A method for estimating road traffic control delay at signalized intersections

Abstract

This invention discloses a method for estimating road traffic control delays at signalized intersections. The method includes collecting license plate data of vehicles passing through a designated signalized intersection in a specified flow direction within a specified signal cycle; extracting the arrival and departure times of each vehicle; determining the single-vehicle cycle travel time for each vehicle when passing through the designated signalized intersection based on the arrival and departure times; calculating the cycle non-stop passage time; calculating the cycle single-vehicle control delay for each vehicle based on the single-vehicle cycle travel time and the cycle non-stop passage time; and averaging all cycle single-vehicle control delays to obtain the cycle-averaged vehicle control delay. This invention provides a method for estimating road traffic control delays at signalized intersections to address the problem that existing analyses of road traffic control delays are mostly theoretical and difficult to apply in practice, thereby improving the accuracy of control delay estimation and achieving cycle-level control delay estimation.

Description

Technical Field

[0001] The invention relates to the field of urban traffic control, and in particular to a road traffic control delay estimation method for a signalized intersection. Background Art

[0002] With the rapid growth of motor vehicle traffic demand, traffic congestion in large cities is getting worse. Accurate, comprehensive and real-time traffic flow detection data is the basis for theoretical research on traffic flow and optimization of intersection signal control. In recent years, with the continuous improvement of image recognition, sensor control, information transmission and storage technologies, the Automatic License Plate Recognition System (ALPR) has become increasingly mature. Compared with other detection methods, the license plate recognition data obtained by ALPR has the advantages of strong continuity, high recognition rate, large sample size, wide coverage, high stability and strong time continuity. With the continuous improvement and deepening of the construction of intelligent transportation systems in large cities, the coverage rate of ALPR continues to increase. In some cities, it has achieved coverage of all signal-controlled intersections on all main roads, and has the basic conditions for extracting traffic state parameters and optimizing signal timing schemes using license plate data.

[0003] In the field of road traffic control, control delay is an important indicator for evaluating signal control and intersection service levels. It is equal to the actual travel time of the vehicle minus the non-stop passing time in that time period. In the prior art, the vehicle control delay model of the signalized intersection is generally established under the condition that the arrival rate is constant and the vehicle acceleration and deceleration delays are not considered. However, in actual applications, vehicles traveling on urban roads are affected by intersection signal control, and the vehicle arrival rate is difficult to remain constant in each cycle. Therefore, the accuracy of the existing vehicle control delay model is low, and the analysis of control delay remains only in the theoretical stage. Summary of the invention

[0004] The present invention provides a method for estimating road traffic control delay at a signalized intersection, so as to solve the problem that the analysis of road traffic control delay in the prior art mostly remains in the theoretical stage and is difficult to be applied in practice, thereby improving the estimation accuracy of control delay and achieving the purpose of cycle-level control delay estimation.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for estimating road traffic control delay at a signalized intersection comprises the following steps:

[0007] S1, collect the license plate data of vehicles passing through the designated signal intersection in the designated flow direction and within the designated signal cycle, and extract the arrival time and departure time of each vehicle;

[0008] S2. Determine the cycle travel time of each vehicle when passing through a designated signalized intersection based on the arrival time and the departure time;

[0009] S3, calculate the cycle non-stop passing time;

[0010] S4, calculating the cycle single vehicle control delay of each vehicle based on the cycle single vehicle travel time and the cycle non-stop passing time;

[0011] S5. Take the average of all cycle single-vehicle control delays to obtain the cycle vehicle average control delay.

[0012] The analysis of road traffic control delay in the prior art mostly remains in the theoretical stage and is difficult to apply in practice. The main reason is that the accuracy of the prior art in estimating vehicle control delay is too low. To this end, the present application proposes a method for estimating road traffic control delay at a signalized intersection. The method is based on ALPR technology. First, the license plate data of vehicles passing through a designated signalized intersection in a designated flow direction and within a designated signal cycle are collected to extract the arrival time and departure time of each vehicle; wherein, the designated flow direction refers to the flow direction of the vehicle whose control delay is to be estimated at the intersection, and the designated signal cycle refers to the operating cycle of the control signal light at the intersection whose control delay is to be estimated; wherein the arrival time refers to the time when the vehicle enters the entrance of the intersection and is captured by ALPR, and wherein the departure time refers to the time when the vehicle leaves the exit of the intersection and is captured by ALPR. After the arrival time and departure time of each vehicle are obtained, the cycle travel time of each vehicle when passing through the designated signalized intersection can be calculated. After that, the cycle non-stop passing time is calculated again, and the cycle non-stop passing time is subtracted from the cycle travel time of a single vehicle to obtain the cycle single vehicle control delay of each vehicle, and then the cycle vehicle average control delay in the cycle is obtained.

