CN107331159A - A kind of traffic major trunk roads velocity estimation apparatus based on coil checker data - Google Patents

Abstract

A device for estimating traffic arterial speed based on coil detector data, including a computer central processing unit and a data memory, and the central processing unit performs the following calculation steps: acquiring coil detector data, road section geometric attributes, signal timing parameters and preset The saturation flow constant is used as an input parameter, and after calculating the travel time of the upstream section and the middle section, the average section speed of the upstream section and the middle section is then calculated.

Description

A Device for Estimating Traffic Arterial Road Speed Based on Loop Detector Data

technical field

The invention belongs to the technical field of intelligent transportation, in particular to a device for estimating the speed of a main traffic road based on coil detector data.

Background technique

In recent years, with the gradual promotion of advanced passenger information systems and intelligent traffic management systems, it is becoming more and more important to provide travelers with accurate route travel time through various methods. Therefore, obtaining accurate real-time traffic data, that is, travel speed or travel time, is of great significance to the successful promotion of these two systems.

In road traffic planning, route travel time is the focus of route selection, departure time, etc., but it is for a specific origin and destination. If a large number of origin and destination needs of users are to be combined, the advanced passenger information system and intelligent traffic management system can provide road information at the road section level. Once down to the link level, speed is a more sensitive and accurate quantity than travel time to describe traffic conditions.

Floating vehicles equipped with advanced positioning systems are widely used to measure road speed or travel time. They can obtain more accurate spatial information than static equipment. However, due to market limitations, for most arterial roads, None of the floating cars are large enough to provide credible data. Loop detectors may be an alternative data source for measuring road segment speed.

Three state-of-the-art velocity estimation models have been disclosed in previous existing documents, namely the British model, the Illinois model and the Iowa ^model0 . The inputs of these three models are the traffic flow and occupancy measured by the coil detector, the timing parameters of signal lights and some geometric parameters of the road. The first two models are linear regression models, and the third model is a nonlinear regression model, so these model parameters require site-specific data for calibration. However, in actual operation, calibration is a difficult step.

Public documents related to the present invention include:

[1] Zhang, H.M.(1998). "A link journey speed model for arterial traffic." Transp.Res.Rec.1676, Transportation Research Board, Washington, D.C., 1998, 109–115.

Contents of the invention

Existing arterial road traffic speed estimation models using loop detector data all require data from specific locations for calibration, which will cause a lot of trouble in actual operation. The present invention proposes a speed estimation model of a traffic arterial road based on coil detector data without calibration, and compares the estimation accuracy with the above three existing models.

The technical scheme of the present invention is, a kind of traffic arterial road speed estimation device based on loop detector data, comprises computer central processing unit and data memory, and central processing unit carries out following computing steps:

Obtain coil detector data, road section geometric attributes, signal timing parameters and preset saturation flow constants as input parameters, calculate the travel time of the upstream road section and the middle road section, and then calculate the average speed of the upstream and middle road sections, where have,

The travel time of the upstream segment and the intermediate segment is divided into two parts, including: cruising time and signal delay time, namely,

Travel time = cruise time + signal delay time

Among them, the cruising time represents the average time for the vehicle to pass through the road section without considering the downstream signal lights;

The signal delay time represents the travel delay time of the road segment due to the downstream signal light,

The cruising time is calculated by formula (1):

Among them, L ₁ represents the length of the research section, u _det is the maximum speed in the data obtained by the coil detectors of the upstream section and the downstream section, or the speed data of any coil detector in the middle section, and the signal delay time is determined by the simplified Webster Formula (2) is calculated to get:

Among them, the detailed expression of the coefficient φ is as follows:

Among them, x represents the saturation degree of the road section; q represents the traffic flow of the road section (pcu/h);

λ represents the effective green light time ratio, that is, the green light time g divided by the total cycle time C.

