JPH04148299A - Traffic signal control method and its execution device - Google Patents

Abstract

PURPOSE:To execute an appropriate and quick traffic control by calculating the optimum cycle time and the split time of the next cycle based on the crossing information of a previous cycle by a fuzzy inference at a real time. CONSTITUTION:The present crossing information obtained by a crossing information grasping part 11 is defined as the input of the antecedent part of each control rule of a control rule description part 13, and collated with the membership function of each control rule, then an adaptation degree toward the control rule is obtained. Next, the obtained adaptation degree is collated with the membership function of a consequent part, and the output toward the rule is obtained. The crossing information changing successively is processed in an inference part 12, and an optimum control amount to the next cycle is obtained by the fuzzy inference at high speed. Thus, it is possible to well correspond even to a congestion which happens suddenly, and the appropriate and quick traffic control becomes possible.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to traffic signal control at intersections, and relates to a technology that is effective when applied to flexible traffic signal control technology in response to constantly changing traffic volumes. .

[Prior art]

Traffic signal control technology can be broadly divided into point control and system control. Point control includes periodic control, which has several control patterns that correspond to the hourly fluctuations in traffic demand during the day, and periodic control, which measures the incoming traffic at intersections using vehicle sensors and responds to increases and decreases in traffic. There is a sensitive control that expands and contracts the green time every time. In addition, system control is the integrated control of multiple traffic lights installed on the road network, and a control pattern that matches the actual traffic situation is automatically created in advance according to the time of day and traffic flow data from the Koryo sensor. There is automatic sensing system control that switches to

Regarding the above-mentioned prior art, for example, see "
1 of ``General Theory of Traffic Engineering'' (Gijutsu Shoin, published in 1989)
It is described on pages 76-203.

[Problem to be solved by the invention]

However, in the conventional program control technology, multiple values of cycle time and split time are set,
Since the time to apply one of these methods is set based on the time, there is a problem in that it is not possible to deal with unexpected and sudden changes in traffic.

Furthermore, with conventional responsive control technology, when an intersection is congested, the exact same display is repeated all the time, making it impossible to perform appropriate control, resulting in traffic jams.

The present invention has been made to solve the above problems, and an object of the present invention is to calculate the cycle time by fuzzy reasoning based on the intersection information of the previous cycle in point control of intersections where traffic volume changes sequentially. The object of the present invention is to provide a traffic signal control method that can perform flexible control by determining the time and split time in real time and combining them.

Another object of the present invention is to provide a technique that allows appropriate and prompt traffic control.

The above and other objects and novel features of the present invention will become apparent from the description and accompanying drawings.

[Means to solve the problem]

In order to achieve the above object, the present invention provides a traffic signal control method that performs point control at intersections where traffic volume changes sequentially. The most important feature is that the system calculates the cycle time and split time, and controls the traffic signal using the optimal cycle time and split time.

The apparatus also includes a vehicle detection means for detecting an approximate number of vehicles waiting at a traffic light at an intersection, and a device for determining the next cycle of traffic flow at the intersection from the vehicle number detection means. A traffic flow detection means, a congestion degree detection means for determining the degree of congestion of an intersection from the traffic flow detection means, and an intersection control parameter based on the output of the congestion degree detection means of the intersection and the degree of adaptation of each control rule of the traffic signal. The present invention is characterized by having an inference means for determining the cycle split time and cycle time in real time using fuzzy inference.

[For production]

According to the above-mentioned means, in point control of intersections where traffic volume changes sequentially, the approximate number of vehicles waiting at the intersection and the degree of congestion at the intersection are determined from information obtained from the vehicle number detection means, and these intersection information and control are used. The suitability of each control rule is determined from the control rules described in the rule description section, and the inference section calculates the cycle time and split time, which are traffic light control, from the determined suitability of each control rule. It becomes possible to perform flexible control by using This makes it possible to carry out appropriate and prompt traffic control.

[Embodiments of the invention]

Hereinafter, one embodiment of the present invention will be specifically described using the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a traffic signal control device according to an embodiment of the present invention, and FIG. 2 is a block diagram for explaining the functions of a control section in the arithmetic processing device shown in FIG.

Fig. 3 is a processing flowchart of the inference section, Fig. 4 is a diagram showing an example of installing vehicle detectors at an intersection, and Fig. 5 is for explaining the membership function representing the expression "many" in the control rule. Figure 6a is a diagram for explaining the calculation of the fitness of the antecedent part of the control rule, Figure 6b is a diagram for explaining the calculation of the output of the consequent part of the control rule, and Figure 7 is a diagram for explaining the calculation of the output of the consequent part of the control rule. FIG. 2 is a diagram for explaining calculation of a control amount based on outputs of consequent parts of a plurality of control rules.

As shown in FIG. 1, the traffic signal control device of this embodiment includes a vehicle detector 1, an arithmetic processing device 2, a signal controller 3. It is composed of a plurality of traffic lights 4.

Said calculation processing bag! ! As shown in FIG. 2, the control section 2 includes an intersection information understanding section 11, an inference section 12, and a control rule description section 13.

As shown in FIG. 4, the intersection information grasping unit 11 includes a vehicle detector 111 for detecting vehicles waiting at an intersection on each road and intersection from the intersection. A number of vehicle detectors 112 for detecting vehicles passing through an intersection and vehicle detectors 113 for detecting vehicles waiting to turn right inside the intersection are installed. The information of these vehicle detectors 111, 112, 113 is input to the arithmetic processing device 2, and the arithmetic processing device 2
1. Understand the degree of mixed pattern at the intersection based on the information of 112 and 113.