[0013] It can be seen that the present application is based on license plate recognition technology, which can realize accurate estimation of road traffic control delays at signalized intersections, and can realize estimation of signal control delays at the cycle level at signalized intersections, which is conducive to the actual application of control delay calculations to road traffic management and analysis, such as accurately determining the time period when traffic conditions suddenly change; in addition, by refining the control delay analysis to the signal cycle through the present application, it is conducive to timely capturing sudden changes in the operating status of intersections, timely back-analyzing the causes of congestion, and implementing congestion relief strategies.

[0014] Furthermore, the license plate data in step S1 includes the license plate data of all vehicles entering the signalized intersection from the designated lane, and the license plate data of all vehicles leaving the signalized intersection from the designated lane.

[0015] The designated lane may be one or more lanes in the same direction.

[0016] Furthermore, step S2 specifically includes:

[0017] S21, matching the license plate data of vehicles entering the signalized intersection from the designated lane and leaving the signalized intersection from the designated lane;

[0018] S22, classifying the vehicles into the following categories based on the matching results: successfully matched vehicles, vehicles with missing entry information, and vehicles with missing exit information;

[0019] S23. Based on the classification results, the data of vehicles with missing exit information are eliminated, and the single-vehicle cycle travel time of successfully matched vehicles and vehicles with missing entrance information are calculated respectively.

[0020] This solution further refines the method for determining the cycle travel time of a single vehicle. The license plate data within a single cycle is divided into a successful import and export matching data set and an unsuccessful matching data set. The unsuccessful matching data set can be divided into two types: vehicles with missing import information and vehicles with missing export information. For the data of vehicles with missing export information, there may be recognition errors or temporary changes in the direction of the vehicle. Therefore, it is meaningless to calculate the control delay of the specified flow direction and can be directly eliminated. For the two types of vehicles with successful matching and vehicles with missing import information, the cycle travel time of each vehicle is calculated separately.

[0021] Furthermore, the single-vehicle cycle travel time of the successfully matched vehicle is calculated by the following method:

[0022] S2311, sorting the order of all successfully matched vehicles entering and leaving the signalized intersection within the specified signal cycle;

[0023] S2312, comparing the sorting results when entering and leaving, and determining the overtaking vehicle and the overtaken vehicle;

[0024] S2313, based on the order in which all successfully matched vehicles leave the signalized intersection, exchanging the entry order and arrival time of the overtaking vehicle with the overtaken vehicle to obtain a revised entry order and arrival time;

[0025] S2314. Calculate the cycle travel time of each successfully matched vehicle based on the departure time of each successfully matched vehicle and the corrected arrival time.

[0026] In actual applications, overtaking often occurs on urban roads, even at intersections; the occurrence of overtaking behavior will interfere with the normal change pattern of travel time. Therefore, when calculating the cycle travel time of a single vehicle of a successfully matched vehicle, this scheme first detects the overtaking vehicle: according to the first-in-first-out principle, when no overtaking occurs, the entry order and departure order of the vehicle should be equal. If the arrival and departure order of the vehicle is different in the actual data, it means that the vehicle has overtaken or has been overtaken; based on this principle, this scheme takes the order of successfully matched vehicles leaving the signal intersection as the basis, exchanges the entry order and arrival time of the overtaking vehicle and the overtaken vehicle, and then corrects the entry order and arrival time, so as to obtain a more reasonable cycle travel time of a single vehicle and provide more reasonable basic data for the subsequent calculation of control delays.