The beneficial effect of the present invention can be obtained from the estimated speed map obtained from the average speed of all vehicles in the road section in the embodiment of the present invention, which is shown in FIG. 5 . The results of this figure show that the second-order mean square error of the velocity estimate is less than 5km/h, which also explains that the error in the absolute velocity value is less than 5km/h at a 95% confidence level.

Comparing the other three models other than the scheme of the present invention, the estimation accuracy of the scheme of the present invention is better than that of the British and Iowa models, and only slightly worse than that of the Illinois model. However, since the technical solution of the present invention does not need to be calibrated, it is more convenient to implement in actual operation.

Description of drawings

The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the invention are shown by way of illustration and not limitation, in which:

Fig. 1 is the node section figure of test network in the embodiment of the present invention

Fig. 2 is the detector position and road section in the embodiment of the present invention;

Fig. 3 is a comparison diagram of the spatial distribution of speeds on three road sections in the embodiment of the present invention.

Fig. 4 is a diagram of signal delays and coefficients in the deterministic queuing theory involved in the present invention.

Fig. 5 is a graph comparing the speed estimated by the model of the present invention with the average speed of all vehicles.

Fig. 6 is a graph comparing the speed estimated by the scheme model of the present invention, the British model, the Illinois model and the Iowa model with the average speed of all vehicles.

detailed description

The invention adopts the data of the coil detector, the geometric attribute of the road section, the signal timing parameter and the preset saturation flow constant as input parameters. Once the travel times for the upstream and intermediate segments are estimated, the average segment speeds for the upstream and intermediate segments can be calculated.

Unlike highway travel time estimation, the main difficulty in arterial travel time estimation is the presence of delays caused by downstream signaling facilities, so any delays caused by traffic signals should be included in the travel time. The travel time of the upstream and intermediate sections can be divided into two parts: cruise time and signal delay time, namely

Travel time = cruise time + signal delay time

Among them, the cruising time represents the average time for the vehicle to pass through the road section without considering the downstream signal lights; the signal delay time represents the travel delay time of the road section caused by the downstream signal lights. The cruise time is longer than the free flow time because of the interaction effects of vehicles following and changing lanes.

Cruise time is calculated by the following formula:

Among them, L ₁ represents the length of the research section, and u _det is the maximum speed in the data obtained by the upstream and downstream coil detectors, or the speed data of any detector in the middle section. The specific details will be described in the specific implementation section below.

The signal delay time is calculated by the following simplified Webster formula:

Among them, the detailed expression of the coefficient φ is as follows:

The specific description details are also set forth in the following specific implementation part.

Regarding the estimation of cruising time, the scheme of the present invention has carried out many simulation experiments to study different vehicle speeds on selected road sections and analyze the relationship between the detection point speed and the spatial average speed. As shown in Figure 3, the input data of this experiment is the average speed value obtained by coil detectors set every 20 meters, and these curves reflect the spatial distribution of vehicle speed on the road section. It can be seen that the speed in the middle section of the curve in the figure is very stable, so the data of any coil detector can be regarded as the cruising speed. At the end of the upstream section and the downstream section, the average vehicle speed decreases within 15km/h, but in general, the decrease is larger at the end of the downstream section. The reason for this is that the vehicle will naturally accelerate on the open road, but on the moderate or even congested road section, the vehicle will wait in line at the crossing due to the signal lights of the downstream section to affect the speed, thereby underestimating the road cruising speed. In contrast, the detector speed of the upstream segment would be a better choice for estimating the cruise speed. In the case of only two detector data, the proposer of this scheme directly substitutes the larger speed to represent the average cruising speed of the road section. The analysis in Figure 3 also confirms that such a choice can select a relatively more accurate average road section estimate. In addition, if the speed of the downstream detector is almost the same as that of the upstream road segment, or even slightly greater than its speed in good traffic conditions, it is also a good representative of the average cruising speed.