Next, the control rule description section 13 will be explained using FIG. 4.

Control rules are created in advance based on past traffic data at intersections. The control rules are as follows.

(1) If more people heading towards direction 1 are waiting at the intersection than in the previous cycle, make the split time for direction 1 a little longer.

(2) If more cars heading in direction 2 are waiting at the intersection than in the previous cycle, shorten the split time for direction 1 a little.

(3) If there are more people waiting in directions 1 and 2 at the intersection than in the previous cycle, lengthen the cycle time a little.

(4) If more right-turning vehicles are waiting at an intersection than in the previous cycle, shorten the cycle time.

(5) If direction 1 is green at an intersection and there is no traffic ahead, shorten the split time for direction 1 a little.

(6) When direction 2 is green at an intersection, if there is no traffic ahead, lengthen the split time for direction 1 a little.

(7) If the flow of seats heading in direction 1 at the intersection is slower than in the previous cycle, shorten the split time in direction 1 a little.

(8) If the flow of traffic heading in direction 2 at the intersection is slower than in the previous cycle, lengthen the split time for direction 1 a little.

Next, the words "many" and "lengthen" expressed in each control rule are expressed using membership functions according to each control rule (see FIG. 5).

Next, the inference process of the inference section 12 will be explained using FIG.

The current intersection information obtained by the intersection information grasping unit 11 is input to the antecedent part of each control rule in the control rule description unit 13, and the degree of adaptation to the control rule is determined by comparing it with the membership function of each control rule (step 21)
.

Next, the obtained fitness is compared with the membership function of the consequent part to obtain an output for the rule (step 22).

These processes are expressed by the following equations.

The subscript i is the i-th rule that has the same control target (split time, cycle time) described in the consequent part of the rule, and the membership function of the antecedent part of the control rule is A1 (X), the consequent part is The number of memberships in the department is B□
(y), the input to the antecedent part is x,', and the fitness of the rule antecedent part is ω.

ω□=A(x□.) It is expressed as

The inference result of the rule is expressed as the membership function B”(y)
Expressed as:

B”s(y)=a+xn(ωx+B i
(y)) (see Figure 6).

Step 23: Combine the outputs obtained from each control rule.
), the control amount (split time, cycle time) is determined by determining the center of gravity of the synthesized function (step 24).

These processes are expressed by the following equations.

The synthesis of the inference results of multiple rules is as follows: B'(y)=max (B'x(y)yB''z(
y)+-B'n(y)), and the control amount becomes y'=fB'(y)ydy//B'(y)dy (see FIG. 7).

In the embodiment described above, the inference unit 12 processes the intersection information that changes sequentially and quickly determines the optimum control amount for the next cycle by fuzzy inference, so that it is possible to sufficiently cope with sudden congestion. Furthermore, by making the control rules more intelligent, intelligent traffic light control becomes possible.

The present invention has been specifically explained above based on examples, but
It goes without saying that the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the spirit thereof.

〔Effect of the invention〕

As described above, according to the present invention, the optimum cycle time of the next cycle is determined based on the intersection information of the previous cycle.
Calculate split time in real time using fuzzy reasoning. Since the combinations of cycle time and split time are infinite based on the intersection information of the previous cycle, more flexible control is possible than with conventional finite pattern classification.

This allows appropriate and prompt traffic control.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a traffic signal control device according to an embodiment of the present invention, and FIG. 2 is a block diagram illustrating the functions of a control section in the arithmetic processing device shown in FIG. 1. Fig. 3 is a processing flowchart of the inference section, Fig. 4 is a diagram showing an example of installing vehicle detectors at an intersection, and Fig. 5 is for explaining the membership function representing the expression "many" in the control rule. Figure 6a is a diagram for explaining the calculation of the fitness of the antecedent part of the control rule, Figure 6b is a diagram for explaining the calculation of the output of the consequent part of the control rule, and Figure 7 is a diagram for explaining the calculation of the output of the consequent part of the control rule. FIG. 2 is a diagram for explaining calculation of a control amount based on outputs of consequent parts of a plurality of control rules. In the figure, 1,111,112,113...vehicle detector, 2
... Arithmetic processing unit, 3... Signal controller, 4... Traffic light, 11... Intersection information understanding section, 12... Reasoning section, 13... Control rule description section.

Claims (2)

[Claims] (1) In a traffic signal control method that performs point control at intersections where traffic volume changes sequentially, the optimal cycle time and split time of the next cycle are calculated in real time by fuzzy reasoning based on the intersection information of the previous cycle, and the optimal cycle time and split time are calculated in real time. A traffic signal control method comprising controlling a traffic signal using a cycle time and a split time of the next cycle. (2) a vehicle detection means for detecting the approximate number of vehicles waiting at a traffic light at an intersection; a traffic flow detection means for determining the next cycle of traffic flow at the intersection from the vehicle number detection means;
A congestion level detection means for determining the degree of congestion at an intersection from the traffic flow detection means, and a split time and cycle for the next cycle which are intersection control parameters from the output of the congestion level detection means at the intersection and the degree of adaptation of each control rule of the traffic signal. A traffic signal control device comprising an inference means for determining time in real time using fuzzy inference.