[0027] Furthermore, the cycle travel time of a vehicle with missing entrance information is calculated by the following method:

[0028] S2321. Divide all vehicles that leave the signalized intersection from the designated lanes within the designated signal cycle into a number of fleets;

[0029] S2322, determining the position of each vehicle with missing entry information within the vehicle fleet;

[0030] If the vehicle with missing entry information is the head vehicle of the convoy, the arrival time of the vehicle with missing entry information is completed by the following formula:

[0031]

[0032] If the vehicle with missing entrance information is the last vehicle in the convoy, the arrival time of the vehicle with missing entrance information is completed by the following formula:

[0033]

[0034] If the vehicle with missing entrance information is in the middle of the convoy, the arrival time of the vehicle with missing entrance information is supplemented by the following formula:

[0035]

[0036] Where: is the arrival time of the vehicle with missing entrance information; i represents the queue order of the vehicle with missing entrance information in the fleet to which it belongs; The departure time of the vehicle with missing entrance information; is the arrival time of the i+1th vehicle in the fleet; is the departure time of the i+1th vehicle in the fleet; is the arrival time of the i-1th vehicle in the fleet; is the departure time of the i-1th vehicle in the fleet.

[0037] In this scheme, the travel time is completed based on the arrival time, so the arrival time of the missing vehicles needs to be estimated first. Since the road section of the signalized intersection is short and the vehicle speeds are similar, vehicles going straight or turning left from the upstream usually leave the downstream of the intersection in the form of a convoy, so this scheme divides all vehicles leaving the signalized intersection into several convoys, where several can be 1, 2 or more; and the arrival of right-turning vehicles not controlled by the signal is relatively random, so this scheme divides the completion of the arrival time into three cases: the vehicle with missing entrance information is the head vehicle of the convoy, the tail vehicle of the convoy, or in the middle of the convoy. For these three cases, different formulas are used to calculate their respective arrival times.

[0038] If the arrival time of the leading vehicle is missing, the information of the following vehicle with known arrival and departure times can be used for calculation; if the arrival time of the tail vehicle is missing, the information of the preceding vehicle with known arrival and departure times can be used for calculation; if the arrival time of the middle vehicle is missing, this scheme uses the preceding and following vehicles with known arrival and departure times for proportional interpolation calculation.

[0039] Furthermore, step S3 specifically includes:

[0040] Determine the flow rate within the specified signal period:

[0041] S31, if the flow rate is 0, the periodic non-stop passing time = the length of the road section in the specified flow direction / the speed limit of the signalized intersection;

[0042] S32. If the flow rate is not 0, the non-stop passing time of the cycle is calculated based on the arrival time of each vehicle and the corresponding cycle travel time of a single vehicle within the specified signal cycle.

[0043] Those skilled in the art should understand that the flow rate of 0 during a cycle is a special case, which generally occurs during part of the night cycle. In this case, the length of the road section in the specified flow direction is directly divided by the speed limit of the signalized intersection to obtain the non-stop passing time of the corresponding cycle. In general, the flow rate during the cycle is not 0, so the non-stop passing time of the cycle needs to be calculated independently.

[0044] Furthermore, when the flow rate is not 0 during the cycle, the method for calculating the cycle non-stop passing time includes:

[0045] S321, using the arrival time of the vehicle as the abscissa and the cycle travel time of a single vehicle as the ordinate, establishing a two-dimensional coordinate system, and using the least squares method to fit, to obtain a vehicle travel time pattern curve within a specified signal cycle;

[0046] S322, in the vehicle travel time pattern curve, taking the average travel time in the last section of the curve as the non-stop passing time in the pending period;

[0047] S323, verifying whether the undetermined periodic non-stop passing time is feasible:

[0048] If so, the undetermined periodic non-stop passing time is used as the periodic non-stop passing time within the specified signal period;

[0049] If not, the cycle non-stop passing time in the adjacent cycle is used as the cycle non-stop passing time in the specified signal cycle.

[0050] This scheme uses the actual collected data as the basis and adopts the least squares method for fitting to obtain the vehicle travel time pattern curve in the actual cycle. Under the ideal condition without considering random factors, the last section of the curve is a straight line, on which all vehicles have the same travel time, and the slope of the curve is 0, which corresponds to vehicles passing without stopping. Therefore, the average travel time in the last section of the curve can be used as the non-stop passing time in the pending cycle.

