the average t1 of the 95% confidence interval of the travel time of vehicles passing through two adjacent gantries as the free flow time to pass through the two gantries.

[0034] S5. Calculate the average value t2 of the 95% confidence interval of the travel time of vehicles continuously passing through gantry AB within the preset time period (5 min) at the current time. Calculate the travel time ratio TTI = t2 / t1. Based on the obtained TTI value of the current time period, judge the current traffic state and continuously iterate to realize the dynamic estimation of the traffic state of the research road segment in the current time period.

[0035] In this embodiment, in step S1, the road segment between two adjacent gantries is selected as the unit road segment for traffic status identification. The elements to be counted include "mileage between the two gantries", "whether there is a service area between the two gantries", and "whether there are entrance / exit ramps between the two gantries".

[0036] In this embodiment, the ETC data elements collected in step S2 include "weather conditions", "highway name", "gantry number", "vehicle passage time", "driving direction", "vehicle image" and "identified license plate number".

[0037] In step S3 of this embodiment, for vehicles passing through the road segment between two gantries, the traffic conditions of vehicles within the road segment are divided into four cases according to whether the road segment includes entrance / exit ramps and service areas: continuously passing through gantries AB, entering the service area after passing through gantry A, exiting the highway from the exit ramp after passing through gantry A, and entering the highway from the entrance ramp and passing through gantry B. For vehicles entering or exiting the highway in the middle of segment AB, only one gantry will record the vehicle's passage, and these vehicles are removed. Next, according to the distribution map of travel time, vehicles entering the service area are judged and removed, and only vehicles continuously passing through gantries AB are retained as the data basis for estimating the free-flow travel time of this road segment.

[0038] In this embodiment, in steps S4 to S5, the average of the 95% confidence interval of the passage time of vehicles continuously passing through gantry AB within n preset unit durations (30 min) of historical time periods without congestion under the same weather conditions is selected as the free-flow time t1 of the study road segment; the average of the 95% confidence interval of the passage time of vehicles continuously passing through AB within a preset duration (5 min) of the current study period is selected as the passage time t2 of the current traffic state, and then the travel time ratio TTI of the current period of the study road segment is calculated as TTI = t2 / t1. This is used as an indicator to estimate the current traffic state of the study road segment, and the dynamic estimation of the traffic state of the study road segment is achieved through continuous iterative updates.

[0039] Furthermore, in step S5, if the estimated current travel time t2 is less than the free-flow time t1, then TTI is defined as 1; if the estimated current travel time t2 is greater than twice the free-flow time t1, then TTI is defined as 2. The specific traffic state level classification based on TTI can be determined according to the actual traffic flow characteristics and road speed limits.

[0040] To facilitate a better understanding of the technical solution of the present invention by those skilled in the art, specific embodiments of the present invention are provided below:

[0041] This invention is based on ETC data from a 13.93km long section of the Chengdu-Nanchong Expressway (downbound). The maximum speed limit is 120km / h, and the minimum speed limit is 60km / h. Within these speed limits, the travel time is approximately 7-14 minutes. Considering the impact of service areas and expressway entrance / exit ramps on traffic flow, a method for dynamic traffic state estimation is established. This invention provides a method for dynamic traffic state estimation of expressways based on ETC data, comprising the following steps:

[0042] S1. Based on the spatial location of the highway ETC gantry, road segments are divided, and the road segment between two adjacent gantries AB is selected as the road segment unit for traffic status identification.

[0043] Specifically, the road segment between two adjacent gantries A and B is selected as the road segment unit for study. It is necessary to determine the basic information of gantries A and B, mainly including: gantry number, gantry station number, the expressway to which they belong, and the direction of travel (up or down). In addition, it is also necessary to determine whether there are service areas or expressway entrance / exit ramps between the two gantries, as shown in Table 1 below:

[0044] Table 1. Information on the studied road sections

[0045]

[0046] S2, obtain the current time period ETC data of the adjacent gantry AB through the highway ETC system and the historical time period ETC data of the gantry AB under the same weather conditions. The elements in the data include weather conditions, vehicle images, license plate recognition images, vehicle passage time through the gantry, and vehicle type.