For the estimation of signal delay time, Webster's formula is usually used to determine the average signal delay of each vehicle near the stop line of the intersection. This proposal proposes a way to factor this delay into the speed estimate for the road segment of interest (typically the road segment 50-100 m upstream of the stop line). The expression of Webster's formula is as follows:

Among them, x represents the saturation degree of the road section; q represents the traffic flow of the road section (pcu/h); λ represents the effective green light time ratio (that is, the green light time g divided by the total cycle time C). The first part of the right half of the formula represents the average delay time when vehicles evenly arrive at the signal intersection with fixed timing, which is obtained according to the deterministic queuing theory. The second part considers the random property of vehicle arrival (queue overflow). When the saturation level of the link is low, the impact of random fluctuations is negligible, but as the saturation level increases, the impact also increases. The third part is an empirical correction factor, which is a subtraction item, and the value ranges from zero to the value of the second part, corresponding to the random arrival situation and the uniform distribution arrival situation, respectively.

Webster's formula estimates the total delay time of the road section controlled by signal lights, but what is needed in this study is only the delay time of the specified road section (the section at the L2 distance upstream of the stop line in Figure 2), so the coefficient φ is introduced to represent the selected road section The ratio between the signal delay and the total signal delay. At the same time, through the inspection of the data set, the first part accounts for 93% of the total signal delay, so for the convenience of calculation, only the first part of Webster's formula (uniform distribution delay) is used in the subsequent derivation.

Figure 4 illustrates how to derive the first part of Webster's formula and how to obtain the coefficient φ. The slopes of the straight lines BC and AC represent the arrival rate and outflow rate of vehicles respectively, and the length of the line segment DD' represents the part of the length L2 of the downstream part of the road segment. Lines I and II are parallel to line BC. It is easy to know that the area of triangle ABC represents the cumulative signal delay of all vehicles on the road section, and the average delay of each vehicle given in the first part of Webster's formula is the area of the above area divided by the total number of vehicles passing through the road section within a cycle C _q . The area of the triangle AB'C' cut by the straight line I represents the cumulative delay of all vehicles in the upstream and middle road sections (the part above the stop line L2), which is also divided by C _q to obtain the average vehicle delay. Therefore, the coefficient φ is the ratio of the area of the triangle AB'C' to ABC, and can also be expressed by the ratio of the lengths of the line segments AD' and AD. If the length of L2 is greater than the length of the line segment AD, the straight line II cannot cut the triangle ABC, which means that the signal lights will not cause queuing delay to the research road section, that is, φ=0. The expression for the coefficient φ is as follows:

A more practical and general two-term Webster formula is,

Use this formula to calculate the signal delay, and then multiply by the coefficient φ to obtain the signal delay time of the section L2 from the stop line.

For verifying the present invention, the simulation road network of the concrete experiment that carries out is as shown in Figure 1, and information such as the road section numbering, road section length, detector position of eight road sections of coil detector is all as shown in table 1, and the implication of each length parameter See Figure 2.

The following is a specific calculation example of two road sections to illustrate the calculation process of the scheme model:

As shown in Figure 3, when the starting and ending point flow rate is 50%, the number of indication of the upstream detector among the two coil detectors on road section 53 is 37km/h, the number of indication of the downstream display is 40km/h, and the total circulation of the crossing is The duration is 100s, the green light time is 60s, and the road flow is 600pcu/h. therefore,

∴Travel time of section 53=18.63+11.55=30.18s

When the flow rate at the starting and ending points is 50%, the reading of the upstream detector among the two coil detectors on road section 117 is 41km/h, the reading of the downstream display is 43km/h, the total cycle time of the intersection is 80s, and the green light time is 55s, and the section flow is 700pcu/h. therefore,

Table 1 is the length of the road section and the position of the detector in the selected road section; Table 2 is the calibration results of the three models other than this scheme; Table 3 is the regression analysis of the four models.

Table 1 Length of road segments and positions of detectors in selected road segments

Table 2 Calibration results of three models other than this scheme

^a The free flow speeds of the selected 10 road sections are different in reality and simulation, and the distribution range is 35.0-55.0km/h.