[0051] However, in actual situations, due to different phase difference settings of trunk signal coordination and different vehicle arrival rates, some cycles may not have vehicles passing without stopping, so the control delay of the last segment of the piecewise linear curve of the cycle travel time including the vehicle may not be equal to 0. Therefore, it is necessary to verify whether the obtained non-stop passing time of the pending cycle meets the requirements of this application.

[0052] If the verification is passed, the non-stop passing time of the pending cycle can be used; if the verification fails, the non-stop passing time of the cycle in the adjacent cycle will be used instead.

[0053] Further, the method for verifying whether the undetermined period non-stop passing time is feasible includes:

[0054] S3231, calculating the period saturation within the specified signal period;

[0055] S3232, judging whether the period saturation is less than the saturation threshold:

[0056] If yes, go to step S3233;

[0057] If not, the undetermined period non-stop passing time is not feasible;

[0058] S3233: Determine whether all travel times in the last curve segment are less than the non-stop passing time threshold:

[0059] If so, the undetermined period non-stop passing time is feasible;

[0060] If not, the undetermined period non-stop passing time is not feasible.

[0061] Under this preferred scheme, only when the cycle saturation is less than the saturation threshold and all travel times in the last section of the fitting curve are less than the non-stop passing time threshold, is the pending cycle non-stop passing time determined to be feasible; otherwise, it is determined to be infeasible.

[0062] Furthermore, the period saturation within the specified signal period is calculated by the following formula:

[0063]

[0064] Where: X is the cycle saturation; Q is the cycle flow rate within the specified signal cycle; C is the cycle capacity; G is the green light duration of the signal intersection; Q max is the maximum flow rate during the cycle.

[0065] In the prior art, the calculation of saturation is generally to multiply the basic saturation flow rate by the adjustment factors such as region, number of lanes, lane width, and driver to obtain the lane saturation flow rate, and then multiply it by the green letter ratio to obtain the traffic capacity. The calculation method is complex for the values ​​of each adjustment factor, and the calculation result may be different from the actual situation if the value is improperly selected. The periodic saturation method proposed in this solution can avoid the above defects.

[0066] Furthermore, the non-stop passing time threshold is obtained by the following method:

[0067] S32331. Obtaining the cycle travel time data of all vehicles passing through the designated signalized intersection in the designated flow direction during the peak period;

[0068] S32332, fitting all the single-vehicle cycle travel time data using a two-component Gaussian mixture model to obtain two fitting curves;

[0069] S32333. The travel time corresponding to the larger point of intersection of the two fitting curves is used as the non-stop passing time threshold.

[0070] This scheme uses a two-component Gaussian mixture model to fit all the obtained single-vehicle cycle travel time data. Here, the meaning of Gaussian distribution is the probability that a certain travel time value belongs to the corresponding state. The intersection of the two fitted state distributions during peak hours means that when the actual travel time of the vehicle is less than the travel time corresponding to the intersection, the probability that the vehicle is in a non-stop passing state is greater than the probability that the vehicle has a control delay. The obtained fitting curve theoretically has two intersections. This scheme selects the intersection on the right side of the fitting curve and uses its corresponding travel time as the non-stop passing time threshold; if the travel time of a vehicle in the last section of the curve is greater than the non-stop passing time threshold, it means that the section of the curve fails to accurately distinguish the non-stop passing vehicles, so the non-stop passing time of the pending cycle is not feasible.

[0071] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0072] 1. A method for estimating road traffic control delay at a signalized intersection of the present invention can achieve accurate estimation of road traffic control delay at a signalized intersection, and can achieve estimation of signal control delay at a cycle level at a signalized intersection, thereby facilitating the actual application of control delay calculation to the management and analysis of road traffic.

[0073] 2. The present invention refines the control delay analysis to the signal cycle, which is conducive to timely capturing the sudden change of the intersection operation status, timely back-analyzing the cause of congestion, and implementing congestion relief strategies.

[0074] 3. The present invention refines the method for determining the cycle travel time of a single vehicle and corrects the overtaking behavior during the period, so as to obtain a more reasonable cycle travel time of a single vehicle and provide more reasonable basic data for subsequent calculation of control delays.

[0075] 4. When the entrance information is missing a vehicle, the present invention can adaptively supplement the travel time according to the specific situation, thus filling the gap in the prior art.