[0047] Specifically, weather conditions can be retrieved from weather websites. The elements included in the vehicle information recorded by ETC passing through the gantry are shown in Table 2 below:

[0048] Table 2

[0049] license plate number Vehicle type Through gantry time driving direction License plate color Weather conditions Sichuan0001 small car 2023-07-14 13:30:01 Downward blue sunny Sichuan0002 small car 2023-07-14 13:30:05 Downward blue sunny Sichuan0003 small car 2023-07-14 13:30:05 Downward green sunny Sichuan0004 small car 2023-07-14 13:30:16 Downward green sunny

[0050] S3 uses a 30-minute sampling period to analyze the traffic flow information required for estimating the free-flow time of the road segment. It collects data on vehicles that continuously pass through gantries A and B within 30 minutes. It also cleans the data source by considering whether the road segment includes entrance / exit ramps and service areas, removing vehicles that pass through highway entrance / exit ramps between gantries A and B and vehicles that enter service areas. Only vehicles that continuously pass through gantries A and B are retained as the data basis for constructing the free-flow time estimation of the road segment.

[0051] Specifically, the sampling period can be determined based on local traffic volume characteristics. The data base required to calculate the free-flow time between gantry A and B includes the following elements: Table 3 below.

[0052] Table 3

[0053] license plate number Vehicle type Time through gantry A Time through gantry B Trip Time Sichuan0001 small car 2023-07-14 13:30:01 2023-07-14 13:38:40 00:08:39 Sichuan0002 small car 2023-07-14 13:30:05 2023-07-14 13:38:42 00:08:37 Sichuan0003 small car 2023-07-14 13:30:05 2023-07-14 13:37:42 00:07:37 Sichuan0004 small car 2023-07-14 13:30:16 2023-07-14 13:43:22 00:13:06

[0054] S4. Statistically calculate the time it takes for vehicles passing through gantry A to reach gantry B within n 30-minute intervals during historical periods without traffic congestion under the same weather conditions. Define the average travel time t1 of vehicles passing through two adjacent gantries consecutively as the free-flow time between the two gantries. This will eliminate the interference of data anomalies and retain the representativeness of data features as much as possible.

[0055] Specifically, in this embodiment, a 30-minute time period with no historical traffic congestion was selected in the study road segment AB under the same weather conditions. The travel time of vehicles continuously passing through gantry A to gantry B within the 30-minute period was recorded. The average value of the travel time statistics within the 95% confidence interval was taken as the travel time under free flow conditions for this road segment. In the formula: t1 represents the travel time under free flow conditions in the research section; This represents the travel time of the i-th vehicle in the j-th sampling; n represents the number of samplings. This embodiment only samples once to study the free-flow time of the road segment. .

[0056] S5, calculate the average of the 95% confidence interval of the travel time of vehicles continuously passing between gantry A and B within the current 5 minutes as the travel time of this road segment under the current traffic conditions. In the formula: t2 represents the current travel time of the road segment under study; Let represent the travel time of the i-th vehicle; calculate the travel time ratio TTI = t2 / t1, and use the obtained TTI value for the current time period to judge the current traffic state, and continuously iterate to achieve dynamic estimation of the traffic state of the road segment under study in the current time period.

[0057] Specifically, when the travel time t2 of the current period is less than the free-flow time t1, TTI is defined as 1; when the travel time t2 of the current period is greater than twice the free-flow time t1, TTI is defined as 2. In different application scenarios, the correspondence between traffic status levels and TTI can be calibrated according to the actual traffic flow characteristics and speed limits. In this embodiment, the time period in which accidents occur on the same road segment under the same weather conditions is selected as the current time period. Using 5-minute time units, a total of 175 minutes of TTI changes are calculated. Figure 6 The current traffic conditions are classified into levels according to the TTI, as shown in Table 4 below:

[0058] Table 4

[0059] TTI [1，1.2] [1.2，1.4] [1.4，1.6] [1.6，1.8] [1.8，2.0] Operating status Smooth Basically unobstructed Mild congestion Moderate congestion Severe congestion

[0060] This invention provides a dynamic estimation method for highway traffic conditions based on ETC data, enabling dynamic estimation of highway conditions. The data foundation of this invention is based on the widely distributed ETC gantry system on highways, offering advantages in ease of implementation and promotion. It considers the impact of weather conditions, service areas within road sections, and highway entrance / exit ramps on traffic flow data statistics, resulting in more accurate traffic condition estimation. The method uses the 95% confidence average of travel times over n historical time periods as the free-flow travel time, eliminating the interference of outliers. The judgment results for vehicles entering service areas can provide a reference for service area management.

[0061] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0062] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for dynamically estimating the traffic state on a highway based on ETC data, characterized in that, The method comprises the following steps: S1, according to the spatial position of the expressway ETC gantry, the road section is divided, and the road section between two adjacent gantries is taken as a research unit for traffic state recognition; S2, the current period ETC data of the vehicle continuously passing through adjacent gantries A-B of the expressway is obtained through the ETC system, and the ETC data under the same weather at the same historical time is obtained; S3, taking a preset unit time length as a sampling period, the distribution trend of the vehicle passing time of the road section under the free flow state at the historical time is analyzed, the data is cleaned in combination with whether the road section contains the expressway entrance and exit ramp and the service area, so as to construct the data basis for estimating the free flow passing time of the road section; S4, the time of the vehicle passing through the gantry A to the gantry B in n preset unit time lengths under the same weather at the historical time is counted, and the average value t1 of the 95% confidence interval of the vehicle travel time continuously passing through the adjacent two gantries is defined as the free flow passing time of the two gantries; S5, the average value t2 of the 95% confidence interval of the vehicle travel time continuously passing through the gantries A-B in the preset time length at the current time is calculated, the travel time ratio TTI=t2 / t1 is calculated, the value of the obtained TTI of the current period is obtained, the current traffic state is judged, and the dynamic estimation of the traffic state of the research road section in the current period is realized through iteration.

2. The method of claim 1, wherein, In step S1, the road section between the adjacent two gantries is selected as the unit road section for traffic state recognition, and the elements to be counted include "mileage between the two gantries", "whether there is a service area between the two gantries", "whether there is an entrance and exit ramp between the two gantries". 3.The method of claim 1, wherein, In step S2, the collected ETC data elements include "weather condition", "expressway name", "gantry stake number", "vehicle passing time", "driving direction", "vehicle image", "identified license plate number".

4. The method of claim 1, wherein, In step S3, for the passing vehicles between the two gantries, according to whether the entrance and exit ramp and the service area are included in the road section, the passing state of the vehicles in the road section is divided into four cases: continuously passing through the gantries A-B, entering the service area after passing through the gantry A, getting off the highway from the exit ramp after passing through the gantry A, and entering the highway from the entrance ramp and passing through the gantry B; for the vehicles entering or leaving the highway in the middle of the A-B section, only one gantry will record the passing of the vehicle, which is excluded; secondly, according to the distribution diagram of the passing time, the vehicles entering the service area are judged and excluded, and only the vehicles continuously passing through the gantries A-B are reserved as the data basis for estimating the free flow passing time of the road section.

5. The method of claim 1, wherein, In steps S4 to S5, the average value of the 95% confidence interval of the vehicle passing time continuously passing through the gantries A-B in n preset unit time lengths under the same weather at the historical time is selected as the free flow passing time t1 of the research road section; The average value of the 95% confidence interval of the travel time of the vehicles continuously passing through A-B in the preset time length of the current research period is taken as the travel time t2 of the current traffic state, and then the travel time ratio TTI=t2 / t1 of the research road section in the current period is calculated, which is used as an index to estimate the current traffic state of the research road section, and the dynamic estimation of the traffic state of the research road section is realized through continuous iteration and update.

Images