^b The measurement parameter γ is 0.5 in this study.

Table 3 Analysis of the regression results of the four models

Fig. 5 is a graph comparing the speed estimated by the model of the present invention with the average speed of all vehicles. Fig. 6 is a graph comparing the speed estimated by the scheme model of the present invention, the British model, the Illinois model and the Iowa model with the average speed of all vehicles. The source of the data set is based on the version 2.0 of the micro-traffic simulation platform INTEGRATION, which randomly inputs different starting and ending traffic and different headway distributions of different road sections, and obtains 288 road speed simulation results. Then it is randomly divided into two equal data sets, wherein, data set 1 is used for the construction of the model of the present invention in the embodiment of Fig. 5 and the parameter calibration of the other three models; data set 2 is used for four in the embodiment of Fig. 6 A comparison of the model accuracy.

It is worth noting that although the foregoing content has described the spirit and principle of the invention with reference to several specific embodiments, it should be understood that the present invention is not limited to the disclosed specific embodiments, and the division of various aspects does not mean that these Features within an aspect cannot be combined, this division is for convenience of presentation only. The present invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims (1)

1. a kind of traffic major trunk roads velocity estimation apparatus based on coil checker data, including central processing unit (CPU) sum According to memory, central processing unit performs following calculation procedure：

Obtain coil checker data, section geometric attribute, signal timing dial parameter and the saturation volume constant conduct pre-set Input parameter, was calculated after the travel time for obtaining upstream section and middle section, and then calculated the flat of upstream and middle section Equal section speed, wherein have,

The travel time in upstream section and middle section is divided into two parts, including：Cruise time and signal delay time, That is,

Travel time=cruise time+signal delay time

Wherein, the cruise time represents not consider the average time that vehicle passes through section during downstream signal lamp；

Signal delay time represent due to caused by downstream signal lamp section go on a journey the delay time at stop,

Cruise time is calculated by formula (1) and obtained：

Wherein, L₁Represent research road section length, u_detIt is the maximum in upstream section and downstream road section coil checker acquisition data Speed, or any coil detector in middle section speed data,

Signal delay time is calculated by the Robert Webster formula (2) simplified and obtained：

Wherein, coefficient φ detailed expressions are as follows：

<mrow> <mi>&amp;phi;</mi> <mo>=</mo> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <mfrac> <mrow> <mo>(</mo> <mi>C</mi> <mo>-</mo> <mi>g</mi> <mo>)</mo> <mi>q</mi> <mo>-</mo> <msub> <mi>L</mi> <mn>2</mn> </msub> </mrow> <mrow> <mo>(</mo> <mi>C</mi> <mo>-</mo> <mi>g</mi> <mo>)</mo> <mi>q</mi> </mrow> </mfrac> <mo>,</mo> </mrow> </mtd> <mtd> <mrow> <mi>i</mi> <mi>f</mi> </mrow> </mtd> <mtd> <mrow> <mo>(</mo> <mi>C</mi> <mo>-</mo> <mi>g</mi> <mo>)</mo> <mi>q</mi> <mo>&amp;GreaterEqual;</mo> <msub> <mi>L</mi> <mn>2</mn> </msub> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mn>0</mn> <mo>,</mo> </mrow> </mtd> <mtd> <mrow> <mi>i</mi> <mi>f</mi> </mrow> </mtd> <mtd> <mrow> <mo>(</mo> <mi>C</mi> <mo>-</mo> <mi>g</mi> <mo>)</mo> <mi>q</mi> <mo>&lt;</mo> <msub> <mi>L</mi> <mn>2</mn> </msub> </mrow> </mtd> </mtr> </mtable> </mfenced> </mrow>

Wherein, x represents section degree of saturation；Q represents link flow (pcu/h)；

λ represents effective green time ratio, i.e. green time g divided by circulation total duration C.