[0076] 5. The present invention proposes a detailed method for calculating the cycle non-stop passing time, which provides a scientific and reasonable basis for estimating the road traffic control delay at the signalized intersection. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0078] Figure 1 It is a schematic diagram of a flow chart of a specific embodiment of the present invention;

[0079] Figure 2 A schematic diagram of a vehicle travel time pattern curve in a specific embodiment of the present invention;

[0080] Figure 3 It is a flow chart of verifying whether the non-stop passing time of a pending period is feasible in a specific embodiment of the present invention;

[0081] Figure 4 It is the fitting result of the Gaussian mixture model for the single-vehicle cycle travel time data in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0082] In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the examples and the accompanying drawings. The schematic embodiments of the present invention and the description thereof are only used to explain the present invention and are not intended to limit the present invention. In the description of the present application, it should be understood that the orientation or positional relationship indicated by terms such as "front", "rear", "left", "right", "up", "down", "vertical", "horizontal", "high", "low", "inside", "outside", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the scope of protection of the present application.

[0083] Embodiment 1:

[0084] like Figure 1 A method for estimating road traffic control delay at a signalized intersection is shown, comprising the following steps:

[0085] S1. Collect the license plate data of vehicles passing through the designated signalized intersection in the designated flow direction and within the designated signal cycle, and extract the arrival time and departure time of each vehicle; the license plate data includes the license plate data of all vehicles entering the signalized intersection from the designated lane, and the license plate data of all vehicles leaving the signalized intersection from the designated lane;

[0086] S2. Determine the cycle travel time of each vehicle when passing through a designated signalized intersection based on the arrival time and the departure time;

[0087] S3, calculate the cycle non-stop passing time;

[0088] S4, calculating the cycle single vehicle control delay of each vehicle based on the cycle single vehicle travel time and the cycle non-stop passing time;

[0089] S5. Take the average of all cycle single-vehicle control delays to obtain the cycle vehicle average control delay.

[0090] This embodiment can refine the control delay analysis to the signal cycle, which is conducive to timely capturing the precise time of the sudden change of the operation status of the intersection, timely back-analyzing the cause of congestion, and implementing congestion relief strategies.

[0091] Embodiment 2:

[0092] A method for estimating road traffic control delay at a signalized intersection, based on Example 1, determines the single-vehicle cycle travel time of each vehicle when passing through a designated signalized intersection by the following method:

[0093] S21, matching the license plate data of vehicles entering the signalized intersection from the designated lane and leaving the signalized intersection from the designated lane;

[0094] S22, classifying the vehicles into the following categories based on the matching results: successfully matched vehicles, vehicles with missing entry information, and vehicles with missing exit information;

[0095] S23. Based on the classification results, the data of vehicles with missing exit information are eliminated, and the single-vehicle cycle travel time of successfully matched vehicles and vehicles with missing entrance information are calculated respectively.

[0096] In this embodiment, the calculation method of the single-vehicle cycle travel time of the successfully matched vehicle is:

[0097] S2311, sorting the order of all successfully matched vehicles entering and leaving the signalized intersection within the specified signal cycle;

[0098] S2312, comparing the sorting results when entering and leaving, and determining the overtaking vehicle and the overtaken vehicle;

[0099] S2313, based on the order in which all successfully matched vehicles leave the signalized intersection, exchanging the entry order and arrival time of the overtaking vehicle with the overtaken vehicle to obtain a revised entry order and arrival time;

[0100] S2314. Calculate the cycle travel time of each successfully matched vehicle based on the departure time of each successfully matched vehicle and the corrected arrival time.

[0101] As shown in Table 1, Table 1 is a correction example of overtaking vehicles in a certain cycle, where the departure orders 3, 4 and 6, 7 are two groups of overtaking and overtaken vehicles. Taking order 3 and 4 as an example, the arrival times of the two vehicles are exchanged, and the time at the intersection is exchanged between vehicles 3 and 4. The total travel time of the cycle remains unchanged, the average travel time of each vehicle remains unchanged, and the vehicle arrival and departure order is restored to first-in-first-out.

[0102] Table 1. Correction example of overtaking vehicles in a certain period

[0103]

[0104] In addition, the calculation method for the single-vehicle cycle travel time of vehicles with missing entrance information is:

[0105] S2321. All vehicles leaving the signalized intersection from the designated lanes within the designated signal cycle are regarded as one convoy;

[0106] S2322, determining the position of each vehicle with missing entry information within the vehicle fleet;

[0107] If the vehicle with missing entry information is the head vehicle of the convoy, the arrival time of the vehicle with missing entry information is completed by the following formula:

[0108]

[0109] If the vehicle with missing entrance information is the last vehicle in the convoy, the arrival time of the vehicle with missing entrance information is completed by the following formula:

[0110]

[0111] If the vehicle with missing entrance information is in the middle of the convoy, the arrival time of the vehicle with missing entrance information is supplemented by the following formula:

[0112]

[0113] Where: is the arrival time of the vehicle with missing entrance information; i represents the queue order of the vehicle with missing entrance information in the fleet to which it belongs; The departure time of the vehicle with missing entrance information; is the arrival time of the i+1th vehicle in the fleet; is the departure time of the i+1th vehicle in the fleet; is the arrival time of the i-1th vehicle in the fleet; is the departure time of the i-1th vehicle in the fleet.

[0114] Embodiment 3:

[0115] A method for estimating road traffic control delay at a signalized intersection, based on Embodiment 1 or 2, a method for calculating a periodic non-stop passing time includes:

[0116] First determine the flow rate within the specified signal period:

[0117] If the flow rate is 0, the cycle non-stop passing time = the length of the road section in the specified flow direction / the speed limit of the signalized intersection;

[0118] If the flow rate is not 0, the following method is used to calculate the cycle non-stop passing time:

[0119] S321, using the arrival time of the vehicle as the abscissa and the cycle travel time of a single vehicle as the ordinate, establishing a two-dimensional coordinate system, and using the least squares method to fit, to obtain a vehicle travel time pattern curve within a specified signal cycle;

[0120] Preferably, this embodiment performs fitting with a four-stage arrival rate model, that is, fitting is performed in four stages to obtain a four-stage vehicle travel time model curve as shown in FIG. Figure 2 shown.

[0121] S322, in the vehicle travel time pattern curve, taking the average travel time in the last section of the curve as the non-stop passing time in the pending period;

[0122] S323, verifying whether the undetermined periodic non-stop passing time is feasible:

[0123] If so, the undetermined periodic non-stop passing time is used as the periodic non-stop passing time within the specified signal period;

[0124] If not, the cycle non-stop passing time in the adjacent cycle is used as the cycle non-stop passing time in the specified signal cycle.

[0125] Embodiment 4:

[0126] A method for estimating road traffic control delay at a signalized intersection, based on Example 3, a method for verifying whether the non-stop passing time of a pending period is feasible is as follows: Figure 3 As shown, specifically including:

[0127] S3231. Calculate the period saturation within the specified signal period:

[0128]

[0129] Where: X is the cycle saturation; Q is the cycle flow rate within the specified signal cycle, in veh; C is the cycle capacity, in veh / cycle; G is the green light duration of the signal intersection, in seconds; Q max It is the maximum flow rate during the cycle, in veh.

[0130] S3232, judging whether the period saturation is less than the saturation threshold:

[0131] If yes, go to step S3233;

[0132] If not, the undetermined period non-stop passing time is not feasible;

[0133] S3233: Determine whether all travel times in the last section of the vehicle travel time pattern curve are less than the non-stop passing time threshold:

[0134] If so, the undetermined period non-stop passing time is feasible;

[0135] If not, the undetermined periodic non-stop passing time is not feasible.

[0136] Preferably, the saturation threshold in this embodiment is 0.9.

[0137] The non-stop passing time threshold is obtained by the following method:

[0138] S32331. Obtaining the cycle travel time data of all vehicles passing through the designated signalized intersection in the designated flow direction during the peak period; preferably, the peak period in this embodiment is 07:00-09:00 and 17:00-19:00;

[0139] S32332, using a two-component Gaussian mixture model to fit all the single-vehicle cycle travel time data, obtain two fitting curves as follows: Figure 4 As shown;

[0140] exist Figure 4 In the figure, state 1 represents the fitting result without control delay through component, and state 2 represents the fitting result with control delay through component; in addition, the horizontal axis is the cycle travel time of a single vehicle, in seconds; the vertical axis represents the frequency, dimensionless;

[0141] S32333. The travel time corresponding to the larger point of intersection of the two fitting curves is used as the non-stop passing time threshold. Figure 4 In the example, the travel time 71.7s corresponding to the right intersection is taken as the non-stop passing time threshold in this embodiment.

[0142] In a more preferred embodiment, if the undetermined periodic non-stop passing time is not feasible, the specific traffic conditions can be further determined:

[0143] Condition A: The cycle saturation is less than the saturation threshold, and the vehicle travel time in the last section of the vehicle travel time pattern curve is greater than the non-stop passing time threshold. It is considered that no oversaturation occurs in the cycle, but there is no non-stop passing vehicle at the end of the cycle.

[0144] Condition B: The cycle saturation is greater than the saturation threshold, and the travel time of all vehicles in the last section of the vehicle travel time pattern curve is greater than the non-stop passing time threshold. It is considered that the cycle state is close to saturation or has been oversaturated, and there is no non-stop passing vehicle at the end of the cycle.

[0145] Condition C: The cycle saturation is greater than the saturation threshold, and the travel times of all vehicles in the last section of the vehicle travel time pattern curve are not all greater than the non-stop passing time threshold. It is considered that the cycle state is close to saturation or is already oversaturated and there are no vehicles passing without stopping at the end of the cycle, but there are vehicles that are subject to minor control delays at the end of the cycle.

[0146] For the road traffic control delay estimation method of this application, the applicant conducted an effect evaluation and verification:

[0147] Since the real value of the control delay cannot be obtained using field data, this embodiment uses the control delay obtained by the VISSIM simulation model of the XX West Road-XX South Road section of a certain city as the standard value, evaluates the control delay estimation error, and uses MAE as an indicator to statistically analyze the results.

[0148] During the verification process, all possible combinations of data of each cycle and each sampling rate with a sampling number greater than 1 are traversed. The sampling results are used as matching data sets, and the unsampled data are used as missing data sets. Under each possible combination, the travel time estimation completion method proposed in this application is used to estimate the travel time of the missing data, and the non-stop passing time threshold of each cycle is calculated at the same time, and the estimated vehicle control delay under various circumstances is calculated. Table 2 shows the verification results of this embodiment.

[0149] Table 2 Verification results

[0150]

[0151] It can be seen from Table 2 that in all verification cycles, compared with the standard value, the estimated error of the control delay is within 10s in 84.94% of the cases, and within 20s in 98.34% of the cases. The verification results can prove that in most cases, the estimation of the control delay by the method of the application can obtain relatively accurate results.

[0152] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0153] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In addition, the term "connected" used in this article can be directly connected or indirectly connected via other components without special explanation.

Claims

1. A method for estimating road traffic control delay at a signalized intersection, characterized in that: The following steps are involved: S1, collect the license plate data of vehicles passing through the designated signal intersection in the designated flow direction and within the designated signal cycle, and extract the arrival time and departure time of each vehicle; S2. Determine the cycle travel time of each vehicle when passing through a designated signalized intersection based on the arrival time and the departure time; S3, calculate the cycle non-stop passing time; S4, calculating the cycle single vehicle control delay of each vehicle based on the cycle single vehicle travel time and the cycle non-stop passing time; S5. Take the average of all cycle single-vehicle control delays to obtain the cycle vehicle average control delay; Step S3 specifically includes: Determine the flow rate within the specified signal period: S31, if the flow rate is 0, the periodic non-stop passing time = the length of the road section in the specified flow direction / the speed limit of the signalized intersection; S32, if the flow rate is not 0, based on the arrival time of each vehicle and the corresponding cycle travel time of a single vehicle within the specified signal cycle, calculate the cycle non-stop passing time; In step S32, the method for calculating the periodic non-stop passing time includes: S321, using the arrival time of the vehicle as the abscissa and the cycle travel time of a single vehicle as the ordinate, establishing a two-dimensional coordinate system, and using the least squares method to fit, to obtain a vehicle travel time pattern curve within a specified signal cycle; S322, in the vehicle travel time pattern curve, taking the average travel time in the last section of the curve as the non-stop passing time in the pending period; S323, verify whether the undetermined periodic non-stop passing time is feasible: If so, the undetermined periodic non-stop passing time is used as the periodic non-stop passing time within the specified signal cycle; If not, the cycle non-stop passing time in the adjacent cycle is used as the cycle non-stop passing time in the specified signal cycle.

2. The method for estimating road traffic control delay at a signalized intersection according to claim 1, characterized in that: The license plate data in step S1 includes the license plate data of all vehicles entering the signalized intersection from the designated lane, and the license plate data of all vehicles leaving the signalized intersection from the designated lane.

3. The method for estimating road traffic control delay at a signalized intersection according to claim 2, characterized in that: Step S2 specifically includes: S21, matching the license plate data of vehicles entering the signalized intersection from the designated lane and leaving the signalized intersection from the designated lane; S22, classifying the vehicles into the following categories based on the matching results: successfully matched vehicles, vehicles with missing entry information, and vehicles with missing exit information; S23. Based on the classification results, the data of vehicles with missing exit information are eliminated, and the single-vehicle cycle travel time of successfully matched vehicles and vehicles with missing entrance information are calculated respectively.

4. The method for estimating road traffic control delay at a signalized intersection according to claim 3, characterized in that: The single-vehicle cycle travel time of the successfully matched vehicle is calculated by the following method: S2311, sorting the order of all successfully matched vehicles entering and leaving the signalized intersection within the specified signal cycle; S2312, comparing the sorting results when entering and leaving, and determining the overtaking vehicle and the overtaken vehicle; S2313, based on the order in which all successfully matched vehicles leave the signalized intersection, exchanging the entry order and arrival time of the overtaking vehicle with the overtaken vehicle to obtain a revised entry order and arrival time; S2314. Calculate the cycle travel time of each successfully matched vehicle based on the departure time of each successfully matched vehicle and the corrected arrival time.

5. The method for estimating road traffic control delay at a signalized intersection according to claim 3, characterized in that: The cycle travel time of a vehicle with missing entry information is calculated by the following method: S2321. Divide all vehicles that leave the signalized intersection from the designated lanes within the designated signal cycle into a number of fleets; S2322, determining the position of each vehicle with missing entry information within the vehicle fleet; If the vehicle with missing entry information is the head vehicle of the convoy, the arrival time of the vehicle with missing entry information is completed by the following formula: If the vehicle with missing entrance information is the last vehicle in the convoy, the arrival time of the vehicle with missing entrance information is completed by the following formula: If the vehicle with missing entrance information is in the middle of the convoy, the arrival time of the vehicle with missing entrance information is supplemented by the following formula: Where: is the arrival time of the vehicle with missing entrance information; i represents the queue order of the vehicle with missing entrance information in the fleet to which it belongs; The departure time of the vehicle with missing entrance information; is the arrival time of the i+1th vehicle in the fleet; is the departure time of the i+1th vehicle in the fleet; is the arrival time of the i-1th vehicle in the fleet; is the departure time of the i-1th vehicle in the fleet.

6. The method for estimating road traffic control delay at a signalized intersection according to claim 1, characterized in that: The method for verifying whether the undetermined periodic non-stop passing time is feasible includes: S3231, calculating the period saturation within the specified signal period; S3232, judging whether the period saturation is less than the saturation threshold: If yes, go to step S3233; If not, the undetermined period non-stop passing time is not feasible; S3233: Determine whether all travel times in the last curve segment are less than the non-stop passing time threshold: If so, the undetermined period non-stop passing time is feasible; If not, the undetermined period non-stop passing time is not feasible.

7. A method for estimating road traffic control delay at a signalized intersection according to claim 6, characterized in that: The period saturation within the specified signal period is calculated by the following formula: Where: X is the cycle saturation; Q is the cycle flow rate within the specified signal cycle; C is the cycle capacity; G is the green light duration of the signal intersection; Q max is the maximum flow rate during the cycle.

8. The method for estimating road traffic control delay at a signalized intersection according to claim 6, characterized in that: The non-stop passing time threshold is obtained by the following method: S32331. Obtaining the cycle travel time data of all vehicles passing through the designated signalized intersection in the designated flow direction during the peak period; S32332, fitting all the single-vehicle cycle travel time data using a two-component Gaussian mixture model to obtain two fitting curves; S32333. The travel time corresponding to the larger point of intersection of the two fitting curves is used as the non-stop passing time threshold.